Session 1-14 Strategic Analytics 0821 PDF
Session 1-14 Strategic Analytics 0821 PDF
Session 1-14 Strategic Analytics 0821 PDF
Term IV, 2013 Indian Institute of Management Kozhikode Instructor: Deepak Dhayanithy
Title explained
Strategic
Sustainable competitive advantage, capabilities (Mintzbergs pattern) Key (big) decisions
Industry connotation
Analytics
Deriving actionable insight from data
Home equity <> jingle mail, loan resolution conversation content, recidivism - impact on borrower/ collateral based resolution approaches
Neither purely deductive, exploratory nor inductive Method agnosticism Analytics in various industries focuses on consumer decisions
Learning objectives
1. 2. Take a systematic approach, anchored in theory (agnostic), to thinking about important [strategic] decisions Help students identify problems [cognition] and structure solution approaches
Equip with models, metrics which have been discussed extensively in research literature Critical thinking and reflection important
3. 4.
Introduce students to the use of statistical and OR, OM methods in thinking about high value decisions
Not about how to build the models but appreciate the scope and important steps involved Provide a glimpse of the analytical methodology at work
5.
Provide exposure to the ways in which the better industry participants have long thought through important challenges they are faced with
Course format
The course is organized around four broad themes, which are
Policy and design Ratings, incentives and performance Strategic interactions Fans, Society and Sport
From 3rd session onward, groups (of 5) will lead off the discussions Class presentations and term papers are key vehicles for us to achieve the learning outcomes Readings from Scorecasting (TMLW) and Stumbling on Wins (SOW) Other readings: abstract, introduction, literature review, methods, data, results, discussion
very important that each of you get your hands around amap Key for our classes to be interesting! Books chosen for readability in addition to richness of the analytics As managers, you would be in a position to run such projects in organizations/ deal with external consultants/ be one such consultant framing the project
Important to ask the right questions
Class presentations
Teams of 5 same team for both this and term paper Class presentations (25)
18 TMLW and 9 + 2 SOW chapters Each team to choose at least 1 chapter from each book to present in class, starting from session 3 [June 26th]. Presentation schedule per course outline, sync with Coco and instructor Please discuss presentations with the instructor
Starting at least a week prior to the date of the presentation in class
Presenting groups and the class to contribute in terms of drawing parallels, discussing limitations
SAIS IIMK 2013 1,2 5
Term paper
Teams of 5 same team for both this and class presentations Term paper (35)
Extension of ideas discussed in the readings to other sports, business or experimental contexts Paper to target business or sports management audience Trade publications, sports management conferences grade material is A+ As a collective, showcase sector focus, strategic analytics Shortlisted topics to be discussed with Instructor by 22nd June after mailing a 1-pager on the same
SAIS IIMK 2013 1,2 6
Both the above have outlets that can be targeted, develop collateral Why the NA sports by and large?
Novelty risk
Keeping communication channels open, n-1, COCO
SAIS IIMK 2013 1,2 7
Topics
# 1P 2P 3R 4R 5R 6R 7R 8R 9I 10 I 11 I SAIS content area Competitive balance, open/ closed league, free agency Doping dilemma Performance metrics when less is more (basketball rebounds) Matching law (going for 3) Stress and performance Assessing adjudicator decisions Measuring and valuing wins produced Hot hand phenomenon or fallacy Signaling, bluffing and bargaining Peloton analytics Monopsony and salary suppression
SAIS IIMK 2013 1,2 9
Topics
# 12 I 13 I 14 I SAIS content area Value of roster flexibility Home advantage Bidding for resources coattail effect
10
12
14
Observation distance is important? What does irrational or sub-optimal mean when used to describe decisions, especially those made by experts? Why sports?
Transparency Cost of failure The business is easy to understand
Difficult to be in a fools paradise for long
SAIS IIMK 2013 1,2 15
Not-interfering-with-play interpretations
SAIS IIMK 2013 1,2 16
Oakland As story
Billy Beane, Moneyball, Michael Lewis Why is the 1999-2002 seasons record worthy? Baseline 1997 John Hakes and Raymond Sauer
Workers paid in line with their expected productivity
What performance characteristics impact wins What teams are willing to pay for the characteristics Are salaries consistent with wins
30 MLB teams, 1999 to 2003 sufficient? Statistical analysis, panel data regressions
On-base % 2x impact on wins as slugging Slugging has a big impact on the salaries of players What happened post 2004?
Recall Ron conversation
Baseball
Like cricket
but you can throw only full tosses Bat is just a club Beamers are illegal here too
Run scored when a batter runners over home base Bases loaded difference between a home-run and a grand-slam Strike out (ball the .299 story), caught, walks If a batter hits into fair territory, he runs
SAIS IIMK 2013 1,2 20
Basketball
Indoor, non-contact sport Duration 4 quarters of 12 minutes each Twos and Threes, free-throws Time-outs
2 100 second [1, 3] 3 100 second [2, 4]
23
Better teams (linked to payroll) win because the system makes upsets that much more difficult
law of large numbers?
SAIS IIMK 2013 1,2 24
For the betting types, where would you prefer to wager on the world champion?
SAIS IIMK 2013 1,2 25
What are the important variables that determine whether a league is competitive or oligopolistic? $$$ per team
Is it a policy/ decision variable?
Games/ season # teams in the playoffs One-and-done or playoff series Salary caps Other key elements of the leagues
Free agency Open or closed leagues
SAIS IIMK 2013 1,2 26
Impact of Free agency in NA pro team sports Maxcy and Mondello (2006) Free agency reintroduced in the 1970s
Concern that the star players would congregate in the big market teams
But how do we measure uncertainty or competitiveness? Within season and between season [NBA, NFL, NHL] measures of competitive balance
Standard deviation of winning percentage (SDWP)
Ratio of actual to ideal (adjusts for number of observations more games in some leagues)
Due to economic, structural and policy differences more across league (sport) differences than within league differences, over time
Leagues depending more on gate receipts (MLB, NHL) and other local revenue Bottom line, league can be designed to maintain competitive balance at the level where it makes most economic sense for the league But, let us understand the Free Agency Competitive Balance
SAIS IIMK 2013 1,2 29
1970 *
1977**
1980
1990
2000
2003***
31
NBADWP t-1
Equals the lagged value of Y
NBASRCC
Y2, Spearmans rank correlation coefficient each year from 1951 to 2004
NBURSFA
Dummy var = 1, representing the sample years of unrestricted free agency in the NBA (1978 82)
NBACAP1
Dummy var = 1, representing the sample years in which the NBA imposed a cap on team payrolls (1983 1999)
NBACAP2
Dummy var = 1, representing the sample years in which the NBA imposed cap on team payrolls as well as individual player salaries (2000 2004)
NBARIVAL
Dummy var = 1, representing the sample years in which the league faced an economic challenge from a rival league (ABA: 1967 1976)
NBASTRIKE Dummy var EXPANSION Dummy var NBATEAMS number of teams in the league in a season
SAIS IIMK 2013 1,2 32
rho 0.605 5.5380.508 4.298 ***p<0.01, Adj R sq= 0.059, =0.276 Model I: Y = SDWP t-1 not included, Model II includes the lagged variable in the model [auto correlation] Similar models estimated using SRCC, with and without lagged variables for NHL and NFL seasons as well
SAIS IIMK 2013 1,2 33
Discussion of results
For NFL there is an improvement in SRCC
Delta over seasons
Introduction of the lagged variable in the regressions does not change the results much Direct effect of free agency on competitive balance is ambiguous
Rottenberg, Coase appear to have gotten it right
SAIS IIMK 2013 1,2 35
Entry (into the cartel) to be approved by a majority of the incumbent teams Competition to the league from a rival league only (ICL)
Open league: integrated in a hierarchical structure where within the national federation (Serie A, EPL, Bundesliga, etc.) which in turn is within the continental (UEFA) and global federation (FIFA)
Entry and exit by the mode of promotion and relegation Rags to riches is a possibility good story
SAIS IIMK 2013 1,2 36
Comparisonscontd.
