Regressions: 2008 MLB ERA -> Wins

Wednesday, February 04, 2009

It's February, so I'm Trying to Find things to Post On. Therefore You're Going to Get Some Simple Regression Tests from the 2008 MLB Season.


The Major High Outlier: Texas with an ERA of 5.37 an 79 Wins
The Low Outlier: Toronto with an ERA 3.49 and 86 Wins

Formula: ERA = 6.5246 + (-0.02728) * Wins
ReFangled Formula: Wins = (ERA - 6.5246) / (-0.02728)

So let's have some fun. Say I want the Yankees to win 100 games next year? Based on last years regression test I expect that they'll need somewhere around a team ERA of 3.8. Meanwhile if I think the Oreos are going to have an ERA close to 5.20, because their starters are miserable, than based on last years stats I can expect them to win about 48 games.

Posted by Simon at 3:33 PM   Digg! submit to reddit BallHype: hype it up!

0 comments:

Post a Comment

Advertise Here!