A0622
Title: Having a ball: Evaluating scoring streaks and game excitement using in-match trend estimation
Authors: Andreas Kryger Jensen - University of Copenhagen (Denmark)
Claus Ekstrom - University of Copenhagen (Denmark) [presenting]
Abstract: Many popular sports involve matches between two teams or players where each team score points throughout the match. While the overall match winner or result is interesting, it conveys little information about the underlying scoring trends throughout the match. Modelling approaches that accommodate a finer granularity of the score difference throughout the match are needed to evaluate in-game strategies and discuss scoring streaks, team strengths, and other aspects of the game. A latent Gaussian process is proposed to model the score difference between two teams and introduce the Trend Direction Index as an easily interpretable probabilistic measure of the current trend in the match as well as a measure of post-game trend evaluation. In addition, the Excitement Trend Index - the expected number of monotonicity changes in the running score difference - is proposed to measure overall game excitement. The proposed methodology is applied to all 1143 matches from the 20192020 National Basketball Association season. It is shown how trends can be interpreted in individual games and how the excitement score can be used to cluster teams according to how excited they are to watch.