In a closed league, the team enjoys absolute exclusivity over a urban area
If the local area turns unprofitable then the franchise can move 48 relocations in the big-four NA leagues (NBA, NFL, NHL, MLB)
ComparisonsCompetitive balance
Labour market regulations is the main instrument in closed leagues While attempted in open leagues, the open-ness addresses some of the re-balancing demands
Considerable effort expended in staying-up and promotion as well as in changing the rules
1st division teams see a 20-40% revenue bump on qualifying for CL Relegation sees a drop of 75 80% in revenues Promotion sees ~ 5x in revenues
ComparisonsLabour mobility
Labour market regulations is the main instrument in closed leagues
Monopsony, reserve clause (1879 in baseball)
Anti-trust exemption for baseball
Lockouts and CBA to renegotiate prices/ clauses Free agency status for veterans (1970)
While attempted in open leagues, the open-ness addresses some of the re-balancing demands
Recall 2005 CL final (Istanbul)
Steven Gerrards get-out clause Belgian league versus Jean Marc Bosman Quotas based on player nationality went [where is IPL?] Labor choice rules in the European market
SAIS IIMK 2013 1,2 39
Rookie draft
Reverse-order-of-finish in NA, with roster limits Bosman deregulation and EU competition policy applies player agents role more pivotal
CBA
Player working conditions
no: games in a week
40
Ownership
Objective function
Closed league: struggling franchises objectives turn financial (do NBA teams tank?) Open league: struggling franchises objectives have to remain sporting Tend to under-invest in sporting talent, its development Tend to trigger off arms races
Chelsea, Manchester City in the EPL FCB, RMA in Spain Financial Fairplay implementation and its effects?
Borussia Dortmund
SAIS IIMK 2013 1,2 41
Investment decisions
Term paper
Teams of 5 same team for both this and class presentations Term paper (35)
Extension of ideas discussed in the readings to other sports, business or experimental contexts Paper to target business or sports management audience Trade publications, sports management conferences grade material is A+ As a collective, showcase sector focus, strategic analytics Shortlisted topics to be discussed with Instructor by 22nd June after mailing a 1-pager on the same
SAIS IIMK 2013 1,2 42
Key Ideas
Competitive balance
Where else would we be interested in maintaining it? Why?
A complex vendor management problem as the one faced by WHO in its global procurement of vaccines Why maintain competition? How should competition be defined?
Rookie drafts, reserve system Free agency Salary caps SAIS IIMK 2013 1,2 Promotion and relegation
43
Avoiding Relegation
From league to club perspective Fairytales are rate Bottom echelon problems are different from Top echelon problems
So are resources
Relegation battles are dramatic Any clues for the clubs in this position?
AIB, Istanbul
44
Avoiding Relegation
Table 1: Summary of clubs promoted to EPL 1993-94 to 2011-12
# SEASONS SINCE PROMOTIO N 1 2-3 4-7 10 + Grand Total NOT RELEGA % TED TOTAL RELEG 3 27 89% 2 11 82% 3 11 73% 2 7 71% 10 56 82%
RELEGA TED 24 9 8 5 46
45
FIRST TIME PROMOTION (0.2 0.34 0.42 1.00 0.52 0.80 (H3) 2) RATIO PURCHASE $ TO LEAGUE AVG (Control) (1.0 0.41 6.13 1.00 0.01 0.37 0)
46
Factors that can shift this baseline curve up or down are those of interest to the club management!
Figure 2: Survival (from relegation) at the mean of covariates
47
Announcements
Groups
Okay, 1 group to give their choice of topics
Please discuss your presentations t 1 (24 hours) at least pretty packed classes schedule
SAIS IIMK 2013 1,2 48
Discussion points
Wider applications of competitive balance discussion
Where to recruit first?
50
51
Mitchell Report
Utility infielders jacking 30 home runs in a season 249/ 274 who tested positive were minor league players Welington Dotel CBA of doping!
