White 25

[cheesehead]

I think it's safe to assume that Black will play 24 ... Bxf6, and continue our analysis on that assumption.

- - -

From my last candidate move analysis, for 24 Rxf6 Bxf6, in the "White 24" thread:

(Note: in many cases, Black can interpolate with ... bxc3 bxc3 before Nxd4, sometimes with advantage. I've included this in some lines, but may have overlooked a possibility.)

If[B] 25 Be3[/B]
Now 25 ... Nxd4 is not dangerous because Black's dark-square rook can't really fork anything. We just re-take with 26 cxd4 or 26 Bxd4 Bxd4+ 27 cxd4.

Verdict: [B]25 [/B][B]Be3[/B] not compelling

-

If [B]25 Ne4[/B]
Now [B]25 ... Nxd4[/B] is not dangerous because of 26 Nxf6+ and there's no bishop fork.
The immediate [B]25 ... Bxd4+[/B] does not fork anything, either: After 26 cxd4 Nxd4 we have three pieces clustered just where we need them to neutralize Black's pawn-push threats in the southwest quadrant of the board. Black's advanced knight may be troublesome, but can't do real harm. We could even bring our king out a bit: 27 Kf2 prevents ... Ne2+. Or just 27 Ne3.
If 27 ... Nb3 28 Rb1 and we threaten both 29 Bxh6 and 29 Nd6+ and 30 Nxc4.
If 27 ... Ne2+ we just bring our king out with 28 Kf2.

If [B]25 ... Kf7[/B] 26 Rxf6+.

If [B]25 ... Bg7[/B] 26 Bf4 e5 27 dxe5 Nxe5 and [I]my previous analysis here was mixed-up[/I] 28 Ne3 threatens 29 Bxe5 and 30 Nxc4.

If [B]25 ... Bxd4+ 26 cxd4 Nxd4[/B] 27 Ne3.
If 27 ... Nb3 28 Rd1 (allows 29 Rxd2 if 28 ... Nxd2, so as not to pull the N/e4 off station). Now if 28 ... Ba6 29 Nd6+ and 30 Nxc4.
If 27 ... Ne2+ 28 Kf2 Nd4 29 Nxc4.
So, there's not much Black's N/d4 can do after 27 Ne3.
We may swap a knight for Black's other bishop via Nd6+, so as to be left with the two bishops in an open position.

Verdict: [B]25 Ne4[/B] has possibilities

-

If [B]25 Bf4[/B]

If [B]25 ... Ra8[/B] 26 Rc1 and we have possibilities like Bxh6, Bc7, Ne3. Our K and N/c5 are still on the d4 diagonal, but as long as we keep that in mind and have a pawn on c3, we're all right.

If [B]25 ... e5 [/B]Black threatens to eat down our pawn chain with 26 ... exd4 etc. Our c3 pawn is an essential part of our defense against the d4-fork threat, so we can't just ignore the e5 pawn ... unless we deliver a forcing check, that is.

If [B]26 Bxc6+ Nxc6[/B], the absence of the knight on e7 might be significant. Two significant differences from other lines we've analyzed are: Black no longer controls the d5 square, and the e-file in front of Black's king is open except only for the e5-pawn.
If we were willing to sac our remaining bishop, we could play Nf2 and Re1 (possibly a check). But after 27 Nf2 exf4 28 Re1+ we have very little force to bear near the Black king. Black can simply sidestep with 28 ... Kf7 or Kf8 or Kd8.
To take advantage of the weak d5 square, we have 27 Ne3 or 27 d5. 27 Ne3 exf4 28 Nd5 Kf7 29 Rf1 doesn't lead to anything but a drawn-out loss.
27 d5 Ne7 28 Bxh6 Nxd5 29 Ne3 Nxe3 30 Bxe3leaves us only a thin positional advantage (both of Black's king-side pawn are isolated) and Black's light-square bishop is likely to be more lively than our knight. A draw at best, I think.
So I see no advantage to 26 Bxc6+.

If [B]26 dxe5[/B] this choice gains us Black's h-pawn (as far as I can see)!

If [B]26 ... Nxe5 [/B]then [B]27 Ne3[/B] is not good: 27 ... g5 28 Bxe5 Bxe5 29 Nxc4 Bc7 looks evenish.Instead, [B]27 Bxh6[/B] and then:

If [B]27 ... Bf5[/B] 28 Be4 Bxe4 29 Nxe4.

