title	Chess Analysis
description	A chess game analysis platform that classifies moves, detects openings, and generates detailed accuracy reports using engine evaluation.
image	/project/chess-analysis.jpeg
technologies	TypeScript Node.js Express Chess.js PGN Parser Stockfish HTML CSS Docker Vercel
github	https://github.com/mr-anvesh/chess-analysis
live	https://chess-analysis.anvesh.dev
timeline	1 month
role	Full Stack
team	Solo
status	completed
featured	true
challenges	Accurate move classification across edge cases Brilliant move detection with sacrifice validation Engine line evaluation and cutoff scoring Opening book integration with PGN parsing Real-time board rendering and eval bar sync
learnings	Chess engine integration and evaluation scoring Move classification logic with centipawn thresholds PGN parsing and virtual board simulation Full-stack TypeScript project architecture Visual report design for game analysis
isPublished	true

Chess Analysis

Overview

A chess game analysis platform that parses PGN game data, simulates positions with a virtual board, evaluates moves via a chess engine, and generates detailed per-move reports — classifying each move and computing accuracy scores for both players.

What Users Can Do

• Analyse Games: Paste a PGN to trigger move-by-move engine analysis with classification and evaluation.
• Review the Report: Navigate through an interactive board replay with an eval bar, eval graph, and engine suggestions.
• Understand Accuracy: See overall accuracy scores and a full breakdown of move classifications (Brilliant, Great, Best, Excellent, Good, Inaccuracy, Mistake, Blunder).

Why I Built This

I built this to deeply understand how chess engines evaluate positions and how platforms like Chess.com generate their game reports:

• Existing tools are black-box — I wanted to build the classification logic from scratch.
• It was an opportunity to explore engine integration, position analysis, and interactive frontend design.

Tech Stack

TypeScript Node.js Express Chess.js PGN Parser Stockfish (engine evaluation) HTML / CSS Docker Vercel

Move Classification System

Each move is classified based on evaluation loss relative to the engine's top line:

Classification	Description
Brilliant	Sacrifice of a non-pawn piece that is tactically sound
Great	Only move to recover from a blunder
Best	Matches the engine's top move
Excellent	Minimal evaluation loss
Good	Small evaluation loss
Inaccuracy	Moderate evaluation loss
Mistake	Significant evaluation loss
Blunder	Large evaluation loss or missed mate
Book	Named opening theory move
Forced	Only legal move in the position

After Launch and Impact

• Built a full move classification pipeline from PGN parsing through engine evaluation to report generation.
• Implemented brilliant move detection with sacrifice and pin validation logic.
• Delivered an interactive frontend with board replay, eval bar, eval graph, and engine line suggestions.
• Deepened understanding of chess engine scoring, positional analysis, and TypeScript full-stack architecture.

Future Plans

• Project Completed