From Outcome Prediction to Advantage Construction: A Temporal and Spatial Attribution Framework for League of Legends
Resumo
Introduction: Prior League of Legends (LoL) studies often emphasize winner prediction from aggregate indicators. This paper instead treats a match as a temporal process in which advantage is built, stabilized, or contested. Objective: The work asks whether a LoL match can be represented as a causal minute-level sequence and analyzed through spatial attribution, temporal state estimation, and short-horizon forecasting. Methodology: Match and timeline data are reconstructed into minute-level states using only information observable up to each minute. The representation combines composition, observed state, lane pressure, objectives, itemization, temporal variation, and gravity-informed descriptors. Gold is treated as an emergent macroindicator rather than as the explanatory center. Results: The state-estimation model reached 0.821 ROC AUC at 15 minutes and 0.872 at 20 minutes. Forecasting improved from 0.536 weighted F1 at a 1-minute horizon to 0.678 at 5 minutes. Spatial artifacts showed outcome-aligned regional control in intermediate windows, while lock-in remained exploratory rather than a validated detector.
Palavras-chave:
League of Legends, temporal modeling, game analytics, competitive advantage, short-horizon forecasting
Referências
Bahrololloomi, F., Smit, M., Ghotbi, E., van der Vegt, W., e van Deursen, A. (2022). Machine learning analysis of player performance for predicting victory in league of legends. In Proceedings of the 14th International Conference on Computer Supported Education (CSEDU 2022), volume 1, pages 63–74.
Battaglia, P. W., Hamrick, J. B., Bapst, V., Sanchez-Gonzalez, A., Zambaldi, V., Malinowski, M., Tacchetti, A., Raposo, D., Santoro, A., Faulkner, R., Gulcehre, C., Song, F., Ballard, A., Gilmer, J., Dahl, G., Vaswani, A., Allen, K. R., Nash, C., Langston, W., Dyer, C., Heess, N., Wierstra, D., Kohli, P., Botvinick, M., Vinyals, O., Li, Y., e Pascanu, R. (2018). Relational inductive biases, deep learning, and graph networks. arXiv preprint.
Birant, K. e Birant, D. (2022). Multi-objective multi-instance learning: A new approach to machine learning for esports. Entropy, 24(1):123.
Bisberg, A. J. e Ferrara, E. (2022). Gcn-wp–semi-supervised graph convolutional networks for win prediction in esports. In 2022 IEEE Conference on Games (CoG), pages 449–456. IEEE.
Chowdhury, S., Ahsan, M., e Barraclough, P. (2025). Applications of linear and ensemble-based machine learning for predicting winning teams in league of legends. Applied Sciences, 15(10).
Cong, S., Wu, H., Xiong, S., Wu, Y., e Zuo, L. (2024). Predicting the outcome of moba game: a case study on pregame prediction for league of legends. In 2024 8th Asian Conference on Artificial Intelligence Technology (ACAIT), pages 1366–1372.
Costa, L. M., Drachen, A., Souza, F. C. M., e Xexéo, G. (2024). Artificial intelligence in moba games: A multivocal literature mapping. IEEE Transactions on Games, 16(1):5–18.
Costa, L. M., Mantovani, R. G., Souza, F. C. M., e Xex’eo, G. (2021). Feature analysis to league of legends victory prediction on the picks and bans phase. In 2021 IEEE Conference on Games (CoG), pages 1–5.
Do, T. D., Wang, S. I., Yu, D. S., McMillian, M. G., e McMahan, R. P. (2021). Using machine learning to predict game outcomes based on player-champion experience in league of legends. Proceedings of the Annual Symposium on Computer-Human Interaction in Play, pages 28–34.
Duan, L., Li, S., Zhang, W., e Wang, W. (2022). Moba game item recommendation via relation-aware graph attention network. In 2022 IEEE Conference on Games (CoG), pages 338–344.
Gudmundsson, J. e Horton, M. (2017). Spatio-temporal analysis of team sports – a survey. ACM Computing Surveys, 50(2).
Guo, C., Pleiss, G., Sun, Y., e Weinberger, K. Q. (2017). On calibration of modern neural networks. In Proceedings of the 34th International Conference on Machine Learning, volume 70 of Proceedings of Machine Learning Research, pages 1321–1330.
Hagelbäck, J. e Johansson, S. (2009). Using multi-agent potential fields in real-time strategy games. In 2009 IEEE Symposium on Computational Intelligence and Games (CIG), pages 191–198.
Hitar-García, J. A., Valls-Vargas, J., Cuesta-Vargas, F. J., e Alba, E. (2022). Machine learning methods for predicting league of legends game outcome. In 2022 IEEE Conference on Games (CoG), pages 406–413.
