Shakmaty Bereolos - The Official Chess Site of Peter Bereolos

9/29/06 - San Francisco 3 :: Tennessee 1

The Tennessee Tempo closed out the first half of the US Chess League season with a 3-1 loss to the defending Western Division champion San Francisco Mechanics. After suffering four straight losses on Board 2, team manager Todd Andrews made the tough decision to sit himself out for a week. This moved myself and Jerry Wheeler up a board and gave Gerald Larson another shot on Board 4. As usual, Ron Burnett held down the top board for us. This was the first week he did not have to face a grandmaster opponent, but IM Friedel proved just as formidable with a powerful exchange sacrifice in the middle game that ran Ron's king all over the board before a final beheading on c4. I thought Friedel deserved the Game of the Week award but the judges only deemed it runner up. On Board 3, something went wrong in Jerry's preparation and he wasn't ready for the line IM Pruess played against the accelerated Dragon although he had played it several times in the past. Despite the ugly look to his position, it was still quite viable for Black. The game got extremely complex and I think things may have still been up for grabs if Jerry had recaptured the piece with 23...gxf2+. Instead, after 23...gxh2+? White won in short order. Gerald got a passed d-pawn in the middle game and drove it home for the win. Unfortunately, that came after we were already down 3 since I played my worst game of the season with White against IM Vincent McCambridge 1. d4 e6 2. c4 Bb4+ 3. Nc3 f5 4. g4!? a home-brewed anti-Dutch weapon that I have had much success with 4... fxg4 5. Bf4 5. e4 also deserves attention 5... Nf6 6. h3 g3 7. Bxg3 Ne4 8. Qd3 d5

9. Be5?! It is better to continue developing and put pressure on e4 with 9. Bg2 I was handicapped by the fact that I had beaten Mark Geist in the 2000 Kings Island Open with the text move. 9... O-O This looks like a simple improvement on Geist's 9... Qh4 after the text White has some trouble getting his king to safety because of the sensitivity of f2. There was some speculation by my teammates that McCambridge had seen my game versus Geist and prepared this, but that wasn't apparent from the time he was spending. Plus, the text move is fairly obvious, especially in a fast time control game. The Mechanics don't mention anything about it in their blog, either. In fact, he spent so much time on his 4th move, I was wondering if he was waiting for a takeback request since g4 looks like it could be a mouse slip. 10. Nf3 Nc6 11. Rh2 an ugly square for the rook, but I didn't want to admit my mistake and retreat the bishop, especially since Black is ready to blast the center open with ...e5 11... Nxe5 12. dxe5 c6 13. Nd2 Nxd2?! The engine points out the simple win 13... Nxc3 14. bxc3 Qg5 since 15. cxb4 Qxe5 forks the White rooks. 14. Qxd2 b5 this was a bit of a surprise, I expected him to increase the pressure on c3 and e5 with 14... Qa5 15. cxb5 cxb5 16. O-O-O Bd7 17. Kb1 Be8 18. e3 Rc8 19. Rc1 Bxc3 20. Rxc3 Rxc3 21. Qxc3? Black is much better after 21. bxc3 Bg6+ 22. Kb2 Qg5 23. f4 Qf5 but the text is just a forced loss. I missed that in addition to back rank mates, a rook landing on c1 also picks up Bf1. 21... Bg6+ 22. Ka1 22. Bd3 Qg5 is killing 22... Qb8 23. a3 Rc8 24. Qe1 Rc2 25. Rh1 b4 this wins, but 25... Qxe5 26. Qb4 Rc1+ 27. Ka2 Bb1+ 28. Kb3 Qc7 seems more obvious. 26. Qxb4 Qxb4 27. axb4 Be4 28. f3 Bxf3?! giving White some unwarrented chances 28... Rc1+ 29. Ka2 Bxf3 30. Rg1 Be2 is clean 29. Bd3 Rd2 30. Rf1 Bg2 31. Rg1 Rxd3 32. Rxg2 Rxe3 33. Rc2 Rxe5 34. Rc8+ Kf7 35. Rc7+ Kf6 36. Rxa7 d4 37. Kb1 White comes up short in the pawn ending after 37. Rd7 Rd5 38. Rxd5 exd5 39. Kb1 g5 40. Kc2 h5 41. Kd3 g4 42. hxg4 h4 (42... hxg4 also wins) 43. b5 h3 44. b6 h2 45. b7 h1=Q 46. b8=Q Qd1# analogous to the finale in my game versus Matthew Marsh in the 2003 Knoxville City Championship 37... Rb5 38. Ra4 e5 39. Kc2 e4 40. Kb3 e3 41. Kc4 Rb8 42. Kxd4 e2 Avoiding 42... Rxb4+? 43. Rxb4 e2 44. Rb6+ with a draw since there is nowhere to hide. Black can't go to the fifth rank because of Rb5+-Re5 and can't cross to the c-file since he must stop the White rook from coming to the e-file. 43. Ra1 Rxb4+ 44. Kc3 Rh4 45. Re1 Rxh3+ 46. Kd4 Rh4+ 47. Kd5?! 47. Kd3 puts up more resistance since Black has to defend his e-pawn from the slightly awkward h2 square, but this shouldn't alter the result. 47... Rh5+ 48. Kc4 Re5 49. b4 Ke6 50. b5 Kd6 51. Kd4 Re8 52. Kc4 Kc7 53. Kd3 Kb6 54. Kd2 g5 [0:1]