If [B]27 ... g5[/B] 28 Ne4 (not 28 h4 gxh4 29 gxh4 Bxh4) and we not only free our bishop but take another pawn in the process:
If 28 ... Kf7 29 Bxg5.
If 28 ... Bh8 29 Bxg5.
If 28 ... either N moves(except Nf3+?) 29 Nxf6+ and 30 Bxg5.
If 28 ... Rb6 29 Nxf6+ and 30 Bxg5.

Verdict: [B]25 Bf4[/B] and if 25 ... e5 [B]26 dxe5 [/B][B][B]Nxe5[/B] [/B]could be good for us.

Backing up,
If [B]26 ... Bxe5 27 Bxh6[/B] but we must be careful until we get our N/d1 and R activated.
If 27 ... Bf5 28 Rc1 and I think our position holds together long enough for us to get our N/d1 out. (If 28 ... Bd3 29 Nxd3 cxd3 30 Nf2 bxc3 31 bxc3 Rb2 32 Nxd3)

Verdict: [B]25 Bf4 e5 [/B](best for Black) [B]26 dxe5 [/B][B]Bxe5 [/B](best for Black) [B]27 Bxh6[/B] gets us a pawn and struggle.

- - -

I'd forgotten why I've left out 25 Ne3. Here's why:

If 25 ... bxc3 26 bxc3 Nxd4 and now if 27 cxd4 Bxd4 28 Rc1 Bxc5 29 Rxc4 Bb6 and we're a pawn down, with a pinned knight that prevents Bxh6 (and we must not try 30 Kf1 to relieve the pin because of ... Ba6).
or
25 ... Nxd4 26 cxd4 Bxd4 27 Ne4 Bxb2 28 R-moves c3 and our extra knight has to try coping with the extra three Black pawns -- all passed, two connected.

[cheesehead]

Tying loose ends:

[QUOTE=cheesehead;370041]If [B]25 Ne4[/B]
< snip >

If [B]25 ... Kf7[/B] [strike]26 Rxf6+.[/strike][/QUOTE]Oops. There's no rook on f1 anymore in this line.
Instead,
If [B]25 ... Kf7[/B] 26 Nxf6 prevents all the d4-fork threats. 26 ... Kxf6 27 Bxh6 and if 27 ... g5 28 Bf8.

- - -

[QUOTE=cheesehead;370041]I'd forgotten why I've left out 25 Ne3. Here's why:

If 25 ... bxc3 26 bxc3 Nxd4 and now if 27 cxd4 Bxd4 28 Rc1 Bxc5 29 Rxc4 Bb6 and we're a pawn down, with a pinned knight that prevents Bxh6 (and we must not try 30 Kf1 to relieve the pin because of ... Ba6).
or
25 ... Nxd4 26 cxd4 Bxd4 27 Ne4 Bxb2 28 R-moves c3 and our extra knight has to try coping with the extra three Black pawns -- all passed, two connected.[/QUOTE]25 Nf2 instead is similar, except that the pinned knight does not prevent Bxh6:
If 25 ... bxc3 26 bxc3 Nxd4 and now if 27 cxd4 Bxd4 28 Rc1 Bxc5 29 Rxc4 Bb6 and we're a pawn down but can regain it with 30 Bxh6.

[cheesehead]

I briefly wondered whether 25 cxb4 might be viable.

But apparently not: 25 ... Bxd4+ 26 Be3 Bxe3+ 27 Nxe3 Rxb4. Black's a pawn up and his rook is activated.

- - -

More than once now, I've had to search back to find out why an apparently attractive move wasn't on the move candidates list. So, instead of erasing refuted moves from the candidates list, I ought to keep them listed at the bottom, in struck-out form, with the refutation following each.

So, merging in from post #2 above, here are the refuted move 25 candidates:

[strike]25 cxb4[/strike] Bxd4+ 26 Be3 Bxe3+ 27 Nxe3 Rxb4. Black's a pawn up and his rook is activated.

