OceanofPDF.

com
An Economist Goes to the Game
OceanofPDF.com
AN ECONOMIST GOES TO THE
GAME

HOW TO THROW AWAY $580 MILLION AND

OTHER SURPRISING INSIGHTS FROM THE
ECONOMICS OF SPORTS

PAUL OYER

OceanofPDF.com
Published with assistance from the foundation established in memory of Amasa Stone Mather of the
Class of 1907, Yale College.

Copyright © 2022 by Paul Oyer. All rights reserved. This book may not be reproduced, in whole or
in part, including illustrations, in any form (beyond that copying permitted by Sections 107 and 108
of the U.S. Copyright Law and except by reviewers for the public press), without written permission
from the publishers.

Yale University Press books may be purchased in quantity for educational, business, or promotional
use. For information, please e-mail sales.press@yale.edu (U.S. office) or sales@yaleup.co.uk (U.K.
office).

Set in Gotham and Adobe Garamond types by Integrated Publishing Solutions.

ISBN 978-0-300-21824-4 (hardcover : alk. paper)

Library of Congress Control Number: 2021949772
A catalogue record for this book is available from the British Library.

This paper meets the requirements of ANSI/NISO Z39.48-1992 (Permanence of Paper).

OceanofPDF.com
 For my father, Calvin Oyer,
who helped develop my loves of sports and learning

OceanofPDF.com
 Contents

Prologue
1 Should You Help Your Kid Become a Pro Athlete?
2 What Do Silicon Valley and Czech Women’s Tennis Have in Common?
3 Why Do Athletes Cheat and Lie?
4 Are Athletes Worth All That Money?
5 Why Do Athletes Use Their Least Successful Moves So Often?
6 How Does Discrimination Lead to a Proliferation of French Canadian
Goalies?
7 How Do Ticket Scalpers Make the World a Better Place?
8 Why Should You Be Upset If Your Hometown Hosts the Olympics?
9 Who Wins When People Gamble?
Epilogue

Notes
Acknowledgments
Index

OceanofPDF.com
 Prologue

On the evening of June 13, 1997, tens of millions of people around the
world were glued to their television sets. Twenty-eight seconds remained on
the clock in the sixth game of the NBA Finals between the dynastic
Chicago Bulls, who led the series three games to two, and the long-
suffering Utah Jazz. The game was tied, and the Bulls were huddling,
planning the play they would use to take the lead and the NBA
championship. Michael Jordan, the Bulls’ star, later recalled: “When Phil
[Jackson, coach of the Bulls] drew up the play at the end, everybody in the
gym, everybody on TV, knew it was coming to me.” Jordan was the best
player in the league, maybe the best in NBA history, and he had won game
1 for the Bulls with a last-second shot. According to a teammate, the plan
on that play was to “give the ball to Michael and get out of the way.”1
When the game restarted, Jordan had the ball and made a move to the
basket as though he were going to take a shot—but, covered by two Utah
players, he dished the ball to Steve Kerr instead. Kerr, now famous as the
coach of another dynasty (the Golden State Warriors of the 2010s), was
then a decent NBA guard best known as an accurate shooter. After taking
Jordan’s pass, Kerr sank a seventeen-foot jump shot to give the Bulls a lead
they would not lose.
The Moroccan-born middle-distance runner Rashid Ramzi became a
citizen of Bahrain and subsequently brought athletic glory to that island
country’s one million people. Shortly after becoming a citizen, Ramzi took
the world of track and field by storm with victories in the 800-meter and
1,500-meter runs at the Helsinki world championships in 2005. More
important, he brought Bahrain its first Olympic gold medal (in fact, its first
Olympic medal of any kind) by winning the 1,500-meter race in Beijing in
2008. But his glory didn’t last long. Ramzi had to give back his gold medal
after a retest in 2009 of his Olympic blood sample showed he had used
EPO-CERA, a banned substance meant to help people with kidney
problems but also used by cyclists, sprinters, and other athletes to create
extra red blood cells.2
 A twenty-three-year-old South Korean named Jeongeun Lee6 burst onto
the golf scene in June 2019, winning the U.S. Women’s Open and a $1
million prize. That’s not a typo in her name; she goes by “Lee6” because
there are so many Lees in the Korean LPGA that it is difficult to keep track
of them all. Lee6’s victory marked yet another milestone in Korean
domination of women’s golf.3
These three events may seem like a random collection of sports-page
headlines and highlights. No doubt they brought joy and excitement to the
Bulls, Ramzi (at least initially), Lee6, and their respective fans. But a
common thread runs more subtly through these examples: we can look at all
of them through the lens of economics. You could even say each is driven
by economics. Economic forces drove the participants’ choices and
strategies in all three examples. Jordan was an amazing player to watch, but
his pass to Kerr showed that Jordan also had a grasp of game theory. Ramzi
spent his life becoming an elite athlete but at some point realized he could
get to the very top of his sport only if he gave in to the prisoner’s dilemma.
And Lee6’s million-dollar payday was the product of growing up in a
country with a high savings rate and limited opportunities for women in the
labor market.
ESPN doesn’t explain these events as natural outcomes of economics.
But having researched and taught economics since before Jeongeun Lee6
was born, I believe that all great athletes (well, most of them anyway) and
fans are also sound economists. They have to understand how to make
investments, how to choose strategies, and how to resolve trade-offs that
separate champions from also-rans.
I could not have been Michael Jordan no matter how hard I tried. But
Michael Jordan could not have become the (arguably) greatest player of all
time without understanding strategy, which in turn relies on economic
principles. He had to know the point at which the costs of driving to the
hoop one more time outweighed the benefits. All else being equal, Jordan
should have taken every shot; he was the best player on the court. But all
else is not equal. In equilibrium, opponents guarded Jordan more closely
than they guarded Kerr. If they had known that Jordan would take every
shot, all five Jazz players would have swarmed him. To truly maximize his
talents, the Bulls had to use them just the right amount. That trade-off
between taking the shot himself and passing to Kerr was an economic
decision that Jordan, through endless practice, coaching, and game
experience, knew how to optimize.
Ramzi also had amazing natural athletic talent, which he developed
through exhaustive training. It made him one of the best runners in the
world. But it probably wasn’t enough; he would not have won a gold medal
solely through his talent and effort. He faced strong incentives to use
banned substances to give himself the extra boost he needed. Although he
ended up being caught and disgraced, he took a rational risk to get a huge
economic payoff in both money and fame. And because all his top
opponents were also taking banned substances, he had no choice if he
wanted to make it to the top.
Lee6 was the product of strong economic forces, as well. She grew up in
a country with an intense education system, an emphasis on skill
development in childhood, and limited labor-market opportunities for
talented women. So she focused her investments on developing golf skills.
It was a risky bet, but it paid off.
Economics won’t make you a great tennis or soccer player, but it
answers some of the questions that sports fans ask every day. Should I
encourage my kid to try to get a college athletic scholarship (a real one, that
is, not the kind people buy)? Why do major sports figures make so much
money? Why do NBA teams give players multiyear contracts that guarantee
them millions of dollars per year even if they get injured or stop playing
well? Why would a pitcher throw the same pitch twice in a row when
mixing up pitches is the key to keeping batters off-balance?
In addition to tackling these questions internal to the game, I also hope
to shed light on some aspects of public policy. Why do cities build stadiums
with public funds, usually with large cost overruns, when the same cities
are unable to provide basic services? Are ticket scalpers good or bad? Why
can’t you get a ticket to a playoff game at a reasonable price?
Along the way, we’ll meet people whose lives are driven by both sports
and economics. Bobby Estalella was a marginal catcher who hung on in
Major League Baseball with the help of unnatural substances; Amy
Stephens was a schoolteacher in suburban Atlanta who built a ticket empire
after a random event forced her to change one evening’s entertainment
plans; Tina Weirather was a downhill skier who hails from the greatest per
capita sports country in the world (and also the dominant country in
producing false teeth). Each of these three people made choices based on
the economic forces they faced—the costs and benefits of the opportunities
in front of them. All of them had to adjust their plans when the competitive
markets in which they were successful evolved in new and more
challenging directions.
Though I hope this book will get readers interested in the academic side
of sports economics, that is not my main goal. As an economist who loves
to play and watch sports, I have written primarily about the economics in
doing those things. This book is for sports fans and players, from casual to
professional. I focus much less on owners and the many others who make
money from sports—such as broadcasters, agents, and advertisers—except
as their decisions affect fans and players.
Sports are meant to be fun and relaxing, mostly. I gather that not
everybody feels the same is true of economics. I hope you will begin to see,
however, as we visit bleachers, broadcast booths, playing fields, and other
arenas, that economics too has its entertaining side. Just as you don’t need
to spend countless hours in the gym to appreciate Michael Jordan’s jump
shot (or Steve Kerr’s), you don’t need to master complex equations to enjoy
contemplating the insights that economics can give you.
And, just maybe, you will never watch or play a game the same way
again.

OceanofPDF.com
 1
Should You Help Your Kid Become a Pro Athlete?

The sun is setting on Hoover Park one afternoon in April. Hengehold

Truck Rental has taken the field for one final stand in a Palo Alto Little
League opening-day game. Masonic Lodge, having fallen behind after
making roughly fifteen errors in the fourth inning, comes to bat with one
last shot to take back the lead. With runners on first and second, the team’s
star player strides to the plate.
The tension mounts. No more than two of Hengehold’s outfielders are
picking their noses. Many of the parents watching from the rickety
bleachers have looked up from their phones. The pitcher rears and fires a
fastball down the middle . . . and David Oyer (my then-twelve-year-old son)
crushes it over the left-field fence for a game-winning home run. He jogs
around the bases with his fist in the air as though he’s just won the World
Series, as any self-respecting Little Leaguer would. His teammates mob
him at the plate. Masonic Lodge has won its first game of the year.
That game still stands as a highlight in his short career as a star of youth
baseball. Both David, whose athletic glory peaked in that final season of
Little League, and I, taking vicarious pride in my children’s sporting
accomplishments the way all overinvolved fathers do, remember the
moment as if it were yesterday. It’s a safe bet that for the rest of his life
David will also remember his first-ever pitching appearance, when as a
nine-year-old he struck out a kid who would eventually become a Stanford
varsity pitcher and pro prospect. It’s equally likely that I will never forget
my only career home run, in the final year of my lengthy stint on the Park
ShopRite squad in the Nutley, New Jersey, Little League.
David (now a lawyer) and I are not alone in reliving these moments.
Youth sports are a cherished memory for more people than might care to
admit it. One reason that is the case, of course, is that sports are fun, and
rational humans are utility maximizing—they choose to do the things that
bring them the most happiness. Playing sports when one is young, just like
going to a movie or an amusement park, increases many people’s utility.
People “consume” sports participation because they enjoy it.
But there is a second consideration here: youth sports can help develop
skills like coordination, teamwork, and decision-making. Think of an eight-
year-old point guard who has to learn to dribble, direct an offense, and
choose which of her teammates to pass to on every possession—as well as
learn that even if she thinks she’s the team’s best player, if she shoots the
ball every time down the floor, her team won’t win. Playing basketball
might make her smarter, more physically adept, and more sociable—all
valuable skills that should contribute to her success and happiness in the
future. Parents who consider these factors when they sign their children up
for teams see organized sports as an “investment.” So youth sports are two
kinds of goods in one: a consumption good, like a can of soda or a
magazine, and an investment good, like a share of stock or a college
education.

How Should Your Kids Spend Their Childhoods?

People often think economics is the study of money, but it’s not. It is the
study of scarce resources. Just as societies have to figure out how best to
use scarce resources like water, meat, and clean air, so people have to
decide how to spend their limited time, money, and energy. When we
wonder how much time kids should devote to sports, we are confronting an
economics problem.
Youth is one of the most valuable resources you will ever possess. To use
it efficiently, you need to prepare for the rest of your life while having
enough fun to avoid looking back with regret. Failure to strike the right
balance in either direction can have long-term repercussions: someone who
spends high school cutting class and smoking dope might enjoy the moment
but will likely pay a price later in life; someone who regularly turns down
party invitations to pound the AP calculus study guide may, as an adult,
mourn their wasted youth.
Seen in this light, youth sports are incredibly expensive. The dollar cost
of soccer balls, hockey sticks, field time, gas used driving to games, and so
on can certainly add up, but it pales next to the cost of time not spent doing
something else. Every afternoon a kid spends at soccer practice is an
afternoon not spent studying—an activity that could pay off more directly
in getting into a better school, earning more money as an adult, and being
able to afford a bigger house and his own kids’ education. On the other
hand, by studying, he would be giving up the chance to play soccer—which
would let him have fun, develop life skills, and get some exercise. Even
boiling the problem down to those two choices ignores all the other forgone
options: he could learn the piano, get valuable sleep to help him grow and
stay healthy, or just enjoy playing with Legos or watching SpongeBob. So
youth sports, like most other childhood pastimes, are very expensive in
“opportunity cost”—the alternative experiences a kid could have during the
few precious years he has to both enjoy childhood and invest in his future.
You might think that economists can just do some sort of clever study to
determine whether playing sports (organized or otherwise) as a child leads
to adult success in the labor market. But it’s not that easy. The ideal way to
determine whether youth sports pay off in the labor market would be to
randomly assign one bunch of kids to play sports and to bar another
randomly chosen group from ever playing them. Then, when the kids had
grown up, the study could look at the incomes of both groups and see which
made more money. If the study found that adults who had played sports as
kids made more, everyone could say confidently that investing in youth
sports yields a financial return. But although randomized experiments are
becoming much more common in the social sciences, nobody has ever done
one on youth sports, and I doubt anyone ever will. Economists have been
able to do conclusive studies about, for example, the effects of more
education, growing up in an affluent neighborhood, and other factors, but
they have not found credible “natural experiments” for youth sports
participation.
Still, economists have done the best they can with the data they have.
Several studies have shown that people who play more sports as kids make
more money as adults. One study used a sample of Americans in their
twenties. Those who had participated in organized sports in high school
made 6 percent more, on average, holding other factors constant.1 A recent
study of German youths also showed that those who play youth sports
obtain better wages and labor-market outcomes.2 But even though these
results are intriguing, we have to remember the mantra of introductory
statistics class: correlation does not equal causation. Although the
economists who conducted these studies made every effort to control for
other factors, they could not clearly separate correlation from causation.
Plausible alternative explanations exist that would not suggest that playing
sports as a kid helps your career when you grow up.
For example, the thrill of victory we got from David’s home run is not
limited to Little League games; David and I are both competitive people.
We measure success in wins and losses when we do pretty much anything.
Watching Jeopardy! is a spirited event in our living room, with score
carefully kept and many shouts of “I said it first!” We also both played a lot
of youth sports. But organized sports did not make us competitive people;
we already were that way. Perhaps a competitive nature, which can’t be
measured in the economists’ analysis, is behind both success in the labor
market and interest in youth sports programs. The researchers would not be
able to identify competitiveness as the cause of the results.
The studies that have been done lead to the conclusion that playing
sports as a kid may bring some benefit later in life. Every study found a
correlation between increased youth sports activity and higher adult wages.
But the effect, if it exists, is small. One recent study looked at data on
American youth sports participation and, consistent with prior work, found
that kids who play sports get more education, get paid more later in life, and
are healthier. But the economists also concluded that these relationships are
almost entirely because kids who participate in sports are different from
kids who don’t and that child athletes would be more successful and
healthier whether they played sports or not.3 If all that mattered was success
in the labor market later in life, studying might have been a better use of
time than long sports practices.
The implications for parents are pretty straightforward. The vast majority
of kids have little or nothing to gain in the future from playing sports. As an
investment, playing sports is hard to justify—if your kids don’t like sports,
don’t force them into it. But for many kids, youth sports offer a great deal
of simple consumption value. Kids really love to play sports, and since you
only get to be a kid once, there is good reason to let them do so. The
economist’s bottom-line advice is simple: let your kid be a kid and play
sports if she wants to.

Sports and College: A Very American Institution

If sports have any investment value for American kids, it can be summed
up in one word: college. College can make youth sports pay off financially
in four main ways.
The first and most direct payoff comes for a hypergifted kid who
becomes a star athlete in college and then a professional. This consideration
is not relevant for the vast majority of people.
The second possible reason was captured well in a New York Times
feature about Matt Skoglund, whose experience might lead you to think that
college can make youth sports pay off even if you never make a nickel as an
athlete. Skoglund spent almost ten years as an attorney for the Natural
Resources Defense Council directing its Northern Rockies operations out of
his office in Bozeman, Montana.4 He didn’t make the kind of money many
big-city lawyers make, but he took his compensation in lifestyle and
working at his dream job for a cause he believes in. It was hockey that
made this life possible. Maybe.
When he was in high school, Skoglund was the focus of a New York
Times article on the value of sports for college admissions.5 A native of the
Chicago suburbs, Skoglund was a senior at Choate Rosemary Hall, a
prestigious boarding school in Connecticut. His middling SAT scores were
well below average for people admitted to Middlebury College, but the
Middlebury hockey coach put Skoglund on his wish list. According to
Skoglund, “Being a hockey player gave me access to a first-rate education.”
Research backs him on this. One study suggests that prestigious college
admissions committees give an advantage to recruited athletes equivalent to
an extra two hundred points on the SAT.6
So Skoglund went to the more prestigious Middlebury instead of, say,
the University of Vermont or Lake Forest College.7 From there, he was
admitted to the University of Illinois law school and went on to a clerkship
and two years as a litigator at a top Chicago firm before heading to the
Rockies. If he had gone to one of those other schools, he might have ended
up at a low-ranked law school and a less prestigious clerkship. The dream
job might have been out of reach.
Or would it? Nobody knows. We can never know the alternative path
Skoglund’s life would have taken had he gone to one of those other schools.
The “correlation is not causation” critique applies to Skoglund just as it did
to the studies of youth sports discussed earlier. Hockey helped him attend a
better college, but research is inconclusive about whether or not going to a
better undergraduate school has long-term positive benefits. People who go
to Harvard do better in the labor market, on average, than those who go to
the University of Massachusetts, but Harvard admits people who are
already on a faster track to success. We can’t tease out Harvard’s
contributions from the personal attributes its students brought with them.
The third way the youth sports–college combination can pay off
financially is through college scholarships.8 In 2019–2020, approximately
$4.2 billion of college scholarships were awarded to approximately 200,000
college athletes, an average of a little over $20,000 each. This means that
about 1 percent of American students receive at least partial athletic
scholarships. (Nearly 400,000 college athletes get no scholarship help.)9
College scholarships can be a huge benefit to some kids. But investing in
a child’s athletic endeavors just for the scholarship money probably isn’t
worth it. First, given that so few students get scholarships, and that most
who do get only partial scholarships, the expected value for any one
aspiring athlete is small. Second, the financial returns to a college education
are huge. For a typical kid, paying the cost of college is a great investment
that pays off. The savings from a scholarship are small compared to the
benefits of the education itself.
None of these first three options provides an especially persuasive
rationale for investing in youth sports, leaving just one way youth sports
can pay off: they provide a set of skills that, together with a college
education, lead to greater earnings later in life. We can certainly find
examples to support this idea; Stanford’s football team, for instance, has a
program to hook players up with Silicon Valley firms and other employers,
and the pipeline from top college lacrosse programs to Wall Street is so
strong that one article claimed “having lacrosse on a résumé is a major
advantage to grads entering the world of finance.”10 The logic comes down
to some combination of two factors. First, sports provide a network of
teammates, competitors, and fans who promote one another’s careers.
Second, the discipline that makes someone a star running back or lacrosse
defender also makes that person a great consultant or trader.
If it seems plausible that sports and college might mix in a way that
makes people more money later in life, and if plenty of anecdotes support
the idea, does that mean you should encourage your kids to play sports in
anticipation of financial benefits later?
 One thing is clear: if playing sports gets kids to stay in school and makes
them more likely to go to college, then youth sports can be justified solely
on financial grounds. As I noted, it is well established that more education
leads to higher income. One study showed that low-income teenagers who
can commute to college, sparing them the cost of housing, fare better than
comparable low-income teenagers who live far from a college and cannot
afford higher education. Another study showed that sets of identical twins
with different levels of education earned incomes that correlated with how
much school they attended. A third study sampled students in states that
allow teenagers to drop out of high school at a fixed age and showed that
people whose birthdays forced them to stay in school an additional year
earned more money, on average, than their counterparts who were able to
skip the extra year.11 Plenty of other cleverly designed studies, covering
many countries and eras, have yielded the same results. Skeptics can
certainly brainstorm confounding factors for any given study, but the
evidence makes it clear that more education causes higher salaries.
The bottom line is that youth sports probably help a small but nontrivial
set of American kids get more and better education than they otherwise
would have, and this education pays off later in life. If hockey led Matt
Skoglund to get more years of schooling than he would otherwise have
gotten (for example, if hockey allowed him to get into Middlebury, and
going to Middlebury led him to go to law school), then hockey improved
his career path.

What If Your Kid Is Kevin Durant?

Living near Palo Alto, I see lots of parents who encourage their kids to
play sports with no illusion that those kids will someday earn livings as pro
athletes. These parents focus on their kids’ formal education, for two
primary reasons. First, the kids come from backgrounds that put them in a
good position to do well at top American universities and quite possibly
continue on to graduate school. These parents have every reason to think
their kids are on track to the good life that goes with education: relatively
high income, low risk of unemployment, better-than-average health, and
other benefits. Second, while many of these kids are good athletes, few
have the natural talents required to excel in professional sports. Some
“option value” exists that any given kid will take his or her mediocre
athletic genes and develop into a skilled athlete. But at some point, usually
before a kid reaches high school, it becomes pretty clear to everyone
(though the kid may take longer to catch on) that he or she will never run,
jump, throw, or kick like LeBron James, Lionel Messi, or Serena Williams.
Privileged kids from the suburbs generally play sports for fun, memories,
and (maybe) the benefits of socialization and discipline.
But what if your kid is Kevin Durant? A working-class kid born in
Washington, D.C., in 1988, Durant had a childhood very different from that
of a Palo Alto Little Leaguer. His mother, Wanda Pratt, was relatively
educated (she eventually earned a B.S. from Strayer University), but shortly
after Durant was born, she became a single mother of two kids. Durant was
raised by his mother and grandmother, and although they did not live in
abject poverty, he did not grow up with many creature comforts or
professional career opportunities.12
Durant’s youth was focused on basketball—surely at some cost to his
schooling. Obviously those investments paid off, as Durant now earns
approximately $75 million per year in NBA salary and endorsements. But
can parents of a kid like Durant justify the decision to expend resources on
the kid’s basketball future? Clearly the investments Durant and his family
made look like great decisions in retrospect. But were they wise at the time?
For Durant, the two main reasons the typical Palo Alto kid should not
focus on preparing for a sports career did not apply. First, coming from an
underprivileged background and also, as a Black man, facing a labor market
marked by systemic racism and historical discrimination, Durant’s career
prospects outside basketball were not promising. Fewer than 20 percent of
African Americans graduate from college.13 A Black man of working age
who holds a full-time job earns an average of about $43,000 per year (half
of what a typical college graduate earns), but only about 60 percent of
Black men work full-time. The other 40 percent earn much less or zero.
Coming from a family with modest resources further eclipsed Durant’s
labor-market prospects, as economic research shows that economic mobility
to a better standard of living is slow in the United States, and much slower
for African Americans. One recent study found that Black Americans “have
substantially lower rates of upward mobility . . . than whites, leading to
persistent disparities across generations.”14 Thus, in the society into which
he was born, Durant’s family might have rationally anticipated that
basketball would be a higher-value play for him than it would be for a white
family that was financially comfortable. Second, and even more important,
Durant showed athletic promise and was unusually tall from a very young
age.
Even so, Durant and his family had to decide that he would invest in
basketball, and this investment surely came at some cost to his schooling.
His godfather recalled asking, “Kevin, is this something you seriously want
to do? OK, we’re going to have to put in a lot of work.” He made Durant go
to daily training sessions and read basketball books, and also made him
“write the six steps to a jump shot 500 times.” He added, “But mostly, there
were drills.”15
As early as his junior year of high school, Durant reached a height of six
feet eight inches.16 That alone changes the odds of earning a living playing
basketball. His height guaranteed he would be at least four standard
deviations above the average height of an American man (he eventually
reached six feet ten inches). That makes Durant taller than more than 99.99
percent of men: a man of his height is one in more than ten thousand.
Let’s do some math to see what Durant’s chances were. When Durant
was in high school, approximately 320,000 African American boys who
were born in 1988 lived in the United States. Of these, the numbers suggest
that approximately forty-five would ever reach the height of six feet eight
inches (Durant’s height at the time, which is also the average height of
players in the NBA).17 Though he and his family could not have known this
when he was in high school, a total of fifty Black men born in 1988
eventually played at least some minutes in the NBA. Twenty-one of these
players had guaranteed contracts for the 2015–2016 season—the peak of
their careers. Eight of them were at least six feet eight inches tall. Given
that eight out of about forty-five Black men born in 1988 who were that tall
had large NBA contracts, others spent time in the NBA, and still others
made a healthy living playing basketball overseas, the odds that an athletic,
hardworking kid as tall as Durant could earn a living playing basketball
start to look pretty good.
Note, however, that this calculation changes dramatically as soon as an
NBA wannabe concludes he will never be taller than about six feet. Almost
no NBA players are that short, and given that lots of people are around six
feet tall, the fraction of them who reach the NBA is essentially zero.
 Another important caveat is that although Durant came from a humble
background, one can’t be sure his nonbasketball opportunities would have
been limited to the low income that his demographics suggested. Given
how hard he worked to develop his basketball skills, it seems safe to
conclude that he has an unusual work ethic that could have paid off in other
fields. Moreover, he is clearly very intelligent, and he is interested in
education—although he left the University of Texas after his freshman
season, when he was the second pick in the NBA draft, he continued to
work toward a degree in the off-season.18 Thus, it’s possible that he could
have gotten a good education and had a successful career in another field,
although his chances of earning as much as he has earned from basketball
were roughly zero. But even if he did not expect the riches he has achieved,
for a six-foot-eight kid with limited family circumstances, investing time
and effort in basketball seems like a pretty reasonable risk.
If we expand our view to include all of Durant’s birth cohort, of the
roughly 320,000 Black males born in 1988, almost 300 played in the NBA
or the NFL at some point, and just under 200 were on an NBA or NFL
roster in the fall of 2015. (I am focusing on that year because complete
information for both athletes and the general U.S. population is available
for research.) The odds that a kid from this cohort grew up to be a pro are
slight (about one in a thousand), but not multimillion-dollar-lottery slight.
When we focus only on those who show signs of talent in high school—the
point at which a focus on sports begins to be more costly—the odds look
much more reasonable.
What are the returns for success? As you might imagine, they are
substantial.19 In 2015, the median income of an African American male
born in 1988 was about $15,600, and the average was a little over $20,000.
For the 2015 NFL season and the 2015–2016 NBA season, 184 African
Americans born in 1988 spent some time on a team’s roster, earning a total
of $410 million in salary. This figure does not include endorsements, but it
also does not deduct the agents’ share of this income (3 to 4 percent in the
NFL and NBA). These calculations lead to some startling conclusions:

1. Between 6 and 7 percent of all the money earned by American Black

men born in 1988 was earned by NFL and NBA players. This suggests
that the top 0.1 percent of the income distribution for this specific
group earns approximately the same share of the group’s total income
 as the top 0.1 percent of all Americans earn as a share of the total
income in America.20
2. Including his endorsement income, Kevin Durant’s annual income
represented almost 1 percent of all the money earned by the more than
300,000 African American men born in 1988.
3. The incomes of the NFL stars Gerald McCoy and LeSean McCoy (no
relation) each represented approximately 0.5 percent of the total
income of African American men born in 1988.

I focused on men who were twenty-seven to twenty-eight years old because

they were old enough to have signed large contracts but few had reached
retirement age. I found no evidence of any other Black men born in 1988
who earned eight-figure incomes.21
Although the data clearly show that investing in sports makes sense for a
small set of athletically gifted (and large) kids from difficult backgrounds,
it’s important to note that Durant was a perfect storm in terms of height and
lack of alternatives to basketball. Two other Black men born in 1988
reached elite NBA status despite not standing out quite as much on one or
both of these dimensions.
Russell Westbrook came from as humble a background as Durant’s.
Westbrook’s father instilled a love of basketball in young Russell, who
devoted a great deal of time to developing his skills. But at the point in his
high school career where colleges start recruiting players, Westbrook was
well under six feet tall and not an especially accurate shooter. His father
advised Russell to “outwork them.”22 This strategy paid off: Westbrook
grew to six feet three inches (big enough, though still small by NBA
standards) and became a great player on all dimensions. But even though
his father’s advice paid off handsomely, it is quite possible that for the
average kid who fit Westbrook’s profile at age fifteen, the better advice
would be to outwork his peers in the classroom rather than on the court.
Steph Curry was also late to grow into his eventual six-foot-three-inch
frame, remaining short and scrawny well into college. But, unlike
Westbrook and Durant, Curry—the son of the former NBA player Dell
Curry—was raised in a well-to-do household. The economist Seth
Stephens-Davidowitz has shown that, among NBA players, African
Americans from relatively wealthy neighborhoods (like Curry) are
overrepresented.23 Although plenty of NBA players like Durant and
Westbrook became superstars despite growing up with limited resources,
many others, like Curry, had affluent childhoods. Unattractive opportunities
outside sports are a good reason for kids like Durant to focus on sports, but
they face the reality that kids from wealthier backgrounds are more likely to
have professional coaches and good equipment.
Kevin Durant’s family made a good decision to invest in basketball on
his behalf, and the Palo Alto parents made a good decision to invest more in
their kids’ formal education. The question is, Where do you draw the line?
Who else can justify going all out to invest in a life in sports?
It’s actually very hard to come up with another group that combines the
two key features highlighted earlier—high probability of success as a pro
athlete and unattractive career prospects in other fields—as well as men
from humble backgrounds who are extremely tall or (focusing on the NFL
prospects) extremely large and muscular or extremely fast and dexterous.
Tall Black women are heavily overrepresented in the WNBA as well, but
the financial value of WNBA success is meager compared to success in the
men’s game. Whereas Kevin Durant earns almost 1 percent of all the money
earned by American Black men born in 1988, no WNBA player earns even
0.01 percent of the money earned by women of her race in her birth year.
Dominican baseball players are heavily overrepresented in Major League
Baseball, and soccer players from Senegal are overrepresented in the
French Ligue 1 (soccer). These people certainly have limited financial
prospects in their own countries. But unlike predicting future NFL and
NBA stars, telling who will become great is much harder when these
players are young. Outliers based on physical size do not have such a clear
advantage in baseball or soccer, and few kids who seem like superstars in
their communities will ever receive a paycheck (or even a college
scholarship) for playing their sport.
Sadly, not everyone understands the low probability of success. Some
parents get blinded by dreams of their kids’ stardom. A New York Times
Magazine article described how one affluent Chinese family moved across
the country to a premier golf training facility, where their three-year-old
son, Xie Chengfeng, began to practice ceaselessly with the help and
encouragement of his parents, plus coaches, tutors, and other adults whose
attention had been acquired for the right price. Xie’s case isn’t an isolated
one: in small but growing numbers, wealthy Chinese families are betting
incredible sums of money and time that their children will someday make
millions on the PGA Tour.24
Xie’s family’s investment in his golf future may be extreme, but many
kids around the world, including the shortstops of Santo Domingo, the
soccer players in Dakar, and girls playing tennis in Russia and Serbia, share
similar dreams. Most of them will look back as adults and wish they had
spent more time preparing for a more conventional life, but a few (maybe
even Xie, who subsequently moved to Southern California and has done
very well on the junior golf circuit) will reach their goals and inspire the
next generation to spend their childhood years trying to become superstars.

OceanofPDF.com
 2
What Do Silicon Valley and Czech Women’s Tennis
Have in Common?

Liechtenstein, a tiny principality between Switzerland and Austria with

just thirty-seven thousand inhabitants spread across sixty-two square miles,
is disproportionately blessed. For starters, the lure of its low corporate tax
rate has given it the highest gross domestic product (GDP) per capita of any
country in the world. Even better, Liechtenstein has no standing army, so its
taxpayers get to keep more of that wealth than citizens of larger countries,
which tend to spend substantially on their militaries. And somehow,
minuscule Liechtenstein manages to be the world’s leading manufacturer of
false teeth.
Of more interest to us, however, is that Liechtenstein is home to Tina
Weirather, a skier who was a fixture on the World Cup circuit in both the
downhill and super G events throughout the 2010s. Weirather, whose
Austrian mother and Liechtensteiner father were both World Cup skiers,
was internationally competitive from an early age. She moved in and out of
the top ten in the women’s Alpine skiing world rankings and took home the
bronze medal in the super G in the 2018 Winter Olympics. Despite being
her home country’s only athlete of international caliber during her career,
she made Liechtenstein, on a per capita basis, the greatest sports
powerhouse in the world.
Liechtenstein’s statistical dominance of global athletics is well
documented, if you know where to look. The country places first on the list
of all-time Olympic medal-winning countries per capita; its nine medals, all
in Alpine skiing, give Liechtenstein about three times more medals per
capita than the second-place country, Norway. China and the United States
churn out hundreds of athletic superstars across dozens of disciplines:
China has earned 0.449 Olympic medals per million inhabitants, while the
United States fares much better at 9.02. But that output pales in comparison
to Liechtenstein’s 274 medals per million.1 The disparity points to just how
unprecedented it is for a group of thirty-seven thousand people to
consistently produce one or two world-beating skiers every decade or so.
But Liechtenstein has a natural advantage that is far more helpful to its
skiers than low tax rates, demilitarization, or even false teeth: the entire
country has easy access to snow and mountains.
Given this geography, it makes sense that Liechtenstein can boast such
high rates of athletic success. Liechtenstein is one of several countries to
channel either natural resources or strategic investment—or both—into an
advantage in a single sport, as the dominance of Norwegian cross-country
skiers, American basketball players, and East African marathoners makes
clear.

Measuring Geographic Dominance

Stories about a country dominating a single sport are interesting, but I
wanted to be a bit more systematic and quantitative in assessing dynasties.
With that in mind, I developed a standardized measure of national athletic
dominance to allow for reasonable comparisons across countries and sports.
The Population-Adjusted Power Index, or PAPI, tells us a country’s strength
in a given sport relative to its population. To measure PAPI, I tracked down
the top 25 rankings in sixteen individual sports that have international
men’s and women’s sports associations.2 Then I counted the nationalities of
the members of the top 25 on each of the 32 circuits. Finally, I found the
ratio between a country’s representation in a given top 25 and its overall
population.
Consider a few groups with high scores. In my data set, Romania has
one of the world’s top 25 women’s skeleton racers; Poland has two top 25
men’s kayakers; and the United States has 16 of the world’s top 25 men’s
golfers. The three countries’ shares of the world population, from smallest
to largest, are 0.28 percent, 0.54 percent, and 4.44 percent. Dividing the
first number (share of the top 25 in the given sport) by the second (share of
world population) yields the countries’ PAPI scores in their respective
sports: all about 14.5. Romania’s single top-25 skeleton racer gives it 14.5
times as many top-25 women’s skeleton racers as it would have if its share
of the world’s top-25 skeleton racers matched its share of the world’s
population. Likewise, if top-25 men’s golfers were distributed across
countries in the same proportions as the world’s population, the United
States would have a little more than one. Its actual number, 16 of the top 25,
is about 14.5 times its expected share.
Thus, although the raw numbers differ greatly, Romanian women’s
skeleton, Polish men’s kayak, and American men’s golf are all, by PAPI, at
about the same strength. For comparison’s sake, the top PAPI in my sample
(Liechtenstein’s women’s Alpine skiers) is 7,718. Note that a PAPI of 1.0
means that a country has exactly the same share of top-25 athletes in a sport
as its share of the world’s population. For example, China has 5 of the top
25 male badminton players, and since it has about one-fifth of the world’s
population, China’s PAPI in men’s badminton is very close to 1. The vast
majority of PAPIs are zero, given that most countries have zero athletes in
any given top-25 ranking.

