Performance analysis plays an important role in sports. Observing and analysing tactical behaviour can generate useful information that can be used for managing training processes and developing match strategies. The technological innovations of recent years - in particular, advances in the field of position tracking - present new challenges when it comes to analysing and interpreting this data. These include such questions as how clubs can best exploit the possibilities on offer to analyse game tactics, manage training processes and make better transfer decisions, how media companies can use this information to offer better and more innovative match coverage products and how new scientific insights into the nature of sporting phenomena in general and the factors that influence performance can be gained.

There has been increased activity in this area in recent years on the part of both the Competition Information Providers (CIP) and the scientific community. Companies are incorporating advanced methods of analysis into their software tools and an increasing number of publications in the academic sphere are reporting success in detecting tactical structures in raw data. Two phenomena can be observed here: the CIPs tend to be quick to launch products - chiefly in an attempt to gain a business advantage by regularly releasing new products - that lack scientific validation, or for which definition is ambiguous or tenuous. There are, on the other hand, approaches being used in the academic community that seem to support scientific profiling, but that do not reflect the framework within which sport managers, coaches and players are compelled to act. The challenge for the data mining and knowledge engineering community lies in using intelligent algorithms in order to derive complex performance indicators from the raw data that add value when it comes to "real" game analysis.

Against this background the talk firstly gives an overview about performance analysis (PA), proposes a structural model of PA and discusses the related epistemological issues. Secondly, the presentation gives insights into the game observation process and the software tools, that were used by the 2016 German Olympic Teams in beach volleyball to improve their performance during the Games. Thirdly, it presents a project being part of an innovation program in German football Bundesliga, which intended to develop smart performance indicators for professional clubs based on spatiotemporal data. Both examples lead to a discussion what kind of information is needed by sport scientists, coaches, analysts and players from data sciences. Finally, potential cooperation models between the scientific communities of sport science and computer sciences are presented.