 This paper proposes the use of analytical distributions, ADs, to compare the performance of four machine learning models. XGBoost, Random Forest, Artificial Neural Network, and Adaptive Boost in predicting heart disease outcomes. ADs are derived analytically from the data rather than simulated, allowing for easier comparison between studies. The authors found that XGBoost had the best overall performance, with the highest OROC and the highest aggregate score considering six other model metrics. Additionally, the ADs derived from the XGBoost model showed a more normal distribution than the ADs derived from the other three models, suggesting that XGBoost may be a better choice for predictive modeling tasks. This article was authored by Alexander A. Huang and Samuel Y. Huang.