Why your puzzle rating is higher than your game rating
A puzzle tells you there’s a tactic and takes the clock away. Why the gap opens, how big a gap is normal and what to do about it.
Search the chess.com forums and the same thread keeps coming back with new numbers. 1500 in puzzles, 400 in rapid. The two numbers measure different things. The gap between them is worth reading, once you know which part of it comes from the rating scale and which part comes from you.
Why is my puzzle rating so much higher than my game rating?
Because a puzzle does the hardest part of a tactic for you before you start. It tells you there is something to find.
| In a puzzle | In a game | |
|---|---|---|
| Is there a tactic? | Yes, guaranteed | Maybe, and nobody says so |
| What you’re asked | Find one line | Play forty moves without a blunder |
| Whose tactics | Mostly yours | Yours and your opponent’s |
| The clock | Depends on the site | Running on every move |
| After a mistake | Next puzzle, fresh start | You play on a piece down, often tilted |
The scale played a part too. In October 2025 chess.com rebuilt its puzzle ratings and admitted the old ones were “significantly inflated”. Under the old system every solved puzzle gave at least 5 points, and incomplete solutions earned partial credit. One mate in one had been rated as high as 1800. On the new scale, the same Glicko-style system used for games, it sits at 963. So if your gap opened before that, a good chunk of it was the scale, and a lot of the forum answers you’ll find were written about the old numbers.
Does time count in puzzle rating?
Depends where you solve. Chess.com’s announcement of the new system doesn’t say whether speed changes your points, and the forum answers disagree, so don’t trust anyone who sounds sure. On Chesslord time doesn’t count. The clock is on screen, but thinking for two minutes costs you nothing.
What does the gap tell you?
Look at the game at the top of this page. Vladimir Kramnik, world champion at the time, walked into a mate in one against Deep Fritz in an equal position. As a puzzle it’s trivial. In the game nobody announced it, and it was the opponent’s threat, while most puzzles ask you to attack.
Your gap measures how much of your tactical skill makes it into your games. A big gap usually means you calculate fine once you know where to look, and the points leak somewhere else.
| What you notice in your games | What the gap is probably telling you |
|---|---|
| You lose pieces to one-move threats | You look for your own tactics and skip your opponent’s |
| The review shows wins you never looked for | You don’t ask whether there’s a tactic during the game |
| Rapid is close to your puzzle rating, blitz is far below | Time pressure. Your tactics work when you have time to look |
| Your puzzle rating is below your game rating | You win games with openings, endgames or your opponents’ mistakes, and tactics are the thin spot |
So a high puzzle rating doesn’t mean you’re secretly stronger than your game rating. Read it as a ceiling, roughly the level you’d play at if you spotted every chance you’re able to calculate. The game rating is what you actually do with a clock running and an opponent trying to trick you.
How big a gap is normal?
Nobody has published a clean answer, and the forum guesses of 400 or 500 points are made by eye. The best public number we found is a 2021 look at 7,500 UK chess.com players. It found a correlation of 0.57 between puzzle rating and best game rating (the highest of bullet, blitz and rapid), with the puzzle rating tending to sit on top. A correlation of 0.57 means the two numbers move together, loosely. Two players with the same rapid rating can sit hundreds of points apart in puzzles. That data also comes from the old, inflated chess.com scale, so gaps there today are likely smaller.
For your own numbers, watch the trend. A gap that stays the same while both numbers climb is fine. A gap that keeps widening means your puzzle work isn’t reaching your games.
How should a puzzle rating start?
From your game rating. On Chesslord your puzzle rating starts at the highest game rating on the chess.com or Lichess account you linked, in the time control you play most. Without a linked account it uses the rating from your games on Chesslord. You don’t begin at an arbitrary number and grind your way up, so the two numbers start level. If they drift 300 points apart after a few hundred puzzles, that gap came from you, and it means something.
Hints can’t prop the number up either. Asking for one counts as a failed puzzle, the same rule chess.com uses, and how to solve chess puzzles explains when that trade is worth it.

What should you do about the gap?
Bring the puzzle question into your games.
- Before every move, ask whether this position would make a puzzle, and look at checks, captures and threats for both sides
- After every move your opponent makes, ask what it threatens, the one question that would have saved Kramnik
- If blitz is the number that lags, play some rapid, because tactics you find with time on the clock are the ones you start finding faster
Then go back over your games. The tactics you missed there are worth more than any random puzzle, because they’re the ones you actually miss. How to analyze your chess games shows how to do it. Chesslord’s insights count the forks, pins and mates you found and missed in your games on Chesslord, and if mates are the problem, read how to stop missing checkmates.
Keep doing puzzles, with a solving habit that carries over. Work out the whole line before you touch a piece, the way you’d have to in a game. How to solve chess puzzles covers the method and how many chess puzzles a day covers how much. If the patterns are new to you, start with chess tactics for beginners.

How Chesslord helps
Solve puzzles