It’s so sad, unfortunately I watched the game, so Carlsen spoilered the solution.
If I remember correctly white allows black to take the bishop and wins by promoting a pawn.
I would have never ever found the combination by myself.
It’s really not that hard. Look at the board … what is the most notable positional feature? If you didn’t say “white’s protected passed pawn”, then look again. Can the white king reach that pawn to help it advance? Certainly if the black king goes after the white bishop … it takes 7 moves to go take the white bishop and return to d7 to protect the black bishop, but only 6 moves for the white king to get to f6 and force the bishop to move, grabbing the h pawn on the way. (Carlsen says he didn’t even bother to count.) And if black doesn’t go after the white bishop, white is none the worse off … and in fact he still wins — he plays Bb7, forcing the black bishop to d7 to defend the c pawn, and the white king is able to infiltrate — the black king doesn’t have enough available squares to prevent it. If black plays Bd5 instead of Bd7 then white plays Bc8, winning the f pawn or forcing a bishop trade, which still wins the pawn because the black king can only defend the f pawn from e6, whereas white can attack it from either g5 or g6.
Like Albi, I had already seen this game (I watched Agadmator’s video yesterday covering this game). Of course, the right idea is to go after the h4 pawn with 1.Kf2. This would be a better puzzle if the initial position had been on white’s previous move of capturing at a6 allowing black’s b5 trapping the bishop and asked if it was ok to capture at a6 given that b5 traps the white bishop.
The point of going after the h-pawn is that white is also threatening to bring the king up the board to support the advance of the e-pawn, thus winning the bishop back or winning the f-pawn and creating connected passed pawns.
Nakamura didn’t play the stingiest defense at that point- but it was probably the best one to play in hoping for Carlsen to err. Nakamura accepted the challenge, and captured the bishop with the king, and Carlsen won the h-pawn and the f-pawn, used the time wasted by Nakamura to put the backward c-pawn on a dark square, and then used the two connected passed pawns to win the black c-pawn- all without ever winning the black bishop. Just an amazing endgame calculation by the World Champion. The technique he used in a rapid game was absolutely flawless- not a single error. I could see that white wouldn’t lose- the idea was instantly spotted by me while I was watching the video, but I couldn’t be sure that it was winning.
The stingier defense would have been to not go for the trapped white bishop, this would have held the h-pawn and kept the white king from infiltrating, but then the white bishop overloads the black bishop and picks up either the c-pawn or the f-pawn, or forces the black king back to protect the f-pawn and allows the white king into black’s half of the board.
Eh? Black can’t hold the h pawn. (But take black’s f5 pawn off the board and he wins easily.) White follows the same plan regardless … Kf2-g2-h3-h4. The black king can block him with Kg6-h6, but white plays c3, Bb7, and e6, winning the c6 pawn and the game.
reply toThe real point is keeping the white king out on the king side. Since I wrote this comment, I spent some time exploring the lines that I found on Chessbomb’s Stockfish analyzer- the key one, black can repel the white king, but it isn’t enough- the white king in the key line makes the long journey to the queenside to win. Fascinating stuff.
Basically, you can sort of solve this puzzle by counting how many moves it takes black to win the trapped bishop and then get back to defend the black bishop against the white king and the passed e-pawn. It is made more complicated by having to consider what happens if black ever has time to play b4 fixing the backward c-pawn on a light square.
It’s so sad, unfortunately I watched the game, so Carlsen spoilered the solution.
If I remember correctly white allows black to take the bishop and wins by promoting a pawn.
I would have never ever found the combination by myself.
It’s really not that hard. Look at the board … what is the most notable positional feature? If you didn’t say “white’s protected passed pawn”, then look again. Can the white king reach that pawn to help it advance? Certainly if the black king goes after the white bishop … it takes 7 moves to go take the white bishop and return to d7 to protect the black bishop, but only 6 moves for the white king to get to f6 and force the bishop to move, grabbing the h pawn on the way. (Carlsen says he didn’t even bother to count.) And if black doesn’t go after the white bishop, white is none the worse off … and in fact he still wins — he plays Bb7, forcing the black bishop to d7 to defend the c pawn, and the white king is able to infiltrate — the black king doesn’t have enough available squares to prevent it. If black plays Bd5 instead of Bd7 then white plays Bc8, winning the f pawn or forcing a bishop trade, which still wins the pawn because the black king can only defend the f pawn from e6, whereas white can attack it from either g5 or g6.
“the h-pawn and the f-pawn, used the time wasted by Nakamura to put the backward c-pawn on a dark square”
Carlsen didn’t have to play c3 at that point … he could have just taken the f-pawn or even advanced e6 first.
Like Albi, I had already seen this game (I watched Agadmator’s video yesterday covering this game). Of course, the right idea is to go after the h4 pawn with 1.Kf2. This would be a better puzzle if the initial position had been on white’s previous move of capturing at a6 allowing black’s b5 trapping the bishop and asked if it was ok to capture at a6 given that b5 traps the white bishop.
The point of going after the h-pawn is that white is also threatening to bring the king up the board to support the advance of the e-pawn, thus winning the bishop back or winning the f-pawn and creating connected passed pawns.
Nakamura didn’t play the stingiest defense at that point- but it was probably the best one to play in hoping for Carlsen to err. Nakamura accepted the challenge, and captured the bishop with the king, and Carlsen won the h-pawn and the f-pawn, used the time wasted by Nakamura to put the backward c-pawn on a dark square, and then used the two connected passed pawns to win the black c-pawn- all without ever winning the black bishop. Just an amazing endgame calculation by the World Champion. The technique he used in a rapid game was absolutely flawless- not a single error. I could see that white wouldn’t lose- the idea was instantly spotted by me while I was watching the video, but I couldn’t be sure that it was winning.
The stingier defense would have been to not go for the trapped white bishop, this would have held the h-pawn and kept the white king from infiltrating, but then the white bishop overloads the black bishop and picks up either the c-pawn or the f-pawn, or forces the black king back to protect the f-pawn and allows the white king into black’s half of the board.
Eh? Black can’t hold the h pawn. (But take black’s f5 pawn off the board and he wins easily.) White follows the same plan regardless … Kf2-g2-h3-h4. The black king can block him with Kg6-h6, but white plays c3, Bb7, and e6, winning the c6 pawn and the game.
Basically, you can sort of solve this puzzle by counting how many moves it takes black to win the trapped bishop and then get back to defend the black bishop against the white king and the passed e-pawn. It is made more complicated by having to consider what happens if black ever has time to play b4 fixing the backward c-pawn on a light square.