Using ELO ratings for match result prediction in association football

Sports betting markets are becoming increasingly competitive. These markets are of interest when testing new ideas for quantitative prediction models. This paper examines the value of assigning ratings to teams based on their past performance in order to predict match results in association football. The ELO rating system is used to derive covariates that are then used in ordered logit regression models. In order to make informed statements about the relative merit of the ELO-based predictions compared to those from a set of six benchmark prediction methods, both economic and statistical measures are used. The results of large-scale computational experiments are presented.

[1]  L. Knorr‐Held.  Dynamic Rating of Sports Teams , 2000 .

[2]  Leighton Vaughan Williams,et al.  Forecasting outcomes in spread betting markets: can bettors use 'quarbs' to beat the book? , 2005 .

[3]  M. Maher.  Modelling association football scores , 1982 .

[4]  Ian H. Witten,et al.  Data mining: practical machine learning tools and techniques, 3rd Edition , 1999 .

[5]  G. Brier.  VERIFICATION OF FORECASTS EXPRESSED IN TERMS OF PROBABILITY , 1950 .

[6]  John Goddard,et al.  Regression models for forecasting goals and match results in association football , 2005 .

[7]  Herman Stekler,et al.  Are sports seedings good predictors?: an evaluation , 1999 .

[8]  H. Stekler,et al.  P redicting the outcomes of National Football League games , 2003 .

[9]  I. D. Hill.  Association Football and Statistical Inference , 1974 .

[10]  M. B. Wright,et al.  50 years of OR in sport , 2009, J. Oper. Res. Soc..

[11]  Mark J. Dixon,et al.  Dynamic modelling and prediction of English Football League matches for betting , 2002 .

[12]  I. Graham,et al.  Predicting bookmaker odds and efficiency for UK football , 2008 .

[13]  F. Galton.  Vox Populi , 1907, Nature.

[14]  David Johnstone,et al.  Economic Darwinism: Who has the Best Probabilities? , 2007 .

[15]  L. Goldman,et al.  Predicting outcomes , 1996, Journal of general internal medicine.

[16]  B Jennett,et al.  Predicting the Outcome , 1987, Journal of the Royal Society of Medicine.

[17]  John L. Kelly,et al.  A new interpretation of information rate , 1956, IRE Trans. Inf. Theory.

[18]  Håvard Rue,et al.  Prediction and retrospective analysis of soccer matches in a league , 2000 .

[19]  David Law,et al.  The Favourite-Longshot Bias and Market Efficiency in UK Football Betting , 2000 .

[20]  Steven D. Levitt,et al.  Why are Gambling Markets Organized so Differently from Financial Markets? , 2004 .

[21]  John Goddard,et al.  Forecasting football results and the efficiency of fixed‐odds betting , 2004 .

[22]  A. Elo.  The rating of chessplayers, past and present , 1978 .

[23]  Sang Joon Kim,et al.  A Mathematical Theory of Communication , 2006 .

[24]  Raymond D. Sauer,et al.  The Economics of Wagering Markets , 1998 .

[25]  Robert Simmons,et al.  Sentiment in the betting market on Spanish football , 2008 .

[26]  S. Clarke,et al.  Using official ratings to simulate major tennis tournaments , 2000 .

[27]  Robert Simmons,et al.  Odds setters as forecasters: the case of English football , 2005 .

[28]  S. Coles,et al.  Modelling Association Football Scores and Inefficiencies in the Football Betting Market , 1997 .

[29]  Herman Stekler,et al.  Testing the efficiency of the National Football League betting market , 2006 .

[30]  Robert Simmons,et al.  Forecasting sport: the behaviour and performance of football tipsters , 2000 .

[31]  Tim Kuypers,et al.  Information and efficiency: an empirical study of a fixed odds betting market , 2000 .