Saturday, October 17, 2009

Friday baseball post

OK, I'm late again (seems to be Saturday morning), but just as well, because instead of talking about the many problems with Brian Sabean I need to talk, instead, about Moneyball, Billy Beane, and an ugly hit-piece by Buzz Bissinger that the New Republic is featuring today.

Bissinger begins with the accurate observation that the four remaining teams in the 2009 postseason are high payroll clubs. No question about that -- in fact, three of the four teams are from the two largest markets in the nation, and Philly ain't chopped liver. He concludes from this that it marks "the effective end of the theories of Moneyball as a viable way to build a playoff-caliber baseball team when you don't have the money."

Except...well, his article is basically a mess. Let's start with "the theories of Moneyball." Bissinger seems to have read the book and sort of understood it:
The basic thesis of the book--the finding of inefficiencies in the marketplace through expert computer analysis--no doubt resonated. As Lewis told it, what Beane and his minions did was usher in the baseball equivalent of a new period of painting, the Age of On-Base Percentage.
Yes, Moneyball is about "finding inefficiencies in the marketplace." That was, in fact, how Beane was successful. But the rest of this conflates different things. One is how to find market inefficiencies. Beane did it not so much through "expert computer analysis" but by performance analysis. Traditional scouting projected future player performance by assessing the skills (and personality) of players; Beane, following Bill James and others, projected future player performance by looking at what players actually did on the field. As it turns out, both of these methods are useful. But in the early Beane era, most teams only used one of them, and therefore players who didn't look good to scouts but had done well on the field were undervalued. Now, that's just one bit of it. Generally, Beane and his staff were good at looking around and noticing what useful skills and traits were going at a premium, and which were undervalued. Most famously, I guess, were (as Bissinger discusses) college pitchers were undervalued in the amateur draft compared to high school pitchers, and among hitters getting on base was undervalued compared to other skills.

Where Bissinger goes wrong is that Beane never believed -- and the Moneyball thesis certainly doesn't rest on -- the idea that scouting is worthless, college pitchers are better than high school pitchers, or OBP is more valuable than other skills. All that Moneyball says is that if a GM can identify one of these market mistakes, he can exploit it -- and that doing so is a big part of being a successful GM. But market mistakes are unlikely to last. Once Beane was successful, other teams copied what he had done, and so the market changed, and the old undervalued things were no longer available. Teams that had traditionally avoided college pitchers in the draft changed their behavior, and soon college pitchers, as a group, were valued correctly. No more market inefficiency. But others were there. The Houston Astros had great success for a while drafting short pitchers, and one of the greatest draft coups of the last decade -- the Giants selection of Tim Lincecum at #10 overall in the 1986 draft -- was available because of a general bias against small pitchers.

(I should note, however, that Bissinger is wrong about OBP, which is in fact just as important as ever, and about high-priced closers, who are just as unreliable and overpriced as ever. Sure, it's nice to have Mariano River, but he's one of the best two or three relief pitchers ever. Look at the World Series winners since the Yankees last won in 2000, and you get lots of rookies and one-year wonder closers).

It may well be true that it's harder to find major market inefficiencies now than it was a decade ago. And that gets back to the "expert computer analysis" part of the equation, and why Billy Beane really did change baseball, or at least represent a major change. What was important about Beane, ultimately, is that he was willing to ignore conventional wisdom and look at things objectively. That sounds easy -- but traditional baseball management just didn't do that. Baseball wisdom was supposed to be acquired from sitting in dugouts and hanging out with baseball people, with traditions and the way things were done passed on from generation to generation. Outsiders, and outside types of analysis, were supposed to be useless. Beane showed that that was wrong. Anyone could sit down and look at the history of the amateur draft and compare the results of pitchers drafted out of high school with the results of pitchers drafted out of college and discover that the latter were undervalued. And so the biggest market efficiency that Beane had over much of the rest of baseball, for a while, was that he was willing to use outsider knowledge and others were limited to insider knowledge.

And then Beane won, and everyone surrendered. The symbol of all that was the Boston Red Sox. Not only did they actually hire Bill James, but their boy genius General Manager, Theo Epstein, is a Yale-educated guy who didn't actually play much baseball, himself. Boston is a big market -- but the Red Sox were copied, and last year surpassed, by the tiny-market Tampa Bay Rays. In fact, most (all?) teams now employ sabermatricians, and GMs are far more educated and a good deal less steeped in baseball tradition than they were a a decade ago.

What of Billy Beane, himself? It's really hard to tell. Beane certainly got the idea of finding undervalued players. But there's a big difference between understanding that, and actually finding those undervalued players -- and yet another big difference between doing it in one era and then being able to once again find the next type of undervalued players. Beane shifted, in the last several years, from OBP to fielding (at least according to the accounts I've read, mostly at Baseball Prospectus), and has had mixed results. Bissinger is right that J.P. Ricciardi was largely a flop in Toronto, although the story of Paul DePodesta in Los Angeles is a lot more complicated than he makes it out to be (DePodesta is responsible for a lot of the talent on the current Dodgers, which makes Bissinger's case a lot weaker). As for Beane, I wouldn't count the A's out yet.

Last bit: is Bissinger right about rich teams and poor teams? Well, this year is certainly a year of big market success. That's also fairly rare, lately; we've had plenty of teams from Denver, Tampa, Miami, and Phoenix in the World Series in the last decade. Parity doesn't mean that New York never wins! I certainly don't think that a single year is proof that exploitable market inefficiencies have disappeared, or that only large market teams are now able to exploit them.

Baseball people were threatened, and rightly so, by Billy Beane and by Michael Lewis. The idea that some nerd armed with a fancy computer program, or some kid fresh out of Wharton, might know things that "baseball people" didn't know, must have been quite scary. Perhaps that's why we've had a succession of comments and articles over the last several years knocking it down. Lewis has moved on, Beane has moved on, sabermatricians have moved on (Baseball Prospectus long ago has moved to reconcile the kinds of things scouts can know with the kinds of things that show up in the stats). I suppose, however, that we'll be hearing from the backlash crowd for years to come. Sad, really.


  1. Great analysis. Dinosaurs didn't die in a day, though, so the Bissingers will be with us for while.

  2. Yup. It is depressing, though...if I look up from my desk I can see my collection of Abstracts, beginning with the 1982 edition (the first "real book" one. Which is, if I check my calendar, 27 years ago.

    BTW, I don't think I mentioned it above, but I really hate the Dodgers.


Note: Only a member of this blog may post a comment.

Who links to my website?