Making Money With Option Strategies - Michael C. Thomsett

Download as pdf or txt
Download as pdf or txt
You are on page 1of 269

Making

Money With Option Strategies


Making Money With Option Strategies
Powerful Hedging Ideas for the Serious Investor to Reduce Portfolio
Risks

By Michael C. Thomsett
Copyright © 2016 by Michael C. Thomsett

All rights reserved under the Pan-American and International Copyright Conventions. This book may not
be reproduced, in whole or in part, in any form or by any means electronic or mechanical, including
photocopying, recording, or by any information storage and retrieval system now known or hereafter
invented, without written permission from the publisher, The Career Press.

MAKING MONEY WITH OPTION STRATEGIES


EDITED BY JODI BRANDON
TYPESET BY DIANA GHAZZAWI
Cover image: Nick M. Do/istock
Cover design by Rob Johnson/Toprotype
Printed in the U.S.A.

To order this title, please call toll-free 1-800-CAREER-1 (NJ and Canada: 201-848-0310) to order using
VISA or MasterCard, or for further information on books from Career Press.

The Career Press, Inc.


12 Parish Drive
Wayne, NJ 07470
www.careerpress.com

Library of Congress Cataloging-in-Publication Data

CIP Data Available Upon Request.


Contents

Introduction: Solving the Time and Proximity Issues


1: The Basics of Options
2: The Hedging Concept and Its Application
3: Option Valuation and Portfolio Risk
4: Speculation With Options vs. Conservative Strategies
5: Charting and Trade Timing
6: The Basic Covered Call
7: The Uncovered Put: Alternative to Covered Calls With Less Risk
8: Hedging With Spreads
9: The Butterfly and Condor
10: Collars and Synthetic Stock
11: Straddles Hedged
12: Rolling and Recovery Strategies
13: Avoiding Early Exercise of Short Options
14: Collateral and Tax Rules for Options Trading
15: The Final Twist: Proximity
Glossary
Bibliography
Index
About the Author
Introduction Solving the Time and Proximity Issues

Options trading is a paradox.


It can be highly speculative or highly conservative, or both, depending on
when and how it is used. This paradox can be employed to respond to the
unceasing symptom of investing in the stock market: Those who buy shares of
stock worry, no matter how price moves. If price moves up, should you take
profits now? If price moves down, should you cut your losses or buy more
shares? Owning stock is a troubling activity because of the uncertainties it
involves. Options—with the paradox they bring to the picture—can solve some
of the risk issues for you.
A lot of focus in the market is on short-term trading opportunities, and these
exist without any doubt. However, the more permanent value of options trading
is not in short-term profit potential, but in how options can reduce risk in your
portfolio. At the same time, reducing risk and generating income is an elegant
combination of features. The flexibility of options is the great advantage; as a
hedge against risk, an options position can also generate income and enable you
to take profits without needing to sell shares. For most traders, identifying high-
risk versus low-risk strategies is where emphasis is placed. This is not a simple
matter, because the rapidly changing stock and options markets define emerging
risks and opportunities that change from day to day and even from minute to
minute. Within this ever-changing situation, today’s high-risk option can be
tomorrow’s low-risk solution.
This suggestion—that “risk” is situational and not position-specific—is one
way to look at options. In this book, the idea is demonstrated through
examination of strategies designed to hedge positions, reduce risk, and generate
income.
The distinction between speculative and conservative is not merely an issue
of which strategy you employ, but when and in what proximity a price is found.
“Proximity” in this sense refers to how close the strike of an option (the price at
which it can be bought or sold) is to the current price of the underlying stock. It
also refers to how close price is to the top (resistance) or bottom (support) of the
trading range.
Reversal is most likely to occur when price is close to resistance or support,
especially when price moves through these range borders. Reversal back into
range is very likely; the trading range brings structure to the price chart, and
significant events such as breakout above resistance or below support point the
way to trading opportunities.
This is where options play a role.
Chapter 1 of this book begins with a summary of the basics: terminology,
terms of each option, and how trading works. For many traders, this is familiar
ground, but for many more it is new and perhaps even mysterious. The purpose
of this chapter is to set the foundation for understanding options. Because these
are complex trading devices with exotic jargon and rules, many traders just stay
away. But this means they miss opportunities to reduce risk and manage
portfolio exposure in all types of markets.
Throughout the book, as terms are introduced for the first time, they are
shown in boldface and italics. Every term is also defined in the glossary at the
end of the book.
The next topic is hedging, covered in Chapter 2. A hedge is nothing more
than a method for offsetting risk, consisting of a limitation or even elimination of
exposure. Following this is an explanation of how option valuation works in
Chapter 3. The “moneyness” of options (referring to the fixed strike price versus
ever-changing value of the underlying stock) is one element in valuation. The
other element is time remaining until the expiration of the option.
Chapter 4 offers an examination of the differences between speculation with
options and conservative application of options. Unlike most products, options
cover the entire risk range and can be tailored to fit any risk profile. In this book,
the assumed profile is that of an investor with a portfolio of stocks, whose
interest is in developing conservative hedging strategies.
Following this, Chapter 5 explains charting and trade timing. Though this
topic is usually of interest only to day traders and swing traders, even a buy-and-
hold investor benefits by being able to recognize the exaggerated price
movement of a stock in specific circumstances; for example, after an earnings
surprise, price tends to overreact, only to correct itself within two to three
sessions. Knowing this, a conservative investor can use options to create short-
term income and even out the momentary price movements—all with little risk.
Chapter 6 begins with what has become the favorite options strategy among
traders: the covered call. This is a strategy combining capital gains on stock,
dividend income, and premium from selling a call. The risks are low, explaining
the popularity of the strategy. However, there is a lot to say about covered call.
Not only should they be opened on carefully selected stocks; there are very real
risks as well. With any strategy, including covered calls, knowing the risks
before entering the strategy is essential as part of a conservative strategy.
An interesting twist on covered calls is introduced in Chapter 7. The
uncovered put seems at first glance to be the opposite of a covered call. In fact, it
is in many ways an identical strategy with the same market risks. This chapter
explains the risks of uncovered puts and compares them to the covered call.
In Chapter 8, the discussion is expanded with introduction of the spread. This
is the simultaneous opening of two different options with dissimilar strikes,
expiration dates, or both. The risk levels vary considerably, so the full range of
spread types is explained here with careful attention paid to the risk level and
where spreads help you to manage your portfolio positions.
Chapter 9 expands on the basic spread with an examination of two types: the
butterfly and the condor. Both are described in this section as 1-2-3, a reference
to opening of three different positions with different expiration dates. This is an
advanced set of strategies, but for those willing to monitor positions, they can
offer a variation on the portfolio that introduces some interesting cash-generation
possibilities.
Chapter 10 introduces more variations: collars and synthetic stock hedges.
These present opportunities to cap risks when stock prices turn volatile.
Chapter 11 explains how straddles work. These are viewed by many investors
as overly risky, but they do not have to be. These devices—opening two options
with the same strike and expiration date—can be designed to provide potential
profits with minimum risks.
Chapter 12 explains the “roll,” a technique in which one option is closed and
another opened with a later expiration date. This is done to avoid exercise. The
roll is appropriate in many circumstances, but not in all.
Chapter 13 moves into new territory by explaining how to recover once the
underlying price declines.
Following this, Chapter 14 covers collateral and tax rules for options trading.
Chapter 15 discusses importance of proximity between price and technical
signals, and how this affects timing of options trades.
The whole range of options trading is complex, but only because so many
investors have not learned the rules, terminology, and risks involved. Once these
are mastered, options trading is more easily understood and mastered. None of
the aspects for options trading is beyond your abilities, even though the
complexity often is what gets emphasized. This book remains basic and non-
technical in how the issues of this topic are explained. The purpose is to
introduce a range of hedging methods and to demonstrate how everyone can
learn options and apply them with full awareness of—and management over—
the risks.
1 The Basics of Options

The options market is characterized by specialized jargon and terminology. This


chapter explains all of the terms used and places them in context for you, as an
investor. Beyond definitions, you also need to grasp the essential trading rules
and to be able to read options listings found online or in the financial press.
This chapter presents a broad overview of the options market as a starting
point for folding an options strategy into an equity portfolio; identifying specific
risks; and understanding how to mitigate or remove an equity-based market risk.

Attributes of the Option Contract


Options are intangible contracts, granting their buyers specific rights (and
imposing obligations on sellers). The amazing attribute of options is that they
can be used in many ways, covering the spectrum from highly speculative to
highly conservative. Most investors cannot be classified as strictly speculative or
conservative, but tend to operate within a range of risk levels. These levels
change based on the circumstances, including market conditions, stock prices,
and the amount of cash in a trading account.
With these variations in mind, options are perfect vehicles due to their
flexibility. The degree of risk you can undertake based on how you use options is
not fixed any more than your risk tolerance. The leverage of options is very
attractive as well. However, depending on how that leverage is applied, it can
increase or decrease your risk.

Key Point
Options are intangible contracts granting specific rights to their buyers and
obligations to their sellers.
For example, options typically cost 3% to 5% of 100 shares of stock. So
buying a single option is a highly leveraged way to benefit from favorable stock
price movement—or to suffer the risk of unfavorable movement. The percentage
of option cost varies due to the specific terms of that option.
The flexibility of options is one of the primary attractions among investors. In
addition to the pure speculator, many conservative investors with a buy-and-hold
portfolio will trade options with a small portion of capital, as a form of “side
bet” on the market. This not only brings up the chance for added profits, but also
allows investors to take advantage of price movement in their stocks. Rather
than sell to take profits, options can be used to capture those profits without
giving up stock. And when a stock price is likely to decline, options can also be
used to limit risks. In other words, options are flexible enough to allow you to
manage portfolio risks while continuing to hold stocks in your portfolio.

The Leverage Benefit (and Risk)


Because option values are determined by price movement in the stock itself,
the skillful use of options as a leverage tool presents many opportunities. For
many, the option is an alternative to actually owning stock, so as a purely
speculative tool, the leverage appeals to this group of traders. However, leverage
also provides hedging value by setting up risk limitation as well as alternative
forms of profit creation based on portfolio positions.

Key Point
Leverage is generally not conservative because it involves borrowing
money to invest. Options are the exception, a form of leverage that does not
involve borrowing.

Leverage is normally associated with borrowing and, in that regard, most


forms of leverage are also high-risk investing strategies. Borrowing money to
invest does not seem to most people like a prudent decision. However, even the
most conservative investors trade on margin, meaning they can buy 100 shares
of stock with 50% cash and 50% leverage. So even when you consider yourself
very conservative, you might be using risky leverage in your own margin
account.
This means that every investor trading in a margin account is exposed to the
risk of leverage through borrowing. This approach might seem wise. You can
buy 100 shares of a $50 stock for only $25 per share; as the stock price rises, the
return on your $2,500 cash investment is twice as much as it would be when
paying all cash. However, if the stock price declines, the loss also is accelerated.
So if the $50 stock falls to $42, you lose $800, or 32% of the $2,500 you put at
risk. However, you still owe your broker $2,500. Your leveraged debt is $2,500,
but the cash portion of your investment has dropped to $1,700.
This demonstrates that leverage based on borrowing money means that both
profits and losses are accelerated. So leverage (meaning borrowing money) can
represent considerable risks. These risks are removed when you trade options as
hedges against your portfolio. You can pay cash to buy options at a small
percentage of the cost of 100 shares, and the most you can lose is the amount
you pay, never any more.

Terms of Options
To completely understand how options provide hedging benefits, you need to
master the jargon of this industry. Every option is uniquely defined by its four
standardized terms. These terms define the option and always work in the same
way, meaning all of the terms apply to all listed options (thus, they are
standardized). So when you buy or sell an option, you know exactly what your
contractual terms are for that asset.
The four terms are:
1. Type of option. There are two, and only two, “types” of options: calls and
puts. A call grants its owner the right, but not the obligation, to buy 100
shares of stock, at a fixed strike price and by or before its expiration. A put
grants its owner the right, but not the obligation, to sell 100 shares of
stock, at a fixed strike price and by or before its expiration.
2. Strike price. This is the fixed price at which a call or a put can be traded.
This price remains fixed for the entire life of the option regardless of the
stock’s price.
3. Underlying security. Every option is tied to a specific stock or other
security (such as a stock index or exchange-traded fund, for example).
This cannot be changed during the limited lifetime of the option.
4. Expiration date. This is the month and date when each option ceases to
exist. Every option is identified by expiration month. In addition, listed
options expire on the third Saturday of that month, and the last trading day
is the third Friday.
Expression of an option is quite specific and is based on these four
standardized terms. Here are two examples:
JNJ Oct 95 c (Johnson & Johnson, call with a 95 strike price, expiring in October)
MCD Mar 100 p (McDonald’s, put with a 100 strike price, expiring in March)

The stock symbol for each stock (JNJ or MCD, for example) is used in an
options listing or description. The expiration month is normally reduced to a
three-digit summary without a period. The strike is always expressed at the price
per share but without dollar signs.

Key Point
Options are expressed in a specific form of shorthand. Mastering these
expressions is essential in options trading.

If the option is not a round dollar per share value, it is expressed as dollars
and cents to two digits, also without dollar signs. So if the strike is 99.50, that
means the strike is equal to $99.50 per share. In describing options and stocks,
the use of dollar signs is always used to explain the price per share of stock, but
never options. So a 99.50 option on a stock currently priced at $99.75 is how the
situation is expressed.

The Price of Options


The price of each option is called its premium and it is always written as the
price per share. So if the premium of a 99 call is $215, it is expressed as 2.15.
Expanding the listing of an option to include the premium value, the following
examples include premium:
JNJ Oct 95 c @ 1.40 (Johnson & Johnson, call with a 95 strike price, expiring in October, with
current premium value of 1.40, or $140)
MCD Mar 100 p @ 7 (McDonald’s, put with a 100 strike price, expiring in March, with current
premium value of 7, or $700)

Figure 1.1 illustrates the terms of the option.

Figure 1.1: Terms of the Option


Like most securities, options also are expressed at both bid and ask prices.
The ask is the price you pay to buy the option, and the bid is the price you
receive for selling the option. For example, the JNJ Oct 95 c @ 1.40 is the bid
price for that option, or what a seller receives. And the MCD Mar 100 p @ 7
describes a put worth $700.

Long and Short Positions


Expanding beyond the listing, every option can be either bought or sold. The
bid price (what sellers receive) and the ask price (what buyers pay) are included
in every options listing. When you buy an option, you are long; when you sell,
you are short. The distinction is one of sequence. A long position is the well-
known “buy-hold-sell” sequence. The short position is the opposite, or “sell-
hold-buy.”
This reversal of the sequence is confusing for many investors accustomed to
first buying a security and then later selling. However, you can open a position
that is either long or short with options, and the risks are different for each. Just
as a buyer has the right to buy or sell 100 shares, the seller is exposed to the
possibility of exercise, meaning a call owner will “call” 100 shares and the seller
is required to deliver those shares at the strike, even when the market value is
much higher. It also means a put owner will “put” 100 shares to the seller,
meaning the seller is required to accept 100 shares at the strike, even when the
market value is much lower.
The buyer of an option enters an opening trade, called “buy to open,” and a
later a closing trade, called “sell to close.” Everyone who has bought and sold
stock is familiar with these definitions. However, for those who enter a short
position by opening with a sale, the opening trade is called “sell to open” and the
closing trade is called “buy to close.” These distinctions are important because
the distinctions—buy versus sell and open versus close—define the action you
take each time you trade an option. Many traders describe the closing of a short
option as “buying back” the option. This is misleading and confusing, because
the buy to close occurs based on the initial opening of a short position. There is
no “buying back” action because the trader never owned the position to begin
with.
To compare buying and selling consider the important differences between
calls and puts and between long and short, illustrated in Table 1.1.

Table 1.1: Terms of the Options


Call Put
Grants the right but not the Grants the right but not the
obligation to buy 100 shares of the obligation to sell 100 shares of the
underlying stock before expiration underlying stock before expiration
and at the fixed strike and at the fixed strike
Increases in value if the underlying Increases in value if the underlying
security’s price rises security’s price falls
Long traders want the underlying Long traders want the underlying
price to rise price to fall
Short traders want the underlying Short traders want the underlying
price to fall price to rise
Long positions are opened with a “buy to open” order and closed with a sell
to close” order
Short positions are opened with a sell to open” order and closed with a buy
to close” order

Where This Gets Confusing


Anyone new to jargon is going to struggle to grasp the concepts. With
options, the opposites—call versus put, long versus short, buy versus sell—can
be very difficult to put into context. To aid in clarifying these differences, keep
in mind that:
• A call is the right to buy, and the put is the right to sell.
• Buyers of calls and puts gain the right to buy or sell, and sellers have an
obligation to accept exercise if and when it happens.
• A long position starts with a “buy” order and ends with a “sell” order.
• A short position starts with a “sell” order and ends with a “buy” order.

Key Point
Options involve a series of opposites, so understand this is the key to
mastering options trading.

These definitions are of crucial importance in developing an understanding of


the many potential hedging strategies you can apply to your portfolio. Among
the difficulties faced by those new to options is the concept of profiting from a
price decline. Most investors grasp the idea that investment is profitable when
prices rise. However, thinking of the put as the opposite of a call, it becomes
clearer why the put becomes profitable when the stock price falls.
The difficulty of jargon becomes clearer when specific strategies are
introduced and aided by examples of outcomes. Price direction defines risk. So
options working as hedges for portfolio positions (usually meaning long stock
held in the portfolio) can involve either calls or puts, opened as either long or
short trades.

The Option Premium’s Three Types of Value


Every stockholder understands that stock has a single value: the price per
share. This changes daily based on many influencing factors, but the value of a
share of stock is universally agreed upon. With options, however, it is not as
simple.
There are three distinct and separate types of value that make up the total
premium of the option. Once you understand how these values interact, you will
have a clearer understanding of why option-based hedging works so well. The
influences on changing option value are not based only on movement of the
underlying stock, but also on volatility and time.
Volatility of stock is often represented by beta, a measurement of how a
stock’s price follows or responds to the larger market. This comparison is made
between the stock and a benchmark index like the S&P 500. A beta of 1.0
indicates that the stock will rise and fall at 100% of the rise or fall in the
benchmark. A higher beta equals higher volatility, and a lower beta equals less
responsiveness or lower volatility.
This is all relevant to option premium because the underlying stock’s
volatility (called historical volatility) is going to show up in the option as well.
Whereas volatility is a clear factor in levels of option value, proximity between
the option’s strike and the underlying stock’s price is another factor. So there are
three key factors adding to value of the option: volatility, time, and proximity
(between strike of the option and price of the stock). This proximity is called the
moneyness of an option.
Every option may be in the money (when the stock price is higher than a
call’s strike or lower than a put’s strike); at the money (when the stock price is
exactly the same as the option’s strike); or out of the money (when the stock
price is lower than a call’s strike or higher than a put’s strike). Figure 1.2
illustrates the moneyness of options.

Figure 1.2: Moneyness of Options

The chart demonstrates the moneyness of calls and puts. With strikes of 95,
calls and puts are at the money (ATM) when the underlying stock is worth $95
per share. The in-the-money (ITM) and out-of-the-money (OTM) status are
opposite for calls and puts.
Key Point
The “moneyness” of options determines option pricing and, more to the
point, also identifies levels of risk.

With moneyness of options, it is easy to determine whether an option is in or


out of the money. This leads to the definition of the first type of option premium:
intrinsic value.
This value is easy to identify. For example, with a strike of 45, a call is 3
points in the money when the stock price is $48 per share. So the call has 3
points ($300) of intrinsic value. If the stock price is $45, the call is at the money;
and any price below $45 per share means the 45 call is out of the money and has
no intrinsic value.
For puts, the same rules apply but in the opposite direction. A 45 put is in the
money by 2 points if the stock price is $43 per share. At a stock price of $45 per
share, the put is at the money. And any time the stock price is higher than $45,
the 45 put is out of the money and has no intrinsic value.
In trading options, the moneyness demonstrates that when proximity is close,
there tends to be a more immediate reaction between overall option pricing and
stock movement. When an option is in the money, intrinsic value changes point
for point with movement in the stock. However, this does not mean the overall
premium value tracks the stock precisely. The two other forms of option
premium, time value and volatility, work together to also adjust the premium
levels of options. With time value, increases in intrinsic value may be offset with
a decrease, so the overall premium is not entirely responsive to stock movement.
And volatility value also affects premium in either direction, depending on
proximity and time.

Key Point
Time value is a depreciating form of price, and the closer the option is to
expiration, the more rapidly it decays.

Time value is the portion of an option’s premium directly related to the time
remaining until expiration. It declines over the lifetime of the option, reaching
zero by expiration day. As expiration approaches, the decline in time value
(called time decay) accelerates as well.
Time value (like intrinsic value) is completely predictable. The curve of time
decay increases toward the end of the option’s life cycle, taking time value down
to zero on the last trading day. This is shown in Figure 1.3.

Figure 1.3: Time Decay of an Option

Time value affects overall premium in predictable ways, and you can see this
effect in a study of an option’s listing. For example, On October 1, 2015, IBM
was trading at $142.64 per share. At that time calls and puts with three different
expirations were trading at the following prices, as Table 1.2 indicates.

Table 1.2: IBM $142.64 – option premium levels


Several interesting observations can be made about this summary. First, the
differences in premium between the 140 and 145 strikes occurred because of the
moneyness of the options. With a stock price between the two strikes, the 140
calls were all in the money; and the 145 puts were all in the money. The large
differences in premium between these two strikes is explained by intrinsic value.
For the 140 calls, intrinsic value is 2.64 (142.64 – 140.00); and for the 145 puts,
the intrinsic value is 2.36 (145.00 – 142.64).

Key Point
A comparison of option values in and out of the money demonstrates how
and why prices vary even for the same expiration month.

Beyond intrinsic value, there remains a combination of time value and


implied volatility. The effect of time value is easily observed. In the case of both
calls and puts, overall premium increases with time. So 15-day, 51-day, and 79-
day time remaining to expiration demonstrates the role of time in setting overall
premium.
The final type of value is where all of the uncertainty comes into play.
Implied volatility (IV) is the portion of option premium that varies based on
volatility in the underlying stock and time remaining until expiration. This type
of premium may also be termed extrinsic value.
Some options insiders lump time and intrinsic value together and call it “time
value” collectively. This is a misleading characterization because time value is
very specific, and intrinsic value may rise or fall during the lifetime of the
option. When the two are classified together, it is possible for time value to
increase based on higher volatility, and this confuses the distinctions between the
roles of volatility and time. By keeping them separate, a focus on analysis can be
limited to implied volatility, where all of the uncertainty will be found.
The distance to expiration determines how implied volatility works. When
many months remain, movement of intrinsic value often is offset by changes in
implied volatility, so very little overall change in premium will be seen. But the
closer to expiration, the more “cause and effect” you see in the IV premium.
Volatility declines toward the end of the option’s life to a point where it—like
time value—has little, if any, impact on the price of the option. By the last
trading day, an in-the-money option will contain almost all intrinsic value
without time or volatility. Any out-of-the-money options will have very little
value remaining, and the further out of the money, the lower the premium value.

Key Point
Implied volatility (IV) is a reaction to the volatility in the underlying stock,
but also to the time remaining until expiration of the option.

For options with several months remaining, volatility can be expected to


change often, and rapidly. When the stock’s volatility increases, so will option
premium (due to its implied volatility). So as a timing strategy, many investors
and traders try to time entry and exit based on changes in the implied volatility
levels, buying at low volatility and selling at high volatility. This makes sense,
but the rapid changes and unpredictability of volatility make this a difficult task
to accomplish consistently. Combining volatility analysis with a study of the
underlying stock’s price chart may add more clarity and predictability to the
timing of trades.
Volatility is studied by a series of calculations collectively called the Greeks.
This name is used due to the definition of each with letters from the Greek
alphabet. These calculations help determine the volatility levels of options and
the level of reaction by options to the price movement in the underlying stock.
The first among the Greeks is Delta. This measures the change in option
premium in comparison to a change in the underlying stock. A reaction level is
affected by time remaining until expiration as well as by the moneyness of
options. So an option whose strike is closer to the stock price is likely to be more
responsive than one that is deep in or out of the money.
Delta is the most-used of the Greeks. It is a measure of the change in option
premium relative to the change in the underlying stock. It ranges from –1.00 to 0
for puts, and from 0 to 1.00 for calls. With volatility at mid-level, an at-the-
money option should have a Delta of 0.50 (for calls) or –0.50 (for puts). This is a
fair measurement of whether a current option premium is reasonable. When at
the money, if the call’s Delta is higher or a put’s delta is lower, it indicates an
increase in volatility.
When the option moves in the money (meaning the stock price moves above
the call’s strike or below the put’s strike), Delta should change by one-half of the
degree of movement. So a 1-point move in the underlying stock would be
accompanied by a 0.50-point change in Delta. As long as volatility levels remain
unchanged, this rule will apply. Any variation points to higher or lower volatility.
So Delta is a good measurement of this relationship.

Key Point
Delta is a reliable measure of the relationship between option price and
stock market value.

As expiration approaches, Delta tends to change its rate of reaction to the


stock price. However, a related concept, volatility collapse, makes volatility
unreliable to measure during the final month of an option’s life.
Closely related to Delta is Gamma, a measurement of the degree of change in
Delta. When options are deep out of the money (usually defined as 5 points or
more distance between strike and underlying price), Delta is usually quite low;
but as those options move toward the money, Delta increases. The degree of
increase is measured by Gamma.
Theta is a separate type of Greek. It measures the decline in time value.
Although the tendency to accelerate time decay is entirely predictable, the speed
of change is not the same for all options. So some experience rapid time decay
while others move more slowly.
Theta normally is higher for out-of-the-money options when implied
volatility is also higher than average. This would imply that volatility affects
time value. However, because time and volatility value are often placed together
as a single attribute of price, this indication is false. The change in what appears
to be time value often is change in IV, which is a separate price element.
The next Greek is Vega. Although not a Greek letter, it is included as a
“Greek” for the purpose of option analysis. Vega measures the rise or fall in an
option’s premium caused by IV. Two factors can be seen with Vega. First, it will
change even when the underlying stock price does not move, due to changes in
volatility. Second, as expiration approaches, Vega tends to decline. This can be
best used to identify the speed of volatility collapse.
Calculation of the Greeks can be complex. The best way to find these
indicators is to use the free calculator provided by the Chicago Board Options
Exchange. This calculator enables you to identify all of the Greeks and to track
their change over time; and also to identify levels of implied volatility.

Resource
To calculate the Greeks and implied volatility, use the free calculator at
http://bit.ly/28dQfkN

Calculation of Option Return


It might seem that figuring out the return on an option trade would be quite
simple. In fact, it is a challenge. In order to compare two or more trades, three
inhibiting factors arise. First, what is the basis for the calculation? Second, how
long was a position open? Third, do you base the calculation on options only, or
do you include dividend income and capital gains as well?
Determining the Basis

Key Point
Return on options is a complex matter that involves several attributes, such
as time and basis; the role of dividends and capital gains is also part of the
difficulty in pinning down how to make these calculations.

The basis price for calculating returns should be used for each and every
trade, so that the outcome will be expressed on the same terms. If you use the
original cost of stock as the basis, you will get many different outcomes based on
how much change has occurred between purchase date and option trade date; so
this is not a reliable basis. The same problem is found in the current value of
stock. This value will change by the time the option is closed or expires, so the
outcome for different positions will also be inconsistent. The only reasonable
basis for options trade is the strike. This price does not change, and it will be the
price used if and when the option is exercised.
Making the Outcome Accurate
The basis is a good starting point, but how long will the position remain
open? In order to reflect outcomes for two or more different options trades, you
need to annualize the return. This means each return is calculated as if the
position was open for one full year. This is necessary because a return achieved
over a short period of time is worth more than the same net return over a longer
period of time.
To demonstrate how this works, refer again to Table 1.2 on page 19.
To determine which option is a more profitable choice, do you pick the 15-,
51-, or 79-day contract? At first glance it looks like the 79-day call with a 140
strike is the best choice because it is worth significantly more money. However,
when you annualize, you discover that shorter-term options yield much better
annualized returns.
The steps in annualizing are:
1. Calculate the net return, dividing option premium by the strike.
2. Divide the return by the holding period.
3. Multiply the result by a full year.
To accurate annualize returns, use the proper value (in the case of a long
option, use the ask price; for short options use the bid). Next, estimate the
transaction costs for options. Most online discount brokers charge approximately
$9 for a single option, so throughout this book that is the trading cost level that
will be used.
It is important to also understand that the cost is reduced for multiple
contracts. For example, after the first option, the cost for additional contracts is
quite small, so the more options traded, the lower the transaction cost. For
example, Charles Schwab applies $8.75 for a single option and $0.75 for each
additional contract traded (per www.schwab.com).
Key Point
Annualizing is necessary to express option returns for different holding
periods in a truly comparable format.

Table 1.3: IBM $142.64 – option premium levels

For illustration of how this works, the following example is based on single
option trades and rounding of transaction costs up to $9.00. Referring to the
chart and assuming the sale of the 140 calls, the bid price for each were reported:
October (15 days) 4.40
November (51 days) 6.60
December (79 days) 7.25

When an option is sold, the transaction cost is deducted from the bid price
(and when an option is bought, the transaction cost is added to the ask price).
Converting these bid price premium values to dollars and then subtracting the
transaction cost:
October (15 days) $440 – $9 = $431
November (51 days) $660 – $9 = $651
December (79 days) $725 – $9 = $716

Next, calculate the return based on the 140 strike for 100 shares:
October (15 days) $431 ÷ $14,000 = 3.08%
November (51 days) $651 ÷ $14,000 = 4.65%
December (79 days) $716 ÷ $14,000 = 5.11%

The initial outcome still makes it seem that the December options are the
most profitable. However, once annualized, the outcome changes:
October (15 days) 3.08% ÷ 15 days x 365 days = 74.95%
November (51 days) 4.65% ÷ 51 days x 365 days = 33.28%
December (79 days) 5.11% ÷ 79 days x 365 days = 23.61%

Once the returns are annualized, it becomes clear that the shorter-term
positions yield significantly higher returns. However, the purpose of annualizing
is to make accurate comparisons, and should not be used to assume the outcome
you should expect in your portfolio. The calculation does not promise you the
amazing return of 74.95% every year on your options trades, but it does provide
a good measurement of which trades are most profitable.
What Else Should Be Included?

Key Point
Capital gains should definitely be considered in the overall outcome, but
option returns have to be kept separate to avoid distortions.

Annualizing returns adds accuracy to comparisons between option trades. But


beyond the option, what about dividends and capital gains? These are certainly
going to be significant in the overall return from options trades, especially if the
trade is based on a combination of stock ownership and options trades. In each
hedge discussed in this book, the assumption is that stock is held in the portfolio,
and options trades accompany these positions. So dividends and capital gains
cannot be ignored.
Capital gains on stock are often a big factor in overall return. However, in
comparing options, the capital gain should be kept separate. If you buy 100
shares of stock at $25 and the value grows to $40 by the time you begin trading
options, the 15 points represents a 60% increase in value. However, if you buy
those shares at $38 per share and trade 40 strike options, you have only 2 points,
or 5.3% gain based on the $38 purchase price. So capital gains can distort an
otherwise-accurate comparison between different options.
Dividends often are included in the calculation of option profits. This is called
the total return method, meaning the total of options and dividends together. In
most calculations, you first calculate the annualized return and then add in the
dividend yield.
To make the point about the impact of dividend, Table 1.4 compares three
different stocks and their prices and option premium in early October 2015:

Table 1.4
Number of options Cost Cost per option
1 $ 8.95 $8.95
2 9.70 4.85
3 10.45 3.48
4 11.20 2.80
5 11.95 2.39
10 15.70 1.57

With 15 days remaining until expiration, the option returns (adjusted for
trading costs) are:
ConocoPhillips (COP) 0.82 ÷ 48.50 = 1.69%
Occidental Petroleum (OXY) 1.42 ÷ 66.00 = 2.15
Exxon Mobil (XOM) 0.66 ÷ 73.00 = 0.90

To annualize and add dividend yield:


ConocoPhillips (COP) 1.69% ÷ 15 x 365 = 41.12% + 6.17% = 47.29%
Occidental Petroleum (OXY) 2.15% ÷ 15 x 365 = 52.32% + 4.55% = 56.87%
Exxon Mobil (XOM) 0.90% ÷ 15 x 365 = 21.90% + 3.93% = 25.83%

Based on this analysis, the October short calls are most productive for
Occidental Petroleum, using the total return method. However, the dividend
calculation has to also be made with awareness of when quarterly dividends will
be paid. It is possible to enter into a trade with no dividend yield. This occurs
when the time between entering the position and the expiration of options
includes no ex-dividend date (the date on which the current dividend is no longer
earned). In spite of the above calculations using 15-day options, none of the
stocks receives a dividend during the period in question.
It is also possible to earn two full quarterly dividends when holding a stock
from just before one ex-dividend date until the following ex-dividend date.
Rather than a full two quarters of 180 days, this could occur in approximately 91
days, or about one-half of the six months associated with two quarters of
dividends. So in calculating total return, it is crucial that the actual number of
quarterly dividends be considered with the timing of dividend earnings dates in
mind.

Key Point
Quarterly dividends are earned as long as you are the stockholder of record
when dividends are earned. This means you can pick expiration dates to
maximize dividend yield.

The Option’s Life Span


Many types of options are issued on company stock. The traditional listed
option lasts as long as eight months. However, many companies also have
weekly options, which expire in between the monthly cycles. On the opposite
end of the spectrum, some options last as long as 30 months. These are called
LEAPS options (long-term equity appreciation securities).
Every company with listed options is classified into one of three annual
cycles. This identifies the months in which options expire. The cycles are for the
months of January, April, July, and October (JAJO); February, May, August, and
November (FMAN); and March, June, September, and December (MJSD). For
each cycle, options exist for all of the quarter months and, in addition, options
will be found for every month during the year with expiration within 30 days. So
there is great flexibility and variety within the range of listings.
Another consideration for options trading is the distance between strikes.
Lower-priced stocks tend to have options at every round dollar amount. Options
trading up to $100 per share have 5-point increments; and above $100, most
trade every $10. However, for shorter-term expirations, the 5- and 10-point
increment stocks may also have increments every 2.5 points or even every point.
The closer increments of strikes adds to the flexibility of trading, especially for
strategies relying on minimal price movement.

Collateral Requirements
Most stock investors are familiar with the concept of trading on margin. You
can buy shares of stock paying one-half and getting the rest on margin. This is a
form of leverage that can be a great advantage, but also comes with risks.
However, “margin” is different for options trading. The margin for options is
calculated depending on the type of option and on whether or not an offsetting
position (in stock or other options) is open at the same time.

Key Point
Margin for options is not a form of borrowing as it is for stock purchases;
option margin identifies the collateral you must maintain.

A covered call (a short call opened when you also own 100 shares of stock)
contains no margin requirement because, if exercised, the shares are called away.
However, for all short options, a margin requirement applies. As a general rule, a
single short option requires 20% to be placed on margin, adjusted by the
premium received and also adjusted based on the moneyness of the option.
For example, you sell an uncovered put with a strike of 30 when the stock is
at $32 per share. You receive a premium of 2 ($200). The margin requirement is
calculated as:

20% of current stock value of $32 per $640


share
Minus proceeds of the option 200
Margin collateral $440

In this example, the put is out of the money. If a put is in the money, the
margin requirement is calculated by adding the proceeds. For example, you sell
an uncovered put for 4 ($400) with a 30 strike when the stock at $29 per share,
or 1 point in the money:

20% of current stock value of $29 per $580


share
Plus proceeds of the option 400
Margin collateral $980

The next chapter expands on the basics by exploring how hedging works and
how options protect stock positions in your portfolio.

Resource
To calculate margin requirement for any position, use the free calculator
provided by the CBOE at www.cboe.com/tradtool/mcalc/. To learn more
about how margin is applied, download the free Margin Manual at
www.cboe.com/learncenter/pdf/margin2-00.pdf.
2 The Hedging Concept and Its
Application

Options traders struggle endlessly with risk. Long-term options are expensive
and short-term options expire too soon. How do you profit in this environment?
Hedging is an expansion of options combination strategies, designed to solve
this problem. This is a conservative program for options used to manage a stock
portfolio. By creating offsetting hedge positions, traders are able to generate
profits in three conditions: underlying price rises, they fall, or they do not move
at all. Safety within the hedge is created by the offsetting long/short and call/put
option structures.

The Most Basic Hedge


With emphasis on profits in the markets, it is easy to overlook the equally
important aspect of investing: protecting capital. This refers not only to the
capital you have investing in building a portfolio, but also to protecting profits
accumulated in well-chosen stocks. The dilemma is unceasing: Do you take
profits when they occur, or do you wait for more?

Key Point
The desire for profits is accompanied by the equally important need to
preserve capital. This is where hedging enters the picture.

If you chose stock well and want to continue holding shares, why take
profits? If you take profits now, where do you invest next? The problem every
stockholder faces is this question of when (or whether) to take profits. Those
who do often end up with a portfolio full of shares that have not become
profitable. This means that, by taking profits, you have removed the best
positions from your portfolio and are left with your capital tied up in paper
losses.
The hedge is the solution to this problem, and the most basic of hedges
involves using options to generate income while protecting paper profits. With
option hedging, it is also possible to take profits without needing to sell shares
you would prefer to keep.
The first hedge to review is the covered call. In Chapter 6, a detailed analysis
of the covered call’s opportunities and risks is presented. For the purpose of the
overview of hedging in general, the covered call is the most popular hedge and
the most obvious way to augment your equity portfolio. This position, involving
100 shares offset by the sale of one call, sets up a very advantageous situation
with three sources of profit: capital gains on stock, premium income from the
short call, and dividends.
For many, the most basic option strategy is to buy either a call or a put. A call
is bought when the stock price declines, on the theory that the price will
rebound. If that occurs, the cycle of price changes in the stock results in added
profits from buying a call and then selling it at a profit. A put is bought when the
stock price rises, based on the same theory: The cycle of stock pricing is likely to
come back around and leads to a decline. At that time, the long put can be sold at
a profit.
For example, consider a choice of buying a call at 3 ($300) or buying a put at
4 ($400) on McDonald’s Corp (MCD). With a strike of 95, the breakeven point
for the call is at $98 per share; and the breakeven for the put is $91 per share.
These are illustrated in Figure 2.1.

Figure 2.1
Considering the general range of this stock over six months, there were times
in which the breakeven was reached on either side. In the case of the put, the
only example occurred in late August, when a marketwide decline occurred
briefly. So buying calls and puts is largely a matter of timing. It also relies on
skillful selection of options based on both price and expiration. For example, if
the call was purchased in late May with a June expiration, it would never reach
its breakeven. Even though the call would be in the money most of this time
(above 95), the cost of the call inhibits the price from ever exceeding the
breakeven. Between purchase date and expiration, time decay erodes premium
value and so the breakeven would never be reached.
For the put, the breakeven at $91 per share was never reached with the brief
exception of August and September. In this case, buying a put in late May with
an August expiration would be a losing trade. The last trading day in August was
August 21, and the big drop in price did not occur until August 24. So even with
three months between the date of purchase and expiration, the 95 put would end
up expiring worthless.

Key Point
Long option trading is a challenge due to time decay and volatility in the
underlying stock, which affects cost. Long options, consequently, by
themselves are poor hedges.

These examples reveal the problem with long options. They are intended to
hedge the short-term volatility in the underlying security, but that works only if
the stock is volatile enough to yield a profit. Even if it is, you still have to decide
whether to speculate on the upside (with a long call) or the downside (with a
long put). In the example of McDonald’s, the stock moved 5 points higher only
in one session over six months; and it only moved more than 2 points below $95
per share in the large market decline, after which the stock price returned into its
established range.
The point is that in deciding to hedge a stock’s price with long options, there
has to be a balance between price and volatility. A low-volatility stock will be
hedged with low-priced options, and a high-volatility stock’s options will be
more expensive. This tendency, a form of point spread, makes it difficult to time
options profitably.
The theory behind speculating in long options is logical at first glance. A
trader may rationalize that there is plenty of time before expiration and
breakeven is only a few points away. However, long options are poor hedges due
to one inhibiting factor: Time value declines, so that the value of the long call or
put will fall as expiration approaches. So even if the stock price moves in the
direction you need in order to make the option profitable, growth in intrinsic
value is going to be offset by decline in time value. If you buy a very short-term
option it will be cheap; however, time decay is accelerated and you will need a
significant change in the stock’s market value just to overcome the time value
problem. If you buy a longer-term option, it will be more expensive, meaning
you have to put more capital at risk. In that case, you still need significant point
movement in the stock to offset the cost of the option.
The problems associated with long options—both short-term and long-term—
make them poor hedges. It is difficult to produce a profitable outcome in the
basic long option, so for consistent hedging you need to look beyond. This is
why the covered call is the most popular option strategy. It solves two problems.
First, because the call is short, a decline in time value is advantageous. A large
number of covered calls are closed at a profit due primarily to the decline in time
value. So it is sold to open and later bought to close for less, producing a net
profit.
The second problem solved by the covered call is profit-taking. When a
stock’s price rises, the temptation is to take profits. However, selling a covered
call at a strike near the current price yields an attractive premium. If the stock
price retreats, the call can be closed at as profit. If the stock price remains at or
below the strike, the short call will expire worthless. If the stock price rises, the
call will be exercised and three forms of profit result: The call premium is yours
to keep after exercise, so it represents a total profit. As long as the strike is
higher than your basis in the stock, you also earn a capital gain when the stock is
called away. Finally, you profit from all dividends earned while you held the
stock.
This seems to present a winning situation no matter what happens. However,
the covered call is not without risk. If the stock price falls below your net basis,
you end up with a paper loss. Net basis in a covered call is the price per share,
discounted by the sale of a covered call. When that occurs, the call will expire
worthless but you end up with depreciated stock. This can be rationalized by
arguing that without the covered call, the stock would have lost value and that
the premium from the call reduces the paper loss.

Key Point
Covered calls are a popular and basic form of hedge, but as a first step, the
company should be picked based on its fundamentals.

This does not mean that covered calls are poor hedges. In fact, they are
excellent hedges when the alternative of simply holding stock is considered. If
you are a long-time holder of shares, the paper loss will be of less concern than
if you just want to get a profit and get out of the position as soon as possible.
The market risk of the covered call is mitigated by careful selection of a
company whose stock you purchase. Every stockholder is wise to qualify a
company based on its fundamental attributes and historical volatility, as a first
step. This applies whether you just hold shares for the long term or you write
covered calls. By picking exceptionally strong companies (in terms of
fundamental strength over many years), the likely gyrations in stock price are
going to be less of a factor than if the stock you buy is a fundamentally weaker
company, whose price volatility is much higher. This is a common mistake made
by investors who want to write covered calls. Intent on getting the highest
possible premium, they are drawn to higher-risk companies. Consequently, they
get a rich premium for the covered call, but they are exposed to higher market
risks as well. A prudent covered call strategy works best when the company is
selected as a first step based on its fundamental strength. The options will not
yield as high a return for these more stable companies, but overall, the lower
market risk makes this a sensible strategy.
A closely related hedge, and another alternative to buying calls or puts, is the
uncovered put. This is counter-intuitive at first glance. Most investors consider
any uncovered option to be high risk, but the uncovered put is not. Any option
that is “uncovered” is not offset by a different position. The covered call is
considered a safe strategy because the short call is covered by 100 shares of
stock. If the call is exercised, 100 shares are called away at the strike. A short put
cannot be covered in this way, so it has to be uncovered.
The uncovered put’s market risk is identical to the market risk of the covered
call. When you consider the risks of both, it is a downside movement that
presents problems, both for the covered call and the uncovered put. In the case of
the covered call, a decline below net basis creates a paper loss. For an uncovered
put, a decline below the strike represents an exercise risk. Most investors think
of covered calls in terms of upside risk, but this is inaccurate. The thinking is
over-simplified. It goes like this: If the stock price declines, the call loses value
and will eventually expire worthless, and that yields a profit. However, if the
stock price rises above the call’s strike, shares are called away and the added
profit (points above the strike) is lost. So this lost opportunity risk is where the
covered call emphasis is placed.
If the call’s strike is higher than your basis in stock, exercise is one of several
possible outcomes and produces a net profit. So it is a positive outcome.
However, you should enter into a covered call strategy only if you are willing to
have shares called away.
For the uncovered put, the perception in the market is also misplaced. The
thinking is that if the underlying price declines below the put’s strike, the put
could be exercise, resulting in being forced to buy shares above current market
value. This perception of the uncovered put as high risk is inaccurate. The put
can be closed or rolled forward to avoid exercise. Considering that a price
decline produces the same loss in either a covered call or an uncovered put, the
characterization of the put as high risk is not accurate. As an alternative to the
covered call but with the same market risk, the uncovered put is more flexible
because it can be rolled without concern for the strike. Because there is no stock
at exercise risk, the put can be replaced with a lower-strike put. The risk of the
short put occurs when the stock price falls many points and results in exercise.
This is the same risk as owning 100 shares of stock; the result of a decline is a
loss of value in both cases.
Table 2.1 summarizes a comparison between covered calls and uncovered
puts.

