Moneyball: The Art of Winning an Unfair Game

Moneyball: The Art of Winning an Unfair Game



The Pequod Review:

Moneyball is ostensibly a book about baseball, and specifically the Oakland Athletics and its general manager Billy Beane, who in the early 2000s used advanced statistical methods to build a winning team at a fraction of the league-average payroll cost. But actually its implications are much broader, and extend to the inefficiencies of the labor market, the nature of human rationality, the pressures of groupthink and tradition, and the role of luck versus skill.  

Michael Lewis made note of these in a 2012 speech:

I wrote a book called Moneyball.… There are poor teams and rich teams in professional baseball, and they spend radically different sums of money on their players. When I wrote my book the richest team in professional baseball, the New York Yankees, was then spending about $120 million on its 25 players. The poorest team, the Oakland A’s, was spending about $30 million. And yet the Oakland team was winning as many games as the Yankees — and more than all the other richer teams.  

This isn’t supposed to happen. In theory, the rich teams should buy the best players and win all the time. But the Oakland team had figured something out: the rich teams didn’t really understand who the best baseball players were. The players were misvalued. And the biggest single reason they were misvalued was that the experts did not pay sufficient attention to the role of luck in baseball success. Players got given credit for things they did that depended on the performance of others: pitchers got paid for winning games, hitters got paid for knocking in runners on base. Players got blamed and credited for events beyond their control. Where balls that got hit happened to land on the field, for example.

Forget baseball, forget sports. Here you had these corporate employees, paid millions of dollars a year. They were doing exactly the same job that people in their business had been doing forever.  In front of millions of people, who evaluate their every move. They had statistics attached to everything they did. And yet they were misvalued — because the wider world was blind to their luck. 

This had been going on for a century. Right under all of our noses. And no one noticed — until it paid a poor team so well to notice that they could not afford not to notice. And you have to ask: if a professional athlete paid millions of dollars can be misvalued who can’t be? If the supposedly pure meritocracy of professional sports can’t distinguish between lucky and good, who can?

Lewis's book is the thrilling story of how Beane (and crucial predecessors before him, such as Bill James) came to discover these discrepancies and exploit them. But it has a major flaw: he doesn’t really dig into the details to prove how Beane & Co did it. His compelling narrative raises far too many questions that go unanswered: How many more walks did Beane’s Oakland teams generate compared to league average? How many additional expected wins did these walks contribute to the team’s overall record? What players were signed at what cost, and how did they perform in terms of WAR relative to their salary? Sabermetrics persuasively claims that talented relief pitchers should be used earlier in the game in high-leverage situations, and that steals and sacrifices are overrated; how frequently did Beane employ such strategies compared to league average, and what was the impact in terms of expected wins? And so on. Moneyball has too many words, and not enough numbers.

Despite all of this, Lewis has written an exceptional book. And these flaws are only apparent because of how masterfully he deals with the underlying subject. He's just so intelligent that he leaves us wanting more.