This study evaluated the performance of five different machine learning programs predicting the tibial intramedullary nail length based on patient demographic data. Data was collected on patients who underwent surgery with a tibial intramedullary nail from October 2022 to October 2024, collecting height, weight, age, gender, and the length of intramedullary nail used. This data was then carefully cleaned before given to the models for analysis, minimizing biases and missing data entries. The five models were: multiple linear regression, decision tree, random forest, support vector regression (SVR) and XGBoost. To assess accuracy, mean absolute error (MAE) and root mean squared error (RMSE) were calculated, smaller values indicating a more precise prediction. Of the five models, XGBoost performed the best, with the lowest RMSE (9.15 mm) and MAE (7.56 mm). The random forest model did score the highest coefficient of determination (0.874), indicating that it captured 87.4% of variance in the data. To evaluate the XGBoost for consistency, cross-validation was conducted five times, and the model scored an average RMSE of 9.21 ± 1.88 mm and a coefficient of determination of 0.811. Overall, XGBoost performed the best of all five models, with both accuracy and consistency despite a slightly lower coefficient of determination than the random forest model. All models additionally indicated that height was the most influential factor in determining nail length, then weight and age, with gender being minimally important.