Olympic Medal Prediction Analysis Based on The Random Forest Model

Abstract

During the 2024 Paris Olympics, both individual events and the overall medal tally have garnered significant attention. As a key indicator of a nation's competitive strength, the number of Olympic medals serves as a critical reference for countries preparing for the Games. This paper addresses three core questions: how to predict the medal rankings for the 2028 Los Angeles Olympics, the likelihood of countries that have never won medals securing their first, and the impact of Olympic events on medal rankings. By combining the Random Forest algorithm and the LSTM model to predict feature variables (such as gender ratio and number of participants), a high-precision medal prediction model was constructed, achieving an R² score of 0.929 and an MSE of 1.78098. For countries that have never won a medal, nine countries were selected, and their probability of winning a medal in 2028 was predicted. Additionally, the Random Forest Regressor model was used to analyze the importance of sports events, identify key events for each country, and reveal the strategic impact of host country event selection. The conclusions indicate that this model provides a scientific basis for countries to optimize their Olympic training strategies and resource allocation.