The Napa Valley of Scandinavia

The example of skiing in Liechtenstein is a classic case of “comparative
advantage”—that is, it is easy for Liechtensteiners to be better at skiing
because, compared with people almost everywhere else, they have easy
access to mountains and snow. That doesn’t mean Liechtenstein was
guaranteed to be good at skiing; it could have limited resources for training
or other constraints that would hold potential skiers back. But the country’s
advantage is “comparative” because, holding other things constant, it is
comparatively easy for Liechtensteiners to be good skiers relative to being,
for example, good badminton players, and relative to how easy it would be
for a Costa Rican to become a good skier.
The idea of comparative advantage extends far beyond sports; it is a
broad and important economic force that drives much of the geography of
business. When asked to name the single most important insight in
economics, a nontrivial number of academic economists will choose
comparative advantage. Chilean copper, Vietnamese fish, Thai rice, and the
wines of California, Australia, and New Zealand (all of which have climates
ideally suited to grape growing) reflect the same economic phenomenon as
Liechtensteiner skiers. Just as resources and climate make it sensible for
Chile to mine copper and Thailand to grow rice (and then to use trade to
make both countries better off), Liechtenstein produces more than its share
of skiers, and the United States an abundance of golfers. Sports fans the
world over are better off because they can watch top athletes who have
come from the best breeding grounds.
Of course, Liechtenstein is not alone (or even unusual) in having a
climate and topography favorable to skiing. Take its not-quite-neighbor
Norway. Norway has plenty of snow and plenty of mountains, giving it a
similar natural advantage in downhill skiing. But other countries, such as
Liechtenstein, have their ski slopes far closer to population centers. In
Norway, the mountains are relatively remote, and that distance is just one of
several obstacles for aspiring Norwegian downhill skiers: Norway is far
enough north that it gets limited daylight, a big problem for downhill skiers,
and it is also extremely cold, which makes sitting in a chairlift pretty
uncomfortable.
So although Norway has not abandoned downhill—in fact, the
Norwegian downhill skiing program is fairly strong, with two men and one
woman among the top 25 skiers—it has not focused on the sport to the
same extent as countries like Liechtenstein. Rather, Norway has parlayed its
challenges in downhill (compared with several other snowy countries) into
strengths in Nordic, or cross-country, skiing. To practice effectively, cross-
country skiers need less daylight than downhill skiers; they stay warm more
easily because they are moving constantly; and they do not need access to
mountains. Norwegian population centers tend to be close to wilderness
(more so than, for example, those in Sweden), which makes it easy for
Norwegians to practice cross-country skiing.3 These characteristics make
cross-country skiing a potential gold mine for the Norwegian athletic
program.
And Norway has cashed in on this comparative advantage. Home to less
than one-tenth of 1 percent of the world’s population (and fewer people
than Minnesota), Norway took home more than half of the gold medals and
more than a third of all medals awarded for cross-country skiing at the 2018
Winter Olympics. Its men’s and women’s cross-country skiers attained the
second- and third-highest PAPI scores, respectively, of all the national
athletic programs I tracked: of the top 25 cross-country skiers worldwide,
ten men and eight women were Norwegian, for respective PAPI scores of
561 and 449. That is, Norway has as many top-25 cross-country men skiers
as we would expect, proportionally, from a country 561 times Norway’s
size. Norwegian cross-country skiers trail only Liechtenstein’s women’s
skiers in the PAPI sample—in other words, they trail only Tina Weirather.
 Although it isn’t surprising that Norway is good at cross-country skiing,
it’s a little surprising that Norwegian skiers are so good, given that Russia,
Finland, and Sweden have similar advantages. Even though each of those
nations has multiple top-25 cross-country skiers, they don’t come near
Norway’s dominance. Why? Because Norway has made strategic use of
some complementary advantages.
As it happens, in addition to snow, Norway owns a bounty of another
valuable resource: oil. Sustained exportation of that oil has allowed Norway
to become nearly as wealthy as Liechtenstein on a per capita basis while
maintaining a much larger (if still fairly small) population. At the same
time, Norway’s government is as socialistic as any in the developed world,
so that the government controls much of the country’s substantial wealth
and can invest it strategically. One of the government’s many social
programs channels a hefty sum (by law, a minimum of about $300 million
per year) into building youth athletic facilities and paying good wages to
highly skilled coaches who identify and develop talented skiers from an
early age.4
The Norwegian Confederation of Sports (NIF), the government agency
that oversees Norway’s Olympic committee and its broader investment in
sports, manages 54 federations, 19 regional sports confederations, 366 local
sports councils, and over 11,000 local sports clubs. These organizations are
funded with the help of revenues from the national lottery.5 Local sports
clubs form the core of Norwegians’ sports experiences when they are
young, training them and identifying those who will move on to elite
programs. Norway has six high schools known as the Norwegian College of
Elite Sport, employing seventy-nine coaches recruited from around the
world.6 From there, many athletes move to full-time training centers.
Although the earlier discussion of youth sports suggests that large
investments are not necessarily a good idea for individuals, the country of
Norway clearly believes they are valuable at a national level.
 Figure 1. Average PAPI for Alpine skiing, Nordic skiing, and skeleton
(Illustration by Julio C. Franco)

This financial and cultural investment gives Norway a big advantage

over other countries in sports. But combining that investment with
Norway’s natural advantages makes it an unprecedented powerhouse.
Plenty of countries have natural advantages in certain sports, but few
exploit them as well as Norway does in cross-country skiing. In this sense,
Norway is the Napa Valley of cross-country skiing. Napa Valley wineries
have excellent natural conditions for growing grapes, and thanks to a
surplus of Northern California millionaires looking for an expensive hobby,
the wineries have the resources to turn those grapes into world-class wines.
Napa Valley wines and Norwegian cross-country skiers may not offer the
best return on investment, but both combine their climate-based
comparative advantage with a wealth of resources to become the best (or
among the best) in the world.
The importance of comparative advantage can be shown graphically
(figure 1). On this world map, the darkness of each country corresponds to
the average PAPI in three sports where wintry conditions confer a large
advantage: Alpine (downhill) skiing, Nordic (cross-country) skiing, and
skeleton. Virtually all the darkness is above the fiftieth parallel, and PAPIs
are basically zero for any country not well above the equator.
The same map can be used to diagram all sixteen of the sports in the
PAPI sample (figure 2). The Nordic countries and Canada still do very well
by this broader measure, but the rest of the world does much better. The two
figures together bring out the importance of comparative advantage; the
Nordic countries and Canada are excellent at sports in general, but they are
particularly excellent at winter sports.
 Figure 2. Average PAPI for sixteen individual sports
(Illustration by Julio C. Franco)

When Running Marathons Is the Best Option

Far away from Norway is the most dominant group in any sport: East
African marathoners. Although Norwegian cross-country skiers are a story
of comparative advantage enhanced by plentiful complementary resources
(that is, oil money), East African marathoners are a story of geographic
comparative advantage enhanced by lack of complementary resources.
Their dominance is truly absolute. A while back, when I first looked up
the top 25 male marathoners, the Italian runner Daniele Meucci was the sole
representative of any country not named Ethiopia or Kenya. Meucci ranked
twenty-fifth. The top 24 male marathon runners in the world all came from
Ethiopia or Kenya. When I checked back a few months later, Meucci had
slipped, and the highest-ranked marathoner not from East Africa was tied
for forty-fifth with two Kenyans. Female marathoners from East Africa are
dominant by any metric other than comparison to their male counterparts:
20 of the top 25 women’s marathoners come from Ethiopia and Kenya,
including the top 12.7
Kenyan male marathoners rank a strong twenty-ninth in PAPI—and
Kenyan women, as well as Ethiopian men and women, also have
respectable scores—but PAPI does not do justice to the dominance of East
African marathoners. Kenya, with 44 million people, and Ethiopia, with 94
million, both have populations large enough to eliminate them from PAPI
competition with countries like Liechtenstein. A more apt measure is the
Herfindahl index, used by economists to measure how concentrated an
industry is. Unlike PAPI, which measures a single country’s dominance
relative to its population, the Herfindahl index measures concentration
within a small set of countries. Among all sport-gender combinations,
marathon has the highest Herfindahl index of all sports for both genders.8 In
fact, the marathon “industry” scored higher on the Herfindahl index than
the beer or breakfast cereal industries, traditionally two of the most
concentrated in the world. As cartels go, Kenyan and Ethiopian
marathoners put OPEC to shame.
A statistic that assesses an entire country’s dominance of marathon,
however, does not do proper justice to the Kalenjin tribe and its close
relatives, which together produce almost all East African runners. David
Epstein, the author of a book about the science of athletes, told an
interviewer: “There are 17 American men in history who have run under
2:10 in the marathon. There were 32 Kalenjin who did it in October of
2011.”9
How can this small part of Africa dominate a sport so thoroughly? Social
scientists, biological scientists, and sports pundits have long debated the
question. To an economist, two factors appear to be critical: geographic
advantage and a lack of alternatives.
First, parts of Kenya and the surrounding region provide a natural
advantage to native runners similar to what Norwegians experience in
cross-country skiing. Kalenjin live at high altitudes, which confers an
advantage for effective and efficient practice of endurance sports. The local
weather is conducive to an active life and being outdoors. Schools are often
far away, so kids get used to walking (and, according to some reports,
running barefoot) for long distances. Some scholars have argued that the
local environment leads to a carbohydrate-heavy diet, which may help in
endurance training.10
There is also a cultural explanation, arguably related to the natural
environment. Many Kalenjin children go through an initiation ceremony
that involves crawling through brush, getting beaten, and anesthesia-free
circumcision. This and other cultural factors, some argue, condition
Kalenjins to be able to run through pain. According to a report by NPR, “In
traditional Kalenjin society, pushing through pain isn’t only a desired trait,
it’s also part of what makes you a man or a woman.”11
In addition to these natural advantages (speculative and otherwise), lack
of resources motivates East African runners’ development. For running, the
equipment and land costs of training are trivial—unlike for sports like
sailing or golf—so that even people from relatively low-income areas are
not at a disadvantage. If Kenya’s weather and altitude (especially in the
areas where Kalenjin live) were conducive to developing skiers, then the
locals, lacking the resources to pay for facilities, equipment, and specialized
coaches, might not be able to take advantage. But even a poor country can
afford to develop runners.
Motivation matters too. Norwegian cross-country coaches may find it
difficult to get a Norwegian teenager to spend excessive amounts of time
skiing when that kid is starting to worry about getting into college and
eventually getting a good job. Rural East African children typically do not
have bright career prospects and thus have strong incentives to become
runners. Daniel Lieberman, a Harvard biology professor who has studied
East African runners, says he thinks the most important reason for their
success is the region’s widespread poverty and (as an economist might put
it) the low opportunity cost of pursuing running. “There are almost no jobs
apart from subsistence farmers,” Lieberman says. The author Adharanand
Finn adds, “In Kenya the life of an athlete is one of relative comfort. Eat,
sleep and run. It beats digging the earth all day with a hand plough.”12
If Norway is the Napa Valley of cross-country skiing—geographic
advantage combined with plenty of resources—then East Africa is the
Nauru of marathon running. Nauru, an island nation in the central Pacific,
had large reserves of phosphates thanks to many centuries of built-up bird
droppings. Early in the twentieth century, Nauru began to mine these
deposits, which meant that the island was literally depleted. The residents of
Nauru didn’t particularly want to turn their island into a strip mine, but
having few other economic opportunities, they went with their comparative
advantage. Many Kenyan and Ethiopian runners, if they had better access to
education and careers, would no doubt choose business, medicine, or
another profession. But given that the most readily available alternative is
subsistence farming, many find that the best option is to pursue running.
 Martina Navratilova Is Hewlett. And Packard.
The Czech Republic is home to about ten million people. The country
has cold winters and plenty of rainfall. It has no natural advantage relevant
to producing tennis players. Yet Czech women’s tennis ranks 25th in PAPI
out of 356. There are enough Czech women in the top 25 to proportionately
represent a country 108 times larger than the Czech Republic.
How did the Czech Republic become such a force in women’s tennis
without any comparative advantage? Simple: over decades, the country has
developed a huge network of Czechs trying to become tennis players. As is
usually the case when a network pops up in an environment that gives it no
natural advantage, the growth was driven by historical circumstance. Back
when the Czech Republic was part of the Austro-Hungarian Empire (and
even before), the local nobility took to tennis. Competitive tennis in what is
now the Czech Republic began in 1879, just two years after the first
tournament at Wimbledon. Tennis, and sports generally, has always been
culturally important among Czechs. There have always been plenty of great
Czech tennis players of both genders, but the country’s most gifted male
athletes often gravitate to hockey or soccer. Although some Czech girls are
drawn away by winter sports, at which Czechs also excel, it seems that a
higher proportion of the most athletically talented dedicate themselves to
tennis.
Thus, three factors have come together to make the Czech Republic a
women’s tennis powerhouse despite the lack of natural advantage: focus on
the game through local clubs, parents, and coaches; investment and focus
during the Soviet era; and Martina Navratilova.
Jan Kodes, a Czech tennis star who won three Grand Slam events in the
early 1970s, described his country’s success this way: “Our history goes
back to 1893 with the first lawn tennis club. And there’s the background of
the small club. There’s always somebody to tell you how to hold the racket
and hit the ball. They may be unknown people, unknown coaches, who
know more about the game than some top-name coaches.”13
Hana Mandlikova, a four-time Grand Slam winner who was born in
Prague, grew up in this system. Her father attributed her success, and the
success of Czech tennis generally, to the quality of coaches, a focus on
basic technique, and “fanatic parents” (including himself) who religiously
took their children to far-flung tournaments run by the tennis federation.
“Nowhere during my travels have I seen things for children so well
organized as here,” he told Sports Illustrated.14 The Czech legend Martina
Navratilova has said that the club system and the early and constant
competitions were responsible for later generations of Czech women’s
success, noting, “In these clubs, you hang out. . . . You become really
competitive and learn the sport the way it should be taught, which is
through competition.”15
Navratilova’s comment points to one thing that makes Czech girls so
good at tennis: other Czech girls. Having a sport on which people focus, be
it cross-country skiing, marathon running, or tennis, brings out kids’
competitive spirit. Not only do they want to win, but they have a supply of
friends and competitors against whom to hone their skills. Competition
leads to innovation, as economists have long emphasized. Tougher
competition makes athletes better just as it makes companies better. They
have to find new ways to beat the competition. Czech girls learn to get
better by playing each other, and they build on each other’s improvements.
Much like the Norwegian government today, the Czech government in
the Soviet era made large investments in these clubs and the tennis system.
It also happily took credit for the players’ achievements, writing that “the
main cause of success is the support of the Czechoslovak Central
Committee.”16 Clubs spread throughout the country, and talent scouts
searched the elementary schools for potential champions as young as six.
The most highly skilled players moved to a network of training centers,
housed within existing clubs, where they were professionally coached,
subjected to rigorous practice regimens, and given the best equipment. At
any time, up to ninety thousand players played in the Czech club system
and about sixty teenagers in the super-elite program.17
But a key reason why the Czech youth system turned out the current
crop of top women’s tennis players is Martina Navratilova. One of the all-
time great female tennis players, Navratilova won 18 Grand Slam singles
titles and another 41 between women’s and mixed doubles. Her 59 total
Grand Slam titles are second all-time to Australia’s Margaret Court, and
Navratilova’s battles with the American player Chris Evert—now an ESPN
commentator—elevated women’s tennis to new heights of viewership and
prize money. Navratilova inspired countless Czech girls to go down to the
local tennis club and give it a try. As a result, the American numerical
advantage was largely undone; there were a lot fewer Czech girls than
American girls, but a large proportion of Czech girls play tennis, while the
Americans play basketball, softball, or golf. A small country can have a big
presence if the infrastructure for success exists and a large fraction of the
population is drawn to the sport.
Petra Kvitova, who has won two Wimbledon championships and has
been ranked as high as second in the world, is a typical example of the
Czech tennis legacy’s impact. She was born in Bilovec, a small town
several hours from Prague. According to her website, “Her family spent a
lot of time playing tennis at the local club,” and her father “used to show
Petra videos of Martina Navratilova playing at Wimbledon, which she
would watch with fascination.” Kvitova writes, “I have so much respect for
Martina, I watched her on television when I was a child, and that’s where I
learnt about Wimbledon and playing on grass.”18 Kvitova showed great
promise and at age sixteen moved to a full-time tennis school (the club
system and elite schools outlived the Soviet era). After becoming a
successful pro, Kvitova paid homage to Navratilova in a series called “My
Hero” for Tennis magazine.19
To get a better sense of Czech tennis players’ geographic dominance in
the absence of any natural advantage, think of the clustering of major
technology companies in Silicon Valley. Like the early Czech competitive
tennis leagues, Stanford University developed a strength in technical and
business education, creating a steady supply of both new companies and
technical talent to staff them; like the Czech leagues, Stanford’s presence
provided infrastructure in a relatively narrow field in a relatively small
region. In 1939, two Stanford graduates, William Hewlett and David
Packard, founded a company in Palo Alto, and the growth of Hewlett-
Packard and the network of technology companies that started up in the
surrounding area made Silicon Valley what it is today. Hewlett and Packard
did for Silicon Valley what Martina Navratilova did for Czech women’s
tennis: they took an existing infrastructure and started a self-sustaining
network around it. In both cases, no natural advantage existed (they don’t
actually mine silicon in Silicon Valley; the industry could just as easily have
developed in Seattle or Saint Louis), but that didn’t stop the region from
becoming dominant in the field.
 An Improbably Dominant Dynasty
If Norway’s cross-country ski dominance shows how natural advantage
can fuse with complementary economic factors to create a niche, and the
Czech Republic’s success in women’s tennis shows how happenstance and
investments can do the same, then South Korea’s unprecedented emergence
in a seemingly random sport falls somewhere in between. South Korean
women do not dominate golf to quite the degree that East Africans
dominate marathons, but they are perhaps the most improbable persistent
sports dynasty of all.
South Korea has more than 1,300 people per square mile, making it the
fourteenth-densest country in the world (and third, behind Bangladesh and
Taiwan, if you ignore microscopic countries like Vatican City and
Barbados); such urban cramming leaves little space for athletic facilities.
South Korea also has few natural resources beyond a handful of mines in
the northern part of the country. It has no obvious advantage in either sports
or economics.
Still, South Korea has become a rich country and has managed to find
success in an array of sports, from archery to modern pentathlon to judo.
Over the last few decades, South Korea has placed in the top ten medal
winners at most Olympic Games, both summer and winter. But its Olympic
success is dwarfed by its dominance of one sport: over a twenty-three-year
period starting with the first Korean woman ever to win a golf major
tournament, twenty different Korean women have won thirty-four major
golf titles (and three more were won by women who were Korean citizens
at birth). In a recent nine-year run, South Koreans claimed twenty-two of
the forty-three (yes, more than half) major championship titles in LPGA
golf.
Granted, South Korean women’s golfers rank a pedestrian fifty-third in
PAPI. PAPI is weighted toward countries with very small populations that
can produce a few elite athletes in one sport. Even so, the South Korean
women’s golf PAPI of 62 is impressive—the country has as many top-25
golfers as we would expect from a country 62 times its size. Their PAPI,
which ranks second among countries with at least 15 million inhabitants—
behind Kenyan male marathoners—is particularly impressive given the
country’s rapid ascension from nonentity in the sport to powerhouse. This
seemingly unnatural phenomenon is explained in large part by a unique
collection of economic and cultural conditions that primed South Korea for
a women’s golf boom.
First, South Korean children are believed to be the most focused and
overworked in the developed world. Perhaps not coincidentally, they are
also the most miserable. An international survey by the Paris-based
Organisation for Economic Co-operation and Development (OECD) placed
Korea dead last in its assessment of children’s happiness in school. Of all
the countries surveyed, Korean parents had the highest expectations that
their children would graduate from college, but the lowest rate of eating
dinner with their children.20 Every November, graduating South Korean
high school students take the CSAT, an aptitude test that largely determines
college placement. Results are considered so important that, on testing day,
the stock market opens late so that parents can support test takers before
going to work, and flights are canceled to avoid noise distractions. Students
who underperform their CSAT expectations have a significantly increased
risk of suicide. South Korea, more broadly, has the highest suicide rate of
any OECD country by far.21
Gender inequality is also dramatic in South Korea. The country placed
108th out of 153 countries sampled in the World Economic Forum’s Global
Gender Gap Report, beating only Turkey among developed nations. The
Economist gives an even harsher evaluation, placing South Korea dead last
among twenty-nine countries in its glass-ceiling index. Though South
Korea elected a female president in 2013, Korean women occupy senior
jobs at astonishingly low rates overall.22
On a more positive note, Korea historically has one of the highest
savings rates in the world. Of thirty OECD countries, its savings rate was
recently second highest (to Norway), and it has been at or near the top of
this ranking for decades.23 This consistent standing likely reflects a cultural
inclination to forgo gratification today for security tomorrow.
Korea thus offers an interesting set of extremes: significant wealth and
high savings alongside unhappy children and staggering gender inequality.
This combination of attributes, it turns out, may be ideal for breeding
world-class female golfers. Nothing increases one’s likelihood of becoming
a pro athlete like consistent, intense work, which is expected of Korean
children in any field. But in Korea, girls get substantially less out of an
investment in school than boys do: the 37 percent pay gap between genders
provides less incentive for girls to push themselves academically.24 For
Korean girls, then, taking time away from schoolwork to practice a sport
might seem like a good investment. Korean boys have reason to study hard
and acquire the skills that will help them in the professional marketplace;
girls have more incentive to forgo that study and do something else, like
practice golf.
This reasoning is borne out empirically by the country’s lack of similar
success in men’s golf. Korean men’s golfing glory is limited to Y. E. Yang’s
anomalous PGA Championship title in 2009—he was ranked 110th in the
world at the time and quickly returned to obscurity—and veteran pro K. J.
Choi’s decent but unspectacular career. Recently, the highest-ranked Korean
man was eighteenth in the world, and the country had four of the top one
hundred men’s golfers. At the same time, a Korean woman (Jin-Young Ko)
was number one in the world; 5 of the top 10 women golfers and 36 of the
top 100 were Korean.25 To be fair, another factor contributing to the gender
disparity in golf performance is Korea’s two-year military service
requirement for men, but this likely isn’t the sole barrier. Choi, for instance,
worked as a sentry for the army during the day and practiced golf at night.
But why golf? Well, for one thing, it’s not just golf. South Korean
women have high PAPIs, for example, in archery (with seven of the world’s
top 25 archers) and short-track speed skating (with four of the top 10). If
you look carefully, you can see that Korea is quite dark in figure 2’s
summary of PAPIs across sports, mainly thanks to its women athletes. But
Korea’s strength in golf may be partially due to the success of Korean
golfer Se Ri Pak. Much like the Czech tennis star Martina Navratilova, Pak
was a trailblazer in the Korean women’s golf world. Her win at the U.S.
Women’s Open in 1998 was the first major championship for a Korean
golfer. Although she was less of a pioneer than Navratilova—other Korean
major champions were well on their way to golf success before Pak’s
breakthrough—her success functioned in a similar way. Many of today’s
Korean golf stars cite Pak as an inspiration. For example, So Yeon Ryu, a
former U.S. Women’s Open champion, said, “After Se Ri won the U.S.
Women’s Open, I really got interested in golf.”26 Seven-time major
champion Inbee Park said that her parents took her to the driving range for
the first time right after Se Ri Pak’s first major victory.27
 In the case of South Korea, then, preexisting economic circumstances
created the potential for geographic success, which was catalyzed by the
success of Se Ri Pak. Together, these conditions likely helped take the
nation’s women’s golf elite to the next level of dominance.
Sometimes an area becomes dominant through natural advantages:
Norwegian cross-country skiers, Napa Valley wineries. But some titans of
sports and industry simply come from the right alignment of stars.
Coincidences happen and even have long-term consequences. South Korean
women dominate golf because of savings, gender discrimination, and a
pioneer. Czech women punch above their weight in tennis because of the
tastes of rich people a few hundred years ago and the success of one woman
forty years ago. A small peninsula in Northern California evolved from the
world’s largest apricot grower into the world’s leading high-tech region
because one local university produced students who opened some computer
and semiconductor firms there.
Once these dynasties become established through some combination of
luck and coincidence, they become hard to topple. I would bet that Czech
women tennis players and South Korean women golfers will still be
dominant ten years from now because advantages, both in sports and in the
economy, tend to persist. For example, the United States has enjoyed a
perennial advantage in world basketball competition in the same way that
Silicon Valley now has a huge lead in technology. Other countries have
developed their basketball programs and made some progress relative to
Team USA, just as technology clusters in Austin (Texas), Skolkovo (outside
Moscow), and Tel Aviv (Israel) have made strides relative to Silicon Valley.
Still, the big leads and reinforcing network effects of Team USA basketball
and Silicon Valley—and Czech women’s tennis—will take decades to
match.

OceanofPDF.com
 3
Why Do Athletes Cheat and Lie?

In the summer of 1998, Mark McGwire and Sammy Sosa were living
legends. Hot in pursuit of Roger Maris’s single-season home run record, the
two sluggers traded long bomb after long bomb as the year wore on and
Maris’s sixty-one-homer mark grew closer. The story helped to rejuvenate
baseball in the wake of its crippling strike in 1994: two media-friendly
stars, each quick to heap praise on the other, were chasing one of the sport’s
most storied records. The pinnacle of the months-long spectacle came on
September 8, when McGwire, playing for the Saint Louis Cardinals, hit his
record-breaking sixty-second home run against Sosa’s Chicago Cubs. As
McGwire rounded the bases, Sosa famously jogged in from the outfield to
congratulate him at home plate, a lasting image of sportsmanship and
respect that was hailed as an encapsulation of everything good about
baseball. McGwire ended the year with seventy home runs, Sosa with sixty-
six, and both were immortalized for their roles in one of the most exciting
baseball seasons ever.
The next time McGwire and Sosa made headlines together was in March
2005. This time, the venue was not the friendly confines of Wrigley Field in
Chicago but the august halls of the U.S. Capitol, where they were testifying
at a congressional hearing. As two of the most prominent baseball stars of
recent years, they had been called before the House of Representatives
Committee on Government Reform to testify about growing concerns over
the use of performance-enhancing drugs in baseball. Neither man, it turned
out, had much to say. McGwire repeatedly told the congressional
committee, “I’m not here to talk about the past,” while Sosa claimed that he
did not speak English well enough to testify and issued a terse statement
through his interpreter that he had never used steroids.1
The congressional hearing only marked the beginning of baseball’s
internal war on drugs. On December 13, 2007, with some fanfare, former
senator George Mitchell issued his commissioned report on the steroid
problem. At the end of the report’s four-hundred-odd pages, he included
among his findings and recommendations the following lament: “The
players who follow the law and the rules . . . are faced with the painful
choice of either being placed at a competitive disadvantage or becoming
illegal users themselves. No one should have to make that choice.”
Here Mitchell hit the nail on the head. A number of factors presumably
go into deciding whether or not to take steroids—including health concerns,
religious beliefs, and willingness to lie—but one consideration clearly at or
near the top of the list is how these drugs will affect the athlete’s
performance relative to his peers. If McGwire was on steroids and Sosa
wanted to keep pace, he would probably have had to take them too—and
vice versa.
In economics terms, McGwire and Sosa faced a “prisoner’s dilemma,” a
situation that can turn otherwise reasonable people into drug users, cheaters,
and liars.

Can a Clean Cyclist Win the Tour de France?

Lance Armstrong, a seven-time Tour de France champion and winner of
the Best Male Athlete ESPY and Sports Illustrated Sportsman of the Year
awards, is no longer a popular man. He makes perennial appearances
(sometimes at the top) on Forbes’s annual “most disliked athlete” list. Once
sponsored by national brands such as Nike, Michelob Ultra, and 24 Hour
Fitness, Armstrong now struggles to find even small bike companies willing
to pay for his endorsement. His already-battered image suffered a knockout
blow when he confessed his longtime use of banned substances to Oprah
Winfrey on national television, confirming an unpleasant truth that was
quickly becoming undeniable in the face of a mountain of evidence. The
title of his best-selling book, It’s Not About the Bike, took on an ironic new
meaning when, for Armstrong, it became all about the doping.
But is Armstrong really such a bad guy? Or simply a victim of
circumstance?
No economist would condone Armstrong’s behavior, but plenty of
economists would call it a predictable response to the incentives he faced.
To see why, imagine you are a talented young cyclist who has been winning
junior competitions for years. You’re about to participate in your first Tour
de France, excited to finally get your shot on the big stage. As soon as the
race begins, however, you realize there’s something different at this level.
It’s no surprise that all the athletes have intense workout regimens, tightly
regulated diets, and cutting-edge equipment. But you quickly gather that
they’re also all using performance-enhancing drugs, something you’ve
never been exposed to before. Should you give up and find a new career, or
suck it up and start injecting yourself?
Some people would walk away. Cheating and lying about illegal drug
use is a psychologically taxing behavior for most people, and those
emotional costs might be so high that some athletes would walk away from
a career in which success requires such consistent dishonesty. But for many
others, the rational choice is easy in a different way. After years of
investment in cycling, it seems only logical to start doping. If everyone is
doing it, then you have no choice if you want to remain competitive. There
are some ethical hang-ups, but cyclists certainly wouldn’t be the first group
of professionals to abandon their collective morals for the sake of their
respective careers.
But how did this happen in the first place? How did the sport get to the
point where all the top cyclists are alleged to be doping? When the Tour de
France started in 1903, performance-enhancing drugs (PEDs) didn’t exist,
but that doesn’t mean everyone rode clean. Ingesting substances has been a
part of the Tour from the beginning. A top competitor was disqualified from
the inaugural race for riding in the slipstream of a car, and nine riders were
bounced the following year for taking cars and trains as shortcuts. They also
used substances. One recent reporter, looking at the Tour’s history,
concluded, “From the earliest days of the race, entrants drank to deaden the
monotony, huffed ether to tamp down the pain, took amphetamines for
stamina, or opiates to ease aches and cramps.” The 1923 winner told a
reporter, “Cocaine to go in our eyes, chloroform for our gums, and do you
want to see the pills?”2
Drugs were not banned until 1965, commencing an ongoing arms race in
which riders seek new and innovative ways to cheat while Tour officials try
(futilely, for the most part) to catch cheaters by means of more sophisticated
testing. We can turn to the notorious prisoner’s dilemma to get some insight
into why riders cheat.
Suppose the police bring in two suspects who are believed to have
jointly committed a robbery and place them in separate rooms. A detective
goes into one room, then the other, and tells each prisoner that he has two
choices: he can keep his mouth shut, which would incur a one-year
sentence, or he can rat the other guy out, which would allow him to go free
while the other guy serves a five-year sentence. The only catch? If both
guys rat, they’ll both be stuck with three-year sentences.
As you may know, the prisoner’s dilemma has an elegant solution. Every
prisoner, every time, should choose to rat. Whatever the other prisoner
does, ratting is the better option: if the other guy keeps his mouth shut, the
first guy goes free if he rats his partner out, but serves a year in jail if he too
keeps his mouth shut. Meanwhile, if the other guy rats, the first guy can
either rat too and take the three-year sentence, or keep his mouth shut and
be stuck doing five years. The reason it’s considered a dilemma is that the
best collective choice is for both prisoners to keep quiet; then they serve a
total of two years between them, versus a minimum of five years under all
other options.3
Now consider a cyclist’s dilemma. Let’s start by making a few
assumptions that probably approximate real-life conditions. First, let’s say
doping increases each cyclist’s speed by enough that if he takes steroids
before the race, he can expect to finish ahead of other slightly faster cyclists
who would otherwise beat him. Second, let’s assume that some chance
exists that a doping cyclist will be caught during a given race, but a much
higher probability exists that he can use masking techniques that let him get
away with PED use. Finally, we’ll assume (safely) that most of the top
cyclists will make a lot more money if they are successful in cycling than
they would in any other profession available to them.
Now consider two equally skilled cyclists deciding whether to use PEDs
under these conditions. Each knows that his competitor faces the same
choice but doesn’t want to risk giving away his own strategy by talking to
him about it. The cyclists face a Hamlet-like quandary: to dope or not to
dope. Just as the prisoners are always best-off ratting on each other, the
cyclists’ best choice (also known as a “dominant strategy”) is always to
dope no matter what the other guy does. If one cyclist dopes, the other guy
is better off doping because that increases his chances of winning from
virtually zero to fifty-fifty. If one cyclist does not dope, the other guy is
better off doping because it increases his chances of winning from 50 to
nearly 100 percent. The only way doping is not a dominant strategy is if the
cyclist in question incurs an extremely high psychological cost from lying
and cheating (that is, if lying and cheating make him feel really bad). In that
case, the cyclist will choose not to dope because his total payoff (winnings
minus psychic costs of lying and cheating) is higher when he does not dope
than when he does. Although such cyclists surely exist, it only takes a few
less-scrupulous types to create a world in which all the top cyclists are
doping.
Why don’t the cyclists simply reach a sportsmanlike agreement not to
dope? After all, given the negative long-term health consequences of
doping, it’s in everyone’s best interests for both cyclists to abstain. If
neither dopes, they each can expect to do about as well relative to each
other, and the fans can enjoy an equally competitive (if somewhat slower)
race.
The answer is a matter of sheer volume. Although the prisoner’s
dilemma contains only two agents, who operate in a kind of vacuum, the
real-world cyclist’s dilemma involves not just two cyclists but hundreds. As
long as many of them have access to performance-enhancing technology,
only one cyclist needs to begin doping before others (eventually, virtually
all others) feel that they have to.
Is cycling a true prisoner’s dilemma? Has the sport degenerated to the
point where you can safely assume that any winner of a major race is
doping? Was Armstrong not a villain but merely the king of an era in which
competition remained fair but was elevated to enhanced levels?
The length and location of the Tour de France course change every year,
but we can compare performance across years by looking at average speed.
As equipment improved, the winner’s average speed increased steadily and
significantly from the race’s early years into the mid-1950s. Between 1956,
when the Frenchman Robert Walkowiak won at an average speed of about
22.5 miles per hour, and 1985, when the Frenchman Bernard Hinault won at
about the same speed, the winning speed rose and fell based mainly on the
course.
After 1985, however, racing speed took off to previously unthinkable
highs. Since 1991, no cyclist has won the Tour with an average speed of
less than 23.8 miles per hour. Certainly, some of the uptick is attributable to
technology, but given the rules put in place in the 1990s by cycling’s
governing body, Union Cycliste Internationale, to limit overly advanced
bikes, it seems unlikely that the unprecedented speeds were due to better
frames and wheels alone.
 The former cyclist Steve Swart recalled the changes in his memoir: “In
1987 and ’88, it was one world. In 1994 it had completely changed. The
increase in speed was unbelievable. . . . They [weren’t] using the same
gears as before.”4 Swart had abstained from doping as he raced in the less
competitive American tour. When he returned to the European circuit, in
1994, he began injecting EPO. “Look,” the once steadfastly antidoping
Swart later told a group of young cyclists that included Armstrong’s future
teammate George Hincapie, “if you’re going to make it in this game, you’re
going to have to do that stuff, simple as that.”5 He seems to have been right:
a report by the United States Anti-Doping Agency concluded, “Twenty of
the twenty-one podium finishers in the Tour de France from 1999 through
2005 have been directly tied to likely doping through admissions, sanctions,
public investigations or exceeding the UCI hematocrit [red blood cell
portion] threshold.”6 The Tour winners in 1996, ’97, and ’98 all used PEDs
as well.
In short, then, the incentives to cheat in cycling are—or at least were—
so strong that virtually everyone at the top of the sport seems to end up
doing it. Although the consensus is that doping in cycling has declined
since more aggressive testing and policing were put in place in 2008, there
is an equally strong consensus that doping remains a substantial factor in
elite cycling. One authority concluded, “The general view is that at the elite
level the situation has improved, but that doping is still taking place.”7 The
average speed at the Tour de France has stabilized, but it has not dropped
from the days of Lance Armstrong and other notorious cheaters. We can
safely assume that many winners of top cycling races are doping—or, as an
economist would think of it, following the dominant strategy.

Does Crime Pay?

For baseball, meanwhile, the Mitchell Report marked a turning point.
Cynics complained that the report was merely a publicity move to distract
from the league’s lack of interest in taking substantive action against
steroids. But the report’s release coincided with a sharp decline in steroid
use. Before 1996, the five league-leading home run hitters had averaged
forty-five home runs or more just three times in over a hundred years. In
twelve consecutive seasons beginning in 1996, the top five in each league
averaged more than forty-five home runs every year—a total of twenty-four
times. At the end of that period, before the 2008 season, the Mitchell Report
came out, and home run production dropped off a cliff. In the following six
years, the five leading home run hitters didn’t average more than forty-five
home runs even once.
The Mitchell Report named about eighty players as steroid users. The list
identified a number of MVPs and Cy Young Award winners, including
Barry Bonds, Roger Clemens, and Jose Canseco. Shortly after the release of
the Mitchell Report, reports spread of a second, sealed list of 104 players
who had tested positive during an experimental testing period in 2003.
Leaks from unnamed sources eventually pointed to the superstars Alex
Rodriguez and David Ortiz as being among the ostensibly anonymous 104
positive tests on the second list.
For big stars like these, the incentives to use steroids were not entirely
financial. Take Bonds, for instance. Reporters who covered his fall from
grace paint a picture of a rabidly jealous man driven to steroid use by his
envy of McGwire’s and Sosa’s spotlight during their home run race in 1998.
Bonds began to take performance-enhancing drugs around then, as he knew
McGwire and Sosa already were, and became noticeably bulked up. But
Bonds was already on track to become a historically great player before his
biceps spontaneously doubled in size. In the 1998 season, he significantly
outperformed Sosa and matched or edged McGwire using advanced
statistical measures of productivity.8 The contract Bonds signed in 1993
was at the time the richest deal ever for a professional baseball player. By
the end of the 1998 season, he had collected three MVP awards, been
selected to eight All-Star teams, and won eight Gold Gloves. Bonds had
little financial incentive to dope. Rather, his steroid use seems to have been
fueled as much by an ego-driven need to secure his place in baseball
history. Even perennial All-Star players have a higher level to aspire to, and
sometimes it can be enough to set them down the path of using drugs.
The public, though fascinated by Bonds, was less interested in his
teammate Bobby Estalella. A career .216 hitter, Estalella shared only two
traits with Bonds: they had baseball in their genes (Bonds’s father and
Estalella’s grandfather were both Major League All-Stars), and both were
linked to steroids in the Mitchell Report.
Estalella’s prisoner’s dilemma had different payoffs from Bonds’s, but
the outcome and the dominant strategy were the same. Estalella was among
a sizable tier of players sometimes referred to as AAAA-players: those who
have bounced between the major and minor leagues more than a couple of
times and spend most of their careers dealing with the fear of demotion.
Most are interchangeable in the eyes of fans, but for these players, the
financial value of staying in the big leagues is huge. The minimum salary in
MLB in 2001 (the year Bonds set the single-season home run record) was
$200,000, and the average MLB salary was around $2 million. A player in
the minor leagues at that time, by contrast, would have been lucky to earn
$75,000 per year. (Today the minimum major-league salary is over
$500,000, and the median is over $1 million.) Estalella played in the minor
leagues in all but two of the nine years that he saw some major-league
action. He earned up to $550,000 per year by staying in the majors much of
the time.
Estalella had the option of abstaining from steroids, but doing so would
have made it more likely that he would play in minor-league obscurity, ride
buses to games, and maybe earn a decent but far from lucrative living. A
second option was to take steroids, collect the much higher paychecks that
went with playing in the majors, and—at least for a few years—travel to
games in chartered jets.9
Estalella knew he would most likely never have the chance to earn so
much money again. Many other marginal players face the same choice. It
doesn’t take long for one borderline pro to start using steroids, and then for
others to follow. They would prefer not to, and all might agree to stay clean
if such an agreement were credible. But when their best financial option for
the rest of their lives is on the line, they have a strong incentive to bite the
bullet and use steroids to avoid getting left behind. It’s a dominant strategy.
When a small difference in quality makes a big difference in income, the
incentive to cheat becomes very strong.

Trust but Verify, or Just Give Up

When the incentive to take performance-enhancing drugs is strong
enough, some athletes will make the wrong choice and corrupt their sport.
What can be done about it? Unfortunately, the answer is probably “not
much.” Two key elements often help clean up prisoner’s dilemma–like
situations: verifiability and repeated interactions with potential penalties.
These approaches, however, are difficult to establish for PED use in sports.
 Let’s start with verification. “Trust but verify” is a mantra often used by
politicians tasked with setting up arms agreements.10 During the Cold War,
for example, the United States and the Soviet Union each built up massive
arsenals of nuclear weapons before eventually signing treaties to reduce
their respective stockpiles. A prisoner’s dilemma ensued in which each
nation’s best natural strategy was to break the deal and keep building its
arsenal—either to use against a weakly defended, treaty-observing enemy
or to protect itself against a similarly treaty-defiant one. The treaties
therefore needed to stipulate how each side could verify that the other was
following the terms. Ultimately the treaties would succeed only if the
verification provisions allowing each side to detect cheating were strong
enough.
Although the stakes are lower, sports with PED problems have their own
verification procedure, drug testing, which they are constantly trying to
refine. Testing as verification should theoretically fix the doping problem:
just check athletes’ urine or blood for illegal substances periodically, and
the cheaters will quickly be driven out. But, as any casual fan knows, it’s
not quite that simple; testing efforts always seem to fall one step behind the
latest drug-masking technology.
Every so often, baseball’s tests nab an All-Star like Ryan Braun.
Meanwhile, though, dozens of other users avoid a positive result. Evidence
of this evasion emerges when cheaters who went unnoticed by testing labs
are caught by investigators or journalists who piece together the players’
connections with distributors. Indeed, MLB reported that of the twenty
players—including Braun—who were caught acquiring steroids in the
Biogenesis scandal of 2014, only four had previously been identified as
users by the league’s drug-testing authority.11
Still, testing has likely served as a deterrent to some degree; the specter
of a failed test must loom especially large for athletes consumed by their
public images. But the trickle of sluggers, sprinters, cyclists, and other
athletes who are nabbed for steroid use remains unchanged. It’s fair to
assume that, at least as yet, verification protocols offer an imperfect
solution for sports officials looking to combat their athletes’ PED prisoner’s
dilemma. What else can they try?
Stronger punishment sounds like the natural solution: change the payoffs
so that cheating is no longer a dominant strategy because the sanctions from
getting caught are overwhelming. But in many cases, that approach is
unlikely to work. Because testing seems to be so easy for clever dopers to
overcome, no penalty, even a severe one, is likely to be sufficient to swing
the balance against doping.
Given the limits of any testing-based system, it’s tempting to turn to an
easier solution: embrace reality. Sports would still be fun to watch if PEDs
were allowed. Athletic governing bodies that try to enforce antidrug rules
could follow the lead of NCAA men’s tennis officials, who simply gave up
on regulating cheating. In college tennis, players referee themselves; they
call balls in or out, and they determine whether the opponent’s serve hit the
net on the way over. If a player hits a fantastic serve that the other player
cannot return, the returning player can, by the traditional rules of tennis,
yell “let.” This indicates that the serve hit the net and must be redone. Some
lets are obvious, but often these nicks of the net are all but unnoticeable.
Many ethically challenged college players fell into the habit of claiming
that clean serves that they could not return were actually lets.
This type of cheating became a prisoner’s dilemma where even honest
kids felt they had to call some dubious lets to keep up with the cheaters they
faced. In response, the NCAA eliminated the let—a staple of tennis at all
other levels—so that any serve that landed in the service box, regardless of
whether it touched the net or not, was considered in play. Surely the NCAA
will soon be able to reverse this rule, as improvements in technology are
making it affordable to install devices that can identify when a serve hits the
net. In other words, once “verify” becomes an option, NCAA tennis can
revert to traditional rules without worrying about cheating.
Low-cost net-cord-detection devices would solve the NCAA tennis issue
once and for all. The situation is more complicated with PED detection,
though, because testers and cheaters are both evolving in a constant cat-
and-mouse game. So the only real alternative to the current system is for
sports officials to simply throw their hands up and give in to the dilemma.
Leaders of leagues particularly plagued by steroids could, like NCAA
tennis officials, concede and allow their players to do what they want. As
some observers point out, the benefits of PED use are not entirely different
from, say, training at altitude (though that argument ignores the long-term
health consequences of PED use). More important, those commentators say,
is recognizing that all other approaches have failed and that only
legalization creates a chance of leveling the playing field for all athletes.12
Although those who favor this “embrace and accept” strategy still seem to
be well in the minority, I suspect that sports fans and athletic governing
bodies may slowly put their ethical concerns aside and move toward it.
Economically speaking, it’s hard to conceive of a more elegant realistic
solution.

