The interest in sports analytics for the past few years has put the focus on analysing data (as opposed to video) to monitor and improve performance. Thanks to the advances of technology, a lot of data on performance is being collected on the pitch (and training grounds). But in order to be useful this data need to be analysed effectively. The methods and tools for doing so have been available for many years (the methods long before computing power caught up with making their application possible). In fact they have been used profitably for years for analysing performance in business and other fields of human endeavour. The problem is that both methods and tools are often unknown or ignored by sports organisations who could greatly profit from their application. This presentation aims to show how some of the most common methods used in business analytics can be usefully applied to the analysis of performance in team sports. It gives examples of each method and its application to a particular aspect of performance - of team and players. Each example is drawn from my experience in analysing performance in ice-hockey. And in particular of the Toronto Maple Leaves, a Canadian team which plays in the NHL - the most prestigious ice-hockey competition in the world. This takes place among teams US and Canada based teams. The application of the following advanced methods to some common performance analysis problems will be shown: 1. Classification trees, 2. Ranking, 3. Hierarchical Clustering, 4. Association (link-play), 5. Visual analysis.
© Copyright 2017 Journal of Human Sport & Exercise. University of Alicante. Alle Rechte vorbehalten.