The Winning Endgame that Can't Be Won

Introduction

Endgames are often seen as the culmination of strategy, where meticulous planning throughout the game finally pays its dividends and delivers the victor a hard-earned point. However, sometimes a chess player can do everything right and still fail to win. These decidedly deserved victories are foiled by endgame oddities. In this blog I'll share three such endgames which defy logic and result in unexpected draws.

#1 - Two Knights vs Lone King

It's tragic, to me, that two knights can't synergize and control enough squares to herd the enemy king and deliver checkmate. Compare that with the two bishops, which dominate the whole board with the power of friendship. Even a bishop and knight can put aside their differences to muster the strength to force checkmate. Knights, though, are much worse in open positions than bishops, and an empty position is pretty open.

It's clear from the above image that we need to buff the knight to make it balanced. I would suggest removing the knight's L-shaped movement and instead allowing it to move orthogonally. We might need to increase its piece value to 5 after this change. If you guys have any other suggestions, leave them in the comments below.

Anyway, forcing a checkmate with two knights against a lone king isn't possible, so you should easily get the half-point playing as the defending side. A few tips:

• Keep your king away from the board's corners and, if possible, edges. It's hard to get checkmated in the center of the board.
• Be patient and stay focused until the 50-move rule takes effect.
• Don't overthink it! It's a pretty simple endgame to draw.

#2 - Queen vs Pawn

The second endgame was even more shocking to me when I first learned of it. A mighty queen fails to defeat a mere pawn? Yes, in some positions that's exactly what happens. The drawn situations usually involve a pawn on the second rank (we're going to assume the defending side is Black) so it can continuously threaten to promote and bring about a fair fight.

First, let's go through the technique to win with a queen against a second-rank pawn:

This concept is simple to master. Even if you've never seen it before, you can probably still win in a real game after clicking through the above diagram. The vast majority of queen vs pawn endgames are winnable, so it would be a great shame to fumble them as the stronger side. I'd recommend practicing theoretical endgame technique with Stockfish or some chess-playing friends, if you have them.

There are some exceptions in which draws occur. If the defending side's pawn is a rook pawn, then the position is almost always drawn because keeping the king in front of the pawn results in many stalemates. Here's a funnier scenario which is a forced draw:

Some general principles about whether this endgame is a draw or not:

• The defending side's king needs to be near its pawn to defend it, or the enemy queen can just win the pawn and play out a trivial win.
• Unless the pawn is on the second last rank, it's generally a simple win for the attacking side as they have an extra tempo.
• If the pawn is on the second last rank and can promote before the attacking side's queen can deliver a check, the endgame will be drawn. Some rare scenarios involve the stronger queen's king blocking its own queen from delivering check.
• If the attacking side's queen can stand in front of the pawn, stopping it from promoting, the endgame will be won as the stronger side can just march over their king and capture the pawn.
• If the stronger side's king is close to the opponent's king and pawn, they can usually win as the king will assist with capturing the pawn or delivering checkmate.
• If the stronger side's king is far from the opponent's king and pawn, but the pawn is on the b-, d-, e-, or g-files, the position is a forced win. See the first example.
• If the stronger side's king is far from the opponent's king and pawn, but the pawn is on the a-, c-, f-, or h-files, the position will be drawn, usually by stalemate. See the second example.

#3 - Bishop and Wrong Rook Pawn

This one is the most striking to me as a "should-be-winning-but-isn't" theoretical endgame. The attacking side has a full bishop (those guys were supposed to be better than knights in endgames) and a pawn which can advance up the board and perhaps promote. Yet the endgame is still drawn!

I'm oddly obsessed with this endgame. Whenever I reach an ending with bishops on the board, I ponder on the possible ways a bishop and wrong rook pawn situation could come about. "How can I use that idea to save myself from this atrocious position?" I assume I'm still completely awestruck that this endgame is drawn, years after learning about it. Here's a real-game scenario in which this knowledge would actually come in handy:

Unfortunately, despite my continued vigilance, I haven't had the chance to find this concept in a real game. The same can't be said for the legendary Mark Taimanov, who could've achieved the wrong rook pawn ending and drawn against Fischer himself in a Candidates match in 1971. He failed to recognize it during the game and ended up losing the game and, famously, the match, 6-0.

The above game illustrates that everyone can falter when faced with these endgame subtleties, even strong Grandmasters and World Champions. It also highlights that true endgame mastery involves not just studying theoretical endgames but also being able to identify them in real games under pressure.

Conclusion

These theoretical endgames can be a very useful weapon in your arsenal as something to strive for when you're trying to draw and something to avoid when you're trying to win. Write in the comments below whether you'd enjoy reading a Part 2 of this blog as I do have some more fitting endgames in mind. Thanks for reading, see you in the next blog.