🏆 World Cup 2026 Monte Carlo Simulator

A complete Data Science pipeline and simulation engine that predicts the winner of the expanded 48-team FIFA World Cup 2026.

Instead of relying on simple guesswork, this project pulls historical match data, builds a custom Elo rating system, calculates team-specific Attack/Defense Strengths, and runs 10,000 Monte Carlo simulations to find the mathematical probabilities of each nation lifting the trophy.

(Example output of 10,000 simulations. Probabilities adjust dynamically based on the latest data).

---

🚀 Key Features

• ETL Pipeline (datafetcher.py): Automatically fetches, cleans, and time-filters historical international match data (from August 2018 onwards to ensure modern form relevance).
• Custom Elo Engine (elo_engine.py): Calculates dynamic Elo ratings for all national teams based on actual historical results.
• Statistical Modeling (team_stats_engine.py): Calculates expected Attack Strength (AS) and Defense Strength (DS) metrics relative to global averages.
• Poisson Match Simulator (match_simulator.py): Uses statistical distribution to simulate realistic match scores (xG).
• Complex 48-Team Bracket Logic (tournament_logic.py): Implements FIFA's 2026 format, including the "8 best third-place teams" logic.
• Monte Carlo Engine (main.py): Simulates the entire tournament 10,000 times.

---

🛠️ Installation & Usage

1. Clone the repository:

   git clone https://github.com/LaxenOsund/world-cup-simulation.git
   cd world-cup-simulation

2. Install dependencies:

   pip install pandas matplotlib

3. Run the Simulation:

   python main.py

📂 Project Structure

world-cup-simulation/
├── main.py                  # Main orchestrator (Monte Carlo loop)
├── match_simulator.py       # Core match logic (Poisson distribution)
├── tournament_logic.py      # FIFA rules and bracket routing
├── datafetcher.py           # Data cleaning & ETL pipeline
├── engines/                 # Mathematical Core
│   ├── elo_engine.py        # Calculates Team Elo ratings
│   └── team_stats_engine.py # Calculates Attack/Defense form (AS/DS)
└── results.csv              # Filtered dataset (International results)

🧠 The Math Behind the Magic

The engine calculates an expected goal (xG) value for each team based on Elo differences and Attack/Defense ratings. These variables are fed into a Poisson Distribution algorithm, generating realistic football scores.

---

Created as a data science portfolio project.