Data science has become an essential component in professional sports, particularly for predicting team performance and outcomes. This study aims to develop and evaluate machine learning models that accurately predict game outcomes in the Chinese Professional Baseball League (CPBL). Method: A total of 859 games from the 2021 to 2023 regular seasons were analyzed, using both traditional baseball statistics and advanced sabermetric indicators such as the Weighted Runs Created Plus (wRC+), Weighted Runs Above Average (wRAA), and Percentage of Leadoff Batters on Base (PLOB%). Five machine learning models—decision tree, logistic regression, Neural Network, Random Forest, and XGBoost—were constructed and assessed through a five-fold cross-validation. Evaluation metrics included accuracy, F1 scores, sensitivity, specificity, and the AUC-ROC. Results: Among the models, logistic regression and XGBoost achieved the highest performance, with an accuracy ranging from 0.89 to 0.93 and an AUC-ROC from 0.97 to 0.98. The feature importance and SHapley Additive exPlanations (SHAP) analysis revealed that the wRC+ and PLOB% were the most influential predictors, reflecting the offensive efficiency and pitching control. Conclusion: The results suggest that combining interpretable machine learning with sabermetrics provides valuable insights for coaches and analysts in professional baseball. Furthermore, incorporating performance weighting based on game context may further enhance model accuracy. This research demonstrates the potential of data-driven strategies in sports analytics and decision-making.

The ball–strike count resets at each plate appearance, continuously shifting strategic advantage between pitcher and batter. Prior research has focused on total count effects, leaving a gap regarding whether different sequential orders of balls and strikes within equivalent count situations produce meaningful performance differences. Purpose: This study investigated whether ball–strike progression sequences significantly affect collegiate batting performance. Methods: T Drawing on 9454 plate appearances from 120 University Baseball League games, one-way ANOVA with Bonferroni post hoc comparisons assessed sequential order effects across AVG, OBP, SLG, and OPS. Results: The results indicate that sequential order produced statistically significant differences across all four indicators at the 2nd, 3rd, and 4th pitch situations (ps < 0.05), while no significant differences emerged at the 1st pitch or full count. Practical Implications: These findings demonstrate that the path through which a count develops systematically influences batting performance, with sequence effects significant at the 2nd, 3rd, and 4th pitch situations but absent at the initial pitch and full count—consistent with the strategic constraints inherent to these boundary situations. Future research should incorporate pitch type, pitch location, and game-context variables to further clarify the dynamic interaction between pitchers and batters.

The literature on speed–accuracy trade-off (SAT) in motor control has evidenced individuality in how individuals trade moments (e.g., mean and variance) of spatial and temporal errors. These individual tendencies could grasp tendencies of the system given previous experiences and constraints of the organism, a signature of the system control. Nonetheless, such tendency must be robust to small perturbations. Thirty participants performed nine conditions with different time and spatial criteria over 2 days (scanning). In between these scanning conditions, individuals performed a practice condition that required modifications of the individuals’ preferred spatial and temporal tendency in the SAT. Our results demonstrated that there were no systematic effects of practice in SAT preferences. However, individual analyses demonstrated significant changes for 25 out of 30 individuals. The latter either attests against a consistent preference or to a more complex characterization of individual SAT tendencies.

The literature of Speed-Accuracy Trade-Off (SAT) in motor control has evidenced individuality in the preference to trade different aspects (mean, variance) of spatial and temporal errors. Nonetheless, to the best of our knowledge, how robust this preference is has not been properly tested. Thirty participants performed nine conditions with different time and spatial criteria over two days (scanning). In-between these scanning conditions, individuals performed a practice condition that required modifications of the individuals’ preferences in SAT. Through Bayesian analyses, we found that, despite individuals demonstrating changes during practice, decreasing movement time, they did not modify how they performed the scanning conditions. This is evidence for a robust SAT individual tendency. We discuss how such individuality could modify how individuals perform within/between SAT criteria, and what this means for interpretation of results.