MONEYBaRL: Exploiting pitcher decision-making using Reinforcement Learning

TL;DR

This paper introduces MONEYBaRL, a reinforcement learning approach that models baseball pitchers' decision-making process as a Markov Decision Process to exploit their 'Baseball IQ' for strategic advantage.

Contribution

It presents a novel MDP-based framework for modeling pitcher decision-making, enabling analysis and potential exploitation of their strategic choices.

Findings

Model effectively captures pitcher decision patterns

Potential to improve batting strategies against pitchers

Provides a new tool for sports analytics and strategy

Abstract

This manuscript uses machine learning techniques to exploit baseball pitchers' decision making, so-called "Baseball IQ," by modeling the at-bat information, pitch selection and counts, as a Markov Decision Process (MDP). Each state of the MDP models the pitcher's current pitch selection in a Markovian fashion, conditional on the information immediately prior to making the current pitch. This includes the count prior to the previous pitch, his ensuing pitch selection, the batter's ensuing action and the result of the pitch.