Variability of GPS-derived running performance during official matches in elite professional soccer players.

Abstract

The aim of this study was to assess the match-to-match variability obtained using GPS devices, collected during official games in professional soccer players. Global positioning system (GPS)-derived data from nineteen elite soccer players were collected over two consecutive seasons. Time-motion data for players with more than five full-match were analyzed (N.=202). Total distance covered (TD), TD>13-18 km/h, TD>18-21 km/h, TD>21 km/h, number of acceleration >2.5-4 m.s-2 and >4 m.s-2 were calculated. The match-to-match variation in running activity was assessed by the typical error expressed as a coefficient of variation (CV%) and the magnitude of the CV was calculated (effect size). When all players were pooled together, CVs ranged from 5% to 77% (first half) and from 5% to 90% (second half), for TD and number of acceleration >4 m.s-2, and the magnitude of the CVs were rated from small to moderate (effect size=0.57-0.98). The CVs were likely to increase with running/acceleration intensity, and were likely to differ between playing positions (e.g., TD>13-18 km/h 3.4% for second strikers vs. 14.2% for strikers and 14.9% for wide-defenders vs. 9.7% for wide-midfielders). Present findings indicate that variability in players' running performance is high in some variables and likely position-dependent. Such variability should be taken into account when using these variables to prescribe and/or monitor training intensity/load.