Trouble With The Curve: Improving MLB Pitch Classification

TL;DR

This paper introduces a new model-based clustering method using Gaussian mixtures and an adjustment factor to improve MLB pitch classification accuracy and interpretability, addressing limitations of previous manual and neural network approaches.

Contribution

The paper presents a novel clustering algorithm and classification method for pitches, enhancing accuracy and interpretability over existing techniques.

Findings

Effective classification of diverse pitch types

Improved accuracy over traditional methods

Applicable to various pitchers and pitch styles

Abstract

The PITCHf/x database has allowed the statistical analysis of of Major League Baseball (MLB) to flourish since its introduction in late 2006. Using PITCHf/x, pitches have been classified by hand, requiring considerable effort, or using neural network clustering and classification, which is often difficult to interpret. To address these issues, we use model-based clustering with a multivariate Gaussian mixture model and an appropriate adjustment factor as an alternative to current methods. Furthermore, we describe a new pitch classification algorithm based on our clustering approach to address the problems of pitch misclassification. We illustrate our methods for various pitchers from the PITCHf/x database that covers a wide variety of pitch types.