Olympic Medal Forecasting through Big Data Analytics: A Sport Competitiveness Index Model

The task of accurately predicting the Olympic medal table stands as a pivotal challenge within sports analysis. This prediction not only serves to evaluate the competitive strengths of various nations but also acts as a barometer for the efficacy of national sports strategies. Recognizing the complexities introduced by factors such as the host country's influence, variations in sports events, and individual athlete capabilities, this study proposes a medal prediction model grounded in the random forest regression algorithm and competitiveness quantification. The model amalgamates multi-dimensional data from all Summer Olympics since 1896. This data set encompasses the medal distributions of past Olympic Games, the participation and achievement records of all athletes, and specifically quantifies five core determinants. The model calculates a competitiveness index, expressed as points, for countries or individual athletes in specific events, offering a quantitative assessment of their advantageous position. Furthermore, the predictive results of this model have been validated on existing datasets from previous sessions, demonstrating excellent reliability. The model accurately forecasted the medal table for the 2028 Los Angeles Olympics, predicting China's victory with 34 gold medals and a total of 83 medals. This outcome aligns with historical trends, underscoring the model's feasibility.