Unlocking Olympic Success: Predictive Modeling and Strategic Insights for the 2028 Games

Abstract

This study aims to predict the medal distribution of the 2028 Olympic Games in Los Angeles by constructing multivariate linear regression and random forest regression models, combining the characteristics of historical data, the advantages of the host country, the number of athletes, and the proportion of gold medals. The results show that the total number of gold medals in the 2028 Olympic Games is expected to be about 255, and the total number of medals is about 714. Feature engineering analysis showed that the average number of gold medals and total medals in the past three years were the key predictors. Model comparisons show that random forests are more advantageous in handling non-linear relationships, but have a tendency to overestimate. The study recommends that National Olympic Committees differentiate the allocation of resources, pay attention to the special needs of the host country, and analyze the data of abnormal years in depth to improve the prediction accuracy.