View the game on Lichess
I recently got bored of my narrow opening base and started playing E4 to spice things up a bit. I quickly fell in love with the King's gambit, it frequently makes for some extremely fun and dynamic games. Possibly the most important thing to remember is that in most positions you have to be hyper-aware of the diagonal to the king opened with 2.F4. Its something that I normally remember, having fallen victim to QH4 a few times, but sometimes I get thrown off by unfamiliar openings and the Falkbeer is one of those (this was only the second time facing it). So instead of the usual NG3 (or BF6 if you're feeling spicy) I instead played NC£ a seemingly logical move to defend the pWN that puts white in trouble right away.

After QE2+, the only response is to bite the bullet, moving the king off the backrank and blocking both my queen and bishop :(

After a few natural moves bringing in the Bishop to the attack and defending with the Horsey, black played a very strange move that seems to be a big blindspot in this opening at my level. 6.C5??. Of course this is a blunder because white has the very natural response D4, kicking the bishop away, disconnecting it from the kingside and allowing the white queen some more room to move. Inexplicably they followed this move up by permanently locking the bishop away with BB6, as long as the D pawn is defended. She had the right idea of just developing and bringing more pieces into the attack, developing the bishop to D6 instead would have been completely reasonable, or even just bringing out the queenside Horsey and long castling to bring a rook to the E file, where my king is a sitting duck.

At this point I did something I'm well aware I have a real problem with, and need to keep working on, and that's placing some visual instinct above concrete (and sometimes straightforward) calculation. The best move is surely 9.BxF4. It develops the bishop to a place where it could dominate the other dark squared bishop if she ever wants to play C6 bring it back in to attack the kingside. It allows me to run my king to D2, breaking the pin and potentially allowing my queenside rook to come to the center in the future, followed by hiding the king behind the pawns. Removing the F pawn also allows H3 to be played at some point without giving G3 to the black Queen. The reason I didn't play it is because I knew that if he captured my Horsey with the Bishop, I would have to recapture with the king to stop my newly developed Bishop from hanging to the Queen. And that just looks very scary! I don't want my king on F3! But in that position there are really no good checks and moving my King means that I'm ready to develop my blocked Queen and Bishop and bring my Rook to the centre before black can get a good attack going.

So the point of Qd3 was that if I captured the pawn with the Queen, the Bishop would defend it and there would be no need to recapture the Bishop with King after it took my Horsey. Not an awful move, but it gives black more time she shoudln't have, and also enables really irritation moves like g5, gluing the pawn in my kingside forever like a thorn.

Fortunately my opponent played c6, presumably to get their bishop back into the game, but its simply far too slow. Qxf4 beats them to the punch and lets me control the diagonal the Bishop would move onto. Then my opponent actually did exactly the thing I was preparing for with Qe2! Bxf3,Kxf3,Qxf4. And then for some godawful reason I completely forgot why I'd played Qe2 and played Kxf4... throwing away all advantage by not bringing my Bishop out quickly and leaving my King a sitting duck for future checks (that's foreshadowing for a few paragraphs time!) The reason Bx is so strong is that Black is 2 moves from castling in either direction. After pretty much any move Re1 will stop the King ever reaching safety and easily bringing her Rook to the centre. At that point, it might even be possible to consider the white King in the center of the board an asset.

In any case, while white has a very slight advantage according the engine, it seems practically much harder for white to play. Black can now capture d4, which they did not have time to do in the better line, potentially threatening Bxc3 to ruin my pawn structure. There is still a way to play the same idea as the better line by developing either Bishop, again threating to give a Rook check to the Black King. At this point I really started to lose the thread though, and played 12.dxc6, a move that solves none of my problems and allows him to naturally develop his Horsey with the recapture. This culminated with 17.Kf3, blundering the obvious Horsey fork, allowing him to take away the Bishop pair from me, and leave me with an isolated pawn.

