A machine learning model predicts the U.S. has only a 1% chance of winning the World Cup final on home soil, after running 100,000 tournament simulations based on team strength, player values, and historical match data.

What happened: Researchers built a machine learning algorithm that combines statistical models, bookmaker odds, player ratings, and market values to forecast World Cup outcomes. The model ran 100,000 simulations and ranked Spain as the favorite with a 14.5% winning probability, followed by England and France at 12.4% each, and Germany at 11.2%. For the U.S., the algorithm gives it a 78% chance of reaching the Round of 32 but only a 1% probability of winning the final at MetLife Stadium on July 19.

Why it matters: The forecast shows how data science can now replace guesswork in sports prediction. The model draws on eight years of national match history, real-time bookmaker opinions, individual player contributions tracked at club and national levels, and even country-level economic factors like GDP per capita. This grounding in multiple data sources makes the probabilities more credible than traditional forecasting methods.

What to watch: The U.S. team's actual performance in the knockout stage will test the model's accuracy. In past tournaments, the researchers' team correctly predicted the winner of the 2019 Women's World Cup but missed the favorites in the 2023 Women's World Cup and 2022 men's World Cup, though both winners were ranked as serious contenders.