O'Bannon, Jr. v. National Collegiate Athletic Association et al

Filing 186

RESPONSE (re #179 MOTION to Strike the Declaration of Daniel L. Rubinfeld ) filed byNational Collegiate Athletic Association. (Attachments: #1 Exhibit A, #2 Proposed Order)(Pomerantz, Glenn) (Filed on 6/6/2014)

Exhibit A

UNITED STATES DISTRICT COURT
NORTHERN DISTRICT OF CALIFORNIA
OAKLAND DIVISION
In re NCAA Student-Athlete Name and Likeness
Licensing Antitrust Litigation

Case No. 09-cv-1967-CW

DECLARATION OF DANIEL L. RUBINFELD
June 3, 2014

HIGHLY CONFIDENTIAL–COUNSEL ONLY: SUBJECT TO PROTECTIVE ORDER

1. I am the Robert L. Bridges Professor of Law and Professor of Economics Emeritus at
the University of California, Berkeley and Professor of Law at New York University.
This declaration contains additional evidence supporting my opinions on competitive
balance, previously expressed in the Rubinfeld Merits Report and the Rubinfeld
Merits Rebuttal Report.1
2. As I explained in my two prior reports, competitive balance is an important driver of
consumer demand for sporting contests.2 I pointed out that the NCAA has achieved a
reasonable level of competitive balance, as demonstrated by the popularity of D-I
men’s basketball and FBS football and in comparison with professional leagues.3
3. There are a variety of measures of competitive balance used in the sports economics
literature, including measures of game or match uncertainty.4 One category of game
uncertainty measures involves the dispersion of winning percentages, a commonlyused variant of which is the ratio of the standard deviation of winning percentages to
the standard deviation of outcomes resulting from a hypothetically balanced league
(hereinafter the “standard deviation ratio” or “SDR”).5 The SDR captures the extent
to which teams in a league differ in their success in winning games, where the
measure of competitive balance is lower the more similar the winning percentages of
the teams.6 If the value of the SDR is equal to one, then actual balance is equal to the
hypothetically balanced league, and if the SDR exceeds one, the actual balance is less

1

2
3
4

5

6

Expert Report of Daniel L. Rubinfeld Regarding Merits, September 25, 2013 (“Rubinfeld Merits
Report”), Section VII; Expert Rebuttal Report of Daniel L. Rubinfeld Regarding Merits, November 5,
2013 (“Rubinfeld Merits Rebuttal Report”), Section VI.B.
Rubinfeld Merits Report, ¶ 88; Rubinfeld Merits Rebuttal Report, §VI.B.
Rubinfeld Merits Report, ¶ 89; Rubinfeld Merits Rebuttal Report, ¶ 235.
See, for example, Fort, Rodney, “Competitive Balance in North American Professional Sports,” in
Handbook of Sports Economics Research, John Fizel, ed., M.E. Sharpe, Armonk, NY, 2006 (“Fort
(2006)”); Szymanski, Stefan, “The Economic Design of Sporting Contests,” Journal of Economic
Literature, Vol. XLI (December 2003) (“Szymanski (2003)”).
See for example: Quirk, James and Rodney D. Fort, Pay Dirt: The Business of Professional Sports,
Princeton University Press, Princeton, NJ (1992) ("Quirk and Fort (1992)"), Chapter 7; Fort (2006);
Szymanski (2003); Baird, Katie, “Dominance in College Football and the Role of Scholarship
Restrictions,” Journal of Sport Management Vol. 18, No. 3 (2004), pp. 217-35 ("Baird (2004)").
Dividing by the hypothetically balanced league standard deviation provides a normalization which
facilitates cross-league and/or cross-season comparisons. See Quirk and Fort (1992), p. 245.
∑
Formally, the measure is calculated as
/
0.5 / / 0.5/√ , where wi/G is team
i’s win percentage, T is the number of teams, and G is the number of games played by each team.

