Black to move.

Black has every piece aimed at the Kside.

White threatens 2. Qxg7#.

1. ... Qf3+ is the obvious first move. There is no chance of 1. ... Rf2+ 2. Kxf2 Qxc2+ since the Rook is defended.

1. ... Qxd5+, with the idea of winning the Knight, is not good due to 2. Qe4+, blocking check and checking the Black King, forcing a Queen trade.

1. ... Rg1+ 2. Kxg1 Qb1+ 3. Kg2 Qf1# fails due to 3. Rc1.

However, the ... Qf1# idea might still be valid, just not this particular variant.

Does Black need to prevent White from reaching h3? What will he do if White does reach h3? Short of 1. ... Qf1#, I don't see a way. And then what does Black do?

1. ... Rg1+ and 1. ... Rf2+ are the only non-Queen checks that exist and both are bad.

Does a wild move like 1. ... Qxg3+ lead anywhere? No.

Since White threatens M1 and there is no Queen move that can defend g7, the only Queen move must be a check [1. ... Rg6 stops checkmate but is very passive so unlikely to be the solution].

Also, even if Black can get a Rook on the 2nd rank and the Queen on h5, White can always play Qh4 to block check, protected by the pawn so no ... Rxh2+ tactic will work to win the Queen [edit: I later figured out this only applied if the Rook moved and opened up the White Queen's sight of h4].

1. ... Qf3+ 2. Kh3 Rg6, threatening 3. ... Rxg3+ 4. hxg3 Rh1# fails due to 3. Qe4, pinning the Rook and attacking the Queen.

Remember to keep open the possibility of combining elements of ideas that don't work because that might form a sequence that does work.

1. ... Qf3+ 2. Kh3 Qh5+ 3. Kg2 Qxd5+ still runs into 4. Qe4+.

Aah, but now a previously "bad" idea becomes good: 1. ... Qf3+ 2. Kh3 Qh5+ 3. Kg2 Rg1+!! 4. Kxg1 Qd1+ 5. Kg2 Qf1#.

.

The most important factor for me was not discarding ... Rg1+ just because it didn't work in my original calculation. I saw that it led to ... Qf1# in one variation so, even though it wasn't forced, it was something to plan towards [I don't read titles so I didn't see the f1 hint until after I solved it].

.

I'm sure some will ask why bother with the "useless extra Queen moves": if Black attempts ... Rg1 from the initial position, even assuming White accepts [it's not forced; White can play 2. Kh3], 2. Kxg1 Qb1+ 3. Rc1 and there will be no checkmate on f1. 3. ... Qb6+ 4. Kg2 Qf2+ 5. Kh3 and Black is lost.

Think of 3. Rc1 like forming a wall that keeps the Black Queen outside the perimeter:

Now compare that to the solution: the "useless" Queen moves allowed it to deliver check on d1, not b1, inside the perimeter:

Here's what I think happened [not just in this puzzle but in most]: someone did not successfully solve the puzzle, looked at the solution, and then worked backwards to ask, legitimately, why the shorter solution didn't also work.

The critical assumption they are making is that the opponent will make the same moves with the variation as in the solution. If 1. ... Rg1+ 2. Kxg1 [ignoring that White could play 2. Kh3 also, since there is no Queen on h5 to prevent it] Qb1+, they assume White will play 3. Kg2 just like in the solution. If so, 3. ... Qf1# indeed works.

But White will not make things easy; he'll instead find the optimal 3. Rc1.

This is why there are so many people wondering why the shorter solution doesn't work.

They could also run the analysis tool and the engine would show them.

Read the original post:

2024/06/30 DPA: "Getting Inside The Castle Wall" - Chess.com