Moneyball: The Art of Winning an Unfair Game, by Michael Lewis, is a book about Billy Beane, general manager of the Oakland Athletics, and how under his management the A's became one of the top teams in MLB on a budget that was a fraction of that of many other teams.
Billy Beane's secret weapon is math. And this puts him at odds with baseball's conventional wisdom. The A's scouts are always second-guessing the work of Beane and his assistant, Paul DePodesta. The scouts travel around the country and recommend players on the basis of what comes down to good looks and wishful extrapolation. Beane and DePodesta, on the other hand, have accumulated massive amounts of data about potential players. They can characterize every aspect of a player's performance and they know how to put a price on it. But they can't explain this to the old guard:
No one in big league baseball cares about how often a college player walks; Paul cares about it more than just about anything else. He doesn't explain why walks are important. He doesn't explain that he has gone back and studied which amateur hitters made it to the big leagues, and which did not, and why. He doesn't explain that the important traits in a baseball player were not all equally important. That foot speed, fielding ability, even raw power tended to be dramatically overpriced. That the ability to control the strike zone was the greatest indicator of future success. That the number of walks a hitter drew was the best indicator of whether he understood how to control the strike zone. Paul doesn't say that if a guy has a keen eye at the plate in college, he'll likely keep that keen eye in the pros. He doesn't explain that plate discipline might be an innate trait, rather than something a free-swinging amateur can be taught in the pros. He doesn't talk about all the other statistically based insights— the overwhelming importance of on-base percentage, the significance of pitches seen per plate appearance— that he uses to value precisely a hitter's contribution to a baseball offense. He doesn't stress the importance of generalizing from a large body of evidence as opposed to a small one. He doesn't explain anything because Billy doesn't want him to. Billy was forever telling Paul that when you try to explain probability theory to baseball guys, you just end up confusing them.
The scouts (and the management of every team except the A's) were looking for personalities. They wanted players who were power hitters and would steal bases (and were, did I mention, good looking?), even though stealing bases doesn't win games. They looked at numbers like RBI and saves, even though RBI doesn't win games. Beane and DePodesta had sophisticated models and they attempted to maximize one thing, the one thing that actually matters: expected runs per dollar, and hence, games won per dollar. So Beane was highly effective in trading away his star (read: overvalued) players and replacing them with unknowns (read: undervalued) who were nearly as good. He knew what traits were important and what deficiencies could be ignored. (Good thing, because, the A's budget being what it was, Beane often had to settle for what were superficially "damaged goods".)
If you can picture all this, consider that Beane managed the team's training and strategy the same way he chose the draft picks. Under Beane's management, the A's had a phenomenal 2002 season, including a 20-game winning streak.
I don't even like baseball (full disclosure: I do like math), and I found this book riveting. Michael Lewis lays out the story very well— even though the book centers around statistics, there is no shortage of the human element here. Moneyball is an easy read, and highly recommended.
(Moneyball is being made into a movie, starring Brad Pitt as Billy Beane!)