[Event "4NCL"] [White "Turner, Lewis"] [Black "Franco, Gaston"] [Site "?"] [Round "?"] [Annotator "Lewis Turner"] [Result "1-0"] [Date "2024.11.24"] [ECO "B34"] [WhiteElo "1984"] [BlackElo "2070"] [PlyCount "61"] [GameId "2115846287810518"] [EventDate "2024.??.??"] {[%evp 0,61,25,16,58,58,65,35,17,31,79,71,49,0,20,9,15,-16,25,30,-4,-9,-35,-5,-132,-40,176,187,188,195,184,201,192,188,189,192,194,199,243,240,255,248,251,249,251,258,279,286,296,297,308,335,339,337,338,312,312,329,396,255,251,255,303,266]} 1. e4 {This game was a real roller-coaster} c5 2. Nf3 Nc6 3. d4 cxd4 4. Nxd4 g6 {Not a line I enjoy facing as my results in the Maroczy are quite poor. I decided upon a fresh approach.} 5. Nc3 Bg7 6. Nb3 $5 Nf6 7. Be3 O-O 8. Qd2 $5 $146 {This was my idea, quite rare to castle long in the accelerated dragon!} a5 $1 9. a4 d5 $1 {Very energetic play by my opponent after some 15 minutes of thought. I was quite scared here.} 10. exd5 Nb4 11. O-O-O $2 {I spent about 10 minutes here and could not see a way to refute this though it did feel dangerous. The engine is very critical of this decision, claiming a winning advantage for black after the correct sequence.} Bf5 $1 12. Nd4 {Before going for this I had calculated:} Nfxd5 $2 {This came as a surprise, I became concerned here of having missed something. Indeed I believe there is only one acceptable line for white. I had to calculate through the technique of elimination.} (12... Be4 13. f3 Bxd5 14. h4 $11) (12... Ne4 13. Nxe4 Bxe4 14. f3 Bxd5 (14... Bxc2 $1 {The engine quickly points out the flaw in my calculation. This is brutal but not obvious.} 15. Nxc2 Rc8 16. Bb5 Rxc2+ 17. Qxc2 Nxc2 18. Kxc2 Qd6 19. Kb1 Qe5 $19 {Engines are brutal, to me this was not obvious and my opponent didn't even consider it. Beware of automatic moves such as Bxd5, this is a common mistake even when stronger players than me calculate.}) 15. c3 Nc6 $13 {I struggled to assess this position}) 13. Nxf5 $1 {Here I was very happy with what I came up with} (13. Nxd5 $2 {The engine confirms I am correct!} Qxd5 $1 {[%CAl Gd5a2]} 14. Nxf5 Qa2 $19 {I get mated so can't allow this.} (14... Bxb2+ $1 {This is even stronger, mate in 18 according to Stockfish})) 13... Nxc3 14. bxc3 Na2+ 15. Kb2 $1 Nxc3 (15... Bxc3+ 16. Qxc3 $1 Nxc3 17. Nh6+ $1 (17. Rxd8 $1 Nxa4+ 18. Kb3 Rfxd8 19. Nxe7+ Kf8 20. Bd3 Kxe7 21. Kxa4 {This is the engine's preference though to a human I think Nh6+ is cleaner}) 17... Kg7 18. Rxd8 $18 {With two extra pieces at least as the knight is also in some trouble} Nxa4+ 19. Kb3 Rfxd8) 16. Nh6+ $3 {The move I believe my opponent underestimated} Bxh6 (16... Kh8 17. Nxf7+ $1 {In the post-mortem he said this was the move he missed} Kg8 (17... Rxf7 18. Qxd8+ Rxd8 19. Rxd8+ Rf8 20. Rxf8+ Bxf8 21. Kxc3 $18) 18. Nxd8 Nxd1+ 19. Kc1 Raxd8 20. Qxd1 Bb2+ 21. Kxb2 Rxd1 22. Bc4+ $1 $18 {Another nice line where white wins all the loose black pieces}) 17. Qxd8 (17. Qxc3 Qxd1 (17... Bg7 $1 18. Qxg7+ Kxg7 19. Rxd8 Rfxd8 20. Be2 $16 {I concluded white is better here but didn't think this was as clean as what I played The engine is more optimistic, awarding this +4}) 18. Bxh6 f6 19. Bc4+ {This was the line I saw first, however I then saw}) 17... Raxd8 18. Rxd8 Rxd8 19. Kxc3 Bg7+ 20. Kb3 Rd1 21. c3 $1 {The final important move, white is completely winning. With time ticking down my opponent played on for a few more moves out of pure inertia.} Rb1+ 22. Ka2 Rd1 23. Kb2 h5 24. g3 Be5 25. Bg2 Rd3 26. Rc1 Rd7 27. Bf3 Kg7 28. Rd1 Rc7 29. Bd4 Bxd4 30. Rxd4 e6 31. Be2 {Black resigned The engine largely agrees that I converted in the best way} 1-0