Table 2.1
Covered Calls Uncovered Puts
The call is matched with 100 shares No stock is held.
of stock.
Stock can be purchased with 50% Uncovered puts require collateral.
margin.
Dividends are earned at long as No dividends are earned.
stock is owned.
Stock is called away when its price Stock is put to the seller when its
is higher than the call’s strike. price is lower than the put’s strike.
In-the-money calls can be rolled In-the-money puts can be rolled
forward, but the strike should be forward to any strike.
higher than the net basis in stock.

Key Point
The covered call and uncovered put are stronger basic hedges than long
calls or puts.

So as a basic hedge, the covered call and uncovered put are the starting points
for a conservative hedging program. The more widely held belief that long calls
and puts represent the basic conservative hedge overlooks the actual risk: With
time decay, the long option is a long shot. If you buy an option expiring in the
near future, time decay is accelerated. If you buy an option with more time
remaining, the time value makes it a more expensive position to enter.

Combining Calls and Puts in a Hedging Position


Some traders recognize that stock prices are likely to move, but they are not
sure of the direction. In this case, the combination of two or more options has a
certain appeal. However, many of these strategies are speculative. There is a big
difference between speculation and hedging.
Speculation is risk-taking, making a move with high risk in the hopes of
making a big profit. In comparison, hedging is a more conservative set of
strategies designed to place a cap on risk and also generating additional income.
So in considering a combination of calls and puts, this distinction has to be
kept in mind. Some market insiders claim that risk cannot be avoided and is a
fact of life. These insiders believe that investors have to diversify their holdings
or enter into a program of asset allocation. Diversification is a worthy concept,
but it does not always work. It often leads to bland returns or even losses. For
example, the mutual fund industry, based entirely on a diversified portfolio
placed under professional management, has a dismal record. Funds traditionally
have performed below the benchmark market indexes. “Because of their
excessive annual fees and poor execution, approximately 80% of mutual funds
underperform the stock market’s returns in a typical year,” according to an
article on Motley Fool (Bill Barker, “The Performance of Mutual Funds”).
So what is all of the fuss about diversification? The professional management
of mutual funds has notoriously been unable to deliver a net benefit better than
market averages, so investing in an index fund that tracks the market is probably
a better way to go.
Asset allocation is a similar tactic to diversification, but attempts to divide a
portfolio into holdings by type (stocks, bonds, real estate, commodities).
However, asset allocation is a responsive tactic, often determining how to divide
up holdings after the problems have emerged. In many applications, asset
allocation is not truly effective, but is used in some financial planning programs
to convince clients that expertise can and does beat the market.

Key Point
Portfolio management based on diversification or asset allocation
acknowledges the need to spread risks. Hedging challenges this and
presents methods for reducing or eliminating risk.

Simply accepting risk as unavoidable is what leads to the decision to buy


mutual fund shares, or to trust an institution to mandate how your holdings
should be allocated. The idea is to accept the inevitability of risk and to spread it
out. In comparison, hedging does not accept this idea, but presents a different
argument: Risk can be limited or eliminated with the right strategies.
So rather than accepting risk as inevitable, or trusting a mutual fund
management team to manage risks for you, the alternative is to seek ways to
actually limit risk. This is by no means a revolutionary idea; it is simply the
recognition that risk can be managed, anticipated, and controlled with hedging.
When options traders consider this idea, do they pick a speculative set of
strategies or do they design a true hedging strategy? A speculative move
involves taking a gamble that a stock price will move enough points before
expiration, to exceed a breakeven price. Some speculative strategies include:
1. Long spreads and straddles. In its most rudimentary form, the spread
involves a long call and a long put with different strike prices and the same
expiration date. The risk level of the straddle is the combined cost of both
options, plus transaction fees. To be profitable, the stock price has to move
more points in either direction than the cost of the position. The point
move must occur above the call’s strike or below the put’s strike.
The straddle is a similar configuration, but the long call and put have
the same strike. So the degree of price movement is not as much as the
spread, but accomplishing a profit remains challenging. The soon-to-expire
straddle is cheap but price has to move quickly; the later-expiring straddle
has more time but is more expensive.
2. Short spreads and straddles. For those investors wary of long options
and the disadvantage of time decay, the short spread or straddle appears to
offer a solution. The fact that you receive money for opening a short
position means that as long as price does not move too far, it will become
profitable. The problem here is that risks are considerable and the
maximum profit is limited.
A short spread involves selling a call and a put with different strikes,
which expands the profit zone. In comparison, a short straddle is opened
with both sides at the same strike. Because both sides are short, the
position is highly speculative.
3. Uncovered calls. The covered call is considered a conservative strategy,
but the uncovered call is one of the most speculative strategies available.
In this position, you sell a call but you do not have stock. The risk is
unlimited, at least in theory. The potential loss occurs when a stock’s price
rises above the call’s strike. The loss is calculated as the difference
between market value and strike, minus the call premium. For example, if
you sell a 35 call for 4 ($400) and the stock rises to $48 per share, the call
will be exercised. The loss is: $4,800 – $3,500 – $400 = $900
All of these speculative applications of options contain a common element:
None of them have any influence on an equity portfolio position. All exist
outside of the stock portfolio and are based on an attempt to make a profit based
on price movement, in exchange of accepting a high (often a very high) risk that
the price will move in the wrong direction or will not move far enough.
In comparison, a combination that hedges portfolio equity positions is more
conservative because risks are managed as part of the options positions that have
been opened. Chapters 8 through 12 examine the combination hedges in great
detail, and demonstrate the differences between high-risk speculation and
conservative portfolio management.

Key Point
Speculation and hedging are entirely different strategies. A speculator takes
risks, whereas a hedger manages risks.

The short straddle is high risk because no matter how the underlying price
changes, one side or the other is always at risk. This problem is solved with the
covered straddle. Even though the call and put are both opened with the same
strike, the risk is not the same as the short straddle’s risk.
The covered call and uncovered put have the same level of market risk. The
actual risk is a downside risk. If the stock price falls, the uncovered put goes in
the money and is at risk of exercise; this makes the uncovered put risk identical
to the risk of owning stock. With the covered call, a decline beneath net basis
represents a net loss as well.
On the upside, the risk is quite different. The uncovered put will expire
worthless as long as the underlying stock’s price is above the put’s strike. The
covered call will be exercised if the stock price is higher than the call’s strike.
This is not going to generate a loss as long as the call’s strike is higher than the
original basis in stock. Even so, the in-the-money covered call can be closed or
rolled forward to avoid exercise.
So the covered straddle introduces two different positions with identical
market risk. Why not just open two uncovered puts or buy another 100 shares
and sell two covered calls? The answer is found in the evolving changes in both
time and intrinsic value. For the covered call and the uncovered put, time value
erodes over time, accelerating as expiration approaches; so both positions are
advantageous due to time decay.

Key Point
The covered straddle creates two forms of option income while enabling
the investor to manage risks in a conservative manner.

On the intrinsic side, the two options act differently. If the stock price rises
above the strike, the call gains intrinsic value. If the stock price falls below the
strike, the put gains intrinsic value. In both instances, the opposite option has no
intrinsic value and can be closed at a profit or allowed to expire worthless. The
risk management attribute addresses the option that is in the money.
The call can be left alone and allowed to exercise, in which case the 100
shares of stock are called away at the strike. Or the call can be closed or rolled to
avoid or defer exercise. The put that has gone in the money can be closed to
avoid exercise, or rolled forward. It can also be left alone and allowed to get
exercised. This makes sense when the in-the-money points are lower than the
premium received for selling the put. For example, you opened a covered
straddle with a 35 strike and received 3 ($300) for selling the uncovered put. At
expiration, the stock price is $34 per share or one point in the money. The put is
exercised and 100 shares are put to you at $35 per share. This is one point higher
than current market value; however, because you received $300 for selling the
put, your net gain is still $200 points below the net:
35 strike – $34 per share = $100 loss
Put premium $300 – $100 loss = $200 net gain

After this, what else can be done? Having acquired an additional 100 shares,
you now have a total of 200 shares. This leads to additional hedging potential in
the form of covered calls or covered straddles based on two options and 200
shares of stock.
Another type of hedge involves long and short positions opened with the
same strike or with different strikes. The synthetic long or short stock strategy
sets up positions on the same strike and value mirrors movement of the
underlying stock. A collar involves 100 shares along with a short call and a long
put, both out of the money. This combines insurance with a covered call. Both
the collar and the synthetic stock strategy are explained in detail in Chapter 10.

Why Hedge?
Options traders are aware of the effect volatility has on option pricing. The
lower the stock’s historical volatility, the lower the option premium tends to be—
the relationship called implied volatility. So a low-volatility stock is going to
experience much different options pricing activity than a high-volatility stock.
You might have examples of both types of stock in your portfolio. So as a
means of managing risk for both types of stock, various options-based hedging
techniques not only help manage risk, but may also generate additional income
without adding substantially to levels of risk. An investor holding a portfolio of
fundamentally strong stocks will not be concerned with short-term volatility, at
least in theory. The idea is to identify quality based on fundamental strength
(revenue and earnings growth, high dividends, low or level long-term debt), and
wait for the long term. However, even the most conservative investor with the
ability to pick and buy high-quality stocks is going to be aware of short-term
price changes and may want to exploit those price movements to generate short-
term income.
This is at times a speculative move made by investors willing to take risks
with a small dollar amount of capital. However, at times, it also contains
elements of hedging as part of a conservative outlook. For example, you own
100 shares of stock in a company you consider a strong prospect for long-term
growth. The company pays a strong dividend of 3.5% and you reinvest your
quarterly in additional partial shares. The history of the stock price has been
strong as well. However, the latest earnings report included a negative earnings
surprise. The company missed its estimates of revenue and earnings by a small
amount. On the day of the announcement, the stock price fell 3 points.
The tendency for stock prices to overreact to earnings surprises (positive or
negative) can be observed reliably. As a stock price moves against the surprise, it
also tends to self-correct within two to five sessions. A fundamentally strong
company with low volatility and consistent returns may fall 3 points when it
misses analysts’ targets, but it is likely that the price will rebound very quickly.
In this situation, a buy-and-hold strategy dictates that no action should be taken.
However, even the most conservative trader can take many actions in this
situation to exploit what is probably an exaggerated price movement of short
duration. These actions can include:
1. Close any open covered call positions and take the profit. When the stock
rebounds, sell a new covered call. This takes advantage of the price decline
on both sides. Taking profits after the decline yields a small profit; opening
a new covered call after the rebound brings in a higher premium.
2. Buy a long call to benefit from a rebound in the stock’s price. This is a
speculative move based on the belief that the rebound is likely to happen.
As with all forms of speculation, it does not hedge your stock position, but
the timing of the move is prudent given the likelihood of a price rebound.
3. Sell an uncovered put to benefit from a rebound. Remember: The
uncovered put has the same market risk as a covered call. This is one of
those situations where the two sides, even with the same market risk, will
behave differently. When the stock price declines, it is not the time to open
a covered call, but it is a good time to open an uncovered put—both
observations based on the likely rebound in stock price.
Key Point
A popular form of hedging is timed to coincide with stock patterns
following earnings surprises. The tendency is for price to move in an
exaggerated manner and to correct itself very quickly.

The same strategic application of options strategies applies after a positive


earnings surprise. If the revenue and earnings exceed expectations, the tendency
is for the stock price to rise and to then retreat within a few sessions. At these
times, you can take advantage of the price change by closing any long calls
currently open. Or you can buy a long put to add insurance to your position, a
popular hedge that cements the gain in the stock price; this is also known as a
married put.
If the stock price does decline, you can sell the long put and take profits,
meaning the benefit from the exaggerated price movement is converted into a
new form of income. Finally, immediately after an exaggerated price movement,
the timing for opening a covered call is excellent. If the stock price declines, the
call can be closed and profits taken; if the stock price does not decline, the
higher strike translates to higher profits if the short call is exercised.
These examples demonstrate that, based on short-term price movement in
your stock positions, the risk management attributes of hedging are preservation
measures that either limit or erase paper losses. Hedging protects your stock
portfolio, and is a more prudent and profitable system than diversifying or asset
allocation. Under those systems, risk and loss are accepted as inevitable. With
hedging, you do not eliminate loss completely, but you do reduce its impact on
your stock positions. Whether you hedge to reduce loss or to generate additional
income, the results are the same: higher value in your portfolio and better
preservation against market risk.
Hedges are not guarantees. Most forms of hedging cost money or limit your
potential profits. So a hedge usually will involve a trade-off between cost and
risk reduction. This is a conservative policy as part of portfolio management,
because it is a means for limiting losses and the willingness to pay for the
protection or offset gained through the hedge. So the choice is not between
hedging and diversification. The real choice is in the degree of hedging you want
to put in place for the cost it involves. A buy-and-hold strategy may be entered
with the long-term view that in spite of short-term price movement, the well-
selected company’s stock will grow. Hedging may not be necessary, but it is a
method for managing the uncertainty of short-term price movement and also for
limiting longer-term risks.
Many theories have been developed to articulate and quantify the value of
hedging. In a “perfect” hedge, for example, the hedge offsets risk by exactly
100%. For example, a long put costs 3 ($300) and the stock declines by a net of
3 points. The loss in stock is offset 100% by the net cost of the put. This “inverse
correlation” is intended only to distinguish between profits and losses. A hedge
that costs more to open than the risk it eliminates is not worth opening, and a
hedge that generates a profit clearly provides added benefits. For the purpose of
comparing outcomes between hedges, the theory of the perfect hedge is useful.
It helps you to decide whether a particular hedging strategy is worth the risk.
The perfect hedge is not possible most of the time, but it sets a standard for
minimum outcomes. However, certain inhibiting factors may create a net loss for
hedges in spite of the intended risk-reducing benefits that hedging provides. In
these cases, the net loss can be thought of as the insurance you pay for to prevent
loss in the event the loss occurs. It is similar to homeowner’s insurance, in which
you protect against catastrophic losses, even though the chance of these
occurring is quite low.

Key Point
Some traders resist long hedge positions because it requires spending
capital. However, it is similar to other forms of insurance: Its value is in the
benefit if and when price moves against the stock position.

In the market, the factors leading to likelihood of negative outcome include:


1. Time decay of long options. Any hedge involving long options has to
consider the impact of time decay. Even when both sides remain at the
same risk levels, time decay erodes the market value of the long option.
This is why offsetting long and short options is desirable as a form of
hedge. Time value works against the long position, but it works in favor of
the short position. The net result is likely to be mitigated well enough so
that you do not suffer from time decay overall.
2. Changes in implied volatility levels. When IV is high, option premium
tends to increase. So if you open a hedge with long options at such times,
some of the value will decline as volatility retreats. If a hedge involves the
use of short options, the opposite problem occurs when the hedge is
opened when IV is low. If IV increases, so does the premium value of the
short option; this is a disadvantage because it means the option is less
likely to become profitable.
3. Price drift. In spite of earnings trends and supply and demand for stocks,
many theorists have observed that stock and stock index prices tend to rise
over time. This so-called price drift affects options as well, especially near
the money. For example, as a stock’s price drifts upward, the premium
value of long puts will decline.

Another View of Hedging


An argument can be made that you do not need to take specific actions in
order to hedge. For example, staying out of the market with a portion of your
capital kept in cash can be called a hedge. In practice, staying out of the market
hedges only the market risk, but not the equity positions you hold. So in that
regard, keeping cash may be conservative in an uncertain market, but it is not a
hedge under the usual definition.
Another way to hedge is through rotating stocks from one sector to another.
Every sector experiences cycles and many are easily identifiable. So the idea of
rotation is to move equity into company stocks most likely to benefit from the
high end of the cycle. As rational as this idea might seem, even if you can call
the cycles correctly, there is no automatic guarantee that stock prices will act or
react to cyclical timing. Beyond sector-specific trends, every individual stock
acts based on its own influences. These may be fundamental or technical, and
economic influences also play a role. For example, a company with a large share
of its revenue and earnings overseas may react more to currency exchange rates
than to its sector cycles. The economic influence is very difficult to anticipate
and is just as uncertain as the rise and fall in currency prices.
So even with the defensive character of rotation in mind, calling a move from
one sector to another is going to succeed or fail based on much more than
cyclical timing. Rotation often becomes more speculative than you would like,
so as a form of hedging, it is as reliable as diversification in other forms.
Rotation is not a true hedge, but merely an attempt to time positions based on
annual cycles.
One strategy is based on shorting equity positions. If you believe the market
in general is heading down, or that an individual stock’s price is weak, you can
either short the stock or invest in an “inverse equity” exchange-traded fund
(ETF). This type of ETF gains value when its basket of holdings loses value. It
is shorting the basket, unlike most ETFs and mutual funds, which hope for
increased value. Some inverse ETFs are also leveraged, meaning that for every
dollar of decline, the ETF holdings gain two or even three times the move. These
are speculative and often high-risk, and though some describe them as hedge
strategies, they do not provide the safer and less expensive benefits found in
options. For example, a long put is likely to provide the desired insurance
against price decline, perhaps more so than an inverse ETF. Even if you own the
basket of securities through ETF shares, most ETFs also allow hedging with
options. So rather than taking higher risks with inverse or leveraged ETF shares,
the simple solution of buying puts is both cheaper and safer.
With hedging in mind as a core strategy for the use of options, how do you
manage your portfolio? More to the point, how do you manage portfolio risks?
Options are valuable in this quest and may be used in a rich variety of strategies.
The elements of time to expiration, proximity between stock price and option
strike (“moneyness”), and even consideration of another type of proximity (stock
price and patterns in proximity to resistance or support) all affect the methods
selected to hedge this risk. In Chapter 5, this second form of proximity is
examined and demonstrated. The next chapter continues the discussion of
hedging with an analysis of option valuation and its relationship to portfolio risk.
3 Option Valuation and Portfolio Risk

Investors are constantly looking for the signals pointing to higher or lower risks.
This is especially true after taking positions in stock. The expression “Wall
Street climbs a wall of worry” is accurate.
The reason for this concern about risk—specifically the risk that a newly
opened position will decline in value—can be traced back to an assumption
some investors have, whether verbalized or not. This assumption is that the entry
price is the zero value or starting point. In other words, if you buy share at $45,
you expect the price per share to rise, but not to fall. The optimistic assumption
is destructive, however. Every price is part of an unending struggle between
buyers and sellers, and this means the price may rise or it may fall.
Everyone knows this, of course. But something more important than
knowledge is how you act on that knowledge, or, when it comes to stocks and
risk, how you may easily overlook some of the obvious forms of risk. For this
reason, options are tools for managing portfolio risk and for identifying the
actions to take using options to mitigate, offset, or eliminate risk.

Key Point
The entry price of any trade is part of constantly changing prices, not your
individual “zero” price.

The Nature of Market Risk


Every investor has experienced the negative side of market risk. A position is
opened and the price immediately declines. Now what? Do you take your losses
quickly; wait out the price, hoping for a rebound; or try to hedge the position?
These are core issues for every investor. Risk often appears when least
expected, even though it should always be expected. It’s the nature of investing.
Even so, one form of risk, which may be called conformity risk, is the tendency
to go along with the majority. In the market, this often is a big mistake. The
“herd mentality” of the market often is wrong, and at the very least, its belief
often is based on the wrong assumptions.
The tendency to conform is strong. Human instinct tells you that the majority
must be right just based on the numbers; but the majority is not always right. In
some cases, the majority is dead wrong, and that’s the point. Why does a
majority make decisions as it does? The answer: The majority acts on emotions
and not on logic. The prevailing market emotions are fear and greed. The
majority (the “herd”) becomes fearful when prices fall, so the emotional impulse
is to cut losses and sell even at a loss. When prices rise, the majority acts
emotionally once again, in greed. With prices on the price, the majority develops
the opinion that they will miss out on more profits if they don’t act immediately.
The outcome of this herd mentality is to buy high and sell low, instead of the
opposite sage advice to buy low and sell high. As obvious and logical as this
sounds, the market acts on emotions and not on logic.
Conformity risk, as much as market risk, explains why so many investors end
up timing their trades poorly. The alternative is to practice contrarian investing,
which is developing a logical basis for when and why you trade. Many think a
contrarian goes against the majority just to be different, but this is not the case.
Contrarians are coldly logical and analytical, and as a result they make fewer
mistakes and time their trades based on facts rather than on feelings. This
discipline also helps the contrarian to ignore and avoid crowd thinking.

Key Point
Contrarians are able to resist the common problems based on opinion,
because they restrict their decision-making to logical analysis.

Contrarian investing brings up a related spectrum of risk. By proceeding


logically instead of emotionally, you can avoid some of the common mistakes
investors make. For example, some investors fall into the trap of feeling superior
to the crowd. Anyone who starts out with a series of winning trades is
susceptible to this misguided sense of superiority. There are numerous problems
with this, including a tendency to ignore warning signs of poor judgment. For
example, a trade is made with poor timing and it becomes apparent right away.
Do you cut your losses and get out? Or do you wait for price to turn around? An
investor with a sense of superiority will not be willing to accept input, whether
from others investors or from the market itself.
Closely related to a sense of superiority is stubbornness. If an investor
develops an opinion and acts on it successfully, how can that stubbornness be
changed? For example, many investors in the past were stubborn about their
loyalty to a specific company, such as General Motors, Eastman Kodak, or
Enron. All of these examples, once considered stellar blue chip investments,
failed. However, long before they failed, warning signs were there. A contrarian
is able to recognize logically what those warning signs reveal; a contrarian is not
blindly loyal to a company or to an idea. The same logic applies to strategies. A
favorite strategy is not going to work in every case, and stubborn loyalty to that
strategy is blinding.
Contrarians are also aware of the problems of confirmation bias. This is the
tendency to seek confirmation of a previously established idea without
remaining objective. For example, you invest in a company’s stock based on
many positive signals. The stock begins a slide downward, and you are
concerned. But in studying the chart, you find many signals that confirm your
bullish opinion about the company and the stock. This is a lack of objectivity,
when the smart choice would be to either cut losses or turn to options to hedge
any further declines. A true contrarian will understand confirmation bias and be
able to avoid it with cold logic.
Contrarians also understand the final type of risk, that of magical thinking.
An investor who follows this thinking believes that when making a trade,
wearing a lucky shirt or undergoing a specific ritual will lead to success. It is not
a rational mindset, but it is common among investors just as it is among faithful
fans of sports teams.

Key Point
Anyone relying on magical thinking to invest makes the mistake of
assuming that being a good person means they deserve profits and being a
bad person means they deserve losses.

Magical thinking also contains an element of religion. The ritual is part of a


belief that being a good person leads to a reward in the form of profits. This is
not faith in the religious sense, but superstition. But it feels like faith, and the
“true believer” may not consciously examine the illogical assumptions being
made.
All of these forms of risk are less tangible than the well-understood market
risk. All contain a similar element, however. They are forms of risk common
among the majority of “the market,” where the herd mentality rules the decision-
making process. In comparison, the contrarian approach involves making
decisions logically and based on analysis of both fundamental and technical
trends; and then making trade decisions based on observed trend movement and
reversal.

Leverage Risk
Another well-known form of risk involves a combination of capital and
borrowed money. Any time you borrow to invest, you are taking on risk. If a
leveraged account rises in value, profits also rise; but if that account loses value,
the losses are accelerated as well.
For example, you can buy stock with 50% cash and 50% margin. This is one
widely used form of leverage. However, ask any investor if they would borrow
to invest and most will say they would not. So what happens to that 50% of
leverage? If you buy 100 shares at $60 and you place $3,000 in your margin
account, the remaining $3,000 is loaned to you by your broker. If the price rises,
this leveraged investment also becomes profitable, as Table 3.1 shows.

Table 3.1

As the value of shares rises, the percentage of the total remaining leveraged
falls. So the net profit is accelerated as well. With the original share price at $60
per share, paying 100% in cash yields a $1,000 profit when the stock price
reaches $70. But with 50% leverage, the same move in price produces $4,000 in
profit ($7,000 total value minus $3,000 borrowed).
This looks attractive, but what happens if the share price declines? Then the
overall picture is far different, as Table 3.2 indicates.

Table 3.2
Now, with the borrowed amount remaining unchanged, the original 50%
leveraged has risen to 60%. If you had paid 100% for the stock, the loss would
be $1,000 ($6,000 – $5,000). However, with leverage of one-half, the net loss
rises to $2,000 ($5,000 – $3,000).
Investors often buy shares on margin without stopping to think about how
much that increases their risk. Leverage in the form of margin provides greater
profit potential while also placing greater loss potential on the position.

Lost Opportunity Risk


The “lost opportunity” of what could have occurred is often on the minds of
investors. This comes in several forms.
For stock positions, a lost opportunity arises when prices rise and profits are
taken too early. The dilemma for anyone taking profits too quickly is not only
that an otherwise-sound investment is lost, but so are profits. If you purchase
stock because you believe the company is fundamentally strong and offers
growth potential in the future, then taking profits too early does bring up a lost
opportunity. If you were to close out all positions as soon as they became
profitable, your portfolio would end up with nothing but shares of stock that had
declined in value.

Key Point
Taking profits as they appear means you end up with a portfolio of
depreciated stocks.

That action—taking profits on worthwhile stocks while holding onto losing


stock—is poor portfolio management. It is self-defeating and so the lost
opportunity is the actual health of the portfolio.
In the alternative, it makes more sense to sell shares of stocks that have
declined or not moved at all, and reinvest in more shares of stock with a robust
growth curve underway. This emphasizes movement and profits, but those
profits accumulate. Once you take profits, you next have to decide where to
reinvest the capital.
Another type of lost opportunity is found on the other side of the portfolio
equation. If you fail to close out positions because the market value has fallen,
what do you do if it continues to fall even further? A stubborn investor might
hold on in the hope of a reversal, so that those paper losses turn into profits.
Unfortunately, this often results only in larger losses. The lost opportunity is the
limitation on losses that could have been put into place.
A third version of lost opportunity relates to options trading in several ways.
The most readily apparent way is the covered call. Those who oppose covered
calls argue that if the stock price rises, the call limits profit and the lost
opportunity represents the higher price per share that you would have had if you
had not written a covered call. However, anyone writing a covered call needs to
understand the pro and con of the position. On the pro side, you keep the option
premium and, upon exercise, you also profit from capital gains. Finally, you earn
dividends as long as you earn the stock. Properly selected covered calls yield
double-digit annualized returns. On the con side, the occasional loss of further
profits makes the covered call less profitable than just owning stock.
The problem with this version of lost opportunity it that is does not occur in
every case. Many covered call writers use the same 100 shares to sell and profit
from calls over and over, generating income from the call premium while being
willing to risk the stock being called away. It is a judgment call, but the emphasis
is on the wrong risk.
The true risk involved in covered call writing is not the lost opportunity risk
on the upside, but the market risk on the downside. This risk is the same as the
basic risk of just owning stock, but it is reduced by the call premium. So it is a
more desirable downside risk than just owning stock. When this is factored in to
the upside lost opportunity risk discussion, the value of covered call writing
becomes clearer, even with that lost opportunity risk.

Other Types of Portfolio Risks


The usual emphasis among investors is on market risk. This is
understandable. Of course, everyone wants their investments to become
profitable and also would like to beat the averages. Three specific additional
forms of risks come into the picture, however, and these cannot be ignored as all
influence whether or not it will be possible to generate profits.
Knowledge and Experience Risk

Key Point
You do not have to become an expert at reading financial statements to
master a few fundamentals. Using long-term trend summaries is adequate.

Knowledge and experience risk refers to every investor’s understanding of


markets. This includes everything from the economic limits on profit based on
supply and demand, to the larger analytical methods by which companies and
their stocks are studied. Fundamental analysis involves a large body of
knowledge about how financial statements are put together and what they reveal.
Fortunately for the average non-accountant investor, online brokerage services
offer nicely summarized fundamental trends in services like the S&P Stock
Report, provided free of charge by many brokerages.
Experience is closely associated with knowledge. For most investors,
“experience” is what happens when they were expecting something else. In other
words, losing money is a great form of experience, but it’s by no means
desirable. However, loss is instructive because it brings up the false assumptions
an inexperienced investor starts out with; the worst aspect of those assumptions
is that they are not obvious until a loss occurs. So some of the best experience is
also expensive, but a small loss early on in an investing career can prevent larger
losses later.
With options trading, knowledge and experience risk is even more severe than
for stock investors. Options are characterized by highly specialized jargon,
trading rules, and restrictions. Because options leverage stock price movement,
both profits and losses can occur rapidly. This is where selection of conservative
strategies is sensible, especially when the purpose is to manage portfolio risk
through hedging, and not just to make fast money through speculation.
Tax and Inflation Risk
Tax and inflation risk surprises many investors, because it is easy to overlook
the significance of these risks. Added together, inflation and taxes mandate the
net return you need from your investments just to break even. This breakeven
yield is higher than most people assume. It is a severe risk, for two reasons.
First, if your net return is lower than your breakeven rate, you are losing
spending power every year. Second, if you end up taking higher market risks to
reach or exceed breakeven, you face the prospect of greater losses as well.
This type of risk is well managed by hedging with options. Even the most
basic hedge, such as a covered call, is likely to offset the double impact of
inflation and taxes. The covered call is an excellent hedge for this reason. It
consistently yields double-digit annualized returns just from option premium, as
demonstrated by the following example.
On October 9, Exxon Mobil (XOM) was trading during the session at $79.68.
The owner of 100 shares may consider a covered call with a strike of 80.
Without considering capital gains or dividends, the premium income from the
covered call would be:
November (49 days) 80 call bid 1.92 (less $9 costs) = 1.83
Yield: 1.83 ÷ 80 = 2.29%
Annualized yield: 2.29% ÷ 49 days x 365 days = 17.06%
December (71 days) 80 call bid 2.35 (less $9 costs) = 226
Yield: 2.26 ÷ 80 = 2.83%
Annualized yield: 2.83% ÷ 71 days x 365 days = 14.55%
January (99 days) 80 call bid 2.78 (less $9 costs) = 269
Yield: 2.69 ÷ 80 = 3.36%
Annualized yield: 3.36% ÷ 99 days x 365 days = 12.39%

These outcomes demonstrate that annualized return for shorter-term options


annualizes to a higher annual return than for longer-term options. But do these
outcomes beat the double impact of inflation and taxes? Yes.

Key Point
Few investors have analyzed what they need to truly break even. The
breakeven yield for most people is higher than their average yield. This is
where conservative hedging with options helps.

To calculate your breakeven yield, identify the inflation rate you believe is in
effect. Divide that by your net income after deducting federal and state income
taxes. If your combined federal and state income tax rate is 36% (33% federal
and 3% state, for example), your after-tax income is 72% (100 – 36). If you
assume inflation is 3%, the calculation of breakeven requirement is:
I ÷ ( 100 – E ) = B

I is inflation and E is effective tax rate; breakeven is the rate you need to earn
to offset taxes and inflation. So in the previous example, the calculation is:
3% ÷ ( 100 – 36 ) = 4.17%
You need to earn 4.17% in your portfolio just to break even. Most investors
realize that to accomplish this rate or better, higher risks need to be taken on.
However, with options used not only to hedge risk but also to generate income
(with covered calls, for example), this breakeven rate can be surpassed easily. In
the example of Exxon Mobil, the annualized yield on a covered call was 17.06%,
well above a typical breakeven yield.
Table 3.3 shows breakeven requirements at various tax rates and inflation
rates.

Table 3.3

The hedge value of options used in even the most basic conservative strategy
is perhaps one of the only ways to beat the inflation and tax breakeven rate
consistently. Alternatives will usually result in higher potential yields
accompanied by much greater risks.
Impatience Risk
Impatience risk is a form of risk that most investors experience. In the desire
to earn profits quickly, and to move from one position to another, the market
does not always move at the speed desired. As a consequence, the impatient
investor closes one position and replaces it with another that has seen more price
action in recent weeks. By chasing profits like this, the likelihood of poorly
timed trade decisions increases. Any investor acknowledging their own
impatience will also recognize the risk that comes with it. An old investing
adage promises that “the market rewards patience.”

Hedging as an Alternative to Diversification


The widely accepted assumptions about investing are that risks are inevitable,
that they have to be spread among many different products, and that a diversified
investment program will beat the market averages.
To the first assumption, inevitability: Yes, risk is inevitable unless action is
taken to ensure positions, place a floor on potential loss, and otherwise hedge
portfolio positions so that risk can be managed or eliminated. If this is
accomplished, then risk and its inevitability are defeated through good
management.
The second assumption, that you have to diversify: There clearly is value in
investing capital in several different places, but diversification itself has to be
done effectively. For example, buying shares of three different stocks with
identical risk factors is not diversification. Secondly, over-diversification leads to
poor overall net returns, meaning that the investor (or mutual fund) ends up with
reports below the market average.
The third assumption is the most dangerous of all, that a diversified program
will beat market averages. As the previous exercise demonstrated, the double
problem of inflation and taxes showed you need an exceptionally large rate of
return just to break even. This means that under the usual program of
diversification, you need to take higher than average risks; unfortunately, this
means making or beating breakeven is much more difficult to accomplish
consistently.

Key Point
The popular assumptions (risk is inevitable, diversification is the solution,
and diversification beats the market) are all incorrect.

Investors relying on professional management through mutual funds might


beat the market in some years, but not in others. Buying shares of ETFs might
accomplish the same thing, but the risk there is that the basket of securities will
net out to an average return, which might not consistently beat the market.
Diversification is not the answer to the problem of wanting to maintain capital
value and reducing risk. The solution to the issues is to figure out how to apply
hedging consistently and effectively to portfolio positions.
When this is designed properly, the outcome is that conservative hedging
does beat inflation and taxes, while yielding returns above the average of the
market. Diversification is a sound practice just to avoid big losses due to
problems in a single security; in spite of how this idea has been sold to the
public, diversification is only one aspect of investing, but it does not solve the
risk challenge.
Coming chapters describe a limited number of conservative strategies to
hedge equity positions, or strategies that can be used as hedges in some
situations, but not in all. Some positions are high risk in some applications and
moderate or low risk in others. This is an essential aspect to recognize in
developing hedges. No single strategy can be labeled in terms of risk in every
case. For example, a covered call is a smart strategy for low-volatility stocks or
for stocks whose price has recently peaked (especially when the signals indicate
coming reversal). However, when a stock’s price has spiked in the opposite
direction and you expect price to turn upward, the covered call is not a wise
strategy. At that time, the uncovered put (with the same market risk as the
covered call) makes much more sense.
Another example is the uncovered spread, which is widely assumed to be a
speculative high-risk strategy. In this strategy, you sell a call and a put, so there
are two uncovered options open at the same time. As long as the price remains
between the OTM (out-of-the-money) strikes of the higher call and the lower
put, a profit will be earned, and both positions can be closed at a profit or
allowed to expire. But if the price moves beyond either strike, the exercise risk
becomes more severe. However, this can be a very low-risk strategy in one
specific situation. When a stock has been trending sideways in a consolidation
trend, it is described as “range-bound,” meaning neither buyers nor sellers are
able to move the price higher than resistance or lower than support.
Many investors view consolidation as a pause between trends, in which
trading cannot be done because there is no dynamic trend to reverse. However,
specific signals do foretell breakouts from consolidation. These include
triangles, wedges, and others. As a basic hedge, the existing consolidation that
lacks breakout signals may last many months or even years. For example, the
one-year chart for Johnson & Johnson (JNJ) shown in Figure 3.01 includes a
four-month consolidation trend from January to May. The end of this trend was
signaled by a double top forming above resistance; and following that, resistance
began declining against level support, forming a bearish descending triangle. In
September and October, price settled into a new consolidation range, including
strong new resistance formed by a flip of the support price. When a flip like this
occurs, the new range tends to be stronger than average.

Figure 3.1: Consolidation Trend


Key Point
Strategies usually considered high risk may be less risky based on the
circumstances in which they are used.

In this example of consolidation, an uncovered spread contains relatively low


risk because of the range-bound nature of the price over four months. The end of
consolidation was clearly marked by the failed breakout above resistance and
confirmed by the descending triangle. So during the period in which JNJ was
range-bound, an uncovered spread could have been opened once the
consolidation was established. For example, by mid-March, a position could
have been opened using short options with April expiration. Keeping expiration
within one month for short positions is wise, because time value falls an
accelerated speed in the final month; with consolidation possibly ending at any
time, the shorter-term options are manageable until the price range evolves and
changes (as it did beginning in June).
An example of an uncovered spread opened in March could include an April
105 call and an April 95 put. This strike range of 10 points is at or beyond the
range-bound area between resistance and support. Because consolidation tends
to be difficult to break from, this is not a high-risk strategy.
In comparison, an uncovered spread on a much more volatile stock would be
high risk. Figure 3.2 is a one-year chart for Alphabet (GOOG). The stock was in
a consolidation trend from February through June. During this period, three
breakout attempts all failed, but there was no specific price decline below the
established support price of $520. In mid-July, a 70-point upward gap moved
price from $580 to $650.

Figure 3.2

The volatility of Alphabet would have made an uncovered spread unwise. For
example, a short call with a 560 strike and a short put with a 520 strike would set
up a 40-point buffer zone, which sounds very safe, especially when that zone is
expanded for the premium received. However, the 70-point move in a single day
demonstrates that volatility stocks are poor candidates for uncovered short
positions.
The comparison between JNJ and Alphabet (GOOG) demonstrates that a
specific strategy may be conservative or high-risk depending on where and when
it is deployed. Alphabet is a high-volatility and high-priced stock, and big price
moves are not uncommon. Johnson & Johnson is a more stable, low-volatility
stock. Its one-year price range was less than 20 points from $90 to less than
$110, compared to Alphabet’s 200-point spread from $480 to $680.

Hedging and Market Inefficiency


In using options to hedge, whether in an uncovered straddle, a covered call, or
any other strategy, the degree of risk relies on volatility in the underlying, but
also on a tendency for markets to behave inefficiently.
As option expiration approaches, notably in the last week, underlying prices
tend to move toward the closest option strike. This phenomenon, call pinning
the strike, does not always occur, but it occurs often enough that it will affect the
value of soon-to-expire hedge positions.

Key Point
Finding trading opportunities does not always take a lot of time. Online
sources provide information you can use to time smart trades.

The short-term inefficiency of the market is even more glaring after even
small earnings surprises. One worthwhile options strategy involves tracking
earnings reports and looking for surprises, and then exploiting the overreaction
in price. With a small amount of research, you can reasonably identify
candidates for earnings surprises.
An example is shown on the chart of Pepsi (PEP) in Figure 03.3.

Figure 3.3

The procedure was not complicated. The day before earnings were reported
(October 5), an analysis of Pepsi’s earnings history revealed consistent positive
earnings surprises:
Quarter Surprise %
9/14 5.4%
12/14 3.7
3/15 5.1
6/15 7.3

Based on this, it was assumed that yet another earnings surprise would occur.
So the recommendation was to buy an October 94.50 call @ 1.33 (net cost
$142). One week later, the call could be sold @ 4.65 (after trading costs, a net of
$456), a profit of 221%. Why sell at this point? The chart revealed a number of
signs that the expected rise in price had peaked. The flip from support to
resistance was followed by a price move above resistance on the day earnings
were reported, including strong upward gaps. However, the Relative Strength
Index (RSI) moved into overbought (above the index value of 70), the first time
that had occurred over the six months shown on the chart. This combined set of
signals was enough to warn that a price correction might be on the way.

Resource
To check a company’s history of earnings, go to the NASDAQ Website
(www.nasdaq.com/symbol) and then add the company name. For example,
for Pepsi, the link is www.nasdaq.com/symbol/pep/earnings-surprise.

The fact that a stock’s price will react strongly to an earnings surprise is one
example of market inefficiency. In most instances, a strong upward move after a
positive surprise (or downward move after a negative surprise) will be followed
within a few sessions by a turn back to price levels near those before the
earnings announcement. By realizing this, you can take advantage of the
inefficiency in price activity by making relatively inexpensive options trades.
A long option expiring in the near future is not generally considered a
conservative strategy. However, in this situation, knowing how prices act after
earnings surprises make the long call an “educated guess” and a good one. The
same would apply after a negative earnings surprise. Anticipating a downward
move, you would buy puts and then sell after a decline took place. This could be
based on a track record of four past quarters in which outcome always fell short
of analysts’ predictions.
Specific to options pricing, volatility collapse, which works as another form
of inefficiency. Implied volatility also tends to spike before earnings
announcements, especially if a surprise (positive or negative) is anticipated.
However, among the many forms of inefficiency, options traders may observe
that near the end of the cycle (especially in the last week before expiration) any
type of reliance on volatility can be misleading. At this point in the cycle, it
makes more sense to time options trades based on technical analysis of the
underlying stock’s price chart.

Key Point
Options traders relying on volatility analysis are at risk as expiration
approaches. At this time, it makes sense to reply more on price charts of the
underlying security.

An example of how knowing this can produce profitable outcomes on very


short-term swing trades: Knowing that both volatility and time value will end up
at zero on the last trading day of the option cycle, analysis of option premium on
the previous day (the third Thursday of the month) will point out potential
bargains. For example, if an at-the-money or out-of-the-money option reports
strong growth in premium on Thursday, you know that it will disappear on
Friday. This points the way to trading opportunities, but these are going to be
limited.
Most instances of such premium spikes are going to be minimal, so that the
trading costs to enter and then exit a trade could consume part or all of the profit.
A single option should cost about $9 to trade, so a round trip cost will be about
$18. So if an option is only valued at 0.18 on the third Thursday, it will not be
worth trading. Making matters worse, the price spread, the difference between
bid and ask, will also reduce potential profits. So a spread of 0.03 with pricing of
0.21 nets out to breakeven at best based on the cost to move in and out of a
position.
Yet another form of inefficiency relates to the option’s time value. You know
that time value ends up at zero by the end of the last trading day. However,
because you cannot trade options when the market is closed, you also know that
time value will depreciate most rapidly on the last Friday. Whereas options
cannot be traded in off hours, time decay should be expected to apply even when
the market is closed. So the change between Thursday and Friday will be
significant. In examining the price of options on Thursday, any premium for at-
the-money or on-the-money options will consist of a mix between volatility and
time value. It is not possible to identify how much of each applies, but it doesn’t
matter either. You know that by the end of trading on the next day, both of these
values will be zero. The entire ATM or OTM premium will disappear the
following day.
The next chapter expands on the question of portfolio risk with a comparison
between speculative and conservative strategies. This comparison is difficult for
some investors, notably those who consider options as high risk in most
situations. The emphasis on identifying safe hedges points the way to success in
portfolio risk management.
4 Speculation With Options vs.
Conservative Strategies

In order to create a system in which options are used conservatively to hedge


portfolio equity positions, it is first necessary to settle on definitions. Many
traders describe themselves as conservative but at times execute highly
speculative trades.
This problem is especially applicable to the options world. With a vast array
of possible strategies, even experienced stock traders may easily fall into the trap
of violating their own investment standards. Options traders need to resist
temptation to veer from their self-imposed risk profile; this is not as a simple as
it sounds.

Key Point
Every investor faces the possibility of violating well thought-out standards
developed as part of a risk profile.

As a starting point in developing an effective program involving options as


hedges against portfolio risk, the definition of conservative has to be articulated.

