Summary:

The European Football Championship (Euro) is different from any other soccer competition in the world in that it has a long and homogeneous qualifying period for all teams. Using data from every tournament that has taken place in history, a step logit model is estimated to quantify the role of team’s earlier qualifying performance in the likelihood of success at the final stage. Considering only the information available at the date preceding each of the last three Euros, we test the model’s ability to forecast the winner at future tournaments. The model correctly predicted Spain to win it in 2008 and 2012, as well as Portugal to take the Cup in 2016. Teams’ efficacy during qualification is found to be a key contributor to the model’s forecasting performance.

Our results have strong implications about the current FIFA ranking system as a way to gauge teams’ relative strength, as well as about which information a sophisticated bettor should process in order to beat the odds and make a profit out of the betting market. In that regard, we conclude that the betting market is possibly not efficient when pricing teams at the Euros.