Saturday, October 29, 2005

Probably one of the best things about the White Sox winning the World Series is that it will shut up all the sabremetricians whoe were gloating that Red Sox 2004 Series win was due to GM Theo Epstein's devotion to sabremetrics.

The impulse is understandable. The Oakland A's, the poster boy for the stathead revolution, had enjoyed considerable regular season success but had failed to win a playoff series. Probably the most striking blow came when Derek Jeter, on whose defesive limitations much sabremetric ink has been spilled, nailed Jeremy Giambi, a prototypical sabremetric player (power hitter, low average, takes a lot of walks, no particular defensive position) at home plate with a spectacular defensive play in which Giambi failed to slide. And those with an interest in holding back the sabremetric revolution enjoyed pointing the A's lack of postseason success out.

I think that the theory that the stathead approach has some merit. Plate discipline and waiting for walks and running up a high pitch count on starters works great in the regular season when teams have to play every day and use their entire rotation. But in the playoffs, when you're usually facing a good team's best three starters (say, Andy Pettitte, Roger Clemens, and Roy Oswalt), waiting for a walk probably isn't the best strategy.

The sabremetric answer to these criticisms is that a three games out of five divisional series divisional series is not a large enough sample size on which to judge these tactics. Which is a fair enough point.

But I'm not sure even the statheads believed it. One of the common criticisms the statheads would make of teams is that they make decisions for sentimental reasons that aren't linked to results on the field. For example, the Astros's current contract with Jeff Bagwell, and their continued use of Brad Ausmus as their starting catcher is subject to much criticism, since Bagwell will probably never be a productive player again, and Ausmus is not a good hitter. So, it seems strange that they would continue to back a srategy that had such poor postseason results. The object of baseball is to win the World Series, and the Oakland A's really hadn't come close.

Enter Theo Epstein and the Boston Red Sox. Epstein took the sabremetric approach to Boston and Voila!, the Red Sox won the World Series, and the curse of the Bambino was lifted! Sweet vindication!

Not so fast.

First, the Red Sox had the second highest payroll in the major leagues, had done so for some time, and third place wasn't that close. If that payroll is spent in even a moderately intelligent way, eventually the team is going to break through and win a championship. It doesn't take advanced analysis to know that Manny Ramirez and Johnny Damon can hit, and that Curt Schilling and Pedro Martinez can pitch -- it takes money to sign them.

Second, it's not as if the Red Sox made a quantum leap in success once Epstein took over (as admittedly, the A's did under Billy Beane). They were perennial playoff contenders before; they're perennial playoff contenders now. If the playoffs really are a crapshoot, or too small a sample size on which to judge things, then the Sox 2004 World Series victory of last year shouldn't have changed the analysis.

But now we have the 2005 White Sox, and their manager, Ozzie Guillen. The sabremetric way says the fields manager should pretty much be a cipher -- simply executing the GM's plan. (Moneyball has a passage in which Billy Beane is furious that Art Howe is using a differnt reliever that the one Beane said should be used in a certain situation.) Before the season the Sox traded Carlos Lee, a hulking slugger, for Scott Podsednik, a leadoff hitter who steals a lot of bases.

And they won the World Series. And I'm glad.

It's not that the sabremetric approach is wrong -- I think they're actually correct about most things. On base percentage is critically important.

But too often, they forget that baseball is played by men, not machines, and that these men, like any men, respond to changes in their environment.

The White Sox used a lot of "small ball tactics" during the year. In the playoffs, some analysts pointed out that the White Sox offense isn't built around taking extra bases, it was reliant on the home run.

But that misses the point -- the occasional use of small ball tactics created anenvironment where everybody was expected to do whatever it took to help the team win. If that meant the clean-up hitter would bunt, then the clean-up hitter would bunt.

It was interesting to read Baseball Prospectus analyze the differences between the Sox's expected and actual 2005 performance. Pretty much, the hitters as a whole performed as expected, except everybody played a little better than expected on defense, which helped the entire starting rotation over-achieve. Hmmm -- defense in baseball is a lot about effort and concentration -- couldn't it be due to Guillen creating a winning attitude?

Since I started drafting this, Paul DePodesta was let go as Dodgers GM, after trying to build a team around Heep Sop Choi, Jeff Kent, Milton Bradley, and J.D. Drew.

It's not that I'm happy to see DePodesta go down. But I think it will be good to see the sabremetric crowd approach the 2006 season with a dose of humility.
Post a Comment