Friday, September 16, 2011

Why "Moneyball" Makes no Sense

John Sickels, a former Bill James staffer, does a review of the results from Oakland's "moneyball draft" in 2002 on his Minor Leagues blog. His conclusion "Overall, the Moneyball draft wasn't a giant success, but it wasn't a massive failure, either."

Oakland had nine choices in the top 100 that year and got three successful major league players out of that, Nick Swisher, Mark Teahan and Joe Blanton. By contrast, the Twins had three choices in the top 100 and ended up with Denard Span and Jesse Crain. What seems clear is that, despite the hype, Billy Beane was not on to something new in his approach to drafting. And that really is no surprise.

The basic idea of "moneyball" is to treat players like the marketplace. This was the spin that got the attention of the Wall Street baseball fans. Yeh, they were smarter than the rest of the world and now there was a baseball GM who really got it. Beane's strategy was to draft players with traits other teams undervalued, much the same way stock traders try to buy stock that other traders undervalue. Unfortunately the baseball draft isn't a marketplace and the strategies that may be appropriate there have little, if any value.

Understand that every player drafted probably is likely rated at least as high by the team that drafted him as the teams ahead of him. This is because each team takes the best player on their list when their turn comes around, at least for the top picks in the draft. Occasionally someone might pass on a player because of money concerns, so it may be more accurate to say they pick the best player they think they can sign. They key thing to understand is at the point a decision is made, a team is trying to take the best player left on their list.

Whatever effect there is from the criteria other teams use to rate players is a given. The other teams passed on all the players still available based on whatever criteria they used. For the "moneyball" team the only question is who is the best player by THEIR own criteria, regardless of how other teams might evaluate that player. So it may be that college players are undervalued by other teams in the draft. But all that means is a college player is more likely to still be available. That they are undervalued by others tells you nothing about which player you should take.

Of course, you could decide that ANY college player is better than ANY high school player. The idea would be that you would ignore your own evaluations with the idea that they must be wrong. If all the teams ahead of you passed on a high school player, you must have them overvalued if you think they are better than the best available college player. I suppose,in effect, that's what Billy Beane did. So it would be interesting to see who was the first high school player drafted after each Oakland choices in the first round:

Nick Swisher - Cole Hamel (Phillies)
Joe Blanton - Matt Cain (Giants)
Ben Fritz - Greg Miller (Dodgers)
Jeremy Brown (35), Steve Obenchain (37,) Mark Teahan (39) - Michah Shilling (Indians-41),Blair Johnson (Pirates 42). Jason Pridie-(Rays-43)

I think I would take Hamel over Swisher and Cain over Blanton. But obviously you would have to take Mark Teahan over Jason Pridie.

Regardless of how you look at it, the basic idea behind Moneyball doesn't really work. And if you shorten it down to "never draft high school players", it is obviously absurd. Just imagine the A's with Cole Hamel and Matt Cain added to their rotation.

No comments:

MLB Twins Updates