At this point it was clear the game was slipping away from me, I was down a pawn and almost certainly going to lose the isolated center pawn, at which point she could use the 2v1 pawn majority on the Queenside to either create a passed pawn, force me to lose material, or both. So I tried to create some counterplay witn 20.Ne4. This offers the b pawn as a sacrifice, with the idea that if he accepts it, Rab1 pins the bishop the Black b pawn. If I could clear off the enemy b pawn, It becomes a lot harder to create a passed flank pawn and I had a reasonable chance of holding a draw, even with the 3v2 on the kingside. Unfortunately this simply allows Be5+, stopping my from taking the pawn...

Ultimately though, black didn't press the advantage, they let me trade enough pieces to remove a lot of pressure, I'm quite pleased that I found 27.a4 to break up his pawn majority. If he takes I was planning on Rb6, but the stronger move I think is Ra1. if Rb6 black can evacuate the Horsey with Ne5, threatining to fork my King and Bishop, and buying enough time to put his Rook behind one of his A pawns. If Ra1, there is no defense of 1 of the A pawns, and the rook will be tied to the defence of the other in the long term. Again, If I can eliminate that pawn, I have at least good drawing chances, and *maybe* some winning chances with my own passed pawn.

In any case, he instead defended the B pawn with Rb8 allowing me to simply capture the pawn. It might be tempting for black to advance the A pawn, allow me to take the Horsey and then take my undefended Rook. But then the pawn will unstoppably remote, thanks to the bishop blocking the Rook from getting behind it. So instead you really have to either take the pawn with the pawn or Rook. Either way is probably a draw, perhaps if the black King were closer to the passed pawn maybe there would be something, but as it is, I can blockade their pawn, and their king is close enough that I shouldn't be able to promote, with careful play it should be a draw.

HOWEVER. 30.Ne5 opens up the possibility of forking the Rook and Horsey. It doesn't work immediately because of Nd3+, which I assume was their idea, but it gives the white king a chance to move closer to the center and maybe push for something. In the game though I discarded Ke3 because of because of Ng4+ followed by capturing the H pawn, but I missed that that falls victim to the same Bishop fork. Black's reaction however, first blundered a pretty concrete draw, then the game. Kf7 allows the fork, the only good defense to which is to defend the Horsey with Re8. Trading off the Horsey and the Bishop, then taking the b pawn is a pretty clear draw. At some point they'll be able to capture the passed pawn, but 3v1 with Rooks on the board isn't enough to win. Likewise if I take the pawn without trading the Horsey and Knight, my king is too far to save the pawn, and the result should be similar.

Instead Kf6 blundered their rook and the game. GG.

