Using the interaction of speed and acceleration to detect repeated-sprint activity in team sports

Purpose To compare the occurrence and characteristics of repeated-sprint (RS) activity in elite team sport competition when classified according to speed and/or acceleration, and their interaction via metabolic power (Pmet). Methods: Elite male hockey players (N = 16) wore player-tracking devices in six international matches. Sprint efforts were defined using four separate classifications: speed >5.5 m∙s−1, acceleration >1.5 m∙s−2, speed-or-acceleration, and Pmet >25.5 W∙kg−1. RS bouts were defined as ≥3 efforts with mean recovery ≤21 s. For Pmet, RS bouts were also classified using a maximal recovery period ≤21 s. The number of sprint efforts and RS bouts, and the number of efforts, effort durations and recovery periods within RS bouts, were compared across each classification method, and between mean and maximal recovery criteria. Results: More RS bouts were identified via Pmet (8.5 ± 2.8) than either speed and/or acceleration, and comprised more efforts (4.0 ± 0.4) with shorter recovery periods (11.5 ± 1.8 s). Fewer RS bouts (7.3 ± 2.8 vs. 8.5 ± 2.8) were identified with a maximum rather than mean recovery criterion. Conclusions: Definitions of sprint efforts and recovery periods which reflect ATP depletion and replenishment via Pmet suggest that RS activity occurs frequently in team sport competition, and is more demanding than when speed and/or acceleration are used to define RS activity in variable-speed locomotion.