What Is “Conservative”?
The definition of a conservative profile contains a degree of variation, based
on several factors. Many readers will be surprised to discover that risk profile
itself contains variables, and that an individual risk profile is going to change
over time based on these factors as well as on changing attitudes toward
investing. The key factors in this equation include:
1. Experience. If you have been trading for many years and have developed a
reliable system for picking stocks and determining when to continue
holding or when to sell, it is more likely that a similar level of discipline
will be applied in the addition of options to hedge. If you are an
experienced investor, you know that research and complete understanding
are essential in the creation of a safe program and in being able to hedge
based on well-articulated risks.
2. Capital. The level of cash available to invest also affects risk profile, even
though it might seem that this should not be a variable. However, if you
have a very small amount of capital, you will have to allocate those scarce
resources cautiously and cannot afford loss. The addition of options as
hedges is desirable but might also be further restricted by capital
limitations. If you have a larger dollar value to the portfolio, you also have
more flexibility to diversify, and can also afford losses more readily.
Adding option hedging is affordable in the larger portfolio. This does not
mean that losses are “acceptable” in a larger portfolio, but they are more
affordable. This changes the risk tolerance level as well.
3. Goals. The purpose for investing has a significant influence on risk profile.
This changes with age, as it should. If you are a younger investor, you are
concerned with job security and being able to start a family and buy a
home. If you are an older investor, you are more likely to think about
investing to build a retirement fund and, once children are grown and on
their own, you may consider some forms of insurance to be less important
than the younger investor. Goals affect your risk profile because as these
change over time, your attitude about levels of risk also changes. The
tendency is that as you grow older, you will tend to become more
conservative and have lower risk tolerance than you did when younger.
4. Portfolio contents and diversification. If you own 100 shares of one
company’s stock, you will benefit 100% from rises in price and suffer
100% from declines. This is not a portfolio but a position, a starting point.
There is not much you can do to diversify a one-stock position, but it can
be hedged. However, if you own shares of many stocks in different sectors,
your diversified portfolio will behave with a mix of price changes. In this
case, the positions in the portfolio can be hedged in different ways,
depending on historical volatility, percentage of holdings in the overall
portfolio, and the effectiveness of diversification. Most investors think a
broadly diversified portfolio makes it a safer and better-managed portfolio.
However, this is not always the case. Being able to hedge a more
concentrated mix of holdings is often more profitable than diversifying
widely and hoping that the mix of holdings will outperform the broader
market. So the diversification policy itself may increase risks, whereas
hedging a smaller, less-diversified portfolio may decrease risks.
5. Market conditions. Finally, all investors are at risk based on market
conditions. If you consider current market conditions relatively stable, you
might think hedging is not as necessary to protect your portfolio. If you
think the market is very volatile, hedging makes more sense. Ultimately, it
is not the volatility in the overall market that matters, but the volatility in
your portfolio positions. If you own stocks that all tend to be volatile, your
market risk is greater than average. However, even in a volatile portfolio,
conservative hedging can improve management over risk and allow you to
limit losses due to fast and unexpected price swings. Your selection of
high-volatility or low-volatility stocks defines your risk profile and
whether or not you are truly investing in a conservative manner.

The Role of Trends and Fundamentals in a Hedging Program


If option hedging can be effective to manage portfolio risk, it has to be based
on analysis of price trends in the underlying stock; and the levels of risk
requiring hedging depend on the fundamentals of each company. This includes
appreciation of overall trends in the market; however, the trend in the single
stock is a separate issue and—apart from market trends and volatility—serves as
the central feature of the hedging program. A starting point for this analysis is
comparison among many companies of the fundamental strength of the
organization and its long-term fundamental trends.

Key Point
A hedging program based on fundamentals of the underlying stock as a
starting point brings order to the selection process and reduces technical
risks.

Your portfolio, consisting of shares of stock in companies, was selected on


some basis. Most conservative investors look at the fundamentals—the dividend,
P/E ratio, revenue, earnings, and debt—in selecting stocks. These are worth
some limited analysis here, if only to demonstrate the differences between
companies in terms of how fundamentals define risk in your portfolio. To many
options traders, this seems contradictory. A preference for technical analysis
does not exclude the fundamental side as a requisite. Both analyses are enhanced
when the fundamentals are also studied.
Analysis of fundamentals is best performed as part of a trend over time.
Looking at any financial result for a single year does not reveal whether the
company’s fundamentals are improving or declining. A longer-term analysis of a
few key fundamentals reveals levels of fundamental volatility, a form of risk
going to the very beginning of how a portfolio is constructed and managed.
Many fundamental indicators are available and can be studied online without
needing to reconstruct financial statements. Many online brokerage services
provide free fundamentals summaries. Many allow their investment customers
free access to data-rich reports like the S&P Stock Reports, which includes 10
years of fundamental data for thousands of publicly listed companies.
These tests are essential to developing a hedging program. The level of
volatility in the company’s fundamentals is reflected in the volatility of the stock
price, which also represents market risk. Finally, the historical volatility in the
stock price affects option premium and, as a result, the cost or profit from a
hedging position. Of course, the higher the volatility, the more valuable the
hedge as long as it truly reduces or eliminates market risk.
Following is a summary of fundamental tests based on dividend indicators,
P/E ratio, revenue and earnings, and debt capitalization.
Dividend Analysis
Three specific and separate dividend indicators are worth studying as part of a
long-term trend. These are dividend yield, dividends per share, and payout ratio.
The dividend yield is the annual percentage represented by the dollar amount
of dividends, divided by price per share. For example, if share price is $75, and
dividend is $3.50 per share, dividend yield is: 3.50 ÷ 70 = 5%. However, in
calculating dividend yield, you should base it on the actual price you paid
originally. As the stock price moves, dividend yield changes. However, your
effective yield is always based on your original price per share no matter how
much price moves. As price declines and dividend per share remains unchanged,
dividend yield rises:
3.50 ÷ $68 = 5.15%
3.50 ÷ $66 = 5.30%
3.50 ÷ $64 = 5.47%
Key Point
A high dividend is not always a reflection of exceptional value. It may be
the result of a large decline in the stock’s price, so the reasons for this
decline should be studied before deciding to purchase share.

This means that a high current dividend yield is not always good news. You
need to check stock price over many years to decide whether the fixed dividend
per share is occurring while the company’s stock is losing value, or whether the
company has increased dividends over the years.
Dividends per share is normally expressed as the total per year. However,
most dividends are paid quarterly. So if a company has declared a dividend of
$3.50 per share, the quarterly dividend payment will be 87.5 cents per share
(one-fourth of the annual dollar amount).
The payout ratio represents the percentage that dividends represent of total
earnings per share. For example, if a company is paying $3.50 per share each
year and total earnings per share was $5.00, the payout ratio is 70% (3.50 ÷ 5.00
= 70%). The ratio is important when studied as part of a trend, because this lets
you see whether the payout is growing or shrinking.
Dividend trends for three companies reveals a comparative summary over 10
years. These companies are AT&T (T), paying 5.66% dividend in October 2015;
ConocoPhillips (COP), paying 5.73% dividend yield; and GlaxoSmithKline
(GSK), with dividend yield of 5.92%. All of these were selected as examples of
companies yielding more than 5.5% dividend, and are summarized in Table 4.1

Table 4.1
The analysis of trends in these dividend indicators over a 10-year period
provides a context to the latest entry in that trend. Comparisons made among
three companies, each paying approximately the same attractive dividend, adds
additional insights into the levels of volatility and hedging potential for each.

Key Point
Most indicators are more accurately understood when studied over a period
of years, rather than only for the latest year.

For example, AT&T (T) increased its dividend every year over the entire
decade. When this occurs, the company is classified as a “dividend achiever,”
because long-term and consistent growth in the dividend is considered highly
desirable as a fundamental test. However, the company clearly has set a policy of
increasing the dividend by four cents per share each year over the past seven
years, and this explains the volatility in the payout ratio. A desirable outcome is
a consistent ratio, meaning the same portion of earnings is paid out each year.
However, when emphasis is on increasing the annual dividend per share but
earnings rise and fall from year to year, the payout ratio is more volatile. In fact,
AT&T’s earnings were also volatile during this period, which explains why
payout ratio is so inconsistent.
ConocoPhillips (COP) increased dividends per share in nine of the 10 years,
and the single year reported no increase or decrease. This places COP in a
category close to “dividend achiever.” However, with earnings reported at a
lower level of volatility, the payout ratio grew over the 10-year period.
GlaxoSmithKline (GSK) was far less consistent in year-to-year dividends per
share. However, the payout ratio rose throughout the period. A degree of
volatility made the GSK results less consistent and more volatile than the other
two companies.
This analysis summarizes how dividend data can be compared and studied
over a decade. The consistent rise in dividends per share is very desirable, and
while all three of these companies paid exceptionally high dividends, the
increases in T and COP were stronger than those in GSK.
P/E Ratio Trends

Key Point
The P/E ratio is a hybrid, combining a technical value (price) with a
fundamental one (earnings). Thus, the time periods for each will be
different.

The price/earnings (P/E) ratio is one of the more popular indicators. However,
it is an oddity in the sense that it compares a technical value (price) to a
fundamental outcome (earnings). With this in mind, the time frame for each side
is also dissimilar. Price represents the latest known price, which changes daily
and is available instantly. Earnings, however, might be many weeks or months
out of date, because the latest reported earnings are for the latest reported fiscal
quarter.
Some attempts to fix this disparity include use of the forward P/E, which
compares current price to estimated current earnings. This is a problematic
method, as the estimate of earnings might be quite inaccurate.
The solution to the odd comparisons and time are found in the long-term
analysis of the annual high and low P/E. This provides you with a more accurate
idea of how a company’s stock has been priced over time. By dividing price by
earnings, the resulting number, called the multiple, represents the number of
years of earnings (based on latest known earnings) reflected in price. For
example, if the P/E is 15, it means the current price per share is equal to 15 years
of earnings for the company.
The generally accepted middle rate for well-priced stocks is a multiple
somewhere between 10 and 25. A multiple under 10 indicates a lack of interest
in the market. And when the multiple moves above, it indicates that the stock is
overpriced. By analyzing 10 years of the annual range from high to low P/E, you
get a good idea of how the stock has been priced over a decade, and how the
price trend has behaved—high or low volatility.
The three companies previously studied for dividend trends can also be
studied for their P/E ranges. Table 4.2 summarizes the range of multiples for
each.

Table 4.2

In the case of AT&T, the multiple was consistently within the moderate range
for most years, with the exception in 2011 with exceptionally high levels of P/E
multiple. The 2014 and 2012 high multiples were also out of the middle range.
However, in addition to checking the high and low levels, the span between the
two is also revealing. AT&T’s high and low span was under 10 points for every
year during the decade, which indicates that the yearly range of changes in both
price and earnings are not volatile. However, the high levels in specific years
does reveal a lot of volatility in earnings levels.
COP reported a range of multiples in the middle range for the entire period,
and the span between high and low never exceeded 6 points. This indicates that
based on volatility in revenue and earnings, COP’s stock price was reasonable
within the decade.
GSK reported spans of 8 points or less every year. The jump in the multiple,
revealing some years in which the stock price was too high, is a reflection of a
natural tendency in the market to see stock prices move too high, only to correct
in following periods. This is good information for any investor with a long-term
portfolio in two respects. First, a big change in a fundamental indicator like the
P/E ratio is likely to self-correct in time. Second, those periods of changes in the
stock price and resulting value through P/E and other indicators, points to the
need for specific types of hedges when the trends begin to change.
Revenue and Earnings
The best-known fundamental indicator is the combination of revenue and
earnings. However, these combined trend values also are easily misinterpreted or
misread.
In an ideal growth pattern, revenue will grow every year, consistently and
with a predictable curve of growth, also reflected as a percentage of year-over-
year dollar value. However, that ideal situation rarely occurs. There is a tendency
in every trend to eventually slow down and level out. This is not a problem as
long as earnings keep track with revenue.

Key Point
An analysis of revenue and earnings should combine dollar value rise or
fall and tracking of net return.

Just as the dollar value of revenue should grow every year, the net return
should remain steady or improve as well. Net return is the percentage that
earnings represent of revenue. It is not realistic to expect to see net return
increase indefinitely because there is a ceiling on the percentage that can be
expected. However, as revenue grows, so should the dollar amount of earnings.
These assumptions are affected by changes in the mix of business and market
share. When one company acquires or merges with another, the trends will
change as well. So in analyzing long-term trends, make sure that the reported
values are adjusted to reflect new realities once a company has merged with or
acquired another, or after a company disposes of part of its operating revenues
by selling off segments.
No one wants to invest in a company that is losing market share. However,
just as declining revenue and earnings are negative trends, some trends are not as
visible. For example, even when revenues and earnings are both increasing, is
the net return keeping pace? When you see increasing revenue with declining net
return (even as the dollar value grows), it signals a decline in internal controls, a
negative trend for the organization.
For example, in Table 4.3, the annual dollar values (in millions of dollars) for
revenue and earnings are summarized over a 10-year period for the three
companies in this study.

Table 4.3

The AT&T (T) trend is strong in terms of revenue, with the dollar amount
growing in most of the years shown. Earnings were more volatile. However, a
great problem is in the net return. In the period between 2005 and 2009, net
return was consistently reported between 10% and 11%. After that, net return
was quite volatile, with three of the most recent years all below 6%. In 2014, net
return was only 4.7%, less than half the average of the net return in the first five
years.
ConocoPhillips (COP) has experienced declining revenues in the most recent
three years, and during this period earnings have also been on a general decline.
Considering the condition of the oil and gas industry, this trend is not surprising.
However, it remains a troubling aspect of the long-term trend that the dollar
value of revenue has declined to levels less than one-third of revenue levels 10
years ago, and earnings have also declined considerably.
GlaxoSmithKline (GSK) has reported no substantial growth in revenue
throughout the decade, although earnings have been inconsistent. In 2014,
earnings were about one-half on the decade’s average. This reveals weakness in
the fundamentals of the company.
Debt Capitalization
The final fundamental indicator is the debt capitalization ratio. This is a
percentage of total capitalization represented by long-term debt (with the
remainder represented by stockholders’ equity).
When you see a trend of long-term debt increasing over several years, it is a
troubling change. The more long-term debt a company is carrying, the more
future earnings will have to be spent on debt service, and the less remaining for
dividends and growth. An examination of the three organizations in Table 4.4
reveals some very different kinds of long-term debt trends.

Table 4.4

Key Point
The debt capitalization ratio studied over many years reveals whether
reliance on long-term debt is rising or falling. A rising level of debt is a
negative signal.
The debt capitalization ratio for AT&T (T) rose during the decade, from
levels around 30% up to more recent levels in excess of 40%. This occurred
during the time of net return declining significantly. So in combination, the net
return and debt capitalization ratio paint a picture of a negative trend.
ConocoPhillips (COP) also saw debt capitalization ratio double from 14.1 in
2005 to 29.2 by 2005. At the same time, levels of both revenue and earnings
were also on the decline. When the debt capitalization ratio is taken into account
with the revenue and earnings trend, the signs were not positive.
GlaxoSmithKline (GSK) reported the most alarming debt trend of these three
companies. Revenue levels did not change much during the decade, and earnings
were stagnant. However, long-term debt grew from less than 40% in 2005 to
more than 70% by 2014.
In summary, all of these companies appear to be strong on a fundamental
level. However, the weaknesses in many fundamentals (especially seeing trends
in combined negative movement) eventually should be expected to affect the
stock price and volatility, and that in turns also affects option pricing and the
need for hedging.

Assumptions Used in Trend Analysis

Key Point
The analysis of fundamentals should be performed before buying an equity
position; these should be monitored periodically to ensure that initial
assumptions still apply.

Once a company is selected as appropriate for your risk tolerance, it may be


added to your portfolio. However, this is not the end of the process, but the
beginning. While trends apply to fundamental strength or weakness, they play at
least as much of a role in tracking stock prices. Your portfolio is rarely
permanent. Yesterday’s strong and seemingly safe stock might be a less
attractive choice tomorrow or next month.
Anyone who doubts this may want to review the history of General Motors,
Eastman Kodak, or any other company that was once considered a bulletproof
blue chip stock, only to end up bankrupt and worth only a fraction of its previous
value.
Assumptions rule trend analysis, and the skillful analysis of the current stock
trend relies on how accurately those assumptions are applied and interpreted.
The first broad assumption worth considering is that some type of trend is
always in effect. Is this even true?
A trend—consistent directional movement in price over time, within a well-
defined range, and continuing until signals forecast an end—may not always
exist for a particular stock. Periods of high volatility often throw trends into
chaos and it is difficult to determine the direction price is likely to move next.
As chaotic as this situation is for stock investors, volatility presents
opportunities for options traders to hedge positions with high probability of
profits. Some specific option strategies are perfectly suited for these times of
price volatility and unclear direction. See chapters 8 through 12 for some of the
many varieties of options strategies well suited to hedging within volatile price
patterns.
In tracking prices of stocks, one assumption that may easily mislead you is
the assumption that your entry price is the “zero price.” In other words, once you
buy shares of stock, you expect the price to rise as if you were starting out at
zero. In fact, whatever price you pay for shares of stock, it is part of an endless
series of increases and decreases in price. These change daily and can move in
either direction. As obvious an observation as this seems, many investors act on
the unspoken assumption that they buy shares at “zero.”
A third popular assumption is that price trends are a reflection of supply and
demand for shares of stock. This is partly true, but there is much more at play in
the overall influence of prices for stock. If supply and demand were simple and
easily identifiable matters, anticipating and forecasting prices would be simple.
A stock’s market value could be appraised just like real estate and a fair price set
based on changes in supply and demand. This does not happen because many
variables come into the pricing of stock. These include forecasting of earnings,
changes in product or service mix, mergers and acquisitions, rumors about the
company, changes in management, and specific management decisions affecting
stock prices. Supply and demand is part of the picture, but the total picture is
much more complicated.
Yet another assumption is that trends are reliable for some predetermined
period of time. In fact, though, trends cannot be predicted. Their duration varies
considerably, some very short and others lasting for years. The only type of
prediction that can be made is a likely end to the trend. This is based on locating
reversal signals and secondary signals confirming a likely change in direction.
Even that is far from completely reliable. There are too many variables that
cause a stock’s price to continue moving or to suddenly reverse and move in the
opposite direction, to be able to call the end of a trend predictably.
Another trend involves perceptions of the market, and tracks how a majority
of traders act based on news. The most important form of news is earnings. A
popular activity among analysts is to estimate the price per share of revenue and
earnings that will be reported. If the outcome is anything other than the
prediction, the earnings surprise is likely to cause a big move in price. Even
missing estimates by one penny per share can cause a 4- or 5-point move in the
stock price. However, whether the news is good (earnings surpass estimates) or
bad (earnings fall short), the resulting jump in price usually self-corrects within
one or two sessions. So a 4-point jump is likely to be followed by a three-point
move in their direction. Stock prices tend to overreact and then to correct very
rapidly. This points to the timing for options-based hedges with exceptionally
good chances to exploit short-term price movement.
A majority of market traders act and react emotionally rather than logically to
surprises. This “crowd mentality” is what causes the short-term price chaos.
However, a contrarian is an investor who understands this overreaction and
tends to make decisions logically rather than emotionally. So the bullish
tendency is to act out of greed and buy shares expecting further price increases;
the bearish tendency is to act out of panic or fear and dump shares before prices
decline further. A contrarian does not make decisions just to do the opposite of
the majority, but based on a rational and unemotional analysis of what is taking
place.
The contrarian tends to follow the advice to “buy low and sell high,” but the
majority does the opposite. Acting out of greed or panic, the more common
tendency is to “sell low and buy high.”

Speculation as an Alternative to Conservative Investing


Contrarians recognize the value of price patterns on charts and are able to act
on what they see. This can be a conservative approach because it is based on
analysis rather than on emotion.
For the purpose of hedging portfolio positions with options, following price
trends of stocks makes more sense than the alternative: calculating the volatility
levels of options. For the options speculator, volatility is the key to timing of
trades. Because volatility changes over time, the theory is that long or short
trades—speculative trades—can be accurately timed. This system may work to a
degree, but it is focused solely on the option.
The alternative of timing conservative hedging trades based on stock charts
and price signals makes more sense—because when tracking the signals in the
stock price, your focus is on the market risk of the stock rather than on the ever-
changing premium value of the option. As a result, option trades are timed based
on exposure to market risk or exploitation of opportunities based on price
movement in the stock.
Speculators share common attributes that do not address the risk reduction
benefits of hedging. These attributes include:
1. Short-term thinking. The speculator often is also impatient. Speculation
is focused on very quick in-and-out trades. A speculator is not likely to
hold onto equity positions unless that is part of a speculative move itself.
Because of this, the speculator—unlike the conservative investor—creates
trades offsetting profits and losses, but does not hold onto equity positions
any longer than they need to in order to play out the speculative purpose.
2. Focus on the immediate pattern, not the long term. For a conservative
investor, the focus is on holding value investments over a period of time
and hedging market risk. So this affects how a conservative investor looks
at trends and price patterns. It is all part of how price behavior affects
value. If the price behavior indicates a change in the long-term trend, a
conservative might decide to sell equity and move cash to a different
position. The speculator is concerned only with the immediate pattern and
how that can be timed to create an immediate profit.
3. Chasing fast profits at the risk of fast losses. The great flaw of the
speculator is focus on fast profits. If you talk to speculators, you will hear
all about how they doubled their money in a single session by getting in at
10 a.m. and getting out by 2 p.m. But you are less likely to hear about all
of the other times when similar trades ended up in big losses.
4. A tendency to keep score of outcomes between trades. A speculator who
loses on today’s trade is likely to double up on tomorrow’s trade to
recapture the loss. This means their risk level is doubled up. A wiser
method is to accept losses as they occur, keep risk levels at the same
profile already established, and move on to the next trade. The “double or
nothing” approach to investing more often than not leads to bigger losses,
not recapture of lost profits.
Conservative investors are quite different in the sense of how price patterns
are interpreted. When option hedging is based on management of equity risk,
patterns in price are more reliable. A longer-term perspective reveals likely price
movement based on historical trends (the past six to 12 months, for example)
and how those price patterns are likely to behave in coming weeks and months.
The speculator is looking at price patterns and anticipating movement over a
matter of days or even hours, so their perspective is very short-term. And short-
term price behavior is invariably chaotic and difficult to understand.
A conservative charting technique is based on recognition of signals and
confirmation through other signals. Speculators certainly use similar techniques,
but such combinations of signals do not always occur within a short span of
time, which is the speculative focus. A conservative analyst understands that
each current price pattern is unique and, even with reliable signals, behavior
depends on circumstances, not only related to the stock being studied but also in
the broader market in the moment.
The speculative assumption often is based on a belief that a correlation will
be found between one set of price patterns and another—in other words, that
price behavior is predictable even in the short term, and price movement reacts
in the same way at all times. This is a blind spot for speculators. Some patterns
are likely to repeat most of the time, but relying on it occurring all of the time is
a high-risk assumption. It makes more sense to look at all of the current signals
and indicators.

Key Point
A common flaw among speculators is to assume a correlation between price
patterns, even when no such correlation can be established.

A final note about speculation: Although it is a method enjoyed by many


traders, the risk of loss makes speculation a difficult system to profit from
consistently. Even speculators who track signals tend to suffer from confirmation
bias. So a speculator who spots a bullish reversal is likely to look for
confirmation of a bullish move. They may easily ignore divergence between
signals or clearly bearish indicators. Contradiction in signals is not unusual, but
it should be taken as a warning signal of confusion in the market. If not
confusion, contrary signals also reveal a fact about price patterns. Some patterns
are simply coincidental and others just fail some of the time. So confirmation
bias is a blind spot to be aware of whether adopting a conservative or speculative
view of a current trade. No one will profit 100% of the time, but prudent chart
analysis and conservative timing of trades—especially applying a contrarian
logic to decisions—will tend to improve trade outcomes more than the
alternative of speculating on short-term price patterns.
To understand investor behavior within a charting method for timing trades, it
is useful to understand the basics of trend theories, including the Dow Theory,
the efficient market hypothesis (EMH), and the random walk hypothesis (RWH).

The Dow Theory


The best-known theory of trends is the Dow Theory, named for the founder of
the Wall Street Journal and one of the two originators of the Dow Jones
Company, Charles Dow. He developed a system aimed at anticipating
fundamental trends within companies. He was a fundamental analyst and did not
anticipate applying his beliefs to stock prices directly. He developed not only the
concept of trend analysis, but also formed one of the first stock averages. This
was an index containing nine railroads, a shipping line, Western Union, and a
small number of other companies traded publicly.
It was not until after Dow’s death in 1902 that the idea of the Dow Theory
was formalized and applied to marketwide trends. The well-known Dow Jones
Industrial Average (DJIA) as well as the transportation, utility, and composite
averages, were devised by Dow’s successor, William P. Hamilton. Today’s DJIA
consists of 30 industrial stocks and, for many market watchers, represents “the
market.” However, it holds only 30 stocks out of thousands, and as of late
October 2015, according to Dow Jones & Company, four companies represented
nearly one-fourth of the total weight of the DJIA:

Goldman Sachs 7.19%


MMM 5.94
Boeing 5.56
IBM 5.29
Total 23.98%

This price weighting is the result of how components are calculated. When a
stock splits, for example, the weighting increases. This means that a few
companies tend to have greater influence over the 30 included in the index. So in
this example, four companies represent nearly one-fourth of “the market,” as
interpreted widely. This points out the importance for equity investors of
focusing on the indicators for individual companies in their portfolio. The
overall market certainly influences price movement in the short term, but for a
more conservative opinion about the value of your portfolio, the fundamental
and technical indicators of those individual stocks you own are more important
in the long term than activity in the overall market, however that is measured.
Key Point
The weighting of the DJIA gives considerable influence to a handful of
companies. Even so, many consider this index representative of “the
market.”

The DJIA and related averages form the basis for how the Dow Theory
works. As currently applied, the Dow Theory has six major points, or tenets:
1. The market consists of three movements. The movements (of duration of
trends) consist of primary, or major trend, lasting up to several years;
medium, or secondary trend, lasting up to three months; and minor, or
swing trend, lasting between a matter of hours and a few days. The normal
explanation of a “trend” is that it can be bullish or bearish. However, a
period of consolidation, in which prices are range-bound and move
sideways, may extend from a matter of weeks to several years. This is a
third type of trend beyond bullish or bearish, but it is clearly a form of
price movement. The consolidation also brings up many options trades and
offers the potential to hedge range-bound prices to create short-term
profits.
2. Trends have three distinct phases. In bull markets, these are
accumulation (the purchase of shares by investors who understand the
markets), public participation (the time when the majority follows the lead
of those investing earlier), and the distribution phase, in which the bull
market comes to an end and investors sell shares. During the public
participation phase, speculation tends to increase in the belief that the
current trend is likely to continue. However, more price movement tends to
occur in phases one and three, which also offers guidance for the timing of
hedge trades. In a bearish market, the first phase is distribution, followed
by public participation, and finally by accumulation, a sign that the bear
market is coming to an end. During a consolidation trend, the three phases
cannot be spotted, making timing of trades more difficult but also
providing good opportunities for hedging with options.
3. News is discounted by the markets and this is reflected in prices. This
portion of the theory assumes an efficient market, in which prices always
reflect current news and even rumor. This might be true, but “efficiency”
should not be confused with “accuracy.” The discounting of news does not
make the current prices fair or accurate, especially given the inclusion of
rumor and gossip as forms of news, and the tendency for markets to react
in an exaggerated fashion to the unexpected, such as earnings surprises.
4. Trends reflected in averages have to be confirmed by the same trend in
other averages. This basic concept of reversal and confirmation is clearly
an effective means for timing hedges based on price behavior in individual
stocks. Under the Dow Theory, the DJIA trend changes direction, but that
is not established as a clear reversal until confirmed by one of the other
averages. This is most commonly viewed in the Dow Jones Transportation
Average; when it reverses in the same manner as the industrials, it serves
as confirmation.
For example, Figure 4.1 summarizes the one-year record of the Dow
Jones Industrial Average.

Figure 4.1: Dow Jones Industrial Average Index, 2014

Compare the pattern in this to the pattern of the Dow Jones


Transportation Average shown in Figure 4.2. The similarity reveals the
accuracy of the theory that the averages confirm one another.

Figure 4.2: Dow Jones Transportation Average Index, 2014


5. Trends are further confirmed by volume. The mood of the overall
market is seen in changes in trend direction and then confirmed by changes
in levels of volume. By the same argument reversal signals in individual
stocks often are confirmed by single-session volume spikes or by one of
several volume indicators. This form of confirmation strengthens the
timing of option trades intended to hedge market risk associated with trend
reversal.
6. Trends continue until signals appear indicating reversal. This might
seem obvious, but it is a key to understanding how trends behave and how
timing of hedge trades can be improved. As long as a clear trend has been
established, it will not just stop without signals. On the basis of an
individual stock, this is valuable information. Timing of hedges should be
based on recognition of a reversal signal and confirmation of a likely
change in direction.

Key Point
The Dow Theory forms the basis for modern technical analysis, notably
about signals within market trends.

The Efficient Market Hypothesis (EMH)


The Dow Theory formed the basis for modern technical analysis. However, it
is focused on the broad averages and the trends established within the 30
industrial stocks. To some degree, these 30 do represent the larger market. Not
only do the 30 stocks represent a large percentage of overall market
capitalization, the trends also tend to lead the market, so trends in the DJIA are
often closely followed by other companies.
The Dow Theory led to some additional theories about the market, including
the efficient market hypothesis (EMH). This theory is based partly on the tenet
of the Dow Theory concerning the efficient discounting of all news. However, an
accurate definition of EMH should clarify that the market is informationally
efficient. In other words, all information known to the public, in theory, is
reflected in current prices. However, this does not mean the market accurately
values stocks, or that it does not contain overreaction to news. In fact, the
tendency for price movement in individual stocks is very inefficient and chaotic
in the short term. However, this is a symptom of a market that is informally
efficient—because not all information is reliable or accurate.
Some observers point to EMH as evidence that stocks are always accurately
priced. They are not. They are efficiently priced based on a combination of
supply and demand coupled with efficient discounting of all news, both accurate
and inaccurate. EMH has always been controversial, partly because it is not
properly understood and partly because a natural conflict has arisen between
those who believe you cannot accurately predict price movement and those who
are devoted to price signals. Once you understand what efficiency means
(specifically how accurate and inaccurate information affects prices), the theory
makes sense. However, EMH is best understood in the context of how markets
react to and discount news, and not as a basis for claiming the markets price
stocks efficiently.
Even if you accept the theory that markets incorporate news efficiently, this
does not mean that participants in the market (whether speculators or
conservative investors) always apply sound judgment. How do investors respond
to information? The overreaction to earnings surprises begs this point. A
predictable tendency of prices to move too far and to then retreat after earnings
surprises reveals that efficiency is not the same as accuracy. In other words, an
efficient reaction to news does not always lead automatically to an efficient or
accurate response by the market.

The Random Walk Hypothesis (RWH)


The Dow Theory also led to the development of another idea closely related
to EMH: the random walk hypothesis (RWH). Under this belief, all price
movement is random and cannot be accurately or consistently forecast by any
methods of study. Under RWH, the initial assumption is that markets are
informationally efficient (as expressed by EMH), so it follows that all price
movement occurs within that informationally efficient environment. In other
words, if the market is efficient, it follows that price activity will be random.
If RWH were accurate, all stocks would move upward 50% of the time or
downward 50% of the time. Overall value would never change. A review of any
stock chart shows that this random tendency is simply not the case. In a truly
random environment, like roulette, nearly half of all spins will land on black and
nearly half on red (outside of this, some spins land on zero or double zero). A
counter-argument can be offered that over the entire market, price movement is a
50/50 proposition. However, history has shown that, over time, stock values
have tended to grow, but to remain in a range of random outcomes.
RWH believers claim that a balance between supply and demand explains the
random nature of the markets. These economic forces tend to balance out against
one another. However, in order for RWH to work in the real world, you would
need agreement between all buyers and sellers about the fairness of current stock
prices. As a result, any movement above or below those levels would be truly
random. This is not how markets can ever operate, however. Too many variables
are in play, including changes in competitive position and earnings, mergers and
acquisitions that change a sector’s membership, and, more than anything else,
the irrational behavior of investors and traders. Yet another reality is that the
forces of demand and supply are rarely equal, and are in constant movement and
change based on trend duration and strength.

An Option Hedging Theory of the Market


In comparison to EMH and RWH, the markets work in very predictable ways.
A new theory of hedging includes the use of options to offset risk and contains a
few observable points:
1. Reversal signals, if confirmed, reliably forecast price behavior. The
general observation about use of reversal and confirmation can be
demonstrated as effective. In spite of EMH and RWH theories about the
impossibility of price predictions, the science is reliable. No one can claim
to generate 100% success, but finding strong reversal and confirmation
signals and timing trades as a result will improve overall trading success.
For an investor with a portfolio of equity securities, this does not mean you
want to move in and out of positions; it does mean you can use option
strategies to hedge risks and generate income at a better than average rate
based on locating strong reversal and confirmation signals. This has to be
qualified by the following key points.
2. The strength of a trend has an effect in the strength of reversal signals.
As you observe how reversal actually occurs, you will discover a statistical
reality. Price movement contains many random variables, but there is a
direct correlation between the strength of a trend and the strength in
reversal signals. “Strength” refers to the speed of price movement as well
as to the rate of price movement. Stronger trends under this definition tend
to end with stronger-than-usual reversal and confirmation signals. It also
follows that weak trends (characterized as slow-moving and with a
marginal degree of price change) tend to lead to weaker and less-reliable
reversal and confirmation signals. It also follows that the levels of strength
or weakness in trends and in reversal and confirmation signals are also
correlated in the strength or weakness of the trend that develops after the
reversal occurs. This observation is crucial to the selection of option
hedges and to the timing of their entry and exit. An understanding of the
stock trend leads to a better understanding of how to hedge with options.
3. The proximity of signals to resistance and support identifies the
likelihood of successful forecasting. Another observable fact about trends
is that the trading range matters. The tendency is for stock prices to adhere
to the defined borders of resistance and support, whether trending upward,
downward, or sideways. Reversal signals occurring at these borders or
moving through them are more likely to succeed than the same signals
occurring at midrange. This proximity effect should not be ignored.
Resistance and support define the trend as well as place limits on how
prices behave. When price moves through resistance at the top of support
at the bottom, it may also set up a new and stronger continuation trend.
This is easily observed by the lack of a reversal signal and the emergence
of strong continuation signals and confirmation indicators.
4. No matter how strong a set of signals, some are going to fail. All signals
may succeed or fail. Even the strongest, best-placed reversals, with equally
strong confirmation with multiple added signals, will not act as expected
100% of the time. This fact points to another essential practice in the use
of options to hedge risk: The level of a trade should be limited to what you
need, hopefully at about the same dollar level for a series of trades. In this
way, a failed reversal is not catastrophic, but merely the expected outcome
in which some portion of trades do not end up profitably. However, even a
strong believer in RWH will admit that creating profits in more than 50%
of trades is “better than random.” Applying the principles of reversal
location and confirmation should produce even better results than a
minimal 51%. It is entirely reasonable to assume that even with some
failures, properly discovered, timed, and executed option trades will hedge
stock position risks, and yield better results than the average of the market.
The next chapter explores the specifics of chart analysis and demonstrates
how to identify strong reversal and confirmation signals. This is based on a
belief that the stock trend is what matters, and that an options trend should be
based on timing to mitigate the market risk in the stock position, not out of a
desire to speculate but in the interest of putting conservative principles into
action.
5 Charting and Trade Timing

Price charts include a wealth of information, which leads to good timing for
both stock trades and option hedges.
Options trading can and should be based on analysis of the stock’s price chart,
on which current trends and reversals can be spotted and acted upon. The
proximity to trades in relation to the current trading range is a critical aspect of
skillful timing. This chapter examines many crucial charting attributes and
explains how they work. This information is at the core of proper timing for
option hedging strategies.
Included in the following pages are basics of chart signals, both Western
(well-known technical price patterns) and Eastern (Japanese candlesticks). The
candlestick chart is today the default charting medium for stock price analysis,
although the specific types of candlestick signals are not always well understood.
Also included is a discussion of options trading based on identified reversal and
confirmation signals involving price as well as volume, moving averages and
momentum oscillators.

Key Point
Charts are the starting point for technical analysis, combining price and
non-price indicators to spot signals and anticipate price behavior.

Most investors are familiar with candlestick charts, as these have gained
popularity in recent years. Before widespread use of the Internet, candlestick
charts were difficult and time-consuming to construct, so most traders relied on
the simpler OHLC (open, high, low, close) chart. On this chart, a vertical line
extends from the high to low range of trading during a session. A small
horizontal line extending to the left identifies the opening price, and a small
horizontal line to the right marks the closing price.
The chart of US Steel in Figure 5.1 shows the appearance of an OHLC chart.

Figure 5.1: US Steel—OHLC Chart, 3 Months

A second form of chart used in the past was a simple line chart. This could be
based on either opening or closing prices. Figure 5.2 is a revised chart showing
closing prices for STOCK.

Figure 5.2: US Steel—Line Chart, 3 Months


Both of these old-style charts are difficult to read in comparison to a
candlestick chart. With this form of chart, you can quickly identify the price
direction (white for upward days and black for downward days); the daily open
and close (marked by the borders of a rectangular box called the “real body”);
and the extent of trading above and below the open and close, better known as
the trading range (identified by the full extent of vertical lines called shadows,
extending above and below the real body).
Figure 5.3 is a chart for US Steel using candlesticks in place of OHLC or line
systems.

Figure 5.3: US Steel—Candlestick Chart, 3 Months

The Meaning of Reversal


Later sections introduce many popular reversal signals that appear on price
charts. This is a sample among dozens of possible reversal signals. By definition,
reversal means a current trend is likely to end and be replaced by a new dynamic
trend. So a current bullish trend will end with a bearish reversal; and a current
bearish trend will signal a change with a bullish reversal signal.
This general rule is easily understood for dynamic trends (bullish or bearish).
For a consolidation trend, the rule is slightly different. The sideways-moving,
range-bound price behavior eventually will break out either above or below. The
breakout identifies a possible new trend and at this point a reversal signal,
accompanied by confirmation, indicates that the new trend will succeed. So for a
continuation pattern, “reversal” means a change from sideways to either bullish
or bearish movement. This observation is controversial, because many technical
analyses describe consolidation as a period in between trends but not as a trend
itself. If this were true, it would be impossible to identify a change (meaning the
beginning of a new bullish or bearish trend) until the new trend was already
underway.

Key Point
The definition of a reversal can be expanded to mean any change in a
trend’s movement, including movement into or out of consolidation.

If you accept the premise that “reversal” is not strictly directional, the use of
reversal signals makes sense. The signal identifies the reversal of a trend, but not
always its direction. For example, a current bullish or bearish trend may end
with a reversal signal, with the resulting trend moving sideways into a
consolidation pattern. So in this case, “reversal” is defined as a change from
dynamic to sideways price movement. These distinctions become important
when identifying the most advantageous option strategy to use for hedging the
price action of the stock held in your portfolio.

The Meaning of Continuation


Not as widely acknowledged as reversal signals is the continuation signal.
This is a specific price pattern, found in either Western technical signals or
Eastern ones (candlesticks). In a continuation signal, the apparent direction of a
trend is confirmed as likely to continue.
The continuation signal is most useful when it appears after a breakout above
resistance or below support. The big question at this moment is whether or not
the price will reverse or continue moving in the same direction. At this moment,
an option hedge makes good sense as long as you know the likelihood of
reversal or continuation. So a continuation signal is a strong forecast that the
breakout is going to succeed. A second location for continuation signals follows
a reversal. For example, a downtrend concludes with a bullish reversal signal,
price begins moving upward, and then a continuation signal appears. This
confirms the initial signal and adds confidence that the new direction is likely to
succeed. The same applies to a bearish reversal after a bull trend. The price
pattern moves downward and is confirmed if a continuation signal appears.

Key Point
A continuation signal works in all types of trends, but often is strongest
following breakout from a continuation trend.

When the trend is not dynamic (bullish or bearish) but is a range-bound


consolidation trend, breakout followed by a continuation signal has even greater
value. A breakout from the consolidation range does not necessarily provide any
reversal indication, so it is not easy to know whether the breakout will reverse or
succeed. However, if a continuation pattern appears, this adds confidence to the
likelihood that the breakout will succeed. In consolidation, many attempted
breakouts are likely to occur, only to fail; so finding a continuation signal with
breakout is perhaps the only sign that should increase confidence. This timing—
breakout from consolidation—is one of the difficult patterns to read, so the
continuation signal has special value as this breakout occurs. The inclusion of a
continuation pattern points to the best timing for an option hedge in what
appears to be a new dynamic trend.
Both reversal and continuation signals may be short term in nature. However,
even though the outcome is uncertain, as a conservative investor you can make
judgment calls about protecting positions and hedging risk as these crucial
patterns emerge. Without the hedge, stock investors are on a roller coaster, but
without knowing how high or low the ride will move. Options hedging timed for
reversal and continuation signals levels out the ride and controls market risk in
the equity position.

Western Technical Signals


The first set of signals is widely understood by chartists accustomed to
spotting signals on charts. As with all chart interpretation, the danger of
confirmation bias affects judgment and may easily lead to incorrect beliefs. So
you need and depend on strong signals and strong confirmation in order to
support a belief about how price is behaving. The purpose in looking for signals
is to anticipate price movement before it happens. With the short list of popular
Western technical signals, a remarkable consistency can be experienced in price
behavior based on testing of resistance or support. As a general rule, when price
tests those price levels without successfully breaking out, the most likely next
step is a price movement in the opposite direction.
The first of these signals is called the head and shoulders. This price pattern
has three parts. The first and third are price spikes and are named the shoulders.
The middle is a higher price spike and is the head. The theory of head and
shoulders is that this three-part attempt at moving price higher fails, so it is
likely to be followed by a price decline.
This can also occur as a bullish pattern, or as an inverse head and shoulders.
In this case, the three parts are at the bottom, and are expected to be followed by
a rise in price. Figure 5.4 shows examples of both head and shoulders and the
inverse pattern.

Figure 5.4: Head and Shoulders/Inverse Head and Shoulders

Another frequently viewed pattern is the double top. This is a double spike in
price at the top of a current price range, and a test of resistance. As this attempt
to break out fails, price behaves as expected and is likely to decline. Likewise, a
double bottom consists of double spikes at the bottom of a downtrend, and is
expected to lead to a bullish result. Figure 5.5 includes a double top and a double
bottom.

Figure 5.5: Double Top/Double Bottom


A gapping pattern is characterized by single or multiple price gaps between
sessions. There are many types of gaps, but the most significant ones for the
purpose of timing option hedges are those that move price through resistance or
support. These two events—the gap itself and the move outside of the
established trading range—set up a likely signal of coming reversal, when price
is expected to retreat back into range.
Figure 5.6 shows an example of the gapping pattern on the chart of
Weyerhaeuser.

Figure 5.6: Gaping Pattern


The gaps followed a short and gradual time of rising support. Gaps appeared
in four consecutive sessions. The final gap is the strongest, from the close of the
black session at about $27.60 to the opening of the larger black session at about
$28.10. A problem with gapping price behavior is that a correction in the
opposite is likely to occur, or the rapid price movement may lead to
consolidation, as it did on this chart. The one-month consolidation ranged only
one point and, after price spiked higher, it appeared to return to this range-bound
trend.
The next form of signal is the wedge. A rising wedge is a bearish pattern.
Both resistance and support trend upward but the trading range narrows. As the
range reaches its narrowest point, price is expected to decline below the rising
level of support. A falling wedge does the opposite, with both sides narrowing
downward until price breaks out and rises. Figure 5.7 shows both wedge types.

Figure 5.7: Rising Wedge/Falling Wedge


The triangle is similar to the wedge and often is only slightly different.
However, whereas the wedge is a reversal pattern, the triangle is a continuation
pattern, meaning it indicates a continuation of the current trend. An ascending
triangle consists of a level resistance price with a rising support price, and a
descending triangle has level support with a declining resistance. As the triangle
narrows, price is expected to break out in the direction indicated, continuing the
trend. An example of both of these patterns is found in Figure 5.8.

Figure 5.8: Descending Triangle; Ascending Triangle


An important distinction between wedges and triangles is their opposite
meaning. Wedges lead to reversal away from the previous trend and triangles are
expected to indicate continuation. Given the fact that the two types of signals are
very similar (but contain different meaning) it is easy to misread what they
forecast. So you need to find strong confirming signals before acting on either a
wedge or a triangle as it appears on a chart.

Candlestick Basics
Beyond the familiar Western reversal and continuation signals are a broad
range of dozens of candlesticks, also classified as Eastern technical signals. In
this and following sections, several popular and frequently seen candlestick
patterns are introduced. These include 26 reversal signals and six continuation
signals. They represent a small portion of the candlestick universe.
As a starting point, you will need to become familiar with all of the
information a candlestick reveals. Figure 5.9 summarizes the attributes of the
candlestick, including the names for each of the parts of the session.

