White to move. It is complicated but can you find the winning plan for White?
Source: ChessToday.net
Chess Daily News from Susan Polgar
Women’s GP chess tactic
About Author
White to move. It is complicated but can you find the winning plan for White?
Source: ChessToday.net
It has to be the deflecting of the Black Q to be able to capture the Black R and keep control of the first rank:
1.Ra7+ Qxa7 (otherwise if 1…Ke8, 2.Ra8 is the same)
2.Qxd1 should win with the material advantage.
Martin
Ra7+ Q*a7 (if king goes to back rank, then its Ra8+), Q*R
Point is: Queen is forced to move off the b file
1. Ra7+ Ke8 or Kf8 (Kd8 Qxd1+)
2. Ra8,
forcing the echange of either the rooks or the queens. In both cases b2 can be stopped and White hopefully wins thanks to the extra bishop.
All I can see as a start is to try to force black to take the rook at a7 to relieve the guard on the the b2 pawn and especially the hidden guard on b1. Let’s see if we make any progress this way:
1. Ra7
Now, black can move the king back to the 8th rank, or black can take at a7 and lose his rook at d1, or black can block the check with Rd7. I am forced to look at all three methods to be sure which is best, but I can already see what is likely to be best for black. Let’s look at the weaker moves first:
1. …..Qa7
2. Qd1 Qb6
3. Qd6 Qd6
4. ed6 Kd6
5. Bd3 and white should win this since black cannot approach b2 before the white king gets there since black must proceed through b6, a5, and a4, and white can just drop the bishop back to c2 to guard a4 anyway.
Now, let’s look at the rook block at d7 before proceeding to what looks to me as the most viable plan:
1. Ra7 Rd7
2. Rd7! Kd7
3. Bb5! and white will win the b2 b-pawn and the game.
So, now let’s look at the king moves at move 1. There are three of them, but we can discount Kd8 since white takes at d1 with check and will mate black in 4 moves total: [1.Ra7 Kd8 2.Qd1 Ke8 3.Qd7 Kf8 4.Qf7#]. However, 1. …Ke8 isn’t much better:
1. Ra7 Ke8
2. Qd1 Qa7
Of course, 2. …b1Q gets black mated after 3.Qd7 and 4.Qf7, too. The other option here was 2. …Kf8 but white is too quick with his own passed pawn that arises after 3.Qd6+: [2. …Kf8 3.Qd6! Qd6 4.ed6 b1Q 5.Ra8! (not 5.d7? Ke7 is more difficult, and might not win at all though white is still ahead in my opinion) 5. …Kg7 6.d7 white will queen without delay or win black’s queen in exchange]. Continuing from move 2 above:
3. Qb3 and white will win the b-pawn and the game.
Finally, let’s look at what seems the most resistant defense for black, but, I think, still losing:
1. Ra7 Kf8!
2. Qd1 b1(Q)
The difference between 1. …Kf8 and 1. …Ke8 for black is that in the latter line, queening at b1 wasn’t possible due to the mate that starts with Qe7+- with the king at f8, white doesn’t have this tempo from the check. Continuing:
3. Qd6
Hard for me to say which plan is best for white here. I will cover the alternatives for white in a further comment, but will look at this one first since it seems the clearest to me:
3. …..Qd6
Very complicated, with lots of variations, some of which are so similar looking that I am having a hard time evaluating them. Here, black could retreat the king to g8 or g7, but are they better or worse? You tell me: [3. …Kg8 4.Qd7 (or 4.Qe7 Qf8 5.a5 Qe7 6.Re7 Kf7 4. …Qf8 5.a5 Qe4 6.a6 Qd5 (or 6. …Qe5 7.Rb7 Qf6 8.a7+-) 7.Qd5 ed5 8.Rd7 Qc8 9.Bb5 Qc3 10.a7 Qe1 11.Bf1+-]; or [3. …Kg7 4.Qd7 Qf8 5.Qe6! Qe4 6.Qf6 Kg8 7.Rd7 Qa4 8.Rd8+-]. Continuing from move 3 above:
4. ed6 Qe4 (what else?)
5. d7
More complications. Again, I am having a hard time distinguishing between variations, and this makes it possible to overlook better plays for both sides. White could push any of the passed pawns and be winning, or throwing the win away. My default is to push the most advanced pawn most of the time, even though it allows the black king to participate. Continuing:
5. …..Ke7 (what else?)