[strike]25 Ne3[/strike] bxc3 26 bxc3 Nxd4 and now if 27 cxd4 Bxd4 28 Rc1 Bxc5 29 Rxc4 Bb6 and we're a pawn down, with a pinned knight that prevents Bxh6 (and we must not try 30 Kf1 to relieve the pin because of ... Ba6).
or 25 ... Nxd4 26 cxd4 Bxd4 27 Ne4 Bxb2 28 R-moves c3 and our extra knight has to try coping with the extra three Black pawns -- all passed, two connected.



and a new one that's not yet refuted:

[B]25 Nf2[/B] is similar to 25 Ne3, except that the pinned knight does not prevent Bxh6:
If 25 ... bxc3 26 bxc3 Nxd4 and now if 27 cxd4 Bxd4 forks rook and knight. If 28 Rc1 Bxc5 29 Rxc4 Bb6 30 Bxh6 Bd7 material is even, but we have a pinned knight.

[cheesehead]

Revised move 25 candidate list (assuming 24 ... Bxf6)

(Note: in many cases, Black can interpolate with ... bxc3 bxc3 before Nxd4, sometimes with advantage. I've included this in some lines, but may have overlooked a possibility.) <== I need to finish reviewing these possibilities.

If[B] 25 Be3[/B]
Now [B]25 ... Nxd4[/B] is not dangerous because Black's dark-square rook can't really fork anything. We just re-take with 26 cxd4 or 26 Bxd4 Bxd4+ 27 cxd4.

What if ... bxc3 bxc3 is interpolated? <==

Verdict: [B]25 [/B][B]Be3[/B] not compelling

-

If [B]25 Ne4[/B]
Now [B]25 ... Nxd4[/B] is not dangerous because of 26 Nxf6+ and there's no bishop fork.
The immediate [B]25 ... Bxd4+[/B] does not fork anything, either: After 26 cxd4 Nxd4 we have three pieces clustered just where we need them to neutralize Black's pawn-push threats in the southwest quadrant of the board. Black's advanced knight may be troublesome, but can't do real harm. We could even bring our king out a bit: 27 Kf2 prevents ... Ne2+. Or just 27 Ne3.
If 27 ... Nb3 28 Rb1 and we threaten both 29 Bxh6 and 29 Nd6+ and 30 Nxc4.
If 27 ... Ne2+ we just bring our king out with 28 Kf2.

If [B]25 ... Kf7[/B] 26 Rxf6+.

If [B]25 ... Bg7[/B] 26 Bf4 e5 27 dxe5 Nxe5 and [I]my previous analysis here was mixed-up[/I] 28 Ne3 threatens 29 Bxe5 and 30 Nxc4.

If [B]25 ... Bxd4+ 26 cxd4 Nxd4[/B] 27 Ne3.
If 27 ... Nb3 28 Rd1 (allows 29 Rxd2 if 28 ... Nxd2, so as not to pull the N/e4 off station). Now if 28 ... Ba6 29 Nd6+ and 30 Nxc4.
If 27 ... Ne2+ 28 Kf2 Nd4 29 Nxc4.
So, there's not much Black's N/d4 can do after 27 Ne3.
We may swap a knight for Black's other bishop via Nd6+, so as to be left with the two bishops in an open position.

Verdict: [B]25 Ne4[/B] has possibilities

-

If [B]25 Bf4[/B]

If [B]25 ... Ra8[/B] 26 Rc1 and we have possibilities like Bxh6, Bc7, Ne3. Our K and N/c5 are still on the d4 diagonal, but as long as we keep that in mind and have a pawn on c3, we're all right.

If [B]25 ... e5 [/B]Black threatens to eat down our pawn chain with 26 ... exd4 etc. Our c3 pawn is an essential part of our defense against the d4-fork threat, so we can't just ignore the e5 pawn ... unless we deliver a forcing check, that is.

If [B]26 Bxc6+ Nxc6[/B], the absence of the knight on e7 might be significant. Two significant differences from other lines we've analyzed are: Black no longer controls the d5 square, and the e-file in front of Black's king is open except only for the e5-pawn.
If we were willing to sac our remaining bishop, we could play Nf2 and Re1 (possibly a check). But after 27 Nf2 exf4 28 Re1+ we have very little force to bear near the Black king. Black can simply sidestep with 28 ... Kf7 or Kf8 or Kd8.
To take advantage of the weak d5 square, we have 27 Ne3 or 27 d5. 27 Ne3 exf4 28 Nd5 Kf7 29 Rf1 doesn't lead to anything but a drawn-out loss.
27 d5 Ne7 28 Bxh6 Nxd5 29 Ne3 Nxe3 30 Bxe3leaves us only a thin positional advantage (both of Black's king-side pawn are isolated) and Black's light-square bishop is likely to be more lively than our knight. A draw at best, I think.
So I see no advantage to 26 Bxc6+.

