Pitchforks

howto
Why would I want one
Drawing them
Reading them
Trading them
 Why do I need a pitchfork

Identify the market context ( up, down,

consolidation)
Identify the market limits (elasticity of price)
identify the market pivots, and the context of
the pivot
provide a framework around price movements
can provide clear trading signals
can provide stop loss and profit levels
can highlight difficult to detect patterns
DRAWING THE PITCHFORK
The standard pitchfork
P(1) and P(2) pivot points, the Median Point
P(0).. the origin point extended to the Median Point
The Support and Resistance tynes

The Schiff fork alignment

The "modified" Schiff fork
The Gap fork alignmment

adding internal divisions

warning Lines
trigger Lines
Sliding Parallel Lines
First select a couple of likely swing points, if you need to,
zoom out to filter out the noise. This is a weeky of the XJI.
P1 is a high, P2 is a low.
Then find the middle of the range for those 2 points. This point is
called the "MEDIAN POINT".
One easy way to find the middle, is to first join the swings with a
line, like this blue one.
Then add a box to the 2 swing points, as I have done in black.
Draw a line between the other 2 diagonals, as I have done in red.
The crossing of the red and blue lines will be the "median point".
Now we use the swing prior to P1 to establish the trend within
this time frame.
We add the "MEDIAN LINE" from the prior swing point through
the median point. Notice how price seems to continually revisit
the median line, and use it as a barrier.
Next we want a price framework, so we draw 2 lines parallel to
the ML from the main swing points
These 2 lines are called the Resistance and Support tynes,
and are sometimes referred to as MLHs. ML being short for
Median Line, and H being Dr Andrews symbol for 2 parallel
lines. Price will tend to reject from the parallels, and move
towards the median line.
The upper parallel is often referred to as the U-MLH, and the other, the L-MLH.
According to Andrews if price does not penetrate the outer MLH, there is an 80%
probability that it will return to the ML. It is also quite normal for price to cross the
ML many times, before heading to the outer ML, usually after repeated testing
without crossing from one side.
There are several different alignment possibilities for median lines which are
used by specialists, usually when here has been a rapid vertical move. All the
variations are based on placement of P0. The remainder of construction is
identical to the standard pitchfork. Their construction is set out in comparison
format below. Dologa's version of the modified Schiff fork (yellow), and the T fork
(black) are only used by Dologa. The orange modified Schiff is the most
frequently used. It appears mostly in FX.
This is the construction of the original Schiff fork preferred by software
vendors, probably because from a programming point of view, the new
P0 is easy to calculate as "halfway along the trigger line"
The software vendors construct the "modified Schiff" fork by moving
the P0 to the mid-point of the vertical originating from the standard P0
Dologa seems to have a variation on this where he constructs the origin
at the mid point in the swing before the standard P0
Dologa also has variations where the mid-point between the highs, or
the lows can be used, as shown in the diagram. Personally I have
never found it necessary to use any of the variations, but I can see that
the software vendor's modified schiff implementation works in many
situations
A Gap Fork can have it's origin pinned anywhere in a gap. Basically the
rules allow you to simply try to link the tops to the U-MLH or the lows to
the L-MLH. In practice, I use gap forks, and I almost always place the
origin at the low above the gap , but this is my own preference, and is far
more consistent than the 'rules' require.
Using NUF, I picked a high/low then rotated the handle till things lined up and
sure enough the handle landed at the top of a gap. I think this fork highlights the
trend pretty well, and it defines the limits, within which price is range trading,
while behaving somewhat erratically. Basically the gap fork is a tool which you
should know how to use with volatile equities.
Here I have overlayed another gap fork which would not be immediately
apparent until the gap down late in the piece and the attempt to break back up all
seem to want to respect the Lower MLH
 ADDING
INTERNAL
SUPPORT AND
RESISTANCE LINES