Figure 5.9: Candlestick Attributes


The candlestick is a powerful charting tool because of its visual aspects. Not
only single candlestick sessions, but an entire chart range becomes immediately
clear as to range and strength as well as general direction of movement. A white
candlestick means the price moved up during the session, from the bottom of the
rectangular real body up to the close at the top. The black candlestick is a
downward-moving day, opening at the top and closing at the bottom. The upper
and lower shadows reveal the full trading range during the session, from high
price to low.
Both reversal and continuation signals should be confirmed independently, by
additional candlestick indicators, or by Western signals (head and shoulders,
double top or bottom, gaps, wedges, triangles, for example). An important
principle about candlesticks is their placement. Many of the writings about
candlesticks report that a reversal signal in the wrong location is a continuation
signal. This is not the case. If a reversal signal appears in the wrong location, it
is not a signal at all. For example, a bullish reversal signal should appear at or
near the bottom of a bearish trend; and a bearish reversal should appear at or
near the top of a bullish trend. These “rules” are basic and can be observed on
actual charts. It also follows that improper location of signals provides no value
at all and might even mislead you into an erroneous interpretation.
Single-Session Reversal Signals

Key Point
The scaling of a chart makes relative size of sessions unique to each chart.
Comparisons between differently scaled charts are not accurate.

The first set of signals involves only single sessions. Although relying on a
single session to find reversal may imply less strength, this is not always the
case. However, single-session reversal signals should be confirmed
independently. Remembering that some formations are going to occur as
coincidence, it is not wise to act on one indicator alone, until confirmation has
been found.
The first signal is called a long candlestick. The definition of “long” is
specific to the chart itself. There is no specific number of points defining a long
candlestick versus a “normal” one. A session is long when it exceeds the typical
size of preceding sessions. This point can be observed in the fact that the scaling
of a chart affects the length of candlesticks. Some charts employ one-quarter-
point increments; others are as high as 10 or even 20 points. So “long” is a visual
outcome based on a comparison between sessions found on the chart and based
on how price is scaled.
A long white candlestick is bullish and a long black candlestick is bearish.
Both are shown in Figure 5.10.

Figure 5.10: Long Candle (White)/Long Candle (Black)


Another single-session reversal is found in the form of a doji. The doji is a
session with the same or very close opening and closing price, meaning that
instead of a rectangle, you find only a horizontal line. The doji includes a
horizontal line and a shadow, but the significance relies on the placement of
both. A dragonfly doji is bullish and has the horizontal line at the top with a long
lower shadow; the gravestone doji is bearish and has the horizontal line at the
bottom with a long upper shadow. Two additional types are significant
depending on where they appear in a trend; at the bottom of a trend, they tend to
be bullish and at the top they are bearish. The long-legged doji has both an upper
and lower shadow; and the spinning top is also known as a near-doji because
instead of a horizontal line, it has a small real body. The upper and lower shadow
should both be longer than the spinning top’s real body. All four of these are
shown in Figure 5.11.

Key Point
Many signals, including the longlegged doji and spinning top, may be
bullish or bearish depending on where they appear and how they are
confirmed.
Figure 5.11: Dragonfly Doji/Gravestone Doji/Long-Legged Doji/Spinning
Top

The hammer and hanging man are examples of single-session patterns with a
small real body of either color and a long lower shadow. Although both are
identical in appearance, they take on different meaning depending on where they
appear in a trend. A hammer appears at the bottom of a trend and is a bullish
signal. The hanging man is bearish and appears at the top of a trend. Both are
shown in Figure 5.12.

Key Point
The hammer and hanging man are identical in appearance, but the
proximity to bullish or bearish trends creates an opposite meaning.
Figure 5.12: Hammer/Hanging Man

Double-Session Reversal Signals


Among two-session reversal signals, one of the strongest is the engulfing
pattern. A bullish engulfing consists of a black session followed by a white
session that extends both higher and lower than the previous session’s real body.
A bearish engulfing is the opposite: a white session followed by a larger black
session. These are shown in Figure 5.13.

Figure 5.13: Engulfing Pattern (Bull)/Engulfing Pattern (Bear)


The engulfing appears often and is one of the most reliable of candlestick
signals. So when you see one of these and can also find confirmation, it is a
strong indicator that a current trend is likely to reverse and begin moving in the
opposite direction.
A harami is the opposite of the engulfing. It consists of a first session
followed by a shorter session of the opposite color. The bullish harami begins
with a long black session followed by a smaller white session, with opening and
closing prices both within the range of the previous day’s real body. The bearish
harami is the opposite: a long white session followed by a shorter black session.
Closely related is the harami cross. This is like a harami, but the second session
is a doji. All four of these are summarized in Figure 5.14.

Figure 5.14: Harami (Bull)/Harami (Bear)/Harami Cross (Bull)/Harami


Cross (Bear)
The harami by itself is not an especially strong signal. However, when
confirmed by other signals, it helps strengthen reversal signals. In addition,
greater strength is seen when the first session is longer than average.
The next two-day signals are the meeting lines and piercing lines. The bullish
meeting lines declines in the first gap, and gaps to open lower on the second day
with a closing price the same as the day before. The bearish meeting lines sees
movement in the opposite direction and reversal of colors.
A very similar formation is seen in the piercing lines signal. The exception is
that the second session closes within the range of the first. The bullish piercing
lines begins with a black day, followed by a white day opening lower but closing
higher than the close of the previous day. The bearish piercing lines starts out
with a white session, with the second day opening with a higher gap but closing
within the range of the previous day. All of these patterns are shown in Figure
5.15.

Figure 5.15: Meeting Lines (Bull)/Meeting Lines (Bear)/Piercing Lines


(Bull)/Piercing Lines (Bear)
Another two-session candlestick signal is the doji star. In this pattern, the
second session follows a gap and forms as a doji. The bullish doji star begins
with a black session, then a downside gap and a doji. The bearish doji star
begins with a white session and is followed by an upside gap and a doji. This
should be found at or near the top of an uptrend. An example of each is shown in
Figure 5.16.

Figure 5.16: Doji Star (Bull)/Doji Star (Bear)


Triple-Session Reversal Signals
The final group of reversal signals consists of three consecutive sessions. In
many respects, the three sessions tend to be stronger than the one- or two-session
signals, simply because they involve more trading sessions to form the
indicators.
The first among these is called three white soldiers and three black crows.
The three white soldiers is a bullish signal made. Each session opens higher than
the opening price of the previous session and also closes higher. The three black
crows is the opposite: It is a bearish session. Each one opens and closes lower
than the preceding one. Both are shown in Figure 5.17.

Key Point
Three-session reversals may be stronger than others; however, many of
these appear often and can reliably forecast reversal in the current trend.

Figure 5.17: Three White Soldiers/Three Black Crows


Another set of three-session reversals appears as a morning star (a bullish
signal) or an evening star (a bearish signal). Both of these appear with gaps
between each of the sessions. Very similar is the abandoned baby. The
difference is that the middle session is a doji in both bullish and bearish versions.
These four reversal signals are shown in Figure 5.18.

Figure 5.18: Morning Star/Evening Star/Abandoned Baby


(Bull)/Abandoned Baby (Bear)
These patterns often are set up as part of a set of trading sessions above or
below the current trading range, set off by gaps both before and after. This
period is known as an island cluster, a series of sessions outside the established
range and marked by gaps.
Continuation Signals
Numerous candlestick continuation signals are found on charts. Following are
the most common categories of continuation, the two-session thrusting lines, and
three-session tasuki gap and gap filled.
The bullish thrusting lines reveals a likely continuation of the current trend.
It consists of a white session that “thrusts” into the price range of the second
session’s black real body. The bearish thrusting lines has the opposite
configuration and the colors of the two sessions are switched. This is found
during a downtrend and forecasts that the trend will continue in the same
direction. An example of each is shown in Figure 5.19.

Figure 5.19: Thrusting Lines (Bull)/Thrusting Lines (Bear)

Continuation may also appear in the form of three-session signals. The tasuki
gap consists of two sessions separated by a gap and then an opposite-color
session moving in the opposite direction. It forecasts continuation of the current
trend. A bullish tasuki gap starts with two white sessions separated by an
upward gap and then a black session moving into the range of the gap but not
closing it. The bearish tasuki gap begins with two black sessions separated by a
downward gap and then a white session moving back up into the gap’s range.
The gap filled is similar to the tasuki gap with the exception that the third
session closes within the range of the second day, absorbing the entire gap that
precedes it. The bullish gap filled starts with two white sessions separated by an
upside gap, and a third, black session moving down to absorb the gap. The
bearish gap filled consists of two black sessions separated by a downside gap
and a third day as a white session moving to absorb the gap. All four of these
continuation patterns are shown in Figure 5.20.

Figure 5.20: Tasuki Gap (Upside)/Tasuki Gap (Downside)/Gap Filled


(Upside)/Gap Filled (Downside)
The reversal and continuation patterns described here are most commonly
found, but they are not complete. Dozens of other signals will also appear, and
the science of candlestick analysis as part of a charting technical theory involves
more than 100 different signals.

Key Point
Continuation is as valuable as reversal, especially when located close to
resistance or support. It indicates a likely successful breakout.

Candlestick Limitations
The candlestick signal is a powerful tool that serves as a guideline for picking
and timing option hedges. However, these are not foolproof. Some specific
limitations of candlesticks should be kept in mind. These include:
1. Interpretation is not a simple matter. The strength or weakness of many
signals might not be well understood or, even more troubling, might vary
from one case to another. A lot depends on the speed and strength of
preceding trends and the amount of time those trends have been in effect.
Interpretation becomes quite complex, however, when a similar pattern is
strongly confirmed but the precise attributes of the preceding trend are
dissimilar. This means that to aid in interpretation, it is useful to seek out
additional confirmation beyond a single indicator, and to also check non-
price indicators as additional forms of signals that may help clarify what
price patterns reveal.
2. Perfect patterns are not always found. The appearance of the perfect
pattern is an ideal discovery, but it does not always occur. Some
conforming patterns meet the criteria but only minimally. The stronger the
shape and relative size of patterns to the ideal, the stronger the pattern
tends to be. A minimally conforming pattern might also tend to be weaker
than many similar signals. With this in mind, the importance of strong
confirmation should be emphasized to ensure that the reading of a pattern
is as reliable as possible.
3. All patterns fail at times, even strong ones. The strong pattern aided by
equally strong confirmation will succeed most of the time and will point to
good timing for option hedging trades. However, there are no absolutely
guaranteed patterns, and at times they will fail. This is true for all
indicators. The purpose of relying on strong patterns and confirmation is to
increase the ratio of profits to losses, striving to beat the averages. The
purpose should never be to try for a “perfect” system for timing trades.
Indicators and confirmation should increase confidence in the timing of the
hedge, but this is not a promise that the timing will yield profits in every
case.

Non-Price Signals and Confirmation


Beyond the numerous price patterns and signals, a comprehensive option-
based hedging strategy should also rely on volume, moving averages, and
momentum oscillators. These indicators are good sources for confirmation, but
they are more likely to not work as leading indicators. For the lead reversal or
continuation signal, it makes sense to rely on price signals and then confirm
them with other price signals or non-price indicators.
The science of charting is intended as a replacement for the better-known
options trading method, based on a study of volatility. This is a fast-moving and
at times unreliable method. For example, in the final month before option
expiration, volatility collapse makes volatility analysis ineffective. Because
many option hedging strategies focus on short-term expiration, the use of
implied volatility of options as a means for timing trades is a speculative and
risky approach to option trading.
Because your purpose is to manage risks in your equity portfolio, the stock
chart and its signals is the most logical source for the timing of hedge trades; and
options are an excellent vehicle to accomplish this purpose, as the timing is
based on patterns in the stock price.
The next chapter introduces the most popular form of hedge, the covered call.
In this conservative strategy, you create current income from options without
adding market risk to your stock position. So with the covered call, you earn
profits from option premium, stock-based capital gains, and dividends.
6 The Basic Covered Call

The most popular option hedge is the covered call. This in its most basic form
has two parts: 100 shares of stock held in your portfolio, offset by a short call. If
that call is exercised, you are required to give up 100 shares at the fixed strike,
even if the current market value is higher. At the time you open the covered call,
you receive the call premium and it is yours to keep, whether the call is
exercised, expires worthless, or is later closed.
This is the starting point for developing a hedging strategy. The premium you
receive for selling the call reduces your net basis in stock and creates a downside
cushion. When you own 100 shares of stock, your market risk level is any price
below your cost. With a covered call, the risk level is reduced by the premium.
For example, if you buy 100 shares at $50 and sell a call for 3 ($300), your
breakeven price is $47 per share.
The strategy is simple at first glance. However, several aspects of writing
covered calls have to be considered in order to convert this strategy into a true
hedge. These considerations include selection of the company whose stock will
be bought and held; your own objectives and attitudes about the strategy;
specific attributes of the covered call; moneyness of the strike you pick; and
many variations on the strategy.
When you first study the covered call, some obvious features are seen at once.
For one thing, it is attractive because it involves three sources of profits: option
premium, capital gains, and dividends. In addition, the covered call is clearly
less risky than just owning stock, assuming the strike selected is higher than your
original basis, because the option premium lowers your breakeven price for
stock. Third, picking stock with a higher dividend appears a sensible idea, even
though this is not always the case.
Key Point
Although covered calls appear very basic at first glance, it contains several
aspects that have to be considered to use it as a true hedge.

As a straightforward strategic use of stock in your portfolio, the covered call


may be used as a “cash cow” in the sense that repetitive positions can be opened
and allowed to expire, closed at a profit, or rolled forward. For the investor
interested in creating additional income, this is a conservative strategy that you
can employ, assuming you also accept the limitations of the strategy. The
covered call limits your capital gain if and when it is exercised. So a
compromise may be to pick very short-term calls out of the money and monitor
them carefully, to allow those calls to expire, or, if they move in the money, to
quickly take action to avoid exercise.
Because the stock price can move either up or down, the outcome of the
covered call can take several forms. However, maximum profit is limited, and
maximum can range considerably farther. The maximum profit is:
Strike – stock basis + premium

Breakeven, however, can in theory be greater than maximum gain. It is equal


to:
Basis in stock – premium

This means that loss could be a big number, assuming the stock price might
fall indefinitely. The overall catastrophic maximum loss is equal to:
Basis in stock – zero + premium

This means that if the stock price falls to zero, you lose the entire investment.
(By the way, just owning stock presents an identical risk of catastrophic loss.)
The only difference here is that the call’s premium discounts the loss.
The breakeven price is itself a hedge. When you buy stock, your breakeven
price is the net amount paid. The call premium discounts this and lowers the
breakeven. However, this points out the risk in covered call writing. If the
underlying price falls below your breakeven (basis – premium), you have a
paper loss. You need to accept this loss, wait out a hoped-for reversal, or devise a
new hedging strategy to recover the loss.
In the straightforward use of the covered call—meaning closing the call once
it loses value and becomes profitable—exercise is clearly a possibility. So as part
of a specific hedging strategy (versus a more basic money-making strategy), you
need to decide whether exercise is an acceptable outcome. Properly structured,
exercise should be profitable in every case, without exception. So one point of
view is that creating double-digit annualized returns justifies the occasional
calling away of shares. With this goal in mind, the ideal covered call situation is
when the underlying price remains in a narrow consolidation range, meaning it
trades sideways with little movement up or down. You just wait out the decline
in time value and then close the call at a profit and replace it with a later-
expiring one. The problem, though, is that exercise remains a possibility if and
when the underlying rises above the call’s strike. If you are happy with the mix
of holdings in your portfolio, having shares called away creates three problems:
1. Tax consequences including capital gains, often short term. This may be
worth avoiding by also avoiding the sale of shares, at least until the one-
year deadline has been passed.
2. Need to replace called-away shares. Do you repurchase shares of the
same company? Or do you look for a bargain-priced replacement? If you
want to hold on to shares, exercise is worth avoiding, either through rolling
or closing short calls; or finding different hedging strategies.
3. Loss of dividend income. A conservative investing strategy is likely to
include selection of companies whose dividend is higher than average. So
a covered call exercised shortly before ex-dividend date also means you
lose the quarterly dividend. The period right before ex-dividend date is the
most likely time for early exercise, so the timing of expiration for your
covered call has to be carefully timed to avoid in-the-money positions in
the month that dividends are earned.

Key Point
You probably would prefer to not have shares called away. However, the
event causes several problems involving taxes, replacement, and loss of
dividends.

As with all hedges involving long positions in stock, the starting point is
selection of the right stock.

Picking the Right Stock


If you already own shares in your portfolio, especially appreciated shares, this
clearly affects your selection of a covered call strategy. This also raises
additional questions. For example, do you want to create a covered call designed
to augment income or have shares called away at a profit? If so, the selection of
a strike and expiration will be likely to change. This possibility is examined later
in this chapter.
If you want to buy shares and, as part of a hedging strategy, sell a series of
covered calls to increase income, the selection of a company is of the utmost
importance. The greater the historical volatility of the stock, the richer the option
premium. So some investors pick stock based on the level of option premium,
without realizing that the market risk of the stock is also much higher. As a
conservative investor, the analysis of fundamental trends and also of
fundamental volatility leads to a more balanced selection of the company.
Covered call premium will be lower, but the entire position will also be safer.
The purpose of a hedge is to reduce and control risks, not to increase them. So
picking a safe stock for covered call writing involves the same basic steps as
those for picking stock to hold in your portfolio.
Chapter 4 included detailed analysis of a short list of fundamental trends and
how evaluation can be conducted. This concept applies to every option hedge,
without exception. All of the positions in your portfolio can be studied based on
fundamental trends and their strength or weakness; and if a once-strong
fundamental investment has changed and moved into negative territory, a smart
move is to sell those shares and replace them with a safer, stronger company.
For covered call writing, a less-volatile stock will offer lower premiums.
However, you will also notice that less-volatile stock prices reflect safer
fundamental trends. Companies whose revenue and earnings are on the rise, with
a healthy and growing dividend, and a stable or falling debt capitalization ratio,
tend to also see long-term stability in the stock price. The two aspects—
fundamental strength and stock risk—are directly connected. However, with
marketwide trends often influencing stock value, especially short-term, it is easy
to overlook this reality.
For example, those stocks classified as “dividend achievers” (companies
increasing dividends every year for 10 years or more) tend to also see consistent
growth in the value of stock over the long term. This reflects the fact that the
company has managed its cash well enough to afford to make dividend
payments, and that profits are more likely to occur than losses. These are basic
ideas.

Key Point
The analysis of dividends involves not only yield, but annual dividend
trends and the payout ratio.

Beyond the dividend per share measured by the dividend achiever standard,
the dividend payout ratio is an equally important test. This is the percentage of
earnings paid out each year in the form of dividends. It is possible to see
dividend payments increase along with a drop in the payout ratio. This means
that the company is growing its earnings every year but paying lower dividends
in terms of the payout ratio. You cannot expect the ratio to increase indefinitely,
but it should remain consistent from one year to the next.
Another fundamental test is more difficult to spot, especially if you focus
solely on dividend payments. A consistently higher dividend paid each year
appears to be a positive trend, but how is this possible if earnings are flat or
falling? What if the company reports a net loss but the dividend is still
increased? This is where comparisons between dividend payments and long-term
debt are crucial to understand how to select a company as part of your long-term
portfolio. If dividends are increasing every year but earnings are flat or falling
(or net losses are reported), where is the money coming from? If you notice that
the debt capitalization ratio has been increasing during the same period, this is a
troubling trend. It means the company is financing higher dividends with higher
long-term debt. So future earnings will have to be used to pay debt service,
meaning less will be available for growth and dividend payments.
The point here is that picking a company for covered call writing involves the
same careful analysis as picking stocks for your portfolio even if you do not
intend to hedge the market risk. Isolating the selection to one set of criteria
without looking at the full picture can easily lead to false assumptions. So
picking a company because dividends are increased every year is one test; but
also check revenue and earnings trends, and the debt capitalization ratio, to make
sure that the dividend history reflects growing earnings and long-term cash
management, and not replacement of equity with debt capitalization.

The Downside Protection Aspect


One hedging attribute of the covered call is quite apparent. Compared to just
buying 100 shares of stock, the combination of 100 shares and a short call
involves a discount. The option premium reduces the net basis in stock.
However, it also places a cap on maximum profit. This is equal to the strike
minus basis in stock plus option premium.
On the downside, this hedge is important but minor in the larger scheme of
risk management. For example, if you buy shares and sell a call for 3 ($300),
your basis is hedged by three points, and so is your breakeven point. However, if
the stock price continues moving downward below your breakeven price, the
result is a paper loss. You either have to wait out the price, hoping for a reversal,
or take further hedging action.
In other words, downside protection matters, but it only reduces risk; it does
not eliminate it altogether.
Waiting out price reversal often works, especially if the company has been
selected wisely and you believe in long-term value. Riding out the price swings
will validate the strategy, but there are cases when the price will simply not
recover or, if it doesn’t, the recovery will take many months. With this in mind,
expansion of the covered call into a balanced strategy makes sense as a more
protective form of hedge. Strategies may include the insurance put for matching
of stock price decline with option intrinsic value growth, a covered straddle to
increase the downside protection, synthetic strategies, or collars.

Key Point
Hedges in the form of bearish defensive moves accompany the covered call
as contingencies in the event of price slide after the call is opened.

These defensive moves can be entered if the stock price begins to slide, and at
the same time the short call can probably be closed at a profit. However, as long
as the market risk works against your equity position, the conversion from
covered call to downside protective hedge has to be acted upon quickly. In
coming chapters, the issue of hedging when prices are on the decline is a core
element of the discussion. Chapter 7 provides an alternative to the covered call
with the same market risk but greater flexibility. A comparison of this topic
(uncovered puts) with covered calls provides a deeper view of the covered call
idea than most investors have. This is true due to the bias against uncovered
options, including uncovered puts.
An example of a fairly simple hedging technique when stock prices begin to
slide is to close the covered call and replace it with an uncovered put. This
replaces one option with another but does not change the market risk. It is a
method for hedging the downside risk without having to sell shares. However, it
is always true that maintaining shares in your portfolio rests on a broad
assumption that you continue to believe in the long-term value of the company
and its stock. If this belief has changed, shares should be sold and replaced with
stronger equity candidates.

Objectives and Investor Attitude


The decision to pick a particular company and write calls, and the policies
you set for yourself about when or if to close, roll, or accept exercise, all reflect
your objectives in writing covered calls, and your attitude concerning the hedge
itself.
Your first objective probably is a combination of two things: achieving
downside protection (to the extent of the call’s premium) while generating
additional income. Writing a series of short-term calls over the period of a year
accomplishes both of these objectives; and selection of a strike out of the money
will yield lower premium but drastically reduces the odds of exercise. A popular
example of this is writing a series of monthly calls; the strategy makes sense
because time value evaporates rapidly during the final month, so chances of
exercise are slim when out of the money calls are used in this manner.
As an alternative, some option series are available on a weekly basis, so you
can write calls expiring in only a few days. In this case, time decay will be
maximized. Whether you use monthly or weekly options, with rapid time decay,
you can create consistent income as long as the underlying stock does not rise
dramatically. Two suggestions to minimize this threat: Avoid the week of ex-
dividend date and the week when earnings will be announced. Both of these
usually come up quarterly, and are likely to occur in different months and weeks.
The period right before ex-dividend is the most likely period for early
exercise, especially the day before ex-dividend. That is the last day to be
stockholder of record and earn the current dividend, so the owner of a call that
has moved in the money is likely to exercise and take shares to earn that
dividend. It does not always happen, but it can happen, so it is worth avoiding.
Once the ex-dividend date arrives, the early exercise threat is over. “Ex-
dividend” means “without dividend,” meaning you do not earn the current
dividend until the following quarter.
Earnings announcement dates are also troubling if you have already opened a
covered call. An earnings surprise can send stock prices higher (positive
surprise) or lower (negative surprise). Although the price usually self-corrects
within a few sessions, the change in the call’s status presents an exercise threat.

Key Point
One requirement for writing covered calls is the willingness to have the call
exercised. It is one of several possible outcomes.

The underlying stock can rise quickly and at any time, but as long as you pick
out-of-the-money calls with strikes above your basis in stock, the risk is
minimal; and if the call is exercised, you profit from option premium and capital
gains. This leads to yet another important objective in covered call writing: A
wise covered call strategy is one in which you accept exercise as one possible
outcome. Because your true objective is to create a hedge against market risk,
you probably prefer to maintain equity in your portfolio and to use covered calls
as hedges. However, one risk to the covered call hedge is exercise, so this has to
be acceptable, even though it is not desirable.
Setting objectives as part of your covered call strategy is good portfolio
management. Without clearly defined objectives, it is too easy to focus on the
three forms of income (capital gains, option premium, and dividends) and to
ignore the reality of downside risk; it is also possible to overlook the clear
advantages of hedging because upside profit is capped at the call’s strike. This
awareness of lost opportunity risk translates to a limitation in thinking. A call
can be closed or rolled forward to avoid exercise, or the profit—even though
limited—is a positive outcome even though not the desired end result.
The lost opportunity risk should not be ignored. In some market conditions,
owning stock without writing calls will perform better than a covered call
position. As a general rule, the covered call reduces the market risk of just
owning stock. However, if the market is strongly bullish and stock prices are
rising, avoiding covered call writing—at least until market volatility ends or
slows down—makes more sense. For example, after a large price decline, stock
prices are likely to rise back into the previous range. Writing covered calls at the
bottom of this trend is poor timing.
For example, on August 24, 2015, the overall market declined in a one-day
drop. Many stocks experienced rapid price decline as well, and the pattern can
be seen on numerous charts. The Boeing chart shown in Figure 6.1 is an
example.

Figure 6.1: Covered Call Timing


The stock had been trading in a consolidation trend with a range of 10 points,
between $137 and $147 per share. The dramatic drop down to as low as $115 on
October 24 was severe; however, as it often occurs in sudden price moves, the
trading range returned to the previous consolidation range within two months.
For the purpose of covered call timing, this chart is instructive. The hedge
works best during a period of consolidation. The range-bound price lends itself
to short-term profits from covered calls that will not move in the money as long
as consolidation holds. So writing a series of very short-term covered calls
hedges the situation well. Consolidation is frustrating for investors; nothing is
moving, so the stock is not profitable. However, generating consistent profits
through short-term options solves this problem. In fact, consolidation adds safety
to the covered call, because the price is range-bound.

Key Point
Consolidation trends are frustrating, but with option hedging this price
pattern presents exceptional opportunities as well.

At the top of this range, at-the-money calls are very profitable. As of early
November (closing values of Friday, November 6), the stock price was still
within the consolidation range established the month before. The stock closed at
$147.94. Table 6.1 shows the covered calls were available at that time.

Table 6.1
Call description Expiration Bid Price
1 week ATM Nov 13 ( 7 days) 1.30
2 week ATM Nov 20 (14 days) 1.94
3 week ATM Nov 27 (21 days) 2.28

These three calls make the point. (The first and third were weekly, and the
middle was the November monthly contract.) As long as Boeing remained in
consolidation, it was possible to generate premium income repeatedly.
Once the price moved below the support price of consolidation, which
happened on August 20, it was time to stop writing calls. Any open calls could
be left to expire worthless, but given the decline in price, no new covered calls
should be opened, especially at lower strikes. The likelihood of a reversal back
into range is quite high once support is violated.
The August 24 decline was unusual, but, as expected, price did return to
range. This occurred in a set of three very strong bullish moves. It took two
months to recover completely from the decline, but this pattern of short-term
volatility points to a period when the covered call hedge did not make sense.
During this time, different hedges would have been more profitable. At the very
bottom on August 24, buying calls (or selling uncovered puts) would have
yielded profits. As a contrarian strategy, making bullish moves at a market
bottom demands nerves of steel, because at such times most investors are
concerned (and some panic) that prices might continue falling. The contrarian
understands that such fast and violent price declines are most like to rise. A
bullish hedge contains minimum risk at this point, because the maximum loss of
a long call is the premium, which will be quite low if a strike is selected slightly
out of the money.
Once price reaches the range of consolidation that held until mid-August, a
short-term covered call program can be resumed. The consolidation resistance
level of $147 to $148 is likely to hold, making the short-term covered call a
worthwhile hedge.

Rules for Writing Covered Calls


A few important rules should always be observed when writing covered calls.
In order for the strategy to work as intended, these are essential:
1. Pick companies as part of a portfolio strategy, not an option strategy.
The starting point for a covered call hedge should be based on your
portfolio priorities and risk profile. So if you prefer low-volatility
companies, avoid high-volatility alternatives just to gain more profits on
covered calls.
2. Always keep original basis in mind. Your basis in stock is a “line in the
sand,” and writing covered calls with strikes at a lower level creates capital
losses in the event of exercise.
3. Be aware of ex-dividend date and earnings date. To avoid early
exercise, avoid having in-the-money calls exposed in the week leading up
to ex-dividend. This is the most likely time in which early exercise can
occur. To avoid short-term volatility, also avoid keeping covered calls open
in earnings week. Any earnings surprises can cause short-term spikes
above or below the strike of the call.
4. Adjust the strategy for marketwide peaks and valleys. Once markets
turn volatile, the covered call hedge is not going to be as advantageous, or
as easy to control, as it is during lower-volatility periods or during
consolidation trends.
5. Accept the lost opportunity risk. As a basic requirement for writing
covered calls, be ready to accept the loss of what could have been without
the profit capping of the call. If stock prices rise above the strike and stock
is called away, this additional profit is lost. A successful covered call
strategy accepts this in exchange for the certainty of the three-source
income (capital gains, option premium, and dividends). Although the
purpose—based on the portfolio objectives—is to not have shares called
away, it is one of several possible outcomes.
6. Take profits when they occur, and write new contracts. Once the
covered call approaches expiration date, it will rapidly lose its time value.
During this time, especially during the last week, the premium value is
likely to decline enough to create a profitable difference between the
opening sold price and current value. At this time, you can wait until the
position expires and then write a new call; or you can take profits by
closing the short call and replacing it with a later-expiring one. There is
never a bad time to take profits; this enables you to generate more short-
term income by continually taking profits and replacing one short call with
another.
7. Choose expiration dates with time decay in mind. The longer-term
expiration tempts the covered call writer with a higher premium; however,
on an annualized basis, more will be earned by focusing on the shorter-
term contracts. This is true for one primary reason: Time decay accelerates
as expiration approaches. With this in mind, picking the shortest term
possible not only generates higher net return, it also makes exercise less
likely due to shorter periods of exposure.

Key Point
To succeed with a covered call hedging strategy, the basic rules have to be
followed consistently.

Another rule worth observing, apart from the strategy itself, is to use the
proper terminology when describing covered calls and other option strategies.
These “style guide” rules include:
1. Use dollar signs for stock prices. The proper expression for stock price is
on a per-share basis and with the use of dollar signs. For example, a stock
is current at $55 per share (not 55 or 55.00); or the price might rise to
$55.25 (not 55.25). Decimal places are always used to two places for
expressing value of stock, and always expressed in price per share.
2. Express option premium in value per share. For options, dollar signs are
never used. So the strike is 55 or 55.25. The decimal places are used only
when premium is at a level other than round dollars. Premium is also
expressed without dollar signs and on a per-share basis. So a premium of
2.30 translates to $230 for one option on 100 shares. If the premium value
is at a rounded number, it is expressed without decimal values. So a
premium of 2 is $200.
3. Use proper buying and selling terms, especially for short options. The
best-known description of opening a long position (in either stock or
option) is “buy to open,” and for a sale it is “sell to close.” When dealing
with a short option, the terminology is reversed. An opening transaction is
“sell to open” and a closing one is “buy to close.” You will see the closing
price described in other places as “buying back” the option. This is
inaccurate and confusing. When you close a short option, you are not
“buying it back” because you never owned it in the first place. Although
this might seem like a minor point, it is important to use the right
terminology to avoid confusion.

Moneyness of the Call


The selection of a particular strike based on its moneyness determines to a
large degree the effectiveness of the hedge. As long as the strike is higher than
your basis in stock, you can decide to write in-the-money, out-of-the-money, or
at-the-money covered calls. Each will have a different hedging profile and likely
outcome.
The chart of Exxon Mobil shown in Figure 6.2 illustrates the status of a
72.50.

Figure 6.2: Moneyness of the Call

The out-of-the-money call contains greater potential profits but lower


downside protection; and vice versa for in the money, in which case profit
potential is reduced but downside protection is greater due to higher call
premium. The final outcome also relies on time remaining until expiration, so
the moneyness is only one-half of the strike selection decision. Perhaps the
optimum strike—assuming your purpose is to hedge downside risk minimally
and to create an income stream with minimal exposure to exercise—is either an
at-the-money call expiring in a few days, or an out-of-the-money call expiring in
less than one month. Both of these contain an element of safety in terms of likely
exercise, and both can be managed to avoid movement in the money where
exercise is a distinct possibility.
Compare premium values for Macy’s (M) as of the closing prices of
November 6, 2015, consisting of two expirations and five strikes, as shown in
Table 6.2.

Table 6.2 – Macy’s call prices, two expirations


Nov 13 (weekly, 7 days) Bid
48 2.40
48.50 2.26
49 2.06
49.50 1.84
50 1.60
Nov 20 (weekly, 14 days) Bid
48 2.94
48.50 2.66
49 2.42
49.50 2.16
50 1.95
Source: Options listings, Charles Schwab & Co.

Key Point
Moneyness and time affect premium levels, but even with this analysis the
picture is not complete.

The differences between the moneyness levels and time to expiration are
clear. All of the one-week options are cheaper, meaning opening one of these
will accomplish faster expiration but less money. However, the difference is
slight. The closest to the money contract of 49 yields 2.06 in the seven-day
option, or 2.42 for the 14-day, a difference of only 0.36, or $36. Is it worth an
extra week of exposure for an added $36 in premium income? Considering how
any stock price can move in a short period of time, the added days of exposure
do not yield enough added income to justify the exposure. The at-the-money
contracts make more sense when opened for the shortest possible term. Earning
2.06 (approximately $197 after deducting estimated $9 for trading costs) in only
one week is not only advantageous income but a 2.06-point hedge as well.
The in-the-money calls yield more premium, but also are exposed to greater
exercise risk. For example, the one-week 48 contracts were big at 2.40 versus
2.94 for the two-week contracts. This raises a strategic question: How do you
analyze the interaction between time decay and potential profits between two
different options? With one point in the money for both of these, the shorter-term
option will expire quite soon but will lose approximately 1.4 points if the stock
price does not move. This means that the time decay will be rapid, so even
though the call is in the money, the potential profit due to time decay is
considerable. In comparison, the two-week option bid at 2.94 will lose nearly
two full points in the next two weeks. The chances of exercise are identical to
the one-week calls, but with more time to act and react, greater profit is also
possible.
With short-term in-the-money covered calls, it is possible to earn a profit even
if the stock does not move, and even if the call remains in the money. This is
because time decay will be rapid, so the position can be bought to close at a
profit even though it remains in the money.
An out-of-the-money strategy is the safest. Even though it yields less
premium, chances of exercise are very remote, notably for the one-week
contracts. For example, you can sell a 50 covered call expiring in one week for
1.60 (net about $151 after $9 trading fees are taken). This call is over a full point
out of the money, and expiration is in one week. With all attributes considered
(time to expiration and moneyness), this is the lowest-yielding dollar amount of
the range of options analyzed; however, it is the one most likely to expire
worthless (or to lose enough value to be profitable with a buy to close order
within a few days).
In selecting one hedge over another, moneyness is a consideration that will
determine the likely profitability of a short-term covered call. Combined with
time to expiration and the certainty of rapid time decay, a covered call hedge can
be structured to yield consistent profits, even if executed using one-week
expiration terms.

Covered Calls With an Expiration Series


Moneyness and time to expiration have to be taken into account when writing
a single covered call against 100 shares. However, this idea can be expanded into
a more elaborate form of hedge when you own more than 100 shares, utilizing a
series of expiration dates and strikes.
Returning to the example of Macy’s, the two next expiration cycles (including
one a weekly option), consider expanding into several. If you owned 300 shares
of Macy’s valued at $48.90 as of the close of November 6, 2015, some
interesting hedge potential is opened up when expanding the basic covered call
idea. Table 6.3 expands the expirations to be considered.
Table 6.3 – Macy’s call prices, three expirations
Nov 20 (14 days) Bid
50 1.95
52.50 1.05
55 0.50
Dec 18 (38 days) Bid
50 2.61
52.50 1.69
55 1.04
Jan 158 (66 days) Bid
50 3.30
52.50 2.42
55 1.67
Source: Options listings, Charles Schwab & Co.

In this expanded list, two dates have been avoided intentionally. Earnings
were scheduled to be announced on November 11, so on the “action date” of
November 6, the possibility of an earnings surprise made this immediate strike
less attractive. The week on December 11 also contained weekly options, but
that same week concluded with an ex-dividend date. This was also avoided as
potentially a time for in-the-money early exercise.

Key Point
An option series allows you to vary the strike while maintaining attractive
premium levels. This may be preferable to writing strictly short-term
options with a single strike.

A series of options is based on current value of the underlying and its basis,
and on the premium values. This is an excellent hedging device as long as you
intend to hold shares for the long term and want to exploit time decay to create
current income. Assuming your basis in this stock was the November 6 closing
price of $48.90, and that you bought 300 shares, the following option contracts
represent one way to set up a series:

November 50 @ 1.95 (minus $9 trading $186


costs)
December 52.50 @ 1.69 (minus $9) $160
January 55 @ 1.67 (minus $9) $158
Total premium $504

This total is lower than the income you could have received for selling three
November 50 calls. These were out of the money by 1.10 points and expired in
14 days, so that would be a reasonable covered call and hedge. The total net
premium for that trade would be $558. However, by expanding into the
subsequent months and changing the strike, you avoid the possibility of short-
term exercise in the immediate term. At the same time, for slightly less income,
you set up the series so that if the stock were to rise higher than the strikes, the
capital gain on the December and January contracts would be higher (by an
additional $250 for December and $500 for January). The lost opportunity risk
of exercise is offset to a degree with this “upside protection” feature of the
option series.
Even if the underlying price were to rise during this period, time decay will
make it very likely that some or all of these positions could be closed for less
premium than the initial sale, making this a profitable hedge based simply on
time decay. This can work as a rolling series as well. For example, once
November options expire or are closed, they are replaced by a new option
expiring in February. However, that will also be a month in which earnings are
again reported, so this is a factor to keep in mind when picking expiration dates.
However, at the point the November contracts were set to expire, February is in
the distant future; so selecting out-of-the-money contracts with attractive levels
of time value offsets this problem.
The option series takes the basic covered call for 100 shares, and expands it
for higher numbers of shares owned. It takes advantage of time decay and
enables you to either roll forward or close at a profit when the underlying price
rises. As a hedge, this reduces current income on an annualized basis but
provides smart management over strikes and creates an income stream that
works well. During times of consolidation, this form of hedge is particularly
attractive as it continues to generate income even though you cannot know when
the consolidation will end on which direction price will move. If price did break
out of consolidation and move downward, currently open calls lose value and
eventually expire; at the same time, if you continue to believe the shares are
worth keeping in your portfolio, you also will assume the price decline is
cyclical. If price breaks out of consolidation to the upside, open calls are likely
to move in the money. If possible, open contracts should be closed to take
profits. If not possible, these can be rolled forward to higher strikes expiring
later.

Calculating Outcomes: The Complexity of Returns


Whether writing single covered calls or a series, one of the big challenges for
every covered call investor is calculating net returns on a consistent basis. The
first question that arises is: What price should you use to calculate returns?

Key point
Return calculations are complex because you first have to decide what
stock price to use. There are three choices.

For example, you have two covered call situations. Both yielded about 3% in
one month, when returns were calculated based on your original cost. However,
one company’s stock was bought at $30 and currently value is closer to $55; the
other was bought at $40 and currently is valued at $42.
Using your original basis in stock would seem an obvious choice, as this is
the normal price for figuring out a net return. However, if your stock has
increased substantially since you bought it, the original price will be
significantly lower than current value; so original basis can distort returns,
especially between two or more situations. In the example, one stock had
appreciated $25 points since purchase and the other had grown by only two
points. Using original basis is not realistic.
Another choice is to use the current value at the time the covered call expires,
is closed, or is exercised. The problem with this is that the outcome is distorted
once again if the current price is substantially higher or lower than the strike.
Using current value is unreliable as a standard, remembering that the calculation
is intended to make reliable and accurate comparisons between two or more
different covered call trades.
The third choice is to use the covered call’s strike price. This is the most
accurate of the three price choices, because this is the price the call will be
exercised if that is the outcome. It is also the price used to determine whether the
call expires worthless, is rolled forward, or exercised.
The calculation of returns is done at two points: first when you are
considering opening a covered call among several choices, and again after the
trade has been closed. This brings up a second issue: Should you include
potential capital gains or dividends?
Capital gains should not be included in the calculation of covered call returns
for the same reason that original basis is not an accurate price for calculating
returns. Though capital gains matter and should be calculated, this should be
kept separate from the calculation of returns on the option hedge.
Dividends are another matter. These should be added into the calculation of
returns, because the dividend yield affects the overall outcome of a covered call
trade. Overall returns has to be annualized to make comparisons accurate, but
timing of quarterly dividends will change the outcome as well. For example, the
following companies offer a similar dividend:
Philip Morris (PM)4.72% (1.18% per quarter)
Consolidated Edison (ED)4.20% (1.05% per quarter)

If dividends were the only consideration in an otherwise equal value for each
of these companies, PM is the choice with a higher dividend. However, an in-
depth analysis of the outcome for covered calls reveals a more complex
outcome.
PM’s ex-dividend was timed for September 28, 2015, and due to repeat at the
end of December, March, and June quarters. ED’s ex-dividend was set for
November 16, with quarterly repeats in February, May, and August. A study of
options trades reveals that the December option closest to the current price but
out of the money yielded a very similar outcome:
Philip Morris (PM), closed Nov. 6 @ $86.37
DEC 87.50 call bid 1.25, net of trading costs = $116
Initial yield: $1.16 ÷ $86.37 = 1.34%

Consolidated Edison (ED), closed Nov. 6 @ $61.95


DEC 62.50 call bid 0.90, net of trading costs = $81
Initial yield: $0.81 ÷ $61.95 = 1.31%

This summary reveals close outcomes: 1.34% versus 1.31%. Based on this,
the PM option appears to be a better choice, especially as it yields more cash.
However, when dividends are added into this equation, the picture changes.
PM’s ex-dividend date occurred on September 28 and was due to repeat at the
end of December, March, and June. ED’s ex-dividend date was set for November
16 with repeats in February, May, and August.
PM did not earn a dividend by expiration of the December contract. The
November 20 option expiration date occurred before the December dividend. So
calculation of return in the window between November 6 and November 20 does
not include a quarterly dividend. Thus, the true annualized return remains at the
outcome based on 1.34% over 38 days:
Annualized: 1.34% ÷ 38 days x 365 days = 12.87%

In comparison, ED does earn a dividend in November, so the return here has


to be annualized and the quarterly dividend added. The annual yield was
4.230%, representing 1.05% per quarter:
Annualized: 1.31% ÷ 38 days x 365 days = 12.58%
Add dividend: 12.58% + 1.05% = 13.63%

Although the difference here is small, the true overall yield of options and
dividends is greater for ED, even with the lower dividend yield, than for PM.
This is due to the timing of the quarterly dividend.

Key Point
The level of dividend yield matters, but equally important is the timing of
quarterly dividends to be earned between opening a position and expiration
of the option.

If you were to hold either of these stocks in your portfolio, both yield a very
attractive annual dividend. However, the point to remember is that once the
timing of quarterly dividends is taken into account, the picture is changed.

Writing Calls on Stock Already Owned and Appreciated


If you buy shares of stock at the same time you open a covered call, the
relationship between underlying price and strike is straightforward. Ideally, the
strike has to be higher than your basis in the stock, meaning the price per share
you paid at the time of purchase. In the event of exercise, you end up with
capital gains or, at worst, a breakeven in the stock. However, covered calls take
on a different meaning and a different hedge profile when you have owned stock
for a period of time and it has appreciated in value.
In this situation, your concern goes beyond generating current income and has
to include the issue of protecting your net gain, even though you do not want to
sell shares. This is an endless problem for even the most conservative investor.
Should you sell and take profits, concerned that the stock price will decline? Or
should you hold onto high-quality shares and ignore the cyclical price changes?
A third choice is to hedge that paper gain with options. There are many
methods for this, including both calls and puts. However, even the covered call
can be used to manage the appreciated stock in your portfolio.
In most covered calls, you are motivated to write at-the-money or out-of-the-
money options. For appreciated stock, this can make sense; if exercised, these
positions yield the option premium and capital gains. But other factors have to
be considered as well:
1. Timing of a covered call. For appreciated stock, are you willing to risk
exercise in less than one year? The long-term capital gains rate is much
more attractive than the fully-taxed short-term gain, so if you do write
covered calls in this situation, pick expirations that occur after the one-year
period. This still contains a degree of risk, especially if ex-dividend occurs
before the year has passed. Early exercise could force a short-term gain.
2. Unintended consequences of exercise. With appreciated stock, you might
consider writing in-the-money calls. This is an attractive scenario. You still
get a capital gain, but much higher call premium represents a nice bump in
your overall return. However, if you write a deep in the money call, you
face a different risk: An unqualified covered call might end up with
capital gains taxed at short-term rates even if the holding period exceeds a
full year.