HIGHLY CONFIDENTIAL–COUNSEL ONLY: SUBJECT TO PROTECTIVE ORDER Page 2

than the hypothetically balanced league.7
4. I have calculated the SDR for the two NCAA sports at issue, FBS football and D-I
men’s basketball and I have done similar calculations for the NFL and the NBA for
the years 2009-10 to 2013-14.8 The results are summarized in Exhibit 1. For FBS
football and D-I men’s basketball, I have calculated SDR in two ways. First, to be
consistent with the literature,9 I calculated the SDR separately for each conference,
and then averaged across conferences. The results are shown in the first and fourth
rows of the exhibit. Second, I calculated the SDR for the all of FBS football and all
of D-I men’s basketball to account for cross-conference competition. The results are
shown in the second and fifth rows. It is clear from the exhibit that FBS football
balance measures are generally comparable to those in the NFL, while D-I men’s
basketball is always closer to hypothetical balance than the NBA.10

This supports

my previous conclusion that the NCAA has achieved a reasonable level of
competitive balance in D-I men’s basketball and FBS football, as compared to
professional basketball and football.
Relationship between Team Financial Resources and Competitive Performance

5. As I explained in the Rubinfeld Merits Rebuttal Report, there are lower-revenue
programs from non-major conferences which have been successful, that would likely
be particularly adversely affected in Plaintiffs’ but for world. I provided examples
including Gonzaga basketball.11
7

8

9
10

11

The standard deviation for the perfectly-balanced league is calculated by assuming a binomially
distributed random variable with a probability of success of 0.5, which can be shown to equal 0.5/(G0.5)
where G is the number of games each team plays. In other words, perfect balance assumes that the
probability of a win is 50% for every team in every game. Note that this is an ex ante concept. It is
possible, although highly unlikely, that all teams would have 50% win record ex post, so that the
numerator of the ratio, and thus the SDR will equal zero.
The raw data for my analysis is the annual win-loss records for each team, available at "College
Football," available at <http://www.sports-reference.com/cfb/conferences/>; "NFL Standings,"
available at <http://espn.go.com/nfl/standings>; "College Basketball", available at <http://www.sportsreference.com/cbb/conferences/>; "NBA Standings," available at <http://espn.go.com/nba/standings>.
Baird (2004), p. 224.
For the NCAA Conference Average, I calculated the SDR by conference based on the teams’
conference records. For the calculations across all FBS Football and D-I men’s basketball, I calculated
the SDR based on team’s overall record. For the NFL and NBA I calculated the SDR across the entire
league. These results are in line with measures for these leagues found in the literature: Baird (2004);
Quirk and Fort (1992); and Fort (2006).
Rubinfeld Merits Rebuttal Report, ¶¶ 252 and 256.

HIGHLY CONFIDENTIAL–COUNSEL ONLY: SUBJECT TO PROTECTIVE ORDER Page 3

6. To demonstrate this more comprehensively, I have collected “Rating Percentage
Index” (“RPI”) data for men’s basketball. The RPI is a ranking of team performance,
based on win percentage and strength of schedule, which is used by the NCAA to aid
in the selection of teams to participate in the men’s basketball tournament.12

I

matched the RPI team rankings to total team revenues and expenses reported in the
Equity in Athletics Disclosure Act (“EADA”) school financials data for the years
2006-07 through 2011-12.13 I then calculated the school ranking by revenue and by
expenditures, and correlation coefficients that determine the extent to which team
performance (as measured by the RPI) is in fact aligned with the ranking of team
financial resources. The correlation coefficients are shown in Exhibit 2. The Exhibit
shows that the correlations are positive, but they are significantly less than one.
7. Exhibit 3 shows a plot of the RPI rank against school total revenue rank for 2012, and
Exhibit 4 shows a plot of the RPI rank against total school expenditure ranks. The
exhibits show that school financial resources and spending do not perfectly determine
team quality rank: there are a large number of schools whose team quality exceeds
their relative financial resources rank; and there are also a large number of schools
whose resources have failed to produce high performance teams.
8. I have also undertaken a similar type of analysis for FBS football. I collected data of
2011-12 “Colley Rankings” for FBS football.14 I then matched these rankings with
school EADA reported revenue and expenditures. Exhibits 5 and 6 show plots of the
2011-12 Colley ranks against football revenue and expenditure ranks.

