Combining video and player telemetry for evidence-based decisions in soccer

• Main Authors: Håvard D. Johansen, Svein Arne Pettersen, Dag Johansen
• Other Authors: The Pennsylvania State University CiteSeerX Archives
• Format: Text
• Language: English
• Published: 2013
• Subjects: big-data soccer analytics; video retrieval; athlete sensor network; data security; Tromsø
• Online Access: http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.484.1310; http://heim.ifi.uio.no/~paalh/publications/files/icSport2013.pdf

Abstract: Technology is changing how soccer clubs train and interact with their supporters. Systems that provide ac-quisition and visualization of low-level player telemetry, like distance covered and speed, are already being widely adopted. A key observation is that such data when correlated with actual in-game video footage is a powerful tool for evidence-based decisions. As data volume and complexity grow, efficient tools for auto-mated high-precision retrieval become essential. This paper describes the unique combination of a radio-based sensor platform and several custom video retrieval systems in operational use at Tromsø Idrettslag (TIL), a Norwegian premier league soccer club. The systems have been developed using an experimental computer-science method where several prototypes were built and deployed for evaluation in close collaboration with the intended users. Although our method of computer-system prototyping has not yielded commercial quality products, it has enabled us to construct several novel systems combining low-level player telemetry with video retrieval, annotations, and user-centric security. 1