Is Usain Bolt for Real?

The only other sport that can claim to have been tainted by PED use
comparable to that in cycling and baseball must be sprinting. Well, okay,
also weightlifting, and football, and field events like discus. But in any
case, steroid use is a major problem in sprinting. Medalists in the sport’s
highest-profile event, the Olympic men’s 100-meter sprint, have repeatedly
been linked to steroids over the past three decades, and those who haven’t
been caught have remained under suspicion simply by association and their
inability to prove otherwise. The public’s cynicism toward sprinters is
reasonable even without the rash of positive tests from big-name sprinters:
the 100-meter sprint has seen a readily observable performance uptick,
comparable to the performance increases in cycling and baseball. Before the
1988 Olympics, only two Olympians had ever broken ten seconds in the
100-meter dash; since that year, only three medalists have not been under
ten seconds, and two of those times were run in the wind-hindered 1992
race in Barcelona.
At some point, runners clearly began to face the same prisoner’s
dilemma as cyclists. That point likely came around 1984, when the
infamous Canadian sprinter Ben Johnson first picked up a bronze medal. He
went on to win gold in 1988 before being brought down in a widely
publicized doping scandal that rocked the sport. The famous American
sprinter Carl Lewis, who beat Johnson for gold in 1984, has never been
proved to have doped, but he did record a positive drug test result in 1988
for which he was eventually cleared.13
Regardless of whether Lewis was or wasn’t guilty, plenty of other gold
medalists—from Linford Christie in 1992 to Justin Gatlin in 2006—were
caught doping at some point in their careers. Even sprinters with completely
clean testing records have had fingers pointed at them. Victor Conte,
ringleader of the notorious BALCO lab, a primary steroid provider to both
sprinters and baseball players, told an Australian newspaper that every
runner in the Sydney Olympics 100-meter final was using steroids. Those
athletes included two prominent sprinters, Maurice Green and Ato Boldon,
who enjoyed successful and ostensibly clean careers. Since the Sydney
Games, a leaked letter supposedly written by Boldon to a former coach
alleged that Green used steroids; Boldon himself once tested positive, but
track and field’s governing body cleared him of wrongdoing.
In recent years, the otherworldly Jamaican Usain Bolt dominated
sprinting from his emergence in 2005 until his retirement in 2017, rewriting
the sport’s record books. An athletic freak who has never tested positive,
Bolt calmly denied allegations while many of his rivals—such as the
Americans Tyson Gay and Justin Gatlin and the Jamaican Asafa Powell—
were caught and suspended for steroid use.
Could Bolt truly be clean? Conte doesn’t think so: he said in the same
interview that he “strongly suspected” Bolt was on steroids. Decisively
beating all competition on a level playing field is one thing. But were Bolt’s
talents so great that he could win decisively while running clean against
competitors who were cheating? If Bolt was indeed clean, his natural
sprinting talent would have had to be like no one else who ever laced up
sprinting spikes.
How much of an outlier would Bolt have had to be? To answer this
question, I need to establish a couple of baselines. First, I will estimate that
using PEDs lowers a given runner’s time in the 100-meter dash by at least a
tenth of a second and maybe close to a quarter of a second. I say this
because winning times in the event dropped by about a quarter of a second
between the late 1970s and the late 1980s, which appears to be the period
during which steroid use became common.14
Second, I will assume that, at any given moment, the fastest man in the
world is not typically faster than the second fastest by more than a tenth of a
second. The winner of the Olympic finals in the 100-meter dash rarely wins
by more than a tenth of a second. Though larger margins of victory were
somewhat more common in the pre-steroid era, when few people were
organizing their lives around the Olympics, winning by more than a tenth of
a second was an anomaly.
Bolt won the finals of the 2008, 2012, and 2016 Olympic 100-meter
dashes by twenty, twelve, and eight one-hundredths of a second,
respectively. Given that his competitors in the first two races have all since
been connected to steroids, to have achieved those winning margins, Bolt
would have to have been a quarter to a half second faster than the clean
versions of all the other runners.
Moreover, Bolt’s best time, 9.58 seconds, is a full fifth of a second better
than the best time ever run by any other runner who has not been credibly
connected to steroid use.15 That margin is the largest attained by any world
record breaker in the 100-meter dash. Bolt also, with his win in 2008, broke
the Olympic record by a wider margin (0.15 seconds) than any previous
record breaker had ever achieved. The last time a sprinter broke prior
records by so much was when Marion Jones tore up the women’s record
book in the late 1990s and early 2000s. Jones was later shown to have used
PEDs and was stripped of all her major titles. In 1988, Florence Griffith
Joyner broke the world records in both the 100- and 200-meter dashes by
about a quarter of a second after showing rapid, drastic improvement. She
has always been under strong suspicion of using steroids to make that leap,
though like Bolt, she has never been proved to have done so and maintains
she has always been clean.
Bolt denies PED use in the strongest possible terms. He has said, “I am
clean, I’m sure about that. I welcome people to test me every day if
necessary to prove it to the world. I have no problem.”16 His coach has said,
“We know he is as clean as a whistle.”
Unfortunately, these kinds of statements mean almost nothing. Such talk
is cheap and is just what athletes who regularly used PEDs said before they
were caught. Lance Armstrong, for example, declared publicly and
defiantly, “How many times do I have to say it? . . . Well, it can’t be any
clearer than ‘I’ve never taken drugs,’” and “I have never doped.”17 Serial
cheater and liar Alex Rodriguez responded to Katie Couric’s question on
national television, “For the record, have you ever used steroids, human
growth hormone, or any other performance-enhancing substance?” with a
simple “No” shortly before he first admitted to taking PEDs. Marion Jones,
months before definitive evidence forced her to admit longtime steroid use,
stated, “I’m confident in the near future my name will be cleared from this
whole situation.”
If Bolt truly never took a PED, it is a shame that the cheaters sowed
doubts in fans’ minds that can never fully be disproved. The world will only
know for sure whether Bolt used steroids if he is caught; there is no way to
ever be certain he is clean.
 Although the odds are astronomically against any sprinter being as much
better than the competition as Bolt is, it is theoretically feasible that he is
just more of an outlier than anyone who came before. As Ralph Mann,
director of sprints and hurdles for USA Track and Field, has argued on
Bolt’s behalf, “Every once in a while you’re going to get a genetic freak
among freaks—a Tiger Woods or a Michael Jordan in his sport. And we’ve
got one in Usain Bolt.”18 Anything’s possible, I guess.

Why Don’t NBA Players Get Caught Using PEDs?

Sprinting, cycling, baseball, football, and weightlifting are all widely
known to have massive steroid abuse issues and all fit the prisoner’s
dilemma framework. With some exceptions when drug testing has been
particularly aggressive and successful, top competitors benefited enough
from PED use that one can surmise that the winners (and, in many cases,
marginal players like Bobby Estalella) are getting chemical assistance.
That’s because these sports have three important characteristics. First, the
value added by PEDs is enough that the best players competing without
them cannot outperform second-tier players competing with PEDs. That is,
PEDs can be the factor that takes the second- (or third- or fifth- or tenth-)
best competitor and allows him or her to win. Second, optimal PED use in
these sports raises the visible suspicion that the person has unnatural muscle
mass. Ben Johnson, the banned sprinter, was given away by his arms, Barry
Bonds by his head and neck, and many Tour de France riders by their
freakish legs. Finally, in all these sports, the people who succeed do much
better (mostly financially) through their sport than they could possibly hope
to do in any other career. The incentives to succeed are overwhelming.
Why do we not hear so much about steroids in hockey, soccer,
basketball, golf, tennis, skiing, and bowling? One possibility is that athletes
in these sports are more honorable than those in baseball and weightlifting,
but that assumption seems pretty naive. And the third characteristic from
the list—these athletes make a lot more than they would make in other
careers—certainly holds in most of these sports. World-class golfers and
basketball players make millions of dollars per year, and even those in
skiing and bowling make a lot more than the average person. Given that
most of these athletes would not otherwise be investment bankers, it’s safe
to say that professional sports offer a large step up from what they can make
elsewhere.
 That leads me to focus on the first two factors: the marginal value of
PEDs and the visibility factor. Many people believe that PEDs do not help
athletes in endurance sports or in sports that are more “athletic” (such as
basketball or tennis) and less muscle based (such as football and cycling).
But PEDs can help athletes improve their strength and their recovery from
difficult workouts in any sport. A tennis player cannot bulk up like Barry
Bonds, because then the player would not be able to chase down drop shots.
But smaller doses could help that player’s muscles recuperate after a long
match or an intense practice session. A well-planned steroid regimen can
help almost any athlete. But it is less clear whether the benefits of PEDs to
a typical athlete in these situations are enough to outweigh the psychic costs
of cheating, the side effects, and the risk of getting caught. The benefits
certainly outweigh the costs for some athletes, as there are credible reports
of PED use in most of the sports mentioned earlier (not bowling). But it
appears that a solid majority of athletes do not think the upside of steroids is
worth it, because they can do fine without them.19 Steroid use in these
sports is not a prisoner’s dilemma, because it is not the dominant strategy
for any given player to choose to use PEDs.
Although the cost-benefit calculus is what determines whether PEDs
create a prisoner’s dilemma in a given sport (yes in cycling and sprinting,
no in skiing and hockey), the visibility issue is likely a crucial factor
affecting whether PED use becomes a scandal. If the athletes look so
freakish that people cannot accept them as natural, fans and analysts take
note. Some basketball players and tennis players probably use PEDs, but
they use them in smaller doses, to improve recovery and endurance rather
than to build huge muscles. Fans and commentators don’t see anything
unusual. Once in a while, an NBA player gets caught (for example, All-Star
Joakim Noah and near-All-Star O. J. Mayo were suspended for PED use),
but the NBA’s testing regimen is not particularly intense, and PED use is
not widely seen as a problem.
Economics offers two conclusions about drug use in sports. First, if you
strongly suspect that an athlete is doping, he or she probably is. At least in
some muscle-focused sports, the rewards for cheating and lying are huge,
and as long as at least a few people are willing to cross that line, others have
little choice but to follow. Lance Armstrong, Marion Jones, and others who
used PEDs in cycling and sprinting were only following their dominant
strategies. Second, just because you don’t hear much about drugs in a sport
doesn’t mean people are not using them to get an edge. Maybe there really
isn’t much to report, but it’s just as likely that nobody is digging deep
enough to find out. For athletes, the problem is simple: if a large number of
your opponents are cheating and you have no effective way to catch them,
you most likely will not win unless you cheat too.

OceanofPDF.com
 4
Are Athletes Worth All That Money?

Professional athletes make a lot of money. In 2019, the median income

for American men twenty-five years or older with full-time jobs was about
$52,000.1 That is, half of the full-time workforce made more than $52,000,
and half made less. The players in the four major American sports leagues,
as well as players in top international soccer leagues, do a lot better. For
instance, the median Major League Baseball salary in 2019 was $1.4
million. Similarly, the median NBA salary in the 2018–2019 season was
$2.65 million.2
It wasn’t always this way. Major League Baseball players in 1964 made
a median salary, adjusted for inflation, of $122,000.3 The median income of
all full-time employed male workers in 1964, again adjusted for inflation,
was $52,000.4 An MLB player made about twice as much as a typical
American male worker in 1964, but twenty-seven times as much as that
same worker in 2019. To put it another way, from 1964 to 2019, the typical
American male full-time worker’s pay has stayed unchanged, but the
typical MLB player’s income has grown by a factor of more than eleven.
A visual sense of the stark difference in the pay trajectories of average
workers and top professional athletes is shown clearly in graph form (figure
3); the graph focuses on MLB players, but other sports look similar. For
both MLB players and what the Census Bureau refers to as “production and
nonsupervisory” workers, average pay is set to one in 1967, and all pay
figures are adjusted for inflation. Over the next fifty years, the nonathletes’
pay barely budged from the 1967 level, but MLB pay went up steeply and
steadily.
 Figure 3. Growth in pay for MLB players and production workers (1967 = 1)
[Illustration by Julio C. Franco]

The growth in athletes’ pay, and some of the details regarding how that
pay has come to grow over time, suggests at least three questions. Why do
professional athletes make so much and, in particular, so much more than
they used to make? Why do baseball and basketball stars get large
guaranteed contracts, while most football stars are one bad season away
from losing their million-dollar paychecks? Finally, why do players and
owners fight so much over money, occasionally getting into such bitter
disagreements that part or all of a season is canceled?
The answers to these questions lie in the stories behind three numbers:
$430 million, $53 million, and $580 million, in that order.

The $430 Million Man

As the 2019 baseball season was approaching, Mike Trout was widely
considered the best baseball player in the world. Ever since he took the
sport by storm in his rookie season of 2012, Trout’s statistics had been
otherworldly. He had also been relatively injury free, missing a substantial
number of games in only one season. He had placed first or second in
American League Most Valuable Player voting in every season of his career
except for the one where he missed a quarter of the season, when he
finished fourth in the voting. On one of my favorite baseball podcasts,
Effectively Wild, the hosts regularly discussed the degree to which Trout
could be handicapped and remain a major-league-quality player. They had
long, spirited discussions, for example, about how good Trout would be if
he had to run the bases backward, how valuable he would be if he were
allowed only one swing each time he came to bat, and (I am not making this
up) whether he could fake his death midway through his career, assume a
new persona, and be inducted into the baseball Hall of Fame twice. In short,
Mike Trout was (and is) an outstanding baseball player.
Shortly before the start of the season, Trout and the Los Angeles Angels
agreed to a twelve-year contract for a total of $430 million. This was the
largest guaranteed contract in the history of North American sports, and it
entitles Trout to an average annual income of almost $36 million until he
turns thirty-nine (an age at which few hitters remain productive).
If Trout played in every single Angels game scheduled during the life of
the contract, he would have made an average of $211,000 per game. Every
time Trout spends a few hours at the ballpark, he takes home four times the
median annual salary of a full-time American male worker.
That average American worker might not think Trout is worth his pay.
That worker might even say it isn’t “fair” that he earns about $25 an hour,
has trouble paying his bills, and can’t do enough to help his kids get ahead
in life, while Trout makes millions for playing a game.
Even economists are human, so I can’t help but sympathize with that
view. A baseball player who excels at his craft does not do nearly as much
to serve others as a farmer, a nurse, or a schoolteacher. But those other
workers also do not create as much financial value as Trout does.
Trout makes so much money for two reasons: the league has lots of
money to spend, and he has scarce and valuable skills that make his team
better.
In the year Trout signed his record contract, MLB teams brought in a
total of almost $11 billion in revenue. ESPN, Fox, and Turner Sports,
combined, paid the league nearly $2 billion for the rights to broadcast
games nationally.5 MLB gets another $800 million annually from its
streaming and technology arm, MLB Advanced Media.6
Ticket sales add billions more. Given that the Angels play in a huge
media market, they usually draw large crowds—and though it is hard to
determine what fraction of those fans come to see Trout play, star players
do draw people in.7 Some teams hope that signing a player like Trout will
pay for itself through a larger local TV deal. This logic doesn’t happen to
apply in Trout’s case, however, because shortly before his rookie season,
the Angels signed a huge TV contract that will last through Trout’s whole
career. But TV revenue is a relevant factor in other large contracts.
Given all that money, what allocates so much of those riches to Trout is
just supply and demand. It may not be “fair,” but it is what the market will
bear. Remember, economics is the study of scarce resources, and the ability
to hit forty home runs per season in Major League Baseball is one of the
scarcest resources there is.
The semiofficial currency in which baseball talent is valued is “wins.”
Teams will pay for anything they think will give them more wins. Winning
games sells tickets, leads to higher fees for TV rights, and brings in revenue
from playoff ticket and merchandise sales.
The market for wins is much like the market for stocks. Investors bet on
a stock because they think they will see future returns based on the
company’s good financial performance. Similarly, MLB teams invest in
players if the expected return in terms of wins (which can be translated back
into money) is high enough.
Back in 1974, one of the first prominent sports economists, Gerald
Scully, was the first (that I know of) to calculate major-league players’
value. He estimated that Hank Aaron, the biggest star of that era, would “be
the difference between victory and defeat in about 20 games” and was
worth $600,000 (in 1971 dollars) to his team. Scully’s calculations were
extremely innovative at the time, but in the decades that followed, the data
on baseball became much better. Sabermetricians—fans and analysts named
for the Society for American Baseball Research who set the early standard
for advanced statistical analysis of baseball—have become extremely
sophisticated at determining players’ relative value in terms of how much
they increase the expected success of their teams. One statistic they
developed that has become a standard for comparing players is “wins above
replacement,” or WAR. WAR estimates how many fewer games a player’s
team would win if that player were unavailable and the team instead used a
“replacement-level” player. Replacement players are generally the best
players in the minor leagues, so they are below average compared with
other players in the top league. Modern sabermetricians assign Aaron a
WAR of about eight for his play during the 1971 season, meaning that the
Atlanta Braves would have won eight fewer games that year had Aaron
been replaced by a right fielder from the top of the minor-league hierarchy.8
Throughout his career, Mike Trout has been at least as good as Aaron
was in 1971, averaging eight to ten wins above replacement for his team
every season. In 2018, as Trout’s previous contract neared its end, the
Angels wanted to ensure they could continue benefiting from his abilities.
Although they had Trout locked in for two more seasons, he could have
become a free agent at the end of 2020, meaning he could go to any team he
wanted. Looking at other recent contracts and projecting future league
revenues, the Angels could have figured that if Trout became a free agent,
he would be offered contracts paying $50 million or more per season. This
is because the baseball market, using methods similar to the stock market,
had determined that the price of one expected win was about $10.5 million,
according to an estimate in 2017 (which also projected the price would rise
to $12 million by 2019).9 Given Trout’s eight-plus win production, the $36
million per year he receives can be viewed as a steep discount (a point I will
return to).
So Trout is well worth his $430 million, or $36 million per year. But
why weren’t Ted Williams and Mickey Mantle, superstars of yesteryear,
worth $36 million per year when they played? Each was the highest-paid
major leaguer at some point, earning $90,000 per year.10 Adjusted for
inflation, that is the equivalent of about $760,000 (Mantle’s 1962 salary) or
$885,000 (Williams’s 1951 salary) in 2019 (the first year of Trout’s
contract). Although Mantle and Williams certainly made good money, they
were paid not even one-fortieth of what Trout earns. MLB players who
earned the equivalent of Mantle’s or Williams’s peak pay in 2019 include
Chris Hermann, a journeyman catcher with a career batting average of
about .200; Alex Wilson, a relief pitcher who generally comes into the
game when the outcome is largely decided; and J. B. Shuck, an outfielder
who has not played in more than half his team’s games in any year since a
promising 2013 rookie season. In that same season, an arbitrator awarded
Tommy Pham, a solid five-year veteran center fielder with career highs of
73 RBI and 23 home runs, a salary of $4.1 million. So a solid modern
player makes about five times as much as Williams and Mantle made at
their peaks.11
 This pay inflation is not unique to baseball. Similar increases have taken
place in all American team sports, in European football leagues, and in
individual sports such as tennis, golf, and auto racing. What changed? Why
has the level of pay in baseball and other sports increased by double-digit
multiples over the last fifty years? The increase can be explained by three
economic factors: technological change, competition, and income
distribution.
From the days of Williams and Mantle to the present, technology has
exerted what the economist Sherwin Rosen dubbed a “superstar effect.”
National and local networks pay billions of dollars to broadcast major-
league games, and MLB makes almost a billion dollars from internet rights.
Obviously it was not always this way. In the 1950s and ’60s, the internet
did not exist yet. Most households had television sets, but broadcasting live
sporting events was difficult and expensive, and the quality was limited.
Instead of watching on forty-plus-inch high-definition screens, people saw
sports on small black-and-white sets. Reception in much of the country was
blurry at best.
Over time, as broadcasts got better through the introduction of color,
more camera angles, instant replay, and other improvements, viewers’ best
option migrated from local, in-person events to watching games on TV.
Most sports fans would agree that the best baseball to watch is the major
leagues. But in the times of Mantle and Williams, a fan’s choice was
between a poor view of the best games and a better view of an inferior
game—like American Legion, semipro, or minor-league baseball—at the
local ballpark. Mantle and Williams were superstars, but watching them
play was not always the best option. With today’s high-definition TV,
though, Mike Trout and contemporary superstars get a much bigger share of
the baseball audience and so can command much larger salaries. As Rosen
and another economist, Allen Sanderson, put it, “A star player is worth only
a few dollars more per spectator than an ordinary player. There are lots of
spectators.”12
Not only is this superstar effect not limited to baseball—local leagues for
other sports have also been overtaken by people watching the world’s best
on TV—but it is not limited to sports. Consider Kevin Hart. He began as a
successful stand-up comedian playing small venues. If he had been born a
few decades earlier, that is probably how he would have spent most of his
career. But thanks to mass media like film and cable television, Hart makes
millions doing movies and stand-up specials on TV. Forbes estimates he
earned $57 million in one recent year, an unheard-of amount for a comedian
from a previous era who could make money only by touring.13 Even
superstars of classical music can cash in on the superstar effect. The cellist
Yo-Yo Ma makes millions of dollars each year from his recordings and his
advertising deals with Apple and Hyundai. In an earlier era, when classical
musicians had to build their audiences one show at a time, there was not
nearly such an extreme difference between the most famous cellist and one
comparably talented; both played concerts before local audiences. In short,
technological changes have expanded the reach of great musicians,
comedians, and athletes alike, raining down riches on the superstars.
The second reason Mike Trout makes so much more than Ted Williams
could ever dream of making is a seminal driver of economic activity:
competition. Williams, Mantle, and baseball players of their era can partly
blame their meager earnings on Oliver Wendell Holmes, who was chief
justice in 1922 when the U.S. Supreme Court ruled in Federal Baseball
Club v. National League. Holmes’s opinion in that case held that the
Sherman Antitrust Act of 1890, which barred monopolistic and
anticompetitive practices by American corporations, did not apply to MLB
because its “business is giving exhibitions of baseball, which are purely
state affairs.”14
This ruling was still in effect in the spring of 1956, when Mantle was on
his way to earning his first Most Valuable Player Award, and Williams—
though in the late stages of his storied career—was having another All-Star
season. While Mantle and Williams were busy on the field, the University
of Chicago economist Simon Rottenberg was busy doing research. That
June, Rottenberg published “The Baseball Players’ Labor Market” in the
Journal of Political Economy, and the academic field of sports economics
was born.15 Rottenberg identified the key constraint on players’ salaries as
the baseball teams’ exclusive rights to players, an agreement among teams
known as the Reserve Clause, which severely limited the players’ options.
Under the Reserve Clause, even after a player’s contract had run out, his
team retained exclusive rights to his services. Such a player, Rottenberg
wrote, “may withdraw from organized baseball and follow some other
calling, but he may not choose freely among bidders . . . within baseball.”
Rottenberg went on to argue that the justifications for limiting player
freedom (for example, that “high-revenue teams will contract all the stars,
leaving the others only the dregs of the supply”) were unsupported, and
even small-market teams could remain competitive. His prediction turned
out to be largely accurate.
Rottenberg’s paper generated an explosion of papers about sports
economics. It also, thanks to Curt Flood and Marvin Miller, presaged an
explosion in baseball salaries. Flood was a very good, if not great, baseball
player of the 1960s who thought that being tied to the team that drafted him
was unfair. When the Saint Louis Cardinals traded Flood to the Philadelphia
Phillies, he sued baseball for the right to be a free agent. The case again
went to the Supreme Court, which again ruled that the Sherman Act did not
apply to baseball. But the Flood decision was more tenuous than the earlier
decision, and Miller, the president of the MLB Players’ Association, was
able to exploit this development. Thanks to Miller’s work on behalf of a
few other players in succeeding years, teams’ rights to permanent control of
a player were eradicated.
Starting in 1975—when Miller’s efforts led to the end of baseball’s
Reserve Clause in an arbitration case—American professional sports
leagues and their players’ unions collectively bargained the rights of players
to move from one team to another. In all sports, after at least a few years of
league experience, players now become free agents when their contracts
expire, and they can then sign with any team they like.
Trout had a third choice beyond the two that were available to Mantle
and Williams—play for the team to which they were assigned, or quit
baseball entirely—he could become a free agent and sign with whatever
team offered him the best deal. Needless to say, that third option made a big
difference in his negotiation with the Angels.
The effects of competition on baseball salaries are clear from the rapid
rise in player pay after free agency was introduced. Jim “Catfish” Hunter,
the first free agent, signed with the New York Yankees in 1974 for $4.5
million over five years. This contract was far more lucrative than any that
had been signed before. Between 1964 and 1974, the top salary in the
league rose by 138 percent, but in the next ten years, with the advent of free
agency, it rose by 700 percent.16 That increase, the fastest the league has
ever seen, was mirrored by increases after free agency began in other
leagues, such as the NFL and the NBA.
The third reason Trout makes so much more than Mantle or Williams (in
addition to technology and competition) is income inequality. At the same
time that technology and the superstar effect were increasing pay for top
athletes and other stars relative to their less accomplished peers, those
trends were having the same effect on the U.S. economy (and many other
countries’ economies as well).
Over the past several decades, the American rich have gotten richer. The
fraction of American income earned by the top 1 percent of all wage earners
rose from around 10 percent in the 1960s and ’70s to nearly 25 percent in
recent years.17 From 1970 to 2011, the average pay for CEOs of firms in the
S&P 500 grew (after adjusting for inflation) from just over $1 million to
just under $11 million.18 The pay of top lawyers, hedge fund managers, and
other executives went up by similar amounts.19
These rich people increase the demand for expensive tickets and
experiences at sports events—both through their enormous wealth and
through their control of the tax-deductible expense accounts of the
businesses they run. Consider the New York Yankees, the team with
perhaps the greatest number of rich fans. Field box seats at Yankee Stadium
are great seats. In 1970, they cost an inflation-adjusted $26 in 2019
dollars.20 The same tickets cost $300 in 2019—more than eleven times as
much.
The Yankees and other teams have figured out even more lucrative ways
to pamper the 1 percent. Luxury boxes at Yankee Stadium, which provide
tickets, food, extremely comfortable seating, HDTV (you came all the way
to the park, but you can still see the game better on TV), and Wi-Fi for
twenty-two people, run between $8,500 and $20,000 per game. If the
Yankees sell a few of those boxes each home game, they can afford to pay
another superstar an eight-figure salary.21
That willingness to pay for great Yankees tickets has skyrocketed as the
wages of fans who want those tickets have skyrocketed—which, in turn, has
been a big contributor to the Yankees’ skyrocketing payroll. It’s a virtuous
circle of the rich getting richer.
In short, Mike Trout came along at the right time relative to Mickey
Mantle and Ted Williams. By the time Trout arrived in the major leagues,
technology had greatly widened baseball’s audience, players had been freed
from the indentured servitude of the Reserve Clause, and the United States
had plenty of extremely rich people looking to show themselves off at
sporting events. Trout has 430 million reasons to be grateful for these
trends.

The $360 Million Contract Hole

In 2020, about a year and a half after Mike Trout signed the largest
contract in baseball history, quarterback Patrick Mahomes signed a $502
million contact with the Kansas City Chiefs of the National Football League
—the largest contract in NFL history.22 Like Trout compared with Mantle,
Mahomes got a huge premium relative to Johnny Unitas and other NFL
quarterbacks of yesteryear. One analysis showed that the average NFL
salary grew by a factor of twelve, adjusted for inflation, from 1969 to
2013.23 The divide has only widened since then.
Even though Trout’s contract seems a bit smaller than Mahomes’s, most
people would (at least financially) prefer to be in Trout’s position. If at
some point during the life of the contract, Trout had an injury that made
him unable to play baseball, he would still get his $36 million per year. If he
decided he didn’t really enjoy playing baseball anymore and was just going
to go through the motions to fulfill his contract, he would still get the $36
million every year. His $430 million is guaranteed.
Mahomes, meanwhile, gets his $502 million only if things fall his way.
The Chiefs can cut him at any time and save about $30 million per year
from that point forward. Mahomes is guaranteed only $141 million in total,
or about $360 million less than the contract’s maximum value.
Why does this $360 million gulf exist? Why can the Chiefs drop
Mahomes and save millions, while the Angels will have to pay Trout no
matter what? The difference in the size and guarantees of NFL contracts
compared with contracts in other sports presents a difficult puzzle.
The natural response to the difference is to say that since injury rates are
higher in the NFL, it makes sense that football teams would be reluctant to
guarantee large contracts. But that response is, in a word, wrong.
Two features of long-term contracts explain the difference in guaranteed
money between Trout and Mahomes. First, such contracts allow teams and
players to plan. The Chiefs’ knowledge that Patrick Mahomes will be their
quarterback for several years, and the Angels’ knowledge that Mike Trout
will be their center fielder for several years, allow both teams to market
tickets and merchandise using the players’ respective images, sell expensive
apparel with their respective names and numbers on it, and make decisions
about other players that complement a great quarterback or center fielder.
Second, teams effectively act as their players’ insurance companies.
Athletes’ skills are like your house or your health: they represent a huge
part of an athlete’s total assets. If something happens to Mahomes’s ability
to play football or Trout’s ability to play baseball and they don’t have long-
term contracts, they’re in big trouble. Just as most people choose to buy
health and home insurance, Mahomes and Trout likely want insurance
policies on their athletic abilities. Those payments might come in very
handy someday—as they did for a prior holder of the NFL’s largest
contract, Colts quarterback Andrew Luck, who earned millions of dollars
despite missing the entire 2017 season.
Restated in economic terms, people who buy health or home insurance
are risk averse. They prefer to pay some amount of money to insure
themselves against a terrible outcome. For example, maybe your house is
worth $500,000, and there is a 0.1 percent chance it will be destroyed this
year. You are probably willing to pay $1,000 to insure your house this year,
even though your expected loss is only $500 (0.1 percent times the loss of
$500,000).
When it comes to player contracts, team owners are equivalent to the
insurance company, and athletes are the homeowners.24 The team holds a
diversified portfolio of football talent assets in the form of a set of players.
Any one player might have an injury or an off year or lose his ability to
play well. If that player has a long-term contract, the team takes the hit. It
hurts the team, but it hurts the team less than it would hurt the player.
Meanwhile, another player might overperform, and the team gets to keep
him at the contracted salary for the next season instead of paying him the
raise he could earn on the open market. Each player gives up some potential
profit so that the team will take on the risk.
That arrangement works for the players, and it works for the teams too—
but only if the long-term contracts pay less than what teams would pay if
the players were free agents each year. That is, players have to give the
team a discount to get the insurance policy, just as you pay an insurance
premium that is, on average, profitable for the insurance company.
Okay, I get it—this is the point that’s hard to swallow, because it is hard
to make the case that $430 million is a discount, even for a player like Mike
Trout. But it is. Writers who analyze baseball contracts have shown that
players sign for less than their current value to guarantee a long-term
income stream. Typically, a star will take a 10 percent discount off his true
value in exchange for more guaranteed years.25
With these ideas in mind, let’s get back to the puzzle of why Trout gets
more insurance than Mahomes. Our model of insurance would actually
predict the opposite; because the likelihood of a serious injury in football is
much greater than in baseball, football teams should be willing to insure
their players more than baseball teams. Remember: though a team will be
extremely unhappy if it owes millions to a player who cannot play, the team
is not nearly as unhappy as the player would be if he had to suffer the
financial consequences. That football is more dangerous, and football teams
have bigger rosters than baseball teams (thus spreading a team’s risk over a
larger set of players), suggests that football teams should provide more
insurance than baseball teams. Mahomes, knowing he has a greater risk of
injury, will give his team more of a discount if the team insures his earnings
than Trout would be willing to give.
Put another way, an insurance company takes all the downside risk out
of your homeownership. Guaranteed contracts take all the downside risk out
of athletes’ earnings. The value of that insurance should be greater for a
football player, and as a result, an economist would expect (all else being
equal) more guarantees to football players.
So if baseball players need insurance less than football players, why do
they get more of it? Why did the Angels take all of Mike Trout’s risk, but
the Chiefs take only a small portion of Pat Mahomes’s downside risk?
Though it’s impossible to prove why baseball players are better insured,
I have a hypothesis that squares with the facts and with basic economics.
Playing football is painful and difficult, and players must be given stronger
extrinsic rewards to willingly endure the ordeal. Perhaps this idea was
expressed best by the former NFL lineman Ross Tucker: “In 2004 I played
the last four games of the season for the Buffalo Bills knowing I had a
herniated disc in my back. I did it because of a sense of duty to my
teammates, because we were winning and, most importantly, because I had
a substantial playing time bonus if I were to play over 80 percent of the
offensive snaps, which I fortunately was able to hit.”26
Teams will be reluctant to guarantee big contracts if doing so leads
players to be less willing to give their all. A player might ask himself why
he works so hard when he gets paid the same either way. So one theory that
would reconcile the difference in the amount of insurance provided in
baseball and football could be that it is simply harder to play football than
to play baseball, and therefore financial incentives to perform are more
important. Mike Trout and his teammates do not need the same financial
incentive as Patrick Mahomes to play their hardest.
Contracts may insure players against financial loss, but they cannot
insure against the deteriorated quality of life that befalls many retired NFL
veterans who took too many hits. Maybe a wide receiver would simply not
be willing to jump for a ball in the middle of the field and expose himself to
a monster hit from the secondary if he didn’t need a few more great plays to
make sure he stayed on the roster the next year.
The NFL has seen several recent retirements by relatively young and
productive football players. The most prominent retiree was Andrew Luck,
a superstar quarterback for the Indianapolis Colts and a hometown favorite
of ours at Stanford, who quit before turning thirty, walking away from the
opportunity to earn tens of millions of dollars. These decisions suggest that
teams need constant incentives to keep players from taking the safer way
out on the field. Mike Trout does not have to put himself in those dangerous
situations, which likely explains why the Angels are less worried about
paying him a ton of money for more than a decade. To be sure, economists
have uncovered some evidence of “shirking” by baseball, basketball, and
soccer players after signing large, long-term contracts—but the evidence is
inconsistent, and the amount of shirking detected is generally not large.27
In sports as in life, the amount and the way people are paid come down
to balancing risk and insurance. The more difficult a job is, the more a
person has to be given incentives to do it. Just as employees with crummy
jobs have to be watched and run the risk of losing their jobs if they don’t
perform, football players have to be given extrinsic motivation to put
themselves in harm’s way. Those economic realities work well for Mike
Trout, but they could cost Patrick Mahomes as much as $360 million.

How to Flush $580 Million Down the Toilet

Professional teams and players can agree that players are worth large
amounts of money and that it makes sense, to varying degrees, to insure
players’ careers. The NBA has the highest average salary of the major
American sports leagues. NBA owners may not love paying out the
millions they do, but the owners are getting a good return on their money.
Why, then, did the NBA owners and players have a fight that led them to,
essentially, throw $580 million into a garbage can in the fall of 2011? The
answer is three simple letters: CBA.
Major American sports leagues exist thanks to arrangements made
between team owners (and the managers they employ) and the athletes who
play the games. The teams no longer have monopoly rights over a player’s
whole career. Players in every sport can become free agents after some
point, and they have other rights to avoid exploitation before they are
eligible for free agency. The players could simply work out contracts
individually with teams, and there would be a labor market not that
different from the market for lawyers, bankers, or architects.
Unlike most of these other groups, however, and largely as a result of the
history of exploitation in sports, athletes throughout American professional
team sports have voted to form unions. This creates two levels of
negotiation in each sport: individual teams and players negotiating
contracts, and the owners as a group negotiating an overall deal with the
players’ union.
Why would either group want that second level? Under American law, it
doesn’t matter if the teams want to do it this way or not. The National
Labor Relations Act of 1935 gives workers the right to form a union. Once
they do, the union bargains on behalf of the players with the teams to iron
out the all-important CBA, or collective bargaining agreement.
The union and the CBA provide a useful service to all involved. The
players have an advocate that acts on their behalf to avoid exploitation, and
the owners, through the commissioner’s office, can negotiate as a group to
make sure their interests are also looked after.
But collective bargaining takes the market out of the equation—and from
an economics perspective, that can cause problems. CBAs generally set
minimum salaries, pension arrangements, maximum contract lengths (at
least in the NBA), restrictions on free agency early in a career, team salary
caps (in the NBA, NFL, and NHL), and many other details.
And then there’s the money. The NBA players’ union and the NBA
owners have to agree on how to split the billions of dollars they bring in.
The NBA CBA expired after the 2010–2011 season, a year in which the
NBA claimed total revenues of $4 billion. That is more than $100 million
per team and nearly $10 million per player. Compare the implications of
splitting that money 50/50 between owners and players versus giving 55
percent to the players and 45 percent for the owners. The second scenario
provides the average owner with $6.7 million less per year and gives the
average player $556,000 more. Both sides clearly have a lot of incentive to
negotiate hard over the CBA.
In deciding how to negotiate, players and owners need to understand
their best alternative to a negotiated agreement, or BATNA. In other words,
they need to know their next best option. Remember how if the Angels
didn’t offer Mike Trout enough money, he could count on another MLB
team making a generous offer? This meant that his BATNA with the Angels
was pretty close to the $430 million over twelve years that he received from
them. If it were very different, he would not have accepted that offer, or the
Angels would not have made it.
The NBA players’ union BATNA when negotiating a CBA with the
owners is not very good. The owners have leases on big arenas, huge TV
contracts, a brand name that brings in customers, and many other assets that
turn players’ talents into cash. The players cannot replicate this outside a
deal with the NBA owners (though basketball players try to do so during a
CBA dispute in the movie High Flying Bird). Any other arrangement, such
as the players doing things other than basketball or going off to play in
other countries or local leagues, would cut their income to a small fraction
of NBA pay.
The owners’ BATNA when negotiating a CBA with the players’ union is
equally unattractive. The players are, by a huge margin, the best and most
marketable collection of basketball talent in the world. If the NBA tried to
put nonunion players on the court (in union parlance, if they hired “scabs”),
their TV contracts would be nullified, ticket revenue would fall off a cliff,
and merchandise sales would collapse. Many would be stuck supporting big
arenas that had suddenly become white elephants.
A favorite metaphor for the value created by two sides that bring unique
assets to an agreement is the “size of the pie.” When the two sides have a
lot to gain by working together, the pie is bigger. But if the people entitled
to slices of the pie argue over how big their slices should be, they often run
the risk of shrinking the whole thing. CBA negotiations fit this metaphor
perfectly (though readers may feel free to use another, equally apt
metaphor: killing the goose that lays the golden egg).
 As the NBA players’ union and the NBA owners negotiated the CBA,
they had to work out a lot of details. But let’s focus on a single point: how
much would owners spend on player salaries? The owners wanted to reduce
those salaries by $400 million per year.
That may seem like a big reduction, but there is no single right way to
divide the pie. If either side were to walk away, the pie would shrink from
$4 billion to a small fraction of that. Each side can thus credibly say to the
other, “You better give us what we’re asking for, because you are nothing
without us,” to which the other side can credibly reply, “The same goes for
you, so you give us what we want.” The inability to agree led to the owners
“locking out” the players—that is, canceling the season until a new CBA
was worked out.
By the time the sides agreed to a new CBA, in December 2011, the
season was shortened. The lost revenue—that is, the shrinkage of the pie for
the season—was in the hundreds of millions of dollars: $580 million, by my
estimate.28 That was money the teams and players would never get back.
The economic damage overall was even greater, as fans lost out on the
value of enjoying games, and people who work at games lost the chance to
make money.
This fighting over the pie within a relationship is certainly not limited to
basketball. Negotiating CBAs is difficult in all major American sports, and
all have seen work stoppages as the players and owners fought over how to
share the pie. The same dynamic goes on in other union negotiations as
well. Steelworkers, autoworkers, teachers, and others go on strike or get
locked out (or threats of strikes and lockouts are made) as new contracts get
negotiated. Occasionally this leads to the shutting down of factories and
employees scrambling to put food on their table. Usually, after a bit of
brinksmanship, a deal gets done before a shutdown occurs. But sometimes
both sides lose a lot of money until they come to an agreement and get the
business back on track.
Five hundred and eighty million dollars is a lot of money to waste, $360
million is a lot for one person in an injury-prone game to put at risk, and
$430 million is a lot of money to pay a man to play a game. But all these
amounts reflect the value athletes create in the modern, internet- and TV-
enabled world. In a world where the rich are getting vastly richer, the very
best athletes are extremely lucky to have been born when they were.
OceanofPDF.com
 5
Why Do Athletes Use Their Least Successful Moves
So Often?