The

correlations are positive, but significantly less than 1: the correlation coefficient
12

13

14

See: "What is the RPI?" RPIRatings.com, available at <http://www.rpiratings.com/WhatisRPI.php>,
accessed May 19, 2014. The RPI data was collected from "2012 NCAA Men's Basketball RPI,"
NCAA.org, available at <http://web1.ncaa.org/app_data/weeklyrpi/2012MBBrpi1.html>, accessed
May 9, 2014.
The EADA data was provided in the Rascher Class Declaration backup materials. They are collected
from each school that participates in the federal student financial assistance program. See: “Equity in
Athletics
Disclosure
Act,”
U.S.
Department
of
Education,
available
at
<http://www2.ed.gov/finaid/prof/resources/athletics/eada.html>, accessed May 16, 2014.
The Bowl Championship Series (BCS) uses the Colley Rankings as one of the components of its
computer
ranking
("BCS
computer
rankings,"
BCSFootball.org,
available
at
<http://www.bcsfootball.org/news/story?id=4765872>, accessed June 2, 2014). Colley rankings for
2011/12
are
available
at
"ColleyMatrix,"
Archive.org,
<http://web.archive.org/web/20130622061106/http://www.colleyrankings.com/foot2011/rankings/rank
16.html>, accessed June 2, 2014.

HIGHLY CONFIDENTIAL–COUNSEL ONLY: SUBJECT TO PROTECTIVE ORDER Page 4

between the Colley rank and the revenue rank is 0.56, and the correlation coefficient
between the Colley rank and the expenditure rank is 0.54.
9. In addition, I have looked at the relationship between both RPI and Colley ranks and
Professor Rascher’s estimated per-player broadcast compensation for 2009-10.15
Exhibit 7 shows a plot of Professor Rascher’s but-for compensation per player in
dollars for men’s D-I basketball against basketball RPI’s, and Exhibit 8 shows a plot
of the rank of Professor Rascher’s but-for compensation against RPI. Exhibits 9 and
10 show analogous plots for FBS football. It is clear from these exhibits that there
are many schools that were competitively strong that year, but which have fewer
resources and therefore would not offer student-athletes compensation as generous as
schools with greater resources. These schools would be further disadvantaged in
plaintiffs’ but-for world.
But-For Compensation Differentials

10. As I explained in the Rubinfeld Merits Report and Rubinfeld Merits Rebuttal Report,
in Plaintiffs’ but-for world, where schools would be free to pay student-athletes, these
financial offers would be an additional, potent recruiting tool that high-revenue
schools could use to attract and retain the best student-athletes. The ability to pay
players would clearly change the recruiting landscape even more in the favor of these
schools.16 In those reports, I provided numerous examples of large compensation
differentials between schools competing for the same recruits.17
11. To demonstrate this more comprehensively, I matched rivals.com recruiting data as
compiled by Professor Noll to Professor Rascher’s estimated per-player broadcast
compensation, by school and year.18

For each recruit, I compared the dollar

compensation figure corresponding to the school to which the recruit committed to
the maximum compensation amount among the set of schools from which the recruit
received offers. Exhibits 11 through 14 show that a substantial number of student-

15

16
17
18

Backup to Exhibits 14 and 15 of Expert Report of Daniel A. Rascher, September 25, 2013 (“Rascher
Merits Report”). 2009-10 is the most recent year in which Professor Rascher calculated damages in
his reports.
Rubinfeld Merits Report, ¶¶ 97-102, and Rubinfeld Merits Rebuttal Report, ¶ 252.
Rubinfeld Merits Rebuttal Report, ¶ 256.
These data overlap for the 2006-07 – 2009-10 school years.

HIGHLY CONFIDENTIAL–COUNSEL ONLY: SUBJECT TO PROTECTIVE ORDER Page 5

athletes would have forgone substantial compensation in Plaintiffs’ but-for world if
faced with the same set of school offers and if they had made the same commit choice
as in the actual world. Exhibit 11 shows that about 45 percent of football recruits
would have forgone at least $10,000 over a four year college career, and Exhibit 12
shows that almost 60 percent of all men’s basketball recruits would have forgone at
least $10,000.19 Exhibits 13 and 14 quantify how much money recruits would have
forgone among those who would have had the opportunity to earn at least $10,000
more over their four-year career. For example, for FBS football Exhibit 13 shows
that 3,108 recruits would have earned at least $10,000 more had they instead gone to
the school that would pay the most among the offers they received. The exhibit also
shows that 758 of these recruits could have been paid $100,000 or more over 4 years
above what they otherwise would have received. For basketball, Exhibit 14 shows
that 1,361 recruits would have earned at least $10,000 more, and of these, 187 would
have made at least $500,000 more over four years had they chosen the maximum
payment school among their offers.
12. Even a relatively small differential in compensation offered could affect the choices
made by student athletes. This is supported by the behavioral economics literature,
which is motivated in part by evidence that many individuals exhibit “hyperbolic
discounting,” meaning that they place heavy emphasis on the near future.20