Lessons from this game: [Event "Rated Rapid game"]
[Site "https://lichess.org/x96cj0dL"]
[Date "2022.06.15"]
[White "MattingtonBear890"]
[Black "Avalanchian"]
[Result "1-0"]
[UTCDate "2022.06.15"]
[UTCTime "15:06:41"]
[WhiteElo "1639"]
[BlackElo "1653"]
[WhiteRatingDiff "+6"]
[BlackRatingDiff "-19"]
[Variant "Standard"]
[TimeControl "900+10"]
[ECO "C31"]
[Opening "King's Gambit Declined: Falkbeer Countergambit, Modern Transfer"]
[Termination "Normal"]
[Annotator "lichess.org"]
1. e4 e5 2. f4?! { (0.12 → -0.55) Inaccuracy. Nf3 was best. } (2. Nf3 Nc6 3. Bb5 Nf6 4. O-O Nxe4 5. Re1 Nd6 6. Nxe5 Be7) 2... d5 3. exd5 exf4 { C31 King's Gambit Declined: Falkbeer Countergambit, Modern Transfer } 4. Nc3? { (-0.28 → -1.50) Mistake. Nf3 was best. } (4. Nf3 Nf6 5. c4 c6 6. d4 Bb4+ 7. Nc3 cxd5 8. Be2 Nc6) 4... Qh4+ 5. Ke2 Bg4+ 6. Nf3 Bc5?? { (-2.49 → -0.52) Blunder. Bd6 was best. } (6... Bd6 7. Qe1 Qh6 8. Kd1+ Ne7 9. Be2 Nd7 10. h3 Bxf3 11. Bxf3) 7. d4 Bb6? { (-0.27 → 0.96) Mistake. Bd6 was best. } (7... Bd6 8. Ne4 Ne7 9. Kd2 Qh6 10. c4 O-O 11. Bd3 Na6 12. h3) 8. Qd2 c6?! { (0.48 → 1.17) Inaccuracy. Ne7 was best. } (8... Ne7 9. Qxf4) 9. Qxf4 Bxf3+?? { (0.89 → 3.69) Blunder. Ne7 was best. } (9... Ne7 10. Be3 Bxf3+ 11. Qxf3 Bxd4 12. g3 Qf6 13. Bxd4 Qxd4 14. Rd1 Qe5+ 15. Qe4 Qxe4+ 16. Nxe4) 10. Kxf3 Qxf4+ 11. Kxf4?? { (3.42 → 0.40) Blunder. Bxf4 was best. } (11. Bxf4 Ne7 12. Bc4 Kd7 13. g3 Bxd4 14. Rhd1 Bxc3 15. dxc6+ Kxc6 16. bxc3 Kc5 17. Bb3 Nbc6) 11... Bxd4 12. dxc6?! { (0.58 → -0.41) Inaccuracy. Bc4 was best. } (12. Bc4 Ne7 13. Re1 f6 14. Kg3 Nd7 15. dxc6 bxc6 16. h4 Ne5 17. Be6 Rd8 18. a4 Nd5) 12... Nxc6 13. Bb5 Ne7 14. Re1 O-O 15. Bd2 a6 16. Bd3?! { (-0.55 → -1.25) Inaccuracy. Ba4 was best. } (16. Ba4) 16... Rfe8?! { (-1.25 → -0.52) Inaccuracy. Ne5 was best. } (16... Ne5 17. Be2 Rac8 18. Kg3 N7g6 19. h4 Nc4 20. Bxc4 Rxc4 21. h5 Ne5 22. h6 Ng4 23. hxg7) 17. Kf3? { (-0.52 → -1.80) Mistake. Re2 was best. } (17. Re2 Rad8) 17... Ne5+ 18. Kg3 Nxd3 19. cxd3 Rad8 20. Ne4? { (-2.00 → -3.60) Mistake. Kf3 was best. } (20. Kf3 Rd7 21. Rac1 Red8 22. Be3 Bxe3 23. Kxe3 Rxd3+ 24. Kf2 Rd2+ 25. Re2 Nc6 26. Ne4 Rxe2+) 20... Bxb2 21. Rab1 Be5+ 22. Kf2 b5 23. Nc5?! { (-3.99 → -5.92) Inaccuracy. Rb3 was best. } (23. Rb3 Nc6) 23... Bd4+ 24. Be3 Bxc5?? { (-5.99 → -2.17) Blunder. Bc3 was best. } (24... Bc3 25. Rec1 Nd5 26. Ne4 Ba5 27. g4 Nb4 28. Rd1 Nxd3+ 29. Ke2 Rxe4 30. Rxd3 Rxd3 31. Kxd3) 25. Bxc5 Nc6?! { (-2.02 → -1.29) Inaccuracy. Nd5 was best. } (25... Nd5 26. Kf3) 26. Rxe8+ Rxe8 27. a4 Rb8? { (-1.48 → -0.35) Mistake. b4 was best. } (27... b4) 28. axb5 axb5 29. d4 f5 30. d5 Ne5 31. Kg3?! { (-0.19 → -0.72) Inaccuracy. Ke3 was best. } (31. Ke3 Rd8 32. d6 Kf7 33. Rxb5 Ke6 34. Rb7 Nd7 35. Bd4 g5 36. Kf2 Ra8 37. Rb5 h5) 31... Kf7?! { (-0.72 → 0.00) Inaccuracy. Rb7 was best. } (31... Rb7 32. Bd4 Ng6 33. Bc5 Kf7 34. Kf3 Ne7 35. Kf4 Nxd5+ 36. Kxf5 Nf6 37. Bd6 Nd7 38. Ba3) 32. Bd6 Kf6?? { (0.00 → 8.66) Blunder. Re8 was best. } (32... Re8 33. Rxb5 g5 34. Bxe5 Rxe5 35. Kf3 h6 36. h3 f4 37. Rb7+ Kf6 38. d6 Re3+ 39. Kf2) 33. Bxb8 { Black resigns. } 1-0