2025/05/03 DPA: "Location, Location, Location!"

.

There are dark square weaknesses for Black but he has a dark-squared Bishop for defense; e7 and d8 are heavily defended currently.

If we could get the Black Queen to a6 or b5, Nc7+ would be a Royal Fork.

Moving the d Knight opens the d file for the Queen, however every square starting at the 4th rank is guarded.

1. b4 attacks the Queen. 

• It can't retreat along the diagonal as every square is covered.
• Moving to a6 or b5 leads to the Royal Fork
• Moving to a3 allows Nc7#

Aah, but 1. ... cxb4  2. cxb4 allows 2. ... Qxd5, supported by the Bishop.

So we'd like to get rid of the Bishop:  1. Bxf5 Bxf5  2. b4.  Hmm, but now 2. ... cxb4  3. cxb4 allows 3. ... Qc5, guarding c7.

How about 1. Bxf5 Bxf5  2. Qa4, threatening the Queen and if the sac is accepted, the Queen no longer controls c7 so 3. Nc7#.

There is no way to defend the Queen to allow it to stay on a5 except 2. ... b6, but then that interferes with the Queen's control of c7.

Black could decline but then White simply wins a piece.

Are there any good alternatives?

Wait:  there's an error in my 1. Bxf5 Bxf5  2. b4 line:  after 2. ... cxb4  3. cxb4, I came up with 3. ... Qc5 as a way to continue defense of c7:  but that's not possible since White's b pawn controls c5 as well as a5.

But both 2. b4 and 2. Qa4 can't be correct so something must be wrong with one [or both] of them.

Oh, I see:  2. b4 cxb4  3. cxb4 Bxb4+ [almost fell victim to RBBS <Remote Bishop Blindness Syndrome>].

So 2. Qa4 must be right.

Yup, that was it.

.

The key was seeing the potential checkmate in Nc7 and working backwards to figure out how to get the Queen away from guarding c7.

• 1. Bxf5 opened up the d1-a4 diagonal for the Queen
• 2. Qa4 attacks the Queen:  if she captures, she allows 3. Nc7#; if she moves to a6 or b5, still 3. Nc7#.  And if she does something else, White wins a Queen

I initially thought the purpose of 1. Bxf5 was to remove the Black Bishop's attack of the Knight which made the subsequent pawn push effective but it fails due to 3. ... Bxb4+.

Like yesterday, the opponent played sub-optimally but that's OK:  the point was to show why 1. Bxf5 was the only winning move.  Which losing response was chosen [either loss of a piece or checkmate] is a matter of aesthetics, not analysis.

Some will balk at saccing their Queen:  they haven't yet learned that material is irrelevant compared to checkmate.  You could take away all of White's pieces except the d Knight and the h Bishop:  as long as Black's Queen does not guard c7, White gets the checkmate.

Nice one by @JohanVA, the prolific puzzle principal.