Loss by resignation: 1317-1305

After months of rating stagnation, the past several weeks saw me reach new highs in Rapid and Daily on Chess.com and Rapid and Blitz on Lichess. However, as was the case the last time I pushed above 1300 Rapid on Chess.com, my wins felt fluky and undeserved. None was flukier than on June 13th, when for no apparent reason my opponent up and resigned on move 11. That boosted my rating to an all-time high of 1317.

I spent most of yesterday, a Saturday, watching chess videos on YouTube, solving tactics puzzles, and playing unrated games on Lichess. When I finally resolved to play a rated Rapid game in the evening, Chess.com matched me with a 1326-rated player named sahil0420 and assigned me the black pieces. We transposed into the French Exchange early and I held my own going into the middlegame, but sahil0420 played faster and the clock pressured me into a few dubious moves that gave away the game.

---

1. e4 e6
2. Nf3 d5
3. exd5 exd5
4. d4 Bd6
5. Nc3 c6
6. Bd3 Ne7
7. 0-0 0-0

After the first seven moves, we arrived at a position familiar to me from SuperChessGURU's video "Spicing up the Exchange French." In the French Defense, it is universally agreed that White's best minor piece is his light-squared bishop. Black's idea in placing his knight on e7 is to control f5 in order to later attack d3, which is typically where White develops his light-squared bishop. Moreover, the knight on e7 rather than f6 creates an imbalanced position, which presumably White sought to avoid in opting for the Exchange Variation.

---

8. Re1 Re8
9. Bg5 f6
10. Bh4 Nf5
11. Bxf5 Bxf5

After 8.Re1, the theory a-la SuperChessGURU calls for 8...Bf5 -- which is also Stockfish's second recommendation, below 8...Na6 and above 8...Re8. White has a tricky response to 8...Bf5, which is 9.Rxe7, and if Black is not careful White can win material. I couldn't remember how the line went, and in lieu of 8...Bf5 I played Stockfish's third choice: 8...Re8.

White proceeded with 9.Bg5, still sticking to SuperChessGURU's script; and I replied 9...f6, both kicking White's bish and controlling g5 so White can't later advance his knight there. After 10.Bh4 I should have continued 10...Bf5 but instead played the top Stockfish move: 10...Nf5, threatening to capture White's dark-squared bishop. Stockfish indicates White should ignore the threat and improve his position elsewhere on the board, but White took with 11.Bxf5 and I recaptured with 11...Bxf5.

---

12. Rxe8+ Qxe8
13. Qd2 Qg6
14. Re1 Nd7
15. Bg3 Bb4

A common idea for Black in the French Exchange is to develop the queen to c7, behind the dark-squared bishop targeting White's pawn on h2. The point of White's next move, 12.Rxe8+, which prompted 12...Qxe8, may have been to draw my queen away from c7. Alternatively, White was uncomfortable with the rook tension on the e-file and sought to resolve it by trading off the rooks. I was worried White would win control of the e-file after, for example, 13.Qe1 Qxe1 14.Rxe1; and that appeared to still be his plan with 13.Qd2. What I saw at this moment, however, was that White's pawn on c2 was weak, and I therefore set out to attack it with 13...Qg6.

Another common idea for Black in the French Exchange is developing the queenside knight to a6. 13...Na6 was the top Stockfish recommendation for move 13, but at this stage in my chess journey I still fail to see the benefit of moving a knight to the edge of the board. White proceeded to take control of the e-file with 14.Re1, and Stockfish again recommended 14...Na6. I agreed it was time to move the knight and let the rook out of its corner, but the move I played was 14...Nd7. White played the top engine move, attacking my dark-squared bishop with 15.Bg6. Stockfish says I should have taken with 15...Bxg6, but the idea of unprotecting the e7 square and inviting a move like 16.Re7 seemed dangerous. I therefore played 15...Bb4, moving my bishop to safety and at the same time pinning White's knight to the queen.

---

16. Re2 Re8
17. Rxe8+ Qxe8
18. a3 Ba5
19. b4 Bb6
20. Na4 Qg6
21. Nc5 Bxc5

Black's best continuation was to fork my queen and bishop with 16.Nh4, which even if not winning material immediately would at least have given him the initiative and complicated the game, forcing me to think defensively, slowing me down, and increasing the chance of my making poor decisions. He either failed to see the fork or was more worried about my battery targeting the pawn on c2; he played 16.Re2. My best move was to pin White's kingside knight to his rook with 16...Bg4; I played 16...Re8, aiming to get White's rook off the second rank. Black should have gone for the knight-fork again, but instead we repeated move 12: 17.Rxe8 Qxe8.

