23. Rxc3 Ra1+ black wins a rook for the bishop.
Giving up the Queen for 3 pieces
I like 21. ... Ng4. Threatening Rxf2+
a) 22. f3?? Ne3+ 23. Kg1 Rxg2#
b) 22. h3 Nxf2
i) 23. Rg1 Nd3 (threatening Rf2#) 24. Rh1 Ng3+ fork.
II) 23. Rh2 Ne4 with Ng3+ coming sealing the king and rook.
"Knights are strong against the queen" is a very general statement, but a pawnless endgame is a very specific situation. Just because knights are good at defending themselves and the king from checkmate doesn't mean they'll fare well in a middle game. I'd guess the queen for 3 pieces trade usually has to be evaluated in the middle game.
knights are actually very strong against the queen....queen vs 2 knights pawnless engame is in general a draw, while queen vs 2 bishops pawnless is in general winning for the queen...queen vs knight and bishop pawnless engame is a draw about half the time :P
Actually, pawnless endgames with queen vs 2 bishops is more often a draw, while queen vs knight and bishop is almost always a loss for the defending side.
Is this idea sound?
Scandanavian Portuguese.
I've used this idea a few times.
It works out well here.
But is it a good idea?
Please advise
--CM