Brazil or Germany - who will win the trophy ? Prediction of the FIFA World Cup 2014 based on team - specific regularized Poisson regression

In this article an approach for the analysis and prediction of soccer match results is proposed. It is based on a regularized Poisson regression model that includes various potentially influential covariates describing the national teams’ success in previous FIFA World Cups. Additionally, similar to Bradley-Terry-Luce models, differences of team-specific effects of the competing teams are included. It is discussed that within the generalized linear model (GLM) framework the team-specific effects can either be incorporated in the form of fixed or random effects. In order to achieve variable selection and shrinkage, we use tailored Lasso approaches. Based on the three preceding FIFA World Cups, two competing models for the prediction of the FIFA World Cup 2014 are fitted and investigated.