6. a5 h6 (only counterplay)
7. a6 g5
8. hg5 h5
9. d8(Q)Kd8
10.Rf7 Ke8
11.Rf4 Qd5
12.g6 and white should win this easily as the queen and king are helpless against the rook and bishop and the three passers.
The only move that seems to create the necessary dynamics is Ra7+.
If Qxa7 Qxd1 and white seems to be winning.
So black doesn’t play that, but goes Kf8. Then Qxd1 b1Q Qf3, threatening mate via Qxf7. So Qe8 for black to protect f7.
Now Qf6 Kg8 Re7 Qf8 Rd7, threatening Re8 which loses the queen for the rook. After all this there’s a winning material advantage for white.
I like 31. Rb5 (as played in the game) but am not sure how to meet 31 … Qd8
The idea is 32. Qxb2 Qd3
Maybe white has a perpetual, but there is no other reasonable defense to the threat at f1.
I missed that Ra7 can be taken after Qf3:)
Damn, I need to get some sleep. Was out playing poker all night, and watching Judit play this morning. Sheesh! I set up the wrong position on my board. Somehow missed those two black pawns on the c-file. Ok, back to the drawing board. Which makes sense now, as I didn’t find my pretend position all that “complicated” to solve. With no bishop hole at d3, much more difficult. My 1.Ra7 definitely won’t win, though it might still draw.
I suspect it starts with 1. Rb5, but I’ll let those who like to write long analysis posts to go further.
I am looking at Ra7+ (if the K moves to 8th rank, then Ra8 to follow) – the idea that if QxR, White takes the Black rook while protecting a4
Ok, let’s try this again with the right position. With no way to play Bd3 at some point, clearly white must either play 1.Qd1, or 1.Rb5 right now. I will cover Rb5 later, but let’s see if I can at least salvage something of my ideas in the erroneous first comment:
1. Qd1 b1(Q) (what else?)
2. Qf3
Ok, I can’t save anything from the earlier effort. I had hoped to be able to utilize the Qd6 idea here, but that draws at best for white: [2.Qd6? Qd6 (or 2. …Ke8?? 3.Qc6 Kf/d8 4.Rb5+-) 3.ed6 Kd6 and only black has real winning chances here since the passed a-pawn and the R + B are just not enough to force a win]. White could try taking at b1: [2.Qb1 Qb1 3.Rc5 Qe4 4.Bc4 and this might be winnable for white]. Also, white could try 2.Rb5 immediately: [2.Rb5 Qd1! 3.Rb8 Qa4 and only black can win this]. The idea behind 2.Qf3 is to threaten 3.Qf6/4.Ra8/5.Qh8 winning one of the queens for the rook and retaining the a-pawn for the victory. Even keeping the king on the seventh rank isn’t going to help since white can take at f7 and get a mating net in some lines:
2. …..Kf8
Nothing here looks like it can hold to me. White is threatening not only Qf6 but Rb5 skewering the queens. If the b8 queen leaves the guard on a7, the white rook and queen will mate starting with Ra7+: [2. …Qe5 3.Ra7! Kd6 (or 3. …Kd8 4.Qf7 and black can’t prevent both Qg7# and Qd7#) 4.Qf7 and now black can’t defend against Qd7# other than delay it]. If the b1 queen gets off the b-file to prevent the skewer with the rook, white will still win: [2. …Qa2/a1/c2/c1 3.Qf6 Kd7 (or 3. …Ke8/f8 4.Qh8+-) 4.Qf7 Kc8 5.Qe8 Kc7 6.Rc5 Kb7 7.Qc6 Ka7 8.Ra5#]. If black brings the new queen back for defense, it won’t help much: [2. …Qb6 3.Qf6 Kf8 4.Ra8! and white will win one of the queens for the rook and retain the a-pawn]; [2. …Qb7 3.Qf6 Kf8 4.Rb5 Qb5 5.ab5 Qb5 6.Qh8 Ke7 7.Qc8 and white will eventually maneuver to win the c4 pawn, and/or force the exchange of queens and win with the extra piece]. Continuing:
3. Rb5 Qb5
4. ab5 Qb5
5. Qe2 and white should win this.
Now, at move 1, white could try 1.Rb5:
1. Rb5 Qb5?
2. Qd1! b1(Q)
3. Qd6 Ke8
4. ab5 Qb5 is similar, however, this isn’t forced as black need not cooperate at move 1:
1. …..Qd8! (planning Qd3)
2. Qc2 Rd2
3. Qb1 Rd1 with a likely draw, though I might be missing a better line or white here, I am just not sure.
Fascinating position, though.
