The Economics of Sports

FIFTH EDITION

Chapter 5

Competitive
Balance

MICHAEL A. LEEDS | PETER VON ALLMEN

 Competitive Balance

• The term means different things to different

people
– Close competition every year, with the difference
between the best and worst teams being
relatively small
– Regular turnover in the winner of the league’s
championship
• More generally, it means degree of parity
within a league

Copyright ©2014 Pearson Education, Inc. All rights reserved. 5-2

 Learning Objectives

• Understand why owners and fans care about

competitive balance
• Be able to use and interpret the different
measures of competitive balance
• Describe and compare the tools that leagues
use to promote competitive balance and the
limitations of those tools.

Copyright ©2014 Pearson Education, Inc. All rights reserved. 5-3

 5.1 Desire for Competitive
Balance
• Fans and owners alike have a conflicted
relationship with competitive balance
• On any given day, seeing one’s team win is
preferable to seeing it lose
• But an uninterrupted string of wins is dull

Copyright ©2014 Pearson Education, Inc. All rights reserved. 5-4

 The Fans’ Perspective

• A game with an uncertain outcome is much

more exciting than a foregone conclusion
• Table 5.1 shows that from 1950 to 1958
attendance for both the Yankees and the
entire American League either stagnated or
fell because of Yankees dominance
• Evidence suggests that in many sports,
fans prefer a game where the home team
has a 60-70% chance of winning

Copyright ©2014 Pearson Education, Inc. All rights reserved. 5-5

 Table 5.1

Copyright ©2014 Pearson Education, Inc. All rights reserved. 5-6

 The Owners’ Perspective

• Competitive balance matters to owners

because it matters to fans
• Leagues adopt policies to promote competitive
balance because they enhance fan demand
• Leagues restrict team behavior if it leads to
teams that are too strong or too weak (see
Table 5.1)
• Balance is hard to achieve if some teams
maximize wins while others maximize profits

Copyright ©2014 Pearson Education, Inc. All rights reserved. 5-7

 Effect of Market Size

• There is considerable debate over the impact

of market size on competitive balance
• There are three primary sources of
disagreement
– How to measure of success
• During playoffs or regular season?
– How to characterize market size
• Market size has become more important with the
advent of broadcasting

Copyright ©2014 Pearson Education, Inc. All rights reserved. 5-8

 Effect of Market Size (cont.)

• The third point of disagreement is how to measure

the impact of policies, such as revenue sharing
• Profit-maximizing leagues do not want total balance
– they want big-market teams to win more
• At minimum, more populous locations will win the
league championship more frequently
• Figure 5.1 shows an additional win is more valuable
in a larger market, so the optimum number of wins is
greater
• Operate where the marginal returns to wins equals
the marginal cost of wins, assuming firms profit
maximize

Copyright ©2014 Pearson Education, Inc. All rights reserved. 5-9

 Figure 5.1

Copyright ©2014 Pearson Education, Inc. All rights reserved. 5-10

 The Effect of
Diminishing Returns
• The impact of another unit of a variable input (when
added to a fixed input) eventually falls
– This effect limits the desire of teams to stockpile –
and pay – star players
– And promotes competitive balance
• Drew Brees has limited value to a team that has
Tom Brady, a Hall of Famer
– Brees adds little to wins, attendance, or revenue
– The added cost exceeds the added benefit
– Other teams can use him more effectively

Copyright ©2014 Pearson Education, Inc. All rights reserved. 5-11

 Is Perfect Balance Profit
Maximizing?
• Winning has a bigger MR, MC
impact in a larger market
– It adds more to gate, MRlarge
media, and venue MRsmall
revenue
– MRwins higher in big cities
• Profit-maximizing leagues MC
and competitive balance
may be incompatible
– Big cities will win more
unless MCwins is also Wins
higher

Copyright ©2014 Pearson Education, Inc. All rights reserved. 5-12

 A History of
Competitive Balance
• Yankee dominance of MLB is not new
– Appeared in 15 World Series between 1947 and 1964
• The LA Lakers and San Antonio Spurs won 9 of 13
NBA championships between 1999 and 2011
• The Montreal Canadiens won 10 Stanley Cups in the
NHL between 1965 and 1979
– They were succeeded by NY Islander and Edmonton
Oiler dynasties in the 1980s
• The NFL is more balanced, but the Browns and
Lions have never been in a Super Bowl