9/22/06 - Seattle 3.5 :: Tennessee 0.5

It was another frustrating night for the Tennessee Tempo, Wednesday in the US Chess League. We had a rematch with our opening night opponents, the Seattle Sluggers. Since they swept us on opening night, Seattle had amassed the best record in the league, while we have headed in the other direction. Our team spirit has remained high and we were ready to reverse the trend. We knew that we had missed a lot of opportunities in that first match as well has having to deal with technical difficulties with our internet connection. We started the same 4 we had in the first match: Burnett, Andrews, myself, and Wheeler while they substituted FM Mikhailuk on Board 2.

It seemed that karma was on our side early when Seattle's internet connection dropped almost as soon as play began. After they finally got back up and running they were all penalized 5 minutes and we ran up a significant time advantage on every board (unknown to us, they were having to relay all their moves through a single computer, much like we had to do with the telephone on opening night). My game was a pretty uneventful draw as we will see below. Todd's game looked like a total mess with an extra piece versus a slew of pawns. Once again the pawns ended up being stronger. Still, down 0.5-1.5 things didn't look too bad for us. Jerry had an extra pawn and Ron's position looked tenable although slightly worse. Both of their opponents were short on time. Things began to go off the track a bit when Jerry allowed a dangerous piece sacrifice instead of taking a clean two pieces for a rook. He used up his time edge, but his opponent misplayed the attack and it looked like Jerry just needed to unwind his position. Meanwhile, GM Serper gave up all of his advantage versus Ron, nearly losing on time and swapping into a drawn rook ending, which he then proceeded to misplay and actually give Black some winning chances. Then, in the span of a few seconds, disaster struck. Jerry locked up and couldn't produce a move and lost on time! Ron offered an exchange of rooks that gave Serper a winning king and pawn ending. From winning or at worst drawing the match, suddenly we had lost by 3 points for the second straight week.

I had black this week in a rematch with FM John Readey. 1. Nf3 Nf6 2. c4 g6 3. Nc3 Bg7 4. e4 d6 5. d4 O-O 6. Be2 e5 7. O-O Na6 8. Re1 c6 9. Bf1 exd4 Although I had lost to both Goletiani and Gurevich with this move in the Spring North American FIDE Invitational earlier this year, I was quite satisfied with my positions out of the opening in those games and did not hesitate to repeat this variation. 10. Nxd4 Ng4 11. h3 Qb6 12. hxg4 Qxd4 13. g5 Qxd1 14. Rxd1 Be5 15. Be3 Nc5 16. f3 a5 17. Rac1 a4 18. Rc2 f6 19. gxf6 Rxf6 20. Ne2 Ne6 21. f4 This was unexpected, 21. Rcd2 looks normal 21... Nxf4 22. Nxf4 Bxf4 23. Bxf4 Rxf4 24. Rxd6 Bg4 24... Rxe4 25. Rd8+ Kg7 26. Rf2 followed by Rff8 will cost Black material. 25. Bd3 Kf7 26. c5 Ke7 27. Rd4 Be6 28. Bc4 Ra5 29. Bxe6 Kxe6 30. Rb4 30. Rd6+ looks like a surer road to equality

