An AutoML-Based Framework for Season-Level League Ranking Prediction in Professional Football

Öz

Accurate forecasting of season-level performance in professional football has gained increasing importance because of its implications for financial planning and strategic decision-making. This study proposes an Automated Machine Learning (AutoML) framework to predict end-of-season point totals and final league rankings in the Turkish Trendyol Süper Lig, conceptualizing cumulative performance as a macro-level forecasting problem. The analysis used data from six consecutive seasons (2018/2019–2023/2024), with out-of-sample validation conducted in the subsequent season. The dataset integrates financial indicators, such as squad market value, with technical performance metrics. Feature engineering expands 17 baseline variables into 32 indicators capturing efficiency, dominance, disciplinary impact, and context-adjusted performance. To mitigate target leakage, variables directly encoding total points are excluded. Two complementary modeling strategies are employed. First, a learning-to-rank approach using XGBoost Ranker directly models league standings. Second, an AutoML-based regression framework estimates the total season points, from which rankings are derived. Model performance is evaluated using Spearman’s rank correlation coefficient and mean absolute error. The ranking-based model achieves a correlation of 0.91, while the AutoML-optimized ExtraTrees Regressor improves this to 0.95, with an average prediction error of approximately 5–6 points. These findings demonstrate the effectiveness of the rank-sensitive predictive framework. Beyond predictive performance, the results indicate that efficiency and dominance-oriented performance indicators play a more decisive role in season-end success than possession-based metrics. This suggests that data-driven recruitment and squad planning strategies should prioritize players who contribute to scoring efficiency and competitive balance rather than purely volume-based attributes.

Anahtar Kelimeler

• AutoML
• Football Analytics
• League Ranking Prediction
• Machine Learning
• Sports Forecasting

Etik Beyan

This study does not involve human participants, animals, or identifiable personal data. Therefore, ethical approval and informed consent were not required.

Profesyonel Futbolda Sezon Düzeyinde Lig Sıralaması Tahmini İçin AutoML Tabanlı Bir Çerçeve

Öz

Profesyonel futbolda sezon düzeyindeki performansın doğru tahmin edilmesi, finansal planlama ve stratejik karar alma süreçleri açısından giderek önem kazanmaktadır. Bu çalışmada, kümülatif performans makro düzeyde bir tahmin problemi olarak ele alınarak Türkiye Trendyol Süper Lig’de sezon sonu puanlarını ve nihai lig sıralamasını tahmin etmeye yönelik Otomatik Makine Öğrenmesi (AutoML) tabanlı bir çerçeve önerilmektedir. Analizde, birbirini izleyen altı sezona (2018/2019–2023/2024) ait veriler kullanılmış ve örneklem dışı doğrulama takip eden sezon üzerinden gerçekleştirilmiştir. Veri seti, takım kadro piyasa değeri gibi finansal göstergeler ile teknik performans ölçütlerini bütünleştirmektedir. Özellik mühendisliği yoluyla 17 temel değişken; verimlilik, üstünlük, disiplin etkisi ve bağlama göre düzeltilmiş performansı yansıtan 32 göstergeye genişletilmiştir. Hedef sızıntısını önlemek amacıyla toplam puanı doğrudan kodlayan değişkenler analizden çıkarılmıştır. Birinci yaklaşımda, XGBoost Ranker ile lig sıralaması doğrudan modellenmiştir. İkinci yaklaşımda ise AutoML tabanlı regresyon ile sezon sonu toplam puanları tahmin edilmiş ve lig sıralaması bu tahminlerden türetilmiştir. Performans, Spearman sıra korelasyon katsayısı ve ortalama mutlak hata kullanılarak değerlendirilmiştir. Sıralama modeli 0,91 korelasyona ulaşırken AutoML ile optimize edilen ExtraTrees Regressor bu değeri 0,95’e yükseltmiş ve yaklaşık 5–6 puanlık ortalama tahmin hatası elde etmiştir. Bulgular, verimlilik ve oyun üstünlüğü odaklı göstergelerin sezon sonu başarısında topa sahip olma ölçütlerinden daha belirleyici olduğunu göstermektedir.

