I was recently playing a 5+2 game online and felt like I messed up the endgame. It's upsetting because I just felt so lost and wasn't sure what I should have been looking for. Anyone wanna help?

For reference, here's the game on chess.com: http://www.chess.com/livechess/game?id=390793063

And here's the pgn:
I (with the computer to support me) think I could have had a small advantage with 18...Bd6, and I regretted not playing that as soon as I played 18...Bf6 instead. He didn't punish correctly, at least. Unless anyone disagrees?

With all that out of the way, here's where I really want to focus from, after 21. Rfe1:


I feel like I, black, should be doing well here. The exposed white king and weak f-pawns look like targets for my queen. With the rooks and queens on board, I didn't think the passed yet isolated d-pawn would make it to promotion. And if all went well, I thought the 2 to 1 pawn advantage on the queenside would be nice for me.

But, I don't know quite how to make the win happen, and as shown in the game I was definitely confused. It's not that I ever expected to lose, but I thought I had a shot at winning.

One more position, after 51. Rh6:


It's a tablebase draw after 51...Rd4 52. Rxb6 as played, but 51...Rd2+ 52. Kf3 Rb2 ? Can I win that?

I think Rd2+ Kf3 Rb2 is winning for black, but that's just me.

It seems to me you dropped the ball after around move 35. Your plan of breaking on the Kingside didn't accomplish much. You need to activate your rooks to penetrate White's posiiton. 35...f5 looks very strong, following up with Rc8, since White can't easily double on the e-file anymore. Or if you want to advance on the Q-side, 35...b5 looks better, to keep more flexibility in the pawn structure.

Annnnnnnnnnnd fight! No, really, the position looks so simple but I don't know how to properly evaluate it, and the computer is fairly useless here.

I think the ending from the diagram is almost positively drawn but slightly better for white. White's weaknesses just don't mean much since you need to keep too much material invested in stopping his pawn. With so many heavies on the board your king isn't going to get meaningfully involved so you'll likely just end up mutually stalemated on the d pawn.

In my experience dynamics play a much more important role in creating winning chances in equal material rook endgames. For instance better king position, active vs passive rook, advanced passers, etc. I think this position is a great example of that as every single static advantage is in your favor, but every single dynamic advantage is in your opponent's.

well, imo black cant win the white a-pawn wiht losing is b-pawn, so easy draw.

white king cant be cut off ,etc..

i really dont see how black can win.

i think black has excellent practical winning chances after 52. ...Rb2 because Rb4 is already a threat and i don't see much else than defending the a-pawn passively with Rh1-a1. I guess theoretically it's still a draw though, would have to analyze some more to be sure.

For the initial position i tend to agree with DIR here, black probably has to block the pawn with Qd6 after which white can just play Qe5, take the queens off (thereby eliminating the danger to his king), go Re7 and how is black ever to make progress then?

Not to sound like a broken record, but I agree with DiR. I've aimed for many endgames like this as black, thinking that because all of the positional factors are in black's favor (king safety, pawn structure, and 2v1 on the queenside) that black should be winning. But white's activity makes it nearly impossible to realize those advantages. The d-pawn will tie black's pieces down I think it'll be really tough for black to ever get active and make a draw. I've had to bang my head against a wall several times in positions just like this one before I finally got a better feel for them.

51...Rd2+ & 52...Rb2 seems winning to me - black's threat is to walk his king to c5 and play Rxa4. White's king can't help so he has to try Rh1-a1 but then he's so passive. The tempos give black time to drive white's king to the second rank, and we end up with something like WRa1, Kc2 versus BKc4, Rb3. This should be winning but some calculation is required.

Yeah, chess is too hard.

Active pieces matter