1. d4 d5 2. c4 e6 3. Nc3 Nf6 4. Bg5 Be7 5. e3 b6 a bit early for this move 6. cxd5 Nxd5 7. Bxe7 Qxe7 8. Nf3 Nd7 9. Bb5 Fritz prefers Nxd5 and is probably right as after ...Nxc3 in the game the c file - along which W usually attacks - is blocked 9... Nxc3 10. bxc3 O-O 11. Bc6 White felt sure there was a win here and spent too long looking for it during the next few moves 11... Rb8 12. Qa4 a5 13. Qb5 Rd8 14. O-O Bb7 15. Bxb7 Rxb7 16. Qc6 Ra7 17. Rfd1 Qd6 18. Qb5 c6 19. Qb3 a4 20. Qb4 Qc7 21. Rab1 b5 22. h3 Fritz prefers 22. c4 22... Raa8 23. c4 Rdb8 24. Rdc1 Qa7 25. cxb5 Fritz reckons 25. Qe7 is just as good 25... Rxb5 is 25... cxb5 any better? It gives B a Q-side majority but b5 is weak 26. Qd6 26. Qe7 is better 26... Rxb1 27. Rxb1 Rc8 28. Ne5 28. Qe7 is definitely better here; it practically forces 28... Qc7 but then 29. Ne5 Nf8 30. Qxc7 Rxc7 31. Rc1 wins the c pawn 28... Nxe5 29. Qxe5 h6 30. Qd6 c5 31. dxc5 31. Rc1 is interesting but complicated 31... Qxc5 32. Qxc5 Rxc5 Notes by PB who threw away a possible win on move 28 but was tired after looking for the win that wasn't there in moves 9 to 13 1/2-1/2 [Peter Brennan]