In a recently played super-tournament in Shamkir, the game between Veselin Topalov and Ding Liren drew worldwide attention as the Bulgarian lost the ‘elementary drawn’ rook vs knight endgame. If a world-class player can lose this endgame, it means that things may not be as simple in practice as they are in theory. Therefore, I decided to investigate rook vs knight endgame in this issue of Endgame series. After reading this article, you should get a better idea about typical ideas and dangers in this endgame, or at least refresh your memory of this endgame.
Of course, if the king and the knight are together somewhere in, or around, the centre of the board, the draw is pretty trivial as the side with the rook can do little to make progress. Therefore, this type of endgame is not interesting from a theoretical point of view, and indeed most such cases end in a quick draw in practice.
However, there are two winning scenarios that we will investigate:
1. King and knight are poorly coordinated on the last rank/rook’s file;
2. King and knight are disconnected from each other.
The above-mentioned high-profile game featured the first scenario and we will see it shortly. But first, let us cover some basics.
King and knight on the last rank
If the king and the knight are on the last rank (or a-/h- file), this endgame is still drawn in most cases. A typical situation can be seen in the following example: