Relationship between game-related statistics in elite men’s beach handball and the final result: a classification tree approach

Abstract

ABSTRACTThis study’s objectives were to (i) to compare beach handball game-related statistics by match outcome (winning and losing teams) and (ii) to develop a multivariate model explaining the performance in elite men’s beach handball. Seventy-two matches of the VIII Men’s Beach Handball World Championship held in Kazan (Russia) in 2018 were analysed. The dependent variable was match outcome (winning and losing teams), and the independent variables were the game-related statistics. A paired sample t-test was used to examine differences between teams. The effect sizes of the differences were calculated. The data were subjected to a multivariate analysis in the form of a regression and classification tree. The game statistics with the greatest effect size (ES≥0.88) and which differentiated the winning and losing teams were the team’s overall valuation and total points. The classification and regression tree model correctly classified 94.5% of the records on the basis of eight variables distributed over 20 nodes: overall valuation, goalkeeper blocked spin-shots, goalkeeper received spin-shots, goalkeeper received one-pointer, spin-shot goals, specialist goals, blocks, technical fouls. Coaches could apply these results to seek better performance in the goalkeeper position and in two-pointer goals.