Birnbaum: The OBP/SLG regression puzzle
From SABR member Phil Birnbaum at Sabermetric Research on May 29, 2013:
I ran regressions in the previous posts to predict winning percentage from on-base percentage and slugging. In those regressions, I had adjusted all teams to the league SLG and OBP. I had to. If you don't adjust, the results vary a lot.
Here's the regression completely unadjusted. (It's all teams from 1961 to 2009, except strike seasons.) Here's the equation. (I'll put the OBP/SLG coefficient ratio in brackets too.)
wpct = (2.19*OBP) + (0.07*SLG) - .2405 [ratio: 30]
That's an OBP/SLG ratio of over 30! We were expecting 1.7. It seems like slugging barely matters at all!
Compare that to the "regular" regression, which adjusts for league-season:
wpct = (2.70*OBP) + (0.89*SLG) - .7843 [ratio: 3]
OK, that's a bit better. The ratio is down to 3.
Read the full article here: http://blog.philbirnbaum.com/2013/05/the-obpslg-regression-puzzle-part-iii.html
This page was last updated May 29, 2013 at 2:25 pm MST.