Relative likelihood of PEDs use X GDP per capita Age profile of PEDs positive tests Repeat offenders Variance even in US based on ZIP code groups Population ecology of the stigma SAIS IIMK 2013 1,2, 3
52
CBA contd.
Positive effect on
Surviving Promoting
More pronounced for baseball athletes from poorer countries Testing protocols matter
Baseball more at the lower levels
Similarities to
Tax filing CV dressing
53
Shermer
Doping Dilemma is the same as the prisoners dilemma athletes instead of prisoners
MY OPPONENT'S STRATEGY
DEFECT (confess)
Three years in jail (Sucker Payoff) Two years in jail (Low Payoff)
54
Model to Reality
Athlete asymmetry
Skill Career stage Economic stakes
Competition structure Who are the players, what are their incentives Are veterans same as debutants
55
Willy Voet
Willy Voet, the soigneur, or all-around caretaker, for the Festina cycling team in the 1990s, explained how he beat the testers in his tell-all book, Breaking the Chain: Just in case the UCI doctors arrived in the morning to check the riders' hematocrit levels, I got everything ready to get them through the tests I went up to the cyclists' rooms with sodium drips-- The whole transfusion would take twenty minutes, the saline diluting the blood and so reducing the hematocrit level by three units--just enough. This contraption took no more than two minutes to set up, which meant we could put it into action while the UCI doctors waited for the riders to come down from their rooms.
56
Joe Papp
a 32-year-old professional cyclist currently banned after testing positive for synthetic testosterone. Recalling the day he was handed the "secret black bag," Papp explained how a moral choice becomes an economic decision: "When you join a team with an organized doping program in place, you are simply given the drugs and a choice: take them to keep up or don't take them and there is a good chance you will not have a career in cycling."
57
Frankie Andreu
superdomestique, or lead pacer, supporting Lance Armstrong throughout much of the 1990s. "Then, around 1996, the speeds of the races shifted dramatically upward. Something happened, and it wasn't just training." Andreu resisted the temptation as long as he could, but by 1999 he could no longer do his job: "It became apparent to me that enough of the peloton [the main group of riders in a cycling race] was on the juice that I had to do something." He began injecting himself with r-EPO two to three times a week. "It's not like Red Bull, which gives you instant energy," he explained. "But it does allow you to dig a little deeper, to hang on to the group a little longer, to go maybe 31.5 miles per hour instead of 30 mph."
58
Frankie Andreu
superdomestique, or lead pacer, supporting Lance Armstrong throughout much of the 1990s. "Then, around 1996, the speeds of the races shifted dramatically upward. Something happened, and it wasn't just training." Andreu resisted the temptation as long as he could, but by 1999 he could no longer do his job: "It became apparent to me that enough of the peloton [the main group of riders in a cycling race] was on the juice that I had to do something." He began injecting himself with r-EPO two to three times a week. "It's not like Red Bull, which gives you instant energy," he explained. "But it does allow you to dig a little deeper, to hang on to the group a little longer, to go maybe 31.5 miles per hour instead of 30 mph."