Key Point
The unqualified covered call can lead to higher tax liability because the
long-term period is tolled while the call is left open.

For example, you bought stock eight months ago at $32 per share. Today the
stock is worth $68, more than twice your initial basis. You would be willing to
sell at a profit, and you do not want to lose these gains. So you consider selling a
covered call with a strike of 45, which is deep in the money. The call expires in
five months and, because exercise at the end of the term would mean you held
the stock more than a year, it would be a long-term gain. However, under the
unqualified covered call rule, the count to full-term status is stopped at the time
you open the deep in-the-money call. Exercise at any time would constitute a
short-term gain in the stock.
The precise definition of “unqualified” varies by price level, time to
expiration, and strike prices. Table 6.4 summarizes then definition of the
unqualified covered call.

Table 6.4: Unqualified Covered Calls


Prior day’s stock Time until expiration Strike price limits
closing price
$25 or below Over 30 days One strike under close
of the prior trading day
(however, no call can be
qualified if the strike is
lower than 85% of stock
price)
$25.01 - $60 Over 30 days One strike under close
of the prior trading day
$60.01 - $150 31-90 days One strike under close
of the prior trading day
$60.01 - $150 Over 90 days Two strikes under close
of the prior trading day
(but no more than 10
points ITM)
above $150 31-90 days One strike under close
of the prior trading day
above $150 Over 90 days Two strikes under close
of the prior trading day

The in-the-money strategy has its time and place. If you are willing to accept
exercise, the in-the-money covered call results in exercise. If times with
expiration occur after the next quarterly ex-dividend date, overall income will be
higher as well. So as a hedge, the in-the-money covered call is a powerful
strategy to generate premium income, capital gains, and dividends. This assumes
that you want to sell.
An alternative outcome creates a different kind of hedge. For example, if the
current underlying price declines, you will see a point-for-point comparison to
intrinsic value in the short call. For each point the stock price declines, a
corresponding decline is seen in the option. Because this is a short position, a
reduced premium value represents profit; the option can be bought to close for a
lower premium than it was sold to open. This is the same level of protection you
gain from buying an insurance put (Chapter 2). In that hedge, downside
protection caps any loss in the underlying with the long put. With the short call,
the same outcome occurs with the difference that you receive the premium
instead of paying it. This advantage is offset by exercise risk.
So when you open an in-the-money put, you either accept the risk of exercise
or exploit intrinsic value in the case of a price decline. This actually is combined
with time value decay, so that the potential profit in the short call contains clear
advantages over the long hedge using the put for downside protection. Your
primary consideration in this strategy is likely to be whether or not you are
willing to risk having shares called away. Remember the basic assumption in
hedging with options: This is designed to protect your equity positions against
market risk, not to give up shares you prefer to keep.

Rolling to Avoid Exercise


Covered call writers avoid exercise by either closing positions as they
approach or move into the money, or by rolling positions forward. Rolling
normally creates a net credit if the roll goes to the same strike, as longer-term
options contain more time value. But there are risks involved. These include:
1. Longer time of exposure. The extended expiration also means your
position remains exposed for a longer time, tying up capital, possibly
without substantially increasing profit potential. So in evaluating a forward
roll of a short call, compare the added credit received to the cost of
keeping the position open longer. Rolling to the same strike is not always a
wise idea.
2. The possibility of creating an unqualified covered call. If the stock price
has increased enough so that the current strike is deep in the money, a roll
to the same strike could set up an unintentional unqualified covered call.
This is a concern only if you have held stock for less than one year; once
the 12-month period has passed, the threat has passed as well.
3. Risk when rolling up. Some big moves in the stock price justify rolling to
a higher strike. This may involve a net debit rather than a credit. The
rationale is that the higher strike represents a higher capital gain if the
position is eventually exercised. However, in this type of roll, keep track of
your true net basis. If you end up rolling to a higher strike but losing on the
option, the profit potential is reduced.
Key Point
Rolling forward helps avoid exercise, but it should be done only with
awareness of time exposure, tax consequences, and added risks.

The forward roll is a smart strategy to avoid exercise. However, if it is


possible to close the position at a profit or even at breakeven, it could make
sense to take that close and replace it with a later-expiring, higher-strike covered
call.
When close to expiration, the roll takes on a different meaning. With time
value gone, should you wait out expiration or close the position and replace it? If
the call’s value is close to zero in the final week, and it is out of the money, there
is little chance of a last-minute move in the money; but it can happen. When at
the money or even slightly in the money, closing a low-premium short call and
replacing it with a new one with higher time value makes sense. When the cost
of buying to close is minimal, the risk of a last-minute price move and exercise
is not worth the small amount it costs to roll forward.

The Ratio Write


One variation of the covered call will be likely to involve more rolling than
the straightforward basic strategy. The ratio write involves writing a higher
number of calls than you cover with stock. So this can be considered as a
partially covered position or a combination of covered and uncovered calls.
The strategy involves more risk than the covered call due to the possibility of
exercise. Refer again to Table 6.3 on page 109. To open a ratio write using the
December options, ownership of 300 Macy’s (M) shares bought at $48.90 could
be converted to a ratio write by selling four 55 calls. The December calls were
bid at 1.04, and trading fees for selling four options would be about $11 (the first
contract costs about $8.75 and additional contracts are priced at about 75 cents
each). The net for these four options would be calculated as: 1.04 x 4 = $416,
less $11 trading costs = $405.
These options with 55 strikes were 6.1 points out of the money, so they have a
comfortable cushion. However, if the underlying prices edged upward toward
$55 per share, some or all of these short calls should be closed or rolled to avoid
exercise.
Key Point
The ratio write increases premium income but also increases exercise risk.

The ratio write provides additional income, thus a greater hedge. However,
the market risk is also increased. So for an additional $104 received for writing
the additional ratio position, the risk is minimal but should not be ignored.
To reduce the ratio write market risk, an alternative is the variable ratio write.
This is the same as the ratio write, but expanded to include two strike prices. For
example, returning to the Macy’s example with an underlying price of $48.90,
owning 300 shares and writing four calls sets up a variable ratio write in the
following configuration:
Sell two 52.50 calls @ 1.69 = $338, less $10 trading costs = $328
Sell two 55 calls @ 0.55 = 1.10 less $10 trading costs = $100

The total net of $428 involves two strikes, both out of the money. This is a
safer hedge because if and when the stock price moves upward, any of these
options can be closed or rolled forward. Because they all are out of the money,
this is a reasonable strategy. Relying on time decay gives you a hedging
advantage when selling options. The covered call is conservative, and the
uncovered call is high-risk. However, the variable ratio write is not as risky as
writing uncovered calls without that ratio protection. The variable ratio write has
risks, but not as high as the ratio write without the two strikes.
The covered call can be written in all of these varieties. The selection of one
level of hedge versus another depends on the volatility of the underlying stock,
exercise and lost opportunity risks, and objectives you hold for your portfolio.
An alternative to the covered call is found in the uncovered put. Although this
is a much different type of option—short put rather than short call and uncovered
versus covered—the market risk of both strategies is the same. The next chapter
examines the uncovered put as a form of hedge that offers another form of
benefit similar to the covered call.
7 The Uncovered Put: Alternative to
Covered Calls With Less Risk

The uncovered put often is perceived as a high-risk strategy. However, it


contains identical market risks to those of the more popular covered call. There
are numerous offsets in the comparison; the point, however, is that the uncovered
put is just as conservative as the covered call.
This chapter describes hedging strategies using uncovered puts. A true hedge
is opened to reduce risks on stock positions owned in the portfolio. So a
conservative investor will not be as interested in an option-only strategy.
However, some circumstances justify writing a put in place of a covered call or
even along with a covered call at the same time.

A Comparison to Covered Calls


The uncovered put and the covered call share the same market value, but the
features of each are not the same. Table 7.1 provides an analysis of the
differences.

Table 7.1
Covered Calls Uncovered Puts
100 shares of stock are held in the No stock ownership is required.
portfolio.
No collateral is needed. Collateral must be held.
When the underlying price rises, the When the underlying price rises, the
call may be exercised unless closed uncovered put will expire worthless.
or rolled.
When the underlying price falls, the When the underlying price falls, the
covered call will expire worthless. uncovered put may be exercised
unless closed or rolled.
In a price decline, the stock loses In a price decline, the put can be
value and has to be held until closed or rolled forward to avoid
recovery or closed at a loss. exercise, using any strike.
In a price decline, a new covered Because no stock is associated with
call cannot be opened below net the uncovered put as a separate
basis without risking a capital loss. trade, the strike does not affect
profitability of stock.
Owners of stock earn dividends as Writers of uncovered puts do not
long as shares are held. earn dividends.

The attributes of each strategy make comparisons difficult, because some of


these (dividends, collateral, specific outcomes of stock price changes) make the
two strategies appear quite different. So equating them in terms of market risks
has to take into account the many variables in attributes.

Key Point
The uncovered put can be rolled with more flexibility than the covered call,
because capital gains in the stock are not at issue.

In any short option position, the desire is for the underlying price to remain
out of the money. The uncovered put is more flexible than the covered call when
the price moves in the money, because it can be rolled without capital gains
consequences in an underlying position. The direction of movement is different.
With the covered call, a rise in the underlying places the short call in the money.
With the uncovered put, a decline in the underlying creates an in-the-money
situation.
The dissimilarity in direction is explained by the nature of the covered call. It
is also termed a synthetic short put because profits accrue in the call as long as
the underlying remains at or below the strike.
The technical description of a covered call as synthetic short put explains why
market risk is identical. However, the attributes (notably dividends and collateral
requirements) make the two positions quite different in many ways. The
flexibility of the uncovered put makes this far more flexible than the covered
call, even though you have to give up dividend income as part of choosing the
put over the call.
A more immediate question for you as an investor seeking hedges against
equity position risks is how the uncovered put fits within your program. As a
basic standard, a “hedge” should be related directly to the stock positions. So
how does this occur? The simplicity and popularity of the covered call is based
on ownership of 100 shares per option, so this naturally acts as a hedge by
discounting the basis in stock while generating premium income; and in the
event of exercise, stock is called away at a profit (as long as the strike is higher
than your basis).
The uncovered put does work as a hedge for positions you own, even though
it is not tied directly to shares. Just as the covered call hedges the equity position
with minimal risks, the uncovered put provides an alternative strategy to
accomplish the same end result. You do not want to write covered calls after the
stock price has turned volatile, and if the price declines, you end up closing the
short call or allowing it to expire. However, if the price decline is below your net
basis, you end up with a paper loss. This means you have to wait for the price to
rebound.
This “wait and see” timing makes covered calls less desirable as a hedge,
especially when compared to the uncovered put. Because the put is not tied to
the stock price, a decline in the underlying can be managed by closing the put or
rolling it forward. In rolling, it does not matter what strike is selected as long as
the roll produces a net credit or a breakeven. This is not possible with the
covered call. If you roll to a strike below your basis in stock, exercise would
produce a net capital loss; and that outcome is unacceptable in a hedging
strategy.
The uncovered put works as a hedge in three specific situations: first, as a
replacement strategy after a covered call has expired worthless due to a drop in
price; second, as part of a recovery strategy after a large decline in the
underlying price; and third, when price is in a consolidation trend.

The Replacement Strategy


The uncovered put, as an alternative to the covered call, has a specific
application when stock prices have declined enough for a covered call to expire
worthless.
For example, you bought 100 shares of Exxon Mobil on October 28, paying
$80 per share. On November 6, the price has risen to more than $84 per share
and you sold a November 83.50 call for a bid of 1.53 (net proceeds after trading
costs were $144). On Monday, November 9, the price of shares has declined to
$83.74 and the November 84 call also fell to an ask price of 0.85. You entered a
buy to close order and paid a net of $94. This generated a profit of $50 ($144 –
$94).
This chart action is summarized in Figure 7.1.

Figure 7.1: Timing for the Uncovered Put

Based on the decline in stock price, you might hesitate at this point to open
another covered call. Given the fact that the next move in this price pattern is
uncertain, this would be a good time to open an uncovered put as an alternative
with greater flexibility. Like the covered call, the uncovered put brings cash into
your account, but closing and rolling are more easily accomplished when you are
not concerned about the underlying price. Support appears at about $80 per
share, which was your original basis. So selling a put anywhere close to the
current value of $83.74 would be one possible way to exploit price behavior.
With the price declining over the last three sessions, chances for a short-term
price upside reaction are positive. So an 83 put is a good candidate. The bid on
the November 83 put was 1.50, so you could sell a contract and net $141.
This short put yields income at close to the premium for the original short
call. However, because expiration was only 11 days away, the chances for rapid
time decay were excellent. This position could be left to expire worthless as long
as the underlying price remained above $83 per share. If price declined, the short
put could be closed or rolled forward.
In this example, the switch from covered call to uncovered put makes sense
because of the underlying price behavior. After buying to close the original short
call, writing another appeared poorly timed, based solely on the risk of a price
surge upward and the consequences to the new covered call. With an uncovered
put, the outcome is more controllable for a position with the same market risk
and greater flexibility.

Recovery Strategy

Key Point
As a recovery strategy after the stock price has fallen, the uncovered put
works where the covered call would not.

The replacement strategy is a sensible way to convert from one hedge to


another. Normally, an uncovered put would not belong in a hedging strategy;
however, in this case, the identical market risk of the two short options made the
uncovered put a smart alternative to use while waiting out the underlying price
behavior. A different use of the uncovered put occurs when the underlying price
has declined below net basis.
The term net basis refers to the original cost of stock, discounted for the net
premium received when selling a covered call. This basic discount is a primary
form of hedging familiar to covered call writers. However, it is also easy to
overlook the very real market risk that shows up when the price decline creates a
net loss.
For example, at the opening on November 6, you bought 100 shares of
Macy’s (M) at $50.40 and sold a November 50.50 covered call at the bid of 1.74
(net $165). This reduced your basis to $48.75 ($50.40 – 1.65). By Monday,
November 9, the underlying price had declined to $46.30. Your net paper loss at
this point was $2.45 per share ($48.75 – $46.30). You bought to close the
November 50.50 call at an ask of 1.05, netting $96. Your profit on the call was
$69 ($165 – $96). So adjusting your basis after closing the call, the net was
changed to $49.71 ($50.50 – $0.69). However, the current value of shares was at
$46.30. The revised net paper loss was $341 ($49.71 – $46.30).
Figure 7.2 summarizes this situation.

Figure 7.2: Timing Uncovered Puts in a Recovery Strategy


At this point, the uncovered put offers a potential solution to the $341 net
paper loss on Macy’s. For example, at this moment, the December 45 put was
bid at 3.80 (providing a net of $371 after $9 for trading costs). This turns the net
loss of $341 into a $30 profit ($371 – $341).
A potential problem with this is exercise risk, of course; and because
expiration for the December contract was 39 days away, the time of exposure
was greater as well. However, you would expect time value to fall rapidly as
long as the stock price remained at or above the strike of 45. At the moment of
this trade, the stock was 1.30 points out of the money. Any change in that
condition could be managed by closing the position and again adjusting the
basis, or by rolling it forward. The forward roll is not ideal in this situation,
because the purpose is not to keep the short put exposure open, but to offset the
loss in net basis.

Key Point
Exercise risk should not be overlooked in the short put position. The put
has to be managed and tracked to avoid exercise.

This provides one example of how the uncovered put is used to manage a
paper loss and create a recovery strategy. If the stock price continues to decline,
the problem is made worse. Not only would the short put move into an exercise
risk status, but the value of shares would also continue to fall. So with this in
mind, the uncovered put as a recovery strategy should be employed only when
you have faith that the fundamental value of shares remains strong and the
likelihood of a price improvement is strong. If this is not the case, additional
hedging to limit downside risk would be justified in the form of longer-term long
puts or other hedges, such as collars or synthetic stock positions.
The possibility that a loss position could be made worse points out one of the
dangers in a recovery strategy. The desire is to absorb the paper loss and return
to breakeven or better. However, as far too many options traders have
discovered, the outcome does not always yield these results. With this in mind, a
recovery strategy should be kept at a modest level. There are instances in which
just taking a small loss is preferable to adding greater risks and ending up with
greater losses. Every investor has to accept loss as a reality; and the idea of
hedging is never to eliminate loss altogether, but to mitigate the effects of losses
on the equity side. Just as stock ownership involves taking a loss in some cases,
the options hedge can also fail to provide the desired result, meaning losses have
to be accepted as a reality in both equity and hedging strategies.

Strategy During a Consolidation Trend


The consolidation trend—in which price is range-bound, often for an
extended period of time—is perhaps the most frustrating of all trends. Price is
not moving outside of the range, so it is easy to think of this period of time as
one between trends, in which neither buyers nor sellers are able to move price in
either direction.
In fact, consolidation as a third type of trend (in addition to bullish and
bearish) presents exceptional hedging opportunities. The stock you hold in your
portfolio is not gaining (or losing) value when prices are fixed in a consolidation,
so you have to either wait out the trend or identify actions you can take to hedge
the situation. The problem, of course, is that you cannot know whether price will
break out to the upside or the downside. Specific signals do exist, and this type
of trend is excellent for writing uncovered puts.
Because you own stock during consolidation, the uncovered put acts as a
hedge in two ways. First, its market risk (identical to the covered call) is
manageable no matter which direction prices move. Upon breakout, a rise in
price leads to the put becoming worthless. A downside breakout leads to closing
the put and either taking profits or rolling it forward. However, as a hedge, the
risk in consolidation is inactivity. The uncovered put generates income in the
same way as the covered call, but without the risk of exercise (if the underlying
prices breaks out above) or paper loss (after a downside move).
Using the uncovered put during consolidation is a safer hedge than the same
move during a dynamic trend. Uptrends, especially fast-moving ones, can
reverse quickly and pose exercise risk. Downtrends clearly are not the time to
write uncovered puts. However, the relatively narrow trading range associated
with consolidation presents not only trading opportunities, but clear breakout
forecasts.
For example, on the chart of Kellogg (K) shown in Figure 7.3, many signals
are present that point to trading opportunities with uncovered puts.

Figure 7.3: Uncovered Puts During Consolidation Trends

The consolidation trend extended for three months, from April to July. During
this time, the price remained in a breadth of only three points. Three attempted
breakouts below all failed, and formed an inverse head and shoulders pattern.
This by itself predicted a breakout in the opposite direction. This was confirmed
in the second half of July when prices narrowed in the top portion of the
consolidation range. This also formed a Bollinger squeeze, a pattern seen within
Bollinger Bands, a system of three volatility bands (a 20-day moving average,
higher band, and lower band, each two standard deviations from the middle).

Key Point
Bollinger Bands uses moving averages to identify price volatility, making
this indicator a type of probability monitor.

The Bollinger squeeze is a narrowing of prices close to one of the sides of the
consolidation range. It is a period of low volatility, which predicts a coming
period of high volatility and potential breakout. On the chart, although the
Bollinger Bands are not shown, the squeeze does occur in the second half of
July, immediately before the breakout. This squeeze confirms the inverse head
and shoulders—and, as expected, price broke out above the consolidation range.
In consolidation, many attempted breakouts fail. So how can you know
whether a breakout will succeed? This pattern, combining inverse head and
shoulders with a Bollinger squeeze, is a convincing start. Price then formed into
an island cluster, described in Chapter 5. This is a period of trading identified by
gaps on either side, either above or below the established range. Once price
retreated back, the long lower shadow on the first white session revealed that
continuation of the consolidation trend was unlikely. In fact, price did begin
trending upward at that point. The island cluster was a final confirmation that the
consolidation trend was over.
Can signals like the inverse head and shoulders apply during consolidation?
Under typical definitions of “reversal,” a signal is valid only in a bullish or
bearish trend, and there is no clear trend to reverse. However, this concept
should be challenged with a new definition of “reversal.” A signal can reverse
the consolidation trend by setting up a new dynamic trend. This includes any
Western or Eastern technical reversal signal located within consolidation, and the
Kellogg chart demonstrates how this works.

Key Point
A careful reading of signals effectively forecasts breakout from
consolidation. This is especially true of the island cluster following
breakout.

So with a series of clear signals, how can this be traded? A hedge consisting
of uncovered puts would be effective in this situation. The three declines below
support that make up the head and shoulders were likely to reverse and move
back into range, as they did. The moment price fell below support, especially in
the case of the shoulders, where the move consisted of shadows but not opening
or closing range, an uncovered put would have effectively hedged the
consolidation of prices. Each of these uncovered puts would have been closed
once price returned into range and rose to the point of resistance or to another
price peak. The first shoulder touched resistance, marking the point to close. The
head peaked at mid-range and then began declining; this decline marked a good
price point to close. The second shoulder rose into range and immediately
trended into the Bollinger squeeze. At that point, the uncovered put should have
been closed, as the squeeze signaled a likely bullish breakout.
The island cluster is always difficult to interpret in the moment. However,
once it completes the formation, it clearly identifies a likely price move to
follow. The long lower shadow at the first session after the conclusion of the
island cluster is bullish (combining the island cluster with the shadow). So it
appears that the breakout from consolidation was succeeding, but could this be
confirmed? The first two sessions after the conclusion of the island cluster,
consisting of the white session with the long shadows and the following black
session, formed a thrusting lines signal, a bullish confirming indicator. This
would be another excellent spot to open an uncovered out. This put could be
closed at the peak with upper shadow occurring in mid-September. Finally, the
long white candlestick in late October identified an excellent spot to open a new
covered call.
This detailed analysis reveals that consolidation can be a signal-rich period in
which option hedges are effective. A covered call writer has to move cautiously,
however, because consolidation can end suddenly and move in either direction.
So in addition to looking for breakout reversal signals (reversal of
consolidation), the use of uncovered puts as a flexible alternative to covered
calls continues the hedge even when price is range-bound.

Contingent Purchase Strategy


Most descriptions of selling uncovered puts address risk levels when shares of
stock are not owned. So in this context, the uncovered put is seen as a
speculative trade. However, it can also be used as a form of hedge when you do
own shares, especially if the price of shares has increased.
In this situation, you face a dilemma: Do you sell and take profits, avoiding
the possibility that the price could reverse and absorb all of the profit? Or do you
hold, hoping for continued profits? A third possibility and a form of hedge is to
buy an insurance put. In this strategy, the long put’s value increases for every
point lost in the stock below intrinsic value. However, because you have to pay
for the put, the true cap on loss consists of the strike of the put minus the cost of
the put. For example, you buy a 35 strike put and pay 2. Your offset will begin
after the breakeven price of $33 per share. So if you paid $30 per share, this
locks in at least $300 in profits even if the stock price falls below your original
basis of $30.
Another way to hedge appreciated stock is by selling uncovered puts. This is
an opposite strategy from the purchase of a long put, but the purpose is also
different. With the long put, you want to protect your paper profits. With the
short put, you set up a hedge with one of two outcomes: profit in the short put, or
purchase or more shares.
For example, you bought 100 shares @ $48 per share. Today, the stock value
has grown to $57 and you are concerned that the value could decline. By selling
a 55 put for 6, you set up a short position three points in the money. However,
the breakeven on this is actually $49 (55 – 6), so that even if the put were
exercised, your acquisition cost will be lower than the strike. The net breakeven
is $49 per share, compared with your original basis of $48. Exercise increases
your position at a price very close to the original. The average basis in stock
(original 100 shares plus another 100 shares put to you upon exercise) is $42.50:
( ( $48 + $49 ) ÷ 2 ) – $ 6 = $42.50

As long as the current price per share at the time of exercise is higher than
$42.50, the overall position remains profitable. Even with a large decline from
the strike of 55 down to $42.50, or 12.5 points, your original basis remains intact
due to the premium from the short put.
Another possible outcome that avoids exercise is closing the short put. At the
time it was opened, the premium of 6 included two points of intrinsic value. A
drop below the 55 strike eliminates all of the intrinsic value, and in addition time
value will decline as well. So the short put can be closed at a profit. It can then
be replaced with other short puts, covered calls, or any number of other hedges.
The short put used to hedge against appreciated stock hedges market risk and
reduces your overall net basis, adding a safety net to lost paper profits. It
provides the same type of hedging value as the covered call, but with one
important difference: With the covered call, replacement involves consideration
of the strike versus original basis in the stock, but with the uncovered put, this
does not matter. Exercise can be avoided by rolling forward to any strike that
produces a net credit. Although early exercise is always possible, it is remote as
long as time value remains part of the put’s premium and as long as the put is at
the money or out of the money.
If you are willing to acquire additional shares as the result of this contingent
purchase strategy, the short put is an excellent hedge. In the event the price of
the underlying does not decline enough to move the put in the money, the
alternative of closing or rolling provides added income along with the hedge
against loss.

Covered Straddles
The covered call and uncovered put can also be combined into an effective
and powerful form of hedging. When two short options are opened at the same
time, it often indicates a high-risk and speculative move. However, the covered
straddle is conservative and presents an exception to this general rule by
providing income and hedging protection together.
The term covered is somewhat inaccurate. The call is covered by the stock,
but a short put cannot be covered in the same sense of the word. So although this
is a straddle (two offsetting options with the same strike and expiration), it is a
combination of two conservative hedges: the covered call and the uncovered put.
The covered straddle can be opened in a consolidation trend, just like the
uncovered put. However, this presents a danger for the covered call. A sudden
price move out of the fixed range may present a particular form of risk
regardless of price direction. So the most sensible point in the price trend for a
covered straddle starts at the bottom of a downtrend, assuming that clear reversal
signals are present and strongly confirmed. This is a point to open the short put;
and once the bullish trend has been established, the second leg, the covered call,
can be opened as well, completing the straddle.
For example, the chart of Alleghany (Y) reveals a clear and strong set of
signals for reversal, as shown in Figure 7.4.

Figure 7.4: Reversal Point for a Covered Straddle


Key Point
Reversal signals found at the right proximity clearly mark the best points
for opening uncovered puts.

On this chart, the bottom of the price trend is marked clearly with the
unusually long lower shadow. This reveals that sellers tried to move price lower
but failed. However, given the signal’s proximity to the marketwide decline of
August 24, this signal by itself is not enough. However, by the end of the month,
two additional signals appeared: a bullish harami and a bullish engulfing. This
reveals that the bottom has been reached and prices will begin moving higher.
This is the point to open an uncovered put.
The signals for the covered call occurred on October 16, with a strong
gravestone doji. This consists of a doji and an unusually long upper shadow. The
signal here is that the bullish rally may be weakening or ending. However, price
continued moving upward. In this mid- to late-October period, opening a
covered call at the 500 or 505 strike would have been well timed, given the
strong gravestone doji pointing to weakness in the trend.
In this case, the two legs created the covered straddle. The uncovered put
would be left intact for the moment because price rose from the $460 range
above $500 per share. So any uncovered put opened with a 460 or 470 strike
would have lost most of its value by this point and could be closed; however,
given the point spread between strike and current value, there is no urgency to
close the put.
A covered straddle also makes sense here because of the lack of clear signals
in either direction. The gravestone doji predicted the end of the bullish trend, but
price did not react as expected. This indicates a likely consolidation. In fact, in
the half-month after this chart, price remained range-bound between $495 and
$505. This lack of movement would take both options down to low levels, so
they could be bought to close at a profit, or left open to expire worthless.
The range of price movement explains the value of the covered straddle.
Opening two uncovered puts does not always make sense and could pose
excessive risk in the case of market decline. Opening two covered calls would
require buying an additional 100 shares and also poses added risk. With the
covered straddle, owning 100 shares allows you to double the hedge value
without changing market risk. In addition, opening the position in legs
maximizes the potential gains to be earned from this position.
Once the bullish reversal appeared strongly, timing for the uncovered put
made sense, but the covered call would not have been well timed. However, once
the bullish trend played out and signs of weakness appeared (meaning either a
coming bearish move or consolidation), adding the covered call made perfect
sense. The subsequent consolidation trend confirmed the wisdom of this
position. However, given the price of the stock in the example and the volatility
seen in historical price movement, this position has to be monitored closely and
one or both sides closed if price begins trending once again.
The uncovered put, by itself or in combination with a covered call, is an
effective hedging strategy. The combined use in the form of a covered straddle is
a timing hedge, whereas alternating between the covered call and uncovered put
can be timed for the peaks and valleys of price movement in swing trends or
secondary trend patterns.
The next chapter expands the hedging strategy by explaining the use of
spreads to protect equity positions and control market risks.
8 Hedging With Spreads

The spread is any position with two offsetting options, which can be bullish,
bearish, or neutral. So as a hedging mechanism, the spread is a flexible and
varied form of option trade.

Key Point
Spreads can be set up in so many varieties that they can be designed as
hedges in all market conditions.

Spreads in their most basic form consist of the same expiration date but
different strikes (the vertical spread). For example, a short call expiring in
October with a 50 strike and a short call expiring the same month with a 45
strike set up a vertical spread.
Another variety is made up of the same strike but different expirations. This
horizontal spread is intended to exploit the time value differences between the
two positions. For example, a long October 50 with a short November 50 put
creates a horizontal spread. This is also called a calendar spread or a time spread.
A final version of the basic spread combines different strikes and different
expirations. This is called a diagonal spread. For example, a long October 50
call and a short November 55 call combine to create a diagonal spread.
Figure 8.1 summarizes the three formations of the basic spread.

Figure 8.1: Types of Basic Spreads


These various formations of the spread work as hedges for your portfolio’s
equity positions just as they may also represent highly speculative option
strategies. An additional variation in this universe of spreads is that positions can
consist of either calls or puts, and may also be long or short. When the attributes
of a spread are evaluated by themselves (often as speculate trades), it is easy to
overlook the true risks. Even a wide distance between strikes in a short position
does not protect you from exercise risk, although it does reduce that risk.
However, opening such a spread just as a matter of estimating outcomes is
purely speculative. However, when spreads are associated with portfolio
positions and intended to offset equity-based market risks, the hedging effect of
spreads is substantial.

The Selection of a Debit Spread or Credit Spread


A spread may be either a debit spread or a credit spread. In a debit spread,
the net outcome is payment of the difference between a long and short premium.
The purchase of the long position costs more than the premium received on the
short side.
A bull spread is designed to yield a profit when the underlying price rises. It
can consist of either calls or puts. A call bull spread is one of a debit spread. It
combines an in-the-money long call with a higher-strike out-of-the-money short
call. In this debit spread, the maximum profit is the net of the short strike value,
minus the long strike value.
The profit and loss are both limited in this strategy. The upside profit is
maximized when the underlying price moves higher than the highest strike, and
will be equal to the difference between the two strikes, less the initial debit:
Maximum profit = Short strike – long strike – net premium paid

Offsetting this limited profit is a limited loss on the downside. A loss occurs
when the price moves lower, but it cannot be lower than the initial net debit:
Maximum loss = net debit of the position

The call bull spread’s breakeven point is the long side strike plus net premium
paid:
Breakeven = long strike + net debit

For example, on November 6, 2015, Exxon Mobil (XOM) closed at $84.47.


At that time, the following options could be opened to create a call bull spread:

Key Point
Understanding the formulas for maximum profit, maximum loss, and
breakeven price is an excellent tool for managing risk and identifying the
viability of a particular spread.

November 84 long call, ask $152


1.43
November 85 short call, bid 70
0.79
Net debit $ 82

Note that in this example, the long position is based on the ask price and the
short position employs the bid. The long is also increased by $9 to reflect trading
costs, and the short is reduced in the same way to reflect reduced premium
income.
In this example, maximum profit is $18 (difference in the two strikes, minus
the net debit, or 85 – 84 – 0.82 = $18). Maximum loss is the net debit of $82.
Breakeven is $84.82 per share (long strike plus net debit).
When puts are used in a bull spread, the position is opened as a credit.
Because puts gain the most when moving against the stock’s price direction, the
bull spread is also going to be a net credit. For example, a put bull spread
consists of a long out-of-the-money put with a higher-strike short in-the-money
put. This position can be opened on XOM using the following positions:

November 84 long put, ask $182


1.73
November 85 short put, bid 201
2.10
Net credit $19

The maximum profit is limited to the net credit received for opening this
position, which occurs if and when the underlying price moves above both
strikes. Maximum loss is also limited, and is equal to the net difference between
strikes, less the net credit:
Maximum loss = short strike – long strike – net credit

Break is equal to the short strike minus net credit:


Breakeven = short strike – net credit

In the example of XOM and a put bull spread, maximum profit is equal to the
net credit of $19. Maximum loss is $81, the net of the short strike minus the long
strike and net credit ($85 – $84 $0.19). Breakeven is the $182, the difference
between short strike and net credit ($201 – $19).
The hedging attribute of a bull spread is a combination of limited profits on
one side, versus limited losses on the other. However, with the spread, you have
an advantage beyond the profit or loss price ranges. With time decay, the short
call or short put can be closed to take profits if price moves down (for the call)
or up (for the put). Although buying to close changes the overall net debit or
credit, it leaves the long side intact with the possibility of accumulating profits if
price continues moving favorably. Eventually (once the net cost has been
exceeded) the long option can be closed at a profit. In this respect, both sides
offer hedging benefits. In the event that price moves up (for the call) or down
(for the put), the short option moves in the money and has to be closed or rolled
to avoid exercise. The maximum profit or loss can be altered by closing the short
side when time decay makes that side profitable; however, as a hedge, the
intention should be to protect the equity position, not only to generate profits. In
closing a short side, it is possible to wipe out any net gain while also losing the
hedging advantage from the bull spread.

Key Point
In deciding to close one part of a hedge, keep the overall purpose in mind,
protecting equity and not merely generating net income.

The bear spread works in the opposite direction. Profits are maximized when
the underlying price declines, so the bear spread works as a hedge against market
risk in the equity, as the overall position’s growing value offsets declining value
in the stock.
Like the bull spread, bear spread profits and losses are both limited. Profit is
maximized when the underlying price advances higher than the high strike. The
maximum profit is equal to the net credit received. Maximum possible loss
occurs if the underlying price moves above the higher strike. This loss is equal to
the difference between strikes, less original credit:
Maximum loss = long strike – short strike – credit received

Breakeven in this strategy is equal to the short strike plus credit received:
Breakeven = Short strike + credit received

For example, on November 6, 2015, Exxon Mobil (XOM) closed at $84.47.


The following options could be opened to create a call bear spread:

November 85 long call, ask $101


0.92
November 84 short call, bid 116
1.25
Net credit $15

The maximum profit is the credit of $15. Maximum loss is the difference
between the long and short strike, minus the credit: $85 – $84 – 0.15 = 0.85, or
$85. Breakeven is equal to the short strike plus credit received, or 84 + 0.15 =
$84.15.
The final version of this basic spread is the put bear spread. This combines a
long in-the-money put and a lower-strike short out-of-the-money put.
Maximum profit is achieved when the underlying closes below the lower
strike. With both options in the money, the higher long strike will have more
intrinsic value than the short side. So maximum profit equals the difference
between the two strikes, minus the debit:
Maximum profit = long strike – short strike – debit

Maximum loss occurs if the underlying rises higher than the higher strike, and
the loss is equal to the net debit received. The breakeven is the long strike minus
the debit:
Breakeven = long strike – debit

For example, XOM’s options as of November 6, 2015, could be used to set up


a put bear spread with the following:

November 85 long put, ask $237


2.28
November 84 short put, bid 201
2.10
Net debit $ 36

The maximum profit equals the difference between the two strikes, less the
debit: 85 – 84 – 0.36 = 0.66, or $66. Maximum loss is the net debit of $36.
Breakeven is calculated as the net of the long strike minus the debit: 85 – 0.36 =
$84.64.

Key Point
The various forms of spread enhance your ability to hedge equity, by
providing ease of switching from one type to another based on market
conditions.

These examples of basic credit and debit spreads using calls and puts, either
bullish or bearish, set up the theme of the put: limited profit and loss with great
flexibility. Although the dollar values used were small for the purpose of the
examples, your imagination can be put to work to devise higher profit and loss
levels with wider gaps between strikes, and with the use of farther-out expiration
dates and higher time value.
As hedges, the basic vertical hedge is attractive because of the tightly limited
profit and loss levels. So this strategy works well during periods of low volatility
and especially during consolidation. Like the uncovered put analyzed in the last
chapter, the vertical spread creates limited potential income in exchange for
limited potential limited loss. Owning stock in your portfolio reduces the short
side risk for calls, making the vertical spread desirable when waiting out the low
volatility.

Box Spreads
Another type of spread expands the vertical spread by combining both call
and put versions. The resulting box spread is either long or short, and the
selection of one over the other depends on your belief about the price direction.
When this is based on chart patterns and confirmed reversal signals, the box
spread can work as one form of hedge.
A potential problem with the box spread is that it employs four different
positions. As a result, the trading costs will be higher for single options than
strategies based on single options or on offsetting two-part positions. For this
reason, the box spread is most likely to net the desired result when it is used for
multiple positions with a lower average transaction cost. However, in the
following examples, single option positions are used to demonstrate how the box
spread is constructed and how its maximum profit and loss are calculated.
The box spread sets up a hedge in which profits can be locked in with no risk.
This could be viewed as a standalone strategy; however, when the box spread is
designed to hedge portfolio positions, it takes on even greater value. For
example, if shares of stock are depressed, a paper loss can be absorbed or
partially absorbed with a box spread.

Key Point
Although the box spread sets up locked-in profits, it can also be viewed as a
worthwhile hedge in consolidation periods or as part of a recovery strategy.

The position is established by opening the same number of long in-the-money


calls, short out-of-the-money calls, long in-the-money puts, and short out-of-the-
money puts. The value of the resulting box at expiration will be equal to the
higher strike less the lower strike, and the profit will equal the expiration value
minus the net premium paid.
For example, referring once again to Exxon-Mobil as of November 6, 2015,
the stock closed at $84.47. The following options could be used to create a box
spread (although single options are used, the trading fees make larger sets of
options more desirable):

November 83 long call, ask 1.97 $206


November 85 short call, bid 0.79 – 70
November 85 long put, ask 2.28 2.37
November 83 short put, bid 1.16 –
1.07
Net debit $266

The expiration value is the difference between the spread prices for each side,
or $200. Profit will be $200 for the call spread plus $200 for the put spread, for a
total of $400. Profit will be $134, or the difference between expiration value and
the debit ($400 – $266 = $134). This outcome is summarized for various
underlying prices at expiration, in Table 8.1.

Table 8.1

In practice, the box spread is likely to be closed in increments. When the


short side of either the call or put spread loses most of its value, the positions can
be closed at a profit, with the long sides left until expiration or shortly before. If
the stock moves several points, either the long calls or the long puts will become
profitable. However, if you close parts of the box spread, be aware of the
maximum net profit and be sure your buy-to-close costs do not exceed that price.

Ratio Calendar Spreads


The calendar spread is another name for the horizontal spread, consisting of a
position with the same strike but different expiration dates (also called a time
spread). The ratio calendar spread combines a short option with two or more
long options with the same strike but a later expiration; or it may combine a
larger number of short options with a smaller number of later-expiring long
options with the same strike.

Key Point
A ratio calendar spread with more long than short options is not a sensible
hedge because it sets up a debit in every case.

When you combine short options with a larger number of long options, the
short spread is not at risk of exercise because it is covered by the later-expiring
long positions. As a hedge, the ratio calendar spread provides benefits in
segments. First, the earlier expiring short position will lose time value more
rapidly than the longer-term long positions; second, the long options remaining
after the short option has expired or been closed provide potential benefits if the
underlying price rises (when calls are used) or falls (when puts are used). The
position will create a net debit because longer-term options have more time
value.
The second version of the ratio calendar spread is more interesting as a hedge,
and it relies on rapid time decay in the shorter-term short options. In this case,
you open more of the short options and fewer of the long. As a hedge, this is
intended to provide a degree of protection for your long positions, not only in the
rapid time decay but also in the longer-term long options that will remain after
the short side is closed or expires. Although there is a risk in opening more short
options than long, exercise risk is managed by either buying to close some or all
of the short options, or rolling them forward. When using calls in this strategy,
be aware of the ex-dividend date and avoid keeping short calls open in ex-
dividend month. This is the most likely time for early exercise in the event the
short calls move in the money.
For example, Exxon Mobil closed November 6, 2015, at $84.47 per share.
Ex-dividend date has occurred the same day, November 6, and the next ex-
dividend would not be until February. The early exercise risk had passed,
because November 5 would have been the last time when this was a possibility.
The term ex-dividend means “without dividend,” so the risk exists up until the
day before.
At the close of November 6, the following options were available:
November 85 calls bid 0.79 and ask 0.82
December 85 calls bid 1.62 and ask 1.78
January 85 calls bid 2.16 and ask 2.37
November 85 puts bid 1.84 and ask 1.99
December 85 puts bid 2.94 and ask 3.15
January 85 puts bid 3.55 and ask 3.70

Key Point
The hedge with a higher number of short-term short positions creates a net
credit. As long as you can manage that credit, the ratio position is effective
and profitable.

In each case, opening a short position with a higher number of long positions
would not make sense, as the overall position would create a large debit. So the
focus in the following examples is on the more practical hedge, opening a larger
number of soon-to-expire short options with a smaller number of later-expiring
long positions. The calls are all slightly out of the money and the puts are all
slightly in the money.
A call-based ratio calendar spread could consist of:
Two short November 85 calls (12 days to expiration), @ 0.79,
= $158 minus $10 for trading costs = $148

One long December 85 call (40 days to expiration), @ 1.78,


= $178 plus $9 for trading costs = $187

This sets up a net debit of $39. The hedge in this case applies if you believe
the stock price will rise in the longer term. Locking in a close-to-the-money long
call with 40 days remaining before expiration at a net cost of only $39 is a great
advantage, assuming the two short calls can be managed for the next 12 days.
Time decay will be rapid, so as long as the stock price remains at or below $85
per share, the entire short premium of $148 will be profitable. However, if the
price trends higher before the November expiration, these can be rolled forward
into either the weekly expirations occurring before December expiration, or
closed at a net profit. However, closing these positions will increase the basis in
the later-expiring long options. The net cost of a buy-to-close of the November
calls increases the net $39. Even so, the position still ends up with one December
85 long call; because the underlying price increased, this long position gains
intrinsic value.
There is a risk in this bullish use of the ratio calendar spread, in two respects.
First, the earlier expiring short positions could move in the money and would
have to be managed through rolling or closing. Second, even after the short
expiration, if the stock price does not rise, the position ends up with the small
$39 net loss. This is a reasonable risk level in exchange for the potential hedge
gained if the price did rise.
If you believe the stock price is more likely to decline, a ratio calendar spread
can be opened using the 85 puts. For example, a put-based ratio calendar spread
could consist of:
Three short November 85 puts (12 days to expiration), @ 1.84,
= $552 minus $11 for trading costs = $541

One long January 85 put (68 days to expiration), @ 3.70,


= $370 plus $9 for trading costs = $379

This alternative sets up a position with a net credit of $162. The three short
puts are slightly in the money, so exercise risk exists for the next 12 days. Two
factors reduce this risk. First, if the underlying price increases by 0.53 points or
more, the November puts will not be in the money. Second, over the next 12
days, the 1.84 points of value (consisting of 1.31 points of non-intrinsic value)
are going to evaporate. These short positions can be closed, even if they move in
the money. Given the 1.84 points of value, the breakeven price is $86.84 per
share.

Key Point
The net outcome that sets up a free or low-cost insurance put is highly
desirable as a hedge and preferable to just buying a long put.

