Institutional Repository of Vilnius University
Mašininio mokymosi metodų taikymas sporto rezultatų prognozavimui /
Abstract
dc:descriptionA lot of statistical and machine learning methods have been used for predicting sports match outcomes. These predictions can be applied in sports analytics, team performance evaluation, or betting strategies. This thesis examines the problem of predicting whether NBA teams will exceed the betting lines set by bookmakers, formulating it as a binary classification problem. The aim of the thesis is to apply machine learning methods to this problem and evaluate whether the predictions produced by these models can be useful in a betting strategy. The classification methods used are logistic regression and XGBoost. The models are evaluated according to classification accuracy and the quality of probability predictions. In addition, the profitability and win rate of the betting strategy are assessed. In the study, the models are trained using chronologically split data; a validation set is used to select hyperparameters and probability thresholds, while the final evaluation is performed on the test set. The thesis also discusses how betting results depend on the selected probability threshold.
Degree
thesis:*- Grantor dc:publisher
- Institutional Repository of Vilnius University
- Year dc:date
- 2026
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Radzevičius, Jovaras,
Rights
dc:rights- Statement dc:rights
-
- info:eu-repo/semantics/openAccess
- Language dc:language
- lit
Identifiers
dc:identifier.*- Repository record dc:identifier
- https://repository.vu.lt/VU:ELABAETD308327678&prefLang=en_US