jlueke, 38. Qa2 Qa7 39. h4 Rb7 does protect the f-pawn. Although this looks very passive to me (and still gives up the a-pawn), I don't see a clear refutation. The question is whether our position would be worse than alternatives after white takes the a-pawn.
Move 38: The Conditional Move
uvang, you might have spent your time better by reading other analysis. 38. Qa2 Kf8 does not appear very good.
Fallacy in my above line(s) - white may play both Bd5 and b4 before h4 - then my line is not so good.
'Shall work on it later when I get time.
savyma,
Rb7 does not work because...
38. Qa2 Rb7
39.h4 (any move)
40.g5 hxg5
41.hxg5 now what?
if you move the bishop, 42.Rxf7
you cannot capture with the queen!
This is supplemental (does not replace) the above diagram (because too many lines make the postings unreadable) with b4 line added before h4. Again there may be moves I have missed (phew!) so please try to find the mistakes:
Savyma,
8. Qa2 Rb7, 39 h4 Kh7, 40. g5 hxg5, 41. hxg5 Bxg5, 42. Bxf7 (seems more effective than Rxf7) looks playable - but I cannot decide whether we will be able to draw or not.
What about moving the rook to b8 and then to f8?
Is there a problem with this? White has a passed queenside pawn but material is equal, and black has time to meet the pawn advance. I don't see any other threats from white here. Why is moving the king the better move?
savyma,
my last move was a mistake - instead oh 46...Kg7, 46...Qd7 should be played. Thanks for pointing it out.
38...Qb7 appears the most forceful line - only if we can find a suitable continuation (which I am unable to find)!
what's wrong with moving the black king back and forth at the end of the above diagram? it doesn't look like any white moves would be a problem.
edit: what about 40) h4, how can the attack at f7 be stopped then?
maybe we need to get the rook to f6 and not worry the a pawn is lost?
themothman, that is Elroch's suggestion, which I have tried to implement in two places in the Kf8 lines. Immediate Qe7 does not work (pls see bottom of my first post).
I find it hard to believe that I need to make this post for the third time... (I've been asking this question since move 34, and still there's been no answer or peer review.)
I noticed that the idea still hasn't registered into the list of possible variations for us, as presented by dsarkar at the start of the present thread:
38.Qa2 a5!? 39.Q:a5 Rc6!? (forcing a queen exchange) 40.Qa8+ Qc8
Well, clearly 38...a5 is a possible move. The question is how good it is, in comparison to other alternatives.
It's not at all clear that white can win this, although they clearly have an advantage. A queen-side blockade seems possible, and it also looks like we can secure against giving up our f7-pawn, so it won't be easy for white to break through. I wouldn't be pessimistic if I were to play black in this position after move 40 -- having given up a pawn but obtained a position without queens on the board seems like a win (not in the sense of winning the game, but in the sense of moving closer to our goal).
[COMMENT DELETED]