Study relationship between chess playing accuracy scores (like those on Chess.com or Lichess) and Elo strength (online or over-the-board).
This repository is dedicated to investigating the relationship between calculated chess engine accuracy scores (e.g., Stockfish evaluations) and established player strength measures (e.g., Elo rating, FIDE/USCF titles).
It serves as a central hub for the source code, analysis notes, and collated data necessary for this research.
The project is organized into four main directories to separate logic from assets:
| Directory | Content | Licensing Note |
|---|---|---|
src/ |
Python scripts for data processing, accuracy score calculation, and statistical analysis. | MIT License |
data/ |
Collated CSV files containing game records and raw data for analysis. | CC BY 4.0 |
binaries/ |
Executable files, including the Stockfish engine binary used for generating accuracy scores. | N/A (External License) |
notes/ |
Plain text logs, research findings, and documentation notes. | CC BY 4.0 |
To use the code and contribute to this project, you will need the following tools installed on your system:
This repository uses Git Large File Storage (Git LFS) to manage the large CSV data files and the Stockfish binary efficiently.
# Install Git LFS (needed on both Windows and Debian)
git lfs install
# Clone the repository
git clone https://github.com/ZAF-prog/chess-accuracy-tools/