Current Match-Analysis Techniques' Underestimation of Intense Periods of High-Velocity Running \ud

To compare the peak 5-min period of high-velocity running (HiVR) during a soccer match using a predefined vs a rolling time interval. Methods: Player movement data were collected from 19 elite Australian soccer players over 11 competitive matches (77 individual match files) using a 5-Hz global-positioning system. Raw velocity data were analyzed to determine the period containing the greatest HiVR distance per match half and the distance covered in the subsequent epoch. Intervals were identified using either a predefined (distance covered in 5 min at every 5-min time point) or rolling (distance covered in 5 min from every time point) method. The percentage difference ± 90% confidence limits were used to determine differences between methods. Results: Predefined periods underestimated peak distance covered by up to 25% and overestimated the subsequent epoch by up to 31% compared with rolling periods. When the distance decrement between the peak and following period was determined, there was up to a 52% greater reduction in running performance using rolling periods than predefined ones. Conclusions: It is recommended that researchers use rolling as opposed to predefined periods when determining specific match intervals because they provide a more accurate representation of the HiVR distance covered. This will avoid underestimation of both match running distance and the decrement in running performance after an intense period of play. This may have practical implica- tions for not only researchers but also staff involved in a club setting who use this reduction as evidence of transient fatigue during a match