White 19

[cheesehead]

They pushed their b-pawn instead of capturing, so we can discard all analysis of ... bxa4.

Extending our skimpy previous analysis after [B]18 a4 b4[/B]:

[B]19 Nd2[/B]
[B]19 ... Ba6 20 Rf2 Rc8 21 Bf1 Na7[/B] and the c4-pawn has two defenders.

Try: 22 b3?! bxc3 23 Nxc4 (threatens Nd6+) Bxc4 24 Bxc4 and now it may not be easy to win the c3-pawn while our d-pawn is vulnerable (e.g., 24 ... Nf5 25 Rc2 Nxe3 26 Nxe3 Nc6 28 Bb5 O-O 29 Rd1 Nb4 30 R2c1 c2 31 Rd2)

Better is 22 cxb4 axb4 23 Rc1 Nd5 24 Nxc4 Nxe3 (removes the R/c1's defender, so now not 25 Ncxe3?? Rxc1 26 Bxa6 losing the Exchange) 25 Ndxe3

(to be continued)

[cheesehead]

Another possible line:

[b]19 Nd2 Ba6 20 Ne4[/b] c3 might be advantageous for Black

- - -

Keep in mind that Black might bring the N/e7 over to the Q-side with ... Nd5-b6 defending c4 and also attacking our a-pawn.

- - -

[cheesehead]

BTW, our deadline is [URL="http://mersenneforum.org/showpost.php?p=365332&postcount=123"]Feb. 1, 0700 GMT[/URL]

---------------------------------------------------------

I'm fairly sure we're going to win a(t least a) pawn by force!

[B]19 Nd2[/B]

If Black doesn't immediately protect the c4-pawn OR castle (e.g. 19 ... Nd5) then 20 Nxc4 Ba6 (to pin the N/c4 against R/f1) 21 Nd6+ K-moves 22 Rxf7 (not 22 Nxf7? Bxf1)

If 19 ... O-O 20 Rf2 (not 20 Nxc4 Ba6) Ba6 21 Bf1 and Black has no way to defend c4 a second time.
E.g., 21 ... Nd5 22 Nxc4
Then we may have quite a fight: e.g. 22 ... Bxc4 23 Bxc4 bxc3 24 bxc3 Rb7 25 Bxd5 (not 25 Rb2 Rxb2 26 Nxb2 Nxe3) exd5 26 Rb2 Rxb2 27 Nxb2 Rb8 28 Nc4 Rb3 29 Bd2 Ne7 30 Kf2 Nd5 31 Rc1 Bf8 32 Ke2 Ba3 33 Rc2 and so on ...
Or 22 ... Nxe3 23 N1xe3 Bxc4 24 Nxc4 bxc3 25 bxc3 and Black can't win control of the b-file (If 25 ... Rb7 26 Rb2 Rfb8 27 R1b1)

If 19 ... Ba6 20 Ne4 O-O 21 Nc5 Ra8 (21 ... Bc8 22 Rf2 Nd5 23 Bf1 Nb6 24 Bg2 Ne7 25 Bd2 Nd7 26 Ne4 Nb6 27 Ne3 and it's likely that some combination with Bf1 and Nd6 will win the c4-pawn. [I]Of course, this remains to be carefully analyzed.
[/I]

[LaurV]

[QUOTE=cheesehead;365359]
[B]19 Nd2[/B]
If 19 ... O-O 20 Rf2 (not 20 Nxc4 Ba6) Ba6 21 Bf1 and Black has no way to defend c4 a second time.[/QUOTE]
Yet, this seems to be their most reasonable reply. All the other versions, black is in big shit there, if we continue with Ne4, as previously analyzed. 19..Ba6 20.Ne4 bxc3 Nd6 check and they lose the pawn in f7 and all the king's flank.

Therefore, I agree that [B]Nd2[/B] is our best move now. (this we analized before, it is no reason to change it)

[WMHalsdorf]

It would appear that all the previous analysis of Nd2 applies just as well if not better in some case now than before. The question now is do we wait to reply or do it now.

[cheesehead]

Continuing the analysis I left off in a hurry:
[QUOTE=cheesehead;365359]

I'm fairly sure we're going to win a(t least a) pawn by force!

[B]19 Nd2[/B]
<snip>

If [B]19 ... Ba6 20 Ne4 O-O 21 Nc5 Ra8[/B][/QUOTE]Black may be able to defend all pawns in this line.

[B]22 Bd2 Nd5 23 Ne3 Nxe3[/B] (not 23 ... Nb6 24 Bxc6) [B]24 Bxe3 Rfc8 25 Rf2[/B] (now working on f7) [B]Ra7 26 Raf1 R8c7 27 Ne4[/B] threatening Nd6, Bxc6 and Rxf7.
Here Black can attempt a b-file invasion with [B]27 ... bxc3[/B] 28 bxc3 Rab7. We can force a retreat with 29 Nc5 Rba7 but now we don't have the N on d6 to menace f7. 30 Bxc6 Rxc6 31 Nxa6 Rcxa6 leaves us no closer to winning on f7.
If [B]27 ... Bf8[/B] 28 Nd6 Bxd6 29 exd6 Rc8 (not 29 ... Rd7 30 Bxc6) 30 Bxh6 (remember him?) we're temporarily a pawn up, but Black can now contend on the b-file and we've expended both of our wonderful knights.

