Abstract
This study explores the application of neural networks to baseball statistics to predict player performance and at-bat outcomes. Traditional baseball analytics rely on statistical models such as linear regression and decision trees. However, recent advancements in artificial intelligence have enabled deep learning models to uncover hidden patterns and provide more accurate predictions. This thesis implements a feedforward neural network trained on historical player and team statistics to forecast key performance metrics. By comparing the model’s results to conventional statistical methods, this research aims to evaluate the effectiveness of neural networks in enhancing baseball analytics. Among several network configurations tested, we found that three hidden layers with upsampling techniques worked best.
Advisor
Visa, Sofia
Department
Computer Science
Recommended Citation
Orona, Tahj Tuari, "Neural Network For Predicting the Next Outcome in a Major League Baseball Game" (2026). Senior Independent Study Theses. Paper 13371.
https://openworks.wooster.edu/independentstudy/13371
Keywords
Neural Network
Publication Date
2026
Degree Granted
Bachelor of Arts
Document Type
Senior Independent Study Thesis
© Copyright 2026 Tahj Tuari Orona