There can be internal support and resistance levels within a fork. This is the SP500, and you
can see that price is bouncing along on the ML. The concept is not really readily visible until I
include a line showing the 23.6% deviation from the ML to the outer tynes.
Once added, the 23.6% variation seems to have acted as resistance for the
first few attempts to rally up from the ML, but that does not explain the 2
emergent rallies, so I will add in the 38.2% variation from the ML
Once I show the 38.2% variation, you can see that this contained the 2
emergent rallies, and that they were followed by a weaker rally which could ony
manage to make it to the 23.6% fib. Was that a sign of the Intermediate Term
rally weakening?.... or was it the result of the fact that price crossed the ML
downwards for the first time since July? Maybe...
The current XJO with no lines on it, does not provide much in the way of
a framework in which to make a decision, so if we take the March low
as P2, then go to the previous high as P1, the obvious choice for P0
would be the November low
So we add a pitchfork. It doesn't really seem to add much does it. We
see the "vibrations" across the ML, a couple of potential breakouts, and
that last rise testing the resistance tyne, but it does not really tell us
much, does it. How about we try a couple of 50% divisions off the ML.
That actually seems to add something. I can see the early rallies limited
by the blue line, then the later rallies supported off it. But it does not do
anything about the little breakouts. So how about we try adding some
external lines
The maroon line is 138.2% from the ML to the outer tyne. It seems to have capped
the breakout rallies pretty neatly. It also seems to highlight how price seems to
want to demonstrate support on the back test of the resistance tyne. But what if
we were to look at another fork, internal to the green fork?
Here I have added the green fork to introduce you to the concepts of drawing
internal forks. It fits pretty well, doesn't it? It certainly gives me lots of information,
and a good framework to work with. If you look closely at where I anchored P0, it
is in that little gap near the March low. Some forks, particularly those near a rapid
move, have their P0 anchored in a gap. Remember gap forks?
And where you can add an extra fork once, you can do it again, to give you
better direction in a shorter time frame. In this case I have added the ST red fork
and I have put 75% divisions in it, to keep the GANN people happy. Anything
with a shorter time frame from this, and you should switch time frames for the
chart down to 4, 2 or 1 hour. The same technique will apply down to 1 minute
charts
 ADDING
WARNING LINES
Warning lines are a special kind of external parallel which is always drawn as an
extra channel. They are used to predict the potential limit of a breakout move
which goes "1 channel higher".
Every now and then we come across a move that requires two warning lines. This
particular warning line puts a potential top on JBH at $23-ish.
Here I have added 3 warning lines to KAR in both directions.
KAR likes to walk along parallels sometimes
TRIGGER LINES
The Trigger Line is drawn from P0 to the either the P1, or P2. In this case
it is P2 and is used to predict a catastrophic style of decline. A normal correction
will rebound off the trigger line to form a new swing point. If the trigger line is
broken, the subsequent decline will usually be rapid and dramatic
Trigger lines can also be used to predict breakouts. In this case you can see the breakout
from the red fork then the parabolic break when it got above the trigger line, then the back
test of the trigger as it fell back through, before the much bigger collapse
So If we draw the next fork on AWB, and its trigger line, we map the rally, then the
breakdown and the back test of the support tyne, then the test of the trigger line, the break
and the collapse. Handy things trigger lines!
 The
sliding parallel line
Sliding parallel lines are used when price fails to reach the outer
tyne before reversing direction and crossing the ML. Quite often
the point of reversal at the opposite tyne will be exactly the
same amount by which price failed to reach the first MLH. It is
as if the entire channel structure has been shifted to meet the
effective support/resistance point
This is the ASX200 in December 2009 with the sliding parallels
drawn in.
This is the FTSE, over the same period, with the sliding
parallels drawn on the ML and the first warning line
Common Patterns