If [B]26 dxe5[/B] this choice gains us Black's h-pawn (as far as I can see)!

If [B]26 ... Nxe5 [/B]then [B]27 Ne3[/B] is not good: 27 ... g5 28 Bxe5 Bxe5 29 Nxc4 Bc7 looks evenish.Instead, [B]27 Bxh6[/B] and then:

If [B]27 ... Bf5[/B] 28 Be4 Bxe4 29 Nxe4.

If [B]27 ... g5[/B] 28 Ne4 (not 28 h4 gxh4 29 gxh4 Bxh4) and we not only free our bishop but take another pawn in the process:
If 28 ... Kf7 29 Bxg5.
If 28 ... Bh8 29 Bxg5.
If 28 ... either N moves(except Nf3+?) 29 Nxf6+ and 30 Bxg5.
If 28 ... Rb6 29 Nxf6+ and 30 Bxg5.

Verdict: [B]25 Bf4[/B] and now
If 25 ... e5 [B]26 dxe5 [/B][B][B]Nxe5[/B] [/B]could be good for us, but .
If [B]26 ... Bxe5 27 Bxh6[/B] we must be careful until we get our N/d1 and R activated.
If 27 ... Bf5 28 Rc1 and I think our position holds together long enough for us to get our N/d1 out. (If 28 ... Bd3 29 Nxd3 cxd3 30 Nf2 bxc3 31 bxc3 Rb2 32 Nxd3)

Verdict: [B]25 Bf4 e5 [/B](best for Black) [B]26 dxe5 [/B][B]Bxe5 [/B](best for Black) [B]27 Bxh6[/B] gets us a pawn and struggle.

-

If [B]25 Nf2[/B] is better than 25 Ne3 because here the pinned knight does not prevent Bxh6:
If [B]25 ... bxc3 26 bxc3 Nxd4 [/B]and now
If 27 cxd4 Bxd4 forks rook and knight, but
If then 28 Rc1 Bxc5 29 Rxc4 Bb6 30 Bxh6 Bd7 material is back to even, but we have a pinned knight.
If 27 other, ?? <==

- - -

[strike]25 cxb4[/strike] Bxd4+ 26 Be3 Bxe3+ 27 Nxe3 Rxb4. Black's a pawn up and his rook is activated.

[strike] 25 Ne3[/strike] bxc3 26 bxc3 Nxd4 and now if 27 cxd4 Bxd4 28 Rc1 Bxc5 29 Rxc4 Bb6 and we're a pawn down, with a pinned knight that prevents Bxh6 (and we must not try 30 Kf1 to relieve the pin because of ... Ba6).
or 25 ... Nxd4 26 cxd4 Bxd4 27 Ne4 Bxb2 28 R-moves c3 and our extra knight has to try coping with the extra three Black pawns -- all passed, two connected.

[cheesehead]

Who-oa! I've left off [B]25 Bxh6[/B] as a candidate, despite having played with it during analysis.

Maybe there are others? Here's a list of all legal moves:

[B]25 Ra2[/B] removes one target of a d4-fork[B] <==[/B] Analyze further

[strike]25 Ra3[/strike] no apparent value as sacrifice

25 Rb1 ? - removes one target of a d4-fork, but lets ... bxc3 force Bxc3 or Nxc3 instead of bxc3 recapture. Maybe the B at c3 would have value threatening the a5-pawn, offsetting the b2-pawn and d4 weaknesses? Maybe the N at c3 would have value, offsetting the b2-pawn and d4 weaknesses?[B] <==[/B] Analyze further

[B]25 Rc1[/B] removes one target of a d4-fork[B] <==[/B] Analyze further

[strike]25 b3[/strike] no apparent value as sacrifice

[strike]25 cxb4[/strike] already struck from candidate list

[B]25 Ne4[/B] already on candidate list

[strike]25 Ne3[/strike] already struck from candidate list

[B]25 Nf2[/B] already on candidate list

25 Bc1 ? - dubious. Allows Bb2 or Ba3 after ... bxc3 bxc3, but perhaps can wait until that's a reality. (Also, our last retreat of this bishop didn't turn out so well.) Shuts in our R/a1 even further.[B] <==[/B] Analyze further

[B]25 Be3[/B] already on candidate list

[B]25 Bf4[/B] already on candidate list

[strike]25 Bg5[/strike] no apparent value as sacrifice

[B]25 Bxh6 <==[/B] [U]Analyze further![/U]