With a net credit of $379, you have flexibility even if you end up having to
close the short puts to avoid exercise. They can be closed or rolled forward to
one of the weekly options. The roll avoids exercise while increasing the net
credit in this overall position. Assuming the short puts are eventually closed or
allowed to expire, you end up with one long January 85 put, probably without
any net cost and a greater likelihood of realizing a net credit. So this cost-free
put provides two months of downside protection, a free insurance put to protect
100 shares in your portfolio.
If you are able to manage the short puts to avoid exercise, this is a “perfect”
hedge because the long put is free, while capping maximum losses in the 100
shares of stock at $85 per share. Any price decline in the stock below $85 will be
offset by a corresponding increase in the value of the long put; and because the
put is “free,” it can be closed at any point before expiration to offset losses in the
stock.
A further expansion of the ratio calendar spread combines both the call and
put positions into a single position. This is called a ratio combination calendar
spread. This would normally involve the two sides, both slightly out of the
money; for example, the XOM 85 calls would be combined with a set of 82.50
puts.
On November 6, the following 82.50-strike puts were available:
November 82.50 puts bid 0.99 and ask 1.14
December 82.50 puts bid 1.83 and ask 2.00
January 82.50 puts bid 2.37 and ask 2.57

The complete ratio calendar combination spread could be set up using:


Three short November 82.50 puts (12 days to expiration), @ 0.99,
= $297 minus $11 for trading costs = $286

One long January 82.50 put (68 days to expiration), @ 2.57,


= $257 plus $9 for trading costs = $266

This creates a credit of $20. To make the combination complete, add in the
call-based ratio side using the 85 strikes:
Two short November 85 calls (12 days to expiration), @ 0.79,
= $158 minus $10 for trading costs = $148

One long December 85 call (40 days to expiration), @ 1.78,


= $178 plus $9 for trading costs = $187

With this debit of $39 on the call side and the credit of $20 on the put side,
the overall net cost is only $19. This creates a hedge in which the longer-term
call and put create profit potential on the upside and the downside. If the
ultimate price of the underlying remains between $84 and $85, you lose $19.
The risk in this position involves the 123-day short calls and puts. As long as the
stock price remains between $84 and $85 per share, all of these short options
will expire worthless—but that is a very narrow window. It is more likely that
one side or the other will move in the money. Closing these to avoid exercise
increases the overall debit; rolling forward to a weekly expiration creates a credit
due to more time value, and helps manage the position. However, it is likely that
eventually one or more of these positions will expire worthless and the other side
will have to be closed. Given the short time remaining for expiration of short
calls and short puts, this is not a long-term risk concern, but a short-term
management issue. The ultimate hedge is the result, a low-cost or cost-free
combination of a long 85 call and a long 84 put, both expiring in January.

Diagonal Spreads

Key Point
The diagonal spread is an effective strategy for managing short-term price
movement, assuming you have a reasonable sense of price direction.

The diagonal spread combines features of the vertical and combination


spread, consisting of different expiration dates and strikes. It is an advantageous
form of spread that takes advantage of price swings, providing a short-term
hedge to equity positions and also managing downside risk.
When you own equity positions in your portfolio, your emphasis is on long-
term growth. However, short-term price fluctuations are also of concern, and this
is where the diagonal spread works as a worthwhile hedge. In fact, the two sides
of this spread can both become profitable if price behaves in a specific manner,
moving first in one direction and then back in the other. The use of different
strikes means it is possible to use the same number of options on either side,
while also creating a net credit. The ability to do this relies on time to expiration
and proximity of strikes to the underlying price.
For example, if you expect the underlying to fall in the short term and then
rebound, a diagonal call spread will be ideal. For example:

This sets up a position close to breakeven, with a net credit of only $7.
However, if the price remains below the short 85 strike for the next 12 days, the
short call will expire worthless. It can also be closed, but the cost of the buy to
close trade will absorb and surpass the net credit, converting the remaining long
option to a net debit. Even so, the new net will be much lower than the cost of
just buying the long call. Because the overall cost for the January long call is
going to be a small credit or a small debit, it is an inexpensive hedge that
exploits any price growth without having to take profits by selling shares. If the
rise does not materialize, it is an inexpensive hedge that loses only a small
amount in the worst case. The short call can also be rolled forward to a weekly
position if the underlying moves in the money during the next 12 days, setting
up additional credit.
This diagonal spread can also be varied by adding a ratio feature. For
example, combining three short December 85 calls and two long January 87.50
calls sets up a diagonal ratio spread:
3 December 85 short call, bid @ 1.62 = $486 – $11 = $475
2 January 87.50 long call, ask @ 1.37 = $274 + $10 = $284

Key Point
The larger the net credit from the diagonal spread, the greater your
flexibility in closing the short side without moving into a debit.

In this ratio version, you create a net credit of $191. Two of the three short
calls are exposed to exercise risk; however, as in all forms of short positions,
these can be closed or rolled if the underlying moves in the money. With a $191
cushion, these can be closed after time decay while keeping a net credit intact.
Clearly, the most advantageous outcome would be for a short-term price decline
(allowing the short options to expire worthless) followed by a price increase,
making the long positions profitable. However, as long as the short options
expire worthless, even if the price does not rise, the net credit is the profitable
result of this strategy.
If you believe the price will rise in the short term and then decline, the
diagonal put spread makes more sense. You can create a net credit even using the
same number of options, and the strategy calls for closing or rolling in the event
the stock price falls below the short put strike. The ratio version of the put
diagonal also works, creating a larger net credit and a cushion in case you want
to close the short puts. The outcome is a longer-term long put and a “free”
insurance put.
Spreads are effective hedges and can be used in many different combinations
and risk levels. The next chapter demonstrates how advanced form of spread, the
butterfly and condor, can create a position with limited profits and losses and
potential hedging benefits when the underlying moves in either direction.
9 The Butterfly and Condor

Spread strategies described in the last chapter create strategies with limited
profit potential, in exchange for maximum and limited losses. Expanding on this,
more advanced hedges can be created in various ways. One of the most popular
of these is the butterfly.
A butterfly is a spread with three strikes, made up of calls, puts, or both. The
high and low strike are both long and the middle strike is short; or high and low
are short and the middle strike is long. Normal construction calls for two middle
strikes versus one each of a higher and lower strike.
The four varieties of the butterfly follow, with examples based on closing
values as of November 6, 2015, for Microsoft (MSFT), which closed at $54.92.
December options are used because MSFT’s ex-dividend date occurred in late
November. Each premium value is adjusted for estimated trading costs: (Note:
Margin for each of these was calculated on the Chicago Board Options
Exchange (CBOE) free margin calculator [www.cboe.com/tradtool/mcalc.].)
Long call—1 long ITM call, 2 short ATM calls, and 1 long OTM call

MSFT 1 DEC 52.50 long call, ask $


2.83 292
MSFT 2 Dec 55 short calls bid 1.17 $ –
224
MSFT 1 Dec 57.50 long call ask $
0.40 49
Net paid (margin $89) $
89

Short call—1 short ITM call, 2 long ATM calls, and 1 short OTM call
MSFT 1 DEC 52.50 short call, bid $ –
2.72 263
MSFT 2 Dec 55 long calls, ask 1.21 $
252
MSFT 1 Dec 57.50 short call, bid $ –
0.29 29
Net received (margin $189) $
40

Long put—1 long ITM put, 2 short ATM puts, and 1 long OTM put

MSFT 1 DEC 52.50 long put, ask $


0.67 76
MSFT 2 Dec 55 short puts, bid 1.59 $ –
308
MSFT 1 Dec 57.50 long put, ask $
3.45 354
Net paid (margin $94) $
122

Short put—1 short ITM put, 2 long ATM puts, and 1 short OTM put

MSFT 1 DEC 52.50 short put, bid $ –


0.64 55
MSFT 2 Dec 55 long puts, ask 1.64 $
338
MSFT 1 Dec 57.50 short put, bid $ –
3.25 315
Net received (margin $89) $
32

The outcomes for profit and loss also vary based on configuration of the
butterfly:
Long Call
Maximum profit = short strike – lower long strike – net premium paid 55 –
52.50 – 0.32 = 2.18, or $218
Maximum loss = net premium paid
= $32
Upper breakeven = high strike – net premium paid
57.50 – 0.32 = $57.18
Lower breakeven = low strike + net premium paid
52.50 + 0.32 = $52.82
Short Call
Maximum profit = net premium received
= $40
Maximum loss = long strike – lower strike – net premium received
55 – 52.50 – 0.40 = 2.90, or $290
Upper breakeven = high strike – net premium received
57.50 – 0.40 = $57.10
Lower breakeven = low strike + net premium received
52.50 + 0.40 = $52.90
Long Put
Maximum profit = high strike – short strike – net premium paid
57.50 – 55 – 1.22 = 1.28, or $128
Maximum loss = net premium paid
= $122
Upper breakeven = high strike – net premium paid
57.50 – 1.22 = $56.28
Lower breakeven = low strike + net premium paid
52.50 + 1.22 = $53.72
Short Put
Maximum profit = net premium received
= $32
Maximum loss = high strike – long strike – net premium received
57.50 – 55 – 0.32 = 2.18, or $218
Upper breakeven = high strike – net premium received
57.50 – 0.32 = $57.18
Lower breakeven = low strike + net premium received
52.50 + 0.32 = $52.82

Key Point
The outcomes of butterfly positions assume all are held open to expiration.
In practice, it is likely that half of these positions will be closed early to
take profits.

These four variations of the butterfly are straightforward in construction and


have predictable outcomes. In practice, traders using butterflies tend to close
legs as they become profitable, in the hope of achieving overall profits higher
than the maximum if held to expiration. As hedges, the long or short butterfly
has limited value but may generate current income. It is a worthwhile part of
recovery strategies for depressed stock prices, or for exploiting volatility. In all
of the configurations, one side or the other will become profitable as the stock
price moves. The challenge is to time trades to protect against downside risk and
to end up with a net profit after a portion of the butterfly has been closed.
The bullish nature of the long call and short put butterfly strategies, and the
bearish nature of the short call and long put strategies, are clear. However, a
butterfly can also be constructed using a combination of calls and puts.

The Iron Butterfly


The iron butterfly is a variation of the basic butterfly, designed to work best
for low-volatility stock situations. It combines calls and puts together, using four
different option positions and three strikes.
The maximum profit on the iron butterfly is the amount of the net premium
received.
Maximum loss is equal to the long call strike minus the short call strike minus
the net premium received:
Maximum loss = long call strike – short call strike – net premium received

This loss occurs in two conditions: first, when the underlying price is greater
than the long call strike, and second, when the underlying price is lower than the
long put strike.
Breakeven occurs in two ways: On the upside, it is the short call strike plus
net premium received, and on the downside, it is the short put strike minus the
net premium received:
Upper breakeven = short call strike + net premium received
Lower breakeven = short put strike – net premium received

The iron butterfly combines an OTM long put with ATM short call and put
and an OTM long call. For example, Microsoft closed on November 6 at $54.92.
An iron butterfly could be constructed using the following contracts:
1 December long 52.50 put @ 0.67
1 December short 55 call @ (1.17)
1 December short 55 put @ (1.59)
1 December long 57.50 call @ (0.40)

The net credit from this position is 1.69 before deducting trading costs
(estimated $36), for a net total credit of $133.
The changes in value from November 9 through the following five sessions
are summarized in Table 9.1.

Table 9.1

Key Point
The hedge from an iron butterfly involves generating profits based on price
activity in the underlying.

The maximum profit or loss and breakeven all assume these positions are
held open to expiration, which in this example would be December 18. However,
in practice, the hedging benefit comes from closing each portion of the iron
butterfly as it becomes profitable. The goal in this is to take profits without
moving the net below the initial amount received. In this example, the net
premium received was $133.
The following reveals what occurred in the week following the opening of the
iron butterfly. Several steps could be taken to take advantage of the changing
price of the underlying. The closing stock price gradually declined through this
week:

December $54.92
6
December $54.16
9
December $53.51
10
December $53.65
11
December $53.32
12
December $52.84
13

In managing this position with a declining stock price, the initial hedge occurs
when the lower-strike long call appreciated in value and the short call declined in
value. As Table 9.2 shows, both of these became profitable due to the stock’s
decline:

Table 9.2
Option Value on Nov. 6 Value on Nov. 13
Long 52.50 put 0.67 1.45
Short 55 call (1.17) (0.49)

In closing both of these positions, profits can be taken. Profit on the long put
is calculated as:
1.45 – 0.67 – 0.18 fees = 0.60

Profit on the short call is calculated as


1.17 – 0.49 – 0.18 fees = 0.50

Key Point
A danger in any butterfly is closing positions that turn a net credit into a net
debit. This is resolved at least partially by closing profitable long and short
sides together.

The total profit is $110 net of transaction fees on both sides of the trades.
However, a net of $96 is spent to close these positions ($145 – $49). This
reduces the initial credit from $133 down to $37. The remaining positions left
open are the short 55 put and the long 57.50 call. In order for these to become
profitable, it will be necessary for the underlying to rise in value.
An alternative to taking profits so quickly would be to let the profits ride and
eventually take profits after the overall net was smaller. Leaving these open to
expiration and then closing all positions would yield a profit of $133, breakeven
of either $53.67 or $56.33, or a maximum loss of $117 ($57.50 – $55 – $1.33 =
1.17, or $117). So with the iron butterfly, both profits and losses are limited and
you have a choice: let positions ride to expiration, or take profits as they
materialize. The position works effectively as a hedge in two ways. First, as long
as the underlying continues to exhibit low volatility, the short middle-strike
options will decline in value and become profitable, while the relatively cheap
OTM lower put and upper call hedge against unexpected underlying price
movement in either direction. Second, if the underlying price moves unexpected
early on, one-half of the positions will become profitable and the other half
(consisting of one long and one short option) will lose value. As the remaining
short option moves farther in the money, it can be rolled forward to avoid
exercise. However, if you own the underlying, this outcome converts the
outstanding short call to a covered call, or the outstanding short put to an
uncovered out.

The 1-2-3 Iron Butterfly


An expansion of the butterfly is intended to set up a special type of hedge
involving three different sets of expirations for the strike ranges of the strategy.
This hedge matrix is designed so that one-half of the positions are always
moving in a profitable direction; and so that time value of short positions
declines as each strike moves closer.
For example, Microsoft options could be applied to create a 1-2-3 iron
butterfly, by opening the following positions:

1 November long $ 25 + $ 34
52.50 put @ 0.25 $9
1 November short 55 $ ( 61) - $( 52)
call @ (0.61) $9
1 November short 55 $ (100) $( 91)
put @ (1.00) - $9
1 November long $ 11 + $ 20
57.50 call @ 0.11 $9
Net credit $ (89)

2 December short $ (128) $


52.50 puts @ (0,64) - $10 (118)
2 December long 55 $ 242 + $ 252
calls @ 1.21 $10
2 December long 55 $ 328 + $ 338
puts @ 1.64 $10
2 December short $ 76 - $( 66)
57.50 calls @ (0.38) $10
Net debit $406

3 January long 52.50 $ 330 + $ 341


puts @ 1.10 $11
3 January short 55 $ (489) $(477)
calls @ (1.63) - $11
3 January short 55 $ (609) $(598)
puts @ (2.03) - $11
3 January long 57.50 $ 228 + $ 239
calls @ 0.76 $11
Net credit $(495)
Overall net credit $(178)
Key Point
A 1-2-3 iron butterfly is designed to create multiple hedging positions over
a series of expirations.

The configuration is designed with the attributes of the 1-2-3 position:


increasing numbers of options (meaning the trading cost was also adjusted and
rounded), with the middle expiration set up as a reverse iron butterfly.
In this strategy, the hedging benefit is derived from the profits to be earned
when the stock price moves in either direction. If the stock loses value, the long
puts and short calls will become profitable; and if the stock gains value, the short
puts and long calls become profitable. In the ideal situation, the stock price will
move in one direction and then in the other, so that all of the positions can be
closed at a profit. However, the goal is not to always achieve profits in all
positions, but to be able to close and take profits without moving the newt credit
below the overall $178. With this in mind, closing long and short positions at the
same time is the most likely strategy to hedge the underlying.
If the stock price remains range-bound and does not move very much at all,
especially within the first and second expiration periods, all of the short
positions, both calls and puts, will decline in value due to time decay. This will
occur rapidly for each set of expirations as the expiration date approaches. This
means that all three of the sets of butterflies (including the reverse iron butterfly)
could become profitable from the combined shorts being closed at a profit or
expiring worthless, and the long positions either being closed along with the
short positions or being closed when they become profitable. Any price
movement in the underlying will make one-half of all the positions profitable.
The many varieties of the butterfly present attractive hedging opportunities.
The expanded 1-2-3 iron butterfly expands on the basic premise for the butterfly
by presenting scenarios of profit when the underlying price moves, regardless of
direction. The short-side risk is reduced when you also own shares in your
portfolio. The short calls are covered and the uncovered short puts have the same
market risk as the covered calls. This combined market risk structure makes the
butterfly an exceptional hedge when you also hold shares of the underlying.

The Condor
The butterfly can also be expanded into a strategy that includes four options
and four strikes. A condor takes the butterfly and its three spreads and adds a
fourth spread.
Key Point
The condor is an expanded version of the butterfly. It uses two middle-level
strikes, both out of the money.

Like the butterfly, the condor is likely to be closed in stages rather than
keeping all contracts open to expiration. Closing profitable long and short sides
together helps maintain a net credit. The condor can be opened in several
configurations, using calls, puts, or a combination of both.
The long condor consists of a long ITM call, a short ITM call, and two
middle-strike calls, both OTM. For example, Boeing closed on November 6,
2015, at $147.94. A long condor could be opened using the following contracts:

This strategy creates a very small net debit, but also offers a limited profit and
loss. One side or the other will always move ITM. If held until close to
expiration, some of these positions will be profitable, and others will not.
However, the time decay of the short sides makes a difference as well.
In addition to this long strategy, a short condor using calls consists of
opposite call positions: a long ITM, short ITM, long OTM, and short OTM.
For example, the previous long condor could be converted to the short version
with the following contracts:

Short ITM call: Nov 147 @ $251 – $(242)


2.51 $9
Long ITM call: Nov 145 @ $405 + $ 414
4.05 $9
Short OTM call: Nov 149 @ $147 – $(138)
1.47 $9
Long OTM call: Nov 152.50 $ 47 + $9 $ 56
@ 0.47
Net debit $ 90

Long or short condors can also be constructed using puts. However, with any
form of condor, the maximum profit or loss is limited by the offsetting positions.
Like the three-strike butterfly, the four-strike condor is a worthy hedge when
volatility is low, but if volatility increases while these options remain open, the
hedge is not as effective. The maximum gain or loss remains in place in any
event, which is what makes the condor an appealing strategy.

The Iron Condor

Key Point
The iron condor is a further expansion, using both puts and calls.

Opening a condor with calls or puts (either long or short) can be expanded to
include both calls and puts. The iron condor is very much like the iron butterfly.
However, it has four strikes rather than three. The normal construction includes a
long OTM put and a lower-strike short OTM put; and a short OTM call with a
higher-strike OTM long call.
In the case of Boeing, an example of the iron condor includes the following,
based on closing values on November 6, 2015:

Long ITM put: Nov 145 @ $110 + $ 119


1.10 $9
Short OTM put: Nov 147 @ $163 - $9 $(154)
1.63
Short OTM call: Nov 149 @ $147 - $9 $(138)
1.47
Long OTM call: Nov 150 @ $115 + $ 124
1.15 $9
Net credit $( 49)

The reverse iron condor can be applied in times when higher volatility is
expected. The positions are reversed from those of the iron condor. For example,
the following positions set up a reverse iron condor for Boeing.

Short ITM put: Nov 145 @ $102 – $( 93)


1.02 $9
Long OTM put: Nov 147 @ $176 + $ 185
1.76 $9
Long OTM call: Nov 149 @ $158 + $ 167
1.58 $9
Short OTM call: Nov 150 @ $106 – $( 97)
1.06 $9
Net debit $ 162

The 1-2-3 Iron Condor


The 1-2-3 iron condor is the same formation as the 1-2-3 iron butterfly, but
with the additional strike added. Each expiration period adds another position
and the middle set is a reverse iron condor.
For example, the Boeing iron condor could be expanded into a 1-2-3 version
with the following positions:

1 Long ITM put: Nov 145 @ $110 + $ 119


1.10 $9
1 Short OTM put: Nov 147 $163 – $(154)
@ 1.63 $9
1 Short OTM call: Nov 149 $147 – $(138)
@ 1.47 $9
1 Long OTM call: Nov 150 $115 + $ 124
@ 1.15 $9
Net credit $( 49)

2 Short ITM puts: Dec 140 @ $264 – $(254)


1.32 $10
2 Long OTM puts: Dec 145 $536 + $ 546
@ 2.68 $10
2 Long OTM calls: Dec 150 $540 + $ 550
@ 2.70 $10
2 Short OTM calls: Dec 155 $194 – $(184)
@ 0.97 $10
Net debit $ 658

3 Long ITM puts: Jan 140 @ $ 672 + $ 683


2.24 $11
3 Short OTM puts: Jan 145 $1,050 – $(1,039)
@ 3.50 $11
3 Short OTM calls: Jan 150 $1,095 – $(1,084)
@ 3.65 $11
3 Long OTM calls: Jan 155 $ 570 + $ 581
@ 1.90 $11
Net credit $ ( 859)

Overall net credit $ (250)

The butterfly and condor are interesting as hedge strategies because they limit
losses, in exchange for also limiting profits. The opportunity based on these
strategies is especially worthwhile when the underlying price is depressed; in
that case, the butterfly or condor can serve as part of a recovery strategy. It also
tends to work effectively during periods of consolidation, when range-bound
stock prices do not move outside of the limited range between strongly held
resistance and support. Like many hedges, this is an opportunity to generate
additional income in exchange for a limited level of risk exposure.

When These Strategies Work as Hedges


For the typical hedge, two elements are required: an equity position requiring
protection and the hedge itself, such as options strategies. However, butterfly and
condor positions are usually employed by traders who seek profits in the options
trade but who do not own stock.
The exception to this occurs when you hold an equity position and you expect
price movement, but you are not certain about the direction this will take.
Timing could be based on:
1. Consolidation. When a stock has been trading in consolidation, the big
challenge is determining when or if it will break out. In a typical
consolidation, repetitive attempts moving price above or below the limited
trading range tend to fail. At some point, a bullish or bearish trend will
take over, but it is not always easy to spot this change, or to identify the
likely direction of price movement. In this case, the butterfly or condor—
especially one designed to produce a small debit or credit—will benefit in
either direction. At the same time, if price does not move, the short
positions will love time value and become profitable or expire worthless.
2. Earnings announcements. You probably have noticed that in the one to
two weeks prior to earnings announcements, a stock’s price might turn
volatile. This is caused by uncertainty about whether revenue and earnings
will be in line with expectations or produce an earnings surprise. If a
surprise does occur, the nature (positive or negative) cannot be known
either, but even a small surprise is likely to have a short-term effect on
price. This is an opportunity to set up profitable hedging outcomes with a
butterfly or condor.
3. Ex-dividend date. Another time when price is likely to act in an erratic
manner is immediately before ex-dividend date and on ex-dividend date
itself. At this time, the price is expected to drop due to the recognized
dividend earned on the day before ex-dividend. However, this may also
create an opposite reaction, anticipating a drop but resulting in a short-term
increase in the price. This is the result of investors buying shares before
the ex-dividend date, to earn the current dividend, creating increased
demand.
4. Changes in volatility. When price volatility occurs but the reason is not
clearly identified, it could be a sign that investors (most likely institutional
traders) are making large buy or sell moves. Mutual funds, for example,
might be likely to take profits immediately before the end of a reporting
quarter, to bolster the current net earnings for the period. Although this
timing is manipulative, it identifies a consequence in price levels
expressed as volatility.
5. Rumors of mergers or acquisitions. When rumors of M&A activity (or
similar fundamentals changes) are floating through the market, price may
reflect the expected volatility that results. This identifies good timing for
short-term butterfly and condor trades to hedge the equity position in your
portfolio.
The butterfly and condor are favored by traders seeking speculative profits
with hedged losses. This form of risk hedge may also translate to a longer-term
form of risk hedge against equity positions. Even when you favor long-term
holdings, awareness of short-term price movement makes trades such as the
butterfly and condor attractive, offering alternatives to closing positions or trying
to time price movement with swing trades in long or short positions.
The next chapter introduces more spread variations in the form of collars and
synthetic stock positions.
10 Collars and Synthetic Stock

You can bring tremendous flexibility to your hedging program by using several
different strategies. Some will be designed to generate income; the covered call,
uncovered put, and covered straddle are examples. Others, like the butterfly and
condor, are more defensive, representing an exchange of limits—limited profit
potential in exchange for limited loss exposure.
The selection of the right strategy relies on the volatility in the stocks you
own as well as in the overall market. This chapter describes several hedges
consisting of spread positions, all designed to accomplish a hedging advantage
and all with effectiveness based on price volatility.

The Collar
One strategy designed to set up conditional exercise of a covered call with an
insurance put is the collar. This is a combination of a slightly OTM (on-the-
money) call and OTM put. However, the collar provides a specific benefit that is
appropriate only if you are willing to exchange downside protection for the
possibility of profitable exercise.

Key Point
The collar is not practical if set up close to your original basis in the
underlying. It is a hedge that works best for appreciated stock.

The collar can be set up at any time. In analyzing it, the position makes no
sense if your stock price is at or close to original basis. With this in mind, it
works only with appreciated stock. For example, Exxon Mobil (XOM) closed on
November 6, 2015, at $84.47. Ex-dividend date was scheduled for November 9,
so entering into a covered call using options on the last day to earn dividends
would not make sense. However, the December options presented an appealing
condition for setting up a collar. The December 85 call closed with a bid of 1.62
and could be sold after trading costs for $153. The December 80 put closed with
an ask of 1.26 and could be bought for $135. The net credit for setting up this
collar would be $18.
If you had bought shares of XOM at or close to the November 6 price of
$84.47, the collar would not make much sense. If the stock price rose and the
call was exercised, stock would be called away at $85, only a few dollars above
your basis. If the stock price declined, the 80 put would cap losses at $80 per
share, but that is of little practical use if your basis in the stock was several
points higher; exercising the put would only be a way to limit losses, not to
offset or eliminate them.
However, consider the hedging value of the collar for appreciated stock. If
you had bought shares in mid-September when the stock price was ranging
between $71 and $74 per share, the appreciated value by November 6 would be
worth protecting. In fact, if you sold shares at the November 6 price, you would
make a handsome profit in less than two months. At this point, it is worthwhile
to protect profits and to set up a contingent sale. The XOM chart in Figure 10.1
shows how this works.

Figure 10.1: Timing of a Collar

The hedging benefits of the collar are of greatest value in a situation like this.
You bought stock at a lower price and you would prefer to keep shares in your
portfolio. At the same time, you are concerned with the possibility of loss, so
you want to hedge your position. In other words, you are willing to risk having
shares called away at a profit, in exchange for preventing a loss.
The entire month of September appeared to have settled into a narrow
consolidation trend. However, looking back to mid-May, the stock price had
been on a downtrend, and this could continue at any time. However, you bought
shares at $72 after the market-wide decline of August 24, in the belief that the
price had stabilized and could be on the rise. You were correct in your timing.
During October and into early November, the price rose about 20%, peaking at
$86. However, November 3 and 4 formed a double top that, though moderate,
did signal the top of the trend. By November 6, the condition of this stock was
uncertain. This was a sensible time to open a collar.

Key Point
A great advantage of positions with offsetting long and short options is that
they tend to be low-cost or zero-cost to open.

If the stock price rose above the 85 call strike and shares were called away,
your capital gain on stock would be $1,300 (per 100 shares), the difference
between basis of $72 and strike price of $85. That would be a desirable outcome
based on the call. However, exercise could also be avoided by closing the short
call or rolling it forward, good moves if it looked as though the stock had
evolved into a bullish trend. If the price declined below $80 per share, you could
exercise the put and sell shares to generate an 8-point capital gain ($80 – $72),
also an acceptable outcome if it appeared that the downtrend has resumed.
Considering that the put had cost nothing (the collar generated a net $18 credit),
exercising the put was acceptable. In comparison, just opening an insurance put
at a net cost of $135 would mean that the exercise price would not break even
with a strike of 80 until price had fallen to $78.65 ($80 – $1.35).
The collar works as a hedge in these conditions. As it turned out, the stock
price, as shown on the chart in Figure 10.1, declined to $78 per share before
returning to $80 by November 18. Considering that the collar at this point had
another full month to run before the December expiration, no action would need
to be taken at this point. As time moved forward, the short call would lose time
value and could be closed to take profits and rolled forward to a later expiration
as a covered call, or just allowed to expire worthless.

Synthetic Stock Positions


The collar, especially if used for appreciated stock, is an effective hedge for
protecting paper profits and for setting up a desirable condition: willingness to
have stock called away in exchange for limitation of loss. A similar benefit can
be gained with the synthetic stock strategy. This is a position set up with options
and that mirrors price movement in the underlying stock on a point-for-point
basis, but for much less cost. This high degree of leverage is effective as a hedge
because maximum loss is limited because the net cost of a synthetic stock
position is at or near zero; it may also be set up as a small credit.
In comparing synthetic stock as a standalone strategy to stock ownership, the
benefits come with short risks. However, when you already own the underlying
stock, the synthetic stock position is very low-risk, and it works in many
situations, including at the time stock is purchased, when it has appreciated, or
when its price has declined.
In a situation when the price has declined, synthetic stock works as a recovery
hedge. This is preferable to what some investors prefer to do: buying more
shares. When you “double down” on a depreciated stock, further decline
accelerates the loss. In comparison, a synthetic stock position combining a long
call and a short put limits potential losses to the cost of the options, and price
decline can be managed by closing the short put or rolling it forward.

Key Point
Synthetic stock positions are conservative hedges because the short option
is manageable as part of the strategy.

When price has appreciated, the synthetic stock adds a similar version of
protection to that of the collar. It combines a short call with a long put. The long
put caps potential downside losses, preferably at the current price of the
underlying when the synthetic stock position is opened. The short call could be
exercised if stock price rises, but this side of the strategy is a covered call, so for
appreciated stock, this status is very similar to the collar.
There are two forms of synthetic stock: long and short. In the long version,
you combine a long call and a short put, setting up the uncovered put with the
same market risk as the covered call. In the short version, you combine a long
put with a short call. The short call is covered because you own stock, and the
long put provides no-cost or low-cost downside protection.
Because synthetic stock positions of both types do not increase portfolio risks
but offer profit potential and loss limitation, they are as close as you can get to a
perfect hedge. Under most popular definitions, this “perfect” hedge contains
100% inverse correlation between the risk on one side and profit on the other.
This is considered difficult to find; most analysts consider positions like covered
calls to be as close as possible to “perfect.” Considering the range of risks, this is
not an accurate definition; however, for synthetic stock in many situations
(appreciated, depreciated, or no-change stock), the synthetic hedge is one way to
manage portfolio risks.
The first of two strategies is called the synthetic long stock strategy. This
combines a long call with a short put. The short put is uncovered, but contains
the same market risk as a covered call, so it is considered a conservative method
for duplicating underlying price movement.
The synthetic long stock strategy is most effective when the underlying price
rises. So after a large price decline in positions held in your portfolio, this
strategy is a low-risk method for setting up recovery. Compared to only opening
a short put or a long call by itself, the synthetic solves many problems. The short
put provides recovery at a maximum equal to its premium value; this does not
negate the value of the short put, but it points out a limitation to this strategy.
The long call by itself is also limited. If the stock price does rise, it can become
profitable; however, time decay is a constant detriment to the potential for profit.
So the long call has to exceed not only the number of points it cost to open, it
also has to overcome and exceed lost time value premium.
The synthetic long stock strategy combines the best features of the long call
with the short put, but without the problems these two sides have by themselves.
For example, on November 6, Microsoft closed at $54.92. A synthetic long stock
position could have been opened using a December 85 call with an ask of 1.78
and total cost of $187; and a December 85 short put with a bid of 2.94, netting
$285. The net credit for this position is $98. The position benefits the most if the
underlying price rises; the combined positions will mirror stock price movement
in both directions, as shown in Table 10.1.

Table 10.1
The net difference of $8 at each price for the stock represents the difference
between the synthetic position’s strike and the share price ($55 – $54.92).

Key Point
When a synthetic stock position yields a net credit, it has an added layer of
hedging value.

The synthetic stock duplicated the movement of the underlying point for point
in both directions. This is among the most conservative of strategies. The long
call did not cost anything; in fact, the overall position was yielding a net credit of
$98. The short put paid for the call. In the event of a price decline in the stock,
the short put can be closed or rolled forward to avoid exercise. An interesting
alternative would be to accept exercise. Even though the exercise price would be
higher than the current value per share, if you believe this company is a
worthwhile one to hold in your portfolio, you might be willing to acquire another
100 shares at the higher strike. However, in a majority of instances, it makes
more sense to avoid exercise through closing of the put or rolling it forward.
Once that is accomplished, if you still want to buy more shares, buy at the
current market value, which is lower than the synthetic strike. This reduces the
overall basis in shares and presents an opportunity to employ hedges on the new,
higher-share position.
In comparison to the synthetic long stock, the opposite synthetic short stock
combines a long put with a short call. This is another very safe strategy because
it combines a low-cost or no-cost long put with a covered call (assuming you
own shares in your portfolio and are using the synthetic position as a hedge).
The synthetic short stock is designed to gain maximum benefit if the
underlying price falls. As a result, the timing for this position is best when stock
has advanced. The hedging benefit here is in the long put, which will gain value
as the underlying price declines. If the underlying advances, the covered call will
be exercised unless it is closed or rolled forward.
Using Microsoft once again as an example, a synthetic short stock position
can be opened with a December 55 long put. On November 6, 2015, ask price
was 1.64, total cost $173; and a December 55 short call bid at 1.17, net of $108.
This is set up to generate a net debit of $65. In the event of price decline, the
short call will expire worthless and the long put will gain 1 point for each point
lost in the stock. In the event of a price advance, the short call can be rolled or
closed.
Table 10.2 summarizes the synthetic short stock based on the Microsoft 55
strike.

Table 10.2

Key Point
Synthetic stock positions combine a certainty of profit in one direction,
with manageable offsets in the other.
The profit and loss in this case are opposite that of the synthetic long. As the
stock price falls, the loss is offset by gains in the long put. Remembering that the
options netted a debit of $65, the overall downside risk is limited to only $65 no
matter how far the underlying price falls. If the stock price advances, the profit
in the stock is offset by a loss in the covered call. The short call can be closed or
rolled to avoid exercise, which is desirable when the stock price is moving
upward.
The synthetic long and short stock positions are both advantageous, and short
side risks are effectively managed through cover or the ability to roll forward.
However, there is more to this strategy. A truly powerful hedge is created to
benefit regardless of the direction of underlying price movement. By “boxing”
the long and short synthetic stock, you set up a larger net credit. Opening both
the long and short at the same time accomplishes this. Based on the two previous
examples, a boxed synthetic stock would protect the stock position with an
overall net effect of zero in both directions. This occurs because long and short
positions are completely offset against one another.
The short sides are also easily managed. The short put on the long synthetic
can be closed or rolled and has the same market risk as a covered call. The short
call on the short synthetic side is a covered call. It can be allowed to exercise, or
it can be closed or rolled to avoid exercise when the underlying has moved
higher than the strike.
All of these synthetic hedges—long, short, and boxed—will work in
protecting portfolio equity positions. Focusing on the use of long synthetics after
a price decline and short synthetics after a price advance, is a sensible and
conservative hedging method. The alternative of speculating in long options or
short options contains much greater risks, either for time decay or exercise. The
chart of Microsoft, used in the synthetic examples, is shown in Figure 10.2.

Figure 10.2: Synthetic Stock Timing


If shares had been purchased as indicated, at a price of $44, the large gap into
price levels 10 points higher would be troubling. The temptation here would be
to sell shares and take profits right after the gap occurred, anticipating a retreat.
However, if you want to keep shares as long-term equity positions, selling is not
a solution. Buying puts at this level is equally speculative, because the put’s
value declines as time passes even if the underlying price does not change. A
covered call is a reasonable strategy in this situation. However, a synthetic short
(long put and short call) gives you the best benefits of the covered call along
with the downside protection of the long put. And the cost is close to zero.

Key Point
Timing of the synthetic stock strategy for conditions in the underlying,
identifies whether to focus on bullish or bearish movement.

Remembering that the purpose in hedging is not to generate income but to


protect portfolio positions, the offset of covered call profits with the cost of a
long put should not be an issue. The timing for opening of the synthetic position
is based on Figure 10.2 partly on the large price jump in late October, and partly
on the level of appreciation since shares were purchased in mid-September.
The next chapter moves into a new form of hedging: the straddle. This is the
purchase or sale of calls and puts with the same strike and expiration.
11 Straddles Hedged

Hedges apply in a vast number of flexible strategies. Among these, some of the
most intriguing fall into the realm of the straddle, a strategy most often
described as highly speculative. When properly constructed, the straddle
provides a conservative hedge for portfolio positions.

Key Point
The straddle can be designed not just as a set of speculative offsetting
positions, but as the more effective hedge attained by combining two low-
risk positions.

This hedge may be long or short, and can also be set up with adjustments in
terms of both strike and expiration. The straddle, as a flexible strategy, can be
configured as a highly speculative trade or as an appropriately conservative
hedge. Just as spreads are intended primarily to manage portfolio risks rather
than to take chances on price movement, straddles are also most appropriate
when used as part of a long-term plan to reduce market risks in equity positions.

The Long Straddle


The long straddle (also called a “buy straddle”) consists of a long call and a
long put, opened at the same strike and expiration. Most investors employing
long straddles select strikes as close as possible to the money.
The problem with the long straddle is the same as that for any long position:
time value. The closer the expiration date, the more rapidly time value declines.
In order to make a profit on the long straddle, the price movement of the
underlying has to be greater than the points spent to open the position.
For example, on November 6, 2015, Exxon Mobil’s close was $84.47. At the
same time, a long straddle using December contracts could be set up using:

1 long December 85 call @ $187


1.78
1 long December 85 put @ $324
3.15

The total cost to open this straddle is $511 ($187 + $324). This means the
underlying price has to move 5.11 points in either direction just to break even.
Price has to move beyond this range to create a profit. This is illustrated in
Figure 11.1.

Figure 11.1: Long Straddle Profit and Loss Zones, Exxon Mobil
This is a daunting challenge. Assuming time value disappears completely by
the time of expiration, the profitability of a long straddle relies solely on intrinsic
value. So as a hedge, the long straddle has little or no value. The risks derive
from two sources. First, time value is going to decline over the life of the long
straddle; this means that even with price movement in the underlying stock, it
takes a lot of change just to break even. Second, prices rarely move in one
direction for long, but rather tend to move both up and down in turn. So time
works against the long straddle in this regard as well.
In looking at the long straddle on paper, these realities are easily overlooked.
However, they make the long straddle an unlikely candidate for portfolio
management. You would be better off to buy a long call at the bottom of the
price swing, and to buy a long put at the top. This involves two long options, but
the danger in the long straddle is that by opening both at the same time, it
becomes nearly impossible to end up profitably.
The purpose of a hedge is to reduce risk, not to increase it. The long straddle
will not accomplish this because not only is it difficult to create a position
profitably, but it does nothing to protect equity positions. The strategy is
designed solely as a speculative, income-generating play with high risks. In the
example, the need for price to move 5.11 points in either direction sets up a loss
zone of 10.22 points.
Is the long straddle ever appropriate in managing a portfolio and hedging its
risks? Probably not. It is wise to know about the differences between long and
short straddles as a form of comparative analysis, but it is difficult to imagine a
scenario that calls for entering a long straddle to hedge a portfolio equity
position. The short straddle is a different story. It contains both elements of risk
and profit potential, but some forms are conservative and can help manage risks
effectively.

The Short Straddle


Compared to the long straddle, the short straddle has potential as a hedge.
Even so, the risks have to be considered. The short straddle can work as a
covered straddle (covered call and uncovered put), making it a conservative
strategy. This strategy was introduced in Chapter 7. Used when the underlying
price is range-bound in a consolidation trend or when the price has risen
significantly in a short time, the short straddle works as an effective hedge.

Key Point
Long positions like hedges are considered low risk because they do not
involve short options. Ironically, they present higher risks and no hedging
benefits.

The long straddle is not going to provide hedging value. In comparison, the
short straddle can work as an effective hedge, especially in low-volatility times
such as consolidation trends. The range-bound price levels provide opportunities
to exploit an otherwise frustrating lack of movement in the underlying. However,
because the short straddle is also a covered straddle (assuming the short call is
covered with equity positions), it is a very low-risk strategy.
In consolidation, looking for signals of a breakout in the form of Bollinger
squeeze or triangle and wedge formations, tells you when to think about closing
out the short straddle completely or, at the very least, closing the profitable side,
either call or put. In addition to the consolidation-specific breakout signals,
reversals (commonly associated with change in direction in either bullish or
bearish trends) can also be redefined as “reversing” consolidation into a dynamic
trend. In this application of a reversal signal, you require strong confirmation to
be convincing. So a breakout without signals is likely to retreat back into the
consolidation range and is one form of retracement. In comparison, once a
breakout is accompanied by a reversal signal and confirmation, it is time to close
the short position.
In thinking about how to manage equity positions in consolidation, the short
straddle makes sense. Consolidation is frustrating for investors, who expect to
see price movement. The lack of movement outside of a limited range is perhaps
even more frustrating than a declining price level. The danger during
consolidation is that once the price does break out, it will be likely to enter into a
dynamic range, possibly one with strong momentum. Knowing which direction
the price will move is essential if you are using a short straddle to generate
income in consolidation.
For example, on November 6, 2015, Exxon Mobil’s close was $84.47. At the
same time, a short straddle using December contracts could be set up using:

1 short December 85 call @ $153


1.62
1 short December 85 put @ $285
2.94

This sets up a total credit of $438, meaning that as long as price moves no
more than 4.38 points in either direction, this is a profitable hedge. Because it
extends in both directions, the total profit zone is 9.76 points wide. This
configuration is shown in Figure 11.2.

Figure 11.2: Short Straddle Profit and Loss Zones, Exxon Mobil
This outcome relies on subsequent price movement in the underlying. There
are two elements favoring this as a conservative trade. First is the safety buffer
of 4.38 points in either direction; recalling that price tends to move back and
forth, first in one direction and then in the other, this buffer provides a
considerable level of protection. Second, regardless of interim price movement,
time value is continually on the decline. So a short-term option straddle will
experience rapid time decay. In the example of XOM entered November 6, the
December contracts expire in about six weeks. So time value will fall rapidly
during this period.
In the case of Exxon Mobil, the chart reveals the long-term and more
immediate price action, shown in Figure 11.3.
Figure 11.3: Short Straddle, Exxon Mobil

On the six-month chart, the long-term consolidation range was identified


between $84.25 resistance and $78 support. At the beginning of the period, daily
breadth was quite small. This expanded as the price declined below support
during August and September. Even though price returned into consolidation
range, the daily breadth was broader than in June and July. This higher daily
breadth is a signal of growing volatility; so any short position is at increased risk
unless price settles into a narrower range. This did occur. The narrower range did
not reappear until the middle of November.
The timing of the November 6 short straddle coincided with the first move
above resistance since mid-June. Even so, the position works as a conservative
hedge and even the in-the-money put can be salvaged. Remembering that time
value works in favor of open short positions, it would be very difficult for any
option to gain much value before expiration.

