At its core, I find the Moneyball hypothesis offensive. I tend to think that, as a general matter, labor markets work quite well, and returns to skill are valued appropriately. But the Oakland example was in opposition to my belief in labor market efficiency. So my colleague Jahn Hakes and I decided to investigate it more in a paper that will be coming out in the Journal of Economic Perspectives. We found that Lewis’ offensive idea was correct. On-base percentage was undervalued, and buying on-base percentage went a long way toward explaining Oakland’s success.
How do we explain this? I think what Lewis found was a very clear-cut example of institutional inertia. A lot of old baseball scouts had a certain idea of which skills made for a good baseball player — and those weren’t necessarily right. Yet once those ideas took hold, they tended to stick. Someone eventually questioned and tested them, and decided there was another way to evaluate talent. Beane was a real innovator, and he was able to exploit the opportunities that he saw. But it’s very hard to do this over an extended period of time. This information will be exploited by others — indeed, we have seen it recently with several other teams. Just about every front office in Major League Baseball has guys poring through data looking for statistical patterns that can give them an advantage. As a result, there will be new innovations in assessing talent that might prove even more effective.
Lots of good stuff.