Weird puzzle

What do you say about this end-game puzzle rated around 2000? I think it's very unnatural the way the black king moves. lichess.org/training/JAmvK

@herr_shaibel said in #1:
> What do you say about this end-game puzzle rated around 2000? I think it's very unnatural the way the black king moves. lichess.org/training/JAmvK

Black plays the king moves with Zugzwang in mind. White will after a while run out of sensible moves and then has to give black access to the weak pawn on the h file. Black will need to take a bit of care to not get trapped there (a and h pawns pretty often need special attention) but the win is not so difficult for black.

That sounds fair and reasonable, but I feel it should be rated much higher. I would never be able to solve similar puzzles at my level. 2300 feels more accurate than 2094?

@herr_shaibel

What achja said... but I put it in a Study for you:

lichess.org/study/ZXssOG1l/LwsFnYHS

If you're having trouble understanding the king moves in this puzzle, I highly recommend you look up the basics of king and pawn endgames. This is really a very elementary technique you simply must know to succeed in endgames.

Opposition, rule of the square, shouldering, triangulation, Zugzwang, etc.

You can find some exercises under the heading "Pawn Endgames" here:
lichess.org/practice

@herr_shaibel said in #3:
> I would never be able to solve similar puzzles at my level. 2300 feels more accurate than 2094?

2094 looks well fitting to me. It may not be trivially easy, but you can readily solve it if you have basic pawn endgame knowledge (particularly about Zugzwang) and take some time to think about it - as @Sefegiru explained very well.
From what you wrote it seems that you don't have that basic pawn endgame knowledge. If other puzzles with a similar rating are considerably easier for you, that probably means you are better at other areas of chess than at pawn endgames. So you might indeed consider reading about pawn endgames, a very worthwhile area of study in my opinion.

Besides, as far as I know, puzzle ratings are determined by how well people do against the puzzles, they are not simply assigned at will. In other words, puzzle ratings are pretty much always correct in the sense that they reflect how well Lichess puzzle users do against them.

@achja said in #2:
> Black will need to take a bit of care to not get trapped there (a and h pawns pretty often need special attention)

I don't know what you mean with this part, in this position I see no risk whatsoever that Black could get trapped in any way with his h pawn. You may be referring to the general fact that King + lone Rook Pawn against King endgames are usually drawn, but that is no issue here at all due to the numerous remaining pawns on the queenside.

@Sefegiru

Thanks alot mate :D Really generous to make that study. This was probably one of my first puzzles with a zugzwang theme.

@Sefegiru said in #4:

> I put it in a Study for you:

Very nice. Thank you.

@JSM2 said in #6:
> I don't know what you mean with this part, in this position I see no risk whatsoever that Black could get
> trapped in any way with his h pawn. You may be referring to the general fact that King + lone Rook Pawn
> against King endgames are usually drawn, but that is no issue here at all due to the numerous remaining pawns
> on the queenside.

I meant, in general, that black could make the mistake to want to win with king and h pawn going for its promotion, and then the white king could "lock" up the black king.

For example here. Black needs to be a bit careful and not play the h3h2 and Kh1 plan.