JRM Vol.36 No.3 pp. 603-617
doi: 10.20965/jrm.2024.p0603


Unveiling Multi-Agent Strategies: A Data-Driven Approach for Extracting and Evaluating Team Tactics from Football Event and Freeze-Frame Data

Calvin Yeung* ORCID Icon, Rory Bunker* ORCID Icon, and Keisuke Fujii*,**,***,†

*Nagoya University
1 Furo-cho, Chikusa-ku, Nagoya, Aichi 464-8603, Japan

**RIKEN Center for Advanced Intelligence Project
1-5 Yamadaoka, Suita, Osaka 565-0871, Japan

***PRESTO, Japan Science and Technology Agency
Kawaguchi Center Building, 4-1-8 Honcho, Kawaguchi, Saitama 332-0012, Japan

Corresponding author

October 19, 2023
March 14, 2024
June 20, 2024
muti-agent, sequential pattern mining, neural point process, deep learning, sports

Studying collective behavior in opposing multi-agent teams is crucial across game theory, robotics, and sports analytics. In sports, especially football, team tactics involve intricate strategic spatial and action behaviors displayed as event sequences during possession. Understanding and analyzing these tactics is essential for successful training, strategic planning, and on-field success. While traditional approaches, such as notational and statistical analyses, offer valuable insights into team tactics, they often lack a comprehensive consideration of contextual information, thereby limiting the holistic evaluation of teams’ performances. To bridge this gap and capture the nuanced intricacies of team tactics, we employed advanced methodologies. The sequential pattern mining algorithm PrefixSpan was utilized to extract tactical patterns from possession sequences, enabling a deeper understanding of how teams strategize and adapt during play. Additionally, the neural marked spatio temporal point process (NMSTPP) model was leveraged to model and predict team behaviors, facilitating a fair comparison among teams. The evaluation of team possessions was further enhanced through the innovative holistic possession utilization score metrics, providing a more nuanced assessment of performance. In our experimental exploration, we identified and classified five distinct team tactics, validated the efficacy of the NMSTPP model when integrating StatsBomb 360 data, and conducted a comprehensive analysis of English Premier League teams during the 2022/2023 season. The results were visualized using radar plots and scatter plots with mean shift clustering. Lastly, the potential applications to RoboCup were discussed.

