F1 Predictor

A Formula 1 prediction engine combining Monte Carlo simulation with Machine Learning for Grand Prix outcomes.

Key Features

• Grand Prix Centric: Organizes data by Season and Round.
• ML / Monte Carlo: Uses Machine Learning to predict probabilistic outcomes (Win/Podium %).
• SHAP Explainability: Explains why a driver is predicted to finish in a certain position.
• Sprints: Full support for Sprint Weekends.

Model Performance

The core ML engine (Random Forest Regressor) is trained on historical lap-time pace and qualifying results from 2022 to the present (96+ Grand Prix events). By evaluating raw car and driver pace, the model is highly effective at identifying the true grid hierarchy:

• Points Finish Accuracy (Top 10): 78.2% (F1 Score: 0.760)
• Mean Absolute Error (MAE): 3.20 positions (Accounting for Unpredictable race-day chaos like incidents, crashes and safety cars)

Clone the Repository

git clone https://github.com/BLShaw/f1-predictor.git

Install the dependencies

pip install -r requirements.txt

Run Application

streamlit run app.py

Screenshots

License

MIT License.