Anahtar Kelimeler

• AutoML
• Futbol Analitiği
• Lig Sıralaması Tahmini
• Makine Öğrenmesi
• Spor Tahminleme
• Spor Tahminleme

Etik Beyan

Bu çalışma insan katılımcılar, deney hayvanları veya kimliği belirlenebilir kişisel veriler içermemektedir. Bu nedenle etik kurul onayı ve katılımcı onamı gerekmemektedir.

Kaynakça

1. Allen, T., Taberner, M., Zhilkin, M., Alexander, J., Harper, D., Rhodes, D., 2025. Is it as simple as run more, win more? Influence of running load on match outcome and final league position across nine consecutive English Premier League football seasons. International Journal of Sports Science and Coaching. https://doi.org/10.1177/17479541251371004
2. Alves, R., 2025. SCORE: A convolutional approach for football event forecasting. International Journal of Forecasting, 41(4). https://doi.org/10.1016/j.ijforecast.2025.02.004
3. Aoki, R. Y. S., Assunção, R. M., Vaz De Melo, P. O. S., 2017. Luck is hard to beat: The difficulty of sports prediction. Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, Part F129685. https://doi.org/10.1145/3097983.3098045
4. Baboota, R., Kaur, H., 2019. Predictive analysis and modelling football results using machine learning approach for English Premier League. International Journal of Forecasting, 35(2). https://doi.org/10.1016/j.ijforecast.2018.01.003
5. Berrar, D., Lopes, P., Dubitzky, W., 2019. Incorporating domain knowledge in machine learning for soccer outcome prediction. Machine Learning, 108(1). https://doi.org/10.1007/s10994-018-5747-8
6. Bornn, L., Cervone, D., Fernandez, J., 2018. Soccer analytics: Unravelling the complexity of “the beautiful game.” Significance, 15(3). https://doi.org/10.1111/j.1740-9713.2018.01146.x
7. Chen, T., Guestrin, C., 2016. XGBoost: A scalable tree boosting system. Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 13-17-August-2016. https://doi.org/10.1145/2939672.2939785
8. Crowder, M., Dixon, M., Ledford, A., Robinson, M., 2002. Dynamic modelling and prediction of English Football League matches for betting. Journal of the Royal Statistical Society Series D: The Statistician, 51(2). https://doi.org/10.1111/1467-9884.00308