59
Sanction rate
Ex hibit 3 : Te s ts and Sanctio ns , U SA D A 2 0 0 1 -'1 2 , N o : o f athle te s (2 /2 ) Santio n rate * # A thle te s Po s t Spo rt te s te d Pre 2 0 0 6 2 0 0 6 To tal
P araly mp ic Row ing P araly mp ic Rugby P araly mp ic Sailing P araly mp ic Shoot ing P araly mp ic Sled H ockey P araly mp ic Soccer P araly mp ic Sw imming P araly mp ic T able T ennis P araly mp ic T ennis P araly mp ic T rack & F ield + P araly mp ic Volley ball Racquet ball Roller Sp ort s + Row ing + Rugby Sailing Shoot ing Skiing & Snow boarding * Soccer Soft ball Sp eedskat ing Squash Sw imming * Sy nchroniz ed Sw imming T able T ennis T aekw ondo + T eam H andball T ennis T rack & F ield * T riat hlon + Volley ball Wat er P olo Wat er Skiing Weight lift ing + Wrest ling 18 40 26 6 43 42 124 15 30 164 49 29 166 502 34 182 202 636 299 158 311 29 1,065 76 40 201 190 53 2,837 368 193 115 79 607 292 0.0% 0.0% 0.0% 11.8% 0.0% 0.0% 0.0% 2.9% 0.0% 0.0% 1.4% 0.0% 0.0% 0.0% 1.6% 0.0% 1.0% 0.0% 0.0% 1.9% 3.2% 0.0% 2.2% 1.0% 0.0% 3.1% 0.0% 0.0% 0.0% 0.0% 1.8% 2.8% 0.0% 3.3% 0.0% 0.0% 3.8% 0.0% 0.0% 0.0% 0.0% 5.4% 0.0% 0.0% 2.1% 0.9% 0.0% 0.0% 1.8% 0.3% 0.8% 0.0% 0.5% 0.0% 0.7% 0.0% 4.3% 3.7% 1.1% 0.0% 1.2% 0.9% 0.0% 1.4% 0.0% 3.2% 2.3% 0.0% 2.5% 0.0% 0.0% 7.0% 0.0% 0.0% 0.0% 0.0% 4.9% 0.0% 0.0% 1.8% 0.6% 0.0% 0.0% 1.0% 0.8% 0.3% 0.6% 0.3% 0.0% 1.2% 1.3% 2.5% 3.0% 1.1% 0.0% 2.0% 0.5% 0.0% 0.9% 0.0% 2.5% 2.4%
Grand To tal
1 4 ,7 8 3
1 .7 %
1 .4 %
1 .5 %
61
Go for it
What is fourth down? What does 3 and 7 mean? What dont coaches go for it (why punt)? Job security Loss aversion Other sporting contexts where loss aversion is key
bubble Keeping the closer for the 9th Keeping Malinga for the 20th
SAIS IIMK 2013 1,2, 3
62
64
Matching Law
Log (B1/B2) Log (B1/B2)
Quasi-experiment
3-pt circle movement gives three eras
1991-94 1994-97 (Brought-in) 1997-00 (Moved out)
X log (3pt scored/ 2pt scored) Y log (3pt att./ 2pt att.) 1991 94
Y = 0.799x + 0.001 [R sq = 0.961]
1994 97
Y = 0.826x + 0.046 [R sq = 0.972]
1997-00
Y = 0.871x + 0.047 [R sq = 0.961]
SAIS IIMK 2013 1,2, 3 67
Quasi-experiment contd.
Mean log relative reinforcement ratios
-1.247 (91-94) -0.797 (94-97) -0.797 (97-00)
68
Soccer
Jabulani introduction
69
SAIS IIMK 2013 1,2 Monopsony and salary suppression Executive pay regulation
72
GVT, 1985
Plethora of sports work
Stats Inside views
When soccer goalies will stop anything Hero calls and tells
73
If skill, streaks
Hot hand
People dont (do) pay much attention to the underlying process when outcomes alternate frequently (infrequently) Greater confidence in subsequent expectations after a series of correct guesses
SAIS IIMK 2013 1-5 74
75
Overestimate the rarity of streaks under the null hypothesis (of randomness) Signal Detection Theory
Hits
Correctly rejecting the nh that data are random
Correct rejections
Correctly retaining the nh that data are random
False alarms
Mistakenly rejecting the nh that data are random
Misses
Mistakenly retaining the nh that data are random
SAIS IIMK 2013 1-5 76
Note that many real world processes are nonrandom/ not purely random
SAIS IIMK 2013 1-5 77
HH and GF in practice
What should be the basis for call routing in a Insurance companys inbound call center with spiraling churn?
Metric for call-type <> agent mapping How real time?
