Predicting basketball game outcomes in elite competitions is a complex task influenced by multiple interacting performance factors. This study applied a supervised machine learning (ML) framework to predict EuroLeague game outcomes using team-level game-related statistics. Four algorithms—Logistic Regression (LR), Support Vector Machine (SVM), Random Forest (RF), and Naïve Bayes (NB)—were trained and compared following recursive feature elimination (RFE) to identify the most informative predictors. The dataset comprised comprehensive in-game statistics describing shooting efficiency, rebounding, ball security, and spatial shot distribution. Model performance was evaluated using accuracy, area under the receiver operating characteristic curve (AUC), precision, recall, and F1-score, ensuring both discrimination and calibration assessment. Among the four classifiers, SVM (AUC = 0.922, Accuracy = 0.841) and LR (AUC = 0.933, Accuracy = 0.818) achieved the highest predictive performance, outperforming RF and NB. Feature importance analysis using Shapley Additive Explanations (SHAP) on the best-performing SVM classifier revealed that true shooting percentage (TS%), defensive rebounds (DR), steals (ST), and turnovers (TO) were the most influential predictors of game outcomes. Teams that demonstrated higher shooting efficiency, greater rebounding control, and fewer turnovers showed a significantly higher probability of winning. These results confirm that well-validated and interpretable ML models can accurately predict game outcomes in professional basketball using readily available box-score statistics. The integration of RFE-based feature selection and SHAP interpretability provides transparent, evidence-based insights that can inform tactical decisions, enhance scouting accuracy, and support coaches in developing data-driven performance strategies within elite basketball environments.

This study aimed to identify game-related performance variables that differentiate winning and losing teams in the Basketball Champions League and determine the most significant predictors of game outcomes. Data were collected from 175 games during the 2023-2024 season. The Mann-Whitney U test and logistic regression analysis were employed to evaluate the influence of offensive and defensive metrics on game outcomes. Significant differences were found in key metrics such as shooting percentages, points per game, and defensive rebounds, with winning teams consistently outperforming losing teams. Logistic regression identified steals and defensive rebounds as the strongest predictors of success. Steals increased the odds of victory by 59.6% and defensive rebounds by 48.8%, underscoring the critical role of defensive metrics. Conversely, two-point attempts negatively predicted winning odds, with each additional attempt reducing the likelihood of victory by 32.9%. The findings highlight the importance of efficient shot-making, effective defense, and offensive coordination in basketball performance. These results provide valuable insights for coaches and analysts seeking to enhance team performance and develop effective game strategies.

Background: Decision-making during late-game possessions in basketball is considered one of the most essential aspects of winning a game. This study examined key determinants of late-game shot-making success in closely contested EuroLeague basketball matches, aiming to optimize offensive decision-making in clutch situations. Material and Methods: An observational analysis from the last two minutes of 83 EuroLeague games recorded 471 clutch shot attempts. Logistic regression and chi-square analyses assessed the impact of offense time, shot range, defensive pressure, game status, and offensive tactical strategies on shot success. Results: The results showed that paint clutch shots (63.9% success) were significantly more effective than mid-range (30.0%) and three-point attempts (33.5%). Defensive pressure was a critical factor, with uncontested shots (58.6%) outperforming contested attempts (36.4%). Quick offensive execution and big player position also influenced the shot outcome. The primary key determinants in predicting clutch shot success included shot range, defensive pressure, offense time, and game status. Conclusions: By prioritizing high-percentage shot opportunities, leveraging cooperative movement-based offensive strategies, and utilizing big players effectively, teams can enhance their performance in high-pressure moments. These findings provide actionable insights for coaches and players, emphasizing the need for data-driven decision-making to optimize late-game performance.