Seeding the UEFA Champions League Participants: Evaluation of the Reform

Abstract

We evaluate the sporting effects of the seeding system reform in the major football club tournament -- the Champions League -- organized by the Union of European Football Associations (UEFA). In the UEFA Champions League, before the 2015-16 season, the teams were seeded in the group stage with respect to their ratings. Starting from the 2015-16 season, national champions of the Top-7 countries are seeded in the first pot, whereas other teams are seeded by their rating as before. We propose a probabilistic model for predicting the score of a single match in UEFA tournaments as well as the whole UEFA season. This model uses clubs' ratings as inputs. Applying Monte-Carlo simulations, we show that the expected rating of the UEFA Champions League winner, as well as the sum of the finalists' ratings, slightly decreased after the reform. At the same time, the difference in the finalists' ratings, which is a measure of competitive balance, increased. The UEFA Europa League became stronger and less balanced. We check the robustness of the results by introducing local fluctuations in the clubs ratings. Also, we study which national associations took advantage of the reform. For seeding rules before and after the reform, we estimate the transition matrix (pij), where pij is the probability of i-th strongest national association moving to j-th position after the season. The effect of reform on a single national association measured by the change in probability to increase or decrease the association's UEFA rank is not more than 3%.

Keywords: tournament; design; seeding; competitive balance; UEFA Champions League; Monte-Carlo simulations

JEL Classification: Z20, C44