Advice on Endgame Defense

Here is a 960 game that I played (my opponent was obviously stronger than me). 

(Don't laugh at the a4 pawn push at move 14; I'm not used to the bishop attacking a4, and apparently I even fought all the way back to a drawing position by move 21 (trade knights then take the a pawn would've been the proper way to continue) before making a fatal miscalculation on move 22.

The endgame is probably lost with proper play, though it does feel like I had several slip-ups that lowered my chances of saving the game. I wonder if there is any feedback on that part here. Advice on defending would be helpful, especially if I get to somewhat closer endgames.

Other than queening a pawn, endgames are essentially all about brining your pieces into contact with weak enemy pawns. Sometimes you have to create a weakness through threats (making their pawns advance which loosens squares / weakens pawns) or by pawn sacrifice (to damage their structure).

So look at the game through this lens. After move 30, when black anchors his a pawn with the knight, how many times does your rook or knight attack an undefended pawn? Also look at your structure, you have 4 isolated pawns, and notice how black's rook goes on tour attacking each of them.

We can start even earlier. Look at move 26. Black's knight ties your rook to the defense of the d pawn.

One idea is to play f4-f5, and whether black captures or you capture you'll split up black's pawns... white is losing in any case, but that's the right kind of idea.

Similarly I probably would have tried 25.Ra2 Nxd4 26.Rb2 because white wants to infiltrate into black's position and start "touching" pawns.

-

-

This isn't enough to turn the game in your favor, but look how much nicer it is that (at least optically) your rook is tying the black rook to the f pawn. Your knight is in contact with the a pawn. This is the sort of thing you want in an endgame, and you have to work hard to find ways to do it.

Unfortunately in this particular endgame, 100% of your pawns were targets while black had a healthy structure plus an outside passed pawn (I'll count the two a pawns as one passer). So black should win, but those are some fundamental endgame ideas.

After 21 Ne5? there is no suitable defence, you can only hope black errs as well.

Agreed that the position is hopeless, but there are probably defenses that give a higher chance of the opponent erring compared to what I actually played. While this position is probably too hopeless even for such techniques to work (ruined pawn structure, 2 pawns down, and vs an outside passed pawn), advice might be more suitable in closer positions which are still likely lost but have a higher chance of opponent erring (such as the same position without doubled kingside pawns, or if my opponent was only up 1 pawn), or drawn or hard to win (equal number of pawns but with worse pawn structure and against the outside a pawn, or down just the a pawn with a normal kingside structure).

Note that computer analysis is generally not good when figuring out how to play lost positions (since computers go for the line that loses the slowest, rather than the line that leads to a higher chance of the opponent erring and giving a chance for you to draw).

I also decided to play on for a while instead of resigning to learn how the opponent would convert and learn how to defend a lost endgame to maximize drawing chances (and there's also the off chance that my opponent would err, though it's less likely since I'm already weaker than my opponent to begin with).

hello guys. How can i analyze my chess 960 games on chess.com?