Analysis of contextualized intensity in Men's elite handball using graph-based deep learning.

Manual annotation of data in invasion games is a costly task which poses a natural limit on sample sizes and the level of granularity used in match and performance analyses. To overcome this challenge, this work introduces FAUPA-ML, a Framework for Automatic Upscaled Performance Analysis with Machine Learning, which leverages graph neural networks to scale domain-specific expert knowledge to large data sets. Networks were trained using position data of match phases (counter/position attacks), annotated manually by domain experts in 10 matches. The best network was applied to contextualize N = 539 matches of elite handball (2019/20-2021/22 German Men's Handball Bundesliga) with 86% balanced accuracy. Distance covered, speed, metabolic power, and metabolic work were calculated for attackers and defenders and differences between counters and position attacks across seasons analyzed with an ANOVA. Results showed that counter attacks are shorter, less frequent and more intense than position attacks and that attacking is more intense than defending. Findings show that FAUPA-ML generates accurate replications of expert knowledge that can be used to gain insights in performance analysis previously deemed infeasible. Future studies can use FAUPA-ML for large-scale, contextualized analyses that investigate influences of team strength, score-line, or team tactics on performance.

Analysis of Motion Characteristics and Metabolic Power in Elite Male Handball Players.

While handball is characterized by repeated sprints and changes of direction, traditional player load models do not consider accelerations and decelerations. The aim of this study was to analyze the differences between metabolic power and speed zones for player load assessment with regard to the player role. Position data from 330 male individuals during 77 games from the 2019/20 German Men's Handball-Bundesliga (HBL) were analyzed, resulting in 2233 individual observations. Players were categorized into wings, backs and pivots. Distance covered in different speed zones, metabolic power, metabolic work, equivalent distance (metabolic work divided by energy cost of running), time spend running, energy spend running, and time over 10 and 20 W were calculated. A 2-by-3 mixed ANOVA was calculated to investigate differences and interactions between groups and player load models. Results showed that total distance was longest in wings (3568 ± 1459 m in 42 ± 17 min), followed by backs (2462 ± 1145 m in 29 ± 14 min), and pivots (2445 ± 1052 m in 30 ± 13 min). Equivalent distance was greatest in wings (4072.50 ± 1644.83 m), followed by backs (2765.23 ± 1252.44 m), and pivots (2697.98 ± 1153.16 m). Distance covered and equivalent distance showed moderate to large interaction effects between wings and backs (p < .01, ES = 0.73) and between wings and pivots (p < .01, ES = 0.86) and a small interaction effect between backs and pivots (p < .01, ES = 0.22). The results underline the need for individualized management of training loads and the potential of using information about locomotive accelerations and decelerations to obtain more precise descriptions of player load during handball game performance at the highest level of competition. Future studies should investigate the influence of physical performance on smaller match sequences, like ball possession phases.

Context is key: normalization as a novel approach to sport specific preprocessing of KPI's for match analysis in soccer.

Key Performance Indicators (KPIs) have been investigated, validated and applied in multitude of sports for recruiting, coaching, opponent, self-analysis etc. Although a wide variety of in game performance indicators have been used as KPIs, they lack sports specific context. With the introduction of artificial intelligence and machine learning (AI/ML) in sports, the need for building intrinsic context into the independent variables is even greater as AI/ML models seem to perform better in terms of predictability but lack interpretability. The study proposes domain specific feature preprocessing method (normalization) that can be utilized across a wide range of sports and demonstrates its value through a specific data transformation by using team possession as a normalizing factor while analyzing defensive performance in soccer. The study performed two linear regressions and three gradient boosting machine models to demonstrate the value of normalization while predicting defensive performance. The results demonstrate that the direction of correlation of the relevant variables changes post normalization while predicting defensive performance of teams for the whole season. Both raw and normalized KPIs showing significant correlation with defensive performance (p < 0.001). The addition of the normalized variables contributes towards higher information gain, improved performance and increased interpretability of the models.