US Chess Defeats Irish Chess Union in Match Play

US Chess defeated the Irish Chess Union by the narrowest of margins – 6½-5½ – during an international friendly match held June 27th on chess.com. The two teams featured competitive matchups on every board between players rated between 2000 and 2450 FIDE. Between the twelve total boards, only two matches ended with a 3-1 or greater. All of the remaining 10 boards were either tied 2-2 or won by only a game: 2½-1½.

The two final matches determined the ultimate winner. With Team USA barely ahead of Ireland, 5.5 - 4.5 - boards 8 and 12 came down to the wire. On board 8, William Aramil trailed WFM Trisha Kanyamarala 2-1 and needed a win to tie the match. On board 12, NM Aakaash Meduri was tied with Peter Carroll, and needed a draw or win to secure the match for US Chess.

Image

In a hard-fought battle, Aramil managed to win an exchange thanks to a dangerous passed pawn and was able to convert the not-so-simple ending.

Image

Aakaash Meduri (photo courtesy subject)

Meanwhile, Meduri, with black, found a way to win Carroll’s queen, though the Irish 12th board had plenty of compensation. Meduri played a simplifying variation to reach an endgame where he had an extra knight against two passed pawns and, after many efforts by both sides, the game was ultimately drawn.

A very special thanks goes to Gerry Graham, team captain and match organizer for Ireland. Gerry has proposed a rematch and more details on round two will be coming soon.

The match was streamed on the US Chess Twitch channel by host NTD Chris Bird and commentator and US team captain Pete Karagianis. You can watch the full replay of the match via our YouTube channel:

https://www.youtube.com/watch?v=puJ3tZkeCI0

Three members of Team USA have also submitted annotated games from the event, which you can replay, with notes, below.

