Questions about puzzles on chess.com

Hello,

I do puzzles frequently on this website and a few questions occured. No big issues, just curiosity about how they're made and work.

It happens frequently that I don't understand why the puzzle stops. I mean it is weirdly cut.

Sometimes it stops after only one move without a checkmate or a capture. My guess here is that the end of the puzzle is when the opponent has several options that are equally good, mediocre, or bad which shows the end of the dilemma and avoid to make a puzzle out of variations. It would make sense as I feel that it is usually - but not always - that abrupt when the puzzle is about a defensive move. When you're under attack, there are a few options that look good but that are actually traps and you have to chose the only good option. Are they puzzles made to train our ability not to fall into traps ?

Sometimes they also stop suddenly two moves before a forced checkmate... The ending is in the middle of a forced sequence. There might be a limitation of the amount of moves you can make in one puzzle, but I feel it also happen after only 2-3 moves. Sometimes the last move is at the beginning of the forced sequence, sometime in the middle... Sometimes it is one move after a capture - which makes sense when it allows you to save the piece you used to capture but sometimes it feels more random,

So my question is, how are they made ? What rules the making of puzzles ?

I was also wondering were the puzzles were from. Are they from games by chess.com users ? Famous games ? Generated algorithmically ? All of the above ?

And finally I was thinking about the rating of the difficulty of the puzzles. I am sure that this has been asked many times but when I search the forum I always find discussions about the rating of our account doing puzzle.

How does it work ? Is it like an Elo rating of the "games" it has against people trying to solve them ? A different type of rating but also based on the % or the level of the people that solve the puzzle ? A system using "objective" or external criterias to rate the puzzles ? Pure randomness ?

Does anybody have an answer or want to give their thought ?

https://support.chess.com/en/articles/8602396-how-do-puzzle-ratings-work

I do quite a few of the tier puzzles on Chess.com and have noticed that now every time I hit a " major benchmark " the bonus points are taken back as soon as I press the " next puzzle " bar . Think this type of thing was reported a while ago & the problem was said to be only affecting on an intermittent basis ! I now look at these major benchmarks as information only , unless Chess.com can resolve this issue . Thank you .

Most puzzles on Chess.com are created by an algorithm that finds appropriate positions from actual games. Puzzles require a unique winning move for your side at every step of the solution, so they stop before alternatives are possible. See this blog for details: Understanding soundness and motivations in chess puzzles, problems, and studies.

Thank you !