30... Rb5 30... Rf7 followed by ...Ke5 may have given Black a small chance to press. 31. Rxb5 31. Rxa4 Ke5 32. Ra8 leads to another equal position 31...cxb5 32. c6 bxc6 33. Rxc6+ Kd7 [½:½]

9/17/06 - Tennessee Open

The Tennessee Open was held over Labor Day weekend at Opryland. It was a somewhat disappointing tournament for me. Although I scored an undefeated 4-1 good enough to tie for second place, a couple of early draws meant that I was never really in contention for the title.

My problems started in Round 1 with Black versus Paul Smith. I wasted a whole tempo in the opening when I thought I was gaining a whole tempo and found myself in some difficulties after 26...Bxc6

27. Qd3 I was more afraid of 27. Qa5 hitting c5, b6, and a6. I thought I had found a good reply with a two pawn sacrifice 27...Rb8 28. Qxa6 c4 29. bxc4 Qc5 and it seems that Black holds everything together. For example,30. Ne1 Ra8 31. Nd3 Qd4 32. Nb4 Qc5; 30. Qa7 Rb7 31. Qa6 Nd7 with the idea Nb8; 30. Rb1 Nd7 31. Rxb8 Nxb8 32. Qc8 Qxc4 33. Qxb8 Qxc2 34. h3 Bxe4. The fly in the ointment seems to be if White does not accept the second pawn, but simply plays 29. Qa5 and the queen is ready to escape and Blacks compensation for the pawn is not clear. He still had a bit of pressure after the text, but I carefully made my way to a draw.

I bounced back in round 2 with White against Ron Seaney. He was under some pressure because of my 2 bishops and more space plus somewhat awkward placement of his minor pieces, when he missed a tactic after 14. Bg5

14...h6? 14...fxe4 is called for, but White has a nice plus 15. Bxe7 Qxe7 16. exf5 when recapturing on f5 loses a piece to 17. d6. I converted the pawn plus without further difficulties.

The third round was key for me with Black against Joshua Suich. I missed several chances in a complex ending after 29. axb4

29... Ne4 This looks natural, but I could have gained a clear advantage with the paradoxical 29... Na4+! going to the edge of the board and blocking the a-file 30. Kb3 (30. Kc2 Nb6) 30... g3 31. fxg3 Re3+ 30. Rd7 Nxf2!? A double-edge winning attempt, more solid was 30...c5 31. Rf1 g3 32. Rxb7 Rad8 33. Nxc6 Rd3 The idea I totally missed was 33... Rd1 34. Rxd1 Nxd1+ followed by Ne3 and Nxg2 giving Black some winning chances. 34. Re7 I had totally overlooked this move 34... Ra8 While not as good as on the previous move, Black could still try to press with 34... Rxe7 35. Nxe7 Rd1 36. Rxd1 Nxd1+ 35. Ra7 Shunning any attempt to try to win with 35. Ne5, 35. c5, or 35. b5, all of which offer White the better chances. 35... Re8 The last try for 35... Rxa7 36. Nxa7 Rd1 37. Rxd1 Nxd1+ which is probably a bit better than when the rooks are traded on e7 since the knight is worse on a7. 36. Re7 Ra8 37. Ra7 [½:½]

The next morning I had White against my clubmate Matthew Marsh. In a complicated game, I won the exchange but it looked like he might have some compensation after 29...Rxe8

