The large amount of features recorded from GPS and inertial sensors (external load) and well-being questionnaires (internal load) can be used together in a multi-dimensional non-linear machine learning based model for a better prediction of non-contact injuries. In this study we put forward the main hypothesis that the use of such models would be able to inform better about injury risks by considering the evolution of both internal and external loads over two horizons (one week and one month). Predictive models were trained with data collected by both GPS and subjective questionnaires and injury data from 40 elite male soccer players over one season. Various classification machine-learning algorithms that performed best on external and internal loads features were compared using standard performance metrics such as accuracy, precision, recall and the area under the receiver operator characteristic curve. In particular, tree-based algorithms based on non-linear models with an important interpretation aspect were privileged as they can help to understand internal and external load features impact on injury risk. For 1-week injury prediction, internal load features data were more accurate than external load features while for 1-month injury prediction, the best performances of classifiers were reached by combining internal and external load features.

The aim of the present study was to analyse the relationships between fluid ingestion, changes in body mass and physical activity amongst elite soccer players. 32 elite French soccer players were divided into six playing positions: goalkeepers (GK), central defenders (CD), full backs (FB), central midfielders (CM), wide midfielders (WM) and forwards (FW) and participated in official friendly matches within two consecutive pre-season periods. Body mass changes and fluid ingestion were recorded before, during and just after the matches. Time-motion characteristics were also analysed. Players ingested 1.4±0.6 L of fluids during matches and lost 1.4±0.6 kg at the end of the games. WM lost more weight than GK, FB and FW (p<0.05) and CM lost more weight than GK (p<0.05). Furthermore, CD covered significantly less total distance, high (HI) and very high intensity running distance than all other playing positions excluding GK (p<0.05); and WM covered greater distances at HI than all other playing positions. No differences were found in fluid ingestion between playing positions. The results of this study indicate that sweat loss was significantly different when comparing across various playing positions. As a result, sweat lost may be more influenced by HI activity during a game than other physical activity within game scenarios. Therefore, players with the highest amount of HI activities during a match should even more pay attention to their rehydration.Key words: performance; fatigue; drinking; dehydration; football; playing position. AbstractCorrespondence/correspondencia: Leo Djaoui Inter-university Laboratory of Human Movement Biology (LIBM), University Claude Bernard Lyon 1, University of Lyon, Lyon Francia Email: leo.djaoui@gmail.com El objetivo del presente estudio consistió en analizar las relaciones entre la ingesta de líquidos, cambios en la composición corporal y en la actividad física en jugadores de élite. La muestra estaba compuesta por 32 jugadores franceses de élite divididos en seis posiciones de juego: porteros (P), defensas centrales (DC), laterales (L), mediocentros (M), Extremos (E) y delanteros (D). Se registraron en dos partidos amistosos jugados en la pretemporada. Los cambios en composición corporal y la ingesta de líquidos se realizaron antes, durante y después de los partidos. Los datos sobre rendimiento físico también fueron registrados. Los jugadores tomaron 1.4±0.6 L de líquidos durante los partidos y perdieron 1.4±0.6 L al terminar los mismos. Los E perdieron más peso que los P, L y D (p<0.05) y los M más que los P (p<0.05). Asimismo, los DC corrieron menos distancia total, a alta (HI) y muy alta intensidad que el resto de puestos de juego, con excepción de los P (p<0.05); y los E recorrieron más distancias a HI que los otros puestos de juego. No se encontraron diferencias sobre la ingesta de líquidos entre puestos de juego. Los resultados del estudio indican que la pérdida de líquidos por sudor fue diferente entre los puestos de juego. De modo que la pérdida de sudor pudo afectar la activ...

No abstract