Copyright ©2014 Pearson Education, Inc. All rights reserved. 5-13

 Competitive Balance in Soccer
from 2000-01 to 2011-12
• In England’s Premier League
– Manchester United, Chelsea, and Arsenal have won
11 times
• In Germany’s Bundesliga
– Bayern Munich and Borussia Dortmund have won 9
times
• In Italy’s Serie A
– AC Milan, Inter Milan, and Juventus have won 11 times
• In Spain’s La Liga
– FC Barcelona and Real Madrid have won 10 times

Copyright ©2014 Pearson Education, Inc. All rights reserved. 5-14

 5.2 Measuring Competitive
Balance
• Within-Season Balance (Variation)
– Compares teams within a season—across a league
– A low dispersion of team winning percentages means
that the teams are evenly matched
• Between-Season Balance (Variation)
– Compares winners (champions) across time
– Some leagues have the same champions year after
year
• Regular turnover is preferred

Copyright ©2014 Pearson Education, Inc. All rights reserved. 5-15

 Within-Season Variation (1)

• We could use the standard deviation of winning

percentage
– The standard deviation gives the dispersion of
performance by teams
– It is the square root of the average squared deviation
from the mean
– See formula on p. 159
• The mean performance is always .5 as there are a
winner and a loser in every game

Copyright ©2014 Pearson Education, Inc. All rights reserved. 5-16

 Application

• In 2011, the standard deviation in the American

League was 0.067
– The typical winning percentage varies by 0.067 from
the mean
• The standard deviation in the National League was
0.054, about three-fourths that of the American
League.
– The National Leagues was more balanced

Copyright ©2014 Pearson Education, Inc. All rights reserved. 5-17

 Within-Season Variation (cont.)

• We cannot compare the standard deviation across

leagues or across seasons with a different number
of games
• As the number of matches rises, winning
percentages cluster around the mean
– If teams are evenly matched, then the probability of
success in any game is close to .5
• We can apply the binomial distribution
• In a short season, a lucky team can have all wins and an
unlucky team no wins
• The league can look unbalanced in a short season

Copyright ©2014 Pearson Education, Inc. All rights reserved. 5-18

 Within-Season Variation (cont.)

– We need a better measure

– We compare a league’s standard deviation to the
standard deviation that would result if teams were
evenly matched
– The “ideal” standard deviation occurs when each team
has a 50% chance of winning a given game
• The better measure is the ratio of the actual to the
ideal standard deviation
– R = A/I

Copyright ©2014 Pearson Education, Inc. All rights reserved. 5-19

 Computing Within-Season
Balance
• The ratio of actual to ideal standard deviation
– N = # Teams
– G = # Games
– WPCTi,t = Winning percentage of team i at time t

N 2
1
N
 WPCT
i 1
i ,t  .500
R A / I 
0 .5
G

Copyright ©2014 Pearson Education, Inc. All rights reserved. 5-20

 Interpreting the Ratio

• The ratio R gives a standardized measure

– Actual and ideal standard deviation fall as G rises
– We can now compare leagues and seasons with a
different number of games
– The formula appears on p. 161
• As a rule, R > 1
• If R = 1, the league is completely balanced
– Outcomes are effectively randomly determined
• As R rises, balance worsens

Copyright ©2014 Pearson Education, Inc. All rights reserved. 5-21

 How Do Leagues Compare?

• English Premier League was the most

balanced in 2011-2012
• The NFL, NHL and MLB have similar balance
• NBA is by far the least balanced
– This has been true in most years
• See Table 5.3 for the actual statistics

Copyright ©2014 Pearson Education, Inc. All rights reserved. 5-22

 Table 5.3

Copyright ©2014 Pearson Education, Inc. All rights reserved. 5-23

 Between-Season Balance

• We can use the standard deviation of each team’s

winning percentage
– Unlike the within-season measure, there is no “ideal”
measure
– It is unclear what is a good or bad value
• We can use the frequency of championships
– It is hard to compare this across leagues
– See Table 5.4

Copyright ©2014 Pearson Education, Inc. All rights reserved. 5-24

 Table 5.4

Copyright ©2014 Pearson Education, Inc. All rights reserved. 5-25

 The Herfindahl-Hirschman Index
• HHI measures the concentration of championships
• In industrial organization, it measures monopoly power
• Let ci = #championships by team i
– T = #teams; N = #Years

T 2
 ci 
HHI   
i 1  N 
– If HHI=1, one team always wins
– If HHI=1/N and N>T, complete competitive balance
– If HHI=1/T and N<T, complete competitive balance
• See p. 164 for computations; What if the league had 10
teams?

