Purpose <p>Slips and falls are leading causes of injury-related incidents, and one of the most effective ways to prevent slips is to use slip-resistant footwear. To evaluate footwear slip resistance, two main experiments are conducted: (1) human-centered tests, and (2) mechanical tests. Human-centered slip resistance tests better capture real-world locomotion dynamics than mechanical tests, but they are costly and time-consuming. This study aimed to develop a predictive model that estimates human-centered slipperiness score called Maximum Achievable Angle (MAA) outcomes using mechanical slip resistance measurements, enabling accessible footwear evaluation.</p> Methods <p>Thirty-seven winter footwear samples were tested using both human-centered MAA testing and the SATRA STM 603 mechanical slip resistance tester on wet and dry ice surfaces. Ten linear and non-linear regression models were trained using a nested 5-fold cross-validation approach with unseen footwear in each fold to support generalizability. Model performance was evaluated using Root Mean Square Error (RMSE), Mean Absolute Error (MAE), Mean Absolute Percentage Error (MAPE), coefficient of determination (R²), and Bland–Altman analysis.</p> Results <p>Across all models, the Ridge model produced the most conservative and consistent results, achieving an RMSE of 2.73° and MAE of 2.27° on wet ice, and an RMSE of 1.58° and MAE of 1.33° on dry ice. Although the results are close to the ± 1° acceptable range of MAA, slight deviations in some cases highlight opportunities for future model improvement.</p> Conclusion <p>Mechanical slip resistance metrics can be used to reasonably predict human-centered MAA outcomes. While prediction accuracy is promising, further refinement is needed to fully meet the precision requirements for footwear slip resistance certification.</p>