I came to the conclusion that holding the d-pawn was the key to my position, worth much more than the exchange, 30. Ng4 Rf8 Taking on g4 lets me throw in Rxe8 getting his queen off of d6, and taking on f3 lets me keep d6 well covered. Of course, taking d6 immediately is hit by Nf6+, so he could try to first bump the knight with 30... h5, but I was ready to sacrifice with 31. Qh6 which appears to give a decisive attack, for example 31...hxg4 (31... Nxf3+ 32. gxf3 Bf7 33. Rxe8+ Bxe8 34. Ne5 Qf6 35. d7) 32. Qxg6+ Kh8 33. Qh6+ Kg8 34. Rxd4 cxd4 35. Rxe6. The other choice seems to be to develop with 30...Nd7, but this takes pressure off of d6 and White has the nice move 31. Qe3 when 31...Nc2 32. Qc3 Nd4 (32...Nxe1 Nh6+ +-) 33. Rxd4 cxd4 34. Qxd4; 31...Bf7 32. Nh6+ Kf8 33. Nxf7; or 31...Nf6 32. Nxf6+ Qxf6 33. Bd5 are all winning 31. Rxe6 Rxf3 31...Nxe6 32. Qd5 Re8 33. Re1 +-32. Qe1 Rf8 33. Rxd4 Simpler is 33. Nh6+ Kg7 34. Re7+ Kxh6 35. Qh4#, but the d4 sacrifice is thematic 33... cxd4 34. Qe5 Qh4 35. Re7 Rf7 36. Qe6 [1:0]

In the final round I had Black against Michael Hornick. After 12. Rb1

Instead of the routine 12...a5, I played for a small trap, which he fell headlong into. 12... Nc5 13. b4? Ncxe4 14. Nxe4 Nxe4 netting a pawn since 15. Qxe4 Bf5 16. Bg5 f6 costs the exchange as well. I ground out a win in a long ending. On the top board, Todd Andrews beat USCF Executive Director Bill Hall. Josh Suich's run came to an end on Board 2 against top seed Ron Burnett who had earlier conceded draws to Andrews and Tony Cao, so Todd took the title outright for the second straight year.

9/16/06 - Tennessee 0.5 :: Miami 3.5

The Tennessee Tempo had another rough night Wednesday in the US Chess League against the Miami Sharks. Before the match, I think we were guardedly optimistic since we had gotten on the scoreboard the previous week and the Sharks were 0-2 coming in. However, they showed why they are the defending division champs.

We had a slight lineup tweak this week with Gerald Larson in on 4th board for Jerry Wheeler. Unfortunately, Gerald dropped a pawn to an opening trap on move 4! Although he gamely fought on for a long time, the other three of us knew early that we had our work cut out for us. The other three games were extremely complicated. Ron went for a highly theoretical line in the Open Ruy Lopez versus GM Becerra. I thought he was at least going to hold a draw, but apparently didn't get out of Becerra's preparation until it was too late. On Board 2, I wasn't really sure what was going on in Todd's game versus IM Moreno Roman, but the tactics did not end up in his favor.

I had White versus NM Miguel Espino on Board 3, and had another adventurous game. 1. d4 g6 2. c4 Bg7 3. e4 Nc6!? A move order trick that I've employed on myself on occassion to avoid the unpleasant lines arising from 3... d6 4. Nc3 Nc6 5. d5 4. d5 Nd4 5. Be3 5. Ne2 c5 6. Nxd4 Bxd4 is probably a better try for an advantage 5... c5 6. Nc3 The point of Black's move order is that 6. dxc6 is met by 6...dxc6 and Black maintains control of d4. 6... e5 7. Bd3 d6 8. Qd2 Ne7 I didn't really like this move, the knight can turn out to be on an awkward here, but I didn't find a precise way to exploit it. I think 8...f5 is better, intending to develop the knight to the more natural f6 square 9. O-O-O O-O 10. Bh6 I wasn't satisfied with this move, I think a better line was 10. h4 f5 11. h5 f4 12. Bxd4 when Black will be left with a bad bishop 10... f5 11. Bxg7 Kxg7 12. exf5 Bxf5 I was going to meet 12... gxf5 with 13. f4 which is roughly level 13. f3 13. Bxf5 gxf5 14. f4 is again roughly level. He could try 13...Nexf5 giving me e4, but keeping his pieces fluid 13... a6 14. h4 I could take control of e4 with 14. g4 Bxd3 15. Qxd3 but Black has ample counterplay after 15... b5 since White has weakened f3 and f4. 14... Bxd3 15. Qxd3 b5 16. Nh3 bxc4 17. Qxc4 h6 stopping the idea of Ng5-e6 although it is not clear if that really accomplishes a whole lot. Now I found a way to unbalance the game and seize the initiative 18. f4 The other choice was 18. h5 g5 19. Nf2 heading to e4, but it wasn't clear to me what I would do after I got there. 18... exf4 Accepting the challenge. The engine prefers 18... Nef5 19. Rxd4 cxd4 20. Qxd4+ Kg8 What really sold me on the exchange sacrifice was that Black has to play this move, leaving g6 a little tender since he can't go for 20... Kh7 21. Ng5+ Kg8 (21... hxg5 22. hxg5+ wins) 22. Ne6 Nf5 23. Qxf4 21. Nxf4 Nf5 22. Qd3 Qf6

