Predicting ERA

Colin Wyers has done something I meant to do earlier this year: a new and in-depth analysis of the effectiveness of various ERA “component” stats. The study seems sound to me, though I would have created a league-specific FIP add-on for each year (which is what we do here at THT), instead of using 3.2 in all cases.

Colin finds that the “DIPS” estimators (FIP and DIPS) are the best estimators of a pitcher’s ERA in the next year. That’s why we carry FIP in our THT stats. Colin also introduces a new ERA component stat, a FIP/Base Runs hybrid, which appears to be the best ERA predictor of all.