DeepShot
NBA game predictor using rolling averages & ML
Listed in categories:
Artificial IntelligenceBasketballGitHub
Description
DeepShot is a machine learning model designed to predict NBA game outcomes using advanced team statistics and rolling averages. It combines historical performance trends with contextual game data to deliver highly accurate win predictions.
How to use DeepShot?
To use DeepShot, clone the repository from GitHub, navigate into the directory, install the required dependencies, and run the model training and evaluation workflow to create the model file. Finally, execute the main program to start predictions.
Core features of DeepShot:
1️⃣
Data-Driven Predictions using advanced rolling averages from Basketball Reference
2️⃣
Real-Time Interface for visualizing upcoming matchups and model predictions
3️⃣
Weighted Stats Engine that calculates long-term form using Exponentially Weighted Moving Averages (EWMA)
4️⃣
Key Stats Highlighting to display important stats and differences between teams
5️⃣
Cross-Platform Support for Windows, macOS, and Linux
Why could be used DeepShot?
| # | Use case | Status | |
|---|---|---|---|
| # 1 | Predicting outcomes of NBA games for fans and analysts | ✅ | |
| # 2 | Assisting sports bettors in making informed decisions | ✅ | |
| # 3 | Providing insights for coaches and teams based on statistical analysis | ✅ | |
Who developed DeepShot?
DeepShot is developed by Francesco Sacco, who utilizes various open-source libraries and historical data to create a predictive model for NBA games. The project is hosted on GitHub and encourages user feedback.