23. Ne4 It was a tough decision between this move and 23. Ne6 which was Ron's preference. I didn't like lines like 23... Rf7 24. g4 Ne7 25. g5 hxg5 26. hxg5 Qf3, but our brief analysis indicated that White may be able to go for h5 at some point. Instead, I decided to bring my other knight into play and keep Nf4 trained on g6. 23... Qd4 24. Qxd4 Nxd4 25. Nxg6 Rfc8+ 25... Rf7 26. Nxd6 Rf6 27. Ne7+ Kf8 28. Ndc8 and the d-pawn is very dangerous. 26. Kb1White has a serious alternative with 26. Kd2 activating the king since 26... Rc2+ 27. Kd3 Rxg2 28. Nf4 is no good for Black 26... Kg7 27. h5 It was better to bring the knight back into play with 27. Nf4 which has the added benefit of covering e2 27... Re8 28. Nxd6 Re2 28... Red8 29. Nc4 29. Rd1 again it was perhaps better to bring the knight back with 29. Nf4 since the game-like continuation 29... Rf2 30. g3 Rb8 31. Nc4 Nb5 does not carry the threat of Nc3+ with a fork on d1. However, if you play a move like 27. h5, it is hard to admit two moves later that you should have moved Nf4 after all. 29... Rb8 30. Nc4 30. Rxd4 Rbxb2+ only Black can win although it doesn't look like he has more than perpetual check. 30... Nb5 31. Ka1 Todd suggested 31. Rd3 but I was wary of the trick 31... Nd6 when it appears Black holds 31... Rc8 32. Nge5 Re8

33. Nf3 It appears that White wins with the ultra-cool 33. a3 simply making luft and ignoring Black's "threat" 33... R8xe5 34. Nxe5 Rxe5 35. d6 Re8 36. a4 Na7 37. d7 Rd8 38. Rd6 and Black is dominated, but I didn't have nearly enough time to find a fantastic variation like that 33... Rc2 34. Nce5?! I rejected 34. Nfd2 because of 34... Rc8 but again the d-pawn proves stronger than the piece after 35. a4 Nd4 36. d6 R2xc4 37. Nxc4 Rxc4 38. d7 Rxa4+ 39. Kb1 Nc6 40. d8=Q Nxd8 41. Rxd8 and White should win. 34... Nd6 35. g4 Re2 36. Nc6 R8e3 37. Ncd4?! The engine suggests 37. Nfd4 Rd2 38. Rf1 as equal. Now, I got in a spot of difficulties. 37... Rf2 38. Nh4? 38. Ne6+ Kf6 39. Nfd4 would at least temporarily shield the d-pawn. 38... Rd3 39. Ne6+ Kf6 40. Rb1 Rxd5 41. Nc7 Rdd2?! a slip that lets White keep fighting. The a-pawn should be preserved with 41... Ra5 42. Nxa6 Kg5 on 42... Nc4 White staves off immediate disaster with 43. Nc5 43. Ng6 Kxg4 44. Nc5 Kxh5 45. Ne5 45. Rg1!? is a trappy move, for example 45... Rxb2?! 46. Nd3 Rxa2+ 47. Kb1 and White wins back the exchange with a drawn position 45... Kg5 46. a4 Kf5 47. Ned3 Rf3 48. Rh1 Nf7 49. Rh5+ Kg6 50. Rd5 I think White is back to equality after this move thanks to the passed a-pawn. Now he just hands me the game, but I make a ton of time pressure mistakes and fail to convert it.

