Inferring Team Strengths Using a Discrete Markov Random Field

TL;DR

This paper introduces a novel Markov Random Field model for estimating team strengths over time in sports, utilizing EM and Loopy Belief Propagation, demonstrated on English Premier League soccer data.

Contribution

The paper presents an original discrete Markov Random Field model for inferring team strengths, addressing non-convex optimization challenges in sports analytics.

Findings

Effective inference of team strengths demonstrated on real soccer data

Model captures offensive and defensive dynamics over time

Addresses high-dimensional parameter estimation challenges

Abstract

We propose an original model for inferring team strengths using a Markov Random Field, which can be used to generate historical estimates of the offensive and defensive strengths of a team over time. This model was designed to be applied to sports such as soccer or hockey, in which contest outcomes take value in a limited discrete space. We perform inference using a combination of Expectation Maximization and Loopy Belief Propagation. The challenges of working with a non-convex optimization problem and a high-dimensional parameter space are discussed. The performance of the model is demonstrated on professional soccer data from the English Premier League.