Interesting Pawn Endgame

An exception to the rule that an outside passer will beat a central passer by distracting the enemy king.

As Pfren pointed out in a different topic, remove the h pawns and white wins easily. The standard outside passed pawn distracting the king.

For example

that was helpful as I have made some bad pawn decisions lately...been looking at pachmans strategy trilogy and some pandolfini stuff, some don't like but I can understand easily...are you a member of THE FIGHTING GLADIATORS by any chance, if I might ask?

Since part of white's difficulty is that he has to capture two pawns while black only has to capture one, if we remove the a pawns in the starting position then white can always draw.

Also notice how in the starting position white's pawn on h3 cripples his pawn majority.

If instead white had pawns on h2, g2, f3, then it's just a draw.

i think that the last position is not draw at all...

1...g5 wins for black, so the main lines are not draws, they're wins (for black).

But if you mean one of the variations, it's a pretty easy draw

And in general Q+p vs Q is a draw unless some conditions are met. In this position black's king is very far away from the pawn, and white's king is too close. Also rook pawns in general are much easier to draw in Q+p vs Q endgames.

You are not explaining most of the moves. Also it's not an exception to the rule, an outside passed pawn will win IF it can become a passed pawn. In this case black can stop it by playing g5 but the position is so hard that I would give the wining chances to white in a OTB match.

Yeah, I didn't really explain stuff.

Basically the idea is since the white king has to keep an eye on the central passer (and since white's majority can't create a healthy passer if black plays correctly), black is (in most variations) the one who decides when to initiate the pawn race. After it starts, Black runs to the kingside, white runs to the queenside, and if someone can queen first with some tempo to spare then they win.

So black gets to set the pawns however he wants before the race begins.
Before it begins Black has two goals:

1) take advantage of white's crippled kingside majority (IMO white's defect is far from obvious, and the first step to exploiting it is by playing 1...g5)

2) Don't let white's queenside pawns get too far up the board (far advanced pawns would white in the pawn race to come)

After that black initiates the pawn race and wins (if he calculated carefully enough).

Yes, trying other moves than 1...g5 I lost as black against the engine, so white can definitely win in practical play with non-GMs.

As for the logic behind certain moves, you'll just have to explore for yourself. I could give an engine dump of long variations on every move, but anyone who cares enough to slog through that can just discover it on their own (with an engine's help). But it's basically the plan I outlined above. So for example 3...Kd6 gets an ! because white is put in zugzwang, and IIRC if Back plays 3...h4 then it's just a draw. One benefit of the central passer and king arrangement is black can often shuffle like this to burn a tempo while white can't. (if the pawns were on h2-g2-f3 then white could burn tempo by walking backwards and giving up ranks to the black king and pawn don't matter IIRC).

White makes a kingside passer on move 4 of the main line.

And there's still lots of craziness I don't understand.

For example I just discovered this.

And this is why

I already knew why g5 wins but you have to calculate many more moves ahead as black that it becomes incredibly hard, this position is too delicate becouse black is just 1 tempo ahead of white and has to make the best move each time to win. Simply too hard.

Too hard if you haven't spent a while studying it, yeah.

But that's how people get better at chess, you study stuff like this.

You can study this position but there are plenty when one side is wining but only if he makes 20 best moves in a row

I'm not claiming studying 1 endgame is going to make you perfect at all endgames.

lol

Even this exact endgame (or one very close to it) I could easily imagine myself screwing it up.

But by analyzing I got new ideas that will help me play better in many related positions.

IMO that's sort of what chess is all about... you learn a little something in one area, and it helps in a few other areas. You still suck and blunder, but you're a little better today than you were yesterday, and this builds up over the years.

Yes and this is the only think we can do. Searching all the wining moves by analyzing with stockfish doesn't help.

Guys, you're both pretty. Btw stiggling i enjoyed your analysis, some nice stuff you found in there.

stiggling qa8 should win

Qa8 loses the queen, I guess you mean Qa1, but there should be perpetuals and such.

This sort of endgame is a pretty well known draw, plus I checked all this with an engine of course.