Copyright ©2014 Pearson Education, Inc. All rights reserved. 5-26

 Applying the HHI to Sports

• See Table 5.4

• the HHI for the Premier League is far greater
than for any other league
• the HHI for the NBA is also large
• the HHI for the NHL, NFL, and MLB are
substantially smaller
• the HHI for the NHL is the smallest, indicating
that the league was most balanced in the first
decade of the 21st century

Copyright ©2014 Pearson Education, Inc. All rights reserved. 5-27

 Illustrating Competitive
Imbalance
• The Lorenz Curve measures inequality in a
population
– It is typically used to measure income inequality
– We use it to measure inequality in winning
• Line up NBA teams by wins in 2010-2011 (p. 164)
– 1230 games were played, so population = 1230
– The 3 weakest teams (the lowest decile) won 58
games
• 58 games correspond to 4.7 % of 1230
• Thus, the bottom 10% accounted for 4.7% of wins
• The next 10% accounted for 5.8% and so on
• The top 10% accounted for 14.7% of wins
– Figure 5.2 presents the results

Copyright ©2014 Pearson Education, Inc. All rights reserved. 5-28

 The Lorenz Curve for the NBA

• Red line shows perfect

balance 1.2

– Adding 10% more teams 1

adds 10% more wins 0.8

• Blue line shows reality 0.6

– Bottom 10% wins less 0.4

than 10% 0.2

• Sags below red line 0

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1
– As we add better teams,
blue curve catches up
– At 100% of teams, we
account for 100% of wins
• The farther the blue line
sags, the greater the
inequality
Copyright ©2014 Pearson Education, Inc. All rights reserved. 5-29
 5.3 Altering Competitive Balance

• All the major North American sports leagues

have developed policies to promote
competitive balance
– Revenue sharing
– Salary caps and luxury taxes
– Reverse-order draft
• Players claim that the policies merely
depress overall salaries
• This section explores the policies’ effect on
competitive balance

Copyright ©2014 Pearson Education, Inc. All rights reserved. 5-30

 The Invariance Principle

• Free agency allows a player to go to the team that

offers the best employment terms
– Players sell their services to the highest bidder
• Owners claim that free agency is incompatible with
competitive balance
– Economic theory suggests otherwise
• Markets direct resources to the most productive
uses
– Property rights do not affect the flow of resources
– They affect only who gets paid for them
– Simon Rottenberg (1956) first applied the principle to
sports

Copyright ©2014 Pearson Education, Inc. All rights reserved. 5-31

 How the Invariance Theorem
Works
• In 2012 Albert Pujols was more valuable to the LA Angels than
to the St. Louis Cardinals in terms of revenue
• With free agency
• The Angels paid Pujols to move to LA
• Without free agency
• The Angels would pay the Cardinals for the “rights” to Pujols
• Pujols moves in both cases—the use of the resource is
unaffected
• The only difference is who gets paid
• The reserve clause did not prevent player movement
• In 1920 Red Sox sold Babe Ruth to Yankees
• Connie Mack twice sold off championship teams in
Philadelphia

Copyright ©2014 Pearson Education, Inc. All rights reserved. 5-32

 With Transaction Costs…

• The Invariance principle breaks down if there are

large costs to making transactions
• Benefits that do not exceed transaction costs are
not realized
• Transactions costs could have prevented the
Angels from pursuing Pujols

Copyright ©2014 Pearson Education, Inc. All rights reserved. 5-33

 Revenue Sharing

• MLB, NBA, NFL, and NHL share network TV revenue

equally
• NFL extensively shares all sources of revenue
– Teams keep only 60% of home gate revenue
– Huge TV package dwarfs other sources
• MLB shares 31% of local revenue (minus
“expenses”)
– Central (non-local) revenue also goes
disproportionately to teams in 15 smallest markets
– They will have to spend this revenue on players

Copyright ©2014 Pearson Education, Inc. All rights reserved. 5-34

 Revenue Sharing (cont.)