So, Zhao did play 1.Rb5. I completely missed the line they actually played into a queen ending:
1. Rb5 Qb5
2. Qd1 Qb6
Guarding d6 square, and white returned her queen to b1:
3. Qb1 Qb3 (guarding c4)
4. Bc4
And Mungutuul continued with 4. …Qc4 and ended up losing the ending, though I am not sure the best play for both sides wasn’t a draw.
Hi Susan Polgar,
Nice puzzle.
White wins the game [ Variations exist ]
Example
=======
1.Q*Rd1 b1(Q)
2.Ra7+ Q*Ra7 [ Variations exist ]
3.Q*Qb1 Q*a4
4.Qb7+ Kf8
5.Qc8+ Kg7
6.Q*c5 Qb3
7.Q*c4 Q*Qc4
8.B*Qc4 f6 [ From here for white its child’s play to win the game ]
9.e*f+ K*f6
10.Kf1 Ke5
11.Bd3 h6
12.Ke1 g5
13.h*g h*g
14.c4 Kd4
15.Kd2 g4
16.Kc2 e5
17.Kd2 Kc5
18.Kc3 Kc6
19.Kb4 Kc7
20.Kb5 Kd6
21.Kb6 Kd7
22.c5 Kc8
23.c6 Kb8
24.c7+ Kc8
25.Ba6+ Kd7
26.c8(Q)+ Ke6
27.Qc6+ Kf7
28.Bc4+ Kg7
29.Qe6 Kh7
30.Qf5+ Kg7
31.Bf7 Kf8
32.Qe6 Kg7
33.Bh5 Kh7
34.Q*g4 Kh8
35.Bf7 Kh7
36.Qg6+ Kh8
37.Qg8++ Mate
White wins the game.
By
Venky [ India – Chennai ]
Yancey –
31. Rb5 Qd8 32. Qc2 Rc1 33. Qxb2 (what else?) Qd1
Tom,
I had not considered that line continuation at all, but I should have- I really don’t know how I missed even looking at it. White might actually be ok with taking at c1 after Qd1- the a pawn can pushed, and quickly, as I found in a lot of the lines I studied in the incorrect comment I posted first this morning:
31.Rb5 Qd8
32.Qc2 Rc1
33.Qb2 Qd1
34.Qc1 Qc1
35.a5
And now black has a decision to make, and I am not really sure what is best for black. He can start immediately bringing the king into play with Kd7, or he can take at c3, or he can bring the queen to the a-file at a1 or a3. I don’t like the looks of 35. …Qc3:
35. ….Qc3
36.Rb7
I like less the push immediately with 36.a6 because it gives black time to take at e5 and prevent white from pushing the black king back to the 8th rank with Rb7+ (white’s e5 pawn prevents black from playing Kf6 or Kd6), a pretty important motif for threatening to actually getting the pawn the 8th rank at all: [36.a6 Qe5! 37.Rb7 (or 37.a7 Qa1 38.Rb7 Kf6! and white has a problem with black’s c-4 pawn, and white cannot get his bishop onto g2 to support a8 because it is pinned to the king, and the king can’t go to h2 to both unpin it and protect it at the same time!) 37. …Kf6 38.Kh2! Qa1 39.Bc4 Qd4 40.a7 Qf2 41.Kh3 Qf3! and black either gets a perpetual with Qh1 and Qxe4, or he will win the pawn at a7 by playing 42. …Qa3 if white tries 42.Rb1 to cover h1. Now, I might have missed something here, but the line looks plenty clean to to me]. Continuing from move 36 above:
36. ….Kd8
37.a6 Kc8
I don’t see any other possible way to hold this for black- the king must get into the play and attack the rook at b7 to prevent an immediate a7. Continuing:
38.Rb5! Qa3
Black has other options here, but I don’t know if any of them are really all that different since I think white plays Rc5 well in all of them. With 38. …Qa3, white can indirectly protect a6 by taking at c4 with the bishop (threatens a deadly discovered attack). Continuing:
39.Bc4
And I think only white has winning chances here, though, what that winning plan my be from this position is a mystery to me right now.