9. Decroos, T., Dzyuba, V., Van Haaren, J., Davis, J., 2017. Predicting soccer highlights from spatio-temporal match event streams. 31st AAAI Conference on Artificial Intelligence, AAAI 2017. https://doi.org/10.1609/aaai.v31i1.10754
10. Dixon, M. J., Coles, S. G., 1997. Modelling association football scores and inefficiencies in the football betting market. Journal of the Royal Statistical Society. Series C: Applied Statistics, 46(2). https://doi.org/10.1111/1467-9876.00065
11. Dubitzky, W., Lopes, P., Davis, J., Berrar, D., 2019. The Open International Soccer Database for machine learning. Machine Learning, 108(1). https://doi.org/10.1007/s10994-018-5726-0
12. Geurts, P., Ernst, D., Wehenkel, L., 2006. Extremely randomized trees. Machine Learning, 63(1). https://doi.org/10.1007/s10994-006-6226-1
13. Gunst, R. F., Mason, R. L., 1977. Biased estimation in regression: An evaluation using mean squared error. Journal of the American Statistical Association, 72(359). https://doi.org/10.1080/01621459.1977.10480625
14. Holmes, B., McHale, I. G., 2024. Forecasting football match results using a player rating based model. International Journal of Forecasting, 40(1). https://doi.org/10.1016/j.ijforecast.2023.03.002
15. Hutter, F., Kotthoss, L., Vanschoren, J., 2017. Automated Machine Learning: Methods, Systems, Challenges. Interdisciplinary Mathematical Sciences, 19.
16. Hvattum, L. M., Arntzen, H., 2010. Using ELO ratings for match result prediction in association football. International Journal of Forecasting, 26(3). https://doi.org/10.1016/j.ijforecast.2009.10.002
17. Jacinto, A. L. B., Praça, G. M., Pinheiro, G. de S., da Costa, V. T., 2024. Identification of indicators that predict victory in the five main CONMEBOL and UEFA leagues. International Journal of Sports Science and Coaching, 19(5). https://doi.org/10.1177/17479541241249475
18. Kasiviswanathan, K. S., Sudheer, K. P., He, J., 2016. Quantification of prediction uncertainty in artificial neural network models. Studies in Computational Intelligence, 628. https://doi.org/10.1007/978-3-319-28495-8_8
19. Khan, M. A., 2024. From Auto ML to Stacking Ensembles Advancing EEG Cognitive State Prediction Techniques. Current Trends in Biomedical Engineering & Biosciences, 23(1). https://doi.org/10.19080/ctbeb.2024.23.556103
20. Koopman, S. J., Lit, R., 2015. A dynamic bivariate Poisson model for analysing and forecasting match results in the English Premier League. Journal of the Royal Statistical Society. Series A: Statistics in Society, 178(1). https://doi.org/10.1111/rssa.12042
21. Link, D., 2018. Data Analytics in Professional Soccer. Data Analytics in Professional Soccer. https://doi.org/10.1007/978-3-658-21177-6
22. Liu, G., Luo, Y., Schulte, O., Kharrat, T., 2020. Deep soccer analytics: learning an action-value function for evaluating soccer players. Data Mining and Knowledge Discovery, 34(5). https://doi.org/10.1007/s10618-020-00705-9
23. Maher, M. J., 1982. Modelling association football scores. Statistica Neerlandica, 36(3). https://doi.org/10.1111/j.1467-9574.1982.tb00782.x
24. Noviandy, T. R., Maulana, A., Idroes, G. M., Emran, T. Bin, Tallei, T. E., Helwani, Z., Idroes, R., 2023. Ensemble Machine Learning Approach for Quantitative Structure Activity Relationship Based Drug Discovery: A Review. Infolitika Journal of Data Science, 1(1). https://doi.org/10.60084/ijds.v1i1.91
25. Pearson, K., 1897. Mathematical contributions to the theory of evolution. On a form of spurious correlation which may arise when indices are used in the measurement of organs. Proceedings of the Royal Society of London, 60(359–367). https://doi.org/10.1098/rspl.1896.0076
26. Peeters, T., 2018. Testing the Wisdom of Crowds in the field: Transfermarkt valuations and international soccer results. International Journal of Forecasting, 34(1). https://doi.org/10.1016/j.ijforecast.2017.08.002
27. Simpson, I., Beal, R. J., Locke, D., Norman, T. J., 2022. Seq2Event: Learning the Language of Soccer Using Transformer-based Match Event Prediction. Proceedings of the ACM SIGKDD International Conference on Knowledge Discovery and Data Mining. https://doi.org/10.1145/3534678.3539138
28. Transfermarkt, 2025. Süper Lig—Club market values. Available at: https://www.transfermarkt.com/super-lig/marktwerteverein/wettbewerb/TR1 (accessed 10 December 2025).
29. Tümer, A. E., Koçer, S., 2017. Prediction of team league’s rankings in volleyball by artificial neural network method. International Journal of Performance Analysis in Sport, 17(3). https://doi.org/10.1080/24748668.2017.1331570
30. Uçar, M., İncetaş, M. O., Bayraktar, I., Çilli, M., 2022. Using machine learning algorithms for jumping distance prediction of male long jumpers. Journal of Intelligent Systems: Theory and Applications, 5(2), 145–152. https://doi.org/10.38016/jista.1078474
31. Van Arem, K., Goes-Smit, F., Söhl, J., 2025. Forecasting the Future Development in Quality and Value of Professional Football Players. Applied Sciences (Switzerland), 15(16). https://doi.org/10.3390/app15168916
32. Yousef, K. M. A., El-Haj, A., Mohd, B. J., Dawoud, N., Alnaimat, S., Saleh, A., 2025. Real-Time Football Match Analysis: Leveraging YOLO for Enhanced Object Detection and Possession Tracking. Procedia Computer Science, 272. https://doi.org/10.1016/j.procs.2025.10.176