Can Statcast Data Improve MLB Player Performance Predictions? — Beating Marcel with LightGBM
📰 Dev.to · YMori
Learn how to improve MLB player performance predictions by 12.1% for batters and 4.0% for pitchers using Statcast data and LightGBM
Action Steps
- Collect MLB Statcast tracking data for exit velocity, barrel rate, Whiff%, and Stuff+
- Preprocess the data for use in a machine learning model
- Train a LightGBM model using the preprocessed data to predict player performance
- Compare the results to the Marcel projection system
- Evaluate and refine the model to achieve the best possible predictions
Who Needs to Know This
Data scientists and analysts on a sports team can benefit from this approach to gain a competitive edge in predicting player performance
Key Insight
💡 Statcast data can significantly improve the accuracy of MLB player performance predictions when used with a suitable machine learning model like LightGBM
Share This
🏟️ Boost MLB player performance predictions by 12.1% for batters and 4.0% for pitchers with Statcast data and LightGBM! 📈
Key Takeaways
Learn how to improve MLB player performance predictions by 12.1% for batters and 4.0% for pitchers using Statcast data and LightGBM
Full Article
I tried to beat the Marcel projection system using MLB Statcast tracking data (exit velocity, barrel rate, Whiff%, Stuff+). Here's what happened: +12.1% improvement for batters, +4.0% for pitchers.
DeepCamp AI