Moneyball

I just finished reading Michael Lewis' book Moneyball. I enjoyed this book even though I know next to nothing about baseball. This is the amazing story of the Oakland Athletics written circa 2002. At that time the Oakland As raised eyebrows because they were one of the lowest budget teams in Major League Baseball, yet, had demonstrated remarkable success. Their accomplishment was dismissed by many as being the product of luck. The book goes on to describe how the ascendancy of the As was the result of a superior strategy at their front office which built predictive models of team wins based on player statistics which were being overlooked by other teams. This allowed them to zero in on the deeply undervalued players, and trade their own overvalued players to build cash reserves and draft pick opportunities. The Oakland As were, in essence, a baseball hedge fund. Two comparisons to things, I have had some connection with in my work, struck me as interesting:

1. It's odd how 100-million dollar businesses -- namely the storied baseball teams -- are run turning a blind eye to correctly valuing their most important assets. In this milieu, a small-pocketed upstart -- in this case a low-budget baseball team --earns disproportionate success by arbitraging these discrepancies. The parallels with statistical arbitrage in hedge funds are striking.
2. Much of the analysis that underlies the Oakland As' predictive models may be traced to the work of pioneering baseball newsletter writer Billy James. James discovered novel statistics to use to value baseball players, instead of the measures that had been inherited from a time when the game was fundamentally different -- for example when fielders were folks on leave from the army who played without mitts. He was faced with poor, inadequate data that was being sold for rip-off prices by established data gathering firms. Pricing aside, these firms were not interested in changing their ways and making a truly useful product offering that was being sought by this new breed of enthusiasts. As a result there arose a small, underfunded volunteer base to gather such statistics. This, to me, is the baseball equivalent of the present-day open-source phenomenon.

One other thing that struck me, upon reading the book, is the distasteful trading of players by the teams. It has the scent of another era. It is as though the players are like cattle that can be freely traded for one another, and sometimes even for cash. The tactics that accompany these transactions are, often, Machiavellian. Oh well, to paraphrase Bismarck, baseball teams are like sausages; it's better to not see them being made.