Hochreiter, S. e Schmidhuber, J. (1997). Long short-term memory. Neural Computation, 9(8):1735–1780.
Hojaji, F., McIlroy, R. E., Dupuy, A., Pedroni, G., Toth, A. J., e Campbell, M. J. (2025). Deep learning techniques for identifying kpis in league of legends: Win prediction, map navigation, and vision control. Computers in Human Behavior Reports, 19.
Jung, C. e Kim, H. (2022). Win prediction from the snowball effect perspectives. In 2022 IEEE Games, Entertainment, Media Conference (GEM).
Junior, J. B. S. e Campelo, C. E. C. (2023). League of legends: Real-time result prediction.
Khatib, O. (1986). Real-time obstacle avoidance for manipulators and mobile robots. The International Journal of Robotics Research, 5(1):90–98.
Lee, J. e Kim, N. (2025). Development of machine learning-based indicators for predicting comeback victories using the bounty mechanism in moba games. Electronics, 14(7).
Lin, T.-Y., Goyal, P., Girshick, R., He, K., e Dollár, P. (2017). Focal loss for dense object detection. In Proceedings of the IEEE International Conference on Computer Vision (ICCV), pages 2980–2988.
Pedregosa, F., Varoquaux, G., Gramfort, A., Michel, V., Thirion, B., Grisel, O., Blondel, M., Prettenhofer, P., Weiss, R., Dubourg, V., Vanderplas, J., Passos, A., Cournapeau, D., Brucher, M., Perrot, M., e Duchesnay, E. (2011). Scikit-learn: Machine learning in python. Journal of Machine Learning Research, 12:2825–2830.
Riot Games, Inc. (2024). Riot games api. [link]. Acessado em: 02 de julho de 2024.
Silva, A. L. C., Pappa, G. L., e Chaimowicz, L. (2018). Continuous outcome prediction of league of legends competitive matches using recurrent neural networks. In Proceedings of SBGames 2018. Computing Track – Short Papers.
Spearman, W. (2018). Beyond expected goals. In Proceedings of the 12th MIT Sloan Sports Analytics Conference, Boston.
Vardakis, M., Margetis, G., Chatzakis, I., Apostolakis, K. C., e Stephanidis, C. (2026). Prediction of moba game events based on in-game data. Entertainment Computing, 57:101091.
Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., Kaiser, Ł., e Polosukhin, I. (2017). Attention is all you need. In Advances in Neural Information Processing Systems 30, pages 5998–6008. Curran Associates, Inc.
Yang, Z., Pan, Z., Wang, Y., Cai, D., Liu, X., Shi, S., e Huang, S.-L. (2020). Interpretable real-time win prediction for honor of kings, a popular mobile moba esport. arXiv preprint.
Battaglia, P. W., Hamrick, J. B., Bapst, V., Sanchez-Gonzalez, A., Zambaldi, V., Malinowski, M., Tacchetti, A., Raposo, D., Santoro, A., Faulkner, R., Gulcehre, C., Song, F., Ballard, A., Gilmer, J., Dahl, G., Vaswani, A., Allen, K. R., Nash, C., Langston, W., Dyer, C., Heess, N., Wierstra, D., Kohli, P., Botvinick, M., Vinyals, O., Li, Y., e Pascanu, R. (2018). Relational inductive biases, deep learning, and graph networks. arXiv preprint.
Birant, K. e Birant, D. (2022). Multi-objective multi-instance learning: A new approach to machine learning for esports. Entropy, 24(1):123.
Bisberg, A. J. e Ferrara, E. (2022). Gcn-wp–semi-supervised graph convolutional networks for win prediction in esports. In 2022 IEEE Conference on Games (CoG), pages 449–456. IEEE.
Chowdhury, S., Ahsan, M., e Barraclough, P. (2025). Applications of linear and ensemble-based machine learning for predicting winning teams in league of legends. Applied Sciences, 15(10).
Cong, S., Wu, H., Xiong, S., Wu, Y., e Zuo, L. (2024). Predicting the outcome of moba game: a case study on pregame prediction for league of legends. In 2024 8th Asian Conference on Artificial Intelligence Technology (ACAIT), pages 1366–1372.
Costa, L. M., Drachen, A., Souza, F. C. M., e Xexéo, G. (2024). Artificial intelligence in moba games: A multivocal literature mapping. IEEE Transactions on Games, 16(1):5–18.
Costa, L. M., Mantovani, R. G., Souza, F. C. M., e Xex’eo, G. (2021). Feature analysis to league of legends victory prediction on the picks and bans phase. In 2021 IEEE Conference on Games (CoG), pages 1–5.