[pgn][Event "Live Chess"] [Site "Chess.com"] [Date "2020.07.01"] [Round "?"] [White "Aramil, William"] [Black "Kanyamarala, Trisha"] [Result "1-0"] [ECO "A07"] [WhiteElo "2070"] [BlackElo "2017"] [Annotator "https://lichess.org/@/infiniteadaption"] [PlyCount "115"] [EventDate "2020.??.??"] [WhiteClock "0:01:00"] [BlackClock "0:00:26"] {Over a month ago, I was asked by friend Pete Karagianis of USCF, to see if I would have any interest in playing a match against Ireland. I decided it would be a nice and friendly exhibition match, but I didn't know I was being led down the primrose path. Ireland had us in their crosshairs, and 9 of the 12 matches would be decided by the 4th and final game, some ending in tied matches. Both the US and Ireland matched us up by the order of our FIDE ratings. As expected, the math held true to form. When I was asked to play in this battle of the two nations, I was playing online a bit more often then. My business and teaching have taken over my time and become my top priorities since then. All of a sudden, it was 15 minutes before the match started, and I took a quick peek at my opponent's repertoire before our match began. I was matched up against young WIM, Trish Kanyamarala from Ireland. Her brother Tarun, played IM Eric Rosen on board 2 in the match.} 1. Nf3 {I wanted to play a psychologically non-commital move.} d5 2. g3 Nd7 {My opponent intends to play e5, but this loses some flexibility.} 3. d4 Ngf6 4. Bg2 e6 5. O-O Be7 6. c4 dxc4 (6... O-O) 7. Qc2 Nb6 $6 (7... b5 8. a4 c6 9. axb5 cxb5 10. Ne5 Rb8 ( 10... Nxe5 11. Bxa8) 11. Nc6) 8. a4 a5 9. Nbd2 O-O 10. Nxc4 Nbd5 11. Rd1 { White already stands better. The main advantage of the Catalan set-ups is the light-squared bishop on g2 compared to its counterpart on c8. I have regained my pawn, and Black still has a less active light-squared bishop. Move 2...Nd7 took away some options of variations with an early Bd7. Also, I believe that some of Black's best set-ups in the Catalan involve taking on c4 combined with an early Rb8.} b6 12. Nfe5 Bb7 13. Qb3 {I can't let Black play Nb4 with a tempo. I need to keep the pin on the b7 bishop to make it more difficult and time-consuming, for my opponent to untangle from their setup.} Rb8 14. Bg5 h6 15. Bd2 {I could've placed my bishop here immediately. I knew that provoking h6 would give me some discovered attack possibilities on that pawn when the center of the board is ripped open.} Ne8 $6 {I was happy to see this move. It makes more fundamental sense to trade the knight with Nd7, especially with less space.} 16. e4 Ndf6 {My opponent had a habit of putting herself in time pressure. I played my next move with pure intuition, and partly to force secretsuperstar7 to calculate more given the circumstances.} 17. d5 exd5 18. exd5 Nxd5 {I sensed I should have a commanding lead here. Where was the win? After the game, I was a bit disappointed that I missed a chance at a significant advantage.} 19. Ne3 $2 {I saw some ideas of playing Ng6, with my queen pinning the f7 pawn to the king. That lured me into this dubious various. For example if 19...Nxe3, then 20. Bxe3 and I only saw Bd6 or Qc8 allowing Ng6 to win an exchange.} (19. Bf4 $1 Bg5 (19... Nef6 20. Ne3 c6 21. Nxc6 Bxc6 22. Nxd5 Bxd5 23. Bxd5 {and the rook is loose on b8.}) 20. Bxg5 hxg5 21. Qb5 Nef6 22. Nc6 Qe8 (22... Bxc6 23. Qxc6 Nb4 24. Rxd8 Nxc6 25. Rxf8+ Rxf8 26. Bxc6)) 19... Nxe3 20. Bxe3 Nd6 {I overlooked the strength of this simple and obvious move. I let my dominating position slip away.} 21. Bxb7 Rxb7 22. Qd5 {I felt I had good compensation for my pawn deficit, and Trisha's low time was going to be the deciding factor. This was still a far cry from a few moves prior...} Bf6 23. Nc6 Qd7 (23... Qe8 24. Rab1 {and Black's rook is temporarily stymied. I feel I can always regain my pawn with Bf4 etc.}) 24. Nxa5 {After watching the video of the match, Chris Bird said on the USCF Twitch channel, "It's moves like this that you fall into when you are low on time."} Ra7 (24... bxa5 25. Qxb7 Nxb7 26. Rxd7) 25. Nc6 Raa8 {I sensed that Bf4 was a bit more accurate. My opponent had only 14.6 seconds at this moment, albeit with a crucial 10-second increment.