Black to move. Can you find the brilliant winning plan for black?
Shankland – Sasikiran (Capablanca Memorial 2017)
- About Us
- Chess Improvement
- Chess Puzzles
- Chess Research
- College Chess
- General News
- Home
- Major Tournaments
- News
- Polgar Events
- Privacy Policy
- Scholastic Chess
- SPICE / Webster
- Susan’s Personal Blog
- Track your order
- USA Chess
- Videos
- Women’s Chess
- Contact Us
- Daily News
- My Account
- Terms & Conditions
- Privacy Policy
1. Kd3 Bc5 2. Be4 and now White must advance his K side pawns because taking the bishop off the a3g8 diagonal allows b4 forcing the creation of a passed a pawn e.g.
2. … Bb6 3. b4 Bc5 4. ba Bxa3 5. Kxe3
so 2. … h3 3. Bf5 Bb3 4. Kxe3 g4 5. hg hg 6. Be4 and White wins.
I meant 3. Bf5 Bb4, Black has 6. … Bd2+ but it’s not enough to draw.
4. Bxh3 was a lot simpler and I would have played that in a game but I was trying to find the fastest line and went wrong.
James,
1. ………Kd3
2. Bc5 Be4
3. b4 axb4 and I think you need to do a bit more work here. I am not saying the first two moves are wrong, but I am not sure they are right either. I can see some potential issues that have to be resolved in the line I just gave.
Yes, I did see the idea of Black simply playing Kd3 Ke2 Kf3 but I was obviously overdoing the amount of willpower I was putting into trying to find a brilliancy and going a bit crazy because I was calling Black White and White Black. Although paradoxically I was probably thinking better than on the last problem which I think was much harder and I was just being lucky on that and making double mistakes that cancelled each other out and stuff.
1. …Kd3
2. Bc5 Ke2
3. Bd4 Be4
4. h4 Bf5
5. Bc5 Kf3
6. Kd2 Kxg3
7. Be7 Kf3
8. Bb4 c1=Q+
9.Kxc1 Kxe3
10. Bd6 d4
11. Kd1 Kf3
12. Kd2 Kg4
13. Be7 Kxf4
It’s a winning position for black but it requires a lot of maneuvering
It is late for me, and I have something I do have to get up for in the morning, but I will add to James’ plan:
1. ………..Kd3
2. Bc5 Be4
3. Bb6 Ke2!?
Going for the kingside pawns. What I can sort of foresee here is forcing white to protect the h-pawn from g5 with the bishop after the pawn has been forced to h4, and protecting the e3 pawn with the king on d2, but then white can sacrifice the c-pawn to win at e3. I think the way to set this up is with the line above, but I can’t quite see myself clear that black can force the position I have in mind. That needs a bit of work, and I don’t have the time for it tonight.
A fascinating position.
And I see Alena has a sort of similar idea, but seems to have diverged from the line I have in my head. Maybe tomorrow afternoon, I will have to time to study this in more detail.
It is simple plan to go with king to g2 and capture pawns. I didnt found a right plan for a pawn to promote, because black controls a1 – h8 diagonal.
1. … Kd3
2. Bc5 Bd7
3. Bb6 b4 ( 3. Bb4 Kxd3 4. Kxc2 d4 and Black’s position looks good to me )
4. ab a3
5. Bd4 a2
6. h4 Ba4 ( 6. b5 or h3 don’t work because Black’s bishop can handle a Kside pawn breakthrough attempt or a b pawn advance and still get back in time to stop the other threat, and 6. Kb2 can be met by a1=Q+ )
And I can’t see any way for White to avoid losing.
Hey, James, this might be the most straightforward solution- I hadn’t really thought about 2. ….Bd7 at all, but it does help in one the issues I was having with your previous lines late last night, which was the issue of the white b-pawn created by black’s b4 move. I don’t have time to do a complete analysis, but I don’t think I will find a flaw in this plan of yours. Right now, I would have to say, nicely done!