Key Point
Rolling short straddles to avoid exercise of one side may require replacing
the straddle with a later-expiring short spread.
As of the close of November 20, the December 85 call had declined and could
be bought to close for an ask price of 0.54, or $63. This represents a profit of $90
based on the November 6 original bid price. The put is a different matter; it has
increased in value to an ask of 5.80, so it will cost $589 to buy to close.
Combined, closing both of these positions creates a net purchase price of $499
($589 – $90).
To close and roll these into positions closer to current value of $79.79 as of
November 20, they can be replaced with a January spread. This consists of a
January 80 call bid at 2.95, net $286; and a January 82.50 put bid at 4.40, net
$441. The total credit would be $727. The overall position’s net basis is $666:
$438 original credit – $499 buy to close + $727 net positions = $666
This recovery strategy takes the put’s strike down 2.5 points and extends the
period of exposure another month. However, based on long-term trends in the
stock, it appeared that the consolidation trend had reestablished itself. If this
turned out to not be the case, the newly set short spread using an 80 call and
82.50 put can be rolled once again to new strikes, either higher or lower. Does it
make sense to roll a short straddle? Given that the idea here is to generate
income during consolidation as a form of hedging (in a situation that is not
yielding any growth through the underlying), does rolling truly hedge risk? The
problem with a forward roll is found in the extension of exposure time. If
consolidation ends suddenly and the price begins moving with strong
momentum, what happens to the rolled short options? This situation presents
risks in place of hedge advantage. With this in mind, it could make more sense to
roll forward to a weekly set of options to reduce exposure time while still
generating a net credit.
Even in a recovery strategy, there comes a point when closing makes more
sense than extending exposure time. A flaw every options trader faces is the
temptation to keep positions open until profits develop, even if this means
escalating risks. It is more conservative to acknowledge the poor timing of a
trade and take the loss.
This recovery strategy is likely to work, however, due to the price buffer and
rapidly falling time value. Even conservative strategies can fail; and the short
straddle opened during consolidation is certainly more conservative than one
opened during a dynamic and volatile trend. This set of events, rolling one side
to the same strike and the other side to a different strike, is an example of how a
short straddle (later converted to a short spread) can be used to hedge a
consolidation trend; and how this can be accomplished with a net credit, even
after rolling out of a loss on the uncovered put.
If the price had risen, a similar strategy could be used to take profits on the
short out and roll the short call forward to a higher strike. When combined with a
new short put, the move creates a net credit in the opposite direction. The same
cautionary risk warning applies: If the stock price enters a new and extended
bullish trend, the rolled short call has to be rolled yet again to avoid exercise.
The strategy makes sense as long as consolidation continues. Once consolidation
concludes, it is time to figure out how to escape from the short positions. The
forward roll is only one possibility; another is simply taking a loss, hopefully a
small one. However, properly managed, the consolidation-based short straddle is
likely to yield profits more often than losses, especially when the time to
expiration is limited and when you keep an eye on evolving breakout signals.
The hedging value in rolling with a bullish breakout is significant. A covered
call cannot be rolled easily as long as the stock’s price has declined below the
strike. For example, Exxon Mobil’s covered call was opened at an 85 strike, but
the stock then fell to below $80 per share. In that case, there is little you can do
but wait out the price, hoping it will recover; or take the profit on the call and
then seek a new strategy, such as a short straddle. However, by starting out with
the short straddle, the outcome can be managed profitably whether the
underlying price advances or declines.

The Strap
A variation on the straddle is the strap. This is a position that adds a ratio
weighting to the call side, by opening two calls versus one put. Like the long
straddle, the long strap has no hedging value and is expensive, involving two
calls in the hope of a price advance, versus only one put protecting against a
price decline. However, the loss zone is increased by the number of points in the
long calls.
Recalling the example of the long straddle, the December 85 Exxon Mobil
options could be used to set up a long trap with the following:

2 long December 85 calls @ 1.78 $356


+
$10
=
$366
1 long December 85 put @ 3.15 $315
+ $ 9
=
$324
Total debit $690

To become profitable, the underlying price needs to move 6.9 points in either
direction, between $91.90 and $78.10. The full loss zone is extended to 13.8
points. Above the higher breakeven of $91.90, the long calls gain 2 points for
each point of movement in the underlying; and for each point below $78.10, the
long put gains 1 point.
In the Exxon example, even the decline from November 6 down to the
November 20 closing price of $79.79 failed to move lower than the breakeven of
$78.10. The long strap is very difficult as a speculative strategy, and it offers no
hedging value.
In comparison, the short strap can work to hedge the underlying, especially
in a time of low volatility or during a consolidation trend. For example, using the
short options as of November 6, a short strap could be set up with the following
positions:

2 short December 85 call @ 1.62 $324


-$10
=
$314
1 short December 85 put @ 2.94 $294
- $ 9
=
$285
Total credit $599

Key Point
The short strap represents a low-risk hedge, assuming the higher number of
short calls is completely covered with shares of stock.

Given the subsequent decline in the underlying between November 6 and


November 20, the profit on the short calls would be substantially higher. The
same rolling technique used for the short straddle could then be used for the
short strap. To work as a truly conservative move, the short calls should be fully
covered, meaning ownership of 200 shares for two short calls.
The Strip
The strap’s opposite cousin is called the strip. This is a variation of the
straddle in which more puts are opened than calls.
The long strip, like the long strap, is highly speculative and offers no hedging
benefits. Based on the original long straddle, the long strip would consist of:

1 long December 85 call @ 1.78 $178


+ $
9 =
$187
2 long December 85 puts @ 3.15 $630
+
$10
=
$640
Total debit $827

In order to become profitable, the underlying price would need to move 8.27
points in either direction. So the loss zone extends from $93.27 down to $76.43,
a total loss zone of 16.54 points.
The long strip does not present a favorable outcome based on the high cost
and required movement required just to break even. In comparison, the short
strip does have hedging benefits.
Based on the original short straddle for Exxon Mobil, the short strip could be
constructed with the following positions:

1 short December 85 call @ 1.62 $162


- $ 9
=
$153
2 short December 85 puts @ 2.94 $588
-
$10
=
$578
Total credit $731
In this example, the profit zone extends 7.31 points above and below the
strike of 85, between $92.31 and $77.69, a zone of 14.62 points. In the Exxon
example, a problem arises because the stock price fell from the $85 range down
below $80. To roll this forward, you would need to create a new straddle, spread,
or strip designed to replace the appreciated short puts with new ones at a net
credit.
The short call could be closed on November 20 at a net profit of $90.
However, the short puts would have grown to 5.80 each, so a buy to close
requires $1,169 ($580 + $580 + $9). These positions could be replaced with a
single 80 call bid at 2.95 and two 82.50 short puts bid at 4.40 each. The total
credit would be:

One 80 call @ $295 $295 –


$ 9 =
$ 286
Two 82.50 puts @ $440 $880 –
$10 =
$ 870
Total credit $1,156

Key Point
The roll that creates a net debit is potentially a problem for short option
positions. It makes sense when an ITM strike is replaced with a later one
that is closer to the money. However, this type of roll has to be entered
cautiously.

The buy to close came out to $1,169 and the two extended short strip yielded
$1,156, a net reduction of $13. This by itself is not a problem, because for a cost
of $13 net, the in-the-money puts are replaced with new short puts and a lower
strike. This means all of the new positions will benefit from time decay until the
January expiration.
The recovery of the original short strip points out a problem in hedging with
short options. This position is recovered and, at the very least, expiration is
deferred. This process of rolling forward can be continued indefinitely, but it
points out the potential risks of this type of hedge. Because of the added shorts
options on one side or the other, when the underlying price moves against the
higher ratio (two options versus one), the replacement and recovery are made
more difficult. This is due to the higher premium value for twice as many
options. Given the case of Exxon Mobil and its long-term consolidation trend,
this hedge is not out of the question, but it does contain greater risk than the
more basic short straddle.

Calendar Straddles
Another variety of the straddle is the calendar straddle. This involves
opening two straddles on the same underlying: a short straddle expiring first and
a long straddle expiring later. Like the long straddle, this position has only
limited hedging value. The short position provides short-term hedging value, but
once this is expired, the long-term long straddle position is strictly speculative
and requires substantial price movement.
The overall speculative nature of the calendar straddle is made more troubling
by the fact that the overall position will yield a net debit. The later-expiring long
straddle contains more time value, so this position is a speculative one, making it
difficult to produce a profit. For that to occur, you rely on short-term low
volatility followed by longer-term high volatility. This is impossible to predict.
As an expanded trade, the short-term short calendar straddle can be combined
with a long-term long spread with a higher call strike and a lower put strike. For
example, the December 85 short straddle could be followed by a January long
spread using a 90 long call and an 80 long put. However, these provide no
hedging value. The long call and put provide some limited protection against
exercise of the earlier short positions; however, if you own shares, the short call
is covered and the short put is easily managed. So the calendar combined
straddle and spread makes no sense compared to the short spread by itself.
The more elaborate forms of straddle may present greater risk and less
hedging benefit. One risk of all options strategies is in the attraction of what
appears to be an elegant strategy. In fact, the simple and more basic hedges are
invariably more effective in managing risk than the more complex ones. In either
case, the need to roll forward as an exercise avoidance technique is unavoidable.
The next chapter explores methods of rolling and loss recovery.
12 Rolling and Recovery Strategies

No matter how cautiously you set up options strategies to hedge equities, some
portion are going to lose rather than gain. In these instances, you will need to
figure out how to manage the paper loss. This can involve just taking the loss,
waiting out the market, or applying a series of rolling and recovery strategies.
A danger in any form of recovery is the potential increased risk. Most forms
of recovery involve replacement strategies, and this can easily be translated into
an exchange of a low risk for a high one, or an exchange between a conservative
strategy and a risky one. Because the purpose of the option hedge is to reduce
risks, one of your primary considerations should be risk awareness as a basic
requirement for portfolio management and for hedging.
With risk awareness as a first requisite for managing equity positions,
especially when they have declined in value, you are better positioned to
maintain a conservative risk profile while still devising strategies to continue
hedging against loss. When options are involved, the first form of management
over paper losses is the forward roll.

Rolling Strategies
Previous chapters have explained rolling forward in various ways, but the
context of the forward roll with risk levels in mind has not been explored
previously. As a form of recovery from a position that has worked against you,
the forward roll contains three key attributes:
1. Exercise avoidance. The most apparent reason for rolling forward is to
avoid exercise. This is also the most legitimate reason for rolling.
Assuming your purpose is to maintain your equity portfolio, exercise
presents a problem: A called-away position has to be replaced. If you
prefer keeping well-selected equities, exercise avoidance makes sense.
Key Point
Avoiding exercise is a sound reason to roll forward. However, the outcome
has to be analyzed to ensure that the overall benefits outweigh added
exposure.

Even so, a few observations have to be made as well. Exercise


avoidance is not itself a hedge against market risk, but a hedge against
exercise. There may come a point where accepting exercise makes more
sense than rolling forward. For example, if the call has moved far in the
money, rolling will yield a small credit and it only delays inevitable
exercise. A covered call may also be opened with exercise in mind. It is
one way to set up an excellent hedging choice. Either the short call expires
worthless (or is rolled), or shares are called away at a profit (and that profit
consists of a capital gain as well as option premium).

Key Point
The roll creates a net credit due to higher time value. However, the added
premium is justified only when the added time of exposure also makes
sense.

2. Creation of a net credit. The forward roll will invariably create a net
credit as long as the new short position has the same strike. This occurs
due to a later-expiring short position containing more time value. As a
hedge against exercise, a forward roll or even a series of forward rolls
generates net income. If enough credit is available to increase the roll to a
higher strike (for short calls) or a lower strike (for short puts), the forward
roll gains hedging value. In the case of a short call, it makes exercise less
likely, or, if exercise does occur, it will be at a higher price per share. The
short put rolling down avoids exercise or sets up a lower cost per share if
the put ends up exercised against you.

Key Point
Increasing exposure to loss with the forward roll is ill-advised, because the
potential for recapturing lost value is offset by an equal risk of doubling the
amount of the loss.

3. Potentially higher risks. Finally, the aspect of forward rolls often ignored
or overlooked is the potential for higher market risks. This occurs
especially when you replace a one-to-one covered call or short put with a
greater number of new contracts, setting up a ratio write or just more
market exposure. This is likely to occur when the option has moved in the
money and you want to set up a net credit while also changing the strike.
So for example, a covered call has moved in the money. So it is bought to
close and replaced with a greater number of higher-strike, later-expiring
calls. The rationale is that with a greater number of contracts, the net is
likely to remain a credit, while eventual exercise is avoided or set to occur
at a higher price per share. However, by converting a covered call to a
ratio write, the risk level is also increased.
This can be mitigated to a degree by setting up a variable ratio write. For
example, with 300 shares and three current covered calls with 40 strikes, you
buy to close and replace with later-expiring sets of two 41 and two 42 strike
calls. This creates higher risk, but remains manageable. However, if your
primary purpose in originally opening the covered calls was to hedge market
risks while earning premium income, what happens with the forward roll and
conversion? The hedge is replaced with a more speculative position because a
portion of the new set of short calls is uncovered. The variable ratio write is a
worthwhile strategy, but moves such as this are most effective when used for
situations with one-point increments between strikes. If the increments are 2.5,
5, or 10 points, the roll forward and conversion is less practical.
The same observation applies to uncovered puts. The market risk of the short
put is the same as that of the covered call. However, if the desire is to roll
forward to a lower strike, it is tempting to also increase the number of contracts.
This also increases the risk that becomes real if and when the short puts are
exercised. This will occur only when the current price per share is lower than the
put strike; so with a greater number of short puts open, the risk is higher as well.
Like the roll of a covered call, the roll of an uncovered put that also includes
increasing the number of contracts replaces the initial hedge with a higher-risk
position.

Is the Forward Roll Worthwhile?


The initial trade that sets up a hedge may effectively avoid losses on the
equity side of your portfolio. However, the forward roll is not part of that hedge,
but a defensive strategy to avoid exercise of a short option. Although exercise
avoidance is desirable, is the forward roll worth the effort? There may be times
when exercise is acceptable and even desirable.

Key Point
When your basis in stock is far below current market value, you gain
greater hedging potential and profits.

For example, if your basis in stock is far below the current market value,
opening a covered call may produce two desirable results. First, it expires
worthless and yields current income, allowing you to repeat the trade. This
hedges the current value of stock to the extent of option premium. A second
outcome is exercise, which combines three sources of profits: capital gain on the
stock, option premium, and dividends.
The forward roll might not have a part in one form of trade: the deep in the
money covered call. For example, you buy 100 shares of stock at $30 per share
and the price has risen to $42. With the possibility of a $1,200 profit, you would
be content to sell but you also see value in keeping those shares. One way to use
these shares to generate income and capital gains is to write an in-the-money
call. For example, a 35 strike yields immediate intrinsic value of $700, a 23%
profit on the original $30 per share investment. If the stock price declines, the
short call hedges market risk. It can be bought to close at a lower priced than the
original sale, so that the profit in the covered call offsets the loss in the stock. As
an alternative, you can wait out the expiration cycle with one of two possible
outcomes. First is exercise and having stock called away at a profit (capital gain
and option premium). Second, if the option premium has declined by the time of
expiration, the short call can be closed at a profit.
A third alternative in this situation would be to roll forward. Even though this
was not part of the original plan, this situation sets up a powerful hedge. The
original deep in-the-money call protects against downside moves in the stock
(acting much like a long put), and the potential of profits from significantly
reduced time value provide added current income. By rolling forward, this
experience can be repeated.
A fourth alternative occurs if and when the stock price falls below the call’s
strike. In this case, the stock loses market value but it is offset by the short call.
It is conceivable in this scenario to alter the hedge by rolling forward and down
to a lower strike (but a strike that is higher than original basis in stock). Or if
expiration is close, the call will expire worthless. In this case, you created a
hedge while earning a rich option premium.
The point here is that appreciated stock contains great potential for hedging.
The covered call, employed as a contingent sale transaction, sets up current
income in double digits, and all outcomes will be profitable while hedging
market risk in the stock.

Types of Forward Rolls

Key Point
The forward roll is not a risk hedge, but an exercise hedge.

The roll forward to the sale strike is the best-known type of roll, and the one
that is likely to be employed in a majority of instances. However, all forms of
forward roll have to be considered as defensive adjustments, in order to avoid
exercise.
Beyond the strictly horizontal forward roll, a diagonal roll is also possible.
When a covered call is involved, avoiding exercise may also involve rolling to a
higher strike. By rolling up, the current in-the-money or at-the-money strike is
increased. This reduces the chance of exercise, and, if exercise does occur in the
future, you will receive a higher capital gain based on the higher strike.
When rolling an uncovered put, the opposite diagonal direction is possible.
The advantage of the short put over the covered call is flexibility. Based on
premium values, you can roll forward to any strike you wish, without concern
for having shares of stock called away. The most desirable roll moves to a lower
strike, rolling down, while still creating a net credit.
This usually involves extending the position out further than desired in order
to set up a net credit. Greater flexibility is going to be found in stock with 1-
point increments between strikes. As a strategy to avoid exercise, any roll that
also changes the strike in a profitable direction is a sensible combination of a
recovery strategy with an exercise hedge.
Rolling may become necessary whenever profitable hedge positions have
been partially closed, but remaining open positions are in paper loss status.
These “orphan” positions should be rolled forward to strikes closer to the money,
creating an overall credit. In rolling, the new position’s basis has to be adjusted
to absorb the loss on the original position.

Other Recovery Strategies


Any recovery strategy should be designed to recapture a paper loss,
preferably without increasing risks. Whenever such a strategy replaces a
conservative hedge with a higher-risk speculative trade, it violates this concept.
The range of strategies is vast, and recovery itself is part of portfolio
management. However, it often is wiser to accept losses and move to the next
trade. The quest for a perfect record of profits invariably leads to greater losses.
So a well-designed recovery trade should be made with hedging in mind.
Because the idea is to hedge risk, a recovery trade may consist of a new current
trade or of a reconstituted hedge, which employs later-expiring positions to
exploit time value without adding market risk.
Some examples of recovery strategies that conform to the hedging standard
without added risk are:

Key Point
Replacement strategies should be made carefully to ensure that risk levels
are not increased.

1. Replacement strategy. In this trade, a current loss is replaced with the


same strategy or a similar strategy but with later expirations. This may
constitute a forward roll or replacement of one type of trade with another.
When using this method, the relative levels of risk should be kept in mind.
The most desirable outcome is to not increase market risks.
An example of a replacement strategy is to accept a loss on a covered
call by closing the position. This avoids exercise. It is then replaced with
an uncovered put with an adjusted strike. The advantage in this is twofold.
First, it offsets the loss on the covered call. Second, it can be set up using
any strike regardless of your basis in stock, but with the same risk levels as
those of the covered call.

Key Point
Losses on an initial hedge may create opportunities for stronger hedges and
higher net income.
2. Alternative strategy with higher credit premium. This involves
accepting a loss on an initial strategy and replacing it with a different
strategy whose market risk is the same or lower. This allows you to keep
portfolio management intact without loss.
An example: You opened a covered call and the stock price has moved
above the strike. With exercise approaching, you want to avoid exercise so
you close the short call at a loss. You then open a later-expiring, higher-
strike covered call that is out of the money and at the same time sell an
uncovered put with the same strike. This creates a covered straddle. The
risk is no higher than a covered call, and creation of higher premium
income further discounts your basis. At the same time, by using a higher
strike, you offset the loss on the original covered call while creating a
higher capital gain in the event of exercise.
3. Increased equity with new hedge. This involves increasing the position in
stock after a decline in market value. This discounts your basis in all
shares and is also called averaging down. It is not advisable when the
company’s value is declining, meaning the drop in the stock price reflects
a fundamental problem. However, when the downward swing is cyclical,
this is an effective recovery strategy that also introduces added hedging
opportunities.

Key Point
Adding more equity positions as part of a recovery strategy has to be done
cautiously, and only when your view of the company justifies adding to
your equity holdings.

For example, you purchased 100 shares at $38 and sold a covered call
at a 40 strike. The stock price has declined to $35 per share. Your covered
call expires worthless. However, the paper loss of $300 creates a problem.
Analysis of the company reveals continued fundamental strength, so you
do not want to dispose of shares. You decide to increase your equity
position by purchasing an additional 200 shares @ $35. This changes your
average basis to $36 per share. You next sell three 35-strike covered calls
@ 3 ($300). Even though the strike is one point below your adjusted net
basis of $36 per share, the covered call premium exceeds this. The one
point in the money is acceptable under these conditions, as the original 3-
point paper loss is entirely absorbed by this trade.
Some trades intended as recovery strategies actually increase your risks and
should be avoided. In the interest of maintaining a conservative profile,
increasing risks is not a wise move. By “doubling down” on a loss, you could
break even or just make that loss much worse. This is a speculative hedge and is
not a conservative trade. Too often, options traders with equity positions are not
willing to take relatively small losses. This is an error, and realistically, you will
have losses some of the time. The purpose in hedging and recovery is to
minimize losses without adding risks.
For example, after opening a covered call for 4 ($400), the underlying price
rises substantially and the call moves in the money, increasing to a valuation of 9
($900). You buy to close and take a loss of $500. Your rationale is that the
underlying price has risen so the loss on call premium is justified or, at the very
least, offset. However, you want to replace the lost value in the short call.
Believing that the underlying will continue to rise, you buy two at-the-money
calls expiring in two months, paying premium of 4 on each, or a total of $800.
This increases your overall net loss to $1,300. Contrary to your belief (and
speculative hope), the underlying price remains below the strike of the new calls,
and you end up losing. In this case, the loss was made far worse. A more prudent
move would have been to accept the $500 loss and sell a later-expiring, higher-
strike covered call. This would generate immediate income and set up a higher
strike, meaning you would earn a higher capital gain if the new short call were
exercised.

Key Point
Increasing risks is a poor way to respond to a paper loss. It is not a recovery
strategy, but an exchange for conservative policies for high-risk positions.

The lesson in this situation is that speculative hedging is a high-risk move. It


can work out, but more often it only ends up making losses worse. Hedging does
not guarantee net profits in every case. However, accepting losses and figuring
out how to offset them with more hedging is a mature and sensible method for
portfolio management. In the example, the attempt to offset losses did not work
out; but by duplicating the conservative hedge (covered call), the loss would
have represented only one part of the overall experience, with a likely increase in
profits from later decisions.
In addition to working as a way to reduce portfolio risks, well-designed
hedges accomplish a second benefit: reducing the degree of losses in the
portfolio. When hedges are timed to coincide with the cyclical moves in equity
positions, they are also more likely to succeed. So timing a covered call when
the underlying is at or near the top of its trading range is one hedging method
relying on proximity of price to resistance. Timing an uncovered put to coincide
with price at or near the bottom of the trading range sets up an identical market
risk but with support proximity ruling the decision.
The covered straddle also works well with proximity in mind, as a form of
hedging. As a recovery strategy, the covered straddle increases current income
without increasing market risk. For many investors, when stock prices move into
consolidation, it is considered a form of “loss” if only because the stock price is
not moving in a dynamic fashion. In this case, consolidation could last several
weeks or months. To recover from this inactivity in the equity value, hedges like
the covered straddle exploit the lack of movement by generating option premium
income. As long as you monitor the position to identify likely breakout points, a
short-term covered straddle can generate attractive levels of income.

Anticipating Price Decline With No-cost Hedging


A second benefit to using short options to hedge equity: This enables you to
keep shares in your portfolio. Assuming you want to keep your equity positions
intact, a dilemma arises when the stock price rises: Do you sell shares and take
profits, or do you wait out the cycles? Another alternative is to anticipate
potential price decline and hedge against it. In this way, you recover a possible
loss of equity value in advance. Positions like the synthetic short stock set up
insurance for little net cost, and the short call pays for the long put. The risk is
minimal in this position because the call is covered and can be closed or rolled to
avoid exercise if the equity price moves that call in the money. If the stock price
does fall, the long put side of the synthetic position offsets losses in the stock
with gains in the put.

Key Point
Synthetic stock positions create hedges not in response to a price move, but
in anticipation of it.

If you just buy an insurance put, you have to consider the cost of the put as
part of the hedge. This means a breakeven does not occur until the stock price
declines a number of points equal to the cost of the put. However, with the
synthetic short stock trade, the net cost of the long put and short call is at or
close to zero; so the insurance benefits of the put are based on a zero cost; the
offset begins as soon as the underlying falls below the strike.

Recovery With Multiple Share Increments


The process of loss recovery is more flexible when an equity position is
higher than 100 shares. With only 100 shares, you are limited to single contracts
for one-to-one positions such as covered calls or covered straddles.
The limitation does not apply to some hedges such as synthetic stock. For
example, a synthetic long stock position combines long calls with short puts. Net
cost is at or close to zero. This eliminates the market risk of the long calls, which
consists of time decay risk. The short put risk is identical to the risk of covered
calls; so in theory, a hedge could consist of any number of offsetting calls and
puts in this synthetic position. However, realistically, the hedging properties are
more important than the theory. A synthetic long stock position duplicates
movement in the underlying even if you do not own stock. This means that the
synthetic is not a hedge, because there is no equity position to offset. However,
when you own shares and the price has declined, the synthetic long stock is an
effective hedging device that exploits expected price swings.

Key Point
When you own several increments of shares, hedging is more flexible and
many opportunities are presented.

This benefit is augmented when you own multiple increments of the


underlying. For example, if you own 500 shares, how do you hedge the position
when the underlying price has dropped? For example, your basis was $47 per
share and today you sold five covered calls with 50 strikes, gaining 2 ($200) for
each, or a total of $1,000. Today, the price has dropped to $44 per share. The five
covered calls expired worthless, and the net outcome is an adjusted net basis of
$45 per share ($47 paid minus 2 points for covered calls). The profit of $1,000
on covered calls is desirable, and current value is only one point below your net
basis.
One way to hedge this relatively small net paper loss would be to sell more
covered calls. However, the timing does not make this ideal, and you believe the
stock price has a good chance to recover. So in place of covered calls, uncovered
puts are more attractive. However, concern about the management of this hedge
if the underlying price continues to fall makes the short put questionable as a
safe hedge.
An alternative to both of these is to identify additional strategies you can use
to create hedges while recovering the loss as well. For example, a covered
straddle with a strike above current value (at $46 or $47, for example) generates
income while offsetting the entire loss. A synthetic long stock consisting of a
long call and a short put will create net income if the stock price rises, while
presenting manageable short put risks (which can be closed or rolled). These are
manageable because the strike is higher than current market value of stock, so
time decay should absorb enough of the current premium to make it possible to
close at a profit.
Even so, the exposure of five short puts can be daunting. With this in mind,
ownership of 500 shares presents an opportunity to combine hedges in a variety
of combined strategies. For example, you may want to sell two covered calls,
two covered straddles, and one synthetic long stock. This creates a combination
overall of four short calls, three short puts, and one long call:
2 covered calls = 2 short calls
2 covered straddles = 2 short calls and 2 short puts
1 synthetic long stock – 1 long call and 1 short put

Although this is a complex series of positions, it does present profit


opportunities no matter which direction the stock price moves. Out of the eight
positions, seven are short. This also means time decay provides exceptional
opportunity for profit.
Why would anyone want to set up such a complex series of trades against 500
shares? It does diversify the short risk, but for many, the alternative of staying
with a single strategy might be seen as more desirable. It is a matter of personal
choice. For example, you could just sell five uncovered puts with a strike above
current price per share. The risks of the short put are identical to covered calls;
however, a drop in share price translates to a risk of having an additional 500
shares put to you upon exercise. So a set of five uncovered puts, opened as a
hedge against an equity position of 500 shares, would have to be monitored with
this is mind. They can be closed to take profits, or rolled forward if the puts
move in the money. Otherwise, as long as the underlying price remains at or
above the strike, they will expire worthless.
With multiple increments of shares, the hedging potential is more flexible
than for sets of only 100 shares. In setting up a recovery strategy, this flexibility
also translates to more effective forms of hedging than for only 100 shares. In
one definition, the market risk of hedges against 100 shares is lower than the
equivalent risk for multiple increments. However, the risk is identical for any
number of shares and hedged options, with a notable difference: The income
from short positions is greater, but the dollar value of a loss is also greater, even
with identical risks.
This risk can be further diversified with calendar positions. This means
opening positions with several different expiration dates as a means for avoiding
the singular risk of several options expiring at the same time. For example, with
500 shares in your portfolio, you could sell uncovered puts in the following
combination:
2 puts slightly out of the money, expiring in under 30 days
2 puts one strikes higher, expiring in 60 days
1 put two strikes higher, expiring in 90 days

Key Point
The potential for varying hedges against multiple increments of stock adds
flexibility, but might not improve the overall hedge.

The higher strikes are further out of the money for uncovered puts, but the
longer expiration timing increases time value. The earliest expirations will have
rapid time decay, and the later ones will decline more slowly. However, the
extended period allows you to manage these positions effectively through
closing or rolling if and when the underlying price takes these puts in the money.
The most important question with this strategy is whether or not the varying of
hedge positions improves the hedging effectively, or not.
The same calendar position method can be used for any hedge, including
covered calls, covered straddles, or combinations of several different hedging
strategies. You could also set up a series of hedges to benefit whether the
underlying moves up or down. Consider the following set of five trades to hedge
500 shares:
1 uncovered put slightly in the money, expiring in under 30 days
1 covered call slightly out of the money, expiring in under 30 days
1 covered straddle one strikes higher, expiring in 60 days
1 synthetic long stock one strike higher, expiring in 90 days
1 synthetic long stock one strike lower, expiring in 90 days
In this set of five hedges over three upcoming expiration cycles, you combine
uncovered puts, covered calls, covered straddle, and synthetic long and synthetic
short stock. Because the two synthetic positions benefit with stock movement in
opposite direction, they both represent low-cost or no-cost hedges with
protection on the upside and on the downside. However, a related concern with
these combined hedges is timing of dividends. When ex-dividend date occurs, it
presents a potential time when short options will be exercised early. So a prudent
hedging policy is to avoid having short calls open during ex-dividend month.
As a matter of practical limits, extending hedges—especially with short
options—for such an extended period of time presents higher risks than those
expiring within 30 days. However, this diversification of hedges also presents
attractive variation to the selection of option hedges against equity positions in
your portfolio.
The next chapter extends the idea of recovery to a different topic, avoiding
early exercise on short positions. One of the problems of the varied hedge is the
potential for exercise of short positions. This is a management issue for the
portfolio as well as for the hedges you design to mitigate risk.
13 Avoiding Early Exercise of Short Options

A general rule of thumb for anyone opening short positions regards the
possibility of exercise: You should be willing to accept exercise as one of several
possible outcomes.
This does not mean that exercise is the best outcome, but it is a possibility.
When using options to hedge equity positions, exercise is not desirable. It
contradicts the basic reason for the hedge, which is to minimize or eliminate
equity market risk. For example, if a covered call leads to having shares called
away, it disrupts the portfolio and presents a problem: Do you get back into the
stock position or find a suitable replacement? It would be more in line with the
hedging goal to avoid exercise as long as that does not mean adding risks to the
hedge.

The Events Leading to Short Call Exercise


Most options traders are aware that exercise is automatic on the last trading
day, as long as an option is in the money, even minimally. Automatic exercise
occurs for all options that are one penny or more in the money by expiration.

Key Point
Automatic exercise is worth avoiding in most hedging strategies, and
exercise at the end of the cycle is the most common timing for this
outcome.

For anyone with an in-the-money option by the third Friday of the month, the
most rational action is buy to close, and either take the profit or loss, or roll
forward. If these actions are not taken, the option will be exercised. For a call,
100 shares of stock are automatically taken from you for each option; and for an
exercised put, 100 shares of the underlying stock are put to you at the strike.
Some covered call positions are designed as a form of contingent sale,
meaning exercise is acceptable and even expected. However, this is not a typical
hedge but a process for selling shares of the underlying at a profit. For the more
typical hedging position, avoiding last-day exercise is desirable. It should also be
profitable to close the short call or put, because time value has evaporated by
this point in the cycle. However, if intrinsic value has increased so that the
position would create a net loss, two choices remain: buy to close and accept the
loss, or roll the position forward to a later-expiring option to avoid exercise.
Beyond the last trading day, early exercise is also a possibility. An option can
be exercised at any time as long as it is in the money. Realistically, the chances
of early exercise are remote with one exception: the few days right before ex-
dividend date. Ex-dividend means “without dividend,” representing the end of
the period in which a dividend can be earned. So as long as you are a
stockholder of record by close of business the day before ex-dividend date, you
are entitled to the current dividend even if you sell stock on the following day.
The details of this event were described in Chapter 6.
As a short call seller, this timing is a likely period when early exercise will
occur. But what are the chances of this? Not every short call is exercised right
before ex-dividend date, because long option holders did not all pay the same
amount for the call. So in order for a call owner to put early exercise into effect,
three values are in play: the original cost of the long call, the current price of the
underlying, and the amount of the dividend.
It makes sense to exercise when the long call owner will be able to make a net
profit by calling away shares. For example, if the strike is 50 and current value
of stock is $53 per share, the owner can buy 100 shares from you at $50 and
realize a $300 profit (because the current price is $53). If the dividend is $35, the
total initial profit will be $335. This makes sense as an early exercise candidate
as long as the initial cost of the call was lower than $335. However, if that long
call owner paid 4.50 ($450) for the call, early exercise would not make any
sense.

Key Point
Early exercise is most likely to occur immediately before ex-dividend date.
However, not all in-the-money calls will be exercised in this situation.

This means that for in-the-money calls, not all will be exercised right before
ex-dividend date. Only those that yield an overall net profit are likely to be
exercised only. Even then, not every long call owner will decide to exercise.
Some will prefer to sell the call at a profit and not want to earn the dividend as
well. If they are uncertain about the investment value of shares, or simply don’t
want to take an equity position, they will not exercise the short call. The decision
will also rely on how many points in the money the call has moved.
The percentage of calls exercised early is not easy to determine. It depends
largely on the amount of capital gain and on the dividend. Because the capital
gain is different for every strike, the portion likely to be exercised early will
vary. The often-cited 75% of all options expiring worthless is wrong. In fact,
75% of all options held until expiration will expire worthless. Most are exercised
beforehand or closed before expiration occurs. Overall, only 17% of all options
are exercised according to the Options Clearing Corporation. Figure 13.1 shows
a breakdown of outcomes.

Figure 13.1: Option Outcomes

Source: Options Clearing Corporation (OCC)

The portion of all options that is exercised, 17%, represent activity occurring
mostly on the last trading day. Some of these exercised options were early, but
only a small portion of all options suffer this fate, even for in-the-money short
calls right before ex-dividend date.
Even so, as a short call seller, you should be aware that some options traders
use the early exercise provision as an intentional strategy. The idea is to buy at-
the-money options immediately before ex-dividend date and exercise them
immediately. This allows them to earn the current dividend. On or after ex-
dividend date, the shares received through early exercise are closed. So the
quarterly dividend is earned for a very short holding period of only a few days.
Early exercise of your short calls can be avoided in three days. First, focus
only on stocks that do not pay dividends (this strategy makes little sense because
dividends represent attractive forms of income). Second, avoid selling calls
expiring in ex-dividend month, which for most stocks occurs only in four
months per year. Third, monitor the situation closely and roll forward when calls
go in the money, to avoid early exercise.
When time value and extrinsic value (implied volatility) are relatively high
right before ex-dividend date, the chances of early exercise are reduced.
However, if the option’s premium is predominately intrinsic value, this greatly
increases your exposure to early exercise. One way to look at early exercise of a
covered call is that, even with the loss of the dividend, it is a positive outcome.
This is true as long as the option’s strike is higher than the basis in stock. Even
so, a hedging strategy should be based on the desirability of keeping portfolio
positions and benefitting from share price growth and dividend income.

Early Exercise and Short Puts


For short put sellers, early exercise related to ex-dividend date is not an issue.
A put is not vulnerable to early exercise in the same way as a covered call. This
brings up an interesting hedging advantage.

Key Point
The risk of early exercise applies to covered calls. For uncovered short
puts, this is not a problem because no dividends are in play.

The uncovered short put has the same market risk profile as a covered call,
with the exception of dividends. First, because you do not have to own shares in
a short put strategy, a standalone short put does not earn dividends. However, as
a hedge, the basic assumption is that you do own shares in your portfolio and
want to use a variety of strategies (including uncovered puts) to hedge market
risk. So you can combine covered calls and uncovered puts to continue this
hedging device each and every month.
In two out of three months, no ex-dividend date is involved, assuming
dividends are paid quarterly. So writing one-month covered calls is a sensible
and advantageous strategy at these times, because time decay is rapid. In ex-
dividend month, when you will want to avoid exposure of short calls to the
possibility of early exercise, leave shares intact but write uncovered puts. This
provides a similar income stream without being at risk of early exercise. The
hedge continues without the exposure.
The covered call is further at risk in months when earnings are announced.
This could be a month entirely different than ex-dividend month. The risk here is
that any earnings surprise, either positive or negative, creates a momentary
overreaction in the underlying price. Because it is most likely to self-correct
quickly, the price reaction to an earnings surprise can be ignored; however, this
may also present another potential timing for early exercise.
For example, in the case of either a covered call or uncovered put, a negative
earnings surprise may cause the underlying price to fall many points. In this
case, the call loses value and can be closed at a profit, but it could end up with a
paper loss in the stock. The put gains value and could be exercised, so it could
become necessary to roll forward to avoid exercise. A positive earnings surprise
causes the stock price to rise temporarily. However, at that price spike, the
uncovered put can be closed at a profit, but a covered call could be exercised to
take advantage of the difference between current market value and a much lower
strike.
With the potential risk related to earnings surprises, it could make sense to
avoid earnings month or to close out of positions before the date of
announcement. It does not matter whether a short option is in the money or not
before earnings are announced. The earnings surprise will change status very
quickly. So if a surprise does occur, a previously safe and buffered short position
could become an exposed one in a single session. Because you cannot know in
advance whether earnings will meet expectations, or end up above or below, this
risk is potentially severe. To hedge against this risk, it makes sense to be out of
short positions ahead of time.
After the earnings have been announced, you can take action to move into
long or short hedges. Knowing that price reaction to earnings surprises is
invariably exaggerated, this moment presents an excellent opportunity. When the
earnings surprise is negative and the underlying price declines, the paper loss is
not likely to last. This presents an opportunity to sell as put or even to buy a call
or other form of hedge. If the earnings surprise is positive, timing is excellent for
selling a covered call or buying a long put.

Key Point
Although price behavior after earnings surprises typically reverses very
quickly, the price spike presents a risk of early exercise.

Even for the most conservative investor focused on the long term, hedges that
are timed for price spikes following earnings surprises can be very profitable
short-term moves. They will not always work out as expected, but they usually
will, so like all hedges, the timing of this move based on likely price behavior is
easily spotted and has a better than average chance of resulting in a profitable
outcome.
When traders do not use options to hedge against price movement, events like
earnings surprises are likely to lead to ill-advised action. When price falls
drastically after a negative earnings report, some equity investors panic and sell,
only to see the price rise in following sessions. When the price jumps after a
positive report, some will buy shares in the belief that a strong bull market has
commenced. They are likely to see a price decline shortly after. This gut reaction
to earnings surprises is typical of market behavior, in which so many people buy
high and sell low instead of following the simple advice to buy low and sell
high.
Another consideration about early exercise is the tax consequence. If you
have greatly appreciated stock, exercise does not appear to present a great
problem; in fact, it yields a profit. However, tax consequences could change this
picture. For example, if your stock has appreciated rapidly but you have only
owned it for 11 months or less, that capital gain is treated as short-term. You
need to exceed a full year before getting the favorable long-term tax rate. So
when ex-dividend date occurs prior to the full-year holding period, rolling
forward makes sense, and when the capital gain is substantial, it makes even
more sense to extend your holding period.

Options: American Style vs. European Style


Most options on stock treaded in U.S. markets are called American style. This
means that exercise is allowed at any time during the option’s life. For a trader of
long options, this is an advantage; for the short seller, it presents a range of risks
related to timing of dividends and earnings.
In comparison, a European style option can be exercised only at the end of
the life of the option. This is most commonly seen for options on stock index
funds, where physical exercise would not be practical. In these forms of
European options, settlement is done on a cash basis, meaning any settlement
involves equivalent cash payments versus exchange of shares.

Early Exercise and Multiple Option Contracts


Some hedges consist of more than single options, and may be set up in a
calendar spread or straddle. These complex hedge positions, designed in some
cases to protect against downside market risk while creating short-term profits,
may also set up cost-free or low-cost situations that also improve your ability to
hedge against the risk of early exercise.

Key Point
The use of multiple options adds flexibility to many hedges, and this raises
potential for added profits in the hedging strategy.

This is most likely in multiple option positions when current in-the-money or


at-the-money options are rolled forward. If this is done only to avoid in-the-
money status or specifically to escape ex-dividend or earnings month, the
flexibility of multiple contracts becomes a hedging advantage to you as well.
This type of complex option rolling demands monitoring, and can provide you
with low-risk hedging advantages. For example, rolling a short option forward to
a higher strike (for calls) or a lower strike (for puts) avoids being at or in the
money, and may also create a breakeven or small credit outcome. These diagonal
moves gain added flexibility when you are able to roll forward and increase the
number of outstanding option contracts. For covered calls, maintaining complete
coverage may involve buying more shares or initially opening positions on only
part of your total portfolio position.
For uncovered puts, the situation is even more flexible. As a first step, the put
will hedge your portfolio position with the same market risk as covered calls.
However, if the put moves closer to the money, it can be rolled forward and
down, with a larger number of outstanding contracts. For example, if you have
200 shares and you sold two uncovered puts, these can be rolled forward and
down to three new uncovered puts.
Unlike the call, for which a full coverage relationship between stock and short
call is desirable, the short put has no such restrictions. The danger in this
diagonal roll is that the stock price could continue to fall. This demonstrates why
selection of companies for your portfolio demands low volatility and strong
fundamentals as a starting point. There is no danger in adding more puts, but if
the stock price continues to decline, this can get out of hand and potentially lead
to losses and a greater demand for higher-risk recovery strategies.
The most conservative version of increased contracts with a diagonal roll is
the purchase of additional shares. This does not always make sense for covered
calls, for which a diagonal forward roll is likely when the underlying price has
increased; so buying more shares means increasing average basis and a higher
risk in the case when the underlying price retreats. For short puts, however,
rolling forward and selling a higher number of contracts does not require buying
more shares. But doing so is a recovery move, because the price per share in this
situation will be lower than the original strikes, so buying additional shares
averages your basis down.

If You Own Long Calls


Within a hedging strategy, there will be instances in which you own long
calls. For example, as part of a spread or straddle, or a synthetic long stock
strategy, one side of the overall trade could involve a long call. One likely
outcome is that a short side loses value and you enter a buy to close to take
profits and remove market risk. What happens to the long call?
The same arguments presented earlier about the risks to sellers may also
translate to advantages to long call owners. There are several instances in which
owning calls present opportunities for hedging. These include:
1. The desire to buy more shares. The contingent purchase is one type of
hedge based on ownership of a long call. Whether this is bought by itself
and represents an orphan position in a more complex hedge, it is the
application of the call to fix the price per share in a future purchase of
stock.

Key Point
Once the short side of a spread or straddle is closed, the remaining long
option can be used to set up additional hedges.

2. Creation of future hedges with other options. The orphan long call can
be combined with other positions, usually short, to set up future hedges.
This outcome is likely to occur with a calendar spread, in which a short
call with earlier expiration is covered by the long call with later expiration.
After the short side has expired, the long side can be used to create new
hedges on the same underlying.

Key Point
Risks to sellers often are offsetting profit possibilities for buyers. With long
options outstanding, you can take advantage the dividend timing.

3. Buying stock to earn dividends. Just as your short option places you at
risk of early exercise, your long option sets up an opportunity to exercise
right before ex-dividend date, and earn the current dividend. This is an
excellent return on investment. Although the current dividend is based on a
three-month period, you can earn it with ownership of stock over only a
few days. This enhances the current position in the same equity within
your portfolio.

If You Own Long Puts


The opposite side of this analysis applies equally well and presents hedging
opportunities. When a long put is left over from an earlier hedge (when the short
side has been closed to take profits, for example), the long put is not a
disadvantage but a potential benefit. For example:
1. The desire to sell shares. The contingent sale is a hedge based on
ownership of a long put. You might desire to sell shares in the future, and
the long put fixes the sales price well in advance in case you decide that
selling shares makes sense.
2. Creation of future hedges with other options. The orphan long put
works like the long call; it can be combined with other positions, usually
short, to set up future hedges. This outcome is likely to occur with a
calendar spread, in which a short put with earlier expiration is covered by
the long put with later expiration. After the short side has expired, the long
side can be used to create new hedges on the same underlying.

Key Point
The insurance put is one happy consequence of closing a short side before a
long side. The outcome is fixing maximum loss at the net basis in the
overall position.

3. Insurance put. The remaining long put serves another hedging purpose,
which is to set up an insurance put. This fixes the maximum loss at the
strike, adjusted for the profit or loss on the original position. If the stock
price declines below that adjusted basis in stock, the put gains intrinsic
value of one point for each point of decline in the stock’s price.
Early exercise is generally a problem within a hedging strategy—unless part
of the hedge accepts a contingent purchase or sale as part of the overall strategy.
When you want to keep equity positions intact, the techniques for avoiding
exercise can be applied. These include avoiding being in a risk situation—
avoiding ex-dividend month for short calls and avoiding earnings month for all
short options.
The next chapter includes a summary of option collateral and tax rules.
Although these do not relate specifically to hedging, the possible advantages or
disadvantages of the tax and collateral rules can affect your selection of one
hedge over another.
14 Collateral and Tax Rules for Options Trading

The strategies used to set up hedges involve a rich variety of option-based


positions. These are not necessarily complicated, and can be put into action
assuming a few basic requirements:

Key Point
Knowing your limitations is essential for successful investing. With
options, knowing the rules of trading and the risks is a wise first step.

