Elo is a rating system that predicts your expected score against a specific opponent. After the game, it compares the result with that expectation and adjusts your rating. A win over a stronger player is worth more than a win over a weaker one. FIDE uses the Elo system; Chess.com and Lichess use Glicko-based systems, although players often use "Elo" as shorthand for any chess rating.
What kind of player are you really?
What is the average chess Elo by platform?
There is no honest single average chess Elo. Each platform has separate pools for rapid, blitz, and other time controls, and each pool uses its own scale. These are the most defensible current reference points, checked on August 21, 2026.
| Player pool | Current reference point | What the number covers |
|---|---|---|
| Chess.com | No stable global median published | Chess.com uses Glicko and separates ratings by time control. Use the percentile shown with your own rating rather than a universal average. |
| Lichess rapid | Median rating bin around 1400 | 479,699 established players in the official weekly rapid distribution. |
| Lichess blitz | Median rating bin around 1475 | 668,630 established players in the official weekly blitz distribution. |
| FIDE standard | Median around 1700 among active published players | 219,505 active players in the August 2026 list. FIDE publishes only ratings of 1400 or higher, so this is not the average of all chess players. |
The Lichess figures come from its live rapid and blitz distributions. The FIDE figure is calculated from the official August 2026 standard rating list, using active players with a published rating. Chess.com explains its Glicko system, but its public help page does not provide one stable platform-wide median.
That is why a 900 on Chess.com cannot be judged against a 1400 Lichess median or a 1700 FIDE median. The useful comparison is your percentile inside one platform and one time control, tracked over time.
How Elo works, in one simple idea
Every rating implies an expected score. If you are rated 1200 and your opponent is 1200, the model expects a 50-50 match. If you are 1200 against an 1800, the model expects you to lose most of the time, so a draw alone would already be a small triumph for you and a small disaster for them.
After the game, the system compares what happened with what was expected. Beat the favorite and you gain more points. Lose to someone you were expected to beat and you lose more. The size of a FIDE rating change is controlled by the K-factor, with higher values for new and junior players and lower values for established players. The current formula and K-factor rules are published in the FIDE Rating Regulations.
The system does not measure effort, intelligence, or the quality of one move. It measures results relative to expectations.
Where the Elo rating comes from
Elo is named after Arpad Elo, a Hungarian-American physics professor and strong amateur player who designed the system in the 1960s. FIDE adopted it in 1970. The word is a surname, not an acronym, which is why you write "Elo" rather than "ELO".
What is a good Elo rating?
There is no universal good Elo. A rating is good only relative to its pool, time control, and your goal. Compare yourself with the percentile shown by the platform, not with a number copied from another site.
If the number you care about is 2000, the dedicated guide to whether 2000 Elo is good and what percentile it represents owns that question. If you want the training plan rather than the level definition, read how to reach 2000 Elo in chess.
Why your Chess.com, Lichess, and FIDE ratings never match
This is the single most confusing part for new players, so let me be blunt: a rating only means something inside its own pool. The same person can be 1500 on Chess.com, 1800 on Lichess, and 1400 in FIDE-rated over-the-board tournaments, all at the same time, and none of those numbers is "wrong".
Three reasons drive the gap. First, the starting point and calibration differ from site to site. Second, the playing pools are different, and a rating is always relative to the people in it. Third, time controls are not the same animal: your blitz, rapid, and classical ratings measure different skills and will not line up even on a single platform. Comparing a Lichess blitz number to a FIDE classical number is like comparing a sprint time to a marathon time.
The practical takeaway: pick one rating, in one time control, as your reference, and track that one over months. Switching the yardstick every week is how players convince themselves they are stagnating when they are actually improving.
What Elo can and cannot tell you
As a coach, here is how I use a student's rating. It tells me, roughly, the level of opposition they handle and the size of the gap to their next goal. It is a useful diagnostic input. What it does not tell me is why they are stuck, and that "why" is where progress actually lives.
Two players rated 1300 can be completely different cases. One hangs pieces under time pressure and would jump 200 points just by slowing down in the last five minutes. The other plays clean but never studies endgames and throws away won positions. Same number, opposite prescriptions. The rating flags that something can improve; it never names the lever. That is exactly the job a structured review, or a coach, does.
So treat your Elo as a thermometer, not a diagnosis. It tells you there is a fever. It does not tell you the cure. If your number has been flat for a while, the problem is almost never "study more", it is "study the right thing", which I unpack in how to improve at chess efficiently.
What to remember
Key takeaways
- Elo is a prediction system, not a verdict. It estimates your chance of beating a given opponent and updates on surprise, not effort.
- "Average" is pool-specific. In the August 21 snapshot, Lichess medians were around 1400 rapid and 1475 blitz; active published FIDE players had a median around 1700.
- Chess.com, Lichess, and FIDE ratings cannot be converted point for point. Compare one time control inside one platform.
- Your FIDE, Chess.com, and Lichess numbers will not match, and that is normal. Track one rating, in one time control, over months.
- Use Elo as a thermometer. It signals that something can improve; it never names the lever.
Your rating is a starting point, not a diagnosis. If you want to turn it into a clear next step, begin with how to improve at chess efficiently, or let me review your games through JD Chess coaching.