78
Nerves of Steel
Two decades of shootouts in the world cup and Euro Stress factors impact on performance
Predictable, anticipated and expected
Crowd, level of the competition, kick sequence
Nerves of Steel
Tortoises fare better than hares
80
SAIS IIMK 2013 1,2 Monopsony and salary suppression Executive pay regulation
83
Blocked shot
Baskets, assists, blocked shots, steals Are all blocks the same? If not, how are they different? Is the difference only a fine one?
84
Going for 3
Field goal kicking in football Who is good at field goal kicking? Measures Issues
Difficulty level and success rate Team/ coach/ game situation and difficulty level
85
86
Situational/ psychological
87
88
Sense-making
89
So what?
Kicker rankings
Difficulty adjusted Added points per attempt
90
So what? contd.
Psychological factors not so important Classifying attempts accurately into hits and misses
Probability cutoff decision Support Vector Machines
Problems
Observational, relatedness of variables Environmental vars at game beginning
Could be more dynamic
91
Measurement issues
Conditions at kick-off Multicollinearity Only distance?
Now video position tracking is possible
92
Limit exposure
To reduce risk capital by $x Who will default perhaps not enough
Authorizations
Marginal $ to be saved
93
Model scrutiny
Concerns
Privacy Fair lending Consumer protection
Efficacy
Recall Prof. Choudhurys talk
94
Cricket
Early mid career selection choices All time lists A dismissal is a dismissal, or?
Opponent batting lineup Home/ away Pitch conditions DRS or not Bowling first, last New ball Bowler type pre or post-Arm extension interpretations
SAIS IIMK 2013 1-5 95
Cricket contd.
Catches Stumpings End of the day question
Would the rankings change with difficulty adjustments? Would our decisions in credit change with difficulty adjustments?
96
$20K
SAIS IIMK 2013 1-5 97
$7.5K
SAIS IIMK 2013 1-5 98
100
101
103
NFL focus on the QB (RELWP100) versus NBA PAWS48 Significant wins produced stats in both sports? What does it mean?
104
106
Questions
Implications of assembling a team of players with high PAWS48 Given PAWS48 measures, what are the areas to look into now? Differences between NBA and NFL wins metrics (for QBs)?
107
108
Paired Pitching
Greg Rubin
Average MLB starter faced a batter 2.8 times in the 2009 season
IP x BF per IP / 9 Starting pitching talent is expensive Batters do better once they see the pitcher
Fatigue, release point
Analysis of covariance
Study interaction between TF and RA ANCOVA on RA using TF, BF for a sample of 83 pitchers standardized residuals for each observation 19.9 RA > 24.8 > 25.5 > 13.3 (why low suddenly?) 75% variance in RA explained by variance in TF, BF
SAIS IIMK 2013 1-5 109
Paired Pitching
Greg Rubin
110
Sport application?
Specialists versus bits and pieces bowlers Penalty corners
111
Elsewhere?
A realtor needs to decide the sequence in which his best salesmen would schedule property inspections with high value customer prospects Salesman problem/ Retail problem
Retailing mixers Of course, the business and the customer has a more congenial relationship
Methodology issues
Endogeneity
Is the n-th play independent of the earlier play?
113
114
Word on probability Games that progress through rounds of revealing cards, of betting Poker
Stud Draw Holdem
Discussion points
Is this a fair game? Two ways of making v1 game fair?
Manipulate
Odds Information dynamics
116
117
Discussion points
Is this a fair game? Two ways of making v1 game fair?
Manipulate
Odds Information dynamics
118
Applications, isomorphism
Settlement between plaintiff and defendant
Defendant offers a generous (fold) or stingy (bet) settlement
Plaintiff folds (accepts the stingy settlement) or calls (rejects the stingy settlement)
119
BargainingInitiating
Tournaments
Increasing blinds and antes
Final table Which processes of bargaining succeed? Who initiates them? Why online tournaments?
120
123
124
Executive pay regulation what can be learnt from major league sports?
Salaries of sports professionals discussed widely, data available
125
Executive pay regulation what can be learnt from major league sports?