The garage door was open one summer afternoon at my home on the
sprawling Stanford University campus. A light breeze kept the garage cool
as the competitive fires heated up toward the end of the fourth game in a
best-of-five Ping-Pong battle between my son, David, and me. Though I
was once the household king of the Ping-Pong table, David had surpassed
me and was a point away from handing me another demoralizing defeat.
After a brief rally, I hit a weak shot to David’s forehand that I immediately
knew he would try to smash for a winner. David’s forehand smash is more
reliable when he aims crosscourt, and I leaned that way, guessing he would
hit to his stronger side. Instead the ball sped down the line behind me.
Game and match to David.
Ping-Pong is just a game in our house. David and I can be overly intense
about it, but ultimately we play just for fun. Nonetheless, the point at the
end of that match illustrates how a game can become game theory. The
choices that David made about whether to drive the forehand crosscourt or
down the line, and I made about whether to lean left or right, were based on
our subconscious application of principles from the economic field of game
theory. In economics terms, we each wanted to maximize our utility—in
this case, maximize our chances of winning the point—while recognizing
that the other player was making calculated, strategic, and selfish choices in
his attempt to do the same.
An economist would say that deciding where to hit and how to lean
constituted a “mixed-strategy equilibrium.” With time and practice, David
and I have learned what probability to assign to each choice the other will
make while trying to vary our own choices in a roughly random matter.
David does not always go crosscourt, even though that’s his better shot,
because if he did, I would always be waiting there. Sometimes he goes
down the line, but since he’s not as good that way, he does it just enough to
keep me guessing. In the equilibrium we have worked out, David is equally
successful, on average, whichever way he hits, and I am equally successful,
on average, whichever way I lean.
It’s not hard to find mixed strategies in places that sports fans actually
care about. Game theory may sound esoteric (and sometimes—okay, most
of the time—it is), but once you have some understanding of it, you will
never again look at a soccer game, tennis match, or baseball game in the
same way. In each of these sports, and most others, success hinges on
optimally implementing a “mixed strategy” that relies crucially on two
things: knowing how often to choose each strategy, and randomly mixing
the strategies so that you remain unpredictable.

Kicking and Diving with a Purpose

One high-profile example of a mixed-strategy equilibrium occurs in the
penalty kicks and shootouts that often decide important soccer matches.
Though penalty shootouts have little in common with the on-field game
action, they are a traditional way of settling elimination games that remain
tied when the clock officially runs out. Italy has twice made it to a World
Cup final only to see the result decided by penalties, losing in 1994 to
Brazil and winning in 2006 against France. The UEFA Champions League,
club soccer’s biggest competition, had eleven finals decided by shootouts
between 1980 and 2020. In international tournaments and other soccer
leagues and tournaments, penalty shootouts are the standard way to break a
tie at the end of regulation time. Given that shootouts decide games with
such high stakes, a fan unfamiliar with soccer might assume them to be
intricately strategized matches that pit opponents’ key players against one
another in a tense battle of pure skill. But that’s a wildly romanticized
description. Shootouts at the highest level of play involve little more than
luck and an ability to control one’s nerves.
In a standard shootout, teams alternate taking penalty kicks. Each team
gets five kicks, though the game can end sooner if one team gains a decisive
advantage, and additional kicks are made as long as the game remains tied.
Kicks are taken twelve yards from the goal, with the goalie the only
defender. Shot takers have essentially three options: shoot left, shoot right,
or shoot straight. Similarly, goalies can either dive left, dive right, or remain
in the center. Because the shooter takes his shot from so close to the goal,
the goalie does not have time to read and react to the shot and must guess
which way to dive while the kicker is still approaching the ball. Game
theorists classify penalty kicks as “simultaneous move” games, because
both players make their decisions without knowing what choice their
opponent will make.
Just like my family Ping-Pong battle, the players in a shootout are forced
to use the information they have about themselves and their opponents to
make a very simple tactical decision. Shoot left, shoot right, or shoot
straight? Dive right, dive left, or don’t dive at all?
Let’s begin with a simplified scenario. Assume the kicker has only two
choices, kick left or kick right, and the goalie also has two: dive left or dive
right.1 Suppose the goalie is equally good at diving each way, and if he
dives in the direction of the kick, he has a 50 percent chance of making the
save. Suppose, too, that all shooters are right-footed and are a bit more
accurate shooting to their left than to their right.2 Let’s assume, then, that a
player will put his shot on target 96 percent of the time if he kicks left and
80 percent of the time if he kicks right. Your first thought might be that the
kicker should always kick left, because that is where he has the best chance
of making the shot.
Of course, that theory doesn’t work, because then the goalie will always
dive left, and the kicker’s success rate will be 48 percent: he will get the
shot in the right place 96 percent of the time, and the goalie will save it half
the time. Now the 80 percent chance on a shot to the right looks like a much
better option.
In reality, however, neither side should ever be a more attractive option
than the other. In equilibrium, the kicker must have an equally likely chance
of scoring whichever way he kicks, and the goalie must have an equally
likely chance of making the save whichever way he dives. The underlying
logic is quite intuitive: if the odds of success were higher in one direction,
the kicker would increase the probability of going in the direction where he
is more successful, and the goalie would do the same.
Starting from the 96 percent and 80 percent accuracy figures for shots to
the left and right, respectively, and the 50 percent save rate for a correct
guess by the goalie, the optimal strategies for this theoretical kicker and
goalie can be derived with a little math. The calculations are fairly simple,
but the concepts, which were first articulated by the mathematician John
von Neumann (who also made seminal contributions to the creation of
nuclear weapons) and the economist Oskar Morgenstern, and refined by
John Nash (who was famously portrayed by Russell Crowe in A Beautiful
Mind), required deep mathematical insight. I’ll spare you the manipulation
of the numbers, but trust me: the math predicts that a goalie playing under
the parameters I have described will dive to the kicker’s left about 64
percent of the time and to the kicker’s right about 36 percent of the time.
Because the kicker is more accurate to the left, the goalie knows that diving
that way is more useful—the goalie’s dive stops 48 percent of shots to the
left and only 40 percent to the right. But, of course, he must dive to the
kicker’s right enough to prevent him from aiming in that direction every
time. The kicker, on the other hand, should find it optimal to kick left 45
percent of the time and kick right 55 percent of the time. Though he is
better at shooting left, the kicker has to consider the goalie’s tendency to
neutralize that strength by diving more often in that direction. Because of
that decision by the goalie, the kicker is actually better off going right—his
weaker side—a bit more than half the time.
So how often does the kicker score? In this example, the probability of
converting the goal is 65 percent no matter which way the kicker aims. If
the goalie dives to the kicker’s left, he is out of luck for the accurate 80
percent of kicks aimed right, but he reduces the kick’s success rate to 48
percent when it goes left. If you add up those probabilities and weight them
by the rates at which the hypothetical kicker goes left and right, you get the
kicker’s 65 percent success rate. If you calculated the probability of success
for the goalie diving to the kicker’s right, you would get the same number.
This is the best both players can do, provided the other is strategic about
it. What would happen if the kicker were, for example, to say to himself,
“I’m good at kicking to my left—I should do that more often,” and began
kicking 60 percent of his shots left while the goalie continued to dive to the
kicker’s left 64 percent of the time? For a while, the change would make no
difference. Remember that the goalie has set his probability such that he is
equally likely to stop a shot no matter which way he dives. In fact, if the
kicker started kicking all his shots left, his probability of success would
remain 65 percent as long as the goalie did not adjust his strategy. But if the
goalie caught on that the kicker went left 60 percent of the time, he would
be able to decrease the kicker’s overall shot success percentage to about 60
percent by diving to the kicker’s left all the time.
 You might think the goalie and kicker can’t work all the math out and
both end up just kicking and diving to the kicker’s left a lot. To some
extent, you would be right, because, of course, the players do not use Excel
Solver, as I did, to figure out how often to go one way or the other.
But through trial and error, as well as advice from more experienced
teammates and coaches, top soccer players do a remarkably good job of
getting the directional breakdown almost exactly right. The economist
Ignacio Palacios-Huerta has become an expert on the economics of soccer,
and his work includes a paper analyzing 1,417 penalty kicks in professional
soccer games around the world.3 He found that soccer players choose their
directions almost as well as they would if they had Excel to help them. As
game theory would predict, kickers are successful at the same rate whether
they kick left or right, and goalies are also equally successful no matter
which way they dive.
To economists, this finding is not too surprising, because we believe that
people who have strong incentives to do something will find the best
possible way to do it. The financial and personal returns to scoring (or
stopping) goals and winning soccer games are very high in many settings.
An athlete who did not maximize his chances of getting the best outcome
on each penalty kick would put himself at a big disadvantage in a world
where the very best get tremendous pay and adulation, while those who just
miss stardom are essentially nobody.
When you next watch a penalty kick on TV or at a stadium, remember
that the players are implicitly doing a lot of math to maximize their chances
of winning. And if you get out on the soccer field, remember to mix it up
when you take a shot on goal.

The Mind Games of the Tennis Serve

From an economist’s perspective, using soccer to demonstrate mixed
strategies is almost unfair. Penalty kicks are exactly the type of environment
we would design in a laboratory to test our models of mixed strategies, so if
the theories don’t work there, they probably never will. Other games,
though, have a similar one-on-one nature but are more complicated. Do our
models work there, too?
They do if two economists play. Let’s turn to tennis, my favorite
competitive sport. Though my glory days as a four-year letterman on the
Nutley Maroon Raiders tennis team and as two-time champion of the Forest
Lake Camp tennis tournament are increasingly distant memories, I still get
out on the court with other middle-aged men a couple of times a week. One
of my regular singles opponents is a world-renowned economist who does
research in game theory. In those matches especially, I have to get my
strategy right, or else I am at a big disadvantage.
Like many players, I use a serve-and-volley strategy on many points.
That is, I hit my first serve and then rush in behind the ball. If I rush the net
quickly enough, my opponent will have to decide what kind of shot to hit
before seeing whether I am coming in or not. But the strategy must be
deployed in much the way that penalty kickers choose which direction to
aim, because if the server uses it too often, his opponent will consistently
execute one of the types of returns designed to beat a serve and volley:
either a lob over the server’s head or a hard shot down one of the sidelines.
One difference exists between this situation and the penalty kicks.
Goalies and penalty kickers have a long time between each penalty kick,
and they almost always face a new opponent the next time it happens. Even
in a shootout to break a tie, the kicker changes with each kick. So, in
practice, it’s really not a big deal if a pattern develops. As long as the
directional choices follow the optimal percentage, things will work out in
the long term. Even if a kicker decides to kick to his left twice, to his right
once, then left twice, right once, and so on, it will take a long time for
anyone to notice this pattern because he faces a different goalie each time.
But in my tennis matches, I serve at least four (and usually more) points
in a row during ten or more service games against the same opponent. In
these matches, I serve and volley on about one-third of the points. If I
adopted a no-no-yes pattern for rushing the net, I would obtain my ideal
ratio of serve and volleys, but my opponent would eventually see the
pattern and start lobbing on every third point. The effectiveness of my serve
and volley would be greatly reduced. So I have figured out a little trick that
I generally reserve for matches with my game theorist opponent: I run the
stopwatch on my wristwatch for a few minutes before each service game
and stop it randomly. The last three digits of the readout give me three
random numbers, and I rush the net on those points in the game. For
example, if I stop my watch at 2:13.47, I rush the net on the third, fourth,
and seventh points.4
 Am I just a compulsive economist, or do professional tennis players,
whose games people actually care about, worry about these issues too? As
you probably guessed, they do. A tennis match is one long sequence of
mixed-strategy games. Before each shot, the player has to decide whether to
go left or right, whether to hit a drop shot or not, whether to charge the net
behind his shot, and so forth.
Given enough data, economists could study all these choices and see
whether tennis players are maximizing their chances of winning points. But
it’s impossible to control for all the relevant factors on some choices tennis
players make. The economists Mark Walker and John Wooders limited their
exploration of mixed strategies in tennis to the server’s placement of his
serve on the left or right side of the service box and this decision’s effect on
the server’s eventually winning the point.5 Walker and Wooders studied
only a handful of matches from major championships in which the stakes
were high and the players were familiar with each other.6 Across these
matches, they found that the players were equally likely to win a point
serving to the right as they were serving to the left. They concluded that
tennis players randomize their serve direction to maximize their chances of
winning the point, ultimately settling into a mixed-strategy equilibrium.
But did players do this in the optimal way? Did they ensure that they did
not settle into a pattern their opponent might discern? This was the question
picked up by Shih-Hsun Hsu, Chen-Ying Huang, and Cheng-Tao Tang,
economists at National Taiwan University. After replicating Walker and
Wooders’s findings across a greater number of matches, this time including
data from major women’s and junior players’ tournaments as well, Hsu and
coauthors focused on the level of unpredictability in the players’ decisions.7
Economists call this “serial independence.”
Serial independence means that the value of one item in a sequence is
not affected by the value of another item in the sequence. In the case of
tennis, serial independence means that the direction of a serve on one point
should not affect the direction of a serve on the next point. People
attempting to simulate randomness generally switch too often from one
option to the other. A tennis player intent on remaining unpredictable might
persistently alternate the direction in which he or she serves, and so actually
become fairly predictable.8
 The economists found that top tennis players closely approximate true
randomness, confirming that tennis players’ serve directions are governed
by a mixed-strategy equilibrium. Note, however, that these studies of tennis
look at the relatively simple decision of serving left or right. A recent study
by economists who considered a wider set of options for servers and also
analyzed the “muscle memory” advantages of hitting the same shot multiple
times in a row show that tennis players don’t fully optimize their serve
strategies.9 To truly maximize her chances of winning a point, a server
would want to remain completely random. It seems that though she may not
have a fully optimal strategy when all factors are considered, she comes
close. Soccer and tennis thus demonstrate two elements of mixed-strategy
equilibrium that are extremely important to a player who wishes to
maximize her success: choosing the right probability of going in either
direction, and doing so in a way that has no discernible pattern. For athletes
with enough experience and incentive to get it right, these are skills that can
be developed with practice.
For the rest of us, who play sports recreationally, getting the
randomization and probabilities right is also useful if we want to win (and
who doesn’t?). But we are less likely to do it as effectively, because playing
the right strategies and randomizing effectively both take practice. Luckily,
your opponent doesn’t have time to get it perfect either.

Wow, This Pitching Can Sure Get Complicated

Like tennis, baseball is one mixed-strategy game after another. Will the
batter bunt? Will he fake a bunt and swing away, having drawn the third
baseman out of position? Will the runner steal? If he does, will the pitcher
throw a “pitchout” to catch him? Each of these decisions must be executed
often enough to be useful but infrequently enough to be unpredictable.
Beyond these many decisions, however, an even more frequent and
elemental mixed-strategy game plays out in every baseball game. On every
pitch, the pitcher must determine what pitch to throw, and the hitter must
decide whether to swing. These choices present mixed strategies to both
players. Like the penalty kick and tennis serve, every pitch in baseball
requires both the pitcher and the hitter to make a decision that will
maximize their outcome contingent on the other’s decision, but without
knowing what the other will do.
 Suppose a batter has such a slow reaction time that he must decide
before the pitch whether to swing or not. When I discussed the idea of a
mixed strategy with my son David, he explained that he has a regular video
game opponent who, when playing MLB: The Show, actually does decide
whether to swing before the pitch. Knowing that his fairly incompetent
friend has already chosen whether to swing, David purposely throws a lot
more balls than most real pitchers do. Given that his friend’s swing will
occur independently of the quality of his pitch, David need not place the
ball close to the strike zone if he expects a swing, and he should throw it
right down the middle if he expects a take.
Of course, if David threw only balls, his friend would never swing and
would walk continuously. But if David threw only strikes, his friend would
swing every time and get a good number of hits. So David’s goal is to throw
just enough strikes that his friend continues to feel as though he should
swing, but not enough that he allows many hits. As it turns out, this is a
fairly easy equilibrium to reach and maintain against an unskilled opponent,
and it yields a lot of swinging strikeouts. This situation almost perfectly
parallels the soccer penalty kick situation and is much simpler than baseball
pitching and swinging decisions faced by good players.
Before I get to mixed strategy in the major leagues, note that pitching
situations exist in which mixed strategies do not apply. If a pitcher is so
good at one pitch that he will always do best with that pitch, no matter what
the hitter is expecting, then the pitcher will use that pitch exclusively. In
that case, an economist would say that the best pitch is a “dominant
strategy” in the same way that using steroids was a dominant strategy for
Tour de France racers.
For example, when David played in the first few games of his final Little
League season, he and several of the other older kids could make every
pitch a fastball down the middle. Most of the opposing batters could not
catch up to it, and pitchers could be extremely effective with only a good
fastball. But as the kids got better over the course of the season, David and
the other kids had to learn to vary the speeds of their pitches and purposely
throw out of the strike zone to keep batters off-balance. At a more advanced
level, Mariano Rivera, the longtime closer for the Yankees, had such an
effective cutter that he threw it almost exclusively—in one season, over 93
percent of the time. Batters knew it was coming but were usually helpless to
do anything about it. Rivera thus did not have to be overly concerned with
strategy. The former Red Sox starter Tim Wakefield came even closer to
having a one-pitch dominant strategy. In 2008 he threw his knuckleball on
99.5 percent of his pitches.
Rivera and Wakefield were exceptions. The element of surprise is
generally important in Major League Baseball and the equivalent
international leagues. A professional hitter has about .27 seconds, by most
estimates, to read the pitch location and allow it to inform his swing
decision. That’s not enough time to make a fully informed decision, so what
the batter is expecting has a big influence on whether and how he swings.
Because the pitcher knows the batter is reading and reacting to the pitch but
has hardly any time to do so, the pitcher gains a large advantage if he fools
the batter.
Rather than the two-option video game example, a major-league pitcher
has several alternatives. New York Mets ace Jacob deGrom, for example,
throws a fastball, a slider, and a changeup. When you add the options of
various combinations of high, low, inside, outside, and in the strike zone,
the set of pitches he can throw becomes extremely complicated. Other
factors make pitcher and batter strategy still more difficult than the tennis or
penalty kick examples. For one thing, pitcher and batter skill is highly
idiosyncratic. Houston Astros ace Justin Verlander’s fastball is much better
than his breaking ball, and New York Yankees starter Corey Kluber’s
breaking ball is much better than his fastball. Verlander throws a lot more
fastballs than Kluber, but they both mix things up to some degree (though
one analysis showed Kluber should throw more breaking balls).10 On top of
that, some batters are much better at hitting fastballs than they are at hitting
breaking balls. The legendary Oakland A’s general manager Billy Beane,
who was once a top pro prospect, apparently never stuck in the big leagues
as a player because he couldn’t hit a slider. If that story is true, then for any
pitcher who had that pitch, throwing a slider was a dominant strategy when
facing Beane.
Another factor that affects pitch strategy is the count. Pitchers should
throw a fastball with a different probability—and batters should adjust their
expectations accordingly—on the first pitch of an at-bat than on a count of
three balls and one strike. A ball on the first pitch has a relatively low cost,
but a ball on a three-and-one count puts a runner on base. Fastballs tend to
be more accurate, so pitchers will throw them more often on three-and-one
counts.
 These and other factors make studying baseball pitch strategy more
difficult than studying penalty kicks and the simple “serve left or right”
studies of tennis. They also make it harder for baseball players to determine
how often they should make each choice. In soccer, a kicker should kick
right with almost exactly the same probability on pretty much every penalty
kick; adjusting a bit for opponents (which is easy to do in tennis because
you are playing one person for a hundred or more points), a tennis player
should hit serves left or right at a constant probability throughout a match.
But a pitcher should throw a fastball at a very different rate to Los Angeles
Dodgers All-Star Mookie Betts on a three-and-one count with the bases
loaded and one out late in a tie game (when a walk would be very costly)
than on the first pitch of an at-bat to a mediocre hitter leading off the fourth
inning with a three-run lead (when a single ball would have little
consequence). Adding in differences in wind, temperature (it’s harder to
control pitches when it is cold), and stadium size, a pitcher needs not just
Excel but a data science team to figure out how often to throw a fastball.
For more insight into these complexities, I talked to Matt Swartz, a
Ph.D. economist and consultant for the Washington Nationals who has
studied the game theory of pitching extensively.11 “For the most part,” he
told me, “batters and pitchers follow mixed strategies pretty well.” For
example, R. A. Dickey was the most recent great knuckleball pitcher.
Dickey also threw a few fastballs, and those fastballs were just as effective
as those thrown by Justin Verlander, who (unlike Dickey) is often
recognized for having one of the best fastballs in the game. Of course, if a
batter knew a fastball was coming, he would typically do much better
against Dickey. But Dickey used his fastball much less frequently. It was as
effective as Verlander’s because of the element of surprise, not its blazing
speed.
But while Swartz finds that hitters and pitchers get the big picture right,
there are details that they miss. Batters, he says, get “swing happy” when
there are two strikes, and “pitchers don’t take advantage of it enough.”
Pitchers, meanwhile, don’t sufficiently adjust their strategies to their own
strengths. “The league as a whole learns how often to throw each pitch in a
given count. However, for example, if you are a great slider pitcher, you
should throw a lot more sliders than other pitchers. But they don’t, actually.
They pitch like other pitchers.” Overall, pitchers are very good at working
out the mixed-strategy equilibrium. But as teams’ analytics departments get
better and better at using the troves of data now available, pitchers will
likely get even better.
One implication of the intricacy of optimal pitch selection is that
experience and intelligence are critical to pitching success. As Swartz puts
it, “I would bet that Jamie Moyer had great mixed strategies.” Moyer, who
was known for being crafty rather than particularly athletic, pitched in the
major leagues until he was forty-seven years old.
Relative to pitchers, soccer players can more easily figure out optimal
strategies when it comes to penalty kicks. But in more complex game-
theoretic situations such as pitch selection, even pros who do the job for a
living with millions of dollars on the line play the game less than optimally.
Put another way, a supercomputer could not beat a soccer player in penalty
kicks, but it could outsmart many pitchers or batters (if it first learned to
pitch or hit).
This “supercomputer” effect in baseball has important implications for
which player should decide which pitch should be thrown. Experience
matters because, to call the ideal mix of pitches, a player needs to know a
lot about the batter, the pitcher, the conditions that day, and the situation
surrounding each at-bat. That’s a lot of information to process. Who should
do it? In the MLB, the catcher generally calls the pitches. The pitcher has a
lot of influence as well, through pregame discussions with the catcher and
by exercising veto power (pitchers often shake their heads in disapproval
until the catcher calls the pitch the pitcher wants to throw). There may come
a day when someone off the field with access to lots of data will call the
pitches. But economics can’t explain everything: it’s possible that catchers
and pitchers can make nuanced observations of their opponent’s behavior at
a given moment, and that additional information is more valuable than
computing power.
Mike Matheny, who has been the manager of a few major-league teams
after a career as a major-league catcher, thinks this on-field information
trumps experience. He advises managers at all levels to let catchers call the
pitches. “There is no way,” he wrote on his blog, “that I could ever have a
better sense of the game, or see the subtle things that only a catcher could
see, from the dugout.”12 But Matheny also understands the value of
experience in calling pitches, stressing the importance of teaching catchers
how to understand each pitcher and “get the most out of them.”
 Other coaches and managers believe that experience trumps the value of
the in-game information available to the catcher, and they therefore call the
pitches from the dugout. This is especially true in high school and college
baseball, where many catchers simply relay the coach’s selections to the
pitcher. Some anecdotal evidence suggests that coaches are becoming more
active in calling pitches even at the higher levels of the sport, which would
not be surprising, given baseball’s increasing use of quantitative analysis. In
any case, economics would predict that more experienced pitch callers are
better at gathering all the relevant factors that determine the probability
with which each pitch should be thrown and then randomizing properly
among the options.
In the end, when it comes to pitching (unlike soccer penalty kicks), the
optimal mixed strategy and randomization are sufficiently complex as to
resist calculation in any useful time frame. In the case of pitch selection,
both sides—the pitcher-catcher team and the hitter—are making incredibly
complex decisions that combine conscious knowledge with intuition. The
irony, and the beauty of the game, is that even if the pitcher and catcher
make exactly the right call, and the pitcher executes the pitch perfectly, the
batter sometimes gets a hit anyway.
Soccer, tennis, and baseball—not to mention Ping-Pong in the garage—
all show us how game theory is important in sports, and how optimizing
your strategy becomes increasingly difficult as the games grow more
complicated. But those sports are far from the only places where knowledge
of the mixed-strategy equilibrium is critical to excelling. Volleyball players
fake spikes, football teams fake punts (though not as often as would be
entertaining), and NASCAR drivers try to surprise their opponents about
when and how they pass each other.
In each of these cases, success hinges on two things: knowing how often
to employ each strategy, and randomizing among the strategies. To win, you
have to be both smart and unpredictable.

OceanofPDF.com
 6
How Does Discrimination Lead to a Proliferation of
French Canadian Goalies?

“Integrating the NFL was the low point of my life,” Woody Strode told
Sports Illustrated, years after becoming the league’s first Black player. “If I
have to integrate heaven, I don’t want to go.”1
Although the story of Jackie Robinson is the most familiar tale of
breaking barriers in sports, baseball was hardly the only American
professional sports league to be tainted by segregation. The NFL fired all of
its Black players in 1934; it was another twelve years before Strode and his
teammate Kenny Washington returned racial diversity to the league.
For Strode and Washington, it was a matter of right place, right time.
When the Rams moved from Cleveland to Los Angeles in 1946, the team
wanted to play in Memorial Coliseum because it seated 100,000 people.
Stadium administrators, supported by a crusading local sportswriter,
demanded that the franchise sign a Black player. The Rams complied,
turning to former UCLA teammates Washington and Strode.
The duo suffered the same sorts of taunting and cheap shots that
Robinson did in his famous first season with the Dodgers a year later; like
Robinson, they were also asked to stay in “racially appropriate” hotels
when the team traveled. Washington, who saw more game action than
Strode, was once held down by opponents at the bottom of a pile while they
rubbed chalk in his eyes. “He could smile when his lip was bleeding,” a
college teammate said of him; in the NFL of 1946, he needed to. Within a
couple of years, both he and Strode were out of the league, but other teams
slowly signed Black players, and the NFL slowly accepted desegregation.
There was not a smooth, fairytale-like ascent from two Black NFL
players in 1946 to today, when Black players make up nearly 70 percent of
the league. It is nice to imagine that Strode and Washington and Robinson
—and Chuck Cooper, Nat Clifton, and Earl Lloyd in the NBA in 1950; Lee
Elder at the Masters golf tournament in 1975; Althea Gibson at Wimbledon
in 1950; and Willie O’Ree in the NHL in 1958—made an immediate and
lasting impact on their sports. To some degree, of course, they did; since
World War II, no league has been resegregated after its color barrier was
broken. But do athletes of color still face discrimination in less obvious
ways? Economists have examined this question for decades, and they have
found that the path to racial equality in sports has been bumpy and remains
incomplete.
Overall, the story of discrimination in sports is uplifting. Social changes
and market forces have helped many female and minority athletes. In
economic terms, team owners learned that discrimination was a competitive
disadvantage. The historical discrimination that came from fans, team
owners, and the players themselves—which affected the racial makeup of
teams, athletes’ pay, and the positions and sports that were open to athletes
—has been substantially reduced, and the explosive growth in sports
revenue has been shared with an increasingly diverse athlete pool. Yet we
still find discrimination in sports around the world. Markets often help
correct injustice, but they can take a long time to do so.

Donald Sterling: Discrimination Poster Child

Economists like to classify discrimination as one of two kinds: either
taste based or statistical. The poster child for taste-based discrimination is
the former Los Angeles Clippers owner Donald Sterling, who was recorded
saying to his mistress, “Why are you taking pictures with minorities? Why?
It’s like talking to an enemy. . . . It bothers me a lot that you’re associating
with black people. . . . You don’t have to have yourself walking with black
people.”2 Sterling’s words implied a “distaste” for people of color that
almost certainly made him less likely to employ them relative to a
comparable white person.
Statistical discrimination, on the other hand, is usually practiced by
employers and others who want to sort a large pool of people based on
empirically supported stereotypes. For instance, a warehouse foreman
might interview more men than women for an open position moving
refrigerators because men are stronger, on average, than women. Of course,
any one woman might be stronger than any one man, but on a large scale,
the foreman’s discriminatory behavior has a statistical foundation. Just
because it makes mathematical sense, however, doesn’t mean statistical
discrimination is legally or morally acceptable. For example, a woman
qualified to move refrigerators would have a compelling legal and moral
case against a foreman who passed her over for a similarly qualified man.3
Taste-based discrimination, not statistical discrimination, dominates the
injustices seen in sports. In the decades after Jackie Robinson’s debut, study
after study showed that across a number of major sports, white players
received preferential treatment compared with Black players. It is not
always clear exactly who is discriminating, however, as taste-based
discrimination can generally come from at least three sources (and, in the
case of sports, a fourth).
First, we have Donald Sterling–style employer discrimination, in which
the people in charge of hiring prefer one demographic group over another.
Employers can also practice statistical discrimination, as in the refrigerator-
moving example. But the distinction between the two is that a statistically
discriminating foreman who brought in more men to interview would
happily hire a woman if she proved herself to be the strongest candidate.
For a taste-based discriminating foreman, a woman would need to be
substantially more productive than a man (or willing to work at a much
lower wage) to be considered for the job, because she would need to
compensate for the foreman’s aversion to working with women.
Second, taste-based discrimination can take the form of employee
discrimination, in which workers from an in-group don’t want to work with
workers from an out-group. For example, consider a perfectly
nondiscriminating manager in charge of hiring workers for a relatively low-
level position at the Lahore Lions cricket team’s office in Pakistan. Suppose
the more specialized team management positions are already staffed by
Sunni Muslims who have a “distaste” for Shia Muslims. Faced with two
equally qualified candidates, one Sunni and one Shia, the completely
unbiased manager has a strong incentive to hire the Sunni candidate. That
manager is aware that if the Shia candidate gets the job, discriminating
Sunnis, who make up the entire active workforce, will make the new
employee’s life unpleasant and may work less efficiently. So the manager
might choose the Sunni candidate to keep other employees happy even
though doing so is, in the bigger picture, unfair and wrong.
Third, taste-based discrimination can take the form of customer
discrimination. In markets where a substantial portion of the customer base
discriminates, a business focused solely on profits might argue that it is
forced to discriminate. Consider the owner of professional hockey’s
Montreal Canadiens. A majority of the team’s fan base speaks French,
making it the only fan base in the league that isn’t primarily English
speaking. Meanwhile the league’s labor force—its players—includes plenty
of French Canadians. Regardless of whether the owner cares about players’
native tongues, customers will probably prefer that the players speak
French, so the owner might be tempted, due to higher profits, to
discriminate in favor of French speakers and against English, Russian, and
Swedish speakers.
The final form of discrimination, and the one largely specific to sports, is
referee discrimination. If all of the officials in a sport are white and favor
others of their race, fielding a white team could have advantages in terms of
getting favorable calls. The economists Joseph Price and Justin Wolfers
have shown that NBA referees are more lenient to players of their own race,
such that the composition of a referee squad likely affects the outcome of
many NBA games. And historically, NBA referees and officials in other
sports have been disproportionately white, which could, on balance, have
led to a disadvantage for minority players.4

Who Is Racist: Fans or Front Offices?

The Minnesota Timberwolves’ 2012–2013 season was not especially
memorable, though the team’s thirty-one wins (against fifty-one losses)
were, depressingly enough, its most in six years. Those Timberwolves,
however, will leave a lasting legacy not for their on-court performance but
for the composition of their roster. Not since Larry Bird’s Celtics teams of
the 1980s had an NBA franchise fielded as many white players as that
year’s Wolves, who rostered white guys ranging from superstar forward
Kevin Love to Montenegrin big man Nikola Pekovic to redheaded former
volleyball standout Chase Budinger.5 In a league that was 78 percent Black,
the Timberwolves were two-thirds white. If you had distributed all of that
season’s NBA players at random among the teams, the odds of having one
team end up with ten white and five Black players were one in ten
thousand.
Confronted by reporters, the Minnesota front office denied that the
team’s racial makeup was at all intentional. But few thought it a
coincidence that the Timberwolves play in the NBA’s whitest market, and
one with an ugly history of taste-based discrimination by its sports team
owners. Calvin Griffith, the owner of baseball’s Washington Senators,
moved the team to Minneapolis in 1961 and renamed it the Minnesota
Twins. In 1978, Griffith told a crowd at a local Lions Club event that he had
chosen to relocate there “when I found out you only had 15,000 blacks
here. . . . We came here because you’ve got good, hardworking, white
people here.”6 Griffith, a holdover from the pre–Jackie Robinson era when,
as the journalist and podcaster Josh Levin put it, “owners hated Black
people more than they liked money,” moved his team from a city with a
majority Black population to a smaller and potentially less lucrative market
to keep his fan base white.7
The Timberwolves, denying any preference for players of a certain color,
claimed that the team’s demographic makeup was a consequence of the
organization’s global scouting efforts, which had hauled in five
international players, all white, to the opening-day roster. And, admittedly,
nobody knows for sure whether the Wolves were going out of their way to
sign white players. The team did become slightly less white by the 2013–
2014 season’s opening night. But it was by no means unreasonable for civil
rights activists to call attention to the roster’s skewed racial makeup; the
Wolves’ whiteness was unprecedented by recent NBA standards.
Implied in these allegations was a charge that the Timberwolves were
suffering from employer discrimination, customer discrimination, or both.
If the whitewashing was deliberate, then either Minnesota fans were
demanding a whiter team and ownership was responding, or ownership
simply preferred white players. This phenomenon appears to have haunted
the league on a larger scale for decades.
The NBA offers a microcosm of larger patterns of discrimination in the
United States. After the NBA was integrated in 1950, a legion of Black
stars quickly inundated the league. By 1965, six of the ten All-Star Game
starters were Black; in 1975, the first year in which fans voted for the All-
Star team, four of the five leading vote getters were Black. Between 1954
and 1970, the Black share of the league’s labor force shot from 4.6 percent
to 54.3 percent.8
But all was not well for the increasingly dominant Black group of
players. A raft of studies, including ones by the economists Gerald Scully
and Lawrence Kahn, showed a long-standing pattern of wage
discrimination: Black NBA players significantly outperformed white ones
at equivalent pay levels, and Black players were more likely to be cut from
teams than equally good white players.9 Put differently, a Black basketball
player in 1985 had to score more points or grab more rebounds than his
white teammate to make the same salary or keep his job.
Were all thirty NBA ownership groups colluding to underpay Black
players in a blatantly racist conspiracy? Not necessarily. It’s certainly
possible that some NBA owners in 1985 held a distaste for Black people—
Sterling bought the Clippers in 1981—but they were also likely responding
to the racist preferences of their fan bases. Several economic studies, while
varying slightly in their specifics, found that fans in the 1980s were more
likely to attend basketball games if more of the home team’s players were
white.10 Researchers also found that white players were not evenly
distributed across teams but clustered in markets with more white fans. So
even if the fans didn’t actually discriminate, the league’s general managers
assembled their teams as if fans did, valuing the white players more in cities
with larger white fan bases. White players were concentrated in white cities:
Bill Walton played in Portland; John Stockton played in Utah; Larry Bird,
Kevin McHale, John Havlicek, and Dave Cowens all played in notoriously
racist Boston; and so on.
But attitudes about race have changed. Although racism is both casually
and empirically observable in the present-day United States, it was more
overt a few decades ago. Americans’ approval of interracial marriages, for
example, has trended in roughly linear fashion from 4 percent in 1958 to
nearly 90 percent in recent years.11 As the social structures and ways of
thinking left over from segregation fade away, logic suggests that
discrimination in the NBA will recede as well. And for the most part, it has.
A number of economists reexamined the data for seasons in the 1990s, and
most found little evidence of salary discrimination. By the mid-1990s,
Black players were earning roughly as much as comparably skilled white
players. The same was true for the findings on exit discrimination.12
Present-day studies have concluded that wage discrimination no longer
exists in the NBA. I don’t know of any formal tests of whether fans still
have discriminatory tastes, but a crude survey of recent opening-day rosters
showed no correlation between the whiteness of a team’s fan base and its
players. In various years, the Warriors, Bucks, Grizzlies, Lakers, and Spurs
have been among the teams with the highest proportions of white players—
but the composition of their home cities has been diverse, ranging from
predominantly white (Milwaukee) to substantially Latino (Los Angeles and
San Antonio) or Asian (San Francisco–Oakland) to the most heavily Black
city in the league (Memphis).
This trend reflects one of the pillars of the Nobel Prize–winning
economist Gary Becker’s theory of discrimination. Becker hypothesized
that markets correct for discrimination: firms that are less willing to hire
minorities will ultimately be driven out of business by firms that hire based
only on talent. As teams realized that ignoring or underpaying Black
players would harm them in the win-loss column more than it would help
attract prejudiced fans, teams ceased to discriminate. In other words,
winning mattered more to fans—and thus to teams’ bottom lines—than the
racial makeup of their rosters. Even Sterling’s Clippers had only one white
player under contract in 2013–2014. The fading of discrimination reflected
the change in underlying social attitudes. If fans had remained as prejudiced
as they were in 1970, a wise owner would have given up a few wins to have
more white players. But as fans became more open-minded, it became
increasingly costly for owners who were prejudiced to indulge their own
discrimination. Overall, competition helped raise the pay and job security of
Black players.
The huge increase in athletes’ salaries has also raised the cost of
discriminating. Other settings (for example, police forces of one race
policing cities dominated by another race) often have no competitive
marketplace to drive out discrimination. So, while good evidence suggests
that competition and economics erode discrimination, Becker’s theory can
be misused to justify government inaction in the face of discrimination. But
at least in some leagues, money and winning come first, and that’s good for
talented athletes from historically underrepresented groups.
As for the Timberwolves, probably due to some combination of
competitive forces and negative publicity associated with their unusually
white roster, the team quickly began to look more like the rest of the league.
The team had a 62 percent Black roster by 2015–2016 and had even
surpassed the league average with an 80 percent Black roster in 2020–2021.
The Timberwolves’ racial makeup may well have held the team back,
but it was not the only factor, as the team has continued to stink after adding
Black players to the roster. While we don’t have strong evidence that racial
composition hurt the Timberwolves, the economist Stefan Symanski
showed, using a broad data set from English professional soccer, that
“teams with a below-average proportion of black players have tended to
achieve inferior playing performance.”13 So, though the Timberwolves’
example suggests that nonprejudicial behavior of owners is not sufficient to
field a winner in the extremely competitive world of modern professional
sports, the data in soccer and elsewhere suggest that nonprejudicial team
formation is necessary to remain competitive.

