A Flexible Approach to Football Analytics: Assessment, Modeling and Implementation
Authors
Philipp Seidenschwarz, Martin Rumo, Lukas Probst, Heiko Schuldt
Type
In Proceedings
Date
2019/11
Appears in
Proceedings of the 12th International Symposium on Computer Science in Sport (IACSS 2019)
Location
Moscow, Russia
Publisher
Springer
Abstract
Quantitative analysis in football is difficult due to the complexity and continuous fluidity of the game. Even though there is an increased accessibility of spatio-temporal data, scientific approaches to extract valuable information are seldomly useful in practice. We propose a new approach to building an information system for football. This approach consists of a method to extract football-specific concepts from interviews, to formalize them in a performance model, and to define and implement the data structures and algorithms in StreamTeam, a framework for the detection of complex (team) events. In this paper we present this approach in detail and provide an example for its use.
Staff members
Research Projects