from Amazon
Thursday, February 03, 2011
FIP challenge: part 4
I keep saying: there’s no reason to only look at one year of data to forecast future ERA. Why not do what Marcel does and look at the last 3 years of data? How does ERA, FIP, SIERA et al do then? Well, Sky did something like that, and.... FIP won. And FIP was further hamstrung in that all the other metrics were park-adjusted and compared to park-adjusted ERA. FIP was not park adjusted, but was (unfairly) being compared to park-adjusted ERA.
Anyway, among pitchers with at least 400 IP, FIP edged out a win. That is, knowing a pitcher’s HR was a bit better than knowing his batted ball distribution (when compared to SIERA), and better than knowing his FB rate (when compared to xFIP).
The other revelation is how good ole ERA (with all its problems) predicted ERA just as well as anything else out there. It seems that once you have at least 300 IP, there’s no value-added in any of the complicated metrics.
Recent comments
Older comments
Page 1 of 344 pages 1 2 3 > Last »Complete Archive – By Category
Complete Archive – By Date