Chess Data Analysis Project

This project analyzes chess games to identify and study player errors (inaccuracies, mistakes, blunders) using Stockfish engine evaluation.

Project Structure

• Clean.py - Main script for analyzing PGN files and detecting errors using Stockfish
• Calculation.py - Processes error data and calculates statistics
• Analytics.py - Generates analytical reports and win rates by error types
• Openings.py - Analyzes opening performance by color
• Prescription.py - (Empty) Future recommendations module

Data Files

Input Files

• MAF13-white.pgn - PGN file with games where you played as White
• MAF13-black.pgn - PGN file with games where you played as Black
• games_raw.csv - Raw game data

Generated Output Files

• games_with_errors.csv - All games with error analysis
• games_with_errors_only_imb.csv - Filtered games with only IMB errors
• errors_imb_with_result_and_phase_player_only.csv - Error analysis by game phase
• phase_error_winrates.csv - Win rates by phase and error type
• opening_stats_by_color.csv - Opening statistics by color

Setup

Quick Setup (Automated)

Run the setup script to automatically configure everything:

./setup.sh

Manual Setup

1. Install dependencies:

   pip install -r requirements.txt

2. Install Stockfish engine:

   Fedora/RHEL:

   sudo dnf install stockfish

   Ubuntu/Debian:

   sudo apt update && sudo apt install stockfish

   Arch Linux:

   sudo pacman -S stockfish

   macOS (with Homebrew):

   brew install stockfish

   Manual installation:

   • Download from https://stockfishchess.org/download/
   • Update the STOCKFISH_PATH in Clean.py to point to your Stockfish executable

3. Verify setup:

   python test_setup.py

4. Prepare your data:

   • Place your PGN files (MAF13-white.pgn, MAF13-black.pgn) in the project directory

Usage

1. Analyze games for errors:

   python Clean.py

   This will process your PGN files and generate error analysis data.

2. Calculate statistics:

   python Calculation.py

   Processes the error data and generates filtered datasets.

3. Generate analytics:

   python Analytics.py

   Creates analytical reports and win rate statistics.

4. Analyze openings:

   python Openings.py

   Generates opening performance statistics by color.

Configuration

Engine Settings (Clean.py)

• DEPTH_BEST - Depth for best move calculation (default: 10)
• DEPTH_PLAYED - Depth for played move evaluation (default: 8)
• STOCKFISH_PATH - Path to Stockfish executable

Error Types

The project analyzes three types of errors:

• Inaccuracy - Minor suboptimal moves
• Mistake - More significant errors
• Blunder - Serious tactical/strategic errors

Requirements

• Python 3.8+
• pandas
• python-chess
• Stockfish engine

Notes

• Update the Stockfish path in Clean.py before running analysis
• The project assumes you have separate PGN files for games where you played White vs Black
• Error thresholds and analysis parameters can be adjusted in the respective Python files