Back at move 35, black could put the queen on the a-file immediately:
35. ….Qa1
I like this better than Qa3 simply because it keeps white’s king and bishop pinned together. Continuing:
36.Rc5
Can white do better with 36.Kg2 freeing the bishop. I honestly don’t know and don’t have the time tonight to look at it in detail, but Kg2 can be played in this line anyway, so it may just all lead to transpositions. Continuing:
36. ….Kd7
Several possibilities here, like h5, Qa3, Qc3, etc. Too many lines to be sure what is best. Continuing:
37.Kg2 Qa3
38.Rc4 Qa5
And I would guess this is about even, but I don’t see how white advances the passed c-pawn while protecting and coordinating the two pieces along with the two e-pawns. This would take a long time to analyze.
Also, at move 35, the plan to bring the king over still looks drawish to me, but is even more complicated as it drops king side pawns for black to win the a-pawn:
35. ….Kd7
36.a6 Kc6
37.Rb7 Qa1 (keeping Bf1 pinned)
38.Rf7 Qa5
39.Rh7 Qa1
40.Re7 Qc3
41.Re6 Kd7
42.Rg6 and who is better here? I couldn’t even guess right now.
Interesting idea, Tom, but I don’t think, on balance, that Rc1 is something black wants to play. The lines above suggest it is still drawish, but might be more challenging for black to actually hold, and I haven’t even looked at white’s other alternatives at move 34, some of which look at least plausible to me, though I haven’t done a look more than a move or two deep.
Yancey —
31. Rb5 Qd8 32. Qc2 Rc1 !? 33. Qxb2 Qd1 34. Qxc1 Qxc1 35. a5 and I agree that this is dangerous for black.
But in addition to your 32 .. Rd2 there is also 32 .. Qd3 immediately!
e.g.
33. Qxd3 (33. Qxd1 Qxd1 34. Rxb2 Qxa4) cxd3 34. Rxb2 d2 35. Kg2 Rxf1 36. Rxd2 Ra2 and after 37 … Rxa4 black looks comfortable
Tom,
Yes, that ending is probably a draw, too. Another line that was invisible to me yesterday morning. In that line, white might try pushing the king back to the 8th rank (to e8 specifically) before capturing at b2, then bring his own king through f3 to e3 to guard e3 (and maybe threaten to get his king somehow into the action on the c-pawns), but then I still have a hard time finding any plan that works:
31.Rb5 Qd8
32.Qc2 Qd3
33.Qd3 cd3
34.Rb7 Ke8
If black plays the king to d8, then the capture at d2 below will occur with check and black won’t be able to take at f1 without losing his rook, and will certainly lose. And if he plays the king to f8, then white can retain the a-pawn by playing Rd8 with check and then play Ra8 (or he can win black’s c-pawn with Rc8 and keep his c-pawn-I have no idea which is better, or if white can even win this even being up 2 pawns. Continuing from move 34 above:
35.Rb2 d2
35.Kg2 Rf1
36.Rd2 Ra1
37.Kf3 Ra4
38.Ke3 Ra3
Now, is this the best move for black? I haven’t a clue. Continuing:
39.Kd3 Kd7
This, of course, is the problem with white’s Kd3- it allows the black king a chance to get across the d-file himself.
40.Kc4 Kc6
I don’t see how white can take advantage of the check at from d6 since the rook will have to return to the 2nd rank to protect f2 and h2. Continuing:
41.f3
To protect e4 from the check from a4 by the rook. Continuing:
41. ….Ra4
42.Kb3 Ra1
And I will stop here since I am already deep into the weeds of speculation. I would not want to be playing black here- it would be torture, though I I have a hard time envisioning any winning plan for white here. Probably best, at this point, to simply put these lines into a high-powered engine like the latest Fritz and see if it can even offer a definitive answer of any kind (there may be a big, big hole that we can’t spot). I am at my limit since I can’t keep track of more than about 4 or 5 variations, even with a pen a paper.