1427-1410, Loss to Benko Gambit (3...b5)

This afternoon's loss featured the Benko Gambit, a 1.d4 defense I have fared well against in the past but to the best of my recollection had never faced before in a rated Rapid game. Itsameeagain, rated 1417, surprised on move 3 and based on his speed of play knew what he was doing early on, while I was in unfamiliar territory and felt the game gradually slip out of my control.

There were a few moments in the endgame where the game review indicates I could have equalized and even seized an advantage, but I failed to recognize them. No excuses, no suspicions of foul play. Like yesterday, my circumstances going into the game were favorably aligned. I got outplayed by an opponent whose opening theory ran deeper than mine and who successfully used a passed pawn in the endgame to paralyze my pieces.

---

physics2112 v. itsameeagain

1. d4 c5
2. d5 Nf6
3. c4 b5
4. b3 g6
5. Bb2 Bg7
6. Nf3 0-0
7. Nd2 d6
8. e4 b4
9. Bxf6 Bxf6

Assigned the white pieces, I opened with 1.d4. Black replied 1...c5, and I knew from experience that 2.d5 was preferable to 2.dxc5. After 2...Nf6, I added defense to d4 with 3.c4, which also allowed my queenside knight to develop without blocking the c-pawn's advance. Then came the move that threw me off balance: 3...b5, attacking my pawn on c4. In considering how to address the threat to my c-pawn, I played the natural-looking 4.b3. The game review says my best move was 4.Bf4, allowing 4...bxc4, keeping my rook on a1 safe, forestalling 4...e5, and preparing a possible 5.Qd2 should Black fianchetto his dark-square bish. A Stockfish analysis at depth 20 suggests other ideas: 4.Nf3, 4.cxb5 and 4.Qc2, respectively.

Having peeled open my queenside, Black began setting up his kingside fianchetto with 4...g6. I continued with the best move, 5.Bb2, to which Black replied 5...Bg7. Considering Black was about to castle, my queenside was exposed to Black's fianchettoed bish, and I had yet to develop my kingside, the game review's evaluation of +0.01 seems generous. After 6.Nf3, Black castled with 6...0-0. The game review indicates my best move here was 7.e3, and I had wanted to advance the e-pawn, but where I wanted to advance it to was e4, which wasn't possible on account of Black's knight. Instead, I continued 7.Nbd2 to prepare my pawn push to e4. Black linked pawns with 7...d6, and I pushed 8.e4. Black advanced on the queenside with 8...b4, and for reasons unknown I traded off my bish for Black's knight with 9.Bxf6 Bxf6. Evaluation: -1.35.

10. Rb1 Nd7
11. Qc2 Qc7
12. Be2 Nb6
13. 0-0 Bd7
14. Rfe1 a5
15. Nf1 a4
16. Bd3 axb3
17. Qxb3 Ra3

As my rook was under attack, I moved it with 10.Rb1, and Black replied 10...Nd7. Already feeling cramped, I developed my queen to one of just three squares she could move to: 11.Qc2. Black copied with 11...Qc7. My bish had only two squares to develop to, and I chose 12.Be2. Black continued setting up on the queenside with 12...Nb6. Prior to playing 13.0-0, I had thoughts of 13.h4, which felt to me like what a GM would play in my position. Game review says pushing on the h-file would have been preferable to castling, but I opted for what seemed the more principled move. Black connected rooks with 13...Bd7, and after my 14.Rfe1 he signaled his queenside attack was imminent with 14...a5. Evaluation: -1.08.

Why I continued 15.Nf1 I cannot for the life of me say. On d2 the knight was adding defense to the pawn on b3, but on f1 it was relegated to a decidedly passive role. Regardless, Black commenced his queenside invasion, attacking my b-pawn with 15...a4. Again I cannot explain what I sought to achieve with 16.Bd3, and the move reduced my time on the clock from 6:54 to 6:25 while opponent still had 8:55. What the game review wanted was 16.Qd1, perhaps with the idea of 17.Qh6 and 18.Ng5. Black captured with 16...axb3, giving his rook a semiopen a-file, and I had a choice of three ways to recapture: 17.Qxb3, 17.axb3 and 17.Rxb3. All were bad, but my choice was the worst, and after 17...Ra3 I was at a disadvantage of -3.39.