50... Rf5? 51. Nf4+ perhaps he only considered 51. Ne5+? Rxe5 52. Rxd2 Rxc5 51... Rxf4 52. Rxd2 Rc4 53. Rd5 h5 54. a5 Rc1+ 55. Ka2 Re1 56. Nd3?! I was getting seriously low on time here, but thought I was going to quickly pick up his h-pawn and then win on the queenside at my leisure, but I should have gotten the pawn rolling immediately with 56. a6 56... Re4 57. b4 57. b3 is better so I don't have to spend an additional tempo to cover b4, but I thought I still had everything worked out. 57... h4 58. Kb3 58. a6 h3 59. a7 h2 60. a8=Q h1=Q was not the type of position I wanted to play with only a minute on the clock. 58... h3 59. Nf2 h2 A bit of a rude awakening. But even the line I looked at 59... Re3+ 60. Rd3 Rxd3+ 61. Nxd3 h2 62. Nf2 "winning", looks like Black holds without much problem after 62... Nd6 63. a6 Nc8 64. Kc4 Kf5 65. b5 Kf4 66. Kc5 Kf3 67. Nh1 Kg2 68. b6 Nxb6 69. Kxb6 Kxh1 70. a7 Kg1 60. Rd1 Rf4 61. Nh1 Ne5 62. b5? With very little time left, I rejected 62. a6 because of 62... Nc6 completely missing the winning 63. Rd6+ 62... Rf3+ 63. Ka4 there is no way to win now 63. Kb4 Nd3+ 64. Rxd3 (64. Kc4? Nb2+) 64... Rxd3 65. a6 Rd1 66. a7 Rxh1 67. a8=Q Rb1+ 68. Kc5 h1=Q 69. Qxh1 Rxh1 70. b6 = 63. Kc2 Nc4 = 63. Ka2 Nc4 64. a6 Ra3+ = 63... Rf4+ 64. Kb3 Rf3+ [½:½]

9/8/06 - Dallas 2 :: Tennessee 2

The Tennessee Tempo got on the scoreboard in Week 2 of the US Chess League against the Dallas Destiny. We avoided the internet problems that plagued us in Week 1. This time all the lightning was on the boards. I think every board ended in an unlikely result. On Board 1, Ron Burnett was trying rather unsuccessfully to construct a fortress with two knights against a queen against GM Panchanathan when his opponent blundered his queen to give us our first points. On Board 4, Jerry Wheeler was up an exchange versus Nelson Lopez, but allowed a perpetual check. My game was also drawn on Board 3 against Keaton Kiewra, but not without adventure as we will see below. That left Todd Andrews on Board 2, who had an extra piece for a bunch of pawns against Andrei Zaremba. Unfortunately, once the time got low, the pawns proved more powerful, so we ended up split. At least getting some points made the drive back to Knoxville more pleasant.

I had Black this week versus NM Keaton Kiewra, who I had faced OTB in 2002 at the US Masters and Chicago Open. 1. e4 e5 2. Nf3 Nc6 3. Bb5 a6 4. Ba4 Nf6 5. d4 He played 5. Bxc6 against me previously, but I thought he might try the text as he had played it in some more recent games 5...exd4 6. O-O Be7 7. Re1 b5 8. Bb3 d6 9. Bd5 Bb7 10. Nxd4 Nxd5 11. Nxc6 Bxc6 12. exd5 Bb7 I wasn't present when my teammates later came up with the suggestion 12... Bd7?! 13. Qe2 O-O?? 14. Qxe7 Re8 so I can't tell you how much beer had flowed at that point, but the -+ evaluation is quickly reversed by 15. Qxd8. Black needs to stay focused on the d-pawn so that Qe2 is met by Bxd5 then Be6 closing the e-file 13. a4

