Applying collective tactical metrics in rugby union: A case study of spatial-temporal team-behaviour during scoring sequences

Abstract

We introduced objective spatial-temporal metrics including centroid, stretch index, convex hull, and approximate entropy (ApEn) to analyse team behaviours during try-scoring events in rugby union. Data collected via 10 Hz global positioning system (GPS) technology during a competitive Super Rugby match were analysed across different field zones and phases of play. At the match level, there were no significant differences in team-level metrics between attack and defence. However, location specific analyses demonstrated higher defensive movement irregularity (higher ApEn) in the D22–40 (attack vs defence; 0.19 ± 0.02 vs 0.27 ± 0.02; mean ± SD) and D40–50 (0.24 ± 0.03 vs 0.30 ± 0.02), indicating field-location specific irregularity and defensive instability. Conversely, attacking sequences in the D22–40 (14.3 ± 1.1 m vs 13.0 ± 0.8 m) and D40–50 (14.1 ± 0.8 m vs 12.7 ± 0.2 m) displayed greater spatial expansion (higher stretch index) associated with creating scoring opportunities. These preliminary findings indicate that spatial-temporal metrics can capture context-dependent team behaviours in rugby union, offering promising tools for tactical analysis. Given the limited dataset, further research with larger samples is necessary to validate and refine these approaches for practical coaching applications.