By Steven Deverteuil
Faculty advisor: Prashant Chandrasekar
11:00-11:50am HCC 328
Background
Player analysis is an integral aspect of the sports analytics world. Player analysis is useful for coaches, scouts, and players alike for enhancing player and team performance. Notably, analysis of player tendencies may lead to player action modeling — a tool which can provide any given team an advantage over an opposing team. This study attempts to provide a new method for analyzing and modeling the general passing tendencies of soccer goalkeepers.
Materials/Methods
A statistical significance test was used in this study. The pass data from several professional goalkeepers in the 2018 World Cup were analyzed. The Mann-Whitney-U test was used to correlate the pass’s time occurrence and pass’s distance. Specifically, the distances of goalkeeper passes were grouped into bins by varying experimental time periods (15-, 30-, and 45-minute bins). These bins of pass distances were then run through the statistical significance test to determine if there is a correlation between the distances of passes per bin.
Results
It was revealed that there is at best some plausibility as to goalkeepers varying their passes by game period. Several of the goalkeepers from this World Cup sample had one or more games where there was statistical significance to one or more of the pass distances by time-bin comparisons. However, this tendency cannot yet be generalized to all goalkeepers as there were many goalkeepers who received few to no significant results across all of their games and subsequent passes. Further, the sample size is too small to generalize the plausible findings.
Conclusions
Player analysis can be used to create models for player behavior. These models can be useful in predicting player behavior and can be generalized by position to reap similar rewards. Position-based models may also lead to a better understanding of the game.
