How not to lose a chess endgame

One of the biggest reasons as to why chess is such a difficult game to master is the sheer amount of possible variations that one must calculate in order to find the best move in a given position. Once you get to the endgame however, where the number of possibilities is dramatically cut down, the whole process becomes much more logical and clear-cut.

Yesterday, I played a very bad game of chess against a computer opponent. It was late, I was tired – and it was easy to see after looking at the game again with a fresh mind how obvious moves can be missed when you don’t look at the big picture, but instead fall into the trap of only considering one possibility.

It was a very back-and-forth type of game. I was behind very early on after playing some very poor moves. I managed to get myself ahead in material in the middle-game, only to then squander my advantage. By the time I reached an endgame position, I was playing for the draw.

The turning point

I now want to show you the key moment of the game. I am playing the white pieces, and it is my move. Take a moment to think of what you would play in this situation and why:

As you can see, I’ve managed to bring the game down to even material. Black and white both have 3 pawns. My pawn on the C file (file is the chess name for a column) is a passed-pawn. This means there are no opponent pawns next to it on either side (none on the B or D files) and so it has free passage to the 8th rank (rank = row in chess) where it can be made into a queen. Therefore the black king must watch this pawn closely.

At the same time, black has 2 pawns on the right side of the board, and I only have one. Therefore, black has the threat of advancing the pawns to H4 and G3, trading my H pawn with his G pawn and making a queen.

Moving the A or H pawn here has no value since the A pawn is blocked by the black A pawn, and the H pawn will eventually get stuck or taken by black’s G and H pawns. I have 2 real choices: to defend my C pawn, or to go after his G and H pawns with my king.

My move

I played Kd4 (meaning king to d4), and lost.

At the time, I didn’t give much thought to the position, because I was tired and was being lazy. My idea was to defend my C pawn, but it cannot be defended and I lost time by moving my king away from the black G and H pawns.

Black responded with g5. I now tried to force my C pawn up the board, but realised that it was no use.

Black has now blocked off my C pawn, and crucially, can make a queen on the G file. I have to go after the G and H pawns at the expense of my C pawn.

I played Ke5 here. Black took my C pawn. And now, because I let him move the G pawn up the board, his king is close enough to come back and defend it after I take the H pawn.

Now I am helpless. I cannot take the G pawn with my king as black’s king defends it. He can shepherd his pawn to the queening square. If I play h4, he can move past it and make a queen by playing g3. I cannot move my king to H3 or G3.

The other crucial factor to this position is the relationship of the A pawns. Because it is my move first, no matter how many squares I move my A pawn forward, black can play ‘copy-cat’ and do the same with his A pawn. Eventually, the pawns will be locked, but it will still be my move. Therefore, I will be forced to play either h3 or Kh5.

I resigned here, but actually I could have managed a draw with Kh5. This is the danger of not looking at all possibilities. My king would move between H4 and H5, and his would move between F3 and F4. He could also play g3 if his king was on F3 and mine was on H5. After the trade of pawns, we would both take each other’s A pawn and the game would be drawn.

The better move

Going back to the original position, the better move would be to go after the G and H pawns straight away with Kf5. This would have created a draw much sooner as I would have been able to get back and take both his G and H pawns whilst black took my C pawn.

If black goes for my A pawn, then we will eventually reach the following drawing position, or one similar to this (I’ll leave it to you to figure out why white cannot win here):

If black tries to move his king towards my H pawn, then it could be a draw, or black could actually lose, depending on how well he plays. Therefore, going after the G pawn with the initial move Kf5 would have changed the entire tone of the following moves, and prevented my losing the game so carelessly.

I hope you can see how easy it is to throw away an endgame by not considering all possible variations and by making just one key error at a critical moment. Endgames are often on a knife edge, and by calculating it through, you can make it easy to see what the best moves should be.

Good luck and enjoy your chess!


Many thanks to Josh Waitzkin for inspiring this article.

Leave a comment

Time limit is exhausted. Please reload CAPTCHA.