[strike]25 d5[/strike] no apparent value as sacrifice

[strike]25 Kf1[/strike] no apparent value

[strike]25 Kf2[/strike] no apparent value

[strike]25 Kh1[/strike] no apparent value

[B]25 Bf1[/B] attacks the c4-pawn, but releases pressure on c6 that ties N/e7 down.[B] <==[/B] Analyze further

[strike]25 Bh1[/strike] no apparent value

[B]25 Bf3[/B][B] <==[/B] Analyze further

[B]25 Be4[/B] prepares Rb1, attacks g6-pawn[B] <==[/B] Analyze further

[strike]25 Bd5[/strike] no apparent value as sacrifice

[B]25 Bxc6+[/B] Nxf6 removes the knight from e7, if that has any value[B] <==[/B] Analyze further

[strike]25 g4[/strike] no apparent value

[strike]25 h3[/strike] no apparent value

[strike]25 h4[/strike] no apparent value

[WMHalsdorf]

I to over looked 25. Bxh6 it seems better than 25. Ne4 in fact Ne4 could be a good follow up to 25. Bxh6

[cheesehead]

Yep, they played 24 ... Bxh6.

Our deadline is [URL="http://www.mersenneforum.org/showpost.php?p=370217&postcount=136"]Thursday. April 10, 11:08 UTC[/URL] = 4:08 PM Thai time = 7:08 AM EDT = 6:08 AM CDT

(Did I get that right? Thai time = UTC+5 ?)

[cheesehead]

[QUOTE=cheesehead;370266]
Our deadline is [URL="http://www.mersenneforum.org/showpost.php?p=370217&postcount=136"]Thursday. April 10, 11:08 UTC[/URL] = 4:08 PM Thai time = 7:08 AM EDT = 6:08 AM CDT

(Did I get that right? Thai time = UTC+5 ?)[/QUOTE]Nope.

Thai time = UTC+7

Our deadline is [URL="http://www.mersenneforum.org/showpost.php?p=370217&postcount=136"]Thursday. April 10, 11:08 UTC[/URL] = [B]6[/B]:08 PM Thai time = 7:08 AM EDT = 6:08 AM CDT

[cheesehead]

[QUOTE=cheesehead;370266]Yep, they played 24 ... [strike]Bxh6[/strike].[/QUOTE]Bx[B]f[/B]6, that is.

- - - - - - - - - - -

[QUOTE=cheesehead;370206][B]25 Bxh6 <==[/B] [U]Analyze further![/U][/QUOTE]We did have a bit of analysis in the Move 23 thread:

[QUOTE=cheesehead;368860]For example, after
[B]23 exf6[/B] Rxf6 24 Rxf6 Bxf6[/QUOTE]... which has been the actual course of play ...
[quote]25 Bxh6 Nxd4 26 cxd4 Bxd4+ 27 Be3 e5 28 Bxd4 exd4, material is still even but Black has transformed his e-isolani into a passer in a queen-side pawn majority. Our N/d1 looks worse than Black's N/e7, and Black's B/c8 is unleashed.

So, we'd need to vary somewhere to prevent that outcome[/quote]Oops. Can we vary on move 26, 27 or 28?

[B]25 Bxh6 Nxd4[/B]

- -

If we don't take the knight, material remains even, and Black threatens ... Nb3 or ... Nc2.
If [b]26 Be3 Nb2 27 Rc1 Nxe3 28 Nxe3 bxc3 29 bxc3[/b] and now
If 29 ... Rb2
If 29 ... Nd5 30 Nxd5 exd5 31 Bxd5 Bf5 32 Bxc4 Rc8 33 Bb5+
If 33 ... Kf8 34 Nb3 and we lose the c-pawn
If 33 ... Kf7 34 Nb7 (threatens Nd6+) Be5 we probably lose the c-pawn
If 33 ... Ke7 (speculation) 34 Nb3 Bxc3 35 Rd1 Bc2 36 Rd7+ and it's a struggle
If 33 ... Kd8 (speculation) 34 Nb7+ K-moves 35 Nxa5 Bxc3 36 Nb3 (avoids losing a rook after ... Bd4+) Bd4+ (anyway) 37 Kg2 (not 37 Nxd4 Rxc1+) and we're two pawns up but must be careful until we can swap a couple of men.
If [B]26 what else??[/B]

- -

If [B]26 cxd4[/B]

- -

If [B]25 ... bxc3 26 bxc3 Nxd4 27 Ne4[/B] <== needs analysis

- -

(Gotta break now. I'd better do a merge of move 25 candidates next.)