13...O-O?! I had prepared the line 13... Qd7 14. axb5 axb5 15. Rxa8+ Bxa8 16. Nc3 O-O from a 1991 game between Romanishin and Adams, but screwed up the move order at the board. The text sacrifices a pawn, but Black has some compensation because of the bishop pair. 14. axb5 axb5 15. Rxa8 Bxa8 16. Qe2 Bf6 A couple of other USCL players were familar with this position as in the 2003 National Chess Congress Ray Kaufman played 16... Bg5 against Lev Milman here. I think my move makes more sense since the bishop pair is Black's trump in the position. 17. Qxb5 Qb8 Blatny beat Ivanchuk in the 1985 World Junior with 17... c6 18. Qc4 giving back the pawn to keep some pressure, rather than trying to hold it and defend after 18. Qxb8 Rxb8 19. c4 Kf8 20. Re2 Rb4 when the 2 bishops should give Black close to full compensation 18... Bxb2 19. Bxb2 Qxb2 20. Nc3 20. Qxc7 Bxd5 21. Qxd6 is completely equal after 21... Be4 20... Qxc2 21. Qxc7 Qb3 22. Rb1 Qc2 23. Rd1 Qb3 The start of a bad plan. The queen was fine on c2, stopping the rook lift to the third rank. I should have just played 23... Re8 24. h3 h6 with a roughly level game. The Ba8 can be activated simply with Bb7, an idea I missed during the game. 24. h3 Qb8 25. Qa5 Qd8 26. Qb4 Qc7 27. Rd3 Rc8 28. Rg3 f5 29. Kh2 Qc5 30. Qf4 Bxd5 I was starting to get short of time here and took the opportunity to finally activate the bishop. Much more solid is 30... Rf8 when Black shouldn't have any further worries. 31. Qxf5 Bc4 31... Ba8 32. Qe6+ Kh8 33. Nb5 looks like it will also end up costing at least the d-pawn. 32. Qd7 Bf7? Forced was 32... g6 33. Ne4 Qc7 34. Qxd6 Qxd6 35. Nxd6 with a similar ending as in the game 33. Ne4 Qc7

34. Qg4? missing the crushing 34. Nf6+ Kf8 (34... Kh8 35. Rxg7! Kxg7 (35... Qxd7 36. Rxh7#) 36. Ne8+) 35. Nxh7+ Kg8 36. Qg4 34... Bg6 35. Nxd6?! He played this move very quickly. Instead, 35. Rc3 Qxc3 36. Nxc3 Rxc3 37. Qe6+ Bf7 38. Qxd6 and it is unlikely Black can construct a fortress35... Rd8 36. Qc4+ This was a real relief, I think Black now has realistic drawing chances. 36... Qxc4 37. Nxc4 Bf7 38. Ne5 Rd5 39. f4 Rd2 40. Rc3 Be6 41. Rd3 Rxd3 42. Nxd3 Kf7 43. Kg3 I think the last chance to press was 43. g4 now Black achieves a pawn trade and has no further difficulties. 43... Kf6 44. Nf2 Kf5 45. Ng4g5 46. fxg5 Kxg5 47. Ne5 Kf5 48. Nc6 Kg5 49. Ne5 Bd5 50. Nf3+ Kf5 51. Nh2 Kg5 52. Nf1 Bc4 53. Ne3 Bd3 54. h4+ Kh5 55. Kh3 Kg6 56. g4 Kf6 57. Kg3 Ke5 58. Ng2 Ke4 59. Nf4 Bc2 60. Ne6 Ke5 61. Ng5 Bd3 62. Nh3 Bc2 63. h5 Bd1 64. Kh4 Kf6 65. Nf2 Bc2 66. Nh3 Bd1 67. Ng5 Bc2 68. Nf3 Be4 69. g5+ Ke6 70. Nh2 Bd3 71. Nf3 Be4 72. Nd2 Bd3 73. Nb3 Ke5 74. Nd2 Ke6 75. Nf3 I wasn't quite sure how to claim 3-times repetition on ICC, so I sent a draw offer, then made my move 75...Be4 and the system instantly called it a draw.1/2-1/2