Peak speed determination in football : is sprint testing necessary?

Purpose: The purpose of this study was to examine the utility of 10 Hz global positioning system (GPS) technology to determine peak speed (PS) during both field-testing assessments and tracking during football matches. Materials and Methods: Twenty-three semi-professional football players wore GPS devices during 14 preseason (non-competitive; n = 6) and in-season (competitive; n = 8) fixtures (GPSMatch), and also during 40-m maximal sprint assessments (GPSSprint) measured concurrently via timing gates at 10-m intervals. Results: A large agreement (r = 0.84; 90% confidence interval [CI]: 0.70-0.92) and trivial bias (−0.30 km ∙ h−1; 90% CI: −0.61 to 0.01 km ∙ h−1) were observed for PS determined via GPSSprint versus timing gates in sprint testing. Absolute PS was faster in matches (31.4 ± 1.5 km ∙ h−1) versus timing gate (+0.80; 90% CI: 0.13-1.47 km ∙ h−1; likely small effect) and GPSSprint (+1.14; 90% CI: 0.47-1.81 km ∙ h−1; likely moderate effect), irrespective of positional role and stage of the season. Conclusions: These data question the relevance of sprint testing in football and suggest that PS can be determined from GPS tracking data collected over a series of matches.