Football, being one of the most popular sports in the world, has been open to the betting and gambling world to try and predict the outcomes of its matches. Yet football is a very complex sport and is notoriously hard to predict. In this study, we attempted to predict the football match outcomes of the Indian Super League team, Kerala Blasters, while comparing three algorithms, Random Forest, Logistic Regression and k-Nearest Neighbors (KNN) to understand which model and features work best across datasets. These algorithms were applied to different sets of features, pre-game and in-game features, to determine which led to the most accurate and precise results. We also applied artificial sampling to some models to test if a more balanced dataset found better results. Results showed that Logistic Regression without using SMOTE achieved the highest accuracy (75%) and precision (42%) compared to the other models. Thus, our findings suggest that Logistic Regression is the best model for this dataset. Future work might extend the comparison to more models and apply oversampling techniques to models that don’t work as well with unbalanced datasets.

This study compared three ML models namely, Random Forest, Logistic Regression and KNN, and applied variations of features to it. The first conclusion we infer is that Logistic Regression is the most suitable model to use for this dataset as it is able to handle class imbalances to study whether just pre-game or including in-game features led to better predictions. We can conclude that Random Forest is not suitable for handling unbalanced classes and that perhaps a Weighted Random Forest model would be more suitable to test on this dataset [10].

[1]Daniel Berrar, Philippe Lopes, and Werner Dubitzky, “Incorporating domain knowledge in machine learning for soccer outcome prediction,” Machine-mediated learning, vol. 108, no. 1, 2019, doi: 10.1007/S10994-018-5747-8. [2]Mike Maher, “Modelling association football scores,”, vol. 36, 1982, doi: 10.1111/J.1467-9574.1982.TB00782.X. [3]Darwin Prasetio, and Dra. Harlili, “Predicting football match results with logistic regression,” 2016 International Conference On Advanced Informatics: Concepts, Theory And Application (ICAICTA), 2016, doi: 10.1109/ICAICTA.2016.7803111. [4]F. Sjöberg and S. Azimi Rashti, “Football match prediction using machine learning,” thesis, Åbo Akademi University, 2023. [Online]. [5]Fátima Rodrigues, and Ângelo Pinto, “Prediction of football match results with Machine Learning,” Procedia Computer Science, vol. 204, 2022, doi: 10.1016/J.PROCS.2022.08.057. [6]Pakawan Pugsee, and Pattarachai Pattawong, “Football Match Result Prediction Using the Random Forest Classifier,” ICBDT2019, 2019, doi: 10.1145/3358528.3358593. [7]Fbref, “Football Statistics and History,” FBref.com, 2024. https://fbref.com/en/ [8]scikit-learn, “scikit-learn/scikit-learn,” GitHub, Apr. 15, 2019. https://github.com/scikit-learn/scikit-learn [9]R. van den Goorbergh, M. van Smeden, D. Timmerman, and B. Van Calster, “The Harm of Class Imbalance Corrections for Risk Prediction models: Illustration and Simulation Using Logistic Regression,” Journal of the American Medical Informatics Association, vol. 29, no. 9, Jun. 2022, doi: https://doi.org/10.1093/jamia/ocac093. A. More, and Dipti P Rana, “Review of random forest classification techniques to resolve data imbalance,” 2017 1st International Conference on Intelligent Systems and Information Management (ICISIM), 2017.

© 2025 International Journal of Engineering and Techniques (IJET).

Digital Object Identifier (DOI)

IJET assigns a unique DOI to every accepted article via Zenodo for persistent linking and citation. Your article’s DOI: https://doi.org/{{doi}}

Open DOI About Zenodo DOI

Close