[Event "Friendswood CC May Round Robin"]
[Site "Webster, TX"]
[Date "1997.05.16"]
[Round "3"]
[White "Lane, Alex"]
[Black "Jouett, Ray"]
[Result "0-1"]
[WhiteUSCF "1729"]
[BlackUSCF "1829"]
[Opening "Blackmar-Diemer Gambit"]
[Variation "Teichmann Variation"]
[TimeControl "40/90;G/30"]

1. d4 Nf6 2. f3 d5 3. e4 dxe4 4. Nc3 exf3 5. Nxf3 Bg4 6. h3
Bh5 7. g4 Bg6 8. Bc4 e6 9. Bg5 Bd6 10. Ne5 Nbd7 11. Qe2
  {To this point the text follows the game Hall-Conquest,
   England 1978.}
11...O-O 12. O-O-O c6 13. h4
  {Hall-Conquest continued 11...c6 12.h4 without the
   intervening castling moves. I like White's position,
   and I think h4 is even stronger against the castled Black
   King position.}
13...h6 14. Nxg6! hxg5 15. hxg5! Nh7 16. Rxh7?!
  {White's last three moves have been played rather quickly,
   and without realizing Black's imminent zwischenzug. Probably
   16. Qe3 would be better, as it protects the pawn on g5,
   and the heavy artillery around Black's king isn't going
   anywhere.}
16...Qxg5+ 17. Kb1 Qxg6 18. Bd3! Qf6
  {Black took 13 minutes for this move. In the end,
   18...Qxh7 was probably better.}
19. Rdh1 g6 20. Ne4?
  {I give this move a '?' because it was played thinking
   the Black Queen could not leave the a1-h8 diagonal,
   which is not true.}
   (20. Qe3)
20...Qxd4 21. Nxd6
  {Here, starting with 21.Rh8, White can get a Queen and Knight
   for two Rooks (and the pawn that just got eaten), but the
   resulting game is not going to be easy to win.}
   (21. Rh8+ Qxh8 22. Rxh8+ Kxh8 23. Nxd6)
21...Qxd6 22. Rh8+ Kg7 23. R1h7+ Kf6 24. Qf3+ Kg5?
  {24...Ke7 is better. 25.Bxg6 is met with 25...Ne5!}
25. Qh3??
   For a long time, I believed this move rated a "!". Then 
   Victor Prupis  dropped me an e-mail 
   pointing out that here, 25. Rh5+! forces mate in two.
25...Rxh8??
  {

   Position after 25...Rxh8??.
   Here, White still has a forced mate in 8, which I did not see at
   the board, but should have.}
26. Qh4+ Kf4 27. Rxf7+
   (27. g5+! Ke5
      (27...Kf3 28. Rxf7+ Ke3
         (28...Kg2 29. Be4+ Kg1 30. Qf2#)
      29. Qf2#)
      (27...Ke3 28. Qe4+ Kf2 29. Rxf7+ Kg3 30. Rf3+ Kh2
      31. Qe2+ Kg1 32. Rf1#)
   28. Qe4#)
27...Ke5 28. Qg5+ Kd4 29. Rf4+
  {This move wins Queen for Rook, but much better is 29. c3!!,
   which forces mate in three.}
   (29. c3+!! Kxd3 30. Rf3+ Kc4
      (30...Ke4 31. Re3#)
      (30...Ke2 31. Qe3+ Kd1 32. Rf1#)
   31. b3#)
29...Qxf4 30. Qxf4+ Kc5
  {Whoopee. The Black Queen is dead, the King still lives, and
   the mating net has evaporated! My plan now is fairly
   straightforward. Kill the pawn on g6 with the Bishop, and
   eventually convert the passed pawn. In the meantime, my
   Queen harasses the heck out of Black on the open board.
   The only problem with this plan is: it's going to take
   a LOT of effort to convert my advantage (if, indeed, I have
   one) into a win.
   Unfortunately, although I have plenty of time on my clock,
   I have a work assignment that is to start in a little bit
   (a Public Affairs support gig for the STS-84 mission),
   so I am not taking the time I should, and I know it.
   The game result, to some extent, reflects my work-related
   "time trouble." Not an excuse, just an observation.}
31. b4+
  {Conveniently buying a tempo and providing the White King with
   "luft."}
31...Kb6 32. Qe3+ Kc7 33. Bxg6
  {33. Qxe6 is probably better.}
33...e5 34. Bf5 Rae8 35. Qxa7?
  {This may have been a needless excursion.}
35...Nb6!
  {Helps isolate the White Queen.}
36. Qa5
  {This or Qa3 is pretty much forced. The threat is ...Ra8}
36...e4 37. c4? e3 38. c5??
  {Blindness. I did not see Kb8 as a resource for Black.}
38...e2 39. cxb6+ Kd8 40. Qa8+??
  {Loses convincingly (although I think I am lost already).
   40. b5 does not help.}
   (40. b5 e1=Q+ 41. Qxe1 Rxe1+ 42. Kb2 c5)
40...Ke7 41. Qxb7+ Kf8 42. Qxc6 e1=Q+ 43. Kb2 Qxb4+ 0-1
  (In the final analysis, I did not work the plan I set forth
   after move 30, thereby allowing Black to consolidate his
   position and eventually permitting him to Queen a pawn.}