GPS Data Reflect Players’ Internal Load in Soccer

Abstract

The use of RPE as a measure of Internal load has become a common methodology used in team sports owing to its low cost. The aim of this study was to build a machine learning process able to describe the players' RPE by the external load extracted from the GPS. In this paper, we propose a multidimensional approach to assess the RPE in professional soccer which is based on GPS measurements and machine learning. By using GPS tracking technology, we collect data describing the training workload of players in a professional soccer club during a season. We show that our Ordinal predictor is both accurate and precise in medium RPE value (i.e., between 4 and 7) but it is not consistent in etreme value (i.e., below 4 and above 7). Our approach is a preliminary study that suggest that it is possible to predict players' RPE from GPS training and match data. However, these are not the only information needed to understand the players' effort perceived after a trainings or matches.