The next three moves involved White chasing my bishop around the queenside. With 18.a3 White asked my bishop if it wanted to trade for the knight on c3, and with 18...Ba5 my bishop said it needed more time to think. With 19.b4 White told my bishop to find another diagnoal to patrol, and with 19...Bb6 my bishop complied and relocated to the a7-g1 diagonal aiming at White's king. With 20.Na4 White again challenged my bishop, but with 20...Qg6 I indicated I was more concerned with White's pawn on c2 than with his cat-and-mouse game with my bish. White, finally tired of harassing my bish, played 21.Nc5, turning his attention to my weak pawn on b7. Since the pawn couldn't be defended, I captured White's knight with 21...Bxc5.

---

22. dxc5 Bxc5
23. Qf4 Bb3
24. h3 Qb1+

White recaptured my bish on the next move, but evidently he did so with the wrong pawn; 22.bxc5 would have left White with three pawn islands and an iso pawn on the a-file, but the engine still prefers it to 22.dxc5 (-0.69/-0.05), which turned my d-pawn into a passer. I made good on my threat against White's underdefended pawn and played 22...Bxc2, my best move. White's best move was to blockade my passed pawn with 23.Nd4, but the move he played, 23.Qf4, ended up impacting the game more than my passer. It was at this point that I noticed I could checkmate White on the back rank if only I could get my queen to an undefended square there. Since my bishop obstructed my queen's path to b1, I moved the bish to b3 with 23...Bb3. The engine much preferred 23...Bd1, taking aim at White's knight.

Needless to say -- since the game did not end on move 24 -- White saw what I was plotting and played 24.h3, creating an escape square for his king. Stockfish says my best move in this position was 24...Ne5, but I played 24...Qb1+, which turned out to be disastrous: the eval bar, which mostly hovered around zero until my check, leapt to +3.11. What happened? One, even though White was triple-guarding e5, the engine must have seen the sequence of captures there after 24...Ne5 as ultimately favoring the chances of my passed d-pawn advancing to d1. Two, not only was my check with the queen pointless, since on h2 White's king was safe from further attacks, but relocating the queen to White's back rank distanced her from the action on the board and decreased her activity. Three, the White queen's invasion on c7 was imminent and I had failed to appreciate the havoc she could wreak on my pawns once there.

---

25. Kh2 Qd3
26. Qc7 Qf5
27. Qxd7 Bd1
28. Qxa7 Bxf3
29. gxf3 Kf7
30. Qa6 Ne5

After 25.Kh2 the engine evaluated all my moves as losing, but it still preferred 25...Ne5 to at least trade off a minor piece. I played 25...Qd3, improving White's advantage to +4.43. White's queen invaded with 26.Qc7, as expected; and after my 26...Qf5 defending the knight, White's advantage swelled to +10.3 because he had a devastating fork with 27.Nd4. White missed the fork and captured a free pawn with 27.Qxb7, but after 27...Bd1 White's advantage grew to +12.3.

White proceeded to capture another free pawn with 28.Qxa7, making his pawn on a3 a passer, but that reduced White's advantage to +2.15. I went for the minor piece exchange with 28...Bxf3; and after 29.gxf3, messing up the pawn structure around White's king, White's advantage shrunk to +1.89. Then I played 29...Kf7, probably hoping to get my king to g6, and White's advantage shot up to +5.70. Targeting my undefended pawn on d5, White continued 30.Qa6 -- but that allowed me to improve my knight with 30...Ne5, defending c6, targeting White's undefended pawn on f3, and reducing the scope of White's bishop. Just like that, and despite White being up a pawn, the eval bar was back at zero.

---

31. Bxe5 Qxe5+
32. Kg2 Qg5+
33. Kf1 Qc1+
34. Ke2 Qc2+
35. Ke3 Qc3+
36. Qd3 Qe5+
37. Kd2 Qg5+
38. Ke1 Qh4
39. Qa6 Qxh3

After 31.Bxe5, I had a choice of recapturing with queen or pawn. 31...fxe5 would have reinforced my precious d-pawn on its voyage across enemy lines, but I and Stockfish preferred 31...Qxe5+ because it came with check and a crucial tempo. After 32.Kg2, I found the only move that wasn't winning for White: 32...Qg5+. White played 33.Kf1, and I checked a third time with 33...Qc1+. White played 34.Ke2, and I checked again with 34...Qc2+.