I thought a bit more about this last night after I went to bed, but I think the right plan is going to require a lot of maneuvering by the black king from one side to the other. I think the plan should start as I and Alena outlined above- go for the king side pawn first with 1.Kd3 and 2.Be4 3.Ke2/2.Ke2 3.Be4. This should result in winning the white g-pawn and force white to advance the h-pawn to h4 in one or two moves (I don’t think it changes things if white goes h3 then h4 rather than just h4). In such a set up, white can protect h4 from the h4-d8 diagonal, but not from g5 specifically- it is important that white retain the time to transfer the bishop to the g1-b7 diagonal to protect the e3 pawn from the black king since black can sacrifice the c-pawn if white tries to protect e3 with the king from d2.
Once black has won at g3 and force h4, then the king marches back to the queenside and gives white a choice- either give up protection of e3 with the bishop, or take the bishop off of the a3-f8 diagonal and allow black to play b4 followed by capture of the new white b-pawn. The creates the second passed pawn on the a-file, and black can advance this pawn to a2. Black then starts to transfer the king the long way through e6 to g4 to attack the h-pawn, during which white must keep the bishop on the long diagonal a1-h8 just to prevent a1(Q) since white can’t put the king on b2 without allowing a1(Q) followed by c1(Q). Once the black king reaches g4, the white bishop is forced to the square of f6 to guard both the h4-pawn and the a1 square. At this point, white is forced to play the king from c1 to d2 just to have a safe move, but then black continues his king first to g3 keeping the attack for a move on h4 forcing the white king back to c1. Then black plays Kf3 attacking e3- white either protects with Kd2 allowing c1(Q) followed by Kxe3, or plays Bd4 allowing a1(Q) followed Kxe3. After e3 falls it should be a straightforward won ending.
When I get back from taking my father to his last appointment on the day, I will outline a concrete example of this plan, but a careful reader should be able to construct it from this brief outline.
I think I have to give the nod to Jame’s first solution, and on deeper consideration, have to ding the comment I praised earlier, though it doesn’t fail to win.
There are a several “lines” to win with here, but they all boil down to the same basic idea- winning the e3 pawn by hook or by crook since e3 supports f4 and is backward itself and, thus, weak. For me, the real issue was deciding when black could force the passed pawn on the a-file, and last night I was unconvinced that this could be done immediately, but I think James was right to force the issue immediately. I will just fill in a few details:
1………… Kd3
This move should be obvious to most players- it attacks the keystone pawn of white’s defense, and basically forces white to play…
2. Bc5
The only way to protect white’s position- if e3 falls, then so does h2 or g3, and later f4. Now, in his first comment, James suggested 2. …Be4 followed by a push of b4. I want to examine this line first before dealing with James’ later line with 2. ….Bd7
2. …………Be4
The idea behind this move is to wait, but it probably isn’t needed- black could probably play b4 right now since white will be forced to play axb4 anyway since Bxb4 allows Kxe3, but I am not completely sure the lines are equivalent, so I will follow the 2. ….Be4 line instead, which forces either a pawn move on the king side, or a bishop move along the d4-a7 diagonal. Continuing:
3. Bb6
It is unlikely that any pawn move on the king side makes a difference here- black can always wait those moves out and white will still be forced to play a bishop move like this since the only king move, Kb2 loses almost instantly to black’s reply of Kd2. Continuing:
3. ………….b4
4. ab4 Kc4
So, black has successfully created the passer on the a-file, but things are not done here- the nature of the ending, bishops of opposite color and the position of the white king means that white still can defend against two passed pawns, and black can’t in this position force the sacrifice of the bishop for one of them, at least not yet. Continuing:
5. Bd4 a3 (should be obvious)
6. Be5 a2
This freezes white’s bishop to the a1-h8 diagonal. White cannot play Kb2 guarding a1 because black will just sacrifice the a-pawn to queen the c-pawn. Black could have taken the b4 pawn before advaning a3 and a2, but it wasn’t necessary since white couldn’t save it anyway. Continuing:
7. Bd4 Kb4
8. Be5
So, now black has to make the next plan- it isn’t possible to win with the two passers right now. Black could try to return to d3, but white can now cut this path off with Kd2, so I think the right approach is to go the long way around and get the king onto the h3-c8 diagonal to reach g4:
8. …………..Kc5
9. Bd4 Kd6
10. Be5 Ke6
At this point, it should be noted, though I think it should be obvious, that white can’t put the bishop on b2 and bring his king to the kingside- black would again just sacrifice one passed pawn to queen the other- it takes both white men to hold up the two pawns, and black has complete freedom to attack the king side. Continuing:
11. Bb2 Kf5
12. h3
Attempting to prevent Kg4, but this is hopeless since black has two paths…..
