A data-driven betting decision-support application that estimates match outcome probabilities using the Poisson model and highlights potential value bets.
Bet-Picker is a web application designed to assist with sports betting analysis by comparing statistically derived probabilities against bookmaker odds.
The system calculates outcome probabilities, converts odds into implied probabilities, and identifies bets with positive expected value — helping transform betting decisions from intuition-based to data-informed.
⚠️ This project is intended for analytical and educational purposes only and does not guarantee profit.
- Probability estimation using the Poisson distribution
- Automatic comparison between model probabilities and bookmaker odds
- Detection of potential value bets
- Profit calculation based on user-defined wager
- Clean and responsive dashboard UI
- Serverless backend with Firebase Functions
- Modular architecture allowing future model extensions
Frontend
- Svelte
- TypeScript
- Vite
Backend
- Firebase Functions
- Node.js
Architecture
- Monorepo structure with separated frontend and backend
- Root-level scripts for simplified development workflow
Bet-Picker/
│
├── client/ # Svelte frontend
├── functions/ # Firebase serverless backend
├── docs/ # README images
└── package.json # Root scripts
Make sure you have installed:
- Node.js (LTS recommended)
- npm
- Firebase CLI
Install Firebase CLI globally:
npm install -g firebase-toolsInstall dependencies for the root project and sub-packages:
npm install
npm install --prefix client
npm install --prefix functionsStart the frontend:
npm run start:devStart Firebase Functions locally:
npm run start:functionsThe frontend development server will typically be available at:
http://localhost:5173
- The application retrieves match odds.
- The Poisson model estimates the probability of each outcome.
- Bookmaker odds are converted into implied probabilities.
- The system compares model probabilities with implied probabilities.
- Bets with positive expected value are highlighted along with projected profit.
- Root-level orchestration scripts for cleaner developer experience
- Clear separation between frontend and backend services
- Serverless architecture for scalability
- Strong emphasis on statistical modeling rather than heuristic predictions
- Designed with extensibility in mind (additional models can be integrated)
- Historical backtesting
- Advanced probability models (xG, ELO, ML-based predictions)
- Arbitrage detection
- Bankroll management tools
- Docker support
- CI/CD pipeline
This project is licensed under the MIT License.
Betting involves financial risk.
This software is provided for research and educational purposes only.