1. You understand option rules and risks. The first rule for options (and all
types of investments) is that you know the trading rules and the risks
involved. When you first look at the terminology and variety of options,
the whole idea seems exotic and risky. And in fact, if you do not first
master the basics, you should not use options, either as a conservative
hedge or for speculation.
2. You are qualified by your broker. When you apply for approval to trade
options, your broker requires that you complete an option application. On
this document, you explain your experience and knowledge about options.
Your broker then assigns a “level of approval” for trading. This defines
how much risk your broker will allow you to take. Every broker defines a
series of levels and the types of options trades allowed. The lower level is
the most conservative strategies, and as the levels advance, the allowed
trades become more complex.
3. Your risk profile is conservative. If you are a speculator willing to
assume high risks, you probably are not at all interested in setting up a
long-term portfolio of equities or in hedging against risks. Speculators are
interested in very short-term profits and tend to favor using high-risk
options strategies. However, conservative investors understand the
importance of setting up a permanent portfolio and protecting it against
market risk.
4. When losses occur, you are able to accept them. In order to succeed in
any investing program, especially one involving hedges, you also need to
be able to accept losses. While the purpose of hedging is to reduce loss
exposure, you will have losses in some percentage of your trades. An
important element of hedging is to know when to walk away and focus on
the next trade. Anyone who “doubles down” and tries to offset yesterday’s
loss with total’s higher-risk trade is only inviting problems.
5. You know your breakeven rate. In Chapter 2, the concept of a breakeven
rate was explained. This is the rate of return you need in order to break
even after accounting for inflation and taxes. Most investors who calculate
this are surprised and disturbed at what they discover: that they need to
exceed the typical net return from investing. In fact, to make a breakeven
rate consistently, you have only two choices. First, you have to increase
the risk levels you are able to tolerate (which also adds the possibility of
higher losses). Second, you need to use options to set up conservative
trades and hedges, not only to increase your typical rate of return but also
to reduce market risks.

Key Point
Many investors are not aware of what they need to break even after
inflation and taxes. This is a troubling fact of life.

Once you meet these initial requirements, entering into a hedging program is
both sensible and profitable. Options are being used more and more as hedging
devices, versus emphasis in the past on speculation. However, two aspects of
options trading add complication to this process: margin rules and taxes.

Margin Rules for Options


When you buy stock, the margin rules allow you to finance 50% of your
purchase in the margin account. This is an attractive feature for stock trading, but
it also adds significant risk. Some traders assume that the same margin rules
apply to options trading, but they do not. In options trading, “margin” is not a
reference to leverage alone, but also defines required collateral that has to be
placed on deposit. The rules are complex, but they can be studied free by getting
a copy of the CBOE Margin Manual. (The link was included in Chapter 1.)
The basic collateral rules are easily mastered. Long calls and puts have to be
paid in full at the time of the trade. Short option trades at least 20% of the
underlying value based on the strike. This is adjusted for the moneyness of the
option and for the proceeds received. For advanced strategies like spreads and
straddles, the calculation is more complex. Use the CBOE Margin Calculator to
determine the amount required to be placed on deposit for each type of trade.
(See Chapter 1 for a link to the Margin Calculator.)

Tax Rules for Options Trading


The federal tax rules for trading options are perplexing and complex. In
Chapter 6, the essentials of a “qualified” and “unqualified” covered call were
explained. The important point to remember is this: If you open a deep in-the-
money covered call before you have reached the one-year holding period to
qualify for long-term gains on stock, the count is tolled as long as the call is
open. This means the period of time required is stopped as long as the
unqualified call remains open. Although this affects only a small portion of
trades, it can become an issue.

Key Point
The term unqualified doesn’t mean you cannot open such a trade. It does
mean you risk losing favorable long-term capital gains rates.

For example, if you have owned stock for less than one year and the value has
increased substantially, the loss of long-term capital gains rates can add to your
tax liability. An unqualified covered call can also be created unintentionally if
you roll a short call forward to avoid exercise after the underlying price has
risen. The two trades—the initial call and the new call—are treated as separate
transactions, and this could result in an unseen tax burden.

Resource
To get a free download of “Taxes and Investing,” link to
http://bit.ly/1WHnCsn.
To fully understand the tax rules, refer to the free booklet provided by the
CBOE entitled “Taxes and Investing.” This runs down the essential rules you
need to know. However, for any level of complex options trades, you should also
consult with your tax advisor and make sure he or she fully understands how
options taxation works.
Capital gains are well understood by most investors. However, with options a
few qualifiers have to be added into the mix. Here are a few of the options-based
rules:
1. The constructive sale rule could apply. This occurs when you are taxed
as if you closed a position even though you did not. When you open
offsetting long and short options on the same underlying, you could create
a constructive sale. For example, you own 100 shares of stock and you
open a long and a short position involving a straddle or spread. This might
or might not set up a constructive sale. This is defined in IRC Section
1259, which explains that offsetting positions against positions already
owned are constructive sales, such as a short sale of stock when you also
own shares of the same stock. An exception arises when the trade is closed
at least 30 days before the end of the year.

Key Point
You cannot move in and out of trades timed for taking losses at the year
ends, unless you wait more than 30 days.

2. Wash sales prevent loss deductions. A wash sale occurs when a position
is closed and then reopened within 30 days. Some traders have attempted
to set up tax deductible losses by selling close to the end of the year and
then moving back in right after the beginning of the new year. A wash sale
also applies when shares of stock are sold and replaced with an in-the-
money out.
3. Exercised long options are treated as part of basis in stock. When you
exercise a long position, it is not treated as a separate trade, but adjusts
your basis in the shares of stock bought or sold.
4. Short option premium is not taxed at the time the trade is opened.
When you sell an option, you receive proceeds. However, this trade is not
taxed until it is closed, expires, or gets exercised.
5. Some losses are not deductible in the year of the loss. A straddle is
treated as an offsetting position. This means that deductions could be
deferred until the so-called “successor position” (the other side of the
straddle) has been closed or expires. In this situation, a loss in the initial
closed position might not be deductible until both sides are closed.
6. Married puts may be treated as adjustments to stock basis. A married
put (an insurance put) is treated as an adjustment to the basis in stock when
the put is purchased on the same day as shares of stock. However, if the
put is sold before expiration, the outcome is treated as a short-term capital
gain or loss.
The many tax rules make options more complex than most forms of investing.
As a hedge against equity positions, the option can provide safety and risk
elimination in many forms; and the best of these is the hedge that also creates
added income. Considering the desirability of capital gains and dividends,
options add another form of income as well as hedging benefits.
Considering the difficulty of balancing a conservative risk profile with
breakeven returns (based on inflation and taxes), the options hedge is one of the
few varieties of investing that enables you to beat the averages without needing
to accept higher risks.
15 The Final Twist: Proximity

In the overall analysis of options as hedges, the biggest question of all involves
when and where the trade should be entered. In Chapter 5, the discussion of
timing for options trades was based on the study of Western and Eastern reversal
signals. However, there is one final point to study: the proximity of signals.

Key Point
Hedges are most effective when they are timed to coincide with price
proximity to resistance and support, where the greatest uncertainty enters
the picture.

The “proximity” of the signal to the trading range determines more than
anything else when a specific trade is likely to be profitable. This is a reference
to the location of price in comparison to two important lines: resistance and
support.
These terms are well understood by most investors; however, for the purpose
of hedging, the range in between resistance and support is the area of agreement
between the two sides, the range of prices agreed upon by the forces of supply
and demand. The widely accepted significance of resistance and support define
the proximity issue for hedging purposes.
Investors understand that if and when price moves above resistance, it can
indicate a coming reversal back into range, or a breakout to a new, higher range;
and that when price moves below support, it may represent a coming upward
reversal or the beginning of a new, lower trading range. However, this widely
accepted view of the trading range and its borders is applied mainly to determine
whether a new trade should be entered. So equity holders will either buy or sell
based on how they perceived the movement of price above resistance or below
support. Ironically, this decision-making process turns buy-and-hold,
conservative equity investors into speculators. If an investor makes a trade based
solely on the movement of price outside of the current trading range but without
understanding what happens next, then the buy-and-hold standard is being
abandoned.

Key Point
Decisions investor make based on price behavior at or near resistance and
support can easily turn conservatives into speculators.

This is one of the many flaws in how many investors operate. However, when
focused not on exploiting what might be a momentary price movement, but on
how to hedge the move, the process of portfolio management becomes more
stable. Assuming that the equity positions in your portfolio are ones you prefer
to keep, the proximity of price to the edges of the trading range are a concern,
but that move should raise hedging questions rather than the question of
changing the equity position.

The Zone of Resistance or Support


When price is in proximity to resistance or support, reversal signals take on
greater meaning. A reversal signal occurring at midrange is less likely to lead to
actual reversal, especially when compared to the proximity at or moving through
resistance or support. This is true whether using a specific price level or a zone
of resistance or support.
For example, the one-year chart of Atmos Energy (ATO) includes more than
nine months of a consolidation trend, ending with a breakout above resistance
near the end of September. The levels of resistance and support varied during
consolidation, but only by a single point. This is typical of the zone approach to
resistance or support. The zone itself tends to be quite small compared to typical
trading price movement, as shown in Figure 15.1.

Figure 15.1
Key Point
Identifying resistance or support zones may make it easier to decide where
price range actually exists.

The use of zones helps identify resistance and support in situations such as
this, as well as clearly identifying the nature of a breakout. This was signaled in
late August with a single session displaying an unusually long lower shadow.
Although the shadow remained within range, it revealed a weakness among
sellers, anticipating a breakout to the upside, which did occur one month later.
The many instances of interim reversal make the point about proximity. The
consolidation range was characterized by offsetting tests of resistance and
support, versus periods of reversal comfortably within range. The period of mid-
March through the end of April, for example, saw reversals of 2.5 points
numerous times. However, the price level remained in midrange and more than 2
points from either zone. In this chart, the interim range reversals were weak
because the price proximity was also weak.
In every instance where price reached resistance or support zones, daily
trading range increased to as much as 3 points. However, because price failed in
each case to break through either zone border, consolidation held. This is a
demonstration of one form of strength and weakness in reversal signals based on
proximity.

Dynamic Trading Range Patterns


The typical trading range is strong. This means that reversal signals at or near
resistance and support carry greater weight. At the same time, once price does
break through, a reversal back into range is the most likely outcome. Eventually,
all trading ranges give way to new dynamic trading either above or below;
however, in the short term, the range tends to rule and gives order to the price
pattern.
This means that hedges can be timed to protect equity positions against
sudden moves in price. When support holds strongly, the hedge against
downside market risk is of less concern. When resistance holds over time, it
indicates a stronger likelihood of a coming price decline. In a sense, the failure
of price to move outside of the established range indicates more likely
movement in the opposite direction. So strong support indicates a bullish
tendency, and strong resistance points to a bearish move in the future.
For example, on the chart of Church & Dwight in Figure 15.2, support held
over the entire period as resistance gradually rose. However, on two occasions,
price levels moved above resistance and then retreated. These failed breakouts
confirm the strength of resistance.

Figure 15.2

It is noteworthy that volume spikes accompanied the larger movements in


price, especially during and preceding the breakout periods. Volume spikes can
identify either opportunity or danger, depending on price behavior at the same
time, and on what types of signals emerge.

Proximity as a Measurement of Strength

Key Point
Reversal and confirmation normally are stronger when in close proximity
to resistance and support, and tend to be weaker at mid-range.

The likelihood of breakout relies on the strength of a reversal signal followed


by strong confirmation. A mid-range reversal signal is weak, meaning less
likelihood (but no certainty) of a change in the range. However, when reversal
leads to immediate breakout, it is likely to succeed if strongly confirmed by
other signals. This further indicates the timing and placement of effective
hedging strategies.
For example, the four-month chart of Canadian Imperial Bank (CM) shown in
Figure 15.3 begins with a narrow trading range and a bullish reversal that occurs
with a breakout below support. This is one of the strongest forms of proximity
for a bullish move. The failure of price to hold below support points the way to
strong moves to higher levels. The bullish piercing lines is the initial reversal,
and this is immediately confirmed with a series of price gaps, moving above
resistance.

Figure 15.3
The move sets up a new consolidation trend lasting nearly two months. This
concludes in a similar pattern with proximity to support. The bearish harami
cross leads to a move of price strongly below support and characterized by
repetitive gaps.
Both of these patterns, the same type of breakout, located in proximity to the
edges of the trading range, lead to strong breakout. Recognizing this pattern also
identifies the timing for effective hedges. Once the first and second gaps
appeared in late September, it looked like a bullish trend was starting above
resistance. This would be a good point to close any short calls or to open long
calls, anticipating a new bullish trend. By mid-October, the daily range
narrowed, pointing to an end to the price move.
When the bearish harami appeared, it pointed to a possible bearish breakout.
However, this was not confirmed until the second gap, which moved below
support. This would have been a good point for a long put to hedge the equity
price, or even to open a covered call in anticipation of the price decline.

Applying Hedges to Recognized Patterns


The entire concept of hedging, you may recall, is based on applying one set of
strategies (options) to reduce or eliminate risk on portfolio positions. Without
this hedge, every investor is at the mercy of unknown market forces. No one
anticipates big corrections, and no one knows how far they go or which stocks
will be affected.

Key Point
Awareness of proximity helps decide which hedge works best, based on the
most likely price behavior to follow.

Hedging for its own sake is no better than just buying shares of stock and
hoping for the best. A hedge is effective only if and when it is based on
recognized patterns seen in the price of the underlying stock. This applies to
even the most basic hedge. For example, opening a covered call when a stock’s
price is at a bottom is ill-advised. When the stock price rises, the call goes in the
money and is at risk of being exercised. Proponents of the strategy argue that as
long as the strike is higher than the basis, this is an acceptable outcome. Even if
this is true, why open the covered call at the wrong proximity?
It makes more sense to open the covered call at the top of the price swing. If
and when the underlying price declines, the call is closed at a profit or allowed to
expire. Meanwhile, the shares of stock cover the risk and you continue earning
dividends. Then, at the bottom of the swing, different strategies are employed,
such as short puts. This version of hedging is timed to exploit the cyclical price
swings, but it relies on pattern recognition and reversal signals.
Another example is the frustrating but common consolidation pattern.
Investors tend to be impatient. They want action, and they want to see profits
accumulate quickly. (Losses also accumulate quickly with high volatility, of
course.) However, in consolidation, prices are range-bound, often for several
months. So impatient investors are likely to sell shares during consolidation,
even at a loss, and look for more volatility elsewhere. This occurs even when the
underlying stock remains a sound investment with strong fundamentals. There is
a solution to this inactivity, and it does not include selling positions. Specific
hedges work in this situation.

Key Point
Strategies (like short straddles) often labeled as high risk may in fact be the
most conservative. It all depends on how and where they are used.
The covered straddle is one of the most effective hedges. It combines a
covered call with an uncovered put. Some will point to a short straddle as high
risk, but this position is one of the lowest-risk hedges possible. The covered call
is conservative and the uncovered put has the same market risk; in consolidation,
whether price trends upward or downward, one of the two sides will lose value
and can be closed at a profit or allowed to expire worthless. Recognized
breakout patterns point to the best timing for closing one or both sides of the
straddle, or rolling forward to avoid exercise. It is, once again, all a matter of
proximity. Recognizing how price behavior changes when it is close to
resistance or support, and then acting on discovered reversal and confirmation
signals, ensures that the timing will work for you. In this situation, the hedge not
only protects equity positions, but uses them to generate added income with very
little added risk. This is the best possible form of hedge.
The hedge takes many forms, and each contains its own element of risk.
However, in order to accept the premise that hedging improves profits while
reducing risks, several popular assumptions have to be examined. These include:
1. Long-term investors should not make short-term trades. To the
contrary, short-term hedge trades are effective at reducing and controlling
market risk. The idea that conservative investors should just buy value
stocks and forget about them is a reckless and dangerous form of advice.
The once-popular blue chip stocks that have fallen out of favor and even
gone into bankruptcy cannot be ignored. Companies like General Motors
and Eastman Kodak once were considered the best investments available;
but nothing is a certainty.
A wiser and safer method is to find high-quality investments based on
strong and long-lasting fundamentals; buy shares and monitor the
fundamentals continuously; and hedge market risk with conservative
option-based strategies.
2. Contrarian investing is high risk. The majority is usually right. In spite
of evidence against this belief, it persists. The majority is wrong more
often than it is right, and when investors follow majority thinking (the
“crowd mentality”) it is a dangerous decision. Because emotions such as
greed and panic rule market thinking, far too many investors buy high and
sell low, instead of doing the opposite.
Contrarian investing is not a reference to only doing the opposite of the
majority. It is a method for determining why to buy or sell, based on cold
analysis rather than on gut reaction, emotion, or wishful thinking.
3. Consolidation is a pause between trends, a period when no one knows
what to do. This is another popular belief, and investors not yet in
positions tend to wait until consolidation ends before putting cash into
positions. This means many opportunities are lost. By failing to recognize
consolidation as a third type of trend (sideways rather than up or down),
those opportunities are lost.
A related belief is that no reversals can be found because there is no
trend to reverse. However, if you change the definition of reversal to
movement away from consolidation and into a dynamic trend, it is easy to
spot signals and confirmation for timely trades.
4. An entry point is “zero.” Although few people say it out loud, some
investors view their entry price as a zero point. They expect equity values
to rise from that price forward. So if price falls, they are taken by surprise.
Everyone knows that any price is part of an unending series of movements
every day. However, with the unstated belief that stock prices will rise, too
many people are taken by surprise when things don’t work out as they
assumed.
This reality points to the essential nature of proximity. By recognizing
price behavior at mid-range compared to how it changes as it moves into
proximity with resistance or support, better-timed decisions are possible in
all types of trends. Reversal and confirmation also are more reliable at
close proximity to the resistance and support levels, improving timing of
equity trades as well as hedges.
5. Diversification is a smart management method. This basic idea makes
some sense, but excessive diversification is a mistake. It is an admission
that risk cannot be avoided, which is not true. Using hedging strategies is
more effective at mitigating market risk than just spreading risks among
many different products. Excessive diversification through mutual funds or
ETFs makes the point: When you rely too heavily on diversification, you
cannot expect a better return than the average of the portfolio, including
the best and the worst. The same argument applies to asset allocation,
which is simply a different name for the same idea but based on industry
sectors rather than on individual stocks.
Most investors can create effective net returns in a less-diversified
portfolio of well-chosen value investments, especially if those positions
are protected with conservative hedges. Options used in this manner are
not high risk, but actually are more conservative than trying to beat the
averages by relying too much on diversification.
6. Options are too risky for conservative investors. A thoughtful analysis
of options strategies reveals the truth: Some strategies are designed as
speculative plays, intended to time the markets to exploit price movement
—even when there is no equity portfolio involved. Speculators often rely
on options to leverage capital, and they often are less concerned with risk,
believing that a few good trades (or lucky trades?) make up for high risks.
Other strategies, which have been described in this book, are designed for the
most conservative purpose, protecting a portfolio of equity positions against
market risk. So creating current income or setting up low-cost or no-cost hedges
creates the effective safety net that every stock market investor wants. The
solution is not based on wise stock selection or broad diversification, but on
timing and proximity, emphasis on the fundamentals for long-term growth, focus
on high-dividend companies, and faithful application of options strategies based
on the specific conditions and the current proximity of price to resistance and
support.
• • • • • •

Key Point
An effective hedging strategy begins and ends with stock selection. The
fundamentals determine the health of your equity portfolio

The selection of a company for your portfolio is a starting point. The


desirability of holding positions for the long term does not mean applying a “get
and forget” strategy, but demands constant monitoring. If strong fundamentals
turn weak, it is time to close out equity positions and look elsewhere. Option
hedging works most effectively when used for equity positions in companies
with strong long-term fundamentals. At the very least, this should include
analysis and comparison of dividends, P/E ratio, revenue and earnings, and long-
term debt trends.
Once you have created a strong portfolio of exceptional companies and their
stocks, the option hedge is the effective measure of risk reduction and
elimination. The same hedge positions are also used effectively to time trades
after exaggerated short-term price movement, or to enter a safe and thoughtful
recovery strategy. There are no certainties in the market, but the combination of
high-quality investments and well-designed hedges reduces the uncertainty that
every investor faces.
Glossary

1-2-3 iron butterfly: A strategy employing three strikes for a series of


butterflies, with an increasing number of options positions at each strike.

1-2-3 iron condor: A condor with three strikes, with the middle expiration a
reverse iron condor; and with each subsequent expiration, the number of
contracts is increased.

American style: The type of option most common for underlying stock in the
U.S. markets. An American option can be exercised at any time.

Annualize: A process of restating a net return as if the position were open for
one full year; this is necessary to make valid comparisons between two or more
options trades with dissimilar periods.

Ask: The price a buyer pays for buying an option.

Asset allocation: A strategy intended to balance risks among several different


products, such as equities (stocks), debt (bonds), and commodities.

Automatic exercise: A procedure by which all options in the money are


exercised at expiration.

Basis: The price used to calculate a return, which for options trades may be the
cost of stock, current market value of stock, or the strike of the option.

Bear spread: A spread designed to increase in value when the underlying price
falls.

Beta: A measurement of how closely a stock’s price follows or responds to the


overall market. A beta of 1.0 is held by a stock that moves the same as the
market. Higher beta indicates higher volatility.

Bid: The price a seller receives for selling an option.

Bollinger squeeze: Within a range of Bollinger Bands, the tendency for prices to
move close to either upper or lower band and to narrow considerably, which
often occurs just before a period of increased volatility.

Box spread: A combination of call and put vertical spreads to create a bullish or
a bearish combination.

Boxed synthetic stock: A hedge combining a synthetic long and synthetic short
stock, opened at the same time. It offsets price movement in both directions.

Bull spread: A spread designed to increase in value when the underlying price
rises.

Butterfly: A spread strategy using calls, puts, or both, with three strikes; the
butterfly can also be either long or short, depending on selection of options at
each strike.

Calendar straddle: A variation of the straddle in which an earlier-expiring short


straddle is combined with a later-expiring long straddle.

Call: An option granting its owner the right to buy 100 shares of stock at a fixed
price or a specific stock, by or before expiration date.

Call bear spread: A credit spread combining a long out-of-the-money call with
a lower-strike in-the-money call.

Call bull spread: A debit spread combining a long in-the-money call with a
higher-strike short out-of-the-money short call.

Collar: A three-part strategy combining long stock, a short call, and a long put.
Both options are out of the money. The collar is designed to limit profit and loss
without cost.

Condor: A strategy with four options and four strikes, offering limited profit and
limited loss, working as an effective hedge for low-volatility stocks.
Confirmation bias: A tendency to find confirmation of a previously set
assumption, even when opposite signals are plentiful.

Conformity risk: The risk associated with following the majority of the market,
and of making decisions with the majority rather than based on logic and
analysis.

Conservative: An investor who bases investment decisions on analysis of risk


and with the idea of reducing risk while maximizing return.

Consolidation trend: A trend moving sideways within a narrow price range,


with neither buyers nor sellers able to move price beyond its current range.

Constructive sale: A tax rule defining some offsetting positions as taxable even
though there was no sale. This applies to some options positions.

Contingent purchase: The use of a long call to fix the price for a future
purchase at the strike, which may or may not come to pass depending on price
behavior.

Contingent sale: The use of a long put to fix the price for a future sale at the
strike.

Continuation: A tendency for a trend to continue in the same direction until a


reversal signal appears. This tendency is confirmed through continuation signals.

Contrarian: An investor who times trades based on logic and analysis rather
than emotion, acting unlike the majority in the market.

Contrarian investing: The practice of making investment decisions based on


logic rather than on emotion, and acting contrary to the prevailing market
tendency to trade based on emotions.

Covered call: A basic option hedge in which a call is sold against 100 shares of
stock. In exchange for receiving a premium, the call writer may be required to
sell stock at the fixed strike.

Covered straddle: A bullish conservative strategy combining 100 shares of


stock, a covered call, and an uncovered put, using the same strike and expiration.
Credit spread: A spread yielding more income from the short side than the cost
for the long side.

Debit spread: Any spread with the long side costing more than the premium
received for the offsetting short side.

Debt capitalization ratio: A fundamental indicator identifying the percentage of


total capitalization represented by long-term debt, versus stockholders’ equity.

Deep in the money: An option far from current value of the underlying stock,
usually 5 points or more.

Delta: A measurement of changes in an option’s premium in relation to changes


in the underlying stock.

Diagonal ratio spread: A diagonal spread employing different numbers of short


and long options.

Diagonal roll: A forward roll to a higher or lower strike, intended to decrease


exercise risk or to improve capital gains in the event of exercise.

Diagonal spread: A spread with different expirations and different strikes.

Diversification: Spreading investment capital among several dissimilar products


in order to manage risks, on the theory that the overall return will be greater than
the return on any one investment.

Dividend yield: The percentage of a dividend per share, calculated by dividing


the annual dollar amount of the declared dividend, by the current price per share.

Dow Theory: A set of beliefs concerning price behavior, based on the writings
of Charles Dow, co-founder of the Dow Jones Company.

Early exercise: The action of exercising a long call or put before the last trading
day.

Earnings surprise: An earnings outcome that is not the same as analysts’


expectations. This surprise often results in short-term exaggerated price reaction.

Eastern technical signals: Broad description of reversal and continuation


signals based on price patterns found in candlesticks indicators.

Efficient market hypothesis (EMH): A belief that stock prices reflect all
known information at all times.

European style: An option for which exercise can occur only at the end of the
option’s life, on a predetermined date or range of dates.

Exchange-traded fund (ETF): A mutual fund with a predetermined basket of


securities, which trades like a stock and may also offer options trading. An
inverse ETF becomes profitable when the basket of securities declines in value;
a leveraged ETF multiplies the effect of profit or loss based on the basket of
securities.

Exercise: The act by an owner of an option to call 100 shares from a call seller,
or to put 100 shares to the put seller. This action is controlled by the option’s
owner, who will choose to exercise when that is profitable.

Exercise hedge: A trade that hedges against exercise rather than against market
risk.

Expiration date: The fixed month and date on which every option expires.
Expiration date is the third Saturday in expiration month, and the last trading day
is the third Friday.

Extrinsic value: Alternative term for implied volatility (IV).

Forward roll: Closing to buy an existing short position and replacing it with a
later-expiring one, with the same strike or a higher strike.

Fundamental volatility: The degree of reliability in fundamental trends over


time, used to determine fundamental risk levels for a company.

Gamma: A Greek that measures the degree of change in Delta, useful in


spotting responsiveness of option pricing to underlying stock movement.

Greeks: A set of calculations defining option price and volatility levels in


relation to the underlying stock.

Hedge matrix: An arrangement of multiple legs to a single strategy, designed to


take advantage of price movement so that profits are realized in either price
direction.

Hedging: Any form of investment made to offset or reduce the risk of loss in
another position. For example, an option may hedge a stock position so that a
loss in the stock is offset by a profit in the option.

Historical volatility: The level of volatility based on price movement over time
and within a range of trading; a basic test of risk.

Horizontal spread: A spread consisting of two options with the same strike but
different expiration months.

Implied volatility (IV): The portion of option premium that changes based on
the historical volatility of the underlying stock and is also affected by time
remaining until expiration.

Informationally efficient: An attribute in EMH reflecting efficient pricing, but


not always accurate interpretation of information.

Intrinsic value: The portion of an option’s premium equal to the number of


points in the money.

Iron butterfly: A butterfly employing both calls and puts, consisting of four
different options and three strikes.

Iron condor: A condor employing both calls and puts with four strikes.

Island cluster: A series of trades occurring after a gap out of range and
concluding with a new gap back into range.

Leverage: A method for duplicating opportunity or risk with a reduced amount


of capital. For example, every option controls 100 shares of stock for a small
percentage of the cost of stock.

Long: Status for an investor who has bought an option, following the sequence
“buy-hold-sell.”

Long condor: A condor consisting of calls, a long ITM, a long ITM, a long
OTM, and a short OTM.
Long straddle: A straddle consisting of the same number of long calls and long
puts, with the same strike and expiration.

Long strap: A strap using long calls in greater numbers than long puts; a
speculative expansion of the long straddle.

Long strip: A strip using long puts in greater numbers than long calls; a
speculative expansion of the long straddle.

Lost opportunity risk: In a covered call position, the risk that the underlying
stock’s price will rise above the call’s strike, and potential added profits will be
lost due to exercise of the in-the-money position.

Magical thinking: A set of beliefs that adhering to some form of ritual leads to
positive outcomes, even when the outcomes are not related to the rituals.

Market risk: The risk faced by investors that a position will lose value rather
than gaining, resulting in either paper losses or realized losses.

Married put: A position involving a long put opened as insurance for 100
shares of stock held in the portfolio. The put protects against losses below its
strike price.

Moneyness: The relationship between the underlying stock price and the strike
price of an option.

Net basis: In a covered call, the net cost of stock, discounted by premium
received for selling the call.

Net return: The percentage derived when dividing net earnings by revenue.

Offsetting position: Description in tax law of straddle legs, in which a loss on


one may not be deductible until the other side has also been closed.

Payout ratio: The percentage of earnings used to pay dividends, best analyzed
over a period of time to determine whether the use of earnings for dividends is
rising or falling.

Perfect hedge: A hedge in which the cost of one side is exactly the same as the
benefit on the other side, so that the net outcome is no gain and no loss.
Pinning the strike: A tendency for underlying stock prices to move toward the
closest option strike in the last few days prior to option expiration.

Premium: The price of an option that a buyer pays or a seller receives;


expressed at price per share without dollar signs, premium translates to dollar
value at 100 times higher (for example, a premium of 1.25 is equal to $125).

Price spread: The difference between bid and ask of an option.

Put: An option granting its owner the right to sell 100 shares of stock at a fixed
price or a specific stock, by or before expiration date.

Put bear spread: A two-part position consisting of a long put with a lower-
strike short put, with the combination creating a net credit.

Put bull spread: A credit spread combining a long out-of-the-money put with a
higher-strike short in-the-money put.

Random walk hypothesis (RWH): A belief that price movement is random and
that it is not possible to beat market averages consistently.

Ratio calendar spread: A horizontal spread consisting of dissimilar numbers of


short options and later-expiring long options sharing the same strike.

Ratio combination calendar spread: A single strategy combining the call-


based and put-based ratio calendar spreads, opened at the same time, and usually
involving out-of-the money strikes on both sides.

Ratio write: A variation of the covered call combining a greater number of calls
than are covered by long stock.

Reconstituted hedge: A form of recovery strategy that consists of extended


expirations and modified strikes, to offset current losses.

Recovery strategy: A trade intended to offset a prior loss by recapturing value


through a subsequent trade.

Resistance: The highest price level in the trading range, identifying the price at
which buyers are willing to buy.
Reversal: The tendency of all trends to eventually end and move in a different
direction, either dynamic (bullish or bearish) or consolidation (sideways).

Reverse iron butterfly: A butterfly employing both calls and puts, but with out
of the money short and middle-range long options.

Reverse iron condor: A condor with the long and short sides of the condor
opened in reverse.

Risk: Exposure to loss, a threat, or a possible negative outcome, which may be


reduced or eliminated with specific options hedging strategies.

Risk awareness: Knowledge about a range of risk involved with initial and
recovery strategies, and the levels of risk involved with a series of decisions.

Rolling down: A forward roll on a diagonal, replacing a current strike with a


later-expiring lower strike.

Rolling up: A forward roll on a diagonal pattern, replacing a current strike with
a later-expiring higher strike.

Round trip cost: The cost to complete both sides of an options trade, entering
and then exiting a position.

Short: Status for an investor who has sold an option, following the sequence
“sell-hold-buy.”

Short condor: A condor reversing the positions of the long condor.

Short straddle: A straddle consisting of an equal number of short calls and short
puts, opened with the same strike and expiration.

Short strap: A strap using short calls in greater numbers than short puts; an
expansion of the long straddle hedge.

Short strip: A strap using short puts in greater numbers than short calls; an
expansion of the long straddle hedge.

Speculative hedge: A form of hedge that increases market risk in the hope that a
loss can be offset with a higher-risk position.
Speculator: A trader willing to take high risks in the desire to out-perform
typical net returns.

Spread: A position combining two options that offset one another in one of
several configurations.

Standardized terms: The four terms defining an option by its characteristics,


which apply to every listed option and cannot be altered or exchanged.

Straddle: An option position combining call and puts with the same strike and
expiration, both either long or short.

Strap: A variation of the straddle combining more calls with fewer puts.

Strike price: The fixed price at which every option can be exercised, regardless
of price movement in the underlying stock.

Strip: A variation of the straddle combining more puts with fewer calls.

Support: The lowest price level in the trading range, identifying the price at
which sellers are willing to sell.

Synthetic long stock: A hedge combining a long call with a short call at the
same strike; it duplicates price movement in the underlying and performs best
when the underlying price advances.

Synthetic short put: Alternate name for the covered call. Combining 100 shares
of stock and one short call creates a position that behaves exactly like an
uncovered put.

Synthetic short stock: A hedge combining a long put with a short call, with the
same strike; it duplicates price movement in the underlying and performs best
when the underlying price declines.

Synthetic stock: A position set up using options that duplicates price movement
in the underlying stock, point for point—but with less market risk.

Theta: A Greek that measures the rate of time decay for specific options.

Time decay: The decline in time value as expiration approaches, which is


accelerates toward the end of the option’s life.

Time value: The portion of an option’s premium related directly to the time
remaining until expiration.

Total capitalization: The combination of long-term debt and stockholders’


equity, the overall capital valuation of a company.

Total return: A calculation of net returns combining option yield with dividend
yield. The rationalize is that stocks may be selected for option trades based at
least in part on dividend yield.

Trend: A tendency for change over time to continue in one direction until it
concludes; this is applicable to fundamental as well as to technical outcomes.

Underlying security: The stock or other security to which every option is tied,
which cannot be changed.

Unqualified covered call: A call whose deep in-the-money status tolls the
period leading to qualification for long-term tax rates.

Variable ratio write: An expansion of the covered call in which more calls are
written than can be covered with stock, but using two different strikes.

Vega: A measurement of changes in an option’s premium caused by implied


volatility (IV). It may be used to measure the speed of volatility collapse during
the final month of the option’s life.

Vertical spread: A spread made up of two options with the same expiration date
and different strikes.

Volatility: The degree of uncertainty in pricing of an asset, as a reflection of


price and market risk. Stock volatility affects an option’s volatility, because
greater uncertainty equals greater potential profit or loss.

Volatility collapse: The tendency for volatility tracking to become unreliable


during the final month of the option’s life. During this month, the tracking
mechanisms such as Delta cannot be used to spot changes in volatility.

Wash sale: The rule preventing deduction of a loss when a sale and replacement
occur within 30 days.

Western technical signals: A set of technical price-based indicators used to


identify reversal or confirmation.

Zone of resistance or support: A price range identifying resistance or support,


rather than a single price level.
Bibliography

Augen, Jeff. Trading Options at Expiration (Upper Saddle River, N.J.: FT Press,
2009).
———. Day Trading Options (Upper Saddle River, N.J.: FT Press, 2010).
Bittman, James B. Trading Index Options (New York: McGraw-Hill, 1998).
Cohen, Guy. The Insider Edge (Hoboken, N.J.: John Wiley & Sons, 2012).
Gidel, Susan Abbott. Stock Index & Futures Options (Hoboken, N.J.: John Wiley
& Sons, 2000).
Hull, John C. Options, Futures and Other Derivatives, 8th edition (Boston,
Mass.: Prentice Hall, 2012).
Kobayashi-Solomon, Erik. The Intelligent Option Investor (New York: McGraw-
Hill, 2014).
Kolb, Robert W., and James A. Overdahl. Financial Derivatives, 3rd edition
(Hoboken, N.J.: John Wiley & Sons, 2003).
McMillan, Lawrence G. Options as a Strategic Investment, 4th edition (New
York: New York Institute of Finance, 2002).
Natenberg, Sheldon. Option Volatility & Pricing (New York: McGraw-Hill,
1994).
Overby, Brian. The Options Playbook, 2nd edition (Charlotte, N.C.: TradeKing,
2009).
Rhoads, Russell. Trading VIX Derivatives (Hoboken, N.J.: John Wiley & Sons,
2011).
Sincere, Michael. Understanding Options, 2nd edition (New York: McGraw-
Hill, 2014).
Index

A
Annualized return, 22, 24-25, 47, 49-50, 98, 105-106, 110-112
Ascending triangle, 83-84
At the money (ATM), 17, 56, 104, 143-144, 146

B
Beta, 16
Bollinger bands, 125-126, 165
Breakeven rate of return, 48-50
Butterfly, 143-149

C
Candlestick, Abandoned baby, 92
Attributes, 85
Chart, 78-79
Doji star, 90-91
Doji, 86-87
Double-session, 88-91
Dragonfly doji, 87-87
Engulfing, 88
Evening star, 92
Gap filled, 93-94
Gravestone doji, 86-87
Hammer and hanging man, 87-88
Harami and harami cross, 88-89
Island cluster, 92-93
Limitations, 95-96
Long white and black, 86
Long-legged doji, 87
Meeting lines, 89-90
Morning star, 92
Piercing lines, 89-90
Single-session, 86-88
Spinning top, 87
Tasuki gap, 93
Three white soldiers and black crows, 91
Thrusting lines, 93
Triple-session, 91-93
Capital gains, 8, 21-24, 30-32, 47-49, 96-99, 102-105, 110-115, 157, 174-179,
184, 187, 193-194
CBOE Margin Manual, 27, 192
Charting, 77-96
Chicago Board Options Exchange (CBOE), 21, 27, 143, 192-193
Collars, 9, 37, 101, 124, 155-157
Collateral requirements, 9, 26-27, 191-192
Condor, 149-152
Conservative trading, 57-59
Consolidation trend, 51-53, 71, 79-83, 98-105, 121-130, 136-137, 165-169, 196-
197, 199-201
Constructive sale rule, 193
Contingent purchase strategy, 127-128
Continuation, 80
Contrarian investing, 201
Covered call
Appreciated stock, 113-115
Basic description, 97
Breakeven price, 98
Comparison to uncovered puts, 119-121
Deep in the money, 20, 113-115, 175, 176, 193
Downside protection, 101
Expiration series, 109-110
Hedging with, 97, 101
Lost opportunity risk, 105, 110
Maximum profit, 98
Moneyness, 106-109
Objectives for investors, 102-104
Portfolio strategy, 105
Ratio write, 116-117
Return calculations, 111-113
Rolling forward, 115-116, 173-177
Rules, 105-106
Strategic value of, 8-9
Timing, 103
Unqualified, 113-115, 193
Variable ratio write, 116-117

D
Debt capitalization ratio, 65-66
Deep in the money, 20, 113-115, 175, 176, 193
Delta, 20-21
Descending triangle, 83-84
Dividends,
Combined income, 8, 120, 175, 186
Earnings, 47-49
Fundamental strength and, 37, 59-62
Future payments, 65
Profits from, 30, 32-33, 96-100, 111-114, 189, 194, 200, 202-203
Return calculations, 24-25
Timing, 182
Double top and bottom, 81-82
Dow Theory, 70-73
Dynamic trading range, 197-198

E
Efficient market hypothesis (EMH), 73-74
Exchange-traded funds (ETFs), 40-41
Ex-dividend, 102-103, 105, 109, 138, 143, 153, 155, 182, 184-188, 190
Extrinsic value, 19-20

F
Falling wedge, 83
Fundamentals, 59-66, 202-203

G
Gamma, 21
Gapping pattern, 82-83

H
Head and shoulders, 81
Hedge,
Application to patterns, 199-203
Attitude toward, 102-104
Basic, 29-34
Butterfly, 144-149, 152-153
Candlestick, 95
Collar, 155-157
Condor, 149-153
Conservative, 57-59
Consolidation, 124-127
Contingent purchase, 127-128
Continuation, 80-84
Covered call, 97-99, 104-106
Covered straddle, 128-130
Diversification and, 50-53
Effectiveness, 106-108
Efficiency and, 53-56
Equity, 194
Exercise, 115-117, 176, 183-189
Fundamental, 59
Long call and put, 189-190
Loss, 192
Matrix, 148
No-cost, 179-180
Options-based, 96
Paper gain, 113-115
Perfect, 39, 140, 158
Potential, 109-111
Price effect of, 66-67
Proximity, 195-196
Recovery strategy, 122-124, 177-182
Replacement strategy, 177
Risk, 100-101
Rolling, 173-177
Speculative, 68-73
Spread, 131-143
Straddle, 163-172
Synthetic stock, 157-162
Theory, 74-76
Timing, 197-199
Types, 61-63
Uncovered put, 119-121

I
Implied volatility (IV), 19-20
In the money (ITM), 17, 114, 143-144. 150-152, 171
Intrinsic value, 17
Iron butterfly, 145-149
Iron condor, 150-152
Island cluster, 92-93, 126

L
Leverage, 12
Line chart, 78
Long positions, 14-15
Lost opportunity risk, 33, 45-47, 105, 110

M
Market risk, 43-45
Married put, 38, 194
Moneyness of options, 8, 16, 17-19, 20, 26, 41, 97, 106-109, 192

N
Net basis, 32
Non-price signals, 96

O
Offsetting positions, 194
OHLC chart- 77-78
Option,
American style, 187-188
Attributes, 11-12
Automatic exercise, 183
Call and put hedges, 34-37
Collateral requirements, 9, 26-27
Combinations, 34-37
Early exercise, 183-190
European style, 187-188
Extrinsic value, 19-20
Hedging theory, 74-76
Implied volatility (IV), 19-20
Intrinsic value, 17
Jargon, 15-16
Life span, 26
Long, 14-15, 30-31, 189-190
Offsetting positions, 194
Outcomes, 185
Price (Premium), 14, 16-21
Proximity, 195-203
Return calculations, 21-25
Short, 14-15
Terms, 13, 15
Time value, 18-19, 39-40
Total return, 24-25
Valuation, 43
Volatility collapse, 21
Zone of resistance or support, 196-197
Out of the money (OTM), 17, 51, 56, 108, 143-144, 146, 148, 150-152, 155
P
Pattern recognition, 199-203
Perfect hedge, 39, 140, 158
Pinning the strike, 54
Premium, 14, 16-21
Price/earnings ratio, 62-63

R
Random walk hypothesis (RWH), 74
Revenue and earnings, 63-65
Reversal, 79-80
Reverse iron butterfly, 149
Rising wedge, 83
Risk,
Experience, 48
Hedging, 8, 37-40
Impatience, 50
Inflation, 48-50
Knowledge, 48
Leverage, 12, 45-46
Lost opportunity, 46-47, 105, 110
Management, 9
Market, 43-45
Portfolio, 48-50
Situational, 7
Tax, 48-50

S
Short positions, 14-15
Speculation, 8, 34-38, 40-41, 51-57, 68-74, 127-128, 153, 161-165, 169-171,
175-179, 191-196
Spread:
Box, 136-138
Call bull or bear, 133-134
Diagonal, 131-132, 141-143
Hedging, 131
Horizontal, 131-132, 138-141
Put bull and bear, 132-136
Ratio calendar, 138-141
Vertical, 131-132
Standardized terms, 13
Stock rotation, 40
Straddle,
Buy, 163
Calendar, 171-172
Covered, 128-130, 155, 179, 200
Hedging, 163
Long, 163-165
Short, 165-169, 200
Strap, 169-170
Strip, 170-171
Synthetic stock, 9, 37, 101, 124, 157-162, 179-182, 189

T
Tax rules, 193-194
Taxes and Investing, 193
Terminology, 8
Theta, 21
Time value, 18-19, 39-40
Total capitalization, 65
Trend analysis, 66-68
Triangles, 83-84
U
Uncovered put,
Bollinger band signals, 125-126, 165
Comparison to covered calls, 9, 119-121
Consolidation trend timing, 124-127
Contingent purchase strategy, 127-128
Covered straddle, 128-132
Early exercise, 186-187
Market risk, 32-33, 119
Net basis, 122
Recovery strategy, 122-124, 177-182
Replacement strategy, 121-122
Risk levels, 119
Synthetic, 120,
Timing, 121, 123
Unqualified covered call, 113-115, 193

V
Vega, 21
Volatility collapse, 21

W
Wash sales, 193-194
Wedges, 83
Western technical signals, 81-84
About the Author

Michael C. Thomsett is author of 12 options books. He has been trading


options since 1978 and has worked as a full-time author since 1984. He has also
written extensively on the topics of technical analysis and candlestick charting.
Thomsett is a frequent speaker at The Money Show, Trader’s Expo, and other
investment seminars. He also teaches options courses with Moody’s and other
institutions. The author lives near Nashville, Tennessee.

You might also like