19

20

This exhibit only counts recruits who received at least two offers as reported in Professor Noll’s
rivals.com data, and for whom there was a match (on both their offers and their commits) with
Professor Rascher’s damages estimates.
Frederick, Shane, George Loewenstein, and Ted O’Donoghue, “Time Discounting and Time
Preference: A Critical Review,” Journal of Economic Literature, Vol. XL (June 2002), pp. 351-401.

HIGHLY CONFIDENTIAL–COUNSEL ONLY: SUBJECT TO PROTECTIVE ORDER Page 6

Exhibit 1
Relative Standard Deviation of Winning Percentages
NCAA versus Professional Sports
2009

2010

2011

2012

2013

NCAA FBS Football (Conference Average)

1.528

1.432

1.477

1.536

1.600

2009-2013
Average
1.515

NCAA FBS Football (Overall)

1.580

1.591

1.581

1.692

1.727

1.634

NFL

1.586

1.474

1.611

1.525

1.527

1.545

NCAA D-I Men's Basketball (Conference Average)

1.708

1.730

1.835

1.723

1.735

1.746

NCAA D-I Men's Basketball (Overall)

1.991

1.980

2.093

1.985

1.956

2.001

NBA

2.902

2.861

2.494

2.762

2.806

2.765

Note: Values closer to one indicate greater balance.
Sources: "College Football," available at <http://www.sports-reference.com/cfb/conferences/>; "NFL Standings," available at
<http://espn.go.com/nfl/standings>; "College Basketball", available at <http://www.sports-reference.com/cbb/conferences/>; "NBA Standings,"
available at <http://espn.go.com/nba/standings>

HIGHLY CONFIDENTIAL - COUNSEL ONLY: SUBJECT TO PROTECTIVE ORDER

Exhibit 2
Spearman Rank Correlations of RPI with EADA
Revenue/Expenditures
Year
2007
2008
2009
2010
2011
2012

Correlation
with Total
0.65
0.60
0.69
0.69
0.67
0.65

Correlation with Total
Expenditure
0.69
0.61
0.71
0.68
0.68
0.64

Observations
(Teams)
329
334
339
343
342
341

Note: Correlations can take a value of -1 to 1. A value of zero means the series of data
are uncorrelated, and a correlation of 1 means they are perfectly correlated (i.e. that the
RPI ranks are always equal to the revenue or expenditure ranks).
Sources: EADA data provided in the Rascher Class Declaration backup materials.
See: “Equity in Athletics Disclosure Act,” U.S. Department of Education, available at
<http://www2.ed.gov/finaid/prof/resources/athletics/eada.html>, accessed May 16,
2014; "2012 NCAA Men's Basketball RPI," NCAA.com, available at
<http://web1.ncaa.org/app_data/weeklyrpi/2012MBBrpi1.html>.

HIGHLY CONFIDENTIAL - COUNSEL ONLY: SUBJECT TO PROTECTIVE ORDER

Exhibit 3
Total Men's Basketball Revenue Rank vs. RPI Rank, 2011-2012
0

50
Teams whose performance
ranks are relatively lower
than their revenue ranks

Revenue Rank

100

150

200

250
Teams whose performance
ranks are relatively greater
than their revenue ranks
300

350
350

300

250

200

150

100

50

0
RPI

Notes: Rating Percentage Index (RPI) ranks teams in sequential order based on their record and their strength of schedule. The best ranking a team can receive is one.
Sources: 2011-2012 Public EADA data; "2012 NCAA Men's Basketball RPI," NCAA.com, available at <http://web1.ncaa.org/app_data/weeklyrpi/2012MBBrpi1.html>.

HIGHLY CONFIDENTIAL - COUNSEL ONLY: SUBJECT TO PROTECTIVE ORDER

Exhibit 4
Total Men's Basketball Expenditures Rank vs. RPI Rank, 2011-2012
0

50
Teams whose performance
ranks are relatively lower
than their expenditure ranks

Expenditures Rank

100

150

200

250
Teams whose performance
ranks are relatively greater
than their expenditure ranks
300

350
350

300

250

200

150

100

50

0
RPI

Notes: Rating Percentage Index (RPI) ranks teams in sequential order based on their record and their strength of schedule. The best ranking a team can receive is one.
Sources: 2011-2012 Public EADA data; "2012 NCAA Men's Basketball RPI," NCAA.com, available at <http://web1.ncaa.org/app_data/weeklyrpi/2012MBBrpi1.html>.