} 26. Ra2 {This was the most practical choice. I knew my young opponent, WIM, Trisha Kanyamarala was feeling the weight of the entire team match and the time pressure in her game. In my mind, Ra2 was the only move considering the circumstances. The advantage of not simplifying in such positions is, it keeps the pressure on the opponent's psyche. For example, if my game ended too soon, that would give Aakaash's opponent more information, and that my alter his decision making over the board.} Qg4 27. f3 Qh3 28. Bf2 Rae8 29. b4 {Create passed pawns! The funny thing is, I knew a would win only tie my individual match at 2-2. But at this exact point, I knew if I won, Aakaash Meduri only needed a draw (tying his individual match) for us to secure the US team victory. I kept a close eye on his match, particularly when we were the last two games. Aakaash is a friend and a roommate at the chess house in Illinois, but unfortunately, I wasn't present at the house for this match. However, even without communicating, I assumed we both knew the scenario. I've been part of high school team chess as a player and coach for 17 years total, and I have always preached EVERYONE should know the team score at all times. It would be irresponsible of me to be unaware of the most current match score. In high school, Aakaash was the top board and captain at Hinsdale Central, and he was part of the final four U of I chess team with Eric Rosen and Michael Auger (another chess roommate). Interestingly enough, Eric finished his last game with a win (tying his individual match), which left only Aakaash and myself as the two remaining boards to decide the fate of the US. As many historians will tell you, the winners of wars write the history books. Ergo, I decided to write this analysis and summary of my game :) I was confident. Actually, I was positive that Aakaash knew the team score and the combinations of results needed with our two matches remaining. If he won outright, we could win the match. Instead, Aakaash made sure that my result had an impact on the team score. Based on our positions at that current time, it looked like I'd need to win and he would need a draw for the team W.} Qe6 30. a5 bxa5 31. bxa5 Nc4 (31... Nb5 {[%cal Gb5c3] The computer has this as equal. With low time, this is such a difficult move to find, and Black must deal with the burden of the passed a-pawn.}) 32. a6 Qxd5 33. Rxd5 Nb6 34. Bxb6 cxb6 35. a7 Rc8 36. Rd6 (36. Nb8 {I knew I had this move in my back pocket, but I wanted to set a trap along the way with 36. Rd6. I'm threatening Rxf6, then Ne7+ and taking the c8 rook. Then a8Q would be the coup de grace.}) 36... Rfe8 37. Kg2 Kh7 38. Nb8 {At this point, I felt that Aakaash's position was going to end in a draw. I told myself I had to win this position no matter what. I delayed playing Nb8 because I wasn't sure if it was objectively winning however.} Rxb8 39. axb8=Q Rxb8 40. Ra6 {I ultimately decided to go with Nb8 due to this variation.} Bd8 {If} (40... b5 41. Rab6 Rxb6 42. Rxb6 { [%cal Gb6b5]}) 41. Ra7 b5 42. Rxf7 Ba5 (42... b4 43. Rdd7 b3 (43... Bf6 44. Rxf6 b3 45. Rff7 b2 46. Rxg7+ Kh8 47. Rh7+ Kg8 48. Rdg7+ Kf8 49. Rh8+ Kxg7 50. Rxb8 {[%cal Gb8b2] Just in time!}) 44. Rxg7+ Kh8 45. Rb7 {This was the idea I spotted in the game to stop the annoying b-pawn.}) 43. Rdd7 Bc3 44. Rb7 Rd8 45. Rxb5 Rd2+ 46. Kh3 Kg6 47. Rc7 Bf6 48. Rc6 Rd3 {My opponent should've kept the rook on the second rank to make it harder for me to untangle.} (48... h5 { [%csl Gb5]} 49. Rbc5 {with the idea of Rc2. It should be winning for White, not without plenty of moves and a decent level of technique.} (49. f4)) 49. Kg4 Rd2 50. h4 Rd4+ 51. f4 Ra4 52. h5+ Kh7 53. Kf5 Ra3 54. g4 Rg3 55. Rc8 Bd4 ( 55... Ra3 {Although I am spoiled for choice here, I saw that this variation was simple and effective.} 56. g5 hxg5 57. fxg5 Rf3+ (57... Bd4 58. g6+ Kh6 59. Rh8#) 58. Kg4 {both the rook and bishop would be hanging.}) 56. Rbb8 g5 57. hxg6+ Kg7 58. Rc7# {InfiniteAdaption won by checkmate 1-0 White wins by checkmate. Aakaash Meduri would go on to draw his final game and his match, sealing a US team victory in our "friendly exhibition" team match against Ireland! Here is a little fun fact that I kept hidden from the USCF. I am also part Irish, and my Mom's last name is Manning. It only seems fitting that I couldn't fully commit to an individual match victory against my adversary. It's as if my subconscious and Irish heritage wouldn't let me pick a side, and a 2-2 tied match against Trisha would be the ONLY proper result. I'm quite convinced I would've been pulled from the lineup by Pete and USCF if I revealed my bloodline!?} 1-0 [/pgn]

