Rugby Union teams offensive style analyze using clustering algorithms over one season

Engineering of Sport 15 - Proceedings from the 15th International Conference on the Engineering of Sport (ISEA 2024)

The use of data based algorithms is more and more common in the daily lives of professional rugby teams. Many performance aspects may be studied: from physical preparation using GPS to analyze physical load, to video analysis of games but also medical and wellness data. It is now crucial to use all this data to give the staff insights into their team but also their opponents. Data from other teams is more and more accessible through data providers which analyze all match video recording to extract information of interest, such as the timelines provided by OptaTM. Moreover, the manner to analyze such data to better understand the strengths and weaknesses of the opponents remains essential. Thus the aim of the study was to proposer an unsupervised learning method to gather teams which had a similar style of play in order to analyze them more easily and to understand the strengths and weaknesses of an opponent's game.