Bayesian estimation of in-game home team win probability for Division-I FBS college football

TL;DR

This paper develops a Bayesian method to estimate in-game home team win probability for Division-I FBS college football using expected possessions and scores, demonstrated on a championship game.

Contribution

It introduces a novel Bayesian approach tailored for college football, incorporating specific game predictors and comparing different models for in-game win probability estimation.

Findings

Effective Bayesian models for college football win probability

Application to 2021 Big XII Championship game demonstrates utility

Comparison of predictors and modeling approaches

Abstract

Maddox, et al. [9, 10] establish Bayesian methods for estimating home-team in-game win probability for college and NBA basketball. This paper introduces a Bayesian approach for estimating in-game home-team win probability for Division-I FBS college (American) football that uses expected number of remaining possessions and expected score as two predictors. Models for estimating these are presented and compared. These, along with other predictors are introduced into two Bayesian approaches for the final estimate of in-game home-team win probability. To illustrate utility, methods are applied to the 2021 Big XII Conference Football Championship game between Baylor and Oklahoma State.