[quote][B](21 ... Bc8 22 Rf2 Nd5 23 Bf1 Nb6[/B][/quote]24 Bd2 and 25 Ne3 seem to win the c-pawn.

- - -

All in all, it looks like we don't have an absolutely guaranteed pawn-win, but we can continue to make Black uncomfortable while maneuvering to find an additional edge, and Black has several ways to go wrong.

- - -

[B]19 Nd2[/B] is my choice, too. We may as well post it now.

[cheesehead]

I keep forgetting how powerful our position is.

[QUOTE=cheesehead;365501]If [B]27 ... Bf8[/B] 28 Nd6 Bxd6 29 exd6 Rc8 (not 29 ... Rd7 30 Bxc6) 30 Bxh6 [/QUOTE]I was so mourning our knights that I overlooked 30 d5!
30 ... Rd7 31 dxc6 Rxd6 32 Rxf7 and Black's h-pawn is still hanging.

[LaurV]

[QUOTE=cheesehead;365501]If [B]19 ... Ba6 20 Ne4 O-O 21 Nc5 Ra8, [/B]Black may be able to defend all pawns in this line.[/QUOTE]
Nc5? This is weak here. He might defend the bishop wit Ra8, or move it to c8, and we didn't gain much, we help them to centralize that bishop, as c4 is not threaten anymore, and we need 2-3 moves to get back the threat. We can be more aggressive here: [B]21.Nf6![/B] and so much for the victory, they can't move the king without losing a rook (and the game), so 21...Bxf6 (must), 22.exf6, Nd5 (must, otherwise black is losing one of the horses), 22.Bxh6, and that is our pawn plus. This continuation (19...Ba6 first, then 0-0 [U]after[/U]) is hell for black. Their only reasonable continuation is what I said: 19...0-0 first. There is no other way out of it, I am almost sure that we can sure-win any other continuation (no deterministic proof, but well... look to the board!)

[cheesehead]

[QUOTE=LaurV;365520][B]21.Nf6![/B] and so much for the victory, they can't move the king without losing a rook (and the game), so 21...Bxf6 (must), 22.exf6 Nd5 (must, otherwise black is losing one of the horses), 22.Bxh6, and that is our pawn plus.[/QUOTE]Right. Thanks.
[quote]Their only reasonable continuation is what I said: 19...0-0 first. There is no other way out of it, I am almost sure that we can sure-win any other continuation (no deterministic proof, but well... look to the board!)[/quote]Here's a continuation I failed to respect earlier:
[B]19 ... bxc3[/B] (not 19 ... b3?) and now Black has opened the b-file in front of his rook.

20 Nxc3 would immediately restore communication between our rooks, but leave our b-pawn exposed and, worse, allow ... Nb4 and ... Nd3. Ugh.

[B]20 Ne4[/B] threatens Nd6+ winning the f-pawn but
20 ... O-O avoids that and now 21 Nf6+ doesn't win the h-pawn because after 21 ... Bxf6 22 exf6, 22 ... Nf5 is feasible. 23 Bf4 Rbc8 24 bxc3 g5 25 Bd2 and 26 Ne3 looks solid for us, but Black may have something better along this way.
If 20 ... Nb4 21 Nd6+ Kd7 22 Nxf7 Rh7 so we can't take the h-pawn, but 23 Bxb7 (to prevent ... Ba6) Rxb7 24 Nd6 Rc7 25 Nb5 cxb2 26 Nxb2 R/c7-moves 27 Rbc1 and we're gonna git that c-pawn after all.

[cheesehead]

After looking at the situation from the Black side, I'm now pretty sure that the response will be
[B]19 ... bxc3[/B], which can cause us more trouble than any other line that we optimistically discussed before.

20 Nxc3 is inadequate, for reasons mentioned above.
20 Ne4 threatens, but we have to always keep in mind that Black will continue to maraud with his advanced pawn via cxb2 or, worse, c2, just as soon as he possibly can.
At first, 20 ... cxb2 looks like it would almost be a favor if it didn't swallow another of our pawns, but 21 Nd6+ Kd7 22 Nxb2 Ba6 23 Rxf7 and now
23 ... c3 24 Rxg7 cxb2 (don't often see that move notation twice in one game) 25 Rb1 Bd3 turns ugly.
Or 23 ... Rxb2 24 Rxg7 Bd3 25 Be4 Bxe4 26 Nxe4 c2 27 Nc3 Nb4 28 Rf7 Nd3 29 R7f1 Nf5 30 Bd2 Nxd4 threatening 31 ... Nb3 and 32 c1/Q 33 Bxc1 Nbxc1 34 R(either)xc1 Nxc1 35 Rxc1 Rc8 etc.
So, we can't snap up the f-pawn right away.
Try: 23 N2xc4 Rb4 24 Rfc1 Rhb8 25 Bxc6+ Nxc6 with even material and Black having initiative.

So, I think we have to do [b]20 bxc3[/b].
20 ... Ba6 gets us to LaurV's line: 21 Ne4 O-O 22 Nf6+ etc. so Black has to do something else.
20 ... O-O unless there's something better.

[WMHalsdorf]

I looked at 19 ... bxc3 and found that 20. bxc3 to be the best move following up with 21. Ne4 for just about all of Black's moves.