Executives and athletes pay
Similar in $ amount Significant differences in the pay structure
Executives
Variable pay dominated Bonus, stock grant, stock options, long-term incentive payouts (60 80%)
Athletes
Mainly fixed, small variable component (~25%) Performance linked, extra-ordinary performance linked
SAIS IIMK 2013 1-5 126
127
130
SDP - 2
Team-up and play 10 hands of SDP (with reraising) From 11th hand onward Antes double There are some extra things that the players can do
131
SDP - 2
Did either of the players negotiate? Who began the negotiation? When did the negotiation begin? Relative position when negotiation began? Was an agreement reached? What were the terms and conditions? Who benefited?
132
Insane volumes
> 1,246 for 1Q, 1 research, 1 range of stakes, 1 room Average $80K per tournament
SAIS IIMK 2013 1-5 133
Play paused to accommodate chat for the deals Aggregators provide player (only nicks of course) stats
Quality, skill level can be researched
134
Initiating Bargaining
David Goldreich and Lukasz Pomorski, 2011, Review of Economic Studies
Data
Table 1: p. 1303 Only a subset of tournaments played on this single site in 1Q of 2007
135
136
137
Why bargain?
David Goldreich and Lukasz Pomorski, 2011, Review of Economic Studies
138
139
140
141
142
143
Summary
David Goldreich and Lukasz Pomorski, 2011, Review of Economic Studies
144
Summary
David Goldreich and Lukasz Pomorski, 2011, Review of Economic Studies
145
Summary
David Goldreich and Lukasz Pomorski, 2011, Review of Economic Studies
Delay in bargaining Identity of initiator Initiators effect on completion of bargaining Deals more likely when initiator in stronger position or experienced/ highly ranked Initiators effect on terms of deal No effect. Leader sacrifice, guy in the middle gets squeezed Characterist ics of the environment conducive to bargaining Larger gains to trade; high risk; equality of outside options; few remaining players; few experienced players
Empirical results
Initiator typically in weaker position; initiator less likely to be experienced or highly ranked
Theoretical lenses complete information bargaining, cib with learning, cib with biases (overconfidence), incomplete information, dissolving a partnership (Cramton et al., 1987)
SAIS IIMK 2013 1-5
146
149
Does a coattail effect influence the valuation of players in the MLB drafts?
Kevin Whitaker, 2013, 7th Annual SSAC
Coattail effect
In elections In drafts
Schools home to top talent see better draft performance at 6+ rounds But, could simply be that top talent plays in good setups
How do you take care of this? Our friend, the FE model could help
150
151
152
2.66 players drafted from a given school in seasons with a possible coattail effect, only .43 otherwise
SAIS IIMK 2013 1-5 153
154
Assumptions
Teams always choose the player they feel to be the best available prospect in the draft But younger players may have more leverage Sums demanded not known unless deal is done
155
156
Coattail Conclusion
Six, Seven figure investments inlottery tickets Additional .76 players drafted from a school in a year where there is a top prospect (.43 Avg) Effect seems to have gotten stronger between 1984 and 2003 (?) But players who were teammates with top players actually outperformed Most draft eligible college players recruited through concurrent programs
SAIS IIMK 2013 1-5 157
158
Estimate
Likelihood and duration of player injuries through a season
Fielding abilities at secondary fielding positions Robust optimization model to measure degradation of team performance due to injuries Measure difference in performance between teams with and without positional flexibility
SAIS IIMK 2013 1-5 159
Player j plays position i with a capability value of vij vij fraction of all innings in a season that player (factories) j is assigned to position (products) i J players, I positions Player j can play up to cj innings per season Each position must be assigned a player up to di innings per season, which can be further divided into L and R Constraints to ensure catchers play at most 85% of the innings per season
SAIS IIMK 2013 1-5 160
Assumptions
Capability values are additive across players Injuries which determine a players capacity values happen before player-innings assignment Innings in which players are injured is controlled by the team - ???
161
Augment basic model to allow an extra decision (by nature) to determine the worst case combination of player injuries I: budget of disruption given to Nature
162
163
164
165