⚽ Player Performance Predictor

A machine learning project that predicts the future performance ratings and statistics of football (soccer) players based on their historical data.

📖 Project Overview

This project uses a trained machine learning model to forecast how professional football players will perform in the next 3 seasons. It analyzes player statistics like goals, assists, and matches played to estimate their future ratings.

Key Features:

• Predicts player ratings for the next 3 seasons
• Estimates future goals, assists, and matches played
• Includes realistic variance and fluctuations in predictions
• Works with any player in the dataset

📊 Dataset

The dataset contains historical player performance data including:

• Teams: The club the player plays for
• Seasons: Season year(s) (e.g., 2024, 2023/2024)
• Players: Player name
• Matches: Number of matches played
• Goals: Number of goals scored
• Assists: Number of assists made
• Seasons Ratings: Player rating/score for that season

Example data:

InterMiami, 2024, Lionel Messi, 15 matches, 14 goals, 11 assists, 8.5 rating

🚀 How to Use

1. Load and Clean Data

python load_data.py

This script:

• Loads the players.csv file
• Removes rows with missing ratings
• Fills missing numeric values with the median
• Shows data preview

2. Make Predictions

python main.py

When prompted, enter a player's name:

Player To Predict: Lionel Messi

The program will output predictions for the next 3 seasons:

         Player  Season  Predicted Rating  Goals  Assists  Matches
0  Lionel Messi    2025              8.3     13        10       14
1  Lionel Messi    2026              8.1     12         9       13
2  Lionel Messi    2027              7.9     11         8       12

🤖 How It Works

1. Model Loading: The pre-trained model (player_performance_predictor.joblib) is loaded
2. Player Search: Finds the latest season data for the requested player
3. Feature Calculation: Computes key metrics:
   • Goals per match
   • Assists per match
   • Total contributions per match
4. Predictions: For each future season:
   • Applies realistic variance (±10% fluctuation)
   • Calculates derived features
   • Uses the ML model to predict rating
   • Returns all predicted statistics

📁 Project Structure

player-performance-predictor/
├── README.md                              # This file
├── main.py                               # Main prediction script
├── load_data.py                          # Data loading and cleaning
├── players.csv                           # Player statistics dataset
├── player_performance_predictor.joblib   # Trained ML model
├── player_rating_model.pkl               # Alternative model file
├── main.ipynb                            # Jupyter notebook for main logic
├── load_data.ipynb                       # Jupyter notebook for data loading
└── model_training.ipynb                  # Jupyter notebook for model training

🛠️ Requirements

• Python 3.x
• pandas
• numpy
• joblib
• scikit-learn (for model)

Install with:

pip install pandas numpy joblib scikit-learn

📈 Example Predictions

The model predicts realistic performance changes considering:

• Player age and career trajectory
• Historical performance trends
• Random fluctuations (like good/bad seasons)
• Minimum thresholds (to prevent unrealistic negative stats)

💡 Notes

• Enter the exact player name as it appears in the dataset
• Predictions are probabilistic and include realistic variance
• The model works best for players with multiple seasons of data
• Future predictions assume similar playing conditions

👨‍💻 Author

Student project for predicting and analyzing football player performance trends.

---

Enjoy predicting! ⚽