Inspiration
We were fascinated by the idea that nations excelling in Olympic sports might also be more likely to reach for the stars. Both Olympic success and space exploration require massive national investments in education, technology, and long-term strategic planning. We wondered: do countries that dominate the podium also dominate space? This curiosity led us to explore whether there's a measurable connection between athletic achievement and cosmic ambitions.
What it does
Our project analyzes the relationship between Olympic performance and space program development across multiple countries and decades. We track how nations' Olympic medal counts, GDP per capita, and other indicators change around the time they first send astronauts to space. The system creates visualizations showing performance trends ±10 years around each country's first space mission and uses machine learning to predict which countries are likely to develop robust space programs based on their Olympic achievements.
How we built it
We integrated three major datasets: Olympic athlete records, astronaut mission data, and GDP indicators spanning over a century. We created weighted medal scoring systems (Gold=3, Silver=2, Bronze=1) and developed "event windows" to examine performance changes around space program milestones. The biggest technical challenge was harmonizing country names across different time periods and political changes. We built twin Random Forest models to predict space program success and created comprehensive visualizations using Python, pandas, matplotlib, and seaborn to reveal patterns in the data.
Challenges we ran into
The most significant challenge was data cleaning and country name matching across datasets from different eras - handling cases like "Soviet Union" vs "U.S.S.R./Russia" vs "Russia" or dealing with countries that changed names or split apart. We also had to account for missing data, different time scales, and ensuring our machine learning models could handle imbalanced datasets where some countries have extensive space programs while others have none.
Accomplishments that we're proud of
We successfully created a comprehensive analysis framework that reveals fascinating patterns between Olympic success and space program development. Our twin machine learning models achieve strong predictive accuracy, and our visualizations clearly show how countries like the United States, China, and Japan exhibit different patterns of Olympic performance around their space program launches. The event window analysis provides compelling evidence that national achievements in sports and space often go hand-in-hand.
What we learned
We discovered that countries with consistent Olympic success do tend to develop more robust space programs, supporting our hypothesis that both achievements reflect similar underlying national capabilities. We learned advanced data integration techniques, the importance of robust data cleaning pipelines, and how to build meaningful visualizations that tell a story across multiple decades. The project also taught us about the challenges of working with historical data and the need for careful consideration of political and temporal context.
What's next for From Kármán Line to Finish Line
We plan to expand the analysis to include more granular sports categories (summer vs winter Olympics, specific sports that might correlate more strongly with STEM education), incorporate additional economic and educational indicators, and extend the machine learning models to predict not just whether countries will develop space programs, but when and how extensively. We're also interested in exploring whether the relationship holds for private space companies and commercial space achievements, not just government programs.
Log in or sign up for Devpost to join the conversation.