How different defensive formations affect physical match demands in the German Handball-Bundesliga: a comparison of the attacking and defending team

Introduction In handball, the effects of different defense playing formations are little studied yet, but play a crucial role for managing training volume and intensity when planning specific drills. The aim of the study was to examine how different playing formations within a positional defense phase influence physical match demands for both the attacking and defending team. Method A fixed installed local positioning system (Kinexon LPS) was used to collect positional data during the 2023/2024 German Handball-Bundesliga season. The physical match demands were operationalized by the normalized number of accelerations and decelerations (events per 10 min) and the average distance covered (m) by the team within a positional defense phase. The playing formations within a positional defense (6:0, 5:1, 4:2, 3:3) were detected using an existing algorithm that was modified accordingly. In total, 11,863 positional defense phases in equal numbers of players (6 vs. 6) were analyzed including 316 players of 18 clubs. Generalized linear mixed models (GLMMs) were fitted to estimate differences between the playing formations for the attacking and defending team. Results Significant three-way interaction effects were observed for deceleration events in defending teams in the 5:1 (0.235 ± 0.063, p < 0.001 ), 4:2 (0.432 ± 0.112, p < 0.001 ), and 3:3 (0.456 ± 0.151, p = 0.003 ) formations. Furthermore, significant differences were found in average distance covered by team for 5:1 (120.0 ± 23.3 m, p < 0.001 ), 4:2 (479.0 ± 45.5 m, p < 0.001 ), 3:3 (776.8 ± 66.0 m, p < 0.001 ) formations compared with 6:0. Conclusion We adapted an algorithm to detect positional defense phases (6:0, 5:1, 4:2, and 3:3). With more aggressive playing formations, total distance and numbers of acceleration and deceleration events tended to increase in both the attacking and defending teams. A stronger increase in deceleration events was estimated for defenders in the 3:3 formation, however this finding should be interpreted cautiously due to the low frequency of this formation. Our outcomes may help practitioners manage training intensity and volume when planning specific drills.