After 35.Ke3 Qc3+, White played the move shown at the top of this report: 36.Qd3. The eval bar went from 0.0 to -0.57 (or -1.30, depending on Stockfish's mood), but I had to find 36...Qe1+ -- the point being after 37.Qe2 Qc1+ I win White's a-pawn. Instead I played 36...Qe5+, and the eval bar reset to 0.0. White played 37.Kd2, I checked with 37...Qg5+ instead of 37...Qb2+ (+1.85/0.0), and after 38.Ke1 I played 38...Qh4 instead of checking with 38...Qc1+ (+1.47/0.0). White used the tempo I gifted him to swoop back into my territory with 39.Qa6; and here I had to position my queen on the h6-c1 diagonal, but instead I captured with 39...Qxh3.

---

40. Qb7+ Kg6
41. Qxc6 Qxf3
42. Qe8+ Kf5
43. Qd7+ Kg6
44. c6 Qxa3
45. Qxd5 Qxb4+
46. Kf1 Qb1+
47. Kg2 Qb4

Having taken my foot off the gas, it was now White's turn to start checking: 40.Qb7+. My king ran for cover behind the f-pawn with 40...Kg6, but his respite was short-lived. After we each took a pawn off the board with 41.Qxc6 Qxf3, White had three connected passers and continued 42.Qe8+. Again using the f-pawn as cover, I replied 42...Kf5. White checked again, 43.Qd7+, and I repeated move 40: 43...Kg6.

The eval bar was at +3.69, but after White's 44.c6, which looked terrifying, I squandered another opportunity to equalize the position: with 44...Qe4+ I could have forced a draw by repetition. Instead I captured with 44...Qxa3 and the eval bar soared to +11.2. Again we each took a pawn off the board with 45.Qxd5 Qxb4+, but the eval bar and clock spelled my doom. White got out of check with 46.Kf1 and I gave another check with 46...Qb1+, but after 47.Kg2 White's king was safe behind his f-pawn and my 47...Qb4 neither threatened White's passed c-pawn nor put his king under any pressure.

---

48. Qd7 Qe4+
49. Kg1 Qb1+
50. Kh2 Qb8+
51. c7 Qb4
52. Qe8+ Kh6
53. c8=Q Qh4+

White continued 48.Qd7, presumably to prevent 48...Qg4, but that allowed 48...Qe4+ and I could yet again force a draw. White retreated his king with 49.Kg1 and all I had to do was find 49...Qe1+, but I played 49...Qb1+ and White regained the advantage. After 50.Kh2 I played 50...Qb8+, but that allowed White to advance his c-pawn closer to promotion and at the same time attack my queen: 51.c7.

With 46 seconds on my clock and 5:51 on White's, my case was hopeless. White had given me numerous opportunities to avoid a loss, but I failed to capitalize on them. With 51...Qb4 I moved my queen to a safe square from which I could at least give another check on the next move and delay the inevitable, but it was White who gave the next check: 52.Qe8+. I replied 52...Kh6, White promoted with 53.c8=Q, and I gave a meaningless check with 53...Qh4+.

Against two queens and with only half a minute on the clock, there was little point in playing on. The game continued:

54.Qh3 Qxh3+
55.Kxh3 1-0

And I resigned. My rating dropped from an all-time high of 1317 to 1305.

---

Key takeaways

The Exchange Variation of the French Defense is notorious for the drawish positions it leads to. Richard James of chessKIDS academy calls it "popular with wimps who want to play it safe." It's therefore not a coincidence that opportunities kept presenting themselves throughout the game to force a draw. Ultimately, though, even when positions are equal on paper, the better player will usually still have an edge in practice. sahil0420 made mistakes, but he made fewer of them, and his quicker play meant I had less time to calculate when the positions demanded deeper consideration.

In reviewing this game, the most frustrating moment came on move 36, because the tactic I missed should have been easy to find. After 36.Qd3, had I found 36...Qe1+, White would have had only three legal moves: 37.Kd4 Qe5# would have been mate in one, 37.Kf4 Qe5+ would have been mate in two, and 37.Qe2 Qc1+ would have won me White's passed pawn and put me in control of the game. Given how consistently I solve tactics puzzles, it feels like I should be recognizing when there are tactical opportunities like 36...Qe1+ on the board. There are games when my intuition does alert me to the probability of there being a tactic in a position, but clearly this is an area where I need to improve.