Abstract | This dissertation explores performance indicators differentiating and discriminating winning and losing teams from official sets of game-related statistics in handball and beach handball. The research comprises four primary studies, each contributing to the use of game-related statistics in a specific performance context. The Study 1 examined sex-based differences in game-related statistics between winning and losing teams in an amateur (hereafter semi-professional) handball league. Performance indicators were selected by predictive models for males (five variables) and females (two variables) from an entire season of statistics, tested for reliability. The findings suggest that special attention should be given to the goalkeeper's performance and overall shooting efficiency. The sexes differed in terms of moderate effect size (d>0.5), 9 m shot efficiency, red cards, 2-minute exclusions, and technical fouls. The Study 2 delved into the final ten minutes of balanced handball games, investigating the factors that discriminate winners from losers during this critical period. The analysis revealed that winning male teams exhibit better shooting efficiency (d=1.114), goalkeepers 9 m saves efficiency (d=1.100), total saves efficiency (d=0.827) and 9 m shots efficiency (d=0.875) with large effect sizes. Results for females identified shots (d=1.414), GB save efficiency (d=1.330), 9 m shot (d=0.933), and GB 9 m save efficiency (d=0.923). The discriminatory model for males selected only technical fouls and 9m shots with 40.4% correct classification. In conclusion, the authors argue that special attention should be paid to 9 m shots in the final minutes due to teams adapting to the high pressure, fatigue, and tactical constraints, possibly by playing more passively. The Study 3 focused on performance profiling in handball, using discriminative variables to develop a performance profile for one successful team using the most impactful performance indicators. The profiles were based on the team´s game-related statistics of the previous five years and split according to opponents' quality. The two profiles against other TOP4 and lower-ranked teams differed, indicating different interactions based on opposition quality. These profiles can aid coaches by identifying current team performance characteristics and areas for improvement. The Study 4 shifted the focus to women's beach handball, comparing game-related statistics between winning and losing teams at the elite international level. Although beach handball has unique sports characteristics, the game principles are comparable to indoor handball, which invites a similar approach to game-related statistics. The findings emphasize that the official data was reliable and performance could be discriminated with 80.6% accuracy using the overall valuation index and GB saves efficiency. The cross-paper synthesis of the findings produced overarching themes related to the steps of using the data. Reliable and valid data collection is fundamental for accurate performance analysis. Selecting appropriate performance indicators based on the context is crucial to inform coaches in developing practical training and match strategies. Creating performance profiles can be a valuable tool to enhance individual and team performance. The dissertation also discusses the limitations of the studies, such as sample size constraints and the variability of game conditions. Future research directions include theoretically guided exploration of the processes behind the game-related statistical events recorded with advanced statistical methods. Moreover, to investigate the applicability of the findings to different levels of competition and other team sports. In conclusion, this dissertation contributes to understanding of performance analysis in handball and beach handball by identifying key performance indicators and demonstrating their practical applications. The findings provide valuable insights for coaches, analysts, and researchers aiming to improve performance outcomes in these sports. |