Race-Free Discrimination in Hockey

Let’s move to a sport where you might not expect discrimination to be
relevant: professional hockey, where a mere 2 or 3 percent of players are
nonwhite. Black players in the National Hockey League are sometimes
subject to shamefully inappropriate taunting and slurs.14 But largely
because of the small sample size, we have little research or empirical
evidence to suggest that they face economic discrimination.
The NHL’s near racial homogeneity doesn’t mean that it has avoided a
somewhat storied history of questionable hiring practices. The NHL may be
America’s whitest major sports league, but it is certainly international, with
substantial numbers of English Canadian, French Canadian, American,
Czech, Finnish, Swedish, Russian, and Slovakian players. Most research on
discrimination in the NHL, however, has been confined to the pay gap
between the league’s two biggest demographic groups: French Canadians
and English Canadians.
Relations between minority French Canadians and the rest of Canada
have often been frosty. The low point occurred in 1969 and 1970, with the
bombing of the Montreal Stock Exchange and the murder of a government
official by separatist French Canadian terrorists. Although those tragedies
were anomalous in their magnitude, language policies in Quebec, Canada’s
lone French-speaking province, continue to generate controversy. Many
residents of Quebec still dream of independence.15
The earliest credible study of discrimination in the NHL, published in
1987 by three Canadian economists, found that French Canadian players as
a group tended to outperform English Canadian players as a group.16 Like
the similar research in the NBA, this study operated on the assumption that
if one demographic group was routinely outperforming another, a player
from the high-performing minority group (in this case French Canadians)
must have had to meet a higher standard than the lower-performing
majority group (in this case English Canadians) to gain a spot on the team.
The study also computed performance differentials by position. It
showed that defensemen apparently faced the most hiring discrimination,
while forwards faced less, and goalies seemed to face none at all. The
authors proposed a clever and sensible explanation for this gap: positions
for which performance is more easily measured are less subject to
discrimination, while positions that require more subjective evaluation
suffer from greater discrimination. Forwards’ play can be measured by the
number of goals and assists they produce, while defensemen’s impact on
the game involves historically less measurable abilities like checking,
passing, and situational awareness. It’s easier for a team to select against
French Canadian defensemen than French Canadian forwards because it’s
harder (or at least it was, before the sports analytics trend hit hockey in
recent years) to make a concrete case that one defenseman is better than
another.
Goalies have an obvious performance metric in save percentage. They
also are more isolated from the rest of the team than the other players,
which likely reduces employee (teammate) discrimination on the basis of
language. Consequently, French Canadian goalies seem to avoid
discrimination altogether.
The study made a fairly convincing argument that the market is better at
driving out racism when performance can be measured more easily. Scouts
evaluating forwards can put aside any preconceived notions they have,
knowing that it will cost the team too much to pass over a French-speaking
player who scores more. But the same team’s coach can tell himself he is
choosing the defenseman who speaks his language because he’s a tougher
or more savvy player, and there is no easy way for anyone to show the
coach was wrong. To borrow from psychology, “confirmation bias”—
talking yourself into believing what you want to be true—is stronger when
the data are more subjective.
To analogize to the business world, an economist would expect to find
that discrimination gets driven out by the market in sales or financial-
analyst jobs, where performance is relatively easily measured, but maybe
not in some middle-manager positions, where performance can be measured
very subjectively.17 Along the same lines, referees subjectively evaluate
players all the time, and as noted earlier, a study showed that NBA referees
favor players of their own race. Those findings received front-page
coverage in the New York Times, putting public relations pressure on the
NBA.18 Since the story was published, the NBA has invested substantial
resources in measuring and standardizing referees’ calls. By scrutinizing
refereeing and adding reviews to make it more objectively measurable, the
NBA has, according to more recent research, eradicated the bias in
refereeing.19

Where Are All the Black Pitchers?

Quick, name a recent African American pitcher besides Cy Young
Award winners C. C. Sabathia and David Price.20 Marcus Stroman? Chris
Archer? Chances are you couldn’t come up with many names; hardly any
players in baseball fit the description. Although African Americans have
typically accounted for at least 25 percent of major-league outfielders, they
have never made up more than 10 percent of pitchers or catchers. In recent
years, a mere handful of African American pitchers, and not a single
African American catcher, have appeared on major-league rosters.21
What causes this positional segregation? No one has a definitive answer.
The first of two primary theories attributes the lack of African American
pitchers and catchers to systemic racism among coaches, scouts, and
general managers, who may believe Black people lack the leadership and
thinking skills necessary to play pitcher and catcher, and therefore don’t
train them for these positions; or at the higher levels, managers shift Black
players to positions they see as requiring less intelligence. Widespread
evidence of racist stereotyping against the intelligence of Black people is
available both empirically and anecdotally. Polling done by the University
of Chicago’s General Social Survey routinely finds that respondents rate
white people’s intelligence higher than Black people’s.22 In the baseball
world, Al Campanis, the white general manager of the Los Angeles
Dodgers, made headlines in the 1980s (and was forced to resign) after he
told the news anchor Ted Koppel that he thought Black people “may not
have some of the necessities to be, let’s say, a field manager, or, perhaps, a
general manager.”23 Given that Campanis had been a general manager for
nineteen years, his comment led many to wonder how pervasive such views
were in baseball generally. Campanis was not alone. Cincinnati Reds owner
Marge Schott was repeatedly suspended for using racist slurs and allegedly
saying things too vile to paraphrase here.24
Other peripheral facts strengthen the case that prejudice plays
a substantial role in positional segregation. For one thing, Japanese players
in Major League Baseball are disproportionately represented at pitcher.
Given the common American stereotype of Asians as especially smart, their
increased presence at pitcher supports the stereotype argument, since
executives seem to think pitcher is a higher-intelligence position.
This apparent prejudice among scouts, coaches of young players, and
voters for the Cy Young Award exemplifies statistical discrimination. These
parties may have no conscious animus or ill will toward African Americans.
But if a coach believes that African Americans are less intelligent, he will
direct young African Americans away from pitching or catching. The racist
coach may think he is acting in the prospect’s best interests by matching
him to a position appropriate for his mental capacity. Here the distinction
between taste-based and statistical discrimination is meaningless because
the player is given no chance to prove he can be an effective pitcher.
As beliefs evolve over time, positional discrimination should be
corrected by the market. For instance, the recent advances of African
American quarterbacks in the NFL may suggest that statistical
discrimination at that position is waning. As Marc Morial, president and
CEO of the National Urban League, recently wrote, “For decades, the
prevailing view seemed to be that while African Americans made good
runners, blockers and receivers, they did not possess the ability or intellect
to be quarterback.”25 But things can change (though it can take a long
time): the New Yorker declared 2014 “the Year of the Black Quarterback.”26
In the five seasons after that article was published, three NFL Most
Valuable Player Awards were bestowed on Black quarterbacks. Before that,
Black players had made inroads at middle linebacker and center, both of
which are leadership and strategic positions that have historically been
dominated by white players.27
In addition to misguided statistical discrimination, another economics-
based explanation for positional segregation exists. Proponents of this
argument believe that pitcher and catcher are more specialized positions
than infielder or outfielder. Pitchers require more focused training and
instruction, and catchers need not only training but also fairly expensive
equipment. African Americans, who have higher poverty rates than any
other ethnic group except American Indians, are less likely to have access
to the instructional tools and coaches that are key to developing pitching
and catching (and quarterbacking) skills. This theory is generally supported
by participation rates of Black people in the sports landscape as a whole.
They are heavily represented in basketball and, to a lesser extent, soccer—
sports that require little equipment—while Black players are less prevalent
in sports that require a good deal of equipment or facility access, like golf,
hockey, and swimming. Structural racism still drives athletic outcomes to a
substantial degree.
The jury is still out on which of these two explanations better explains
positional segregation. Clear correlation exists between the costs of a sport
and the proportion of players from underrepresented groups. On the other
hand, quarterbacks don’t need more specialized equipment than wide
receivers. It seems likely that stereotypes and resource differences have
both contributed to segregating certain sports and positions.

Gender and Prize Money

It’s no secret that men dominate the mainstream sports scene. For
example, a quick comparison of the NBA and WNBA attendance records
demonstrates which league currently generates greater fan interest. Does the
fact that the minimum NBA salary is more than seven times as high as the
maximum WNBA salary constitute discrimination? Discrimination could be
part of the difference, but the difference in itself is not evidence of
discrimination.
Even the Women’s Sports Foundation abstains from advocating equal
salaries across gender lines for sports like basketball. Women’s basketball
leagues just do not draw as much interest as men’s basketball leagues and
therefore cannot raise the revenue necessary to compensate their players as
well as the men. In economics terms, the WNBA’s product (the games)
attracts less demand (fewer fans), keeping the league’s profits well below
those of the prosperous NBA. One can certainly argue about the cause of
the disparity: some fans would say men’s basketball is simply more
entertaining, while others would argue that women’s basketball is in the
process of establishing a fan base and brand just as the NBA did decades
earlier. This second argument would reflect some version of taste-based
customer discrimination. Regardless, most women’s professional sports
leagues are governed by the same facts: they have fewer fans than the men’s
equivalent, and their players receive much lower salaries. Even if
progressively minded WNBA owners wanted to pay their players NBA-
level salaries, they would quickly go bankrupt if they tried.
By contrast, one sport where women have a good case for discrimination
is one where they are paid very well: tennis. Indeed, in terms of pay, tennis
is the dominant women’s sport. It draws the highest ratings and biggest
crowds, and its players make the most in both tournament winnings and
endorsement money. According to Forbes magazine, the nine highest-paid
female athletes in the world in 2020 were all tennis players (soccer star
Alex Morgan ranked tenth).28 Whereas all-time women’s basketball greats
like A’ja Wilson and Breanna Stewart may not be familiar to the average
observer, most casual sports fans can name at least a handful of top active
women’s tennis players at any given time.
The biggest tennis tournaments are run by the same governing body for
men and women. The four Grand Slam events—the U.S., Australian, and
French Opens and Wimbledon—are conducted simultaneously for both
genders, with equal-sized 128-player fields. In tennis tournaments hosted at
the same venue by the same organizer, at which tickets earn admission to
both draws, it’s nearly impossible to parse out interest in the men’s event
from the women’s. The women might, at least theoretically, generate half
the revenue, so it would certainly seem like outright discrimination to
award them less prize money.
It took until 1973 before any Grand Slam event seriously considered
equalizing its prize money. In that year, women’s players at the U.S. Open
threatened to strike if they were not compensated equally to men. The
tournament’s organizers acquiesced to the demand, and the U.S. Open has
awarded equal purses to men and women ever since. The Australian and
French Opens soon followed suit. It took until 2007, however, for the last
holdout, Wimbledon, to finally raise the women’s purse to equal the men’s.
Until that time, the All England Lawn Tennis and Croquet Club continued
to distribute more money to the men than to the women in the same
tournament. Was Wimbledon discriminating?
For Wimbledon’s behavior to have been justified, the event’s organizers
would have needed evidence that the men’s tournament drew a greater share
of the revenue than the women’s tournament. But such evidence is hard to
come by. Television ratings were far more difficult to access in the 1970s
and 1980s than they are today, but one economist quotes a CBS executive
in the mid-1970s as saying the women’s tournament actually drew larger
viewing audiences than the men’s. For that study, the economist also
obtained Nielsen ratings for the 1988 Wimbledon finals, which showed that
the women drew a substantially larger share than the men. Interpreting
ratings is complicated, though, given that men play best-three-out-of-five-
sets matches, while women play best two out of three, so men’s matches
typically have more commercial breaks, which generate revenue.29
If these results held even somewhat true for all the years between 1973
and 2007, then women’s tennis players would probably have done well to
unionize and sue Wimbledon. If the body that oversees major tennis
championships were more subject to competitive pressures, the women
would probably have done better sooner.
Now that Grand Slam purses are equal by gender, is discrimination gone
from tennis? Maybe in terms of prize money at Grand Slam events. But
even though tennis players are the highest-paid female athletes, the top
women make noticeably less than the best-paid men. Tennis magazine
reported the following after the 2018 U.S. Open: “The women’s final
attracted 3,101,000 viewers, while the men’s final between Novak Djokovic
and Juan Martin del Potro drew in 2,065,000 viewers.”30 In that same year,
four male tennis players and no women made Forbes’s list of highest-paid
athletes.
Market forces have been a valuable weapon for reducing discriminatory
treatment of women and minorities in sports, and in labor markets more
broadly. But there are limits. If customers are prejudiced, historically
underrepresented groups will remain at a disadvantage. Eliminating that
sort of discrimination requires social, not economic, change. At other times,
no competitive pressure exists (such as when all MLB or NFL owners
agreed to have white-only leagues; or when Wimbledon, being such a
dominant tennis institution, could do whatever it wanted), and thus no
market incentive for things to change. But the other limit—and maybe the
greatest enemy of fairness—is that when animus is deep and widespread,
markets can be painfully slow to help those on the wrong side of
discrimination. Minorities and women have come a long way in the U.S.
labor market, thanks in part to market forces and competition. But
economists have shown quite clearly that civil rights laws were essential in
creating and sustaining momentum for women and minorities in the U.S.
labor market when market forces were moving much too slowly.31
Economics is powerful, but intolerance and stereotypes are strong
competitors.

OceanofPDF.com
 7
How Do Ticket Scalpers Make the World a Better
Place?

In 2001, Amy Stephens was a middle-school teacher in the Atlanta

suburbs. One weekend she took a trip to New York with her husband. The
couple planned ahead and bought tickets to The Producers, a very hot
Broadway musical at the time. But a family complication arose, and Amy
opted to sell the tickets rather than go to the show. That simple decision
changed her life forever. “We bought the two tickets for a hundred bucks,
sold them on eBay for five hundred, and there it came. I was like, ‘Wow,
that was easy.’”1 Amy didn’t know it yet, but she was on her path to
becoming a full-time professional ticket scalper.
When they returned to Atlanta, Amy and her husband started “picking up
tickets here and there, then throwing them on eBay.” They still had no plans
to make this more than a sideline, but things changed a year after the New
York trip, when Amy had her first child. She had been planning to take a
long maternity leave anyway, and she and her husband realized she could
make money at the same time by expanding the ticket resale venture. She
has been running Amy’s Tickets ever since. “It happened by accident,” she
says, “but it became a huge source of income.”
From the start, Amy’s business was built on a simple system. She bought
tickets from college and pro teams when they became available, and later
offered them through eBay. “We buy season tickets to good college teams,
and we have contracts with arenas—I get first right of refusal for certain
seats for every event there. These places work with brokers to shift
inventory.”
The teams whose tickets Amy sold didn’t necessarily like her; they
would have preferred to get the extra money she was making as profit. But
they tolerated her because she could reach fans that they couldn’t, and she
could vary her prices in ways the teams could not. The priority for teams—
especially college teams—was simply to sell bunches of season tickets.
They were happy to leave to Amy the work of breaking them up and
finding buyers for each game. She was equally happy to be the profiting
middlewoman.
Things couldn’t have been running more smoothly or profitably. “It used
to be that, anything I tried to sell, I made money hand over fist,” she recalls.
A few years after she got her eBay-based business up and running, Amy
evolved it further with the help of a new entrant into the ticket marketplace:
StubHub. Initially, the new site seemed to make her system work even
better; it took care of customer service issues, and it attracted fans in
hordes.
Yet what StubHub giveth, StubHub taketh away. After several years of
making her business bigger and better, StubHub shifted its focus and
became far less kind to Amy.
When Amy started her business in 2001, StubHub was still a fledgling
start-up. But by 2014, StubHub was the first place most fans would go to
buy and sell tickets. As it developed its technology to make directly
transferring tickets easier and easier, and as buyers and sellers became
comfortable with StubHub as a middleman, Amy wasn’t needed as much.
As she put it to me in 2014, “StubHub has been our partner, but they are
getting to be our competitor. . . . I doubt there’s a future in selling twenty or
thirty season tickets for a college football team. We have built our lives on
this business, and I don’t know what the future holds.” Yet even as StubHub
is making the newly democratized ticket business more difficult for small
resellers, it faces pressures of its own. Teams, event planners, and even
speculators are trying to take back the market from the people.

The Evolution of the Ticket Market I: The Past

One of the first recorded uses of tickets was for events at the Roman
Colosseum—gladiator contests, chariot races, and the like. Seats were
assigned; pottery shards served as indicators of where to sit. Although entry
was free, demand for seats was often huge, so people had to line up for
tickets. Citizens of higher class got preferential treatment and seating. We
have no surviving record of ticket resale in that era.2
Sometime in the intervening centuries, the people who organized
sporting events began to see tickets as a source of profit, which led to the
first big change: tickets started to cost money. (Also at some point, the
pottery shards were replaced by paper tickets; paper reached the Arab world
in the eighth century A.D. and Europe in the eleventh.) Fans began to take it
for granted that if enough people wanted to see a sporting event, they would
be charged. But for a long while, sporting events did not have advance
ticket sales.
This trend was broken in Renaissance times. Sixteenth-century English
theater operators are the first group documented to have sold tickets at
different prices and to allocate better places to people willing to pay more.
They also realized that when tickets were scarce, it was useful to give
patrons a chance to plan ahead.3 For their most hotly anticipated events,
they forged an additional innovation: they introduced advance sales, to
allocate tickets through a combination of time and money.
Over the centuries, advance ticket sales became more common, and at
some point they gave birth to the ticket resale business. Although people
liked the opportunity to buy in advance, plans sometimes changed. People
bought tickets to games and then found they couldn’t go, or decided shortly
before a sold-out big game that they wanted to attend and needed a way to
get in.
Enter ticket brokers—and, as a subset of them are less affectionately
known, scalpers. Economists love scalpers, or at least one primary aspect of
scalping. Ticket resale creates what economists call “gains from trade.”
Suppose you had tickets to an NFL game, but your daughter’s soccer team
made the playoffs, and her game created a conflict. The football tickets
would now be worth less to you than they were when you had originally
planned to attend the game. By selling your tickets to someone who valued
them more, you could make some money and also make the world a
marginally happier place. If you valued the ticket price at zero once you
decided to go to your daughter’s game, and someone out there was willing
to pay $100 for it, then you could sell the ticket to that person for anything
less than $100 and make you both better off. You get some money instead
of a ticket that would go to waste, and the other guy gets a ticket for less
than what he was willing to pay for it. It’s a win-win.
Although that’s a nice story, it wasn’t always so easy to get to the happy
ending. In the days before StubHub, it was hard for someone with an extra
ticket to find someone else who wanted it. That’s where scalpers and ticket
brokers came in. Once your daughter’s game was scheduled, you could take
your ticket down to (the imaginary) Vinny’s Ticket Shop at the local
shopping center. You could sell Vinny your ticket for $50, and the guy who
was willing to pay up to $100 could come in and buy it from Vinny for $80.
You and the new buyer could have done better if you had found each other
independently of Vinny and you sold him the ticket for, say, $60. But in pre-
internet times, that transaction was usually either impossible or so time-
consuming to set up that the money you made was not worth the hours you
put into it. So the broker—or, if you prefer, scalper—was valuable: Vinny’s
presence made you, the guy who ultimately bought the ticket, and Vinny
himself all better off.
In pre-internet times, scalpers and brokers also provided another
valuable service: they acted as simple distributors. Depending on where you
lived, a trip to the stadium box office could be inconvenient. If Vinny went
to the box office, bought a bunch of tickets, and sold them in the
neighborhood of Vinny’s Ticket Shop, everyone was better off. Just as
Walmart provides a service by selling soap so that you don’t have to go to
Cincinnati to buy it directly from Procter and Gamble, Vinny provided a
valuable service by cutting down travel and wait times.
Although economists love ticket scalpers, lots of other people hate them.
If you type “ticket scalpers” into Google’s search bar, the autofill engine
comes up with entries like “ticket scalpers are bad actors” and “fans hate
being gouged by ticket scalpers.” Ticket brokers don’t get as bad a rap—the
distinction being that scalpers operate outside arenas at game time, and
brokers buy and sell tickets leading up to the game—but they rarely get a
lot of praise.
One simple driver of scalpers’ bad reputation is bad experiences. As
Chris Tsakalakis, former president of StubHub, told me, “Anyone who ever
paid more than face value for a ticket doesn’t believe in economics, and
anyone who ever received less than face value for a ticket doesn’t believe in
economics.” In other words, people used to get indignant (and often took
their anger out on scalpers) when they felt they paid too much as buyers or
got too little as sellers. They were not the ones getting a good deal; the
scalpers were.
Another downside of doing business with ticket resellers was the
possibility of outright fraud. Some scalpers, it’s true, sold counterfeit
tickets. If a scalper sold you a ticket in front of the arena and the ticket
turned out to be a fake, you were very unlikely to get your money back.
Outright dishonesty is indeed a bad thing. But we have little evidence to
suggest that fake tickets ever posed such a problem that counterfeiting alone
should have shut down the secondary market.
Another reason people were wary of brokers and scalpers was the belief
that scalpers often bought up all the tickets to a game and drove up the
resale price. This scenario could theoretically lead to seats sitting empty
despite there being people who would have occupied them if the tickets
were reasonably priced. But while fans and teams regularly accuse scalpers
and brokers of trying to corner the market, essentially no evidence indicates
that they have ever tried—at least not at games with a large capacity. At a
baseball or football game with forty thousand seats, a group of scalpers
would have to buy thousands of tickets to corner the market, a feat for
which they likely lack both the opportunity and the resources. I don’t know
of a single documented example of market domination by scalpers.
In their heyday, the biggest argument against scalpers and brokers was
probably that they wasted time and resources. For any given fan, a scalper
might save that person the time of standing in line. But, as a group, scalpers
made it harder for fans to get access directly to tickets when they became
available, and forced fans to search around before eventually finding the
scalper. If fans could have bought and sold without a middleman,
presumably they could have kept more of the money, and the potential
scalpers and brokers could have done something more productive with their
lives.
Consider the buildup to the Boston Bruins’ appearance in the 2011
Stanley Cup Finals, their first in more than twenty years. On the night
before the team box office put tickets up for sale, hundreds of fans camped
out in line; when the box office opened the next day, the games sold out in
minutes.
But a lot of the buyers weren’t fans at all. Many of them walked directly
across the street to a row of ticket broker offices to sell their tickets for a
profit.4 These resellers’ decision to camp out for tickets that would have
sold out either way wasted the time of everybody in line by creating a
longer wait. One economic analysis of rock concert tickets (a market very
similar to sports tickets) showed that the effect of resellers in the early
arrival game was truly costly, in the sense that they wasted a lot of people’s
time.5
 Overall, ticket brokers did hurt fans who regularly bought tickets early.
But those costs were more than offset by the gains brokers provided to fans
who wanted to buy tickets closer to game day, avoid the trouble of waiting
in line, or easily sell tickets if their plans changed. So while a few fans—
those who waited extra long in ticket release lines or accidentally bought
counterfeit tickets—had good cause to dislike ticket resellers, the
antiscalper vitriol was unwarranted for the vast majority of ticket buyers.
Like fans, teams have also historically opposed scalping. From a selfish
perspective, this attitude is perfectly understandable. For one thing, when
demand for a game was low, resellers could purchase tickets below face
value from season ticket holders and fans who found out they could not
attend a game after prepurchasing tickets. If they resold these tickets to
other fans below face value, that discount hurt a team’s reputation and
encouraged fans not to bother purchasing tickets directly from the team.
They could just wait until they found tickets below face value, which, in
turn, discouraged people from buying season tickets at full price in the first
place.
But the main reason teams hated scalpers is that the teams wanted to be
more like airlines. Although airplane tickets and sports tickets are in most
ways similar, airlines have a big advantage that teams do not: their tickets
cannot be resold. A person who buys a flight ticket has to show
identification to prove that she is the one entitled to use it. Thus, as the date
of a flight approaches, the price of an airline ticket goes up, often
dramatically. Prices are higher, sometimes a lot higher, for flights that leave
tomorrow than for those that leave in a week; flights that leave next week
are, in turn, more expensive than those that leave in a few months.
For sports, these trends are inverted. The economist Andrew Sweeting
crunched the numbers for baseball ticket sales on eBay and found that a
ticket that sold for $50 two months before a game was worth about $38, on
average, a week before the game, and $30 on game day.6 Unsurprisingly,
the game-approaching discount is much greater for tickets sold by
individuals—those who are just trying to get rid of tickets they can’t use,
and by game day will often give them away for free—than for brokers.
Ticket brokers undermined teams’ pricing power by buying tickets at the
advance purchase price and undercutting their desire to raise prices as the
event neared. A team couldn’t charge a lot for a last-minute ticket when a
scalper in the parking lot was offering a lower price. But while teams have
always hated scalpers for this reason, fans should love them for it. They
create competition in the market, and for buyers, more competition among
sellers is always better.
In any case, until recently, sports teams disliked brokers and scalpers, but
apparently not enough to get rid of them. Remember, it’s easy to get rid of
ticket resale if you sell tickets only at the door, just before the event. Teams
tolerated brokers as a side effect of advance sales.

The Evolution of the Market II: The Present

This tenuous relationship of grudging tolerance among fans, teams, and
scalpers continued unchanged for pretty much the entire second half of the
twentieth century. But the internet transformed the industry. The arrival of
sites like StubHub put significant pressure on brokers like Amy’s Tickets,
since people with extra tickets could now easily find people who wanted to
buy them.
How did StubHub’s entry change brokers’ and scalpers’ business
models? First, StubHub tore down the barriers to entry for people who
wanted to enter the ticket resale business. Ticket brokers once needed to
invest in storefronts, or at least spend money developing and marketing
their websites. But “how to become a ticket broker” articles now abound on
the internet, and most of them suggest operating as a StubHub super-user.
This may be the one promise that you can get into a business opportunity
with “no money down!” that is actually true. But this ease of entry makes
competition fierce, so new entrants are likely to have trouble earning much
of a living.
A second effect of StubHub is that it has made the secondary market for
tickets much “thicker”; that is, a lot more tickets are being offered for sale,
all on a single site. This has put serious pressure on ticket resellers’
margins. More tickets are great if you’re looking to buy, but not so much if
you’re the seller.
A third downside for brokers is that they are getting cut out: consumers
now buy and sell on StubHub without middlemen. According to one
estimate, 60 percent of StubHub sales in the company’s early days were
made through brokers, but a few years later, 60 percent were direct sales
from one fan to another.7 A scalper summed up the situation succinctly to a
New York Times reporter: “StubHub is killing us.”8
 Finally, StubHub has trained consumers to be smarter. The pattern of
resale prices dropping as a game approaches wasn’t a problem for brokers
and scalpers when they could count on selling tickets at high prices way in
advance to fans who didn’t want to risk getting shut out of a game. But
many of those fans have now figured out that, thanks to StubHub, tickets
will still be widely available in the days leading up to the game.
StubHub has thus displaced almost all the economic value once provided
by scalpers and brokers. There’s a new middleman in town, and it has made
the market more competitive while lowering transaction costs for buyers
and sellers. Vinny’s Ticket Shop, the scalper in the parking lot, and Amy’s
website are under siege and may not survive. If they don’t, what will
replace them?

The Evolution of the Market III: The Future

Technology has already dramatically changed the economics of ticket
resale. But it has not made much of an impact on how teams sell tickets in
the first place. To be sure, there are some changes: Philips Arena, for
instance, now goes directly to customers through Ticketmaster, when it
once offered a first-sale contract to Amy. But as teams have been doing for
decades, they are selling individual and season tickets to fans at the highest
prices they can get.
Things are changing, however, as teams and leagues busily work on two
initiatives—also borrowed from long-successful airline strategies—that
they hope will noticeably increase ticket revenue. One of these ideas is
good for the teams, and maybe for fans too. Economists are all for it. The
other is an economic disaster waiting to happen.
First, the constructive venture: “dynamic pricing.” This term is heard
regularly from team ticket sales departments nowadays, but other industries
have practiced the idea far longer. Hotels, for example, have known for
years that it makes financial sense to vary prices based on factors like the
day of the week, the time of year, and how many visitors are expected in
town. Prices also change as the hotel gets an idea of how much availability
remains on a given approaching day. In short, hotel rooms are a scarce
resource, and when demand is high, hotel owners take advantage of this
scarcity. More recently, Uber and other rideshare platforms have used
dynamic pricing (known in this context as “surge pricing”) to equate supply
and demand based on current conditions.
 Sports teams are only now starting to vary prices based on realized, up-
to-the-minute demand. The first step teams took was to price different
games based on expected willingness to pay. So the New York Yankees, for
example, set much lower prices for a weekday game against the Oakland
A’s—my favorite team, but apparently not the country’s—than for a
Saturday night game against their archrival, the Boston Red Sox. This type
of price discrimination, which follows the same logic as hotels charging
more for days on which they expect less-price-sensitive business travelers,
has slowly been creeping into the ticket market for years and is now fairly
widespread.
Taking the same line of thinking a step further, dynamic pricing involves
making a given game’s tickets more and less expensive as demand ebbs and
flows. Teams don’t want to be perceived as unfair, and they especially don’t
want to alienate season ticket holders by making tickets available at low
prices for those who wait, so they have been slow to adopt dynamic pricing.
But teams in British soccer leagues and Major League Baseball have started
to use dynamic pricing, and it has become a hot topic among sports
executives.9
As the Chicago Tribune reported, “Want to go to the Cubs game against
the Colorado Rockies this week? A check of the Cubs website Tuesday
showed that a bleacher ticket for Monday night cost $19. On Wednesday,
the same ticket cost $21. Why the $2 increase? Welcome to the world of
dynamic pricing.”10 As with hotel room pricing, dynamic pricing models
use a mix of computer algorithms and analyst decisions to generate a
tailored ticket price based on the demand for any particular game.
Dynamic pricing will increase teams’ ticket profits. One baseball
executive says, “Dynamic pricing can lift average ticket prices 10 to
15 percent over the course of a season. During a playoff race, dynamic
pricing can lift that percentage over the 20 percent average.”11 The
economist Andrew Sweeting estimates the revenue increase at 16 percent.
That is a valuable boost, but not a true game changer, given that ticket sales
make up less than 30 percent of Major League Baseball revenues. The
Arizona Diamondbacks, for example, reaped $54 million in ticket revenue
in 2019, so a 15 percent bump in that number owing to dynamic pricing
would yield an additional $8 million: good money, sure, but only about a
quarter of the annual salary Arizona paid the pitcher Zack Greinke that
year.12
Although the added revenue may not be dramatic, dynamic pricing
allows teams to achieve another important goal: selling more tickets.
Selling more seats matters to teams more than optimizing the price of
individual tickets. Tickets can certainly be expensive, but they are often
sold below the price the market would bear. Teams are willing to forgo
some ticket revenue in the short term because having sellouts helps create
interest in the team and, in turn, more ticket sales in the long term. Also, an
empty seat is extremely costly to a team because of the loss of parking,
concessions, and merchandise revenue.
It’s probably not a surprise to hear that having high attendance numbers
helps a team’s reputation. As anyone who has seen an Ottawa Senators or
Florida Panthers highlight on SportsCenter knows, it doesn’t reflect well on
those teams to have more people watching from the penalty box than from
the stands.13 Moreover, full stadiums heighten the overall fan experience,
just as eating in a full restaurant can enhance one’s enjoyment of the food.14
Teams want more people at their games because each additional fan
brings a marginal increase in utility for the rest of the crowd. Additionally,
benefits arise from large stadium audiences beyond the mere feeling that
one is partaking in something popular. Many fans enjoy engaging in
routines such as “the wave” or “Let’s Go [Home Team]” chants. Both of
these rituals require a critical mass of fellow fans and increase many
people’s enjoyment of the game.
Furthermore, teams speculate in creating bigger fan bases in the future.
Teams have a vested interest in inflating attendance today, even at the cost
of a few dollars in short-term profit, because fans who have a good time at
the game are more likely to return. Sports fandom can have a nearly
addictive power; serious fans may feel compelled to check in on their
favorite team even during dinner parties or weddings.15
In short, teams have strong incentives to get fans into the seats. And
dynamic pricing may turn out to be a critical way for teams to keep their
stadiums full without underpricing tickets as much as they might have
needed to in the past. Moreover, dynamic pricing helps teams without
hurting fans on the whole. True, a few people will pay more—those who
used to line up to get Yankees–Red Sox tickets when they cost the same as
Yankees-A’s tickets—but teams are only helping fans allocate the tickets the
way brokers helped allocate them in the old days. Dynamic pricing is here
to stay, and most of us should be fine with it.
As for the second idea some teams are hoping to adopt from the airline
industry—nontransferable tickets—I am not so optimistic.
You cannot sell your airline ticket to another passenger. Your name is on
the ticket and boarding pass, and some combination of airport security and
airline staff will stop you from getting on a plane with a ticket sold to
someone else. Logistically, that kind of enforcement is tricky. Airlines can
only stop you from using another person’s ticket because they demand
identification multiple times before you board.
Entry to games under this system would become more complicated than
it is now, but there’s no reason stadiums can’t enact the same requirement:
they could sell a ticket with your name on it and then ask for identification
as you enter the stadium. Teams are reluctant to impose this burden on fans,
however, both because it could slow stadium entry and because they don’t
want to alienate loyal season ticket holders, who would lose the ability to
sell off or give away their tickets. Still, nontransferable tickets are
becoming more common at concerts, where season ticket holders are not an
issue.
Promoters generally appeal to the evil of scalpers to explain
nontransferability. Resellers do impose extra costs on fans, either by
requiring them to pay more than the listed price for tickets, or by making
them spend time and resources to compete with brokers to buy tickets as
soon as they become available. This problem has been exacerbated in recent
years by the rise of bots—computer programs that buy tickets online as
soon as they become available, strictly for the purpose of reselling those
tickets at a profit later. Given how easy it is to write this kind of computer
program, their proliferation has made the scalping market extremely
competitive. Resellers lucky enough to get high-value tickets right away
might make a nice margin on them, but for many others, the business is not
very profitable.
A fan really has no reason to prefer to pay high prices directly to the
teams rather than to brokers who manage to get tickets early and cheap, but
somehow the creators of bots and other resellers continue to draw much
more of the fans’ wrath than team owners do. So teams are working to
severely restrict resale, arguing that they are looking out for fans against
those greedy ticket brokers.
More and more sports teams and leagues are making deals with a single
resale channel to make them the “exclusive” ticket outlet. If you’re
interested in buying tickets to the NCAA Final Four men’s or women’s
basketball tournament, for instance, you’re likely to end up at
PrimeSport.com. That’s not because you will find their site more user-
friendly than other options; odds are you’ve never heard of PrimeSport, and
there’s really nothing special about it relative to its competitors. You will
end up at PrimeSport because the company has an exclusive arrangement
with the NCAA. The NCAA does not sanction the selling of Final Four
tickets on StubHub, and there is no guarantee of authenticity. StubHub
cannot provide tickets to customers as seamlessly as PrimeSport.
Is that really a problem? Aren’t tickets still available for resale? Well,
yes. Because it has a quasi-monopoly, PrimeSport can charge a higher
commission than it would if it had to compete on equal footing with
StubHub. This added profit margin enables PrimeSport to pay the NCAA a
handsome fee to be the exclusive agent, and the NCAA in turn increases its
profits.
These types of arrangements are not as bad as pure monopolies, but they
do stifle competition and hurt fans. Also, although teams claim to be
looking out for fans’ interests by cutting brokers out of the process, the
teams are not trying to lower prices for fans but rather trying to capture the
brokers’ profits. The law professor Chris Sagers has glimpsed the future of
ticket sales, and he warns that if laws are not passed to protect consumers, it
may not be pretty. “Artists, promoters, teams and ticket companies,” he
argues, “want to restrict what fans can do with their tickets, and ultimately
want to corner the now-lucrative resale market. They claim that their goal is
to eliminate scalping, but the realities are that they will profit from ticket
resale if they can control it, and they want to eliminate competition in resale
markets in order to keep their initial ticket prices high. Sports teams are so
committed to the latter goal that some secretly impose price floors on their
affiliated resale exchanges, and some . . . have threatened to confiscate
season tickets of fans who resell on competing exchanges.”16
The “price floors” Sagers refers to are restrictions that prevent fans from
selling tickets on an exclusive resale site for less than face value. Although
these rules ostensibly protect season ticket holders’ investments, in practice
they lead to empty seats and many fans not going to a game that they would
happily have attended if they could have bought a ticket at the market price.
StubHub, the official resale site of Major League Baseball, allows people to
sell tickets well below face value. This leads to more seats being sold but
some unhappiness among sellers. Chris Tsakalakis, the former StubHub
president, remembers being harassed at a cocktail party by an investment
banker who owned San Francisco Giants season tickets. “He asked me,
‘Can you do something so prices don’t go below face value? I know how
supply and demand work, but I don’t like it for myself.’”
As teams and leagues impose more restrictions on resale and use
dynamic pricing to fill seats without scalpers, what is the future of ticket
scalping and brokering? I foresee two recent innovations turning into larger
trends.
Amy’s Tickets provides the first answer: value-added services that go
beyond simple tickets. Economic forces are turning Amy’s Tickets into
Amy’s VIP Events. Amy explained that Amy’s VIP Events offers not only
tickets but hospitality, hosting, and transportation. If you want to go to a
major game or tournament and have it feel more like a wine mixer, you
should call her. It’s an upscale business for an expanding market—the
American leisure class, with money to spend in search of a unique and
memorable experience. This is Amy’s future. “I am doing bigger events,”
she says, “and getting away from the day-in-and-day-out of selling tickets.
The VIP side is definitely an area where I can grow.”
The big ticket resellers are not far behind. StubHub throws a party for all
its Super Bowl customers, for example, and PrimeSport hosts an array of
“VIP Hospitality” offerings at the NCAA Men’s Basketball Final Four,
including sports bar games, drinks and food, and a chance to meet legends
of college basketball. Amy’s Tickets has a more personal touch, but it may
be tough for her to stay ahead.
The other innovation with a promising future is yet another example of
teams borrowing an idea from the airline industry: the upgrade. A few
Major League Baseball teams have already rolled out an in-game upgrade
program. If you get to the stadium and don’t like your view—and you’re
not daring enough to try to sneak past the ushers guarding the better seats—
you can now pull up a league-run app on your phone and pay to move to an
empty seat in a better location. Every game has a few no-shows or some
people who never succeeded in reselling their extra tickets. Through in-
game resale apps, the “invisible hand” of the market can better use those
resources while also earning teams a little profit.
Upgrades can also help make the stadium look and feel more full,
enhancing the game experience. Teams that have a no-show problem, like
the Yankees, are especially likely to employ this scheme. The Yankees
charge an arm and a leg for good seats because New York is full of rich
financiers who give the seats to clients, entertain at the stadium, and even
occasionally take their families to a Yankees game. There’s only one
problem: these ticket holders often don’t bother showing up. The journalist
Chris Connors lamented “the glaring number of empty seats in the lower
levels of the stadium” and noted that “the Yankees continue to undermine
many of their most dedicated fans by pricing them out of the most coveted
seats in their palatial residence.”17 The no-show “fans” with great seats
alienate the poorer but more avid fans sitting in the upper decks.
If the no-show seats can be filled with those more devoted fans through
upgrades, the atmosphere at Yankee Stadium will improve, and the Yankees
can siphon a bit more money from their loyal followers. There is, however,
an impediment that may keep upgrades from becoming the norm. The
journalist Tom Ley concludes that the Yankees do not “want the kind of
riff-raff that buys cheap tickets on StubHub mixing with rich fans who pay
full price to sit in the same section.”18 For now, upgrades are making
inroads with teams that have fewer super-rich fans, like my local favorite
Oakland A’s.19
Perhaps some team will think outside the box and try a promising
solution to the empty-seats problem: give tickets away. Could a team do
better by just giving tickets away on a first-come-first-serve basis and
making up the lost ticket revenue from concession sales and the goodwill of
its fans, which would presumably pay off in higher viewership and more
expensive TV rights? Although I would love to see a team experiment with
free tickets, it is hard to see it becoming the hot new trend. Ticket sales are
small compared with TV rights, but they do bring in a big chunk of revenue
that concessions would not replace. Also, if people didn’t pay for their
tickets, they might not feel as invested in the game and might even skip it if
they got busy or the weather was not perfect. Perhaps, in this scenario, no-
shows would make the stadium emptier than when prices are high. But,
most important, free tickets would probably not end up being free; bots and
brokers would gobble up the good seats and then resell them. Owners have
learned to tolerate letting brokers have some of the profit from ticket sales
—but don’t count on their giving all the profit to brokers.
The past and the present have been good to ticket holders: the thicker
market has made it easier to buy and sell. We can hope the future looks
even brighter, but as teams get more sophisticated, be ready for them to
continue to figure out ways to squeeze more money from ticket sales and to
cut out the scalpers. Perhaps governments around the world will protect
buyers and resale markets, but given the teams’ long history of shaping
public opinion against scalpers, I wouldn’t count on it.

OceanofPDF.com
 8
Why Should You Be Upset If Your Hometown Hosts
the Olympics?

Nursultan Nazarbayev has always been good at getting what he wants.