TEST AND RE-TEST

TEST AND BACK TEST
GAPPING THE MEDIAN
ZOOMING THE MEDIAN
The TEST/RETEST PATTERN is the single most useful pattern.
When you see it you can have added confidence that you have
drawn the fork correctly
This is the same pattern on a downward facing fork

TEST/RETEST
DOWNWARD FORK
Back testing is a pattern which confirms that you are not the
only person who sees the line. It confirms the existence of the
line and it's alignment
This graphic shows the template for a gap across the ML in both
directions. Gapping across the ML gives you confidence that an
area of high energy has been crossed, with enthusiasm.
Sometimes you get the back test of confirmation 3 or 4 days
later
Zooming the ML is commonly seen where there is a big move
afoot, but the ML is not gapped. Normally zoom bars come in
sets of 3, both up and down, and cross the ML, not the outer
tynes.
The Test/Retest pattern is the single most important trading pattern. The entry is to look for
the candle with the long upper shadow which touches the resistance tyne. Safest short entry
is at the close of trade, stop placed above the high of the original test candle. Target is the
ML, but a trailing stop is used in case the target is not reached. The retest often occurs at
3.25 trading days after the original test.
This slide is to demonstrate how common the pattern is, why you use a trailing stop to exit
the trade, and why you short a downwards facing fork (and go long on upwards facing
forks)
ILU with a penetrating test of the resistance tyne. The long upper candle and the close at or
below the resistance tyne is what you are looking for. Enter on the close and place your stop
above the swing point. With a penetration shadow this long, maybe above that (reduce
position size). Here I have shown how to use yesterday's high as the trailing stop. A better
strategy is to use a high from before 3.25 days ago, using an intraday chart.
This illustrates the breakout play, where you wait for the back test. If the re-test bar
penetrates the line it's OK, you are looking for a close at or above the bar, or an open next
day above the bar. This slide conveniently shows gapping behaviour. Put the stop below
the gap.
This slide of MQG was taken on 25/11/2009 after the close to illustrate a trade as it
appeared at the time. This provides the context where the red fork provided a test/retest
trade which should have been stopped out, but the green ST fork has now presented a
long trade.
This is the close up view. You can see the candle where the retest of the green support tyne
occurs, with a long probing shadow. The entry is on the close, although an initial position
could have been taken on the tyne, and a pyramid on the close (and possibly another
pyramid if the next open was higher) . The stop is positioned below the pivot point low, the
target is the red resistance tyne, or the green ML, probably both. Risk is $1, reward is $5.
Here we see a sliding parallel. These are typically seen where the retest pattern fails to
reach the tyne. They are drawn parallel to the tyne at the limit of movement and will
commonly be tested again. The stop should be placed outside the tyne, but that's really up
to you. This one also shows a test/retest on the pink resistance tyne, and several ML
reversals, which are more risky to trade.
This is MPO. If you study this slide you should be able to identify nearly
every pattern including at the end, zooming the ML followed by a
breakdown of support on one fork while confirming support on another
 Some "STUFF" you should keep in mind

If price does not breach the boundary tynes of the fork,

there is an 80% probability that it will re-test the Median
Line
If price cannot reach and test the median line, it will go in
the other direction to test the outer tyne, and will have a
much higher probability of breaching the outer tyne.
If price does breach the boundary of the fork, it is time to
draw a new fork to tell you what the new framework for
price movement is. Normally we would use the old P2 as
the new P1 and the old P1 as P0.
If price breaches the outer boundary, we would normally
draw a "warning line" from the old fork. This is an extension
of an extra channel on the old fork.
Some
Interesting
Examples
This is an example of a median line being gapped in both
directions, producing an island reversal pattern. Look for gaps
across median lines. It is a sign the fork is correctly aligned.
As an altenative to gapping behaviour, look for the zoom, a
rapid move across the ML with a minimum of 3 longer than
normal bars
Look for the test/retest pattern after a tyne is first formed. The
more a tyne is tested, the more confident you can be in it. This
pattern can frequently yield very rewarding trades.
Some extra lines and
some extra forks can
help
This is the CAC40 over 10 years. Note how it followed the support tyne of the
blue fork, but never reached the median line despite several attempts to rally
towards it. It eventually went the only way it could and tested the support tyne.
Note how it gapped across, then back tested the old support tyne. This gapping
behaviour is very common when breaching a line. It is one of the signs you look
for.
In this view of the CAC I have added the first warning line, which is an extra
channel added to the fork in the direction of the breakdown. It approximates
where you can expect price to go. In this case there was a small temporary
breach, then a reversal and re-test. Remember the trigger line, it may have
been a rejection off a trigger line as well
In this slide I have added a downwards facing fork in pink, note that price has not reached
the pink/purple median line but has reversed, probably off the warning line from the blue
fork. It will now probably test the outer pink tyne.

To get an idea of the probable path, we could put in an extra fork or two, so we might as
well zoom in to an 18 month time frame
Here I have left the edge of the old blue support tyne in view, and the 1st warning line at the
bottom. I then picked a couple of low/high combinations to go with a P2 of the absolute low.
As you can see, the forks are strangly parallel. This gives me quite a bit of confidence that we
are correctly framing the trend of price. I notice how the green fork is tending to keep price
within it's resistance tyne, but that it seems supported by the ML of the slightly longer time
frame pink fork. There is, of course, nothing to say there cannot be several concurrent
influences on price.
 ADDITIONAL READING

Integrated Pitchfork Analysis: Mircea Dologa

Andrews original notes

http://le-blog-de-reklats.over-blog.com/article-29976882.html

Tim Morge's website

http://www.marketgeometry.com/

My email
[email protected]