[cheesehead]

Here's the all-legal-moves move 25 candidate list, reformatted and incorporating all previous posts in this thread.

At top are candidates with significant analysis.

Below those are moves that, on brief analysis, might have some merit, pending further analysis.

Below those are the struck-through moves that seem to have no apparent value after brief analysis, but could be resurrected by further analysis.

At the bottom are ruled-out moves with refutations.

[B]25 Bxh6
[/B]If [B]25 ... bxc3 26 ??
[/B]If [B]25 ... Nxd4 26 ?? [/B](or maybe 26 cxd4 Bxd4+ 27 Be3)If [B]25 ... e5 26 Ne4 ??
[/B]If [B]25 ... Nd5 26 ??
[/B]If [B]25 ... Nf5 or [/B][B]... Ng8 26 Bxc6+
[/B]If [B][B]25 ... Bxd4+ 26 cxd4 Nxd4 [/B][/B]27 Kf2 prevents Ne2+ fork, or 27 Ne3 prevents Nc2 fork, or 27 Bg7 prevents e5, (but not 27 Rc1 Ne2+) <==
If [B][B]25 ... g5 26 Ne4 Kf7 27 Bxg5 [/B][/B](not Nxg5+ Kg6) and we're two pawns up, but must watch the queen-side[B]
<==[/B] [U]Analyze further![/U]

- -

[B]25 Ne4[/B] Now [B]25 ... Nxd4[/B] is not dangerous because of 26 Nxf6+ and there's no bishop fork.
The immediate [B]25 ... Bxd4+[/B] does not fork anything, either: After 26 cxd4 Nxd4 we have three pieces clustered just where we need them to neutralize Black's pawn-push threats in the southwest quadrant of the board. Black's advanced knight may be troublesome, but can't do real harm. We could even bring our king out a bit: 27 Kf2 prevents ... Ne2+. Or just 27 Ne3.
If 27 ... Nb3 28 Rb1 and we threaten both 29 Bxh6 and 29 Nd6+ and 30 Nxc4.
If 27 ... Ne2+ we just bring our king out with 28 Kf2.

If [B]25 ... Kf7[/B] 26 Rxf6+.

If [B]25 ... Bg7[/B] 26 Bf4 e5 27 dxe5 Nxe5 28 Ne3 threatens 29 Bxe5 and 30 Nxc4.

If [B]25 ... Bxd4+ 26 cxd4 Nxd4[/B] 27 Ne3.
If 27 ... Nb3 28 Rd1 (allows 29 Rxd2 if 28 ... Nxd2, so as not to pull the N/e4 off station). Now if 28 ... Ba6 29 Nd6+ and 30 Nxc4.
If 27 ... Ne2+ 28 Kf2 Nd4 29 Nxc4.
So, there's not much Black's N/d4 can do after 27 Ne3.
We may swap a knight for Black's other bishop via Nd6+, so as to be left with the two bishops in an open position.

Verdict: [B]25 Ne4[/B] has possibilities
[B]
- -

25 Nf2[/B] is better than 25 Ne3 because here the pinned knight does not prevent Bxh6:
If [B]25 ... bxc3 26 bxc3 Nxd4 [/B]and now
If 27 cxd4 Bxd4 forks rook and knight, but
If then 28 Rc1 Bxc5 29 Rxc4 Bb6 30 Bxh6 Bd7 material is back to even, but we have a pinned knight.
If 27 other, ?? <==

- -

[B]25 Be3[/B] [B]N[/B]ow 25 ... Nxd4 is not dangerous because Black's dark-square [strike]rook[/strike] bishop can't really fork anything. We just re-take with 26 cxd4 or 26 Bxd4 Bxd4+ 27 cxd4.

What if ... bxc3 bxc3 is interpolated? <==

So far, not compelling. <== needs more analysis

- -

[B]25 Bf4

[/B] If [B]25 ... Ra8[/B] 26 Rc1 and we have possibilities like Bxh6, Bc7, Ne3. Our K and N/c5 are still on the d4 diagonal, but as long as we keep that in mind and have a pawn on c3, we're all right.