Team tactics performance radar plot

Team tactics performance radar plot

Cite this article as:
C. Yeung, R. Bunker, and K. Fujii, “Unveiling Multi-Agent Strategies: A Data-Driven Approach for Extracting and Evaluating Team Tactics from Football Event and Freeze-Frame Data,” J. Robot. Mechatron., Vol.36 No.3, pp. 603-617, 2024.
Data files:
  1. [1] K. Fujii, “Data-driven analysis for understanding team sports behaviors,” J. Robot. Mechatron., Vol.33, No.3, pp. 505-514, 2021.
  2. [2] I. Palacios-Huerta, “Professionals play minimax,” The Review of Economic Studies, Vol.70, No.2, pp. 395-415, 2003.
  3. [3] M. Kubo, S. Mikami, Y. Kakazu, and M. Wada, “Adaptive formation plays in simulated soccer game based on pheromon as communication media and reward resources,” J. Robot. Mechatron., Vol.11, No.1, pp. 72-77, 1999.
  4. [4] S. Liu et al., “From motor control to team play in simulated humanoid football,” arXiv:2105.12196, 2021.
  5. [5] M. Tominaga, Y. Takemura, and K. Ishii, “Behavior learning system for robot soccer using neural network,” J. Robot. Mechatron., Vol.35, No.5, pp. 1385-1392, 2023.
  6. [6] G. Yasuda, “Modular distributed control architecture for cooperative soccer-playing robot agents,” J. Robot. Mechatron., Vol.12, No.1, pp. 29-34, 2000.
  7. [7] C. C. K. Yeung and K. Fujii, “A strategic framework for optimal decisions in football 1-vs-1 shot-taking situations: An integrated approach of machine learning, theory-based modeling, and game theory,” arXiv:2307.14732, 2023.
  8. [8] J.-F. Gréhaigne, P. Godbout, and D. Bouthier, “The foundations of tactics and strategy in team sports,” J. of Teaching in Physical Education, Vol.18, No.2, pp. 159-174, 1999.
  9. [9] S. Plakias et al., “Identifying soccer teams’ styles of play: A scoping and critical review,” J. of Functional Morphology and Kinesiology, Vol.8, No.2, Article No.39, 2023.
  10. [10] T. Decroos, L. Bransen, J. Van Haaren, and J. Davis, “Actions speak louder than goals: Valuing player actions in soccer,” Proc. of the 25th ACM SIGKDD Int. Conf. on Knowledge Discovery & Data Mining (KDD’19), pp. 1851-1861, 2019.
  11. [11] M. Van Roy, P. Robberechts, W.-C. Yang, L. De Raedt, and J. Davis, “A Markov framework for learning and reasoning about strategies in professional soccer,” J. of Artificial Intelligence Research, Vol.77, pp. 517-562, 2023.
  12. [12] C. C. K. Yeung, T. Sit, and K. Fujii, “Transformer-based neural marked spatio temporal point process model for football match events analysis,” arXiv:2302.09276, 2023.
  13. [13] K. Fukunaga and L. Hostetler, “The estimation of the gradient of a density function, with applications in pattern recognition,” IEEE Trans. on Information Theory, Vol.21, No.1, pp. 32-40, 1975.
  14. [14] C. Yeung and R. Bunker, “An events and 360 data-driven approach for extracting team tactics and evaluating performance in football,” StatsBomb Conf. 2023, 2023.
  15. [15] J. Fernandez-Navarro, L. Fradua, A. Zubillaga, P. R. Ford, and A. P. McRobert, “Attacking and defensive styles of play in soccer: Analysis of Spanish and English elite teams,” J. of Sports Sciences, Vol.34, No.24, pp. 2195-2204, 2016.
  16. [16] J. Bekkers and S. Dabadghao, “Flow motifs in soccer: What can passing behavior tell us?,” J. of Sports Analytics, Vol.5, No.4, pp. 299-311, 2019.
  17. [17] Q. Wang, H. Zhu, W. Hu, Z. Shen, and Y. Yao, “Discerning tactical patterns for professional soccer teams: An enhanced topic model with applications,” Proc. of the 21th ACM SIGKDD Int. Conf. on Knowledge Discovery and Data Mining (KDD’15), pp. 2197-2206, 2015.
  18. [18] P. Cintia, F. Giannotti, L. Pappalardo, D. Pedreschi, and M. Malvaldi, “The harsh rule of the goals: Data-driven performance indicators for football teams,” 2015 IEEE Int. Conf. on Data Science and Advanced Analytics (DSAA), 2015.
  19. [19] R. Rein and D. Memmert, “Big data and tactical analysis in elite soccer: Future challenges and opportunities for sports science,” SpringerPlus, Vol.5, No.1, Article No.1410, 2016.
  20. [20] J. Pei et al., “Mining sequential patterns by pattern-growth: The PrefixSpan approach,” IEEE Trans. on Knowledge and Data Engineering, Vol.16, No.11, pp. 1424-1440, 2004.
  21. [21] H. Eggels, R. v. Elk, and M. Pechenizkiy, “Expected goals in soccer: Explaining match results using predictive analytics,” Proc. of the Workshop on Machine Learning and Data Mining for Sports Analytics 2016 (MLSA 2016), 2016.