[pgn][Event "USA vs. Ireland Friendly Match"] [Site "Chess.com"] [Date "2020.06.27"] [Round "-"] [White "Carroll, Peter"] [Black "Meduri, Aakaash"] [Result "1/2-1/2"] [BlackElo "1925"] [BlackTeam "USA"] [CurrentPosition "8/7b/8/6k1/1K6/8/8/8 w - -"] [ECO "B23"] [ECOUrl "https://www.chess.com/openings/Sicilian-Defense-Closed"] [EndDate "2020.06.27"] [EndTime "20:36:54"] [EventDate "2020.06.27"] [Link "https://www.chess.com/live/game/5066484937"] [StartTime "19:43:37"] [Termination "Game drawn by insufficient material"] [TimeControl "15m+10s"] [Timezone "UTC"] [UTCDate "2020.06.27"] [UTCTime "19:43:37"] [WhiteElo "2009"] [WhiteTeam "Ireland"] 1. e4 c5 { I essayed the 3...Qd6 Scandinavian in Game 2, which ended in a dramatic draw with pawns on the seventh rank for both sides! With our match score tied 1.5-1.5 and the overall match score in flux, I chose the Sicilian so that all three results were on the table.} 2. Nc3 a6 3. g3 b5 4. Bg2 Bb7 5. d3 e6 6. Nh3 Nc6 ( 6... Nf6 { is more common} )7. O-O Nd4 8. Ne2 Ne7 9. c3 Nxe2+ 10. Qxe2 Nc6 11. f4 Be7 12. f5 { A thematic thrust in the Closed Sicilian. White seeks to generate some attacking potential directed at the Black king.} 12... Ne5 13. Be3 { At this point in the match, USA led Ireland 4-3. My opponent no doubt felt some pressure to deliver a victory for Team Ireland. Based on Peter's previous games and the nature of the position, I knew an attack was imminent.} 13... Rc8 14. Qh5 Bf6 15. Ng5 $2 { An ambitious attempt, but White's pieces lack enough space to secure the breakthrough.} ( 15. g4)15... g6 $1 $19 { Disrupting the harmony of White's pieces. The queen is tied to the defence of her knight. Black will gain a material advantage.} 16. Qh4 ( 16. Qh6 Ng4 17. Qh4 Nxe3 18. Rf2 Nxg2 $19 )16... h6 17. fxg6 $5 { Queue fireworks.} 17... hxg5 18. gxf7+ Kf8 { White's queen is trapped. Although my opponent receives some compensation for the queen, Black is much better.} ( 18... Nxf7 19. Qg4 Ne5 20. Qe2 g4 { A slight improvement over the line I chose. The g4 pawn clamps down on White's kingside, and the extra piece is healthy.} )19. Bxg5 Rxh4 20. Bxf6 Qc7 $6 { A poor decision. The queen serves no purpose on c7. I played this move too quickly, and this is when I started to go wrong.} ( 20... Qxf6 $1 { Giving up the queen is a nice simplifying resource. After Black minimizes White's counterplay, the f7 pawn will fall.} 21. Rxf6 Rh7 )( 20... Qb6 21. Bxh4 Nxd3 22. Bg5 { With the idea of Bh6+ looming.