Sabermetric Research

I had forgotten to mention that Phil Birnbaum, editor of SABR's By the Numbers newsletter, has a new blog simply called "Sabermetric Research" where he includes "Links to and reviews of sabermetric studies and sports research". I see that Phil's including some interesting studies on sports other than baseball as well.

To me this is a particularly welcome addition to the baseball blogosphere since I've written before that what I think is needed at this point in the performance analysis revolution is a clearinghouse where researchers are able to search and access all previous studies on a particular topic. At one point I had started laying the groundwork for such a site but then of course fell into other pursuits. At this year's SABR Statistical Analysis Committee meeting there was some talk on this very topic and I believe several members volunteered to get the ball rolling. I for one, would be a big supporter.

But back to Phil's blog...

One of my favorite posts thus far was this one where he cites a study that discusses the true boost in attendance from interleague play. The gist is that while MLB touts a 13% increase, the real increase is along the lines of 5% once you factor in the months in which interleague games are played, the prevalence of weekend matchups, and the boost from the inaugural 1997 season. It had occurred to me this was likely the case but never had the gumption to break it down.

I also enjoyed the post where Phil summarized the work of previous studies related to hit batsmen. That's a subject near and dear to my heart since I took a look at the historical trends in a series of three articles earlier this summer on BP. What I find most interesting about the subject are the three trends; a) the increasing number of hit batsmen over time, b) the difference between the leagues from just before the introduction of the DH to 1993 and c) the way in which the rates have evened out across the leagues since that time. I look at all three in my series and conclude:

...from an initial difference of nearly 21% in the rate of hit batsmen between the two leagues in the 1973-1993 period, just over 7% can be accounted for by the presence of more true hitters in the lineup and another 4% by two hitters [Don Baylor and Chet Lemon both of whom played in the AL] who were exceptionally "gifted" at getting plunked. This still leaves ample room for the moral hazard theory, a theory that incorporates differences in the two leagues relating to strike zone or styles of play, or a combination of all of the above to operate.

In any case, it's a fascinating subject that provides ample room for speculation and the proposal and testing of various theories.

Finally, I (like many others) enjoyed both posts related to attempts to measure the improvement of players over time. In other words, attempting to answer the question of just how good a Babe Ruth or a Rogers Hornsby would be, given the skills they possessed when the played, in the modern game. The relevance is that a chapter of Baseball Between the Numbers includes this very discussion written by Nate Silver. There Silver creates a league difficulty factor based on Davenport Translations by examining the performance of players in successive seasons. He then uses these factors to translate statistics across eras. His analysis of Ruth concludes as follows:

Ruth's career EqA would be .274. He probably would have made the All-Star team a couple of times, with an EqA in his best seasons approaching .300. But he'd be remembered as merely a good player and certainly wouldn't be a credible candidate for the Hall of Fame. In modern terms, Ruth might be a Tino Martinez (career .274 EqA) or Raul Mondesi (.278).

Birnbaum seems to think that studies like this are inherently flawed:

But can you design an experiment, like [Dick] Cramer tried to do, that will find an answer without looking to physics? I can't find the reference, but I'm pretty sure Bill James once speculated that there's no way to do it. I think I agree.

His view is that performance data (not data from the world of physics) is so intertwined with variations in the underlying difficulty level of the sport that one likely cannot devise a study that disentangles them. I'm not so sure but will have to ruminate on it for awhile. I'm of the opinion that Silver is essentially correct but these discussions have sent me back to the drawing board in terms of thinking about we could measure it.