  22. [22] B. Macdonald, “An expected goals model for evaluating NHL teams and players,” 2012 MIT Sloan Sports Analytics Conf., 2012.
  23. [23] P. Lucey, A. Bialkowski, M. Monfort, P. Carr, and I. Matthews, “‘Quality vs quantity’: Improved shot prediction in soccer using strategic features from spatiotemporal data,” 2015 Annual MIT Sloan Sports Analytics Conf., 2015.
  24. [24] D. Link, S. Lang, and P. Seidenschwarz, “Real time quantification of dangerousity in football using spatiotemporal tracking data,” PLOS ONE, Vol.11, No.12, Article No.e0168768, 2016.
  25. [25] G. Liu, Y. Luo, O. Schulte, and T. Kharrat, “Deep soccer analytics: Learning an action-value function for evaluating soccer players,” Data Mining and Knowledge Discovery, Vol.34, No.5, pp. 1531-1559, 2020.
  26. [26] G. Liu and O. Schulte, “Deep reinforcement learning in ice hockey for context-aware player evaluation,” Proc. of the 27th Int. Joint Conf. on Artificial Intelligence (IJCAI-18), pp. 3442-3448, 2018.
  27. [27] W. Spearman, “Beyond expected goals,” 2018 MIT Sloan Sports Analytics Conf., 2018.
  28. [28] W. Spearman, A. Basye, G. Dick, R. Hotovy, and P. Pop, “Physics-based modeling of pass probabilities in soccer,” 2017 MIT Sloan Sports Analytics Conf., 2017.
  29. [29] M. Teranishi, K. Tsutsui, K. Takeda, and K. Fujii, “Evaluation of creating scoring opportunities for teammates in soccer via trajectory prediction,” U. Brefeld, J. Davis, J. V. Haaren, and A. Zimmmermann (Eds.), “Machine Learning and Data Mining for Sports Analytics,” pp. 53-73, Springer, 2022.
  30. [30] K. Toda, M. Teranishi, K. Kushiro, and K. Fujii, “Evaluation of soccer team defense based on prediction models of ball recovery and being attacked: A pilot study,” PLOS ONE, Vol.17, No.1, Article No.e0263051, 2022.
  31. [31] R. Umemoto, K. Tsutsui, and K. Fujii, “Location analysis of players in UEFA EURO 2020 and 2022 using generalized valuation of defense by estimating probabilities,” arXiv:2212.00021, 2022.
  32. [32] I. Simpson, R. J. Beal, D. Locke, and T. J. Norman, “Seq2Event: Learning the language of soccer using Transformer-based match event prediction,” Proc. of the 28th ACM SIGKDD Conf. on Knowledge Discovery and Data Mining (KDD’22), pp. 3898-3908, 2022.
  33. [33] M. J. Quinlan and S. K. Chalup, “Impact of tactical variations in the RoboCup four-legged league,” Proc. of the 2006 Int. Symp. on Practical Cognitive Agents and Robots (PCAR’06), pp. 27-38, 2006.
  34. [34] M. J. Quinlan, O. Obst, and S. K. Chalup, “Towards autonomous strategy decisions in the RoboCup four-legged league,” Proc. of the 7th IJCAI Int. Workshop on Nonmonotonic Reasoning, Action and Change, 2006.
  35. [35] L. de Koning, J. P. Mendoza, M. Veloso, and R. van de Molengraft, “Skills, tactics and plays for distributed multi-robot control in adversarial environments,” H. Akiyama, O. Obst, C. Sammut, and F. Tonidandel (Eds.), “RoboCup 2017: Robot World Cup XXI,” pp. 277-289, Springer, 2018.
  36. [36] T. Fukushima, T. Nakashima, and V. Torra, “Team classification with tactical analysis using fuzzy inference in RoboCup soccer,” 2020 Joint 11th Int. Conf. on Soft Computing and Intelligent Systems and 21st Int. Symp. on Advanced Intelligent Systems (SCIS-ISIS), 2020.
  37. [37] M. Prokopenko, O. Obst, P. Wang, and J. Held, “Gliders2012: Tactics with action-dependent evaluation functions,” RoboCup 2012 Symp. and Competitions, 2012.
  38. [38] X. Xie et al., “PassVizor: Toward better understanding of the dynamics of soccer passes,” IEEE Trans. on Visualization and Computer Graphics, Vol.27, No.2, pp. 1322-1331, 2021.
  39. [39] A. Vaswani et al., “Attention is all you need,” Proc. of the 31st Int. Conf. on Neural Information Processing Systems (NIPS’17), pp. 6000-6010, 2017.
  40. [40] D. P. Kingma and J. Ba, “Adam: A method for stochastic optimization,” arXiv:1412.6980, 2014.
  41. [41] K. G. Derpanis, “Mean shift clustering,” 2005.
  42. [42] K. Fujii et al., “Adaptive action supervision in reinforcement learning from real-world multi-agent demonstrations,” Proc. of the 16th Int. Conf. on Agents and Artificial Intelligence (ICAART), Vol.2, pp. 27-39, 2024.

*This site is desgined based on HTML5 and CSS3 for modern browsers, e.g. Chrome, Firefox, Safari, Edge, Opera.

Last updated on Jul. 12, 2024