STGCN-LSTM for Olympic Medal Prediction: Dynamic Power Modeling and Causal Policy Optimization

TL;DR

The paper tackles the challenge of forecasting Olympic medal distributions by introducing a hybrid STGCN-LSTM framework that captures both spatio-temporal country relationships and long-term performance dynamics, augmented with a Zero-Inflated Compound Poisson model to handle structural and random zeros. It advances the field with a dynamic national power weighting scheme, a PageRank-based event influence index, and host-factor analyses, complemented by a robust triple-validation framework (historical backtesting, policy shock simulation, and causal inference). The approach yields actionable insights into policy optimization, such as the impact of coaching mobility, event prioritization, and investment strategies, while providing quantified projections for the 2028 Los Angeles Olympics and emerging medal-winning countries. Practically, the work offers data-driven guidance for allocating resources, prioritizing events, and designing targeted coaching interventions to maximize medal yield under complex, time-varying conditions. The combination of deep learning, causal analysis, and economic-policy modeling positions this framework as a versatile tool for national sports agencies pursuing efficient, evidence-based strategies across varied Olympic contexts.

Abstract

This paper proposes a novel hybrid model, STGCN-LSTM, to forecast Olympic medal distributions by integrating the spatio-temporal relationships among countries and the long-term dependencies of national performance. The Spatial-Temporal Graph Convolution Network (STGCN) captures geographic and interactive factors-such as coaching exchange and socio-economic links-while the Long Short-Term Memory (LSTM) module models historical trends in medal counts, economic data, and demographics. To address zero-inflated outputs (i.e., the disparity between countries that consistently yield wins and those never having won medals), a Zero-Inflated Compound Poisson (ZICP) framework is incorporated to separate random zeros from structural zeros, providing a clearer view of potential breakthrough performances. Validation includes historical backtracking, policy shock simulations, and causal inference checks, confirming the robustness of the proposed method. Results shed light on the influence of coaching mobility, event specialization, and strategic investment on medal forecasts, offering a data-driven foundation for optimizing sports policies and resource allocation in diverse Olympic contexts.

Figures (7)

• Figure 1: Paris Olympics (2024) Medal Table Gold Medal Top 10 Countries
• Figure 2: Medals Prediction Model Flowchart
• Figure 3: STGCN-LSTM Hybrid Model Flowchart
• Figure 4: Zero-Inflated Compound Poisson Model Flowchart
• Figure 5: Model Validation and Assessment Flowchart