Although sports on-field activities occur sequentially, traditional performance indicators quantify players` activities without regard to their sequential nature. Nowadays, movement patterns are used to sequentially quantify players` activities to understand match demands on players. However, the specific behavioural (i.e., signature) movement patterns of rugby league players per playing position remain unknown and the prediction of rugby league players into all nine playing positions based on their movement patterns is largely unexplored. Hence, this study identified the signature movement patterns of elite rugby league players per position and revealed the contribution of movement patterns towards the prediction of players into positions during the 2019 and 2020 seasons. Varying numbers of signature movement patterns were identified across playing positions with centres having the highest number of signature patterns (i.e. 1241). Random Forest best predicted elite rugby league players` positions at 73.41% accuracy, 0.74 recall, and 0.73 f1-score and precision scores based on movement patterns relative frequency values and top contributing movement patterns were identified. Therefore, we recommend sports stakeholders recognize the signature and contributing movement pattern of players per playing position while making decisions regarding training programmes, talent identification and recruitment.
© Copyright 2024 Machine Learning and Data Mining for Sports Analytics. MLSA 2023. Communications in Computer and Information Science. Veröffentlicht von Springer. Alle Rechte vorbehalten.