In 2013, the opening day payroll for the New York Yankees was nearly $230 million USD. Only the Dodgers were in the rarified air above $200M with them. The eventual winner of the 2013 World Series were the Boston Red Sox. Boston achieved this success for a paltry $159M—much less than the Yankees, but still the fourth largest opening day payroll in Major League Baseball. The Red Sox’s opponent in the World Series, the St. Louis Cardinals, were outspent by New York and the Los Angeles Dodgers by more than $100M. As a Red Sox fan, I can't throw stones. We've had some very expensive, very bad years recently. The point about the Red Sox and the Yankees is that if they haven't been getting enough bang for their buck because aside from the essentially random events that influence baseball—injuries, weather, etc.—if their theories for winning were equal to those of other teams, they would win many more games and championships than they do. For many years the dominant theory was that better players lead to better outcomes. But everything from pre-season team-building exercises to pre-game warm ups to individual superstitions are theories that a team and its members make use of to achieve their goals. These theories can be well or poorly founded, but they are nonetheless theories.Each year between 1999 and 2013 the Yankees paid the most for their players. Indeed, one of the theories that seemed to operate in the Yankees organization was that aggressive pursuit of the most expensive free agents in MLB would result in success. But in those 14 years, this strategy—this theory—yielded only three World Series wins—a considerably lower success rate than they have had since the days of Babe Ruth’s migration to the Bronx in the 1920s. Not only have the Yankees been experiencing a World Series drought, their win percentage is also not as high as one would expect. One of the most famous cases of superior theory building in sports is from baseball and comes to us from Michael Lewis’ Moneyball. Lewis has a knack for identifying the theory-building process as it plays out (pun intended) in sports—he unlocks the secrets of emerging theories in sports and looks to gaps in existing theories and innovative new theories that yield better results. In Moneyball Lewis tells the story of how one manager, Billy Beane of the Oakland A’s—whose team is perpetually under resourced compared to the Red Sox, Yankees, Dodgers and most every other team in the league—has figured out new ways to win. The A’s under Beane’s leadership have consistently outperformed their spending. Over the years, Oakland has consistently been able to find undervalued players for its teams. Beane (and others in the organization, most notably John DePodesta) used a theory that violated the instincts and practices of baseball scouts, by relying heavily on statistics that were either uncommon or not valued. Put simply, the traditional theories of baseball scouts had a lot to do with how a player looked and acted. Beane and the A’s acted on a theory that looked into the data about player’s history to decide whether he was worth hiring. The A’s theory has proven quite effective over the last decade or so. In fact, it’s not just that they figured out a particular strategy, their governing theory is that they can always find the next statistic(s) that will enable them to hire undervalued players and later sell off those players at a premium—sort of a buy low sell/high strategy. The graph below clearly shows that despite the huge gap in payroll, the winning percentages of the the Yankees and the A’s are not greatly different. However, the A’s still have been unable to lay claim to to a World Championship—maybe they need an even better theory.Finding a better theory is often the most durable strategy available in sports, often—as with the A’s—by finding undervalued players, which requires looking at data differently or finding new data. In basketball, Shane Battier has recently received a great deal of attention following an article in the New York Times Magazine by—who else?—Michael Lewis. There, Lewis starts, as any good investigator should, with a puzzle (what we will later refer to as an anomaly).“Here we have a basketball mystery: a player is widely regarded inside the N.B.A. as, at best, a replaceable cog in a machine driven by superstars. And yet every team he has ever played on has acquired some magical ability to win.”This mystery spurs deeper investigation. Because there are data that purport to assess the quality of a player—and because we can track the valuation of players in a fairly straightforward way, namely by how much they are paid—until recently, there was no widely used theory that reflected what Shane Battier does for an NBA team. One of the false lessons that people can learn from Shane Battier and the Oakland A’s is that the explosion of data is what is leading to winning. But that’s a dangerous half-truth that we’ll address in more detail later. For our purposes here, what matters is that data alone is meaningless. Smart researchers—including those employed by sports teams—use rich data sets as, in a sense, observations that allow managers and players to develop more effective theories. Bill James the famed father of sabermetrics was not satisfied with the vast troves of data that baseball has collected for decades. Rather, he sought to understand which statistics actually contributed to winning—he was looking to the data to find better theories. Without his drive to understand the causes of winning, the data available to him would have made little difference, except there would just be a lot more undifferentiated, meaningless numbers. A good example of how data without a theory is useless is in basketball. For many years basketball has gathered data, mostly the stuff that appears in a box score. But those data had, as serious scientists have shown, not been terribly useful. They were, more often than not, measuring the wrong things—the common theories in basketball lacked the process of thinking we will describe in this book. Consequently, evaluation of basketball players languished in the realm of modest rules of thumb.Returning to the example of Shane Battier, now of the Houston Rockets, we see that the Houston Rockets organization has found a superior theory for assessing the value players add to a team’s winning, and that Battier himself behaves like a scientist—adhering to the dictates of sound theory in his play. He is constantly putting into actions theories supported by data—regardless of whether the theory of success against an opponent conforms to his intuitive sense or not. “The numbers either refute my thinking or support my thinking,” he says, “and when there’s any question, I trust the numbers. The numbers don’t lie.” Even when the numbers agree with his intuitions, they have an effect. “It’s a subtle difference,” [Daryl Morey, General Manager of the Houston Rockets] says, “but it has big implications. If you have an intuition of something but no hard evidence to back it up, you might kind of sort of go about putting that intuition into practice, because there’s still some uncertainty if it’s right or wrong.”Knowing the odds, Battier can pursue an inherently uncertain strategy with total certainty [that it’s better than the alternatives]. He can devote himself to a process and disregard the outcome of any given encounter.Battier accords an appropriate amount of belief to the odds he’s got. So, on the one hand, there are often no dominant strategies—that is, often there’s no way to shut down an opponent’s scoring completely. Moreover, there’s no certainty that pursuing the best strategy will work every time. But Battier can know for certain that there is a best way to defend another player. What’s more, as these theories have evolved, Battier and management had begun to discover under which conditions forcing an opponent to his left is better than forcing him right, or when it is best to box out your teammate’s man rather than your own. The interesting thing about what the Houston Rockets did with Battier and continue to do now is that they are delving deeper into these formerly unknown or at best mysterious facets of the game. By creating robust theories and acting on them—testing them—they follow the fundamental cycle of designing and testing theories against the real world.