Rated Puzzles Question, What Determines When You Get Rating Points?

Hi everyone 👋

I’ve got a question about Rated Puzzles since the rating system change, and I’m hoping someone might be able to shed a bit of light on how it works now.

As many of you know, I spend a lot of time doing puzzles, and I’ve noticed that rating points are no longer awarded for every solved puzzle. That’s totally fine in principle, but it’s made me wonder if there’s a specific pattern or criteria behind when points are actually given.

For example, is it influenced by things like:

• Time taken to solve the puzzle
• Accuracy or efficiency of the solution
• Certain themes or ideas, such as underpromotion, castling, quiet moves, etc.
• Or is it simply random, based on internal calibration

I’ve got my selector permanently set to Extra Hard, so difficulty isn’t changing on my end. Some puzzles feel just as challenging as others, yet one gives rating points, and another gives nothing, even when solved correctly.

I’m not complaining, just trying to understand the logic behind the system a bit better. If anyone from the community or staff has insight, or if others have noticed similar patterns, I’d be keen to hear your thoughts.

Thanks in advance, and happy puzzling ♟️
@JosephReidNZ ♟

@JosephReidNZ Great questions! I have no answers but I've been wondering similarly (but at lower level, of course). Thanks for asking the question!

👋 The number of rating points gained/lost for a puzzle attempt depends on the difference between your rating and the puzzle's rating. This is similar to how ratings change after a game between two players - but in this case, one of the "players" is the puzzle itself. If you correctly solve the puzzle, you've won the game.

It is possible that you can solve a puzzle and still receive no rating points; this would happen if you were already much higher rated than the puzzle. In this case, the player is strongly expected to solve the puzzle correctly - so solving it does not indicate that they are better than their current rating suggests.

Time is no longer a factor in the new rating system. It's totally up to the player how long they want to calculate.

Hope that clarifies things!

I am always using the default-difficulty. From my experience, I only gain points when I solve a puzzle without hints and mistakes. If the puzzles have many moves, a single mistake could result in losing less points than usual. I don’t think, that time ever affected my rating. Usually I gain 2 or 3 points by solving a puzzle. I think a streak increases my chance of gaining 3 points instead of 2 points and also increases my chance of gaining 4 points or more, the higher my streak gets. I think, the same applies for a failing streak. I am not familiar with the balance of my rating by solving / failing a puzzle. My experience might be a bit mixed up with this new feature with these puzzle-points (Not the rating points) but I still hope I was helpful!

Thanks everyone for the replies, much appreciated 👍

@isolani-d4, glad it’s not just me wondering about this. Nice to know others are noticing it too.

@DeepGreene, thanks for the clear explanation, that makes a lot of sense. Thinking of the puzzle as the 'other player' in an Elo-style system actually helps frame it really well, especially the part about being strongly expected to solve lower-rated puzzles. It's also good to know that time is no longer a factor, which answers one of my main questions.

@Ein-Schachspieler, thanks for sharing your experience as well. The streak aspect and clean solves explanation is interesting, and I agree things feel a bit mixed at the moment with puzzle points vs rating points alongside the new system.

Overall, this helps clarify what’s going on. I think most of my confusion just came from being used to the old system, where almost every correct solve felt more directly rewarded. Still enjoying puzzles a lot, just adjusting expectations a bit 😄

Cheers all ♟️
@JosephReidNZ ♟

what

I have a puzzle rating of 1617 and Chess.com says that is at the 93.9 percentile. I saw someone else who had a puzzle rating 1516 and was at the 96.4 percentile. Yet someone else who has a somewhat similar rating was at the 99+ percentile. I have also seen other ratings that don't appear to correlated with my math knowledge. Is there a logical reason for this?

Percentile hasn't adjusted as to your current rating and so the percentiles reflect to your puzzle rating before the huge adjustment chess.com did... I think.

okey