Women's beach handball game statistics: differences and predictive power for winning and losing teams

Abstract

The objectives of the present study were: (i) to compare beach handball game-related statistics by match outcome (winning and losing teams), and (ii) to identify characteristics that discriminate performances in the match. The game-related statistics of the 72 women’s matches played at the VIII Women’s Beach Handball World Championship (2018) were analysed. The game-related statistics were taken from the official website. A validation of the data showed their reliability to be very good (the inter-observer mean reliability was α=0.82 and the intra-observer mean was α=0.86). For the differences between winning/losing teams a parametric (unpaired t-test) or non-parametric (Mann-Whitney U test) test was applied depending on whether the variable met or did not meet normality, respectively. A stepwise discriminant analysis was then performed to determine the variables that predicted performance (victory or defeat). Five variables showed differences between the winning and losing teams: total points (p<.001; ES=1.09), technical faults (p<.001; ES=‑0.96), the number of players with either negative (p<.001; ES=‑0.86) or positive (p<.001; ES=1.05) valuations and overall valuation (p<.001; ES=1.29). The predictive model correctly classified 80.6% of the matches using two variables (Wilks’s λ=0.618; canonical correlation index=0.618): overall valuation and GK shots.