} 22... c4+ 23. Kh1 Nf2+ 24. Rxf2 ( 24. Kg1 $4 Nh3+ 25. Kh1 Qg1+ 26. Rxg1 Nf2# )24... Qxf2 25. Rf1 Qxf1+ 26. Bxf1 Bxe4+ 27. Kg1 Kxf7 $19 )21. Bxh4 Nxf7 $2 { As often happens, one bad move follows another. I went into the think tank, going from six minutes on the clock to a minute and a half. The time deficit would haunt me later on. IM Eric Rosen scored a win in his game to tie his match, bringing the overall match score to 5.5-4.5 USA. I was paying close attention to my friend and roommate, NM William Aramil, as he battled against WIM Trisha Kanyamarala. He held a better position and a time advantage, and I expected him to win his game and tie his match. Which meant, I needed at least a draw for USA to clinch the match.} ( 21... Qb6 { The best move offered by Stockfish, and a psychologically difficult move to make after having moved the queen to c7 just before. However, the situation is not so clear.} 22. Rae1 Nxf7 23. Rf6 Kg7 24. Ref1 Nh8 25. R1f4 Rg8 26. Bg5 b4 { Apparently, Black still holds a slight edge.} )22. Rf4 { Despite being down material, White has some serious initiative. My task is to hold down the fort.} 22... Kg8 23. Raf1 Ne5 24. d4 $2 ( 24. Bf6 $1 Ng6 25. Rg4 Kh7 26. Rg5 Rf8 27. Rh5+ Kg8 28. Rg5 Kh7 29. Rh5+ $10 )24... cxd4 25. cxd4 Qb6 26. Bf6 Ng6 27. Rg4 Kh7 28. Rg5 Rf8 29. Rh5+ Kg8 30. Rh6 Rxf6 $1 31. Rxf6 Qxd4+ 32. Rf2 Ne5 33. Bf1 Qxf2+ $2 { An incorrect decision, fueled by time pressure, to simplify into an endgame. The "simplification" actually makes my task of winning far more challenging.} ( { The clearest road to victory. White is toast. } 33... Qxe4 34. Bg2 Qe1+ 35. Bf1 ( 35. Rf1 Qe3+ 36. Rf2 Bxg2 37. Kxg2 Qxh6 )35... Nf3+ 36. Kg2 Nxh2+ 37. Kxh2 Qxf2+ )34. Kxf2 Ng4+ 35. Kf3 Nxh6 36. Kf4 Kg7 37. h4 Nf7 38. e5 { The endgame is objectively winning, but it's not an easy practical conversion.} 38... Bd5 39. a3 Bb3 40. Bg2 Nh8 41. h5 a5 42. Bb7 Nf7 43. Bc8 d6 44. exd6 Nxd6 45. Bd7 Kf6 46. g4 e5+ 47. Ke3 Kg5 48. Kd3 Bd5 { I had some hopes of mating with the bishop and knight. But the Twitch chat did not have as much faith in my mastery of this endgame. I'll have to prove them wrong next match :)} 49. b4 axb4 50. axb4 Bc4+ 51. Ke3 Bd5 52. Kd3 Bf3 53. Ke3 Bg2 54. Be6 Bc6 55. Kd3 Ne8 56. Ke3 Nc7 57. Bf5 Nd5+ 58. Ke4 Nxb4+ $2 $10 { At this point in the game, Will had won, and the match score was 6-5. Therefore, I only needed a draw to secure the victory for our team. I am grateful that Will won and put me in a situation where I did not have to win, especially since my endgame technique was not the greatest here.} ( 58... Ne7+ 59. Kxe5 Nxf5 60. gxf5 Kxh5 61. Ke6 Kg5 62. f6 Be8 63. f7 Bxf7+ 64. Kxf7 Kf5 $19 { The Black king shoulder checks his White counterpart, guaranteeing that his pawn will promote.} )59. Kxe5 Bf3 60. Kd6 Na2 61. Kc5 Be2 62. Bd7 Nc3 63. Kb4 Nd5+ 64. Kc5 Nc3 65. Kb4 Na2+ 66. Ka3 Nc1 67. Kb4 Nd3+ 68. Ka5 b4 69. Be6 Bd1 70. Bc4 b3 { The final attempt to win.} 71. Bxd3 Bc2 72. Bc4 { Only move.} 72... b2 73. Ba2 b1=Q 74. Bxb1 Bxb1 75. Kb4 Kxg4 76. h6 Kg5 77. h7 Bxh7 { A hard-fought draw gives USA the 6.5-5.5 victory!} 1/2-1/2 [/pgn]

