The way that puzzles are rated on this website does not make any sense to me. Where do these puzzle ratings come from? I just got this puzzle, which is in incredibly basic back rank mate, has a 99.4% pass rate, and a rating of 2200? How is this a 2200 puzzle? It should be a 400 rating puzzle. Shouldn't puzzle ratings be adjusted the same way my rating is when I pass or fail one? Any time someone solves a puzzle correctly or incorrectly the puzzle's rating should be adjusted based on the rating of the person solving (or not solving) it. No puzzle should have a 99%+ pass rating because that would mean it's being given to way too many people who are too advanced for it. I also see the opposite happen frequently, where a puzzle has a pass rate of <20% and a rating that is far too low. Statistically, I would think a well-rated puzzle should always have a pass rate of roughly 50%? Or at least be relatively consistent puzzle-to-puzzle. Anyone else notice this?
https://support.chess.com/article/2016-how-do-puzzle-ratings-work
Puzzles get ratings based on the rating of the members that fail/pass and gain/lose rating as if it was a game.
If you see one that is much too high rated or too low, you could always report it.
https://support.chess.com/article/1223-how-do-i-report-a-bad-puzzle
The way that puzzles are rated on this website does not make any sense to me. Where do these puzzle ratings come from? I just got this puzzle, which is in incredibly basic back rank mate, has a 99.4% pass rate, and a rating of 2200? How is this a 2200 puzzle? It should be a 400 rating puzzle. Shouldn't puzzle ratings be adjusted the same way my rating is when I pass or fail one? Any time someone solves a puzzle correctly or incorrectly the puzzle's rating should be adjusted based on the rating of the person solving (or not solving) it. No puzzle should have a 99%+ pass rating because that would mean it's being given to way too many people who are too advanced for it. I also see the opposite happen frequently, where a puzzle has a pass rate of <20% and a rating that is far too low. Statistically, I would think a well-rated puzzle should always have a pass rate of roughly 50%? Or at least be relatively consistent puzzle-to-puzzle. Anyone else notice this?