Improving the Characterization and Comparison of Football Players with Spatial Flow Motifs

Alberto Barbosa, Pedro Ribeiro and Inês Dutra

2022

Abstract

Association Football is probably the world’s most popular sport. Being able to characterise and compare football players is therefore a very important and impactful task. In this work we introduce spatial flow motifs as an extension of previous work on this problem, by incorporating both temporal and spatial information into the network analysis of football data. Our approach considers passing sequences and the role of the player in those sequences, complemented with the physical position of the field where the passes occurred. We provide experimental results of our proposed methodology on real-life event data from the Italian League, showing we can more accurately identify players when compared to using purely topological data.

Keywords

Sports analytics; Subgraphs; Network motifs; Spatial data

Reference (text)

Alberto Barbosa, Pedro Ribeiro and Inês Dutra. Improving the Characterization and Comparison of Football Players with Spatial Flow Motifs. Proceedings of the 11th International Conference on Complex Networks and their Applications (CNA), pp. 579-591, Springer, Palermo, Italy, November, 2022.

Bibtex

@inproceedings{ribeiro-CNA2022b,
  author = {Alberto Barbosa and  Pedro Ribeiro and Inês Dutra},
  title = {Improving the Characterization and Comparison of Football Players with Spatial Flow Motifs},
  doi = {10.1007/978-3-031-21131-7_45},
  booktitle = {11th International Conference on Complex Networks and their Applications},
  pages = {579-591},
  publisher = {Springer},
  month = {November},
  year = {2022}
}