[pgn][Event "USA vs Ireland Team Match"] [Site "Chess.com"] [Date "2020.06.27"] [Round "2"] [White "Melaugh, Shane"] [Black "Andersen, Gunnar"] [Result "0-1"] [ECO "E73"] [WhiteElo "2161"] [BlackElo "2158"] [Annotator "17196"] [PlyCount "80"] [EventDate "2020.??.??"] 1. c4 {I was feeling confident since I had won the last game, but I was a bit tired. The amount of danger and uncertainty of this game woke me up pretty quickly.} Nf6 2. Nc3 g6 3. d4 Bg7 4. e4 d6 5. Be2 O-O 6. Be3 e5 7. d5 Na6 8. g4 {This is a really hot line right now, my friend Brian Wall plays it pretty frequently, so I had looked a bit up. I knew that black has to be ready to give some material in this line.} Nc5 9. Bf3 {Looks a bit strange but it's logical to protect both g4 and e4- note that playing f3 for example loosen's white grip over the h5 square} a5 10. b3 {an apparent novelty- I don't personally know that this move is so good.} c6 $1 {Thematic in these structures- black agrees that he's going to lose a pawn now after this move.} 11. g5 Nh5 {This move is basically forced} (11... Ne8 $2 12. dxc6 bxc6 13. Bxc5 $18) (11... Nfd7 12. dxc6 bxc6 13. Qxd6 {White is up a pawn and retains all of his positional advantages- ...Nh5 forces a concession from white.}) 12. Bxh5 gxh5 13. Qxh5 Nd3+ 14. Kd2 Nf4 {This was the point of playing ...Nh5. Now black's knight is on one of its dream squares in the King's Indian, and capturing this knight will only lead to black's bishop on g7 become a monster.} 15. Bxf4 (15. Qf3 f5 $1 16. gxf6 Rxf6 $15 {Black has immense compensation for the pawn in the form of his two bishops, more active pieces, and safer king.}) 15... exf4 16. Nge2 f5 $2 {A big mistake- it's easy in hindsight to understand why. In positions where he does not take on f4, playing ...f5 is a strong idea- but here black's king is just too weak concretely speaking} (16... a4 $1 {Opening up lines on the other side of the board was called for.} 17. bxa4 $2 Qb6 {a double attack now on f2 and b2}) 17. gxf6 Qxf6 18. f3 b5 $2 $18 { Attacking quickly is in the nature of this position- however here it's just too crazy unfortunately, black is lost now. Such is rapid chess!} (18... a4 $16 {was much better, and moreso in the spirit of the position. White's pretty stable and black's king is a bit shaky, so white is better here still.}) 19. Rab1 $4 $11 (19. dxc6 bxc4 20. bxc4 Be6 21. Rab1 Bxc4 22. Nd5 $18 {Black is completely dominated and white's king has no problems, there are still practical chances to hold but black is objectively lost.}) 19... b4 20. Na4 cxd5 {Since white's king is weak and I have two bishops, trading pawns should help me conceptually since the open lines will likely benefit me more than my opponent.} 21. Qxd5+ Be6 22. Qd3 Rab8 23. h4 Bd7 {Kind of an artificial looking move- I spent about 5 minutes here. It was better to play more defensively first. I wanted to play ...d5!? in some position to open up more lines to the white king.} (23... Kh8 24. h5 Qe5 25. h6 Bf6 $11 {The game is even- it's hard for either side to really break in}) 24. h5 Kh8 $2 {I was pretty low on here- only about 4 minutes, so I played a move that looked safe. My opponent had about 10 minutes here} 25. Qd5 $2 (25. h6 $1 {This idea was really strong, I kind of underrated it. Because this isn't possible on the last move, it would've been better to play ...Kh8 then} Bxh6 26. Nb6 $3 Rxb6 27. Rxh6 Qxh6 28. Qd4+ $16) 25... Be6 {I was trying to repeat a bit and then make a decision- I had won the previous game with white, so a draw with black was totally fine. My opponent needed to play for a win here, so he had to go for the gusto. This choice is not objectively bad, the position is even now.