Ratings
What Is a Good Chess Rating?
Published Updated
A rating is good only in relation to a defined group. Ask which organisation, which time control, which player pool and which date before comparing two numbers. There is no defensible universal table that turns 800, 1200 or 1800 into the same skill label everywhere.
On this page
The cleanest answer is a percentile
If a platform shows your percentile, use it. The 60th percentile means your current rating is above 60 per cent of the players included in that platform’s comparison group. That says far more than an imported label such as “intermediate”. Check who qualifies for the distribution: some sites include all accounts, while others require recent rated games.
Keep each pool in its own column
Two pools can use the same mathematical family and still produce different scales. Starting values, active-player rules, player strength and implementation details all matter. Conversion charts based on a sample can describe that sample, but they are not permanent exchange rates.
- FIDE standard
- Over-the-board games that meet FIDE rating regulations
- FIDE rapid and blitz
- Separate lists; neither is interchangeable with the standard list
- Online rapid or blitz
- A site-specific pool with its own players and implementation
- Puzzle rating
- Performance against rated positions, not opponents managing a whole game
FIDE numbers have a few firm landmarks
FIDE publishes ratings from 1400 under the regulations in force when this page was checked. Below that floor a player appears as unrated on the next list. That is an administrative threshold, not a verdict on whether the player can play good chess.
On the open-title ladder, the rating-only thresholds are 2200 for Candidate Master and 2300 for FIDE Master. International Master normally requires a 2400 rating threshold plus norms, and Grandmaster normally requires 2500 plus norms. The women’s title ladder begins with Woman Candidate Master at 2000. These are title landmarks, not a complete ladder of ordinary club strength.
Use rating as evidence, not a report card
- Compare the same pool and time control over a run of games, not after one result.
- Treat a provisional or high-deviation online rating as a wide estimate rather than a precise rank.
- Track decisions alongside the number: missed tactics, time trouble, opening uncertainty and endgame technique point to useful work.
- Expect ordinary variance. A short fall can happen even while the quality of your play is improving.
A practical definition of good
A useful rating is one that helps to arrange competitive games and provides a consistent reference within the same pool. A good result is not a particular four-digit number; it is making stronger decisions against comparable opposition. If you want a social benchmark, use your current percentile. If you want a training benchmark, review a meaningful sample of serious games and count the errors you can actually work on.
Sources and scope
- FIDE Rating Regulations, especially articles 7 and 8 for the published list, rating floor and calculation.
- Lichess's rating-system guide, which explicitly cautions against direct comparison between servers.
Regulations and platform implementations can change. These notes were checked on 16 July 2026; follow the linked first-party source for the current wording.