18. Qc2 Rfa8
19. Re2 Rxa2
20. Qxa2 Rxa2
21. Rxa2 Na4
22. Re1 Nc3
23. Ra8+ Kg7
24. Nfd2 Qb7
25. Raa1 Nxe4
26. Bxe4 Bxa1
27. Rxa1 b3

My queen was under attack and only had two squares to escape to. After 18.Qc2, Black doubled rooks with 18...Rfa8. I was squirming in my chair trying to find a move, and the best I could come up with was 19.Re2. Material was equal, but I was on the defensive; I figured if I could trade Black's attacking pieces off the board, surely it would be a better deal for me than giving a pawn away for free. Black accepted my terms, and we got a massive exchange on the a2 square: 19...Rxa2 20.Qxa2 Rxa2 21.Rxa2.

With the elimination of my a-pawn, Black's pawn on b4 was officially a passer. Compounding my problems, with 21...Na4 Black was threatening a fork on my rooks. Surprisingly, given how poorly I was playing, I spotted it and safetied a rook with 22.Re1. Black nonetheless attacked with 22...Nc3, but that freed up the a-file for me to give a check with 23.Ra8+. Black effortlessly get out of the way with 23...Kg7, and that was the end of my counterattack. What to do next? Game review says doubling up rooks on the a-file was my best move, and that makes sense: Black's bish on f6 couldn't see a1 on account of the knight on c3, and after 24.Rea1 maybe I could harass Black's queen and pressure Black's position. Instead, though, I played 24.Nfd2 to activate my passive knight. Black attacked my rook with 24...Qb7, and after I dropped back with 25.Raa1 Black captured a pawn with 25...Nxe4, in the process revealing an attack on my rook. I captured Black's knight with 26.Bxe4, Black captured my rook with 26...Bxa1, and after I recaptured with 27.Rxa1 Black pushed 27...b3. Evaluation: 0.00.

28. Rb1 b2
29. Ne1 Ba4
30. Bc2 Bxc2
31. Nxc2 Kf6

I was down to 4:14 on the clock while opponent still had 7:05. Even if the position was objectively equal, the speed with which Black played made me feel like I was getting crushed. Would my pieces be enough to stop Black's pawn from promoting? I attacked b3 with 28.Rb1 -- but this was a mistake. "You ignored an opportunity to eventually win a pawn," says the game review. My best move was to attack b3 with 28.Ra3. Why? Because after attacking from a3, Black wouldn't be able to defend with 28...Ba4; the pawn would advance with 28...b2, and after 29.Rb3 the queen would have to vacate the b-file. Instead, Black pushed with 28...b2, I brought another knight into the action with 29.Ne1, and Black brought in his bish with 29...Ba4. Evaluation: 0.00.

Again, Black was the one doing the attacking, so I offered to trade off Black's active bish for my inactive one. Black obliged, and we got another exchange with 30.Bc2 Bxc2 31.Nxc2. Black's only remaining piece on the board now was his queen; I still had a rook and two knights. My fear at this point was Black's queen attacking my rook from a1 and sacrificing to promote the pawn. In actual fact, Black's queen couldn't access a1 on account of my knight having captured on c2. That would explain 31...Kf6: Black's only hope of making progress was to activate his king. But in so doing, he gave me an advantage of +1.24. My king, after all, was closer. With the right sequence of moves, I could capture Black's passer and use my numerical advantage to start picking off Black's pawns.

32. Na3 Qb4
33. Ne4+ Ke5
34. Re1 Qxe1#

My next move gave away all my advantage and dropped me to -5.03: 32.Na3. Black, to his credit, immediately found 32...Qb4, forking my knights. I played 33.Ne4+ endeavoring to gain a tempo, but Black's mistake on move 31 proved fortuitous for him. With 33...Ke5 his king counterattacked my knight, and that meant I had two undefended knights under attack. I tried 34.Re1, lining up with the king in the hope that with a discovered check after 34...Qxa3 I might find a target and win material back. But Black was sharp. Rather than capturing the knight the queen had forked, he captured my rook with 34...Qxe1#. Checkmate.