Born to nomadic parents, he rose through the political apparatus in his
native Kazakhstan, and in 1989—when Kazakhstan was still part of the
Soviet Union—he became the leader of the state. Four years later, he
became the first president of independent Kazakhstan, a position he retained
for thirty years by regularly getting reelected with over 90 percent of the
vote. He was, of course, functionally a dictator, though opinions vary on
whether he was a benevolent one. Under his control, the resource-rich
country did reasonably well economically and formed close ties with
Russia.1
Yet Nazarbayev could not attain one thing he seemed to truly crave:
international recognition for himself and his country. Kazakhstan is best
known, at least in the United States, as the home of Sacha Baron Cohen’s
fictional character Borat. Although Kazakhstan is larger in area than
western Europe, it’s likely that few western Europeans could pick it out on
a map. Even fewer have heard of Nazarbayev. His attempts to bring renown
to his country, which include paying the former British prime minister Tony
Blair millions of dollars for advice and exchanging public praise with
former Italian prime minister Silvio Berlusconi, don’t seem to have helped
much.
Consequently, Nazarbayev tried another tack: Almaty, Kazakhstan’s
biggest city, put together a bid for hosting the Winter Olympics in 2022.
The effort was initially dismissed as a long shot; observers noted the
country’s relative anonymity, remote location, and sketchy human rights
record.2 But the bid began to attract attention. Kazakhstan hoped to use the
Olympics to “make a big splash on the world stage,” the New York Times
reported. “It is a springboard for conquering new heights,” Nazarbayev
crowed to the press. “Kazakhstan should be known as a nation of winners.”
In the end, though, Nazarbayev’s big moment was snatched away. In the
summer of 2015, the Olympic Committee announced that the Games were
being awarded to Beijing. The decision gave some indication of the
committee’s low esteem for Kazakhstan, given that Beijing’s drawbacks
include terrible air pollution, getting so little snow in its nearby mountains
that it needs tons of environmentally concerning manmade snow, a plan to
spread the events over hundreds of miles, and having recently hosted the
Summer Olympics.
How and why did the choice for 2022 come down to two flawed cities in
countries ruled by repressive regimes? As it turns out, the warped economic
incentives behind hosting big sporting events almost gave Nazarbayev his
big shot. Only maligned or ignored nations, anxious to display their
greatness, were willing to waste the money and effort.
This international lack of interest in hosting is a new phenomenon; in
2013, when the Olympic Committee first requested bids for the 2022
games, many cities seriously began the process of bidding, and six cities
ultimately prepared detailed proposals. Soon, however, the trouble began.
First Stockholm backed out because of political opposition in Sweden. In
May 2014, Krakow, Poland, held a referendum on its bid and, after finding
its citizenry opposed, withdrew as well. Lviv, Ukraine, withdrew shortly
afterward in light of a domestic political crisis. That withdrawal left Oslo as
the leading contender: Norway had once hosted a successful Olympics, was
swimming in money thanks to high oil prices at the time, and is a perennial
leader in the Winter Olympic medal count. But in October 2014, after being
named a finalist along with Almaty and Beijing, Oslo also dropped out
owing to overwhelming local opposition.3
What the residents of Oslo, Stockholm, and Krakow recognized is that
hosting the Olympics means three weeks of revelry followed by years of
economic fallout. Massive public investments that could be directed toward
needed government programs are instead used for constructing large
stadiums and other athletic facilities. The economist Alan Blinder once said
that “economists have the least influence on policy where they know the
most and are most agreed.”4 He had in mind practices like rent control and
trade barriers, both of which are derided by virtually every economist, yet
still persist. But as the economists John Siegfried and Andrew Zimbalist
have written, “Few fields of empirical economic research offer virtual
unanimity of findings. Yet, independent work on the economic impact of
stadiums and arenas has uniformly found that there is no statistically
significant positive correlation between sports facility construction and
economic development.”5 Economists’ skepticism regarding Olympic and
World Cup bids is decades old. Only recently, however, as the dearth of
bids for the 2022 Games shows, have politicians and taxpayers shown signs
of waning enthusiasm for these investments.
The problem comes down to two pervasive economic phenomena:
“agency problems” (that is, misalignment of interests between the people
who make a decision and those who pay for it) and old-fashioned
corruption. Consider three examples:

• Vladimir Putin essentially decided, on behalf of all of Russia, to spend

without limit on the 2014 Winter Games in Sochi. The final price tag
of $51 billion made the 2014 Games the most expensive Olympics
ever, summer or winter. In preparation for the Games, for example,
Russia built a rail link from the Olympic Village to the ski events. It
extended thirty-one miles and cost $8.7 billion, three times the
amount that NASA spent on the newest generation of Mars rovers.
The head of the railway agency that built the link is a close associate
of Putin’s and has ties to local banks and to the two companies that
were given the contract without competitive bidding.6
• The state of Wisconsin agreed, in 2015, to pay $250 million toward
building an arena for the NBA’s Milwaukee Bucks. After accounting
for interest and many hidden costs, the true price to taxpayers will be
much greater. The team is owned by the hedge fund managers Wes
Edens and Marc Lasry, each of whom is worth a couple of billion
dollars, but they nonetheless threatened to move the Bucks if the
arena was not publicly funded. The state’s commitment to the arena
was approved by Governor Scott Walker in the same year that he
signed legislation authorizing $300 million in budget cuts for
Wisconsin’s public universities.7
• Perhaps the only sports event that can compare with the Olympics in
terms of wasted public resources is the World Cup. Qatar won the
rights to host the World Cup in 2022 through outright bribery of the
 officials responsible for deciding where the event would be held. The
cheating was necessary because there is literally no other reason to
hold the World Cup in Qatar: it has sweltering temperatures, very few
appropriate athletic venues, and no noteworthy soccer tradition. It has
never previously qualified for the World Cup, though by virtue of
hosting will do so in 2022. The event had to be scheduled for the late
fall to avoid the worst of the summer heat. Qatar naturalized a number
of talented African players to try (unsuccessfully, it turned out) to put
together a team that could qualify for the 2018 World Cup, and the
government has set about building a collection of new stadiums at
fabulous expense.8 Zimbalist estimates that Qatar will lose
approximately $200 billion on the event.9

The common thread in each of these and countless other examples is that
individuals had the power to decide how a much larger group of people
spent their money. This misalignment of interests led to an elected or self-
declared leader making financial decisions that hurt constituents while
making himself or herself better off.
Still, hopeful news is on the horizon. Fans and citizens are pushing back
against the idea of taxpayers shelling out for sports events and facilities.
Will they save cities from frittering away money? Maybe, although those
who champion Olympic bids are trying to stay a step ahead of those who
insist on fiscal responsibility.

A Walk in the (Olympic) Park

The people who lead Olympic bids, as well as those who advocate
building new stadiums, argue that the events create lasting economic
vitality. But these claims simply hold no water. To get a sense of the
problems with the hosting logic, consider what is widely accepted as one of
the more financially successful Olympic Games: the London Summer
Olympics in 2012.10
The newly built venues in the area the London organizers called
“Olympic Park” include the Aquatics Center, at a cost of $460 million; the
Basketball Arena, $62 million; the Velodrome, for cycling, $180 million;
the Riverbank Arena, for field hockey, $32 million; the Water Polo Arena,
$46 million; the Copper Box, for team handball preliminary matches, $75
million; and, of course, the Olympic Stadium, for track and field, which
cost $832 million. For those keeping score at home, that’s a grand total of
$1.69 billion—and that money did not cover hosting sports like fencing,
surfing, or beach volleyball, all of which required upgrades to existing
facilities.11
What has happened to those facilities since the Olympics? The Aquatics
Center, after significant and costly post-Olympics renovations, opened to
the public in 2014. The Basketball Arena, which was designed to be
portable, was dismantled and put up for sale, but it has never been reported
as sold. The Riverbank Arena and Water Polo Arena were demolished
shortly after the conclusion of the Games. The Velodrome remains in place
as a potentially top-flight indoor cycling venue, but the British national
team decided to keep its training base in Manchester. The Copper Box has
hosted events such as home games for club basketball and handball teams
and a mixed martial arts showcase. The Olympic Stadium was given for
free to the soccer team West Ham United F.C. and subsequently converted
into a full-time soccer stadium for an additional $400 million, most of
which was covered by the British government. It’s worth noting that even
before the Olympics, London already had several large stadiums, each of
which was used about twenty times per year.
The revenues from those two and a half weeks of the Olympics needed
to be otherworldly to compensate for the costs of hosting them. The Games
brought some added attention to London and may well have brought some
added tourism and new business deals. But it is hard to make the case that
the city of London would not have been better off spending $14 billion on
something else (like improving transportation or retraining workers who
have lost manufacturing jobs, both of which have longer-lasting benefits),
rather than hosting a glorified showcase of archery and fencing. And
remember: the London version has been generally accepted as a relatively
successful Olympics, financially; such events often waste even more
money.

The Faulty Logic in Favor of Public Financing

Team owners, much like politicians justifying an Olympic bid, also push
cities to build stadiums with promises of financial windfalls. A team
wanting its home city to build a new facility for it obviously does not take
the position that the venue’s construction is likely to incur massive losses.
Instead it rolls out a series of reports by economists and developers—all, of
course, on the team payroll—showing why the new structure is guaranteed
to bring an economic boom. Given that it’s not hard to brainstorm ways in
which a sparkling new arena or ballpark might help a city’s economy, we
can charitably suppose that the people who write these reports believe their
claims hold at least some water.
Consider Golden 1 Center, the arena built for the Sacramento Kings in
2016. The city of Sacramento, which is on the hook for $255 million of the
estimated $477 million construction cost, created a website to promote the
deal to its taxpayers. One of the FAQs concerned the new arena’s financial
benefits; the city’s answer was that the project would “increase the vitality
of downtown areas” through the specific avenues of “investment from
businesses,” “sales, property, hotel, and utility tax revenues,” and
“thousands of jobs during construction . . . and hundreds of permanent
positions.”
The Kings’ idea was that businesses would want to move to the same
neighborhood as the hot new arena because it would attract fans, who
would then frequent the new businesses on their way to or from a game,
thus growing the local economy. Add in the resulting tax gains for the
government and the extra jobs created for residents, and this proposition
sounds pretty promising.
But do these benefits actually accrue to the area around the sports venue?
In the case of Golden 1 Center, things have gone much better than you
might expect. First of all, the final cost was about $560 million, just 17
percent over budget. Also, because Sacramento’s city government did an
unusually good job of negotiating the contract, the Kings’ owners paid for
all of the $83 million cost overrun. That, sadly, is a huge victory by the
standards of American professional sports stadiums.
As to the arena’s benefits, the Downtown Sacramento Partnership issued
a fact sheet listing Golden 1 Center’s accomplishments on its second
anniversary. The document notes tremendous property sales and
investments, as well as large increases in employment, in downtown
Sacramento in the first two years of the arena’s life. It also notes that
downtown office vacancy rates are at a five-year low and that downtown
foot traffic is up substantially.
Those facts are all true, and it certainly seems that the new arena did not
hurt the local economy. But did it help? Answering this question requires
knowing two things: what would have happened if the city had not built the
arena, and how did the economic activity in the area around the arena affect
the economic activity of nearby areas?
From 2015 to 2018 (the time of Golden 1 Center’s construction and its
first two years of use), several trends that had nothing to do with the arena
affected the Sacramento business environment. The economy was booming,
with the city’s unemployment rate dropping from 6.0 percent to 3.9 percent;
demographics favored downtown markets such as the area around Golden 1
Center; people were fleeing the increasingly expensive Bay Area to smaller
cities like Sacramento; property values boomed; and office vacancy rates
dropped in all major cities anywhere near Sacramento. (To illustrate that the
phenomenon of local property value growth was not confined to
Sacramento, note that the vacancy rate in Reno, a somewhat smaller city
130 miles to the east, fell at the same rate as Sacramento’s, while the second
tech bubble in San Francisco, ninety miles to the west, made Sacramento’s
vibrant center look like a one-horse town.) In short, Sacramento boomed at
the same time that the Golden 1 Center was being built and opened, but it
would have gone through a strong period of economic development in any
case.
The Downtown Sacramento Partnership had lots of good numbers to
back up its argument. But the right way to evaluate the city’s $255 million
public investment isn’t to look at how the area did; it’s to ask if that $255
million was used effectively. The true economic cost of the arena includes
both the direct cost and the opportunity cost. Any given purchase is
worthwhile only if its benefits outweigh its cost and if it is the most
efficient use of the resources used to make it. Imagine entering a thrift shop
in which everything costs exactly $1 with a single dollar bill in your pocket.
Suppose you find a valuable baseball card worth $25 and an old Super
Bowl ticket worth $20. If you only had a dollar to spend, you would
obviously make a profit buying the ticket, but you would forgo an even
greater profit by not buying the card. Buying the ticket is thus a poor
allocation of your very scarce resources—your one dollar.
The same applies to stadium deals. Sacramento’s $255 million
commitment to the Kings is modest on the grand scale of publicly financed
arenas, and its population is on the smaller end of major-league sports
cities. To the city’s credit, Sacramento made sure it would not pay for cost
overruns and seems to have done other things to encourage economic
development in the area. Just as London was an example of an Olympics
that minimized the damage, Golden 1 Center did not bleed ungodly
amounts of public money.
Still, rather than build a new arena for a wealthy NBA team, the city
could have used that $255 million to employ more than two hundred public
workers for $50,000 a year for twenty-five years. It could have built better
roads or better water systems to manage the area’s frequent droughts; it
could have funded a 1,000 percent increase in the endowment of
Sacramento State University, which already generates over $800 million for
the local economy every year; or it could have made a large dent in the
city’s $2 billion pension plan deficit. Sacramento didn’t really have $255
million to spend, but since it decided to do so, any of these investments
would almost certainly have returned more to the city’s economy than the
$255 million contribution to the new arena. And yet the city government
pushed the deal, and Golden 1 Center now hosts lots of Kings’ games and
other events every year.
Sacramento made a mistake. Study after study by economists shows that
for the investment involved, stadiums and arenas never produce a favorable
return on taxpayer dollars. Even when the area around a facility does well,
it only draws economic activity that would have gone to other forms of
entertainment a few miles away. As Siegfried and Zimbalist put it, “While
sports teams may rearrange the spending and economic activity in an urban
area, they are not likely to add much to it.”
Why don’t stadiums and arenas provide a return on taxpayer dollars? We
can identify several possible reasons, all of which probably contribute.
First, the stadium almost always costs more than initially estimated, and
the government is usually expected to come up with the extra money. Those
pushing the deal have an incentive to underestimate the cost to get it
approved, and for some reason, city after city falls for this trick. As a result,
the city is implicitly in debt even before the venue opens its doors. But this
phenomenon may be slowly going out of style. As noted, Sacramento drew
the line here and ensured that the team covered the cost overruns. Likewise,
when voters in the city of Arlington, Texas, approved $500 million of
public funding for a baseball stadium for the Texas Rangers, they did so on
the condition that the Rangers would be responsible for all cost increases. If
cities continue to fund sports venues, let’s hope they will at least follow
Sacramento and Arlington and not write a blank check.12
 Second, the jobs that a team supposedly adds to the community are
minimum-wage and part-time jobs. A basketball team like the Kings plays
forty-one home games a year, not counting playoff games, for perhaps three
hours a night. Even adding concerts and other events at the venue, a
concessions or parking lot employee will only get a few hundred work
hours per year—not exactly the kind of job that’s likely to keep someone
gainfully employed for long periods of time. And those job gains are likely
offset by fewer people needed (or lower tip earnings) at entertainment
venues a few miles away that are less crowded each night the arena is in
use.
Third, the supposed ripple effect in the surrounding community caused
by the stadium’s presence simply doesn’t exist. Businesses within a block
or two of the stadium inarguably see a sales boost from incoming and
outgoing fans. But no study has ever shown that a new stadium pumps
additional money into the city as a whole, and several have shown
conclusively that it does not.13
Finally, cities have no negotiating power because they are on the wrong
side of a supply-and-demand imbalance. There is only one Olympic
Committee; there are only two Olympic Games every four years. Similarly,
there are only about thirty teams in each major sports league. In the game of
musical chairs, mega-events and teams are the chairs, while the cities are
the kids rushing not to get left out.
For a team, the threat of moving to a more welcoming city is a valuable
trump card. A natural consequence of the relative scarcity of teams is that
several large cities don’t have one. Four top-20 metropolitan areas (San
Diego, Seattle, Saint Louis, and Baltimore) lack a basketball team, and five
top-30 metropolitan areas (Charlotte, Portland, San Antonio, Orlando, and
Sacramento) lack a baseball team. Meanwhile, only twelve metropolitan
areas host teams in all four major sports, meaning that many of the
country’s largest cities are a constant threat to poach a team from an
underperforming smaller city.
Because so many cities want a team, relocation becomes an easy
bargaining chip anytime an existing team wants a new stadium. The Kings,
for instance, had all but packed their bags for Seattle before a new
Sacramento ownership coalition came together to promise a state-of-the-art
new venue. Several well-known relocations have been driven by a hunt for
stadium deals. Back in the 1950s, the desire for a new stadium—more
accurately, a difference of opinion over where to build it—was at least part
of the reason that the Dodgers moved from Brooklyn to Los Angeles. In
1997, the lure of new facilities led the Houston Oilers to become the
Tennessee Titans and the Hartford Whalers to become the Carolina
Hurricanes.14 Few teams actually move for a new facility, however, because
the incumbent city usually ponies up when a team threatens to leave.
If a city is being realistic, building an arena can never be justified on
solely financial grounds. Economists Roger Noll and Andrew Zimbalist
once suggested that everybody would be better off if the government took a
fraction of the money it might spend on a stadium and just gave it directly
to the team as a bribe to not leave town.15 But does another valid reason
exist to back a stadium, an Olympic bid, or a World Cup? After all, the
games are a lot of fun.

Can You Really Put a Price on a Day at the Park?

Although publicly funded sports facilities invariably prove to be awful
investments, an economist can still rationalize the cost if a stadium provides
the public with enough nonfinancial benefits. Happiness, or utility, carries
legitimate economic value that you can’t always put a price on. If you buy a
$1,500 ticket to sit in the front row above the Yankees dugout for a playoff
game, it’s not because you expect to catch a foul ball and toss it to Aaron
Judge for a $2,000 autograph. It’s because the enjoyment and memories
you’ll get from attending the game are worth $1,500 to you. Similarly,
maybe cities know full well that they aren’t going to recoup the cost of a
$500 million stadium, but they think that the happiness their citizens derive
from having the team in town is worth $500 million.
If this were true, it would provide an elegant solution to the riddle of
publicly financed stadium deals; maybe citizens know that the financial
contribution of stadiums and events is a fiction, but they feel the
nonfinancial benefits outweigh the costs.
Maybe, but probably not. A group of economists did a study of the 2012
London Games using surveys that measured the happiness of people in
London (relative to other cities) around the time of the Olympics. The
economists found that people in London did feel happier during the
beginning and end of the Olympics. But their calculations suggested that
the happiness the Games created, if translated into dollars using economic
techniques for determining people’s willingness to pay to be made happier,
was not enough to make up for the cost of the Games unless the benefits
extended to much of the United Kingdom. They also found that the
happiness people felt was limited to the time of the Games themselves and
did not last through a follow-up survey a year later.16
Another reason to think the warm glow of stadiums and Olympics does
not outweigh the financial costs is the growing public expression of
disapproval. As mentioned earlier, negative public opinion led Oslo and
other cities to withdraw their bids for the 2022 Winter Olympics. Recent
polls and an accumulating pile of “no” outcomes on voter referenda further
suggest opposition to public expenditures on sports investments. A
nationwide poll of Americans in 2014 showed that 69 percent opposed
public financing of NFL stadiums. This trend has been reflected in recent
votes. The last three new venues for MLB, NBA, NFL, or NHL teams to
receive voter approval for public financing were all in the sports-crazy
Dallas metropolitan area: the aforementioned Arlington stadium vote in
2016, the Cowboys’ stadium in Arlington in 2004, and the NHL Dallas
Stars arena in 1998.17 Outside of Texas, voters simply do not want to spend
their tax dollars on expensive new venues for professional sports teams.
In truth, however, public approval is not necessary for arena-hungry
investors. Case in point: the Charlotte Hornets’ arena deal was rejected by
57 to 43 percent in a public referendum, but city council members
nevertheless pushed through a different plan, estimated to cost the public
more than $250 million.
A classic example of well-connected people going behind the backs of
taxpayers to build a publicly financed stadium is SunTrust Park, home of
the Atlanta Braves baseball team since 2017. John Schuerholz, president of
the Braves, said that the plan succeeded only because “it didn’t leak out. If
this deal had leaked out, it would not have gotten done.” He made this
statement after the public announcement of the deal prompted groups as
diverse as the Atlanta Tea Party and the local chapter of the Sierra Club to
hold an unsuccessful joint demonstration against it.18
Another noneconomic idea that might contribute to politicians pushing
for sports facilities and events is unrealistic optimism. Did the Braves’
management team and the Cobb County commissioners (or, for that matter,
the people behind the London Olympic bid, the Sacramento arena, and
other previous publicly funded sports investments) know they were trying
to pull the wool over citizens’ eyes? Or had they talked themselves into
believing that they were doing the best thing for their communities and that
they had to keep it quiet to avoid letting shortsighted activists defeat a great
idea? Nobody will ever know, but it also doesn’t matter; either way,
selfishness and self-deception are equally wasteful.

Fat-Cat Local Politicians

The idea that citizens might derive enough joy from a stadium or event
to justify it seems unlikely at best. Why, then, do cities keep bidding for the
Olympics and offering to build mind-numbingly expensive new sports
venues? One simple answer is that, like most other public policy decisions,
the choice of whether or not to fund the stadium isn’t made by citizens. It’s
made by politicians.
The disparity between politicians’ and voters’ attitudes toward stadium
deals underscores a larger economic issue known as “agency problems.” An
agency problem arises when a person or group of people is given the power
to decide how to use resources that are not their own, and whose use affects
some other group’s well-being. A good example is a CEO and the board of
directors whose primary responsibility is to increase their company’s value
for shareholders. Despite that mandate, CEOs engage in pet projects and
negotiate lavish pay packages all the time. Just as CEOs get board members
to go along with their schemes at shareholder expense, team owners get
politicians to pony up the taxpayers’ money.
Representative democracies create agency problems at every level.
Spending taxpayers’ money is, of course, necessary: society would struggle
to form fire departments, distribute welfare, or preserve public land if
elected officials did not govern the administration of those processes. But it
is impossible to perfectly align officeholders’ incentives with those of their
constituents.
Although voters almost never think that using tax money to fund a
team’s new digs is a good idea, their elected representatives often feel
differently. If a politician thinks he will personally derive enough utility
from constructing the new stadium to make up for whatever small minority
of voters will base their voting decision on their dislike for the stadium
deal, he will do it.
 The personal utility of a new stadium to a politician could come from a
couple of different sources. For example, voters who did not approve of
increasing the hospital tax may nonetheless like their mayor better after
seeing him break ground for their favorite team’s fancy new arena. Perhaps
rabid fans base their voting decisions on these sorts of events, whereas most
voters, though they don’t want a new stadium, don’t care enough to let such
issues be a driving factor in their voting. Or perhaps the politician is a fan
of the team. Imagine you were elected mayor of the city in which you grew
up, and your favorite team started griping about its antiquated facility and
wondering in public whether Las Vegas might be willing to build the team a
fancier one. Might you round up a few hundred million of your
constituents’ tax dollars and offer to build one right there in town?
Taxpayers in Hamilton County, Ohio, home to the city of Cincinnati,
were whipped by one particularly cruel display of agency problems. In
1995, the NFL’s Cincinnati Bengals started making it known that if they did
not get a new, publicly funded stadium, they would buy a one-way ticket to
Baltimore. At the same time, baseball’s Cincinnati Reds also decided to
demand a new home. For reasons related to “downtown development” and
other familiar catchphrases, local politicians began to negotiate with both
teams.
Representatives from the Reds and Bengals, the city of Cincinnati, and
Hamilton County eventually reached a deal to finance the two stadiums for
a combined cost of $500 million by increasing the county sales tax by 0.5
percent. So far, not too bad: the tax increase was small, the teams agreed to
stay in town, and a lot of people came away happy. Very quickly, however,
it became clear that neither stadium was going to come in at its expected
cost. The completed Bengals field eventually cost nearly $500 million by
itself, and with the Reds’ field over budget as well, the final price tag for
the two venues was $792 million. Here’s how the New York Times described
the terms of the county’s deal with the Bengals:
The team had to pay rent only through 2009 on its 26-year lease, and has to cover the cost of
running the stadium only for game days. Starting in 2017, the county will reimburse the team
for these costs, too. The county will pay $8.5 million this year to keep the stadium going. The
Bengals keep revenue from naming rights, advertising, tickets, suites and most parking. If the
county wants to recoup money by taxing tickets, concessions or parking, it needs the team’s
approval.19
 Now that the deal has tanked so badly, you might imagine that the people
responsible for it are long gone from local government. And they are, in a
way; Hamilton county commissioner Bob Bedinghaus, the most
enthusiastic member of the county board that approved the deal, was voted
out of office in 2000. But he didn’t spend much time looking for a job:
shortly after his failed reelection campaign, he was hired as director of
business development by none other than the Bengals. He worked in the
team’s front office for eighteen years.

Corruption
When the stakes are high enough, the misalignment between decision
makers with big power and the constituents they represent can make the
leap from “agency problems” to outright corruption.20 Many in the United
States like to think that corruption is something that happens in developing
countries, Russia, and maybe in the negotiation of a few New Jersey
sanitation contracts. But even Salt Lake City, with its squeaky-clean
reputation, is not immune.
In 1991 the city finished just four votes behind Nagano, Japan, in its bid
for the 1998 Winter Olympics. Officials supporting the Salt Lake bid were
convinced that the loss was a consequence of their failure to court the
representatives of developing countries, an emerging voting bloc under the
International Olympic Committee’s globalization efforts. Determined to
land the 2002 Games, the Salt Lake City bid’s backers lavished African and
Latin American IOC officials with all-expenses-paid trips to Utah, tuition
help and internships in the United States for their children, and campaign
donations. When the committee convened in 1995 to select a host, Salt
Lake won a decisive fifty-four-vote majority over the other four candidates
on the first ballot.
The scandal that ensued as details emerged of Salt Lake’s voter courtship
efforts underscored a predictable problem with many bidding processes: the
ultimate decision was made by a group of people who didn’t have much
interest in the outcome. In the case of the Olympics, host cities are elected
by the general body of the IOC, which is composed of one hundred or more
members from a wide range of countries around the world.
In theory, the IOC’s members should all share an interest in maximizing
the quality of the Olympics as a whole, but in practice, many of them are
probably more concerned with their country’s or their own welfare. After
all, why would a representative from Samoa or Sudan care about whether
the Winter Olympics are held in Utah or Sweden? No one from his country
is going to participate, and it’s unlikely that either site will do a vastly better
job of hosting than the other.
The main thing people on the Salt Lake bid committee were guilty of—
besides maybe a federal crime or two—was participating in a system of
broken incentives. Only two ways exist to win an Olympic bid: either you
can put together a great, detailed proposal that makes it clear your city is the
best choice for a host, or you can scratch the backs of the people who make
the decision. The former approach is more noble, but it’s really just a
crapshoot if you don’t do some of the latter as well. If you think there was a
big difference between Salt Lake City’s winning bid and the bids of its top
rivals Östersund, Sweden, and Sion, Switzerland, think again. Both cities
put together strong bids, and then politics and luck sorted the rest out.
After the Salt Lake scandal, the International Olympic Committee
changed the rules to try to discourage corruption. For example, site
selectors may no longer fly to potential locations to be courted by an
aspiring city. One theory about why Beijing was given the 2022 Winter
Games instead of Almaty, Kazakhstan, is that IOC voters could not travel to
Kazakhstan and therefore had only uninformed and vague images of the
country. This theory suggests that the change in IOC rules may lessen the
opportunities for corruption, but at the cost of leading to overly
conservative host city selections.
On the other hand, it would not be unreasonable to worry, based on
Nursultan Nazarbayev’s reputation, about the persuasive techniques he
might have used had the committee visited. If the IOC had not changed its
rules, perhaps some Kazakh oil money would have found its way into the
bank accounts of IOC officials and selectors. In any case, Nazarbayev did
not get his Olympics, which means you will have to get used to artificial
snow if you go to the Beijing Games in 2022, and you will have to find
some other reason to visit Kazakhstan.
Getting cities to bid for the Olympics or to build a stadium continues to
get harder as voters have woken up to the costs. Voters continue to stifle
attempts to build large new stadiums. For the 2026 Winter Olympics,
despite much early interest, voters in Calgary (Canada), Graz (Austria), and
Sion (Switzerland) all squashed bids led by local politicians looking to
make a name for themselves, and other cities also withdrew under public
pressure. The Covid-19 pandemic will surely make overcoming citizens’
objections even more difficult, given the massive losses Tokyo faced when
forced to delay the 2020 Olympics by a year and to hold the Games with no
international spectators, and the huge losses of money to cities as stadiums
sat idle for more than a year during the pandemic (leading Sacramento to
dip into its rainy-day fund to pay off bonds on the Golden 1 Center).
Meanwhile, Nursultan Nazarbayev is enjoying the quiet life of retirement,
and his country’s citizens are lucky he didn’t leave behind a large Olympic
bill.

OceanofPDF.com
 9
Who Wins When People Gamble?

The Cleveland Browns, long one of the doormats of the National

Football League, got off to an encouraging start in the 2020 season, having
won four of their first six games. But things were not looking good for the
Browns in their seventh game when, with a little over a minute left to play,
the promising Cincinnati Bengals rookie Joe Burrow threw a touchdown
pass to give his team a 37–34 lead. Yet the Browns’ upstart quarterback
Baker Mayfield used that final minute to complete four straight passes and
drive the Browns seventy-five yards, culminating with Mayfield’s fifth
touchdown pass of the day. With eleven seconds to go, the Browns led by
three points, and the victory was almost iced. Cody Parkey, the Browns’
placekicker, came on and missed the extra point. The missed kick had no
bearing on the game. The Browns won, and Cleveland fans celebrated.
But some people, far from Paul Brown Stadium in Cincinnati, ignored
the Browns’ victory and instead focused on Parkey’s seemingly
meaningless missed kick. The Browns had been favored to win by three and
a half points, and the missed extra point attempt prevented them from
covering the spread. Some of the most vocal reactions to the Browns’ win
on Twitter came from people who had bet on the Browns and lost because
of Parkey’s miss. One gambler tweeted, “Cody Parkey . . . thanks man, that
missed extra point only cost me a couple thousand. No problem.”1 Of
course, lots of people who had bet on the Bengals were very happy (though
few of them took the trouble to thank Parkey for their winnings).
If Parkey upset many bettors, he was a godsend to Las Vegas
sportsbooks, for whom that day in October was overall the worst of the
season. Gamblers tend to pick favorites at higher rates than underdogs, and
as a result, a disproportionate share of money was bet on the Browns,
whose failure to cover the spread meant a loss for gamblers as a whole but a
win for Las Vegas sportsbooks. The sportsbooks don’t care much about any
given game: they expect to lose money on some games every Sunday but
win on enough others to make money anyway.
This particular Sunday, however, was different. Popular favorites Tampa
Bay (with their self-proclaimed “golden boy” quarterback Tom Brady),
defending Super Bowl champion Kansas City, and the Los Angeles
Chargers all won that day. Gamblers who submitted a sheet filled with
nothing but the day’s strongest favorites won big; those who did so as part
of a parlay—a gambling strategy in which each bet’s winnings are levied
into a bigger bet for the next game—made a fortune. According to one
bookie, “We lost four of our top five decisions. That pretty much decided
the day for us.” He reported a six-figure loss for his casino. In one day,
businesses took big losses, while a few lucky guys sitting on couches or in
bars had the most lucrative day of their lives by doing very nearly nothing.2
If you think about it, it’s odd that one split-second bad decision or one
amazing stroke of luck by a man on a football field or basketball court can
deeply affect the finances of a completely unrelated person thousands of
miles away. But that’s the nature of sports gambling. The only somewhat
comparable industry is the stock market. But at least the stock market
arguably performs a valuable economic function by helping to finance
commerce, and in any case, few people who don’t work in finance spend
hours each week watching stocks move up and down. Sports gamblers,
however, do exactly that—and if they can’t do it in Nevada or the growing
list of municipalities where gambling on sports is legal, they do it illegally
online. All of this gambling is driven by economic forces, has a major
economic impact, raises ethical issues, and allows some well-informed
groups to take advantage of (or, if you look at it more optimistically,
entertain and excite) naive thrill seekers. There’s a lot at stake. So should
people be allowed to gamble legally on sports? The answer is a clear yes.
Or no. But one thing is for sure: you should only gamble if you can afford
to lose, because the odds are stacked against you.

Why Do People Gamble?

Just about every casual gambler falls into one of two categories: rational
gamblers, whose inevitable losses are compensated for by the thrill of
betting; and problem gamblers, whose ever escalating losses pile up as they
feed an unhealthy addiction. The big difference between the two is the level
of regret: rational gamblers lament their losses but accept that they are part
of the overall experience, while irrational gamblers are inevitably surprised
that they have lost and regret the choices their earlier selves have imposed
on their later selves.
How can gambling be rational? The simple answer is “utility.” If
someone has a lot more fun watching a game after putting ten dollars down
on one of the teams, the bet might be worth it even if he loses. Many people
would pay ten dollars to enjoy a movie, so why shouldn’t those people
spend ten dollars to increase their enjoyment of a basketball game they
otherwise wouldn’t care about?
Rational-pleasure gambling accounts for the bulk of money that flows to
bookies during the Super Bowl. The Super Bowl is an event that most
Americans feel almost compelled to watch, but it can be hard to care about
the actual game. Relatively few Americans feel strongly about, for instance,
the San Francisco 49ers or the Kansas City Chiefs, the contestants in Super
Bowl LIV.3 A random fan in Oklahoma, watching at home with friends and
family, might have a lot more fun if he bets a couple of bucks on the 49ers
or on either team scoring a touchdown in the first five minutes of the third
quarter. For that fan and millions of others who make similar bets, that
couple of bucks generates enough added fun, win or lose, to be worth the
wager. This phenomenon leads to the large array of available Super Bowl
prop bets, ranging from which team will win the coin flip to the color of
Gatorade poured on the winning coach.
Casual bets between friends can be fun and on balance don’t cost people
anything. One friend wins and the other loses. But as soon as a bookie gets
involved, be it a guy at the office or a casino with a legal sportsbook, the
expected return becomes negative. The primary reason is that on every bet,
the bookie keeps a “vig” and gives you less than a full return if you win. If
you pick the Green Bay Packers to beat the spread on a given Sunday, for
example, you have to accept that you lose $11 when the bet fails but win
only $10 when the Packers cover. Bookies stay in business by charging a
fee. If a bookie has equal numbers of bets on both sides of a game, the
bookie reaps 10 percent of the amount won by people who win the bet.
That 10 percent fee is money well spent if it makes the average fan a
little bit happier, and in many situations, betting works that way. Wagers on
landmark events like the Super Bowl and NCAA Basketball Tournament
are social opportunities that enhance the experience of watching the event.
In addition, lots of people enjoy investing in teams or people they support.
Taking this pursuit to an extreme, the golfer Rory McIlroy’s father
famously scored £100,000 after his £200 bet on his son to win the British
Open within ten years, placed when Rory was fifteen, came out a winner.4
McIlroy’s father may have believed he had inside information about his
son’s abilities that actually made his bet rational, but he just as likely
thought that betting on the family’s fairy-tale scenario would be a fun,
slightly wasteful, but certainly affordable way to spend a few hundred
pounds.

The Bookies Bet, Too

In some betting markets, the fee is all the bookmaker takes in—though
in those cases, it is usually more than 10 percent. In this system, known as
“pari-mutuel betting,” the odds are determined mathematically by the ratio
of bets. Most horse racing in the United States, as well as the few remaining
greyhound racing and jai alai facilities, use pari-mutuel betting. For any
given type of bet (for example, win, place, show, or a trifecta in horse
racing), all the money in that type of bet is pooled, and after the house takes
its cut, the pool of money is divided among those who bet correctly in
proportion to their bet. In pari-mutuel betting, the person laying down a bet
does not know what odds he or she will get until the event actually starts,
because additional bets affect the size of the pool. Each time someone bets
on a horse other than the one you picked, you get a higher payoff if your
horse wins. But each time someone bets on your horse, you stand to win
less. The house takes no risk, and its cut can be quite high. Churchill
Downs, the home of the Kentucky Derby, takes about 20 percent off the top
of each betting pool, as do horse tracks in New York State.5
You might think you are getting a better deal from the sportsbook in a
casino because the casino only takes 10 percent off the top. But unlike the
racetracks, casinos and other bookmakers don’t simply balance their bets so
that they pay out to winners exactly what they take in from losers (minus
their cut). They set up the betting markets so that they are, in effect, making
bets of their own.
That tactic wouldn’t necessarily hurt you if the house were just one more
bettor in the market. But the house has way better information than you
have, and as in the stock market, better information is worth a fortune in
betting markets. You may have worked out an intricate system for betting
the NFL games this weekend or think you know when the Reds’ starting
pitcher looks like he has a gimpy arm, but the house has reams of data on
who bets what and how well it pays off. The house knows what people tend
to overvalue, such as favorites and big-city teams. Sportsbooks do
sometimes lose if a few things line up wrong, as they did on the day of the
Browns’ missed extra point. But over the long run, bookmakers take the
right side of a majority of bets.
Until recently, betting on major professional sports was a bit like
drinking during Prohibition: it was technically illegal, but people found a
way. Local bookies were often shady guys running solo gambling
operations. If a bookie didn’t pick up the phone, you were more or less out
of luck (though actually better off, since you were likely to lose). But today
the sports gambling business is big, getting bigger, and highly sophisticated.
The business has exploded for two reasons—one somewhat dicey from a
legal perspective, and the other fully legal (though questionable as public
policy). First, like the secondary ticket market, gambling has been
revolutionized by the growth of the internet. Online international
sportsbooks, like Bovada, Sportsbetting.ag, and 5Dimes, dominate the
industry. These sites are able to dodge U.S. restrictions on sports betting by
establishing official headquarters offshore (the Caribbean and Central
America are popular locations). They can also instill confidence in their
legitimacy that the bookie on the phone could never approach. Make a few
clicks, enter some credit card information (or, more likely these days,
deposit some cryptocurrency), and you’re signed up to bet as much as you
please, knowing that the site will cash your winnings out at any time you
choose. The problem is, if you play longer than a few lucky days, you’ll
probably have no winnings to cash out.
The second reason sports gambling is growing quickly is that, for
municipalities, it is an easy source of revenue. Over the past few decades,
state budgets have come to rely more and more on legalized gambling,
often in the form of slot machines and blackjack tables in new casinos in
cities including Detroit and New Orleans, on riverboats throughout the
country, and in lotteries. But the Bradley Act of 1992, named for its chief
proponent in the U.S. Senate, New Jersey senator and former NBA star Bill
Bradley, prevented any significant expansion of legal sports betting for
many years, limiting it to Nevada casinos—and state-sponsored lotteries in
Oregon, Delaware, and Montana, where lottery winners are tied to sports
game outcomes. But once the Bradley Act was declared unconstitutional by
the United States Supreme Court in 2018, it immediately led to an
expansion of sports betting at New Jersey and Delaware casinos and, in the
2020 elections, successful referenda to legalize sports betting in Maryland,
South Dakota, and Louisiana.
Las Vegas casinos and online sportsbooks offer many forms of betting.
For example, matches between two teams can be bet in terms of money
lines or point lines. If you want to bet, for instance, on a Seahawks-Vikings
game in which the Seahawks are slight favorites, you can either put money
on the Seahawks at a –130 money line, which means betting purely on a
Seahawks win but at a rate that requires a $1.30 bet for a $1.00 payout, or
you can bet on the Seahawks at –2.5, which means that your bet wins if the
Seahawks win by more than 2.5 points. If you can’t decide which team you
like, you can bet the “over” (or the “under”), where the bet pays off if the
total points scored by the two teams exceed (or fall short of) some number.
Bookies, like stockbrokers, just make markets in bets. The “prices”
(known in this context as “lines”) are flexible to reflect bettors’ demand and
the possibility of receiving new information. If bookies begin taking bets on
the Seahawks-Vikings game with the Seahawks as a 2.5-point favorite, and
customers are putting $5 on the Vikings for every $1 on the Seahawks, the
line will move to correct that disparity; the Seahawks might drop to 1.5-
point favorites. Unlike pari-mutuel bettors, however, people who placed
their bet when the line was 2.5 points are paid off based on that
predetermined line (rather than the 1.5-point line at game time). The line
might also move if the bookie receives word that the Seahawks’ starting
quarterback Russell Wilson will sit out the game, information the bookie is
most likely to receive just before most bettors. The adjustments continue
until the bookie feels confident about the risk he is assuming on each side.
So why does the bookmaker take a position rather than just balance the
bets and keep the 10 percent guaranteed take? Taking all those bets gives
bookmakers an information advantage that they can exploit, just like a stock
trader with a direct line to a firm’s CEO. The casino gets a point spread
estimate from its mathematical model, which takes into account past
behavior and can generally anticipate which team will be more popular with
bettors, then manually adjusts it toward the team that tends to draw more
money. The guys that sportsbooks pay to set the lines are some of the most
informed and advanced gamblers in the world. Few people can consistently
think a step ahead of the bookies (and the vast majority of those who claim
they can are lying to you or to themselves). In other words, all but the best
gamblers will lose in the long run.

