#5353rd of 70 in Open models
nba-machine-learning-sports-betting
NBA game winner and over/under predictor using XGBoost and neural nets
- Stars
- 1.8k
- License
- no license
- Last commit
- Nov 2026
- Last release
- Jan 2026
- Language
- Python
1,605 on 21 Apr 20261,777 on 10 Nov 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
Predicts NBA moneyline and totals outcomes from team stats (2007-08 onward) and sportsbook odds scraped from SBR, storing data in SQLite. It trains XGBoost, neural network and logistic regression models, then prints win probabilities, expected value and optional Kelly Criterion stake sizing for today's games. A small Flask app browses outputs.
Who it is for: Python users building NBA betting models and expected-value tools
Strengths
- Full pipeline in one repo: data pull, feature build, training and daily prediction
- Fetches live odds from seven named sportsbooks, with manual odds entry as fallback
- Outputs expected value and Kelly Criterion fractions, not just win probabilities
- Backfill flags fill missing stats and odds by date or season
Weaknesses
- No license file, so reuse terms are unclear
- NN training scripts use hard-coded dataset and model paths
- Depends on scraped NBA endpoints and SBR odds, which can change or break
- No Docker setup; requires Python 3.11 with TensorFlow installed manually
What it needs
- Needs Python 3.11, TensorFlow, XGBoost, SQLite, Flask, NBA stats endpoints, SBR odds data
- Models: XGBoost, TensorFlow neural network, Logistic regression
Quick answers
Is nba-machine-learning-sports-betting open source?
Its code is public on GitHub, but the repository has no license, so by default nobody may reuse it.
Also in Open models
See all 70| Rank | Project | Score |
|---|---|---|
| 1 | ruviewWiFi CSI sensing on ESP32 for presence, breathing, heart rate and pose | 94 out of 100 |
| 2 | layaSingle-pass multilingual classifier for typed choice, score and yes/no decisions | 88 out of 100 |
| 3 | rf-detrReal-time transformer models for object detection, segmentation and keypoints | 79 out of 100 |
| #4 | timesfmPretrained time-series foundation model for zero-shot forecasting from Google Research | 78 out of 100 |
| #5 | yolov5PyTorch YOLOv5 object detection, segmentation and classification models with training code | 76 out of 100 |