Do, T. D., Wang, S. I., Yu, D. S., McMillian, M. G., e McMahan, R. P. (2021). Using machine learning to predict game outcomes based on player-champion experience in league of legends. Proceedings of the Annual Symposium on Computer-Human Interaction in Play, pages 28–34.
Duan, L., Li, S., Zhang, W., e Wang, W. (2022). Moba game item recommendation via relation-aware graph attention network. In 2022 IEEE Conference on Games (CoG), pages 338–344.
Gudmundsson, J. e Horton, M. (2017). Spatio-temporal analysis of team sports – a survey. ACM Computing Surveys, 50(2).
Guo, C., Pleiss, G., Sun, Y., e Weinberger, K. Q. (2017). On calibration of modern neural networks. In Proceedings of the 34th International Conference on Machine Learning, volume 70 of Proceedings of Machine Learning Research, pages 1321–1330.
Hagelbäck, J. e Johansson, S. (2009). Using multi-agent potential fields in real-time strategy games. In 2009 IEEE Symposium on Computational Intelligence and Games (CIG), pages 191–198.
Hitar-García, J. A., Valls-Vargas, J., Cuesta-Vargas, F. J., e Alba, E. (2022). Machine learning methods for predicting league of legends game outcome. In 2022 IEEE Conference on Games (CoG), pages 406–413.
Hochreiter, S. e Schmidhuber, J. (1997). Long short-term memory. Neural Computation, 9(8):1735–1780.
Hojaji, F., McIlroy, R. E., Dupuy, A., Pedroni, G., Toth, A. J., e Campbell, M. J. (2025). Deep learning techniques for identifying kpis in league of legends: Win prediction, map navigation, and vision control. Computers in Human Behavior Reports, 19.
Jung, C. e Kim, H. (2022). Win prediction from the snowball effect perspectives. In 2022 IEEE Games, Entertainment, Media Conference (GEM).
Junior, J. B. S. e Campelo, C. E. C. (2023). League of legends: Real-time result prediction.
Khatib, O. (1986). Real-time obstacle avoidance for manipulators and mobile robots. The International Journal of Robotics Research, 5(1):90–98.
Lee, J. e Kim, N. (2025). Development of machine learning-based indicators for predicting comeback victories using the bounty mechanism in moba games. Electronics, 14(7).
Lin, T.-Y., Goyal, P., Girshick, R., He, K., e Dollár, P. (2017). Focal loss for dense object detection. In Proceedings of the IEEE International Conference on Computer Vision (ICCV), pages 2980–2988.
Pedregosa, F., Varoquaux, G., Gramfort, A., Michel, V., Thirion, B., Grisel, O., Blondel, M., Prettenhofer, P., Weiss, R., Dubourg, V., Vanderplas, J., Passos, A., Cournapeau, D., Brucher, M., Perrot, M., e Duchesnay, E. (2011). Scikit-learn: Machine learning in python. Journal of Machine Learning Research, 12:2825–2830.
Riot Games, Inc. (2024). Riot games api. [link]. Acessado em: 02 de julho de 2024.
Silva, A. L. C., Pappa, G. L., e Chaimowicz, L. (2018). Continuous outcome prediction of league of legends competitive matches using recurrent neural networks. In Proceedings of SBGames 2018. Computing Track – Short Papers.
Spearman, W. (2018). Beyond expected goals. In Proceedings of the 12th MIT Sloan Sports Analytics Conference, Boston.
Vardakis, M., Margetis, G., Chatzakis, I., Apostolakis, K. C., e Stephanidis, C. (2026). Prediction of moba game events based on in-game data. Entertainment Computing, 57:101091.
Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., Kaiser, Ł., e Polosukhin, I. (2017). Attention is all you need. In Advances in Neural Information Processing Systems 30, pages 5998–6008. Curran Associates, Inc.
Yang, Z., Pan, Z., Wang, Y., Cai, D., Liu, X., Shi, S., e Huang, S.-L. (2020). Interpretable real-time win prediction for honor of kings, a popular mobile moba esport. arXiv preprint.
Publicado
29/09/2026
Como Citar
SILVA, Henrique de Macedo Airoso da; LOPES, Yuri Kaszubowski.
From Outcome Prediction to Advantage Construction: A Temporal and Spatial Attribution Framework for League of Legends. In: SIMPÓSIO BRASILEIRO DE JOGOS E ENTRETENIMENTO DIGITAL (SBGAMES), 25. , 2026, Goiânia/GO.
Anais [...].
Porto Alegre: Sociedade Brasileira de Computação,
2026
.
p. 882-893.
DOI: https://doi.org/10.5753/sbgames.2026.25804.