HIGHLY CONFIDENTIAL - COUNSEL ONLY: SUBJECT TO PROTECTIVE ORDER

Exhibit 5
Total Football Revenue Rank vs. Colley Rank, 2011-2012
0

20
Teams whose performance
ranks are relatively lower
than their revenue ranks

Revenue Rank

40

60

80

Teams whose performance
ranks are relatively greater
than their revenue ranks

100

Spearman
Correlation = .56
120
120

100

80

60

40

20

Colley Rank 0

Notes: The Bowl Championship Series (BCS) uses the Colley Rankings as one of the components of its computer ranking. The best ranking a team can receive is one.
Sources: 2011-2012 Public EADA data; "Colley Matrix 2011 Rankings," available at
<http://web.archive.org/web/20130622061106/http:/www.colleyrankings.com/foot2011/rankings/rank16.html>.

HIGHLY CONFIDENTIAL - COUNSEL ONLY: SUBJECT TO PROTECTIVE ORDER

Exhibit 6
Total Football Expenditures Rank vs. Colley Rank, 2011-2012
0

Teams whose performance
ranks are relatively lower
than their expenditure ranks

20

Expenditures Rank

40

60

80

Teams whose performance
ranks are relatively greater
than their expenditure ranks

100

Spearman
Correlation = .54
120
120

100

80

60

40

20

Colley Rank 0

Notes: The Bowl Championship Series (BCS) uses the Colley Rankings as one of the components of its computer ranking. The best ranking a team can receive is one.
Sources: 2011-2012 Public EADA data; "Colley Matrix 2011 Rankings," available at
<http://web.archive.org/web/20130622061106/http:/www.colleyrankings.com/foot2011/rankings/rank16.html>.

HIGHLY CONFIDENTIAL - COUNSEL ONLY: SUBJECT TO PROTECTIVE ORDER

Exhibit 7
Professor Rascher's But-For D-I Men's Basketball Broadcast
Compensation vs. RPI Rank, 2009-2010

$300,000

Professor Rascher's But-For Compensation

$250,000

$200,000

$150,000

$100,000

$50,000

$0
350

300

250

200

150

100

50

RPI 0

Notes: Rating Percentage Index (RPI) ranks teams in sequential order based on their record and their strength of schedule. The best ranking a team can receive is one.
Sources: 2009-2010 Public EADA data; "2010 NCAA Men's Basketball RPI," NCAA.com, available at <http://web1.ncaa.org/app_data/weeklyrpi/2010MBBrpi1.html>; Backup
to Exhibits 14 and 15 in Rascher's Merits Report.

HIGHLY CONFIDENTIAL - COUNSEL ONLY: SUBJECT TO PROTECTIVE ORDER

Exhibit 8
Professor Rascher's But-For D-I Men's Basketball Broadcast
Compensation vs. RPI Rank, 2009-2010
0

Professor Rascher's But-For Compensation

50

100

150

200

250

300

350
350

300

250

200

150

100

50

RPI 0

Notes: Rating Percentage Index (RPI) ranks teams in sequential order based on their record and their strength of schedule. The best ranking a team can receive is one.
Sources: 2009-2010 Public EADA data; "2010 NCAA Men's Basketball RPI," NCAA.com, available at <http://web1.ncaa.org/app_data/weeklyrpi/2010MBBrpi1.html>; Backup
to Exhibits 14 and 15 in Rascher's Merits Report.

HIGHLY CONFIDENTIAL - COUNSEL ONLY: SUBJECT TO PROTECTIVE ORDER

Exhibit 9
Professor Rascher's But-For FBS Football Broadcast Compensation
vs. Colley Rank, 2009-2010
$60,000

Professor Rascher's But-For Compensation

$50,000

$40,000

$30,000

$20,000

$10,000

$0
120

100

80

60

40

20

Colley Rank 0

Notes: The Bowl Championship Series (BCS) uses the Colley Rankings as one of the components of its computer ranking. The best ranking a team can receive is one.
Sources: "Colley Matrix 2009 Rankings," available at <http://web.archive.org/web/20111211074129/http:/www.colleyrankings.com/foot2009/rank16.html>; Backup to Exhibits
14 and 15 in Rascher's Merits Report.

