Improved Component Predictions of Batting Measures

TL;DR

This paper introduces a random effects modeling approach to improve the estimation and prediction of batting performance measures by decomposing them into component probabilities, applicable also to pitchers.

Contribution

It presents a novel application of random effects models to decompose batting measures into component rates for more accurate estimation and prediction.

Findings

Enhanced accuracy in estimating batting component probabilities.

Improved season-to-season performance predictions.

Applicable to pitcher on-base and FIP measures.

Abstract

Standard measures of batting performance such as a batting average and an on-base percentage can be decomposed into component rates such as strikeout rates and home run rates. The likelihood of hitting data for a group of players can be expressed as a product of likelihoods of the component probabilities and this motivates the use of random effects models to estimate the groups of component rates. This methodology leads to accurate estimates at hitting probabilities and good predictions of performance for following seasons. This approach is also illustrated for on-base probabilities and FIP abilities of pitchers.