If [B]25 ... e5 [/B]Black threatens to eat down our pawn chain with 26 ... exd4 etc. Our c3 pawn is an essential part of our defense against the d4-fork threat, so we can't just ignore the e5 pawn ... unless we deliver a forcing check, that is.

If [B]26 Bxc6+ Nxc6[/B], the absence of the knight on e7 might be significant. Two significant differences from other lines we've analyzed are: Black no longer controls the d5 square, and the e-file in front of Black's king is open except only for the e5-pawn.
If we were willing to sac our remaining bishop, we could play Nf2 and Re1 (possibly a check). But after 27 Nf2 exf4 28 Re1+ we have very little force to bear near the Black king. Black can simply sidestep with 28 ... Kf7 or Kf8 or Kd8.
To take advantage of the weak d5 square, we have 27 Ne3 or 27 d5. 27 Ne3 exf4 28 Nd5 Kf7 29 Rf1 doesn't lead to anything but a drawn-out loss.
27 d5 Ne7 28 Bxh6 Nxd5 29 Ne3 Nxe3 30 Bxe3leaves us only a thin positional advantage (both of Black's king-side pawn are isolated) and Black's light-square bishop is likely to be more lively than our knight. A draw at best, I think.
So I see no advantage to 26 Bxc6+.

If [B]26 dxe5[/B] this choice gains us Black's h-pawn (as far as I can see)!

If [B]26 ... Nxe5 [/B]then [B]27 Ne3[/B] is not good: 27 ... g5 28 Bxe5 Bxe5 29 Nxc4 Bc7 looks evenish.Instead, [B]27 Bxh6[/B] and then:

If [B]27 ... Bf5[/B] 28 Be4 Bxe4 29 Nxe4.

If [B]27 ... g5[/B] 28 Ne4 (not 28 h4 gxh4 29 gxh4 Bxh4) and we not only free our bishop but take another pawn in the process:
If 28 ... Kf7 29 Bxg5.
If 28 ... Bh8 29 Bxg5.
If 28 ... either N moves(except Nf3+?) 29 Nxf6+ and 30 Bxg5.
If 28 ... Rb6 29 Nxf6+ and 30 Bxg5.

Verdict: [B]25 Bf4[/B] and if 25 ... e5 [B]26 dxe5 [/B][B][B]Nxe5[/B] [/B]could be good for us.

Backing up,
If [B]26 ... Bxe5 27 Bxh6[/B] but we must be careful until we get our N/d1 and R activated.
If 27 ... Bf5 28 Rc1 and I think our position holds together long enough for us to get our N/d1 out. (If 28 ... Bd3 29 Nxd3 cxd3 30 Nf2 bxc3 31 bxc3 Rb2 32 Nxd3)

Verdict: [B]25 Bf4 e5 [/B](best for Black) [B]26 dxe5 [/B][B]Bxe5 [/B](best for Black) [B]27 Bxh6[/B] gets us a pawn and struggle.

- - -
[B]25 Ra2[/B] removes one target of a d4-fork[B] <==[/B] Analyze further

[B]25 Rc1[/B] removes one target of a d4-fork[B] <==[/B] Analyze further

[B]25 Bf1[/B] attacks the c4-pawn, but releases pressure on c6 that ties N/e7 down.[B] <==[/B] Analyze further

[B]25 Bf3[/B][B] <==[/B] Analyze further

[B]25 Be4[/B] prepares Rb1, attacks g6-pawn[B] <==[/B] Analyze further

[B]25 Bxc6+[/B] Nxf6 removes the knight from e7, if that has any value[B] <==[/B] Analyze further

25 Rb1 ? - removes one target of a d4-fork, but lets ... bxc3 force Bxc3 or Nxc3 instead of bxc3 recapture. Maybe the B at c3 would have value threatening the a5-pawn, offsetting the b2-pawn and d4 weaknesses? Maybe the N at c3 would have value, offsetting the b2-pawn and d4 weaknesses?[B] <==[/B] Analyze further

25 Bc1 ? - dubious. Allows Bb2 or Ba3 after ... bxc3 bxc3, but perhaps can wait until that's a reality. (Also, our last retreat of this bishop didn't turn out so well.) Shuts in our R/a1 even further.[B] <==[/B] Analyze further

[strike]25 Ra3[/strike] no apparent value as sacrifice
[strike]25 b3[/strike] no apparent value as sacrifice
[strike]25 Bg5[/strike] no apparent value as sacrifice
[strike]25 d5[/strike] no apparent value as sacrifice
[strike]25 g4[/strike] no apparent value
[strike]25 h3[/strike] no apparent value
[strike]25 Bd5[/strike] no apparent value as sacrifice
[strike]25 Bh1[/strike] no apparent value
[strike]25 h4[/strike] no apparent value
[strike]25 Kf1[/strike] no apparent value
[strike]25 Kf2[/strike] no apparent value
[strike]25 Kh1[/strike] no apparent value

[strike]25 cxb4[/strike] Bxd4+ 26 Be3 Bxe3+ 27 Nxe3 Rxb4. Black's a pawn up and his rook is activated.

