Opendatabay APP

Chess Game Analytics Data

Product Reviews & Feedback

Tags and Keywords

Chess

Elo

Pgn

Fen

Blitz

Trusted By
Trusted by company1Trusted by company2Trusted by company3
Chess Game Analytics Data Dataset on Opendatabay data marketplace

"No reviews yet"

Free

About

A large collection of over 60,000 chess matches sourced directly from the Chess.com platform. This data product was generated utilizing the Chess.com API and encompasses more than twenty variables related to player performance and game dynamics. It is designed for chess enthusiasts and members interested in detailed analytics and study of competitive play. Key attributes include individual player ELO ratings, detailed game results, and information on time controls and rulesets.

Columns

The dataset contains 14 fields detailing game specifics and player performance:
  • white_username, black_username: The unique account identifiers for the participants in the match.
  • white_id, black_id: Links that point to additional external details about the respective players.
  • white_rating, black_rating: The ELO rating of the player at the time of the match. The mean rating for both colours is approximately 1,250, with ratings extending as high as 3,172.
  • white_result, black_result: The specific outcome condition of the game for each player, such as a win, loss, checkmate, draw, or resignation. 'Win' is the most frequently recorded result for both white (50%) and black (47%).
  • time_class: The speed category of the match, divided into bullet, blitz, rapid, or daily. Blitz is the most common category, accounting for 44% of the records.
  • time_control: Represents the combination of the total time allowance and any time increment used during the game.
  • rules: Indicates the type of chess played. The vast majority of games (98%) follow standard chess rules, though variants like crazyhouse or chess960 are also present.
  • rated: A Boolean value indicating if ELO points were contested. Roughly 97% of the matches are rated.
  • fen: The Standard Notation used to describe a specific board position at any given moment in the game.
  • pgn: The standard plain text format for recording the entire sequence of moves in a chess match, allowing for the extraction of many more features.

Distribution

The data is provided in a tabular format, specifically a CSV file titled club_games_data.csv. The file is 166.67 MB in size and consists of 14 distinct columns, holding approximately 66.9 thousand valid records. The structure is suitable for analysis by beginner and intermediate data users. Updates to this product are expected on a monthly basis.

Usage

This data product is highly valuable for various analytical and modelling projects:
  • Machine Learning Development: Training models, such as LSTMs, to predict final game results based on input move sequences.
  • Outcome Prediction: Developing models that forecast the match outcome when supplied with the FEN board notation.
  • Strategic Analytics: Performing detailed statistical analysis concerning current ongoings and trends in chess to determine strategies for increasing win probability.
  • Rating Forecasting: Building models designed to predict a player’s ELO rating based on the moves they execute.
  • Data Visualization: General data visualisation projects related to board games and online competitive play.

Coverage

The scope is focused entirely on online matches hosted on the Chess.com platform. It captures a large distribution of player ELO ratings, ranging from the low hundreds up to 3,172. Coverage includes various online play speeds, categorised as bullet, blitz, rapid, and daily matches. The file predominantly features games adhering to standard chess rules, with 97% of records being rated games where ELO points were at stake.

License

CC0: Public Domain

Who Can Use It

The dataset is appropriate for several user groups:
  • Data Analysts: Seeking statistics on player ratings, game outcomes, and time control impacts.
  • Machine Learning Engineers: Looking for structured PGN and FEN data suitable for developing advanced predictive models.
  • Chess Researchers: Interested in investigating game theory, move patterns, and the effectiveness of different strategic choices.

Dataset Name Suggestions

  • Chess.com Match History (60K+ Records)
  • Online Chess ELO and Game Record Repository
  • Monthly Chess Performance Statistics
  • Chess Game Analytics Data

Attributes

Original Data Source: Chess Game Analytics Data

Listing Stats

VIEWS

5

DOWNLOADS

2

LISTED

07/11/2025

REGION

GLOBAL

Universal Data Quality Score Logo UDQSQUALITY

5 / 5

VERSION

1.0

Loading...

Free

Download Dataset in CSV Format