There Are Smarter People Than You

So far, we’ve discussed two reasons you will lose when you gamble: the
house takes a cut, and the house doesn’t set fair odds, because it has better
information than you have. Now let’s add a third factor. Not only does the
house know more than you do, but so do the top gamblers. There are not
many of them, but they move a lot of the money in betting markets.
One important bettor, a Berkeley dropout named Bob Stoll—known
professionally as Dr. Bob—had a huge impact on the gambling industry.
According to the Wall Street Journal, sportsbooks were thrown into a panic
when Dr. Bob’s clients, who paid for an email subscription service that sent
them specially recommended weekly football gambling picks, bet millions
of dollars a week on a few obscure college games.6 In his best season, Dr.
Bob’s NCAA football picks hit at an unbelievable 71 percent; even people
who make their living gambling on sports rarely pick at better than 55 or 56
percent. In eight college football seasons from 1999 through 2006, Dr. Bob
picked at over 60 percent in four of them, and in only one of those seasons
were his picks unprofitable.7
Sportsbooks hated Dr. Bob. He was cleaning up at their expense because
he had better information than they had. But other bettors (other than those
who subscribed to his service) should also have hated Dr. Bob. Dr. Bob was
in the same market as casual bettors and taking their money, with the casino
acting as middleman. Dr. Bob was the Billy Beane (as in the book and
movie Moneyball) of NCAA betting. He had a better analysis of the
situation than his competitors, and he was able to take advantage of it at the
expense of casinos and other bettors.
Just as other teams caught up to Beane’s Oakland A’s and now have just
as much analytical power, so the casinos caught up to Dr. Bob. They hired
more analysts and used more computing power to get their line-setting
systems up to his standards. Since the end of the 2006 season, Dr. Bob has
had decent success picking college football games, but his results no longer
blow competitors out of the water. In his eight standout years, he amassed
212.3 “net stars” (a unit he developed to rate himself); in the subsequent
seven years, from 2007 to 2013, he added just 13.8 total net stars. That’s a
drop from 26.54 net stars per season in the first period to 1.97 in the second.
His fifth-best season between 1999 and 2006 was more profitable than any
year he has had since.
Whatever advanced model that made Dr. Bob so successful was
eventually replicated by the bookies’ mathematicians. The linemakers were
able to build Dr. Bob’s clientele into their predictions of what bets would be
placed. Dr. Bob was unusually forward about his methods in write-ups on
his site (though maybe not as forward as Beane, who gave away his secrets
in a book and a movie). Dr. Bob lost the advantage that came with having
better information, while the casinos became ever better informed and
upped their advantage over regular gamblers.
Well-informed bettors in general (and Dr. Bob in particular) are
becoming less of an issue as betting markets get bigger and more
sophisticated. So much money has flooded these markets that, like stock
markets, they have attracted highly sophisticated people using analytics to
seek out even a small opportunity that they can exploit. The bookmakers, in
turn, have invested more to stay right behind the smart money. While
“touts” like Dr. Bob get hot from time to time, beating the sportsbook as
badly or for as long as he did is extremely rare. “Years ago, the technical
stuff really worked,” Dr. Bob says. “The lines have gotten so much sharper
now.” But don’t worry about Dr. Bob; his real money comes from
subscriptions to his newsletters and his website where he reveals his
predictions.

If All Else Fails, Cheat

Dr. Bob and the bookmakers make it hard for you to make money
because they know more and, in the case of the bookmakers, take some
money off the top. But there’s yet another group of people out there who are
literally stealing your money. It’s not news to anyone who has seen Field of
Dreams or On the Waterfront that sports betting markets can be
manipulated. When that happens, the loser isn’t the bookmaker, if it
managed the line properly; it’s the uninformed bettors who did not know
what was going on—that is, you and me.
The gentlemanly country-club sport of tennis has been a center of match
fixing in recent years. Almost every day, professional matches all over the
world pit good but unknown players with world rankings in the hundreds or
thousands against one another. Betting sites will take bets on these matches,
and sometimes the bets are more specific than simply who will win the
match. For example, in 2016, the Australian tennis player Oliver Anderson
reached a world ranking of 639, which is the best he has ever done. That
October, he played his fellow Australian Harrison Lombe, who was ranked
1,634th in the world and was not expected to give Anderson much of a
match. Before the match, CrownBet—an Australian betting site—was
offering to take bets that Lombe would win the first set. However, the site
refused to take a $10,000 (Australian dollars) bet because it was suspicious.
It did accept a bet of $2,000.8
Anderson, who was the reigning Australian Open Boys’ champion at the
time, lost the first set 6–4 before wiping out Lombe 6–0, 6–2, and winning
the match. He later admitted to accepting money to throw the first set.
This minor scandal raises two questions. The first one—the easy one—is
why Anderson would accept this payment. His incentives were pretty
strong. At the time he was suspended for match fixing, he had amassed
career earnings of about $20,000. That is not nearly enough to cover the
expenses of being a touring professional, much less to live on. It certainly
cannot support the luxurious lifestyle of Roger Federer, but it is typical for a
tennis pro who is well outside the world top 100. A few hundred dollars
here and a thousand dollars there to drop a set (or even a match) from time
to time can be a strong temptation when that money means a decent hotel
room or a good meal or even just staying afloat on tour while hoping to
break through.
The harder question to answer is why in the world does CrownBet (the
Australian betting site) take bets on who will win the first set of some
match that nobody beyond the players and their mothers cares about? Who
would bet on those matches if they were not fixed? Nobody cared whether
Anderson or Lombe won. The match had all the hallmarks of a bad market
for regular bettors (that is, those who are not cheaters with inside
information), given that there was not much money in the betting pool, the
players were vulnerable to financial inducements to throw the match, and it
is pretty easy to drop a set on purpose. The Wimbledon finals (or pretty
much any match in a tournament offering serious prize money) are a lot less
likely to put uninformed bettors at a big disadvantage, because the players
have too much to lose in prize and endorsement money to risk being caught
throwing a match. CrownBet must set the odds with a large vig (that is, so
that its expected take is very high), because otherwise the company could
never make a profit setting a market on matches so prone to fixing.
A much larger betting market than the lowest reaches of professional
tennis is college basketball. People bet large sums in Vegas casinos and
other legal venues, as well as illegally with bookies, on NCAA games. For
both coaches and players, winning games has a major effect on their
futures, so the costs of throwing a game are extremely high. But maybe a
smart player can figure out how to win the game and win a bet at the same
time.
College basketball might be an easy sport to fix: the players are young,
impressionable, and often from poor backgrounds; a place on the team
might be the most high-leverage position some players ever hold; and it
only takes paying off one or two players to have substantial control over
outcomes. Also, like tennis players who throw a set but win the match,
basketball players on heavily favored teams can lose against the spread
while winning the game—a practice known as “point shaving.” Imagine,
for example, that Michigan is favored by thirty-three points against
Binghamton (an actual recent spread). Michigan can coast to a twenty-point
win while those who know the game is fixed can bet that Michigan will not
cover the spread.
The economist Justin Wolfers analyzed the outcomes of more than forty-
four thousand NCAA games and concluded that point shaving is fairly
common. Although subsequent studies have cast doubt on some details of
his findings, college basketball has a long history of fixed games as young,
unpaid athletes sought to cash in on what might be their best moneymaking
opportunities. For example, in 1997, Arizona State guard Steven Smith
employed the “win but don’t cover” strategy quite successfully for a few
games before ultimately ending up in prison.9
The bottom line is that, in many sporting events, someone actually in the
game has better information than you do. In at least a few games, that
person may be using it to bet on himself (as Pete Rose famously did) or to
fix the outcome. But either way, “smart money” (that is, the people who
know how a game is fixed) will win out against regular bettors, so you
might be out some money before the game you bet on even starts.

Should Sports Betting Be Legal?

 If your goal is to make money, betting on sports is a fool’s errand. You
might have a good day from time to time. But in the end, you will lose. The
bookmakers take a cut, other people in the betting market have better
information than you do, and some of them are cheating. The financial
return on betting is so lousy that some people might conclude that all sports
betting should be illegal because it’s just a way for a few businesses to take
money from naive fans.
But let’s slow down and take the arguments for a ban on sports betting
one at a time. One argument is that gambling lowers productivity. People
spend time at work gambling, with the NCAA’s March Madness and Super
Bowl pools often cited as particular distractions. One survey showed that
March Madness bracket activities ranked third among “tech-related office
distractions” (behind Facebook and texting); the authors of another study
estimated that March Madness lowers U.S. productivity by $6.3 billion each
year. This argument does not get you very far, though, because other studies
suggest that March Madness activities increase office morale.10 Wasted
time is a real concern, but not enough to demand policy interventions.
Another argument is that gambling is unfair, for all the reasons I noted
earlier. More-informed bettors take money from less-informed bettors, both
legally and illegally. That is true in the stock market, as well, and nobody
(well, few people) would outlaw stock markets. Some stock investors have
better information because they do better research. Others have inside
information, though it’s technically illegal to use it. However, the stock
market, unlike sports betting markets, is an important financial tool that
allows companies to raise capital and creates incentives for executives to
manage companies well. Stock markets are unfair and cause problems—and
therefore require regulation (arguably more than they currently get)—but
they serve a purpose: they make the whole economy work better.
Sports gambling confers no such society-wide benefit. Or does it?
Remember that gambling brings people utility—they enjoy it. As long as
gamblers are reasonably informed about the disadvantages they face in the
betting market, there’s no reason not to let them spend their money on
gambling, just as society lets people spend their money on clothes,
furniture, and sports paraphernalia.
Indeed, if all gamblers gambled for fun and could afford the bets they
made, then the answer would be easy. Sports betting should be legal, and
the market would set the odds for each contest. But the best argument for
banning sports betting centers on a small set of the betting population:
gambling addicts.
Addiction to gambling destroys lives. There’s no way around it.
Statistics are hard to come by, but there are many cases of gambling-related
suicide.11 Most estimates put the number of people with pathological
gambling problems at 0.5 to 1 percent of the population in most Western
countries, though many of those people are addicted to nonsports gambling.
Almost all problem gamblers are men.
From a policy perspective, it should (in principle) be relatively easy to
decide whether to legalize sports betting. Society just needs to decide if the
dramatic costs to the problem-gambling population and their families
outweigh the benefits of gambling to the much larger responsible-gambling
public. While one could imagine banning sports betting on that basis, doing
so becomes less realistic when you consider that many problem gamblers
will still gamble if it is illegal. They might even be made worse off, because
they would now owe money to shady bookies instead of casinos, or they
might just sit at home and gamble illegally over the internet. Making
gambling easier to access increases the number of problem gamblers, but
not necessarily by a lot. Studies show that people who live near casinos are
more likely to be problem gamblers, for example, but the effect is not
dramatic, and it is possible that people who choose to live near casinos are
more prone to problem gambling to begin with.12
Perhaps the best argument against outlawing sports (and other) betting
comes from the historical example of Prohibition, which banned liquor
sales in the United States from 1920 to 1933. Even though it lowered the
total amount of liquor consumed, Prohibition also resulted in liquor
production and distribution being controlled by a costly and destabilizing
criminal element. Moreover, the people whose consumption was most
dramatically lowered were the casual and social drinkers for whom drinking
did not create significant problems. Addicts found a way to drink, often
getting sick on unregulated booze in the process. By the same logic, serious
gamblers are likely to gamble even if it is illegal. Banning gambling will
most affect casual gamblers for whom the benefits are relatively high (or for
whom the costs are low) while not necessarily stopping the problem
gamblers, who will still seek out bookies and other unsavory betting
mediums.
 In the end, therefore, I don’t think there is a credible case for outlawing
sports gambling. But can we make legal sports gambling less damaging? It
may not make a big difference, but it seems sensible to at least insist that all
gambling-related advertising be realistic and show winners and losers in
their proper proportions. Ads for casinos and racetracks (as well as
nonsports gambling ads for lotteries and the like) invariably show people
celebrating after winning some money. But most people lose. So we might
benefit from a law that says, if you want to show people celebrating their
winnings, you have to show even more people going home disappointed.
The country of Singapore, not known for its progressive social policies,
uses an innovative method to keep people from gambling if they cannot
afford it. Casinos in Singapore will happily let any adult foreigner enter. It
would be illogical to charge foreigners to gamble, given all the other
destinations they can choose to visit. But residents have to pay 150
Singapore dollars (about $110 U.S.) to enter a casino for a day, or 3,000
Singapore dollars for an annual pass.13 This system creates a nice source of
revenue for the government and dissuades people with less money from
gambling. Singapore undid some of this policy’s advantages, however, by
allowing residents to use two online gambling sites without a minimum
payment (which makes one wonder if Singapore is fine with letting poor
people gamble but just doesn’t want them doing it in sight of tourists).14
Still, charging people fixed fees to be allowed to gamble is a potentially
useful way to lower compulsive and spontaneous gambling—though it does
nothing to deter addictive gambling by people rich enough to pay the entry
fee.
One optimistic take on the rise of sports gambling is that it is a sign of
overall increases in wealth. Most sports bets are simply expenditures on
leisure activity, like buying a ticket to a movie or going to a restaurant.
These expenditures become possible as societies get richer. When most
people are worried about survival, they don’t spend money on gambling
and other leisure goods (though, again, it is important to keep in mind
problem gamblers). So even though purist sports fans may prefer games
without betting, sports bets are likely here to stay, and even though many
more socially useful ways exist for people to spend their money, there is
nothing wrong with people enjoying themselves.

OceanofPDF.com
 Epilogue

Sometimes the connection between sports and economics is simple,

direct, and obvious. But only after the fact.
Real Madrid has been one of the world’s most successful soccer teams
for more than a century, both financially and on the field. In 2000,
Florentino Perez took over as president of the team while continuing to run
Spain’s largest construction company.
Perez, a very wealthy man, saw a way to grow his soccer club to even
greater heights. In the United States, major sports leagues such as the NBA
and MLB have a constant and relatively small set of teams that play only
against one another. These top leagues get the vast majority of TV airtime
and fan attention, and they bring in revenue that dwarfs any other league in
their respective sports. European soccer, on the other hand, comprises
hundreds of teams in a loose collection of leagues, with talent spread out
more evenly across all of them. Perez led an effort to build a “Super
League” on the apparently sensible premise that if his league could
assemble the best teams from across Europe, it could sell broadcast rights
for huge amounts. By emulating an American sports league, the Super
League would get the lion’s share of fan attention, broadcasting time, and
revenue. As the richest soccer league, it would inevitably accumulate the
best players. Its dominance would be self-sustaining.
The business model was simple: bring together the best teams to play
each other regularly. The best players would face off more often, and there
would be fewer mismatches between top-level teams and local squads with
lower budgets. Fans would pay the most to see top teams playing each
other. And indeed, if the owners had been starting European professional
soccer from scratch, the Super League would probably have been a huge
success.
After years of planning, Real Madrid and eleven other top European
teams announced the formation of the Super League in the spring of 2021.
Perez and the league’s other backers knew contracts and business, but it
turns out they didn’t know economics very well. They ignored the
economic value of tradition and underestimated the potential backlash from
local fans, politicians, and others whose favorite teams, no longer playing
against the best competition, would be overshadowed by the Super League.
Many politicians condemned the Super League, including British prime
minister Boris Johnson. This outcome was entirely predictable. Although
the league would have been great for the players and owners of a few top
teams, far more would have been poorer, and most fans would have been
less happy. Johnson and his party count on votes from the whole
population.
Within two days of being announced with great fanfare, the Super
League collapsed. Perez and his colleagues had misread the market. The
market almost always wins.

OceanofPDF.com
 Notes

Prologue
1. Aaron Dodson, “On This Day in NBA Finals History: Steve Kerr’s 17-Foot Jumper Clinches
Bulls’ 1997 Title,” The Undefeated, June 13, 2017.
2. “CERA: The New EPO Discovered at the Tour de France,” Cycling Weekly, July 17, 2008;
“Rashid Ramzi Stripped of Beijing Olympic 1500m Gold After Failing Dope Test,” The Telegraph,
November 18, 2009.
3. “Lee6 Wins U.S. Women’s Open, Pockets $1M,” Associated Press, June 2, 2019.

Chapter 1. Should You Help Your Kid Become a Pro Athlete?

1. Bradley T. Ewing, “The Labor Market Effects of High School Sports Participation: Evidence
from Wage and Fringe Benefit Differentials,” Journal of Sports Economics 8 (2007): 255–265.
2. Michael Lechner, “Long-Run Labour Market and Health Effects of Individual Sports
Activities,” Journal of Health Economics 28 (2009): 839–854.
3. Michael B. Ransom and Tyler Ransom, “Do High School Sports Build or Reveal Character?
Bounding Causal Estimates of Sports Participation,” Economics of Education Review 64 (2018): 75–
89.
4. Natural Resources Defense Council, www.nrdc.org, December 31, 2015.
5. Edward B. Fiske, “Gaining Admission: Athletes Win Preference,” New York Times, January 7,
2001.
6. Thomas J. Espenshade, Chang Y. Chung, and Joan L. Walling, “Admission Preferences for
Minority Students, Athletes, and Legacies at Elite Universities,” Social Science Quarterly 85
(December 2004).
7. Full disclosure: I am a Middlebury graduate. But various college rankings have considered
Middlebury more prestigious than these other schools.
8. I am only considering legitimate athletic scholarships, not the type given by the Stanford sailing
coach and others exposed during the 2019 “Varsity Blues” scandal.
9. “Average per Athlete 2020,” www.scholarshipstats.com/average-per-athlete.html; “Athletic
Scholarships: Everything You Need to Know,” www.ncsasports.org/recruiting/how-to-get-
recruited/scholarship-facts; “Scholarships,” www.ncaa.org/student-athletes/future/scholarships;
Hanna Muniz, “How Many College Students Are in the U.S.?” www.bestcolleges.com/blog/how-
many-college-students-in-the-us/. One percent is based on 200,000 as a fraction of 20 million college
students. All sites accessed April 23, 2021.
10. Bryce Druzin, “The Cardinal Connection: Why Joining Stanford’s Football Team Is a Great
Career Move,” San Jose Business Journal, August 11, 2014; Paul Wachter, “Wall Street’s Lacrosse
Mafia,” Bloomberg Business, March 22, 2012.
11. David Card, “Using Geographic Variation in College Proximity to Estimate the Return to
Schooling,” Working Paper, National Bureau of Economic Research, 1993; Orley Ashenfelter and
Alan Krueger, “Estimates of the Economic Return to Schooling from a New Sample of Twins,”
American Economic Review 84 (1994): 1157–1173; Joshua D. Angrist and Alan B. Krueger, “Does
Compulsory School Attendance Affect Schooling and Earnings?” Quarterly Journal of Economics
106 (1991): 979–1014.
12. “Kevin Durant Biography Facts, Childhood and Personal Life,” storytell.com, accessed June
6, 2021; Sam Anderson, “Kevin Durant and (Possibly) the Best Basketball Team of All Time,” New
York Times Magazine, June 2, 2021.
13. It’s important to note that the statistical focus on Durant’s race is not ideal, as other (possibly
more important) factors are correlated with race and limit someone with Durant’s background. For
example, having a single mother and being poor are likely to be more important than race. I use race
because the data are reliable and compelling. But note that individual circumstances beyond race
should drive the calculation of whether or not to focus on sports at a young age.
14. African American wage and employment information from quarterly releases by the U.S.
Bureau of Labor Statistics; latest version available at www.bls.gov/news.release/pdf/wkyeng.pdf; Raj
Chetty, Nathaniel Hendren, Maggie R. Jones, and Sonya R. Porter, “Race and Economic Opportunity
in the United States: An Intergenerational Perspective,” Quarterly Journal of Economics 135 (2020):
711–1014.
15. David Wharton, “Sweet Youth,” Los Angeles Times, March 18, 2007.
16. Charles Nuamah, “Tall NBA Players Who Had Relatively Short Parents,” howtheyplay.com,
accessed June 6, 2021.
17. Calculations based on the height calculator at Tall.Life, https://tall.life/height-percentile-
calculator-age-country, accessed April 23, 2021.
18. Jonny Hughes, “Top 15 Little-Known Facts About Kevin Durant,” TheSportster, December
24, 2014. The first pick in that NBA draft, Greg Oden, was a very sad story. Because of recurring
injuries, Oden has never played in the NBA for any sustained period of time. But all the calculations
for Durant apply to Oden, who was also born in 1988 and made millions of dollars in guaranteed
NBA contracts despite being unable to play.
19. I look at income statistics for Black men born in 1988 using a representative sample from the
U.S. Census Bureau’s American Community Survey. This huge survey is widely used by social
scientists, as well as by federal and local government to allocate many billions of dollars of funds
each year.
20. Emmanuel Saez and Gabriel Zucman, “Wealth Inequality in the United States Since 1913:
Evidence from Capitalized Income Tax Data,” Quarterly Journal of Economics 131, no. 2 (2016).
21. Two other professions where men often earn large incomes at relatively young ages are acting
and music. I did not find anyone born in 1988 making an eight-figure salary in either of these fields
or any other field.
22. Chris Palmer, “From the Bottom to the Top: The Russell Westbrook Story,” Bleacher Report,
November 12, 2015.
23. Seth Stephens-Davidowitz, “In the N.B.A., Zip Code Matters,” New York Times, November 2,
2013.
24. Brook Larmer, “Golf in China Is Younger Than Tiger Woods, but Growing Up Fast,” New
York Times, July 11, 2013.

Chapter 2. What Do Silicon Valley and Czech Women’s Tennis

Have in Common?
1. Calculations based on medal counts and populations from www.medalspercapita.com/ as of
March 20, 2018.
2. The sports are Alpine (downhill) skiing (combined ranking for all events), archery, badminton,
bowling, golf, judo (the heaviest weight class for each gender), kayak (one kilometer), marathon,
modern pentathlon, Nordic (cross-country) skiing, short track speed skating (1,500 meter), skeleton,
swimming (400-meter individual medley), table tennis, tennis, and trampoline. Rankings were
collected in the latter part of 2015.
3. Matthew Futterman and Kevin Helliker, “The Mystery of Norway,” Wall Street Journal,
February 24, 2010.
4. “Joy of Sport—for All: Sport Policy Document, 2011–2015,” Norges Idrettsforbund, 2015,
www.sportanddev.org/sites/default/files/downloads/sportpolicydocument2011_2015_1.pdf.
5. Norwegian Olympic and Paralympic Committee and Confederation of Sports (NIF),
www.idrettsforbundet.no/english/.
6. Elsa Kristiansen and Barrie Houlihan, “Developing Young Athletes: The Role of Private Sport
Schools in the Norwegian Sport System,” International Review of the Sociology of Sport 52 (2017):
447–469.
7. Marathon rankings from www.all-athletics.com, which are no longer available there.
8. Unlike PAPIs, which are calculated for each country in a sport, Herfindahl indexes are
calculated to measure the concentration for the top 25 as a group.
9. Gregory Warner, “How One Kenyan Tribe Produces the World’s Best Runners,” National
Public Radio, November 1, 2013.
10. Jackie Dikos, “The Simple Staple,” Runner’s World, June 28, 2012; Warner, “How One
Kenyan Tribe”; Max Fisher, “Why Kenyans Make Such Great Runners: A Story of Genes and
Cultures,” The Atlantic, April 17, 2012; Brendan Koerner, “Why Are Kenyans Fast Runners?”
Slate.com, 2003.
11. Warner, “How One Kenyan Tribe.”
12. Adharanand Finn, “Kenya’s Marathon Men,” The Guardian, April 8, 2012.
13. “Czechoslovakia Strives to Maintain Tennis Tradition,” AP Story, December 21, 1985.
14. Sarah Pileggi, “Fanatics and Fools,” Sports Illustrated, January 12, 1981.
15. John Branch, “Czech Women Continue Wimbledon Onslaught,” New York Times, June 30,
2014.
16. “Czechoslovakia Strives to Maintain Tennis Tradition,” AP Story, December 21, 1985.
17. Pileggi, “Fanatics and Fools.”
18. Petra Kvitova’s bio, http://petrakvitova.net/petra/, accessed May 24, 2019.
19. “Petra Kvitova: Martina Navratilova,” Tennis.com, December 2, 2015, www.tennis.com/pro-
game/2015/12/petra-kvitova-martina-navratilova/56930.
20. This particular survey sampled eleven nations, including Hong Kong, Germany, Italy,
Belgium, and Portugal.
21. The OECD has thirty-four members, including virtually all of western Europe plus most of the
rest of the world’s developed countries.
22. The Global Gender Gap Report, World Economic Forum, 2020; “Iceland Leads the Way to
Women’s Equality in the Workplace,” The Economist, March 4, 2020.
23. OECD, “Saving Rate,” https://data.oecd.org/natincome/saving-rate.htm.
24. Pay gap from OECD, https://data.oecd.org/earnwage/gender-wage-gap.htm, accessed
September 19, 2021.
25. Official World Golf Ranking, www.owgr.com/ranking; Rolex Women’s World Golf Rankings,
www.rolexrankings.com/rankings.
26. Randall Mell, “Pak’s Influence on Game Immeasurable,” Golf Central blog,
www.golfchannel.com/news/golftalkcentral/paks-influence-game-immeasurable/.
27. Alan Shipnuck, “A Seminal Win at the 1998 U.S. Women’s Open Has Triggered Nearly Two
Decades of South Korean Dominance on the LPGA Tour,” NCGA Golf, Summer 2015.
 Chapter 3. Why Do Athletes Cheat and Lie?
1. McGwire has since admitted to regular steroid use throughout his career, and Sosa’s name was
leaked from an infamous confidential list of positive drug tests in the early 2000s.
2. Benjamin Soloway, “Scandal on the Tour de France,” Foreign Policy, July 3, 2015.
3. Of course, if the prisoners are friends and legitimately care about each other’s welfare, they
might help each other out. The athletes I focus on, however, are likely to be more like the self-
interested actors in economic models, given the importance of winning relative to any other outcome.
4. David Walsh, From Lance to Landis (Ballantine Books, 2007), 66.
5. Walsh, From Lance to Landis, 69.
6. United States Anti-doping Agency, “Report on Proceedings Under the World Anti-doping Code
and the USADA Protocol, United States Anti-doping Agency vs. Lance Armstrong,” August 24,
2012, www.usada.org/wp-content/uploads/ReasonedDecision.pdf.
7. Dick Marty, Peter Nicholson, and Ulrich Haas, “Cycling Independent Reform Commission,
Report to the President of the Union Cycliste Internationale,” 2015.
8. FanGraphs’ “wins above replacement” (WAR) measure, for instance, gives Bonds’s 1998
season an 8.5 rating, tied with McGwire and far exceeding Sosa’s 7.1. Baseball Reference WAR
gives Bonds an 8.1, McGwire a 7.5, and Sosa a 6.4.
9. Nate Silver, “Lies, Damned Lies: The Steroid Game,” Baseball Prospectus, May 7, 2009.
10. Suzanne Massie taught President Ronald Reagan this phrase, which he made famous. The
phrase had a rebirth in 2015 during negotiation of the Iran nuclear deal. See Barton Swaim, “‘Trust
but Verify’: An Untrustworthy Political Phrase,” Washington Post, March 11, 2016.
11. Jennifer Laaser and John Fauber, “Baseball’s Drug Testing: Thorough or Easily Thwarted?”
Milwaukee Journal Sentinel, July 14, 2013, http://archive.jsonline.com/news/health/up-to-20-major-
league-players-to-be-suspended-but-not-because-of-stringent-tests-b9939132z1-215413631.html.
12. J. Savulescu, B. Foddy, and M. Clayton, “Why We Should Allow Performance Enhancing
Drugs in Sports,” British Journal of Sports Medicine 38 (2004).
13. In fairness to Lewis, that test might truly have been a case of inadvertent use, as he claimed:
the relatively low stimulant levels that triggered the positive result would not be considered high
enough to fail the test by today’s standards.
14. John Brant, “The Marriage That Led to the Russian Track Team’s Olympic Ban,” New York
Times Magazine, June 22, 2016. Focusing on doping by a middle-distance woman, this article
suggests that doping lowers the best times of women in the 800 meters by at least five seconds, or 4
percent. Four percent for a 100-meter dash is about .38 seconds. Some extremely unreliable guesses
suggest 0.2 second savings from steroids for top 100-meter sprinters.
15. In the summer of 2018, the American newcomer Christopher Coleman ran a 9.79-second 100
meters. The fastest 100 meters by someone other than Usain Bolt with a long history of running and
testing with no positive tests is the 9.82 by Richard Thompson of Trinidad and Tobago.
16. Oliver Pickup, “Usain Bolt Denies Using Performance-Enhancing Drugs Ahead of Paris
Diamond League Meeting,” The Telegraph, July 3, 2013,
www.telegraph.co.uk/sport/othersports/athletics/10157145/Usain-Bolt-denies-using-performance-
enhancing-drugs-ahead-of-Paris-Diamond-League-meeting.html.
17. For the Armstrong denial quotes, see Stephen McMillan, “Lance Armstrong’s Doping Denials
—in Quotes,” The Guardian, January 18, 2013, www.theguardian.com/sport/2013/jan/18/lance-
armstrong-doping-denials-quotes.
18. Charlie Gillis, “Too Fast to Be Clean: Why the World’s Fastest Man Can’t Run Clear of
Controversy,” Maclean’s, September 1, 2009, www.macleans.ca/society/too-fast-to-be-clean/.
19. On the use of steroids in swimming, see Donald McRae, “Michael Jamieson: ‘Swimming Has
a Problem. Micro-dosing Is a Huge Issue,’” The Guardian, January 28, 2018; in soccer, Ferdinand
Dyck, “An Ex-Pro Soccer Player Explains How Easy It Is to Dope,” Vice, July 6, 2018; in skiing,
Gordy Megroz, “Lindsey Vonn and the (Vast) Potential for Doping in Ski Racing,” Outside, June 5,
2013.

Chapter 4. Are Athletes Worth All That Money?

1. U.S. Bureau of Labor Statistics, “Economic News Release,” table 1,
www.bls.gov/news.release/wkyeng.t01.htm, accessed June 13, 2021.
2. MLB from “Scherzer Highest-Paid Player, Red Sox Top Payroll List Again,” Associated Press,
March 28, 2019; NBA from “2020–21 Player Contracts,” Basketball Reference, www.basketball-
reference.com/contracts/players.html. When quoting statistics on pay for a typical worker or athlete, I
use medians wherever possible, because the U.S. Census Bureau reports medians in most cases, and
because medians give a better sense of the typical person, while averages can be skewed by a few
outliers. I could not find median pay for European soccer leagues, but the averages there are also in
the multi-millions of dollars.
3. “Sports List of the Day,” December 5, 2011, https://sportslistoftheday.com/2011/12/05/major-
league-baseballs-average-salaries-1964–2010/. Only average (as opposed to median) salaries are
available for the 1960s, so I can only provide an upper bound based on the fact that averages in
sports salaries tend to be higher than medians.
4. United States Census Bureau, “Current Population Reports: Consumer Income,” September 24,
1965, www2.census.gov/prod2/popscan/p60-047.pdf, accessed June 13, 2021.
5. Jabari Young, “Major League Baseball Revenue for 2019 Season Hits a Record $10.7 Billion,”
CNBC; James Wagner, “M.L.B. Extends TV Deal with Fox Sports Through 2028,” November 15,
2018; Maury Brown, “MLB Sees Record Revenues of $10.3 Billion for 2018,” January 7, 2019,
Forbes.com.
6. Shalini Ramachandran, “MLB’s Streaming-Tech Unit Goes Pro,” Wall Street Journal, February
23, 2015.
7. Dave Cameron, “Big Ticket Signings Don’t Drive Attendance,” FanGraphs.com, December 9,
2011.
8. Gerald W. Scully, “Pay and Performance in Major League Baseball,” American Economic
Review 64 (1971); Henry Aaron Baseball Reference page, accessed December 14, 2020.
9. Matt Swartz, “The Recent History of Free Agent Signings,” FanGraphs.com, July 11, 2017.
10. Michael Haupert, “MLB’s Annual Salary Leaders Since 1874,” Society for American Baseball
Research, http://sabr.org/research/mlbs-annual-salary-leaders-1874-2012.
11. Marc Topkin, “Tommy Pham Wins Arbitration Case over Rays, Gets $4.1M,” Tampa Bay
Times, February 5, 2019.
12. Sherwin Rosen and Allen Sanderson, “Labour Markets in Professional Sports,” Economic
Journal 111 (2001).
13. Hayley C. Cuccinello, “The World’s Highest-Paid Comedians of 2018,” Forbes.com,
December 19, 2018.
14. Federal Baseball Club v. National League, 259 U.S. 200 (1922).
15. Simon Rottenberg, “The Baseball Players’ Labor Market,” Journal of Political Economy 64
(1956).
16. Haupert, “MLB’s Annual Salary Leaders Since 1874.”
17. Emmanuel Saez, “Income and Wealth Inequality: Evidence and Policy Implications,”
Neubauer Collegium Lecture, University of Chicago, 2014.
18. Kevin J. Murphy, “Executive Compensation: Where We Are, and How We Got Here,” in
Handbook of the Economics of Finance, edited by George M. Constantinides, Milton Harris, and
Rene M. Stulz (Elsevier, 2013).
 19. Steven N. Kaplan and Joshua Rauh, “It’s the Market: The Broad-Based Rise in the Return to
Top Talent,” Journal of Economics Perspectives (2013).
20. Benjamin Kabak, “The Economics of Yankee Tickets,” River Avenue Blues, March 14, 2008,
http://riveraveblues.com/2008/03/the-economics-of-yankee-tickets-2327/.
21. “New York Yankee Suites,” Suite Experience Group, www.suiteexperiencegroup.com/all-
suites/mlb/new-york-yankees/, accessed May 24, 2019.
22. Tadd Haislop, “Patrick Mahomes Contract Details: Here’s How Much Guaranteed Money
Chiefs QB Will Make in ‘Half-Billion’ Dollar Deal,” Sporting News, September 10, 2020.
23. Debra Bell, “US News Questioned Football’s Future Nearly 45 Years Ago,” US News and
World Report, February 1, 2013.
24. I use Mahomes as an example because of his huge contract. That he does ads for an insurance
company is a happy coincidence.
25. Dave Cameron, “Big Ticket Signings Don’t Drive Attendance,” FanGraphs.com, December 9,
2011; Craig Edwards, “Mike Trout Leaves Money on the Table Again,” FanGraphs.com, March 19,
2019.
26. Ross Tucker, “No Guarantees in the NFL,” July 6, 2016,
www.sportsonearth.com/article/188169010/guaranteed-contracts-will-not-work-nfl.
27. I have focused on baseball and football because the contract forms lie at opposite ends of the
risk-versus-incentive spectrums. Basketball contracts look closer to MLB contracts in that they are
fully guaranteed, but they are shorter than the longest MLB deals because of restrictions imposed by
the collective bargaining arrangement between the NBA and its players’ union. NHL contracts also
tend to be fully guaranteed and can be long-term. The classic study of “shirking” by athletes is
Kenneth Lehn, “Property Rights, Risk Sharing and Player Disability in Major League Baseball,”
Journal of Law and Economics 25 (1982). Other baseball studies find less moral hazard, including
Anthony C. Krautmann, “Shirking or Stochastic Productivity in Major League Baseball?” Southern
Economic Journal 56 (1990). On soccer, see Bernard Frick, “Performance, Salaries, and Contract
Length: Empirical Evidence from German Soccer,” International Journal of Sport Finance 6 (2011);
on the NBA, see David J. Berri and Anthony C. Krautmann, “Shirking on the Court: Testing for the
Incentive Effects of Guaranteed Pay,” Economic Inquiry 44 (2006).
28. This estimate is based on comparing actual NBA revenues in the lockout-shortened 2011–
2012 season to the average of 2010–2011 and 2012–2013 NBA revenues. See “National Basketball
Association Total League Revenue from 2001/02 to 2019/20,” Statista.com,
www.statista.com/statistics/193467/total-league-revenue-of-the-nba-since-2005/, accessed June 8,
2021. An alternative estimate, from Forbes, is that the total losses to the NBA and players were $800
million. See Patrick Rishe, “NBA Lockout Costs League $800 Million . . . and Counting; Players
Justified to Fight in Courts,” Forbes.com, November 16, 2011.

Chapter 5. Why Do Athletes Use Their Least Successful Moves So

Often?
1. Adding the shoot-straight option for the kicker and the don’t-dive option for the goalie makes
things more complicated, but the logic remains exactly the same.
2. Left is the natural side for a right-footed kicker to aim and is, indeed, a consistently better bet
for right-footed kickers. The opposite holds true for left-footed kickers.
3. Ignacio Palacios-Huerta, “Professionals Play Minimax,” Review of Economic Studies 70
(2003): 395–415.
4. That’s only 30 percent of the time rather than one-third—close enough for me, especially given
that my optimal percentage of one-third was obtained somewhat arbitrarily.
 5. I did not use this example in my case, because I aim for the center of the box on every serve and
let general inaccuracy generate the random strategy.
6. Mark Walker and John Wooders, “Minimax Play at Wimbledon,” American Economic Review
91, no. 5 (2001): 1521–1538.
7. Shih-Hsun Hsu, Chen-Ying Huang, and Cheng-Tao Tang, “Minimax Play at Wimbledon:
Comment,” American Economic Review 97, no. 1 (2007): 517–523.
8. Kudos to the astute readers who have figured out that my stopwatch method actually introduces
a small amount of serial dependence in my serve-and-volley strategy. A truly random strategy would
have me serve and volley an average of three times every ten points, but the stopwatch method
means that I serve and volley exactly three out of every ten points.
9. Axel Anderson, Jeremy Rosen, John Rust, Kin-Ping Wong, “Disequilibrium Plan in Tennis,”
working paper, February 2021.
10. Neil Paine, “Game Theory Says R. A. Dickey Should Throw More Knuckleballs,”
FiveThirtyEight.com, August 13, 2015.
11. Matt Swartz, “Bayes at the Plate: Game Theory and Pitch Selection,” 2013 SABR Analytics
Conference Presentation, http://sabr.org/latest/2013-sabr-analytics-conference-research-presentations,
accessed July 3, 2014. Interview with Matt Swartz on July 13, 2014.
12. Mike Matheny, “Calling Pitches,” blog post, www.mikematheny.com/mikes-blog/calling-
pitches, accessed July 5, 2014.