[strike]25 Ne3[/strike] bxc3 26 bxc3 Nxd4 and now if 27 cxd4 Bxd4 28 Rc1 Bxc5 29 Rxc4 Bb6 and we're a pawn down, with a pinned knight that prevents Bxh6 (and we must not try 30 Kf1 to relieve the pin because of ... Ba6).
or 25 ... Nxd4 26 cxd4 Bxd4 27 Ne4 Bxb2 28 R-moves c3 and our extra knight has to try coping with the extra three Black pawns -- all passed, two connected.

[cheesehead]

(A) Correcting a format error so as not to overlook one possibility

[QUOTE=cheesehead;370362][B]25 Bxh6
[/B]If [B]25 ... bxc3 26 ??
[/B]If [B]25 ... Nxd4 26 ?? [/B](or maybe 26 cxd4 Bxd4+ 27 Be3)If [B]25 ... e5 26 Ne4 ??
[/B][/QUOTE]... should've been

[B]25 Bxh6
[/B]If [B]25 ... bxc3 26 ??
[/B]If [B]25 ... Nxd4 26 ?? [/B](or maybe 26 cxd4 Bxd4+ 27 Be3)
If [B]25 ... e5 26 Ne4 ??

[/B][B]* * * * * * * *

[/B](B) Expanding on[B] [B]25 Bxh6[/B][B] bxc3 26 ?? :[/B][B]
[B]
25 Bxh6
[/B]
[/B]If [B]25 Bxh6[/B][B] bxc3 26 bxc3[/B][/B]
But now 26 ... Nxd4 does not merge into the analysis for [B][B][B]25 Bxh6 Nxd4[/B][/B][/B] because now the d4-fork _does_ threaten our rook, and Be3 will not suffice to answer it.
If [B]26 ... Nxd4 27[/B] ?? <= needs further analysis

[B]* * * * * * * *[/B]

(C) Expanding on [B]25 Bxh6 Nxd4 26 ??[/B] : The ?? could be cxd4 or Ne4 because the d4-fork has no danger for us.

[B]25 Bxh6
[/B]
If [B]25 ... Nxd4

26 cxd4 Bxd4+ 27 Be3[/B]
If [B]27 ... Bxc5 28 Bxc5[/B] we're up a bishop for a pawn. I don't think Black has any tricks to change that for the better.
(speculation) If 28 ... Nd5 or ... Nf5 29 Ne3 threatens to take the c-pawn unless (or even if) Black swaps another piece.
If 29 ... Nb6 30 Bxb6 Rxb6 31 Nxc4. We're a whole knight up and two of Black's pawns are isolated.
If 29 ... Nxe3 30 Bxe3. Soon Be4 will both attack a K-side isolani and guard against Q-side Black adventurism. E.g., 30 ... c3 31 bxc3 bxc3 32 Be4.
If 29 ... c3 30 Nxd5 exd5 31 bxc3 bxc3 32 Bxd5 c2 33 Be4 Bf5 34 Bxf5 gxf5 35 Rc1 Rb1 36 Be3 and the c-pawn will not queen.
If [B]27 ... Bxd3+ 28 Nxe3[/B] we're up a knight for a pawn (soon to be up a whole knight).
(speculation) If 28 ... c3 29 bxc3 bxc3 30 Rc1.

or [B]26 Ne4
[/B]If[B] 26 ... Ne2+ 27 Kf2 Nxc3 28 Nxf6+ Kf7 29 bxc3 Kxf6 30 cxb4 [/B]and we're up a whole bishop at this moment.
If 30 ... g5 31 b5 Kg6 32 Bf8 Kf7 33 Bxe7 Kxe7 34 Rc1 and we have a winning endgame.
If [B]26 ... ??[/B] <== needs more analysis