bszek213 / ml_mlb

Deep Learning to Predict MLB Outcomes

MLB Game Predictions

Machine learning that predicts the outcomes of any MLB game. Data are from 2014 - 2023 seasons. Current accuracy on test data:

• Teams that I have a >70% prediction probability: 'KCR', 'ATL'
• Regression DNN has a validation RMSE of 1.15 runs

Usage

python mlb_ml_classify_deep_learn.py tune or python mlb_ml_classify_deep_learn.py notune
#check accuracy of running averages
python mlb_ml_classify_deep_learn.py test

Correlated features   >= 0.90: ['RBI', 'onbase_plus_slugging', 'ER', 'strikes_total']
Number of samples: 62718

### Current prediction accuracies - DNN
Validation Accuracy: 0.985
Train Accuracy: 0.986
Validation Loss: 0.03
Train Loss: 0.037

Running average outcomes

Contributing

Pull requests are welcome. For major changes, please open an issue first to discuss what you would like to change.