Hierarchical Bayesian Bradley-Terry for Applications in Major League Baseball

Authors: 
Gabriel C. Phelan John T. Whelan
Journal: 
Mathematics for Applications
Volume: 
Page number: 
Publication date: 
December 16, 2017
E-print ID: 
arXiv:1712.05879

A common problem faced in statistical inference is drawing conclusions from paired comparisons, in which two objects compete and one is declared the victor. A probabilistic approach to such a problem is the Bradley-Terry model, first studied by Zermelo in 1929 and rediscovered by Bradley and Terry in 1952. One obvious area of application for such a model is sporting events, and in particular Major League Baseball. With this in mind, we describe a hierarchical Bayesian version of Bradley-Terry suitable for use in ranking and prediction problems, and compare results from these application domains to standard maximum likelihood approaches. Our Bayesian methods outperform the MLE-based analogues, while being simple to construct, implement, and interpret.

Home | Links | Credits | Search | Terms of Use | Disclaimer | My Account 

Center for Computational Relativity and Gravitation, School of Mathematical Sciences,
Rochester Institute of Technology,
One Lomb Memorial Drive, Rochester, New York 14623, USA
Phone: (585) 475-7752
Fax: (585) 475-7340
Education - This is a contributing Drupal Theme
Design by WeebPal.