2025/01/24 DPA: "Navigation By Triangulation"
White to move:
.
It took me a few seconds to recognize this was my puzzle.
White's up in material but his f Knight is hanging and his c Knight is pinned and doubly-attacked. Defending with 1. Kb1 loses to 1. ... Qxf5+.
And playing the immediate 1. Qxh8+, while tempting, leads to 1. ... Kc7 and the c Knight will fall.
The Solver must find 1. Rc2, which not only defends the Knight a 2nd time, it attacks the Queen and X-Ray's the c file [namely, c7 and c8] so 2. Qxh8+ is a potential checkmate threat.
1. ... Rxc1+ solves nothing: 2. Rxc1 and Black faces the same problem as before, only now he's down more material. Note that White must choose 2. Rxc1, which maintains the attack on the Queen and the threat of 3. Qxh8+; if 2. Qxc1, those threats are gone.
So Black opts for 1. ... Qxf5 which is the puzzle solution.
.
Yes, Black played sub-optimally. I'm aware of that. Puzzle curator Dane Mattson is aware of that. Stockfish is aware of that. The optimal idea is to sac the Queen for the Rook [as happened in the actual game] but it wouldn't have been much of a puzzle then, would it? Choosing a sub-optimal allowed the motifs to be better showcased.
Also, due to the initial material edge for White, Black is lost no matter what he does. So 1. ... Qxf5 was not a blunder; it was merely sub-optimal, which is acceptable for puzzles.
https://www.chess.com/blog/EnPassantFork/no-the-puzzle-is-not-wrong
Here is the game upon which the puzzle was based:
https://www.chess.com/game/live/127198742393
I'm actually Black in the game: I came up with two puzzles - one as White [this puzzle] and one as Black [possibly to be published in the nebulous future].
In the game, I captured a pawn on c4, leaving myself vulnerable to the Rook move and subsequent Qh8+. I had to sac my Queen for the Rook and while I was only down 1 point of material, I was hugely down on the clock: 1:44 vs 15:23.
But I played on and he counter-blundered by allowing a Royal Fork, showing how "it ain't over 'til it's over."