12. …………Bd3! (opening the e4 square for the king)
13. Bd4 Ke4
And white is pretty much in zugzwang- black is going to play Kf3 if white plays Kd2 and is going to take at e3 if white moves the bishop. With the kingside pawns basically toast, the best I see for white is to try to exchange them off……
14. g4 h4!
I think is clearly best for black- having the h-pawn on a dark square is no danger to it- the white bishop can never leave the a1-h8 diagonal, and now the h-pawn is completely indefensible. White could continue here with f5, but it wouldn’t change the outcome. The only plausible move here is…..
15. Kd2 Kf3
Black could probably play Bf1 here and give up the c-pawn, but why take the risk of that? Continuing:
16. Bb2 Kg3
17. Bd4 Kh3
And it should be obvious that black wins from here.
So, Jame’s first idea was ok, but had to be fleshed out. My main concern last night when I looked at it was having white put the king on d2 and leaving it there while cutting out g4 with h3, but I hadn’t yet thought things through in detail- in the end white couldn’t cut the black king off from the king side. So lets walk through the 2. ….Bd7 line for a brief moment
1. ………….Kd3
2. Bc5 Bd7
I gather James’ idea was to leave the king on d3 for as long as possible after playing b4- the bishop covers b5 when pawns are exchanged at b4, but I think James might have underestimated white’s potential reply at move 3 with….
3. Bb4
This is still a winning position for black, but I am not sure black can trade the c-pawn immediately for the e-pawn:
3. ………….Ke3
4. Kc2 d4
I have not considered this position in great detail, and don’t plan to, but I think d4 is an obvious move here, but just not as convincingly winning as the lines above. Continuing:
5. Bc5 Ke4
6. Kd2 d3
7. Bd6
And I think black has to reorganize things to protect d3 with the bishop before continuing over the king side. It may still be won, but I am not convinced of this yet. Just saying- I like the clarity of 1. …..Kd3 2. …..Be4 line even if it is rather long.
I just realized this wasn’t a designed puzzle but was an actual game at the Capablanca Memorial. One should study Sasiskiran’s method- it employs many of the motifs discussed above, but is in many ways far more elegant- he goes for the king side g-pawn and leaves the king over there, then creates the passed a-pawn with b4 daring white to take with the bishop at b4 instead of the a-pawn, but later forces the issue by blocking up the a1-h8 diagonal by pushing d4 at a critical point forcing the loss of the h-pawn.
White ends up using the passed b-pawn to get the black bishop off the guard of c2, and white eventually even wins the pawn on a2 at that point, but it matters not since white never has enough time to clear a line for the bishop on the h2-b7 diagonal, and the h-pawn can’t be stopped. For an over the board finish, it was an outstanding effort.
That was my comment from “y”. I forgot to accept the autofill of my name.
It looks to me like 1….Kd3 2. Bc5 Bd7 3. Bd4 b4 4. ab a3 5. Bf6 Ba4 6 Bd4 a2 leaves white with an insoluble dilemma.
His B needs to stay on the a1h8 diagonal.
His K can’t move except to b2, and as several people have noted …a1Q+ is a killer if he does that.
His b-pawn can’t move while the black K guards the c2 pawn.
It looks like the only rational move is 7. h4 (7.h3 Bd7 8. h4 Ba4 just leads to a worse zugzwang dilemma since white now has to unguard the e-pawn while the B guards the c2 pawn) when black just shuffles the B to, say, b5 forcing the B off of d4, then grabs the e-pawn in exchange for the c2 pawn and the march of the d-pawn decides. Whether White moves his B to a1 or g7 is irrelevant, in either case the B must give itself for the d-pawn when it gets to d2.