HIGHLY CONFIDENTIAL - COUNSEL ONLY: SUBJECT TO PROTECTIVE ORDER

Exhibit 10
Professor Rascher's But-For FBS Football Broadcast Compensation
Rank vs. Colley Rank, 2009-2010
0

Professor Rascher's But-For Compensation Rank

20

40

60

80

100

120
120

100

80

60

40

20

Colley Rank 0

Notes: The Bowl Championship Series (BCS) uses the Colley Rankings as one of the components of its computer ranking. The best ranking a team can receive is one.
Sources: "Colley Matrix 2009 Rankings," available at <http://web.archive.org/web/20111211074129/http:/www.colleyrankings.com/foot2009/rank16.html>; Backup to Exhibits
14 and 15 in Rascher's Merits Report.

HIGHLY CONFIDENTIAL - COUNSEL ONLY: SUBJECT TO PROTECTIVE ORDER

Exhibit 11
Percentage of Football Recruits Who Would Have Forgone at Least $10,000
70%

60%

Percent of Recruits

50%

40%

30%

20%

10%

0%
5

4

3

2

0

Total

Rivals.com Player "Star" Rating

Note: The plotted distribution was calculated by matching Professor Rascher's per-player broadcast "damages" by school with Professor Noll's recruit data which indicate the recruit
year (covering recruit years 2007-2010), where each student received offers, and the school to which they committed. Payments forgone for each student are the difference between
the maximum but-for payment they might have received among all the schools from which they received an offer and the payment they would have received at the school they
committed. The chart is limited to recruits that received more than one offer and includes compensation over four years.
Source: "Offers_and_commits_std" and "player_details" datasets from rivals.com backup to Noll Merits Report; Backup to Exhibits 14 and 15 in Rascher's Merits Report.

HIGHLY CONFIDENTIAL - COUNSEL ONLY: SUBJECT TO PROTECTIVE ORDER

Exhibit 12
Percentage of Basketball Recruits Who Would Have Forgone at Least $10,000
70%

60%

Percent of Recruits

50%

40%

30%

20%

10%

0%
5

4

3

2

0

Total

Rivals.com Player "Star" Rating

Note: The plotted distribution was calculated by matching Professor Rascher's per-player broadcast "damages" by school with Professor Noll's recruit data which indicate the recruit
year (covering recruit years 2007-2010), where each student received offers, and the school to which they committed. Payments forgone for each student are the difference between
the maximum but-for payment they might have received among all the schools from which they received an offer and the payment they would have received at the school they
committed. The chart is limited to recruits that received more than one offer and includes compensation over four years.
Source: "Offers_and_commits_std" and "player_details" datasets from rivals.com backup to Noll Merits Report; Backup to Exhibits 14 and 15 in Rascher's Merits Report.

HIGHLY CONFIDENTIAL - COUNSEL ONLY: SUBJECT TO PROTECTIVE ORDER

Exhibit 13
Cumulative Distribution of But-For Live Broadcast Payments
Forgone By FBS Football Recruits Over A Four-Year Career
3,500

Number of Recruits Who Would Have Foregone
at Least the Indicated Amount

3,108
3,000
2,669
2,500
2,087
2,000
1,581
1,500

1,000

758
382

500

192

107

15

0
>$10,000

>$25,000

>$50,000

>$75,000

>$100,000

>$125,000

>$150,000

>$175,000

>$200,000

Amount Forgone: Difference Between But-for Maximum Offer Payment and Commit Payment
Note: The plotted distribution was calculated by matching Professor Rascher's per-player broadcast "damages" by school with Professor Noll's recruit data which indicate the recruit
year, where each student received offers, as well as the school to which they committed. Out of a total of 7,154 football recruits identified in Professor Noll's rivals.com recruits
dataset (which covers recruits from 2007-2010), for those "damaged" students who received 2 or more offers, 3,108 would have been able to earn an additional $10,000 over a 4year career if they had chosen a different school. Payment forgone for each student is calculated as the difference between the maximum but-for payment they would have received
and the payment they would have received at the school to which they committed. The cumulative distribution of this difference is plotted for the cases in which the student would
have been better off in terms of but-for payments had they committed to one of the other schools from which they received an offer. The difference calculated applies to Professor
Rascher's estimated alleged "damages" for one year: the recruiting year and the corresponding year's "damages." This difference is multiplied by 4 to estimate a four-year total
payment.
Source: "Offers_and_commits_std" and "player_details" datasets from rivals.com backup to Noll Merits Report; Backup to Exhibits 14 and 15 in Rascher's Merits Report.

