1. e4 e5 2. Nf3 Nc6 3. d4 exd4 4. Bc4 Bc5 5. c3 Nf6 6. cxd4 Bb4+ 7. Nc3!?

Scott said he had chosen the famous Moeller Variation on the spur of the moment, having enjoyed looking at it with a student a few days ago. But he confessed that he actually did not know more about the line than what he had learned over a decade ago from Andrew Soltis's Winning with the Giuoco Piano and Max Lange Attack. I am convinced he was telling the truth, because I had recently completed a review of the theory myself and so was asking him if he were familiar with various published analysis, none of which he knew.

 

7... Nxe4 8. O-O Bxc3 9. d5!

This is one of my favorite moves in chess theory.

 

9... Ne5!

The current GM recommendation, discussed by Jude Acers in Italian Gambit and Glenn Flear in "Calming the Romantics" in Dangerous Weapons 1.e4 e5, both of which Scott confessed were among the unread volumes in his enormous chess library. He was only vaguely aware, also, that Tim Harding had done a three-part series on the Moeller at ChessCafe. He was familiar with Soltis, and maybe that's all he needed anyway.

 

The line Scott had been reviewing was 9... Bf6 10. Re1 Ne7! (10... O-O?! 11. Rxe4 Ne7 12. d6 cxd6 13. Bg5 Ng6 14. Qd5) 11. Rxe4 d6 12. Bg5 Bxg5 13. Nxg5 h6 14. Qe2! hxg5 15. Re1 Be6! 16. dxe6 f6! 17. Re3! c6 18. Rh3!? Rxh3 19. gxh3 g6! (given as "the refutation" by Mihail Marin in Beating the Open Games) 20. Qd2 (Soltis in Winning with the Giuoco Piano and Max Lange Attack) 20... d5 21. Qc3! Qc7 22. Qxf6 O-O-O 23. Bd3.

 

10. bxc3 Nxc4 11. Qd4

Scott said he had only the vaguest recollection of the theory on 9...Ne5, but this looked right. The rest of the game was completely figured out on his own at the board, which impresses me.

 

11... Ncd6

a) 11... O-O! is theory's current "refutation": 12. Qxe4 (12. Qxc4 Nd6 pretty much transposes) 12... Nd6! (12... b5!? 13. a4! c6 14. axb5!? cxd5 15. Qd4!? (15. Qxd5 Nb6 and Black eventually gets Bb7 with play) 15... Nb6 16. Be3 and White could have improved on 1/2-1/2 Dzindzichashvili, R-Karpov,A, Mazatlan 1988) 13. Qf4 (13. Qd3 b6 14. Ba3 Qf6 15. Qd4 Qxd4 16. Nxd4 Bb7 17. Bxd6 cxd6 18. Nf5 g6 19. Nxd6 Bxd5 20. Rfe1 Be6 21. f4 0-1 Fritz 6-Anand,V/Frankfurt GER 1999 (45)) 13... Ne8 (13... Re8! was Spassky's choice and superior, as given by Glenn Flear in Dangerous Weapons, 1.e4 e5.) 14. d6 Nxd6 15. Ba3 b6 16. Bxd6 cxd6 17. Nd4 Bb7 18. Rfe1 Qc7 19. Re3 Rae8 20. Rg3 Qc5 21. Nf5 Qe5 22. Rxg7+ Kh8 23. Rxh7+ Kxh7 24. Qh6+ Kg8 1/2-1/2 Hardarson,R-Gutierrez Castillo,P/Reykjavik ISL 2006.

 

b) 11... f5!? 12. Qxc4 d6 might also be playable for Black but 13. Nd4 O-O 14. f3 causes a lot of trouble.

c) 11... Ned6?? 12. Qxg7

 

12. Qxg7 Qf6 13. Qxf6 Nxf6

14. Re1+ Kf8

a) 14... Nfe4! is the only way to make a game of it: 15. Nd2 O-O! (15... f5 16. f3 is good for White after 16... O-O 17. fxe4 Nb5 18. c4 Nd4 19. Rb1 b6 20. e5 c5 21. Bb2 Ba6 22. Re3 Rfe8 23. Rf1 b5 24. Nb3 bxc4 25. Rg3+ Kf7 26. Nxd4 cxd4 27. Rxf5+ Ke7 28. Bxd4 Kd8 29. Rf7 Rb8 30. Rgg7 Rb1+ 31. Kf2 Bb5 32. e6 dxe6 33. Rxa7 1-0 Tempel-De Visser/corr NLD jub 1991) 16. Nxe4 Nxe4 17. Rxe4 f5 18. Re7 Rf7 19. Re8+ Rf8 20. Re3 (20. Re1!?) 20... Kf7 (20... d6) 21. d6 Re8 22. Rxe8 Kxe8 23. Bf4 cxd6 24. Re1+ Kf7 25. Bxd6 b6 26. Re7+ Kg6 27. Be5 d5 28. Rg7+ Kh6 29. Rg8 Bb7 30. Bf4+ Kh5 31. Rg5+ Kh4 32. h3 h6 33. Rxf5 1-0 Buerger,H-Matriciani,R/Trier op-RL 1992.

 

b) 14... Kd8? was the line that entertained Scott the most in his calculations: 15. Bg5 Nde8 16. Rxe8+ Kxe8 (16... Rxe8 17. Bxf6+ Re7 18. Re1) 17. Re1+ ( Scott pointed out also 17. Bxf6!? Rg8 18. Re1+ Kf8 19. Ng5 Rg6? 20. Nxh7+ Kg8 21. Re8+ Kxh7 22. Rh8#) 17... Kf8 18. Bh6+ Kg8 19. Re5 Ne4 20. Re8#

 

c) 14... Nde4 15. Nd2 Kf8 16. Nxe4 Nxe4 17. Bh6+ Kg8 18. Rxe4 1-0 Bil,A-Ciemniak,K/Poland 1952.

 

15. Bh6+ Kg8 16. Re5 Nfe4

16... Nde4? 17. Nd2!! d6 18. Nxe4 dxe5 19. Nxf6#

 

17. Re1!

a) 17. Nd2? f6 18. Re7 Nf5

b) 17. Nh4!? followed by f3 is the computer's pick.

 

17... f5 18. Re7 b6 19. Nh4!? Bb7 20. Rg7+ Kf8 21. Rxh7+ Ke8

21... Kg8 22. Rg7+ Kf8 23. Ng6+ Ke8 24. Nxh8

 

22. Rxh8+ Kf7 23. Rh7+! Kf6

 

24. Rxd7?!

Scott will be sad when I show him what the computer found instead, which would have saved him another hour on the clock:

24. Bg7+!! Ke7 (24... Kg5 25. f4+ Kxf4 (25... Kg4 26. Ng6 mates) 26. Nf3 Nf6 27. Rh4+ Ng4 28. Bh6# or 24... Kf7 25. Be5+ Kg8 26. Rg7+ Kf8 27. Ng6+ Ke8 28. f3) 25. Be5+ Kd8 26. f3 and White picks up a piece due to the back rank threats, making for an easy win.

 

24... Bxd5 25. f3 Nc5!

a) 25... Ng5?? 26. Bg7#

b) 25... Nxc3? 26. Bg7+ Kg5 27. Bxc3 Kxh4 28. Rh7+ Kg5 29. f4+ Kg6 30. Rg7+ Kh6 31. Re3 with mate to follow.

 

26. Rxc7 Nf7 27. Be3 Ne6 28. Bd4+! Nxd4 29. cxd4

and White went on to win the ending. A rather spectacular example of how you can still win even slow games with the old theory.

 

1-0