Formula 1 Fans Using AI to Predict Race Results
Ahead of a grand prix weekend, Formula 1 enthusiasts often share their predictions about who will emerge victorious on Sunday. Data scientist Mariana Antaya took this common practice to the next level by developing a machine learning model designed to predict F1 race results. So far, her model has successfully called the winners of three loving grand prix events this season.
“I’m a huge Formula 1 fan,” Antaya explained in an interview with Motorsport.com. “Machine learning algorithms are widely used in Formula 1 by teams. I wanted to see how accurately we could predict race winners using available data.” To create her model, Antaya started by comparing 2024 race results with 2025 qualifying performances, using lap times from the previous year’s Australian Grand Prix as a data source. After removing rookies from the program due to lack of benchmark data, she trained her model using a gradient boosting tool. The model correctly predicted Lando Norris as the winner at Albert Park.
The project gained momentum as the F1 community rallied around Antaya’s efforts, suggesting additional features to improve the model. “I wanted it to be a crowdsourced project where the audience could suggest features like weather data or practice session inclusion,” Antaya added. The model has continued to predict race winners correctly, including Max Verstappen’s victory at the Japanese Grand Prix after incorporating weather data and wet-weather driver performance.
The model’s accuracy has impressed even F1 teams, with several engineers reaching out to Antaya. “I’ve been shocked by the response. I have no idea how teams do it, but I hope I’m on the right track,” she said. Antaya plans to experiment with more complex machine learning processes to increase accuracy and reduce the mean absolute error of her model before the Miami Grand Prix.
While the model has shown promise, Antaya acknowledges the inherent unpredictability of F1. “There’s always going to be a barrier to perfect prediction. Events like safety cars can trigger unforeseen outcomes.” To address this, she may incorporate additional data such as past crash percentages during races.
The success of Antaya’s model demonstrates the growing intersection of AI and Formula 1, both in team strategies and fan engagement. As the technology continues to evolve, it will be interesting to see how it shapes the future of the sport.
