1. e4 e5 2. Nf3 Nc6 3. Bb5 a6 4. Ba4 Nf6 5. O-O Be7 6. Re1 b5 7. Bb3 O-O 8. c3 d5 The first time I have tried the Marshall gambit in over-the-board chess, after having had some practice on the internet correspondence site chessworld. net 9. exd5 Nxd5 10. d4? The critical test of the Marshall attack is the capture 10.Nxe5 Nxe5 11.Rxe5 c6 12.d4 Bd6 when white will have to work hard to defend his kingside, but equally black must be very aggressive or else his pawn deficit will show. Hill's move for me improves black's prospects, as I think that black now has all the attacking prospects without being a pawn down. 10... exd4 11. Nxd4 Nxd4 12. Qxd4 c6 rather hastily played, I now wonder if 12. .. .Be6 may have been better, but I played the move that is normal in the Marshall to defend this knight. 13. Nd2 White has to solve his development problems, and this is ultimately just too slow. Better would be 13.Bxd5 cxd5 14.Bf4 and then Nd2 and Nf3, rather than this order, which gives black a little time to sort himself out. 13... Bf6 14. Qd3 14.Qc5, exploiting the hole, is a far better choice. Instead white's retreat allows black to finally seize the game, although the line 14.Qc5 Rb8! may give white other problems of black found this ultra-special move idea. The point is that 15.Qxc6 Bb7 16.Qc5 Nf4 throws white on the back foot and may be a total crush since black lines up against the white king. However, White's 15.Nf3 is safe and roughly equal. Another amazing discovery, the first of a couple in this complex game. 14... Be6 Here I was looking at the idea of ...g6 and Bf5, but I wanted first that white play Bc2, threatening mate, and couldn't find a useful move. If white fell for this he would lose the c2-bishop unless the knight moved. But this was too much to ask for from an opponent like T Hill, so I am glad that I played this flexible move. 15. Nf3 not an active square. 15.Ne4 is best. 15... c5 The idea of 15. ...Nxc3 briefly flashed across my mind in my attempt to search for a back-row mate idea, but after 16.Qxd8 Raxd8 I couldn't carry the train of thought far enough to see that white had no defence to some slight gain in material, although Fritz's verdict is about equal after 17.bxc3 Bxc3 18.Bg5 Rc8 19.Rxe6! (the move that I think makes the line viable for black; white may not find this move, while black's moves seem obvious) Bxa1 20.Rd6. The chosen move is fine, since white must address the threat of ...c4 in ways that I think are too passive. 16. c4?? The critical blunder, and the end of the game. 16... Nb4 17. Qe2 Qd3?! I think tha t although black will be able to win this, 17. ...Nd3 which was my alternative idea, followed by 18.Rd1 bxc4 19.Bc2 Qb6, the line I looked at, is better although I assessed it in the end as "difficult" which means that I felt easily capable of messing it up. My chosen move I played to try and avoid the pin on the knight that caused me some concern. 18. a3?? Presumably a desperate attempt to get out of the pressure, but 18.Bg5 at least forces black to do some work. From now on, black plays almost flawlessly, although this isn't the hardest of positions to do so in. 18... Qxb3 19. axb4 Bxc4 20. Qe4 Bd5 this is the only move, I think, that is not the best in the remainder of the game for black. White has now got 21.Nd2 as a saving grace that makes the black win difficult, although it should be achieved in the end. 21. Qg4 Bxf3 22. gxf3 Qxb4 23. Re4 Qb3 24. Ra3 Qd1+ 25. Kg2 Qxc1 26. Rf4 Qxb2 27. Re3 Bd4 28. Rxd4 Qxd4 29. Re4 Qf6 30. Qh3 c4 31. Qg3 c3 32. Rf4 Qxf4 33. Qxf4 c2 0-1 [James]