HIGHLY CONFIDENTIAL - COUNSEL ONLY: SUBJECT TO PROTECTIVE ORDER

Exhibit 14
Cumulative Distribution of But-For Live Broadcast Payments Forgone By
Division I Men's Basketball Recruits Over a Four-Year Career

Number of Recruits Who Would Have Foregone
at Least the Indicated Amount

1,600
1,400

1,361

1,200

1,084

1,000
873
800
587

600

425
400
267
187

200

126

81
23

9

>$800,000

>$900,000

0
>$10,000

>$50,000

>$100,000

>$200,000

>$300,000

>$400,000

>$500,000

>$600,000

>$700,000

Amount Forgone: Difference Between But-for Maximum Offer Payment and Commit Payment
Note: The plotted distribution was calculated by matching Professor Rascher's per-player broadcast "damages" by school with Professor Noll's recruit data which indicate the recruit
year, where each student received offers, as well as the school to which they committed. Out of a total of 2,332 basketball recruits identified in Professor Noll's rivals.com recruits
dataset (which covers recruits from 2007-2010), for those "damaged" students who received 2 or more offers, 1,361 would have been able to earn an additional $10,000 over a 4year career if they had chosen a different school. Payment forgone for each student is calculated as the difference between the maximum but-for payment they would have received
and the payment they would have received at the school to which they committed. The cumulative distribution of this difference is plotted for the cases in which the student would
have been better off in terms of but-for payments had they committed to one of the other schools from which they received an offer. The difference calculated applies to Professor
Rascher's estimated alleged "damages" for one year: the recruiting year and the corresponding year's "damages." This difference is multiplied by 4 to estimate a four-year total
payment.
Source: "Offers_and_commits_std" and "player_details" datasets from rivals.com backup to Noll Merits Report; Backup to Exhibits 14 and 15 in Rascher's Merits Report.

HIGHLY CONFIDENTIAL - COUNSEL ONLY: SUBJECT TO PROTECTIVE ORDER

List of New Materials Considered
Declaration of Daniel L. Rubinfeld, Ph.D., Regarding Competitive Balance
In re NCAA Student‐Athlete Name & Likeness Licensing Litigation
Bates Stamp/Title
Academic Texts and Articles
Frederick, Shane, George Lowenstein, and Ted O’Donoghue, “Time Discounting and Time Preference: A Critical Review,” 
Journal of Economic Literature,  Vol. XL, June 2001, pp. 350–401

News, Press and Websites
“2012 NCAA Men's Basketball RPI,” NCAA,  2012, available at 
http://web1.ncaa.org/app_data/weeklyrpi/2012MBBrpi1.html
“Conference Index: College Basketball,” Sports‐Reference.com,  2014, available at http://www.sports‐
reference.com/cbb/conferences/
“Conference Index: College Football,” Sports‐Reference.com,  2014, available at http://www.sports‐
reference.com/cfb/conferences/
“Colley's Bias Free College Football Rankings,” Colley Rankings,  n.d., available at 
http://web.archive.org/web/20130622061106/http://www.colleyrankings.com/foot2011/rankings/rank16.html
“NBA Standings: 2013–14,” ESPN,  2014, available at http://espn.go.com/nba/standings
“NFL Standings: 2013,” ESPN,  2014, available at http://espn.go.com/nfl/standings
“What Is RPI?,” Collegiate Basketball News,  2014, available at http://www.rpiratings.com/WhatisRPI.php

HIGHLY CONFIDENTIAL–COUNSEL ONLY: SUBJECT TO PROTECTIVE ORDER

Date

Last Accessed 
On:
3‐Jun‐14
3‐Jun‐14
3‐Jun‐14

3‐Jun‐14
3‐Jun‐14
3‐Jun‐14
3‐Jun‐14

1 of 1