17

Arbiting an IM norm tournament last week I came across this position in one of the games. Both White's queen and bishop on a4 were en prise and it was Black's move!

My initial reaction was that I must have missed a touch-move violation which the players had sorted out themselves. Next I wondered why White hadn't just resigned. Finally, when I saw that Black was deep in thought, I wondered why he didn't just whip the queen off immediately.

[Title "Jonah Willow vs Neil Berry 4th EJCOA IM 2022"]
[fen ""]
[Startply "19"]

1. Nc3 Nf6 2. e4 e5 3. Nf3 Nc6 4. Bb5 Nd4 5. Ba4 c6 6. Nxe5 d6 7. Nf3 Bg4 8. d3 Bh5 9. Rg1 b5 10. Nxd4 Bxd1 11. Nxc6 Qc8 12. Bxb5 Nd7 13. Kxd1 Qb7 14. Be3 Be7 15. e5 a6 16. Ba4 Bd8 17. Ne4 O-O 18. Nxd6 Qxb2 19. Bd4 Qa3 20. Bb3 Nxe5 21. Bxe5 Qc5 22. Ba4 Qxf2 23. Bd4 Qf4 24. Ne4 f5 25. Nd2 Bf6 26. Nf3 Bxd4 27. Ncxd4 Qd6 28. Re1 Kh8 29. Rb1 Rab8 30. Rxb8 Rxb8 31. Bb3 g6 32. Re6 Qf4 33. g3 Qg4 34. Kd2 Rd8 35. Kc3 Rc8+ 36. Bc4 Qh3 37. Re7 Qh5 38. Ne5 Rb8 39. Ndf3 f4 40. g4 1-0

As you can see, White went on to win. Overhearing the postmortem, Qc8 was a mistake. Qd7! would have been better and, according to the white player, the computer gives a long line ending with an evaluation of 0.00. Quite remarkable.

What is the history of this stunning queen sacrifice?

Michael West
  • 4,947
  • 12
  • 17
Brian Towers
  • 92,895
  • 10
  • 227
  • 372

1 Answers1

10

Seems to be a relatively unknown sacrifice - the chessbase online database gives me three games after 10.Nxd4, all of which were correspondence and within the last decade. The root game is M.Dos Santos-A.Dutra, correspondence 2014. So it appears the game you mentioned will be the first OTB game featuring this sacrifice.

Somewhat interestingly, none of the moves after 4.Bb5 are rare themselves - they are all either the most common or second most common choice. The second most popular moves are: 4...Nd4, 5...c6, 8...Bh5, 9.Rg1, and 10.Nxd4.

Inertial Ignorance
  • 19,788
  • 22
  • 68
  • Lichess's master games database has 2 games that reached the position after 9 Rg1 b5 (Ivanov-Panitevsky, Dec 2012; Hild-Thoma, Dec 2012). Both continued 10 Nxd4 Bxd1 11 Nxc6 Qd7. – Rosie F Nov 02 '22 at 10:07
  • 1
    @RosieF Yup, those are the two other correspondence games that featured this line after the 2014 one (I think the date is Dec 2017 not 2012 though). As a sidenote, it's odd these games appear in lichess' masters database, since lichess describes it as containing OTB games of 2200+ FIDE players. But they show up for me as well. – Inertial Ignorance Nov 02 '22 at 10:14
  • What about the same sacrifice if Black plays 8...b5 at once, without including Bh5 Rg1 ? Has it already been played ? – Evargalo Nov 02 '22 at 10:58
  • 1
    @Evargalo In the chessbase online database I have two games featuring the immediate 8...b5. One of the games continued with 9.Be3!, which is probably best. The point is that on 9...Nxf3+ 10.gxf3, it's fine that the a4-bishop hangs, since Black's bishop is still on g4 (so the f3-pawn attacks it). None of the games have continued with 9.Nxd4, although White also is a little better here. The reason simply seems to be that it's obviously favourable for White's rook to not have moved to g1. – Inertial Ignorance Nov 02 '22 at 13:05
  • As for the number of games for the more popular 8th moves, I have the following: 8...d5 (132 games), 8...Bh5 (34 games), and 8...Nd7 (23 games). The former two seem to be objectively best as well, as Stockfish gives equality in both. SF at depth 36 recommends 8...Bxf3 as the third best move, though it appears this one is a novelty. – Inertial Ignorance Nov 02 '22 at 13:08
  • 1
    @InertialIgnorance I stand corrected, Dec 2017, not Dec 2012. (I must've misread the final 12 as 2012.) – Rosie F Nov 02 '22 at 16:07