King and pawn vs. king endgame

Oops, I have to change that.  I meant 2.Kc2

These are just baby endgames though.  Easy for an engine to brute force them.  And yes sometimes they can choose the right move in spite of a poor evaluation.  Other endgames all engines are totally clueless... unless they can calculate and store the ludicrously large number of positions it takes to reach a depth of a few hundred when their 50 move rule kicks in and tells them which lines to avoid because they are a draw.

And that's after the endgame has been reached.  In middlegame positions when deciding which endgame to go to the engine is helpless... but yes it will beat humans in a match because humans make more errors overall.

The original analysis of Drtina's position (given above):

 "Theorie pĕšee proti králi" from Časopis českých šachistů(1907)

http://chessskill.blogspot.com/2012/03/algebraic-notation-language-of-chess.html

This position, also in Dvoretsky's Endgame Manual sometimes proves difficult:

Wafflemaster´s first diagram is particularly interesting, as it contains several pitfalls for White and an important winning move (4.Kc7!). Interestingly, taking the opposition immediately is not the shortest way to success; 1.Kd4 is faster:

Thanks for all answers !!! 

You are awesome  ! 

You need some work on your endgame, sabir4795.

Rook's pawns are always drawn if the enemy king can make it to the pawn's file anywhere ahead of the pawn or take opposition to the supporting king along a rank ahead of the pawn across the file adjacent to the pawn. 

The key squares for other pawns depends on the rank the pawn has reached. These are on the pawn's file or an adjacent file.

2nd. rank: 2 or 3 ranks ahead of the pawn. 

3rd. and 4th. ranks: 2 ranks ahead of the pawn.

5th. and higher ranks: 1 or 2 ranks ahead of the pawn.

If the supporting king occupies one of these squares he wins irrespective of the move and the position of the enemy king, except when the enemy king to move can immediately capture the pawn or if the side with the pawn is to move, the pawn is on its starting square and the enemy king stands directly in front of the pawn.

For pawns behind the 5th. rank the squares adjacent to the pawn 1 rank ahead are almost key squares. If the supporting king can reach one of these squares he wins unless the enemy king can immediately take close opposition ahead on the same file.

There are exceptions for knight's pawns where stalemate can occur. E.g. this position is drawn whoever has the move even though White to play can reach the "key" squares f7 and f8.

Similar key squares for a rook's pawn don't exist until the pawn has passed the crease, when the squares on the 7th. and 8th. rank on the file adjacent to the pawn are usually key squares, but there are exceptions, e.g.

This position is drawn even though White to move can reach the key square g8. (Black to move draws by Kf7 as in my 1st. paragraph.)

Of course key squares are irrelevant if the enemy king can't enter the square of the pawn, when the pawn doesn't need support.

On nomenclature, Averbakh notes that there is some disagreement in the use of the terms "key square" and "critical square". He uses "key square" to denote a square which if occupied by the supporting king guarantees promotion and "critical square" to denote a square which if occupied by the enemy king threatens capture. 

Has wafflemaster been here since 2013?

Play the knight in the opposite direction: Ne3-d5 with subsequent capture of both white pawns. Why sacrifice a good knight?

Ariskotle is right, though it may be tricky to see at first. You may have thought to lead the white king away so you could capture the pawns and figured that king + pawn should win, after all there’s plenty of time to manoeuvre, right?

But King + pawn doesn’t always win. The trick is to know whether your own King can get to the three squares, two rows *before* your pawn and before the enemy King can prevent that (there are exceptions, see Marattigan’s post above). If your King is too close to its pawn (same line or one line before) and the enemy King can prevent you from moving forward (through a method called “opposition”), the enemy King can force a draw. Of course, the white King should not make a mistake (here, 4. Kg3 was the only move that leads White to force a draw, anything else and you could’ve won).

Sacrificing your Knight here allowed the draw. Since your King already guarded the white pawns, there was nothing wrong with hopping around with your Knight until those pawns were taken, the enemy King couldn’t have taken your Knight anyway as upon capturing your Knight is guarded by your King.

Pawn endgames are tricky and require a lot of practice. A good starting point is at Lichess, where they explain the concept of “key squares” and “opposition” much better than here on Chess.com. It’s free: https://lichess.org/practice/pawn-endgames/key-squares/. And there is a lot of videos on YouTube as well.

Every player rated 400 or better should learn these basic pawn endgames. If you don't it is your own fault.

ha I see plenty of 600s playing wayward queen attack I'm not sure you should expect people to study king and pawn endgames

also thanks a lot, I didn't even notice that! and here I thought I was clever, thanks aris and mister

I just realized I don't really know how to play K vs K + P, I just use intuition. I guess puzzles taught me something

But the basic problem was not knowing how to play KPK. 

@fripallofyou obviously didn't realise that if he gave away the knight White only had to reply to the capture of his last pawn with Kg3 and the game was drawn. 

Had White not had a stalemate defence against Black's pawn queening @fripallofyou's knight trap would have been a nice risquée finish.

@brodyrogers, what do you mean? It’s a clear win, unless you sacrifice the Knight (as was done).

Even after you threw away the win, your opponent gave you another chance in the pawn ending.

if the king is In front of the pawn,win,if not,draw.