tag:blogger.com,1999:blog-12133335.post5952277349131619497..comments2014-05-29T00:31:30.902-04:00Comments on Walk Like a Sabermetrician: Leadoff Hitters, 2009phttp://www.blogger.com/profile/18057215403741682609noreply@blogger.comBlogger5125tag:blogger.com,1999:blog-12133335.post-20763930348067657702010-02-19T13:57:18.788-05:002010-02-19T13:57:18.788-05:00I just realized that if you substitute 2(or 1.8 or...I just realized that if you substitute 2(or 1.8 or 1.6)*OBP +SLG for wOBA, then wOBA*OBP/SLG becomes 2*OBP^2+OBP, something that will rank hitters no differently than OBP. If, however, you use OBP-SLG instead you get something a little more interesting: 2*OBP^2-OBP*SLG-SLG^2Marchttp://www.blogger.com/profile/08473194081491054706noreply@blogger.comtag:blogger.com,1999:blog-12133335.post-12695527285965542922010-02-19T13:24:43.477-05:002010-02-19T13:24:43.477-05:00Marc, it's true that 2*OBA + SLG approximates ...Marc, it's true that 2*OBA + SLG approximates wOBA, but Tango's work and my own have both indicated that 1.6-1.8 times OBA is the best value. Assuming that is the case, using 2 is a very slight extra weight on OBA. I agree, though, that it is something of a general hitter method.<br /><br />OBA/SLG ratio is a similar concept to the Run Element Ratio--I wouldn't be crazy about it because of the different denominators. wOBA*OBA/SLG is interesting, but my personal preference is for metrics with easily explicable units.phttp://www.blogger.com/profile/18057215403741682609noreply@blogger.comtag:blogger.com,1999:blog-12133335.post-84102175931613095762010-02-19T09:46:38.141-05:002010-02-19T09:46:38.141-05:002*OBP+SLG is mentioned in THE BOOK as a good estim...2*OBP+SLG is mentioned in THE BOOK as a good estimator of wOBA. So using that or Runs Created is simply a general measure of hitter production, not specific to leadoff. I like to think in terms of hitters that derive a larger proportion of their output from OBP rather than slugging so OBP/SLG ratio or wOBA*OBP/SLG might be a better stat for choice of leadoff man.Marchttp://www.blogger.com/profile/08473194081491054706noreply@blogger.comtag:blogger.com,1999:blog-12133335.post-52470506547659325762009-12-16T23:36:02.900-05:002009-12-16T23:36:02.900-05:00I'm defining outs made in this case as AB-H+CS...I'm defining outs made in this case as AB-H+CS, and the average outs/game when outs are defined in that manner is right around 25.5 (it was actually 25.43 in the AL and 25.38 in the NL this season).<br /><br />So using 25.5 outs puts it on roughly the same scale as team runs scored/game.phttp://www.blogger.com/profile/18057215403741682609noreply@blogger.comtag:blogger.com,1999:blog-12133335.post-5587301516651592082009-12-16T15:03:05.392-05:002009-12-16T15:03:05.392-05:00for the uninformed here, mostly me, why 25.5 outs?...for the uninformed here, mostly me, why 25.5 outs?Ron Rollinshttp://www.blogger.com/profile/16852012772573977515noreply@blogger.com