Interchangeability of optical tracking technologies: Potential overestimation of the sprint running load demands in the English Premier League.

Abstract

The purpose of this study was to assess the agreement between match-derived running load outputs; total distance (TD), high-speed running (HSR) and sprint distance (SPR) obtained by two optical tracking systems. Data were collected from 31 elite footballers from the first team and under-21 squads of an English Premier League (EPL) football club across three competitive matches. One EPL game (game 2) and one under-21 Premier League game (game 3) were played at the team’s home stadium and one EPL game (game 1) at an away venue. All matches were tracked concomitantly using eight colour cameras sampling at 10Hz (PROZONE®) and six high definition motion cameras sampling at 25Hz (TRACAB®). TD displayed a perfect (r = 0.99) correlation while HSR and SPR displayed very large (r = 0.81 and r = 0.73) correlations between TRACAB® and PROZONE®. Mean bias were 5% for TD, -3% for HSR and 61% for SPR. Between games, mean biases for TD were 6% for game 1, and 5% for game 2 and game 3. For HSR, 9% for game 1, -5% for game 2 and 6% for game 3 and for SPR, 31% for game 1, 71% for game 2 and 84% for game 3. TD, and HSR can be interchanged between PROZONE® and TRACAB®, to allow accurate interpretation between the two systems. PROZONE® overestimated SPR compared to the TRACAB®, with the magnitude of difference considered meaningful to alter interpretation of historical match outputs, sprint volume trends across the EPL and forecasts of the modern game.