EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Fractal Ball Movement Complexity: Spatiotemporal Analysis for Team Performance Evaluation in Association Football

Ishara Bandara, Sergiy Shelyag, Sutharshan Rajasegarar, Daniel B. Dwyer, Eun-jin Kim, Dat Le, Maia Angelova

SN Computer Science · 30.03.2026 · 7 (4)

DOI: 10.1007/s42979-026-04878-w ↗

Kurz gefasst

Fachartikel in der Zeitschrift SN Computer Science, online erschienen am 30.03.2026. Verfasst von Ishara Bandara und 6 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 32 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Deakin University.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Spatiotemporal techniques are increasingly integral to football analytics for assessing team performance, yet the inherently dynamic and interactive nature of association football continues to hinder fully objective evaluation. Although existing research has examined the randomness of match events and patterns of player interactions, the contribution of ball movement trajectories has received comparatively limited attention despite its tactical significance. This study addresses this shortcoming by introducing a novel framework to quantify the spatial complexity of team ball movement as an indicator of offensive performance. Using a time-series feature extraction methodology, the fractal dimension of two-dimensional ball trajectory representations is calculated over predefined temporal windows to capture spatial complexity. The findings indicate a positive relationship between higher spatial complexity and match success, particularly in the early stages of play. In addition, a multi-scale fractal analysis reveals scale-dependent effects, with coarse- and mid-scale spatial complexity showing positive correlations with match-winning outcomes, while fine-scale complexity exhibits an inverse relationship, indicating that spatial structure at broader scales may be more influential for effective team performance than high-frequency local variations in ball movement. Finally, a Random Forest classification model trained solely on spatial complexity features achieved an AUC–ROC score of 0.8180 in predicting match outcomes, demonstrating that spatial complexity serves as an informative, interpretable, and effective time-series metric for evaluating team performance in association football.

Förderung

  • Deakin University

Zitieren

APA

Bandara, I., Shelyag, S., Rajasegarar, S., Dwyer, D. B., Kim, E., Le, D., & Angelova, M. (2026). Fractal Ball Movement Complexity: Spatiotemporal Analysis for Team Performance Evaluation in Association Football. SN Computer Science, 7(4). https://doi.org/10.1007/s42979-026-04878-w

Vancouver

Bandara I, Shelyag S, Rajasegarar S, Dwyer DB, Kim E, Le D, et al. Fractal Ball Movement Complexity: Spatiotemporal Analysis for Team Performance Evaluation in Association Football. SN Computer Science. 2026;7(4). doi:10.1007/s42979-026-04878-w

BibTeX
@article{Bandara2026,
  author = {Bandara, Ishara and Shelyag, Sergiy and Rajasegarar, Sutharshan and Dwyer, Daniel B. and Kim, Eun-jin and Le, Dat and Angelova, Maia},
  title = {Fractal Ball Movement Complexity: Spatiotemporal Analysis for Team Performance Evaluation in Association Football},
  journal = {SN Computer Science},
  year = {2026},
  volume = {7},
  number = {4},
  doi = {10.1007/s42979-026-04878-w}
}
Kostenlos starten →