} 26. Qd3 (26. Qxd6 Rbd8) 26... Bd7 27. Rh2 {This move comes with a sneaky idea which I missed} (27. h6 $1 {was strong again. I totally overlooked this idea unfortunately}) 27... d5 $4 {I was wanting to play this for several moves, that was the reason I played ...Bd7, but white's idea is more powerful than black's. Of course, from a practical perspective this move was unexpected, so white had to use some time here.} 28. h6 $1 {The point of Rh2. The h7 pawn will now be lost.} Bxh6 29. Rbh1 $4 {A mistake in kind! It is sometimes the most natural looking move that causes the biggest problems. Black is equal now} (29. cxd5 $1 Bb5 30. Qd4 Bg7 31. Qxf6 Bxf6 32. Rbh1 Rf7 33. Nc5 $18 {White's knights are incredible here, and black has no choice but to flounder around. White should easily round up the pawn on f4, and then play f3-f4, e4-e5, and black will get steamrolled.}) 29... dxe4 {As mentioned previously, black needs to try and open up lines to have any hope. This also prevents him from taking on d5 and having a strong positional advantage like was possible on the previous move.} 30. fxe4 (30. Qxd7 $4 Rbd8 $19) (30. Qxe4 Bf5 31. Rxh6 Rbd8+ 32. Ke1 Qg7 $13) 30... Bg7 31. Rxh7+ Kg8 {I had 3 minutes to 6 in this position, I think my opponent felt a lot more confident due to this fact. The position looks very good for white- however, the rooks on the h file actually turn out to be somewhat of a hinderance, moving one of them will yield problems. For example:} 32. Rxg7+ (32. R7h2 f3 33. Nc1 (33. e5 Qg5+ 34. Qe3 Qxe3+ 35. Kxe3 fxe2 $19) (33. Rg1 fxe2 $19 {Black is winning}) 33... Rbd8 { The white king is significantly weaker than his black counterpart, the game is lost. White cannot avoid serious material concessions.}) 32... Qxg7 (32... Kxg7 33. Qxd7+ $18) 33. Rg1 Bg4 {Only move- perhaps he missed this.} 34. Kc2 $4 $19 {And just like that the tides swing in black's favor.} (34. Qd5+ $8 {Only move} Rf7 35. Nd4 $14 {White has a small plus} Kh8 {threat: ...Rd7} 36. Nf3 Rd7 37. Rh1+ Qh7 38. Rxh7+ Kxh7 39. Ne5 Rxd5+ 40. cxd5 Bf5 41. exf5 Rb5 $16 {Might be tenable}) 34... Rbd8 {White's queen is trapped} 35. Qxd8 Rxd8 36. Nxf4 { In the time trouble, I was looking for a simple way to convert. Black has a queen for a knight, so he is objectively winning, but somehow it felt a bit strange with low time and the knights coming in} Kf7 {not objectively best but easy enough.} (36... Qd4 $1 {I actually did consider this and it's apparently the best, it was super messed up though} 37. Rxg4+ Kf7 38. Nd5 {At this point, I thought that my attack is over, but the computer points out:} Rh8 $3 { White's checks are fruitless} 39. Rf4+ Ke8 40. Nc7+ Ke7 41. Nd5+ Kd8 {...Rh2+ is lights out}) 37. Nd5 Qg6 (37... Qd4 $1 {Again very strong} 38. Rxg4 Rh8) 38. Nc5 Qh5 {this move is fine} (38... Rh8 $1) 39. Rf1+ Kg7 40. Rg1 $2 {An unfortunate blunder, but the position is lost anyway.} (40. Kb1 Qh2 {Black is untangling nicely here} 41. Rc1 Qg3 $19 42. a3) 40... Qh2+ 0-1 [/pgn]