Looking for the perfect comp
1
|
The Hardball Times found this 7/16/2007 on www.hardballtimes.com [flag] |
Tags:
MLB
Mark Buehrle
Comments (4)
-
justbob898 Buehrle through 2006 (age 27): W-L 97-66, IP 1428.0 ERA 3.83 (ERA+ 123), BB/9 2.07, K/9 5.22.
Glavine through 1993 (age 27): W-L 95-66, IP 1357.0, ERA 3.45 (ERA+ 114), BB/9 2.88, K/9 4.96.
Close enough for government work. I don't know why others have missed this comp.
-
tango tiger At the bottom of the Lichtman blog, I provided a couple of different comp-sets. I agree with John that you can pretty much get anything you want.
In my case, I provided one set of comps (post 34 or 36, I think) that would exactly match the 4/57 deal that Buehrle got.
-
Chris Jaffe John, good points. I probably would've been a bit more careful, but his K's and walks have always struck me as his defining characteristics more than his homers. I thought about cofining the search just to lefties, but it narrowed the sample size too much for my tastes.
I also liked this group because I was struck by what seemed to me to be a clear pattern: those guys all had a rocky season in their late 20s and recovered.
Really, what one would need to do would be to compare a pitchers's W/9, K/9, HR/9 or whatever else to league average for when he played, but that was far more work than I could do when writing this.
Links (1)
THT: Jaffe: Looking for the perfect comp
Published 7/16/2007 at BBTF's Baseball Primer Newsblog
THT: Jaffe: Looking for the perfect comp Chris Jaffe takes a more in-depth look at the pitchers being compared to Mark Buehrle in the wake of his contract extension. Unfortunately, it turns out most of the guys in that study were the long deceased. Only a third were liveballers with the remainder fairly evenly divided between deadballers and pre-1893-that s-hardly-even-pitching guys. I mean, are Charlie Buffinton and Jim McCormick really good comps for Mark Buehrle?
I get the feeling that these comparision studies (and I've done a few myself) can spit out just about any answer depending on the precise criteria you use to build your comps. You wrote:
Figure out what his main traits are, find guys with similar characteristics, and see how they aged.
Sure, but what if I don't agree with your main traits? I might choose left-handedness as a trait. Or pitching in the high-scoring 2000's. Isn't HR/9 an important trait? And even if I agree with IP, SO/9 and BB/9 (which of course are quite reasonable), I might make different cuts which would yield a different set of pitchers and, possisbly, a different conclusion.
I can't avoid the feeling that if you asked 10 smart analysts to do this study, they would come up with 10 different comp lists and 10 different answers.
What we really need to do is show that this method actually works. It could be tested, of course. Put yourself in 1980 and do an aging study based on comps for Rick Langford, Scott McGregor, Pat Dobson, etc.