2015/01/23 by Jean‐Louis Foulley, Foulley, Jean-Louis
Computer Science · Decision Sciences · Economics, Econometrics and Finance · #62 #Applications (stat.AP) #Data Analysis with R #FOS: Computer and information sciences #Forecasting Techniques and Applications #Sports Analytics and Performance
paper · pdf · doi:10.48550/arxiv.1501.05831
openalex publication_date 2015/01/23 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
This article presents a Bayesian implementation of a cumulative probit model\nto forecast the outcomes of the UEFA Champions League matches. The argument of\nthe normal CDF involves a cut-off point, a home vs away playing effect and the\ndifference in strength of the two competing teams. Team strength is assumed to\nfollow a Gaussian distribution the expectation of which is expressed as a\nlinear regression on an external rating of the team from eg the UEFA Club\nRanking (UEFACR) or the Football Club World Ranking (FCWR). Priors on these\nparameters are updated at the beginning of each season from their posterior\ndistributions obtained at the end of the previous one. This allows making\npredictions of match results for each phase of the competition: group stage and\nknock-out. An application is presented for the 2013-2014 season. Adjustment\nbased on the FCWR performs better than on UEFACR.\n