Chapter 6. How Does Discrimination Lead to a Proliferation of

French Canadian Goalies?
1. Alexander Wolff, “The NFL’s Jackie Robinson,” Sports Illustrated, October 12, 2009.
2. Timothy Burke, “Your Complete Quotable Guide to Decades of Donald Sterling’s Racism,”
Deadspin, April 26, 2014.
3. For a high-stakes example of the problems caused by statistical discrimination, see Mike Baker
and Nicholas Bogel-Burroughs, “How a Common Air Freshener Can Result in a High-Stakes Traffic
Stop,” New York Times, April 17, 2021.
4. Joseph Price and Justin Wolfers, “Racial Discrimination Among NBA Referees,” Quarterly
Journal of Economics 125 (2010): 1859–1887.
5. Jerry Zgoda and Dennis Brackin, “Timberwolves: Pale in Comparison to Rest of NBA,”
Minneapolis Star Tribune, October 28, 2012.
6. Howard Sinker, “Recalling Calvin Griffith’s Bigoted Outburst in Southern Minnesota,”
Minneapolis Star Tribune, April 30, 2014.
7. Hang Up and Listen podcast, “The Major League Baseball Needs to Reckon with the Negro
Leagues Edition,” August 17, 2020.
8. “1965 NBA All-Star Game,” Basketball Reference, www.basketball-
reference.com/allstar/NBA_1965.html, accessed June 10, 2021; “1975 All-Star Game Voting,”
Basketball Reference, www.basketball-reference.com/allstar/NBA_1975_voting.html, accessed June
10, 2021; growth in black players, 1954–1970, Gerald W. Scully, “Economic Discrimination in
Professional Sports,” Law and Contemporary Problems 38 (1973): 67–84.
9. Scully, “Economic Discrimination in Professional Sports”; Lawrence M. Kahn and Peter D.
Sherer, “Racial Differences in Professional Basketball Players’ Compensation,” Journal of Labor
Economics 6 (1988): 40–61; Lawrence M. Kahn, “The Sports Business as a Labor Market
Laboratory,” Journal of Economic Perspectives 14 (2000): 75–94.
10. Mark T. Kanazawa and Jonas P. Funk, “Racial Discrimination in Professional Basketball:
Evidence from Nielsen Ratings,” Economic Inquiry 39 (2001): 599–608.
 11. Frank Newport, “In U.S., 87% Approve of Black-White Marriage, vs. 4% in 1958,”
Gallup.com, July 25, 2013.
12. Ha Hoang and Daniel A. Rascher, “The NBA, Exit Discrimination, and Career Earnings,”
Industrial Relations: A Journal of Economy and Society 31 (1999): 69–91; Peter A. Groothuis and J.
Richard Hill, “Exit Discrimination in the NBA: A Duration Analysis of Career Length,” Economic
Inquiry 42 (2004): 341–349.
13. Stefan Szymanski, “A Market Test for Discrimination in the English Professional Soccer
Leagues,” Journal of Political Economy 108 (2000): 590–603.
14. For example, the reaction by Boston Bruins fans after losing a playoff game in 2014 on an
overtime goal by star Montreal Canadiens defenseman P. K. Subban, who is of African Caribbean
descent. Bruins supporters on Twitter used Subban’s name and the n-word in conjunction more than
seventeen thousand times in the hours after the goal. “P. K. Subban Targeted by Racist Tweets After
Habs Win,” CBC News, May 2, 2014.
15. Philip Authier, “The Dream of Independence Lives on in a New Generation,” Montreal
Gazette, October 24, 2020.
16. Marc Lavoie, Gilles Grenier, and Serge Coulombe, “Discrimination and Performance
Differentials in the National Hockey League,” Canadian Public Policy 13 (1987): 407–422.
17. W. Bentley MacLeod, “Optimal Contracting with Subjective Evaluation,” American Economic
Review 93 (2003): 216–240.
18. Alan Schwarz, “Study of NBA Sees Racial Bias in Calling Fouls,” New York Times, May 2,
2007, A1.
19. Devin G. Pope, Joseph Price, and Justin Wolfers, “Awareness Reduces Racial Bias,”
Management Science 64 (2018): 4988–4995; “Ref, You Suck!” Against the Rules with Michael Lewis
podcast, April 2, 2019.
20. It’s necessary to distinguish “African American” (or, more specifically, U.S. born with African
heritage) from “Black” here, as many major leaguers of African descent were born in Latin America.
21. Major League Baseball also has fewer Black players in general: 7 percent of major leaguers
were black in 2021, down from 19 percent in 1981. Brandon Jones, “73 Years After Robinson Broke
Barrier, Baseball Still Struggles,” Cronkite News, January 6, 2021; Earl Smith and Marissa Kiss,
“Why Are There So Few Black American Players in MLB 74 Years After Jackie Robinson Took the
Field?” Philadelphia Inquirer, April 1, 2021.
22. University of Chicago General Social Survey, http://www3.norc.org/GSS+Website/. Select
“whites” and “blacks” from the subject index.
23. Richard L. Harris, “For Campanis, a Night That Lived in Infamy,” Los Angeles Times, August
5, 2008.
24. “Key Events in Marge Schott’s Tenure as Owner of Cincinnati Reds,” Associated Press, June
12, 1996.
25. Marc H. Morial, “Black Quarterbacks Leading More Teams in the NFL,” Huffington Post,
September 30, 2013.
26. Samuel G. Freedman, “The Year of the Black Quarterback,” New Yorker website, January 30,
2014.
27. Roland Laird, “White Up the Middle: How Pro Football Changed the American Racial
Psyche,” PopMatters, January 19, 2011; Jason Reid and Jane McManus, “The NFL’s Racial Divide,”
The Undefeated, April 26, 2017.
28. Kurt Badenhausen, “Highest-Paid Female Athletes 2020: 50 Years After Creation of Women’s
Tour, Tennis Dominates Earnings List,” Forbes.com, August 17, 2020.
29. Lawrence M. Kahn, “Discrimination in Professional Sports: A Survey of the Literature,”
Industrial and Labor Relations Review 44 (1991): 395–418.
 30. Kamakshi Tandon, “US Open Ratings Increase with Big Numbers During Chaotic Woman’s
Final,” Tennis.com, September 11, 2018.
31. John J. Donohue III and James Heckman, “Continuous Versus Episodic Change: The Impact
of Civil Rights Policy on the Economic Status of Blacks,” Journal of Economic Literature 29 (1991):
1603–1643.

Chapter 7. How Do Ticket Scalpers Make the World a Better

Place?
1. Quotes from Amy Stephens come from an interview with the author on August 14, 2014.
2. “Tickets to the Colosseum,” www.tribunesandtriumphs.org/colosseum/tickets-to-the-
colosseum.htm.
3. Pascal Courty, “Some Economics of Ticket Resale,” Journal of Economic Perspectives 17
(2003): 85–97.
4. Jim Armstrong, “Legal Scalping of Bruins Stanley Cup Tickets,” CBS Boston, May 31, 2011,
http://boston.cbslocal.com/2011/05/31/legal-scalping-of-bruins-stanley-cuptickets/.
5. Phillip Leslie and Alan Sorensen, “Resale and Rent-Seeking: An Application to Ticket
Markets,” Review of Economic Studies 81 (2014): 266–300.
6. Andrew Sweeting, “Dynamic Pricing Behavior in Perishable Goods Markets: Evidence from
Secondary Markets for Major League Baseball Tickets,” Journal of Political Economy 120 (2012):
1133–1172.
7. Ken Belson, “As Economy Sagged, Online Sports Ticket Market Soared,” New York Times,
January 14, 2011.
8. Adam Davidson, “How Much Is Michael Bolton Worth to You?” New York Times, June 4, 2013.
9. See, e.g., Patrick Rishe, “Dynamic Pricing: The Future of Ticket Pricing in Sports,”
Forbes.com, January 6, 2012.
10. Ameet Sachdev, “Baseball Teams Get Dynamic with Pricing,” Chicago Tribune, May 12,
2013, http://articles.chicagotribune.com/2013-05-12/business/ct-biz-0512-stub-hub-
20130512_1_stubhub-bleacher-ticket-ticket-reselling.
11. Quote from Larry Baer, president and CEO of the San Francisco Giants, in an email exchange
with the author, April 18, 2021.
12. Gate receipt data from statistica.com.
13. Filling the stadium is more of an issue for MLB teams, where the average stadium is about
two-thirds full. Owing to smaller arenas or fewer games per season, NBA, NFL, and NHL teams
generally sell out or nearly sell out.
14. I am alluding here to a theory well known to economists, first espoused by the Nobel Prize
winner Gary Becker. Becker suggested that he and his wife would willingly wait in line for a table at
a local seafood restaurant despite the existence of a comparable but less busy restaurant directly
across the street because of their (and other people’s) desire to patronize businesses that have been
given a social stamp of approval. Gary S. Becker, “A Note on Restaurant Pricing and Other
Examples of Social Influences on Price,” Journal of Political Economy 99 (1991): 1109–1116.
15. This is the premise for more than a few sitcom gags and real-life breakups.
16. Chris Sagers, “Why Fans Can’t Win When It Comes to Buying Concert, Game Tickets,”
Cleveland.com, October 24, 2014.
17. Christopher Connors, “New York Yankees: The Rich and Poor Seating Divide at Yankee
Stadium,” Bleacherreport.com, September 21, 2012.
18. Tom Ley, “Yankees COO Defends New Ticketing Policy Like a True Rich Asshole,”
Deadspin, February 18, 2016.
 19. Caleb Garling, “MLB Puts Squeeze on Sneaky Fans with App,” sfgate.com, March 16, 2013.

Chapter 8. Why Should You Be Upset If Your Hometown Hosts the

Olympics?
1. Richard Orange, “Berlusconi Lavishes Praise on ‘Absolutely Justifiably Loved’ Kazakh
Leader,” The Telegraph, December 2, 2010; Robert Mendick, “Tony Blair’s Five-Million Pound Deal
to Advise Kazakh Dictator,” The Telegraph, April 23, 2016; biography on thefamouspeople.com,
accessed June 11, 2016; C. J. Chivers, “Kazakh President Re-elected; Voting Flawed, Observers
Say,” New York Times, December 6, 2005; Victor Mather, “2022 Winter Games Vote Down to Two
Cities and Some Major Concerns,” New York Times, July 29, 2015; Paul Bartlett, “Disappointment
for Almaty as Winter Olympics Go to Beijing,” Eurasianet.org, July 31, 2015; Nadezhda Khamitova,
“Almaty to Bid to Host 2022 Winter Olympics,” Astana Times, August 21, 2013.
2. Jules Boykoff, “Beijing and Almaty Contest Winter Olympics in Human Rights Nightmare,”
The Guardian, July 30, 2015, www.theguardian.com/sport/2015/jul/30/china-kazakhstan-winter-
olympics-2022.
3. “Stockholm Drops Its Bid to Host the 2022 Winter Olympic Games,” BBC, January 17, 2014;
“Krakow Withdraws 2022 Winter Olympics Bid,” Associated Press, May 26, 2014; “Ukraine
Withdraws Bid for 2022 Winter Olympics,” Associated Press, June 30, 2014.
4. Alan S. Blinder, Hard Hearts, Soft Heads (Perseus Books, 1988), 1.
5. John Siegfried and Andrew Zimbalist, “The Economics of Sports Facilities and Their
Communities,” Journal of Economic Perspectives 14 (2000): 103.
6. Joshua Yaffa, “The Waste and Corruption of Vladimir Putin’s 2014 Winter Olympics,”
Bloomberg Businessweek, January 2, 2014.
7. Alice Ollstein, “Scott Walker to Cut $300 Million from Universities, Spend $500 Million on a
Pro Basketball Stadium,” thinkprogress.org, February 2, 2015; Michael Powell, “Bucks’ Owners
Win, at Wisconsin’s Expense,” New York Times, August 14, 2015; “Walker Signs Bill to Fund New
Milwaukee Bucks Arena,” WMTV Milwaukee, August 12, 2015; Don Walker, “Milwaukee Bucks
Dramatically Expanding Ownership Group,” Milwaukee Journal Sentinel, October 16, 2014; Bruce
Murphy, “Cost for Taxpayers in Latest Bucks’ Deal?” urbanmilwaukee.com, July 23, 2015.
8. John Duerden, “How Qatar Is Trying to Build a Team to Qualify for 2018 World Cup,”
ESPN.com, January 10, 2015, www.espn.com/soccer/club/name/4398/blog/post/2184455/headline.
9. Aaron Schacter, “Why Does Qatar Even Want to Host the World Cup?” Public Radio
International’s The World, June 12, 2015.
10. “Olympics 2012: A Spectacular Triumph for London,” CNN, August 12, 2012,
www.cnn.com/2012/08/12/opinion/hooper-london-triumph/index.html; “Lon-don 2012: How the
World Saw the Olympic Games,” BBC News, August 13, 2012, www.bbc.com/news/uk-politics-
19238284.
11. The total cost to London for hosting was $14 billion.
12. Golden 1 Center sources: Max Resnik, “Report: Sacramento Sees Economic Boost Thanks to
Golden 1 Center,” KCRA-TV, October 17, 2017; Downtown Sacramento Partnership, “Golden 1
Center Fact Sheet,” downloaded November 13, 2018 from
www.downtownsac.org/about/reports/golden-1-center-creates-economic-spark/; Dale Kasler, “Cost
of Building Golden 1 Center Just Went Up Again,” Sacramento Bee, November 29, 2016. Arlington
Stadium: Mac Engel, “Price of New Rangers Stadium Up $200 Million; New Ticket Prices and Seat
Relocation Coming,” Fort Worth Star-Telegram, October 10, 2018.
13. For details on previous studies, see Andrew Zimbalist, Circus Maximus (Washington, D.C.:
Brookings Institution Press, 2015).
 14. Paul Hirsch, “Walter O’Malley Was Right,” SABR.org, accessed June 11, 2021; Gary
Jeanfaivre, “Why the Whalers Left Connecticut, and Why It’s Important Now,” Patch.com, January
14, 2014; John Royal, “It Was Bud Adams, Not the Fans, Who Caused the Oilers Move to
Nashville,” Houston Press, February 6, 2017.
15. Roger G. Noll and Andrew Zimbalist, “Build the Stadium—Create the Jobs!” in Sports, Jobs,
and Taxes: The Economic Impact of Sports Teams and Stadiums, edited by Roger G. Noll and
Andrew Zimbalist (Washington, D.C.: Brookings Institution Press, 1997), 29.
16. Paul Dolan, Georgios Kavetsos, Christian Krekel, Dimitris Mavridis, Robert Metcalfe, Claudia
Senik, Stefan Szymanski, and Nicolas R. Ziebarth, “Quantifying the Intangible Impact of the
Olympics Using Subjective Well-Being Data,” Journal of Public Economics 177 (2019).
17. The San Francisco 49ers’ new stadium site in Santa Clara was also approved by voters, but
with the stipulation that no public funds be spent on the construction. See the Sports and Urban
Policy Initiative at Georgia State University for data on efforts to use public money to fund stadiums,
https://education.gsu.edu/kh/khresearchoutreach/center-for-sport-and-urban-policy/#stadiatrack,
accessed June 11, 2021.
18. Charlotte Arena: “Bobcats Unveil $265M Downtown Charlotte Arena,” Associated Press,
October 19, 2005. Braves Stadium: Patricia Murphy, “Tea Party Strikes Out Against the Atlanta
Braves,” Daily Beast, November 27, 2013; Tom Ley, “Braves President: Stadium Deal Had to Be
Done in Secret,” Deadspin, May 27, 2014; Matthew Pearl, “Braves President: Cobb Deal Had to Be
Kept Under Wraps,” 11Alive, May 22, 2014.
19. Ken Belson, “Stadium Boom Deepens Municipal Woes,” New York Times, December 24,
2009; see also Eric Roper, “Taxes to Pay for Now-Open U.S. Bank Stadium Rebound, Thanks to
Gamblers,” Minneapolis Star Tribune, July 22, 2016.
20. Of course, one can easily argue that it is merely a quirk of American politics that deals such as
those in Cobb County and Cincinnati are not considered illegal corruption.

Chapter 9. Who Wins When People Gamble?

1. Charles Curtis, “A Missed Late PAT from Browns’ Cody Parkey Was a Horrible Bad Beat for
Bettors,” USA Today, October 26, 2020.
2. Todd Dewey, “Las Vegas Sportsbooks Suffer Worst Sunday of NFL Season,” Las Vegas
Review-Journal, October 25, 2020.
3. Some years are different, given that the broader population does seem to have strong feelings
about the New England Patriots, Tom Brady, and perhaps the Dallas Cowboys.
4. Thomas Johnson, “Rory McIlroy’s Father Wins $171,000 Bet on His Son to Win the British
Open,” Washington Post, July 21, 2014.
5. New York State Gaming Commission, “New York State Racetracks and Applicable Takeout
Rates,” www.gaming.ny.gov/pdf/NYSTakeoutRates0417.pdf.
6. Sam Walker, “The Man Who Shook Up Vegas,” Wall Street Journal, January 5, 2007.
7. Stoll uses a star-rating system whereby he assigns more stars to games about which he is most
confident. He calculates his winning percentages by weighting four-star games more heavily than
two-star games, but a more traditional straight-wins-and-losses calculation would still yield
abnormally high win percentages from Stoll’s picks.
8. Nino Bucci, “Former Junior Tennis Champion Oliver Anderson Avoids Conviction After
Pleading Guilty to Match-Fixing,” Sydney Morning Herald, May 23, 2017.
9. Justin Wolfers, “Point Shaving: Corruption in NCAA Basketball,” American Economic Review
Papers and Proceedings, May 2006, 279–283; Richard Borghesi, “A Case Study in Sports Law
Analytics: The Debate on Widespread Point Shaving,” Journal of Sports Analytics 1, no. 2 (2015):
87–89; “The Most Notorious Sports Betting Scandals of All Time,” SportsHandle.com, n.d., accessed
December 14, 2020.
10. Jessica Dickler, “March Madness Takes a Toll on Productivity,” CNBC.com, March 18, 2018.
11. A. Blaszczynski and E. Farrell, “A Case Series of 44 Completed Gambling-Related Suicides,”
Journal of Gambling Studies 14 (1998).
12. Martin Young, Bruce Doran, and Francis Markham, “Too Close to Home: People Who Live
Near Pokie Venues at Risk,” The Conversation, December 5, 2013.
13. Linette Lai, “New Rules on Advance Payment of Entry Levies from August,” Straits Times,
April 4, 2019.
14. Danson Cheong and Melissa Lin, “Online Betting to Be Allowed in Next Two Months via
Singapore Pools, Singapore Turf Club,” If Only Singaporeans Stopped to Think blog, September 30,
2016.

OceanofPDF.com
 Acknowledgments

David Oyer (my son, the hero of the chapter 1 Little League game and
my Ping-Pong nemesis in chapter 5) was a major contributor to this book.
He assisted with much of the research and drafted and edited several
chapters. We had many discussions (and more than a few arguments) about
the sports and the economics aspects of the book. Since I took David to
Comiskey Park when he was four years old, it has been a great joy to
experience sports with him, as both a fan and a participant. We spent time
together at many Stanford sporting events, Oakland A’s games, and
countless Little League and other games as he was growing up. Now that he
is grown, we don’t get to do that as often, but he keeps me informed and
amused with regular text updates about the A’s and other sports topics.
Spending time with David made all the research for this book far more
enjoyable than it would otherwise have been.
I also spent a lot of delightful time playing and enjoying sports with my
daughter, Lucy Oyer. Though my years of coaching her baseball and soccer
teams did not lead to athletic stardom, she did go on to be an economics
major, which seems to have prepared her well for nonathletic career
success.
My wife, Kathryn Stoner, has been a wonderful companion and source
of joy throughout this project. She is always encouraging, and she makes
me feel lucky every day. She also knows enough economics to get by and,
despite more than her share of injuries, holds her own in the sports world.
Thanks to my stepchildren, Abby and Adam Weiss, for brightening up
our household at various times throughout this project. They take their
athletic pursuits seriously and, hopefully, are learning some economics
along the way.
Bill Frucht has been a terrific editor. I enjoyed the many conversations
we had about the book, even when they inevitably devolved into lamenting
the condition of the Mets. Bill was extremely patient (too patient!) and had
excellent insights on the big picture of the book, as well as the little details.
Speaking of patient, my agent Zoe Pagnamenta has taught me a great deal
about shaping a book. Thanks, also, to Julio Cesar Franco Ardila for help
with the figures.
I have had many economics colleagues, those at Stanford as well as
others in the profession, who have taught me a great deal about economics
over the years and have also engaged in many great discussions of sports.
They are too numerous to single out here, but two are worthy of special
mention. My friend and co-author Scott Schaefer has influenced my
economic thinking for over twenty-five years and, more specifically, made
a key point about the difference between football and baseball players that
was crucial to chapter 4. Ed Lazear also made me a much better economist
over the last few decades. That happened mostly through seminars, working
on papers, and traditional academic pursuits, but we also spent countless
hours discussing sports (with and without mentioning the underlying
economics), sometimes while skiing and watching football. Sadly, Ed
passed away toward the end of this project.
Josie, our faithful flat-coated retriever, was by my side throughout the
entire project, including watching the Ping-Pong game in chapter 5. She
was a great athlete and the most optimistic distraction from work a person
could ask for, though her economics were never very good. Sadly, Josie
passed away as I concluded the book. Her younger sister, Phoebe, also
brings great joy to our household.
Finally, I want to thank my late mother, Alice Oyer, for all the support
she gave me for over forty years. And I thank my father, Calvin Oyer, for
playing sports with me and taking me to games when I was little, but never
overdoing it. Neither of my parents knew much about economics, but they
laid the groundwork that made this book and all my other academic
endeavors possible.

OceanofPDF.com
 Index

Aaron, Hank, 68
African Americans, 19
in baseball, 113–14
earnings of, 15–16, 18–19
equipment and training unavailable to, 115–16
agency problems, 140, 152, 154
Almaty, Kazakhstan, 138–39, 140, 156
Amy’s Tickets, 120, 125, 135
Anderson, Oliver, 168, 169
Apple Computers, 71
Archer, Chris, 113
archery, 38, 41
Arizona Diamondbacks, 131
arms control, 54
Armstrong, Lance, 45–46, 49, 51, 59–60, 62–63
athletic scholarships, 12
Atlanta Braves, 151
Australian Open (tennis), 117, 118
Austro-Hungarian Empire, 35
auto racing, 70, 100–101

badminton, 24
Bahrain, 1
BALCO (Bay Area Laboratory Co-operative), 57
Baltimore, Md., 148
Baron Cohen, Sacha, 138
baseball: Black players in, 113–14
Dominican players in, 20
Japanese players in, 114–15
 league structure in, 175
performance-enhancing drugs in, 44–45, 51–53, 60
pitch selection in, 94–100
racial integration of, 102, 103
racism in, 113–15, 119
Reserve Clause in, 72–73, 74
revenue in, 67, 131
salaries in, 53, 65–75
technological change and, 70–71
ticket sales and, 130
base stealing, 94
basketball, 1, 23
Black players in, 15–20, 103, 106–7, 108
collective bargaining in, 81–83
discriminatory officiating in, 106, 113
fan interest in, 116, 117
free agency in, 73
international competition in, 42–43
league structure in, 175
performance-enhancing drugs in, 61, 62
salaries in, 64, 80, 108, 109, 110, 115
Beane, Billy, 97, 166, 167
A Beautiful Mind (film), 88
Becker, Gary, 109–10, 190n14
Bedinghaus, Bob, 154
beer industry, 32
Beijing, China, 139–40
Beijing Winter Olympics (2022), 139, 140, 156
Berlusconi, Silvio, 138
best alternative to a negotiated agreement (BATNA), 81–82
Betts, Mookie, 97
Blair, Tony, 138
Blinder, Alan, 140
bluffing, 100–101
Biogenesis scandal (2014), 54–55
Bird, Larry, 106, 109
Boldon, Ato, 57–58
Bolt, Usain, 58–59, 60
Bonds, Barry, 51–53, 61
bookies, 158–72
Boston Bruins, 125–26, 187n14
Boston Celtics, 106, 109
Boston Red Sox, 130
Bovada, 163
bowling, 61
Bradley, Bill, 164
Bradley Act (1992), 164
Brady, Tom, 159
Braun, Ryan, 54
Brazil, 87
breakfast cereals, 32
broadcast rights, 70
Brooklyn Dodgers, 149
Budinger, Chase, 106
Buffalo Bills, 78
bunting, 94
Burrow, Joe, 158

Calgary, Alberta, 156

Campanis, Al, 114
Canada, 29–30, 111–12
Canseco, Jose, 51
Carolina Hurricanes, 149
casinos, 162, 164–65, 167, 172
centers, in football, 115
changeup, 96
Charlotte, N.C., 148
Charlotte Hornets, 151
Chicago Bulls, 1
Chicago Cubs, 44
Chile, 25
China, 23, 24
Choi, K. J., 41
Christie, Linford, 57
Churchill Downs, 162
Cincinnati Bengals, 153–54, 158–59
Cincinnati Reds, 114
civil rights laws, 119
Clemens, Roger, 51
Cleveland Browns, 158–59, 163
Cleveland Rams, 102
Clifton, Nat, 103
climate, 25, 29, 33
Coleman, Christopher, 182n15
collective bargaining agreements (CBAs), 80–83
college basketball, 169–70
comparative advantage, 25, 26, 29, 30–31, 34
competition, 36, 42, 70, 71–73, 110, 119, 127–28
complementary resources, 27, 30
confirmation bias, 113
Connors, Chris, 136
consumption goods, 7, 10
Conte, Victor, 57, 58
Cooper, Chuck, 103
copper mining, 25
correlation vs. causation, 9, 12
corruption, 154–57
Couric, Katie, 60
Court, Margaret, 37
Covid-19 pandemic, 156–57
Cowens, Dave, 109
cross-country skiing, 26–29, 33
Crowe, Russell, 88
CrownBet, 168, 169
Curry, Dell, 19
Curry, Steph, 19
cycling, 45–51, 59–60, 62
Cy Young Award, 51, 113, 115
Czech Republic, 34–38, 42–43

Dallas Cowboys, 151

Dallas Stars, 151
defensemen, in hockey, 112
deGrom, Jacob, 96
Delaware, 164
del Potro, Juan Martin, 119
Detroit, Mich., 164
Dickey, R. A., 98
discrimination, 102–19
in baseball, 113–15, 119
by customers, 105–6, 107–8
by employees, 105
by employers, 105, 107–8
exit discrimination, 109
positional, 113–16
by referees, 106, 113
statistical, 104–5
taste-based, 104–5, 117
against women, 116–19
Djokovic, Novak, 119
Dr. Bob (Bob Stoll), 166–67
Dominican Republic, 20
dominant strategy: doping as, 48–49, 51–53, 55, 62, 63
in pitch selection, 95, 97
doping. See performance-enhancing drugs (PEDs)
drug testing, 2, 53–56, 62
Durant, Kevin, 15–20
dynamic pricing, 129–32, 135

Edens, Wes, 141

Effectively Wild (podcast), 66
Elder, Lee, 103
endorsements, 18
English-speaking Canadians, 111–12
EPO-CERA, 2
Epstein, David, 32
ESPN, 67
Estalella, Bobby, 4, 52–53, 60
Ethiopia, 31–32
Evert, Chris, 37

fastball, 96–98
Federal Baseball Club v. National League (1922), 71
Federer, Roger, 168
Final Four, 133–34, 135
Finland, 27
Finn, Adharanand, 34
5Dimes, 163
Flood, Curt, 72
Florida Panthers, 131
football, 20, 100
betting on, 158–59
Black players in, 17–18, 115
free agency in, 73
performance-enhancing drugs in, 57, 60
racial integration of, 102–3
racism in, 102–3, 119
salaries in, 65, 75, 81
forwards, in hockey, 112–13
Fox Sports, 67
France, 87
free agency, 73, 80, 81
French Canadians, 111–12
French Open (tennis), 117, 118

gambling: bettors at a disadvantage in, 160, 165–67, 170–71

on college basketball, 169–70
fixed, 167–70
on football, 158–59
as government revenue source, 164
legality of, 163, 170–74
rational vs. impulsive, 160–61, 172–74
on tennis, 168–69
types of, 162–63
game theory, 2, 85–101
prisoner’s dilemma in, 45, 47–49, 54–57, 60. See also dominant strategy
Gatlin, Justin, 57, 58
Gay, Tyson, 58
Gibson, Althea, 103
Global Gender Gap Report, 40
goalies: in hockey, 112, in soccer, 87–92
Golden 1 Center, 144–47, 151, 157
Golden State Warriors, 109
golf, 20–21, 25, 61
earnings in, 70
Korean women’s dominance in, 2, 38–42
racial integration of, 103
U.S. men’s dominance in, 24
Graz, Austria, 156
Great Britain, 1
Green, Maurice, 57–58
Green Bay Packers, 161
Greinke, Zach, 131
greyhound racing, 162
Griffith, Calvin, 107
Hart, Kevin, 71
Hartford Whalers, 149
Havlicek, John, 109
Herfindahl index, 32
Hermann, Chris, 69
Hewlett, William, 38
High Flying Bird (film), 82
Hinault, Bernard, 50
hockey, 35, 61, 81, 103, 105–6, 111–13
Holmes, Oliver Wendell, 71
home runs, 44, 51–53, 67, 69
horse racing, 162
hotel prices, 129, 130
Houston Astros, 96–97
Houston Oilers, 149
Hsu, Shih-Hsun, 93
Huang, Chen-Ying, 93
Hunter, Jim “Catfish,” 73
Hyundai Motor Company, 71

Indianapolis Colts, 79
industry concentration, 32
inequality, 73–74
International Olympic Committee (IOC), 155, 156
internet, 70, 163
interracial marriage, 109
investment goods, 7, 10
Italy, 86–87
It’s Not About the Bike (Armstrong), 46

Jackson, Phil, 1
Johnson, Ben, 57, 61
Johnson, Boris, 176
Jones, Marion, 59, 60, 62–63
Jordan, Michael, 1, 3, 60
Joyner, Florence Griffith, 59
judo, 39

Kahn, Lawrence, 108

Kalenjin tribe, 32–33
Kansas City Chiefs, 74, 159, 161
kayaking, 24
Kazakhstan, 138–39, 156
Kentucky Derby, 162
Kenya, 31–33, 39
Kerr, Steve, 1, 2
Kluber, Corey, 97
knuckleball, 98
Ko, Jin-Young, 41
Kodes, Jan, 35
Koppel, Ted, 114
Krakow, Poland, 139, 140
Kvitova, Petra, 37

lacrosse, 13
Lasry, Marc, 141
Las Vegas, Nev., 164
Lee6, Jeongeun, 2, 3–4
Levin, Josh, 107
Lewis, Carl, 57
Ley, Tom, 136–37
Lieberman, Daniel, 33–34
Liechtenstein, 22–23, 25–27, 32
Ligue 1 (French soccer league), 20
Little League, 6–7, 95–96
Lloyd, Earl, 103
lockouts, 83
Lombe, Harrison, 168, 169
London Summer Olympics (2012), 142–43, 146, 150, 151
Los Angeles Angels, 66–69, 73, 74, 81
Los Angeles Chargers, 159
Los Angeles Clippers, 104, 108, 110
Los Angeles Dodgers, 114, 149
Los Angeles Lakers, 109
Los Angeles Rams, 102
Louisiana, 164
Love, Kevin, 106
Luck, Andrew, 75, 79
Lviv, Ukraine, 139

Ma, Yo-Yo, 71
Mahomes, Patrick, 74–80
Major League Baseball (MLB). See baseball
Mandlikova, Hana, 35
Mann, Ralph, 60
Mantle, Mickey, 69, 72
marathon running, 23, 30–34, 39
March Madness, 171
Maris, Roger, 44
Maryland, 164
Matheny, Mike, 99–100
Mayfield, Baker, 158
Mayo, O. J., 62
McCoy, Gerald, 18
McCoy, LeSean, 18
McGwire, Mark, 44–45, 52
McHale, Kevin, 109
McIlroy, Rory, 161–62
Memphis Grizzlies, 109
merchandise, 67, 76, 82
Meucci, Daniele, 31
Middlebury College, 11, 14
middle linebacker, 115
Miller, Marvin, 72–73
Milwaukee Bucks, 109, 141
Minnesota Timberwolves, 106–7, 110–11
Minnesota Twins, 107
Minnesota Vikings, 164–65
Mitchell, George, 45
Mitchell Report, 45, 51, 52
mixed-strategy equilibrium: in pitching and batting, 94–95, 98, 100
in soccer, 88, 91, 94
in tennis, 88, 92–94
modern pentathlon, 38
Moneyball (book and film), 166
Montana, 164
Montreal Canadiens, 106, 187n14
Morgan, Alex, 117
Morial, Marc, 115
Moyer, Jamie, 98–99
muscle memory, 93

Nagano, Japan, 155

NASCAR, 100–101
National Basketball Association (NBA). See basketball
National Collegiate Athletic Association (NCAA), 55–56, 133–34, 169–70,
171
National Football League (NFL). See football
National Hockey League (NHL). See hockey
National Labor Relations Act (1935), 81
natural experiments, 9
Natural Resources Defense Council, 11
Nauru, 34
Navratilova, Martina, 35–38, 41
Nazarbayev, Nursultan, 138–39, 156, 157
New Jersey, 164
New Orleans, La., 164
New York Mets, 96
New York Yankees, 73, 74, 96, 130, 136
Noah, Joakim, 62
Noll, Roger, 149
Norway, 23, 25–28, 33, 139–40

Oakland A’s, 76, 130, 137, 166

Oden, Greg, 179n18
Olympic Games, 2, 22–23, 26, 27, 39, 57–59
bids to host, 138–41, 155–56
as financial sinkhole, 142–43
opportunity costs, 8, 34, 146
option value, 14
O’Ree, Willie, 103
Oregon, 164
Organisation for Economic Co-operation and Development (OECD), 39
Orlando, Fla., 148
Ortiz, David, 51
Oslo, 139–40, 150
Östersund, Sweden, 156
Ottawa Senators, 131
Oyer, David, 6–7, 9–10, 85–86, 95–96
Packard, David, 38
Pak, Se Ri, 41–42
Palacios-Huerta, Ignacio, 90
pari-mutuel betting, 162, 165
Park, Inbee, 42
Parkey, Cody, 158–59
Pekovic, Nikola, 106
penalty shootouts, 86–90, 91–92, 97, 99
Perez, Florentino, 175, 176
performance-enhancing drugs (PEDs): in baseball, 44–45, 51–53
in cycling, 45–51, 59–60, 62–63, 95
marginal value of, 61–62
testing for, 2, 53–56, 62
in track and field, 2, 57–60, 62–63
 visibility and, 62
PGA Tour, 21
Pham, Tommy, 69
Philadelphia Phillies, 72
Philips Arena, 129
phosphates, 34
Ping-Pong, 85–86
pitch count, 97
pitch selection, 94–100
who decides, 99–100
point shaving, 170
Poland, 24, 139
Population-Adjusted Power Index (PAPI), 23–32, 34, 39, 41
Portland, Oregon, 148
Portland Trail Blazers, 109
Powell, Asafa, 58
Pratt, Wanda, 15
Price, David, 113
Price, Joseph, 106
price discrimination, 129–30
PrimeSport, 133–34, 135
prisoner’s dilemma, 45, 47–49, 54–57, 60
Prohibition, 172–73
psychic costs, 49, 62
punting, 100
Putin, Vladimir, 140–41

Qatar, 141–42
quarterback, 115
Quebec, 112

Ramzi, Rashid, 1–2, 3

randomization, 92–94
in pitch selection, 100
Real Madrid, 175, 176
referees, 106, 113
Reserve Clause, 72–73, 74
rice cultivation, 25
ridesharing, 129–30
risk aversion, 75
Rivera, Mariano, 96
Robinson, Jackie, 102, 103, 104
Rodriguez, Alex, 51, 60
Romania, 24
Rose, Pete, 170
Rosen, Sherwin, 70–71
Rottenberg, Simon, 72
Russia, 27, 138, 140–41
Ryu, So Yeon, 42

Sabathia, C. C., 113

Sacramento, Calif., 144–49, 157
Sacramento Kings, 144–45, 148, 149
Sagers, Chris, 134
Saint Louis, Mo., 148
Saint Louis Cardinals, 44, 72
salary caps, 81
Salt Lake City, Utah, 155–56
San Antonio Spurs, 109
Sanderson, Allen, 70–71
San Diego, Calif., 148
San Francisco 49ers, 161, 192n17
savings rate, 40
scalpers, 4, 120–37
Schott, Marge, 114
Schuerholz, John, 151
Scully, Gerald, 68, 108
Seattle, Wash., 148
Seattle Seahawks, 164–65
Senegal, 20
serial independence, 93
serve and volley, 91, 92
Sherman Antitrust Act (1890), 71, 72
shirking, 79
Shuck, J. B., 69
Siegfried, John, 140, 147
Silicon Valley, 37–38, 42–43
simultaneous move games, 87
Singapore, 173–74
Sion, Switzerland, 156
skating, 41
skeleton racing, 24, 29
skiing, 4, 22, 25–26, 61
cross-country, 26–29, 33
Skoglund, Matt, 11, 14
slider, 96, 97, 98
Smith, Steven, 170
soccer, 61, 130, 141–42, 175–76
baseball contrasted with, 97, 99
Czech men in, 35
penalty shoot-out strategy in, 86–90
racism in, 111
salaries in, 64, 70
Senegalese players in, 20
Sochi Winter Olympics (2014), 140–41
Society for American Baseball Research, 68
Sosa, Sammy, 44–45, 52
South Dakota, 164
South Korea, 2, 38–42
Soviet Union, 35, 36
speed skating, 41
Sportsbetting.ag, 163
sprinting, 56–60, 62–63
stadiums, 4, 144–49, 150–54
Stanford University, 13, 37–38, 42, 85
Stanley Cup, 125–26
Stephens, Amy, 4, 120–22, 135
Stephens-Davidowitz, Seth, 19
Sterling, Donald, 104, 108
steroids. See performance-enhancing drugs (PEDs)
Stewart, Breanna, 117
Stockholm, 139, 140
stock market, 69, 159, 163, 167, 171
Stockton, John, 109
Stoll, Bob (“Dr. Bob”), 166–67
Strode, Woody, 102–3
Stroman, Marcus, 113
StubHub, 121–22, 128–29, 134, 135
Subban, P. K., 187n14
Sun Trust Park, 151
Super Bowl, 135, 160–61, 171
Super League, in European soccer, 175–76
Swart, Steve, 50
Swartz, Matt, 98–99
Sweden, 26, 27, 139
Sweeting, Andrew, 127, 131
Symanski, Stefan, 111

Tampa Bay Buccaneers, 159

Tang, Cheng-Tao, 93
television revenue, 67
Tennessee Titans, 149
tennis: Czech women’s dominance in, 34–38, 42–43
earnings in, 70, 117
Grand Slam events in, 117, 118, 119
match fixing in, 168–69
performance-enhancing drugs in, 55, 61–62
 racial integration of, 103
serve strategy in, 91–94, 97
women in, 117–19
Texas Rangers, 147
Thailand, 25
Thompson, Richard, 182n15
Ticketmaster, 129
ticket sales, 67
advance, 122–23, 127
by airlines, 126–27, 129, 132, 135
in antiquity and Renaissance, 122
attendance vs. revenue in, 131–32
for concerts, 126, 133
dynamic pricing in, 129–32, 135
nontransferable, 132–33
no-shows and, 136, 137
quasi-monopolistic, 133–34
resale restrictions in, 134–35
by scalpers and brokers, 4, 120–37
upgrades and, 136–37
Tokyo Summer Olympics (2020), 157
Tour de France, 45–51, 61
track and field, 1–2, 56–60
Trout, Mike, 73–82
athletic accomplishments of, 65–66, 68
record-setting contract of, 66–67, 69, 71, 73, 75–82
Tsakalakis, Chris, 124, 135
Tucker, Ross, 78
Turkey, 40
Turner Sports, 67

Uber, 129–30
UEFA Champions League, 87
Ukraine, 139
Union Cycliste Internationale, 50
unions, 80–81
Unitas, Johnny, 74
U.S. Anti-Doping Agency, 50
U.S. Open (tennis), 117, 118, 119
U.S. Women’s Open (golf), 2
Utah Jazz, 1, 109
utility, 85–86, 149–50, 160

Verlander, Justin, 97
volleyball, 101
von Neumann, John, 88

Wakefield, Tim, 96
Walker, Mark, 92–93
Walker, Scott, 141
Walkowiak, Robert, 49–50
Walton, Bill, 109
Washington, Kenny, 102–3
Washington Nationals, 98
Washington Senators, 107
weightlifting, 57, 60
Weirather, Tina, 4, 22, 27
Westbrook, Russell, 19
West Ham United F.C., 143
Williams, Ted, 69, 71, 72
Wilson, A’ja, 117
Wilson, Alex, 69
Wilson, Russell, 165
Wimbledon championship, 35, 37, 103, 117, 118–19, 169
winemaking, 25, 28–29
Winfrey, Oprah, 46
wins above replacement (WAR), 68
Wisconsin, 141
Wolfers, Justin, 106, 170
Women’s National Basketball Association (WNBA), 20, 115–16
Women’s Sports Foundation, 116
Wooders, John, 92–93
Woods, Tiger, 60
World Cup (soccer), 86–87, 141–42
World Economic Forum, 40

Xie Chengfeng, 21

Yang, Y. E., 41

Zimbalist, Andrew, 140, 142, 147, 149

OceanofPDF.com