• The NBA is expected to vastly increase sharing

– Teams will share up to 50% of local revenue (minus
“expenses”)
• The NHL transfers income to teams
– In bottom 15 smallest media markets
– If the market has a base population under 2 million

Copyright ©2014 Pearson Education, Inc. All rights reserved. 5-35

 Revenue Sharing (cont.)

• Revenue sharing equalizes revenue across teams

• Goal is to reduce incentive of big teams to pursue
talent
• This will not work if
– Sharing shifts down MR of a win for all teams equally –
big-market teams still have higher MR
– Teams that receive revenue do not spend their added
revenue on talent
• Some teams might pursue profit over wins

Copyright ©2014 Pearson Education, Inc. All rights reserved. 5-36

 Salary Caps

• NBA, NFL, and NHL all have salary caps (not MLB)
– Salary caps are neither a salary limit nor a cap
• They set a band on salaries: both upper and lower
limits to payrolls (not individual salaries)
• Take qualifying revenue (QR) of league
– Not all revenue “qualifies”
– Definition varies from league to league
• Players get a defined share of the QR
• Divide total player share by # of teams
• Add & subtract a fudge factor (5-20%) to get the
bounds

Copyright ©2014 Pearson Education, Inc. All rights reserved. 5-37

 NFL Example

• Players receive
– 55% of national broadcast revenue
– 45% of NFL Ventures (merchandising) revenue
– 40% of aggregate local revenues
• Each team must spend at least 89% of the
cap
• Overall, players must receive at least 95%

Copyright ©2014 Pearson Education, Inc. All rights reserved. 5-38

 Hard Caps and Soft Caps

• The NFL has a hard cap

– Sets a firm limit on salaries without exceptions
• The NBA has a soft cap with many exceptions
– Mid-level exception
• Team can sign 1 player to the league average salary
• Even if it is over the limit
– Rookie exception
• Team can sign a rookie to his first contract
• Even if it is over the limit
– Larry Bird exception
• Named for former Celtics great who was its first
beneficiary
• Team can re-sign a player who is already on its roster
• Even if it is over the limit

Copyright ©2014 Pearson Education, Inc. All rights reserved. 5-39

 The NBA and Soft Caps

• All the exceptions have undermined the cap

• This has led to further rules
– The NBA now caps individual salaries as well
– The NBA has a luxury tax to prevent teams from
abusing the exceptions
• This has nothing to do with luxury boxes
• Teams pay a tax that increases for every $5 million over
the cap
• A team $15 million over the cap must pay a $37.5
million tax

Copyright ©2014 Pearson Education, Inc. All rights reserved. 5-40

 MLB’s Luxury Tax

• Tax starts at 17.5% for first-time offenders

– Threshold is $178 million in 2011-2013
– Rises to $189 million in 2014
• Tax rises with the number of abuses
• NY Yankees have paid the tax every year

Copyright ©2014 Pearson Education, Inc. All rights reserved. 5-41

 The Reverse-Order Entry Draft

• Ideally, it levels out talent over time

• Teams select new players according to their
order of finish in the previous season
– Weakest teams get the first choice of new talent
– Strongest teams get the last choice

Copyright ©2014 Pearson Education, Inc. All rights reserved. 5-42

 What Was the Point of the Draft?

• Did teams just want to keep salaries low?

• Was is a cynical move by weak teams?
– Eagles’ owner Bert Bell proposed the draft
– The Eagles happened to have the NFL’s worst record
• Was it an idealistic move?
– The NY Giants & Chicago Bears agreed to the
draft
– They were the dominant teams & had the most to
lose
– Tim Mara (Giants owner): “People come to see
competition…. We could give [it to] them only if the
teams had some sort of equality.”

Copyright ©2014 Pearson Education, Inc. All rights reserved. 5-43

 Weaknesses of the Draft

• It can lead to “tanking”

– Teams lose intentionally to improve draft position
– That is why the NBA has a draft “lottery”
• Under a lottery
– The weakest team has the best chance of choosing
first
– But it might not
• It works only if teams can identify talent

Copyright ©2014 Pearson Education, Inc. All rights reserved. 5-44

 Identifying Talent: Moneyball

• Billy Beane, the Oakland A’s general manager,

found underrated players
• He saw that teams
– Overrated physical skills
– Underrated on-base percentage
• Using different criteria in player selection kept his
small market team competitive
• Other teams eventually caught on
– A’s have fallen on hard times as a result

Copyright ©2014 Pearson Education, Inc. All rights reserved. 5-45