Jerome Gambit: Staggering (Part 1)

As I stagger to the finish line in the Italian Game tournament at Chess.com, I can count no wins and no draws with the Jerome Gambit, against 5 losses. 

Rather discouraging - Or time to improve my game. 

I will stick with the latter.

Perhaps, next time, I will pick on someone my own size - or rating. As I wrote in an earlier post

One interesting point is that my rating is in the bottom 10 of the tournament’s 50 players, 8th out of 10 in my group, to start. Not quite the same thing as "giving Jerome Gambit odds," but things should become interesting, quickly.

It looks like I will finish 8th in my group.

perrypawnpusher - mathgk

3 d/move, Italian game tournament, Chess.com, 2024

1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+ 

4...Kxf7 5.Nxe5+ Nxe5 6.Qh5+ Kf8 

Introduced by Alonzo Wheeler Jerome in his article in the July 1874 issue of the Dubuque Chess Journal, titled "Jerome's Double Opening: Third Variation".

The move 6...Kf8 has a lot of numbers attached to it

According to The Database, this is a popular move (3,600 games), and more successful for Black than White (Black scores 53%).

Personally, I have faced 6...Kf8 in 64 games, scoring 70% with White. That is an impressive number.

Also impressive is that playing an opponent  200 point higher than me, my chances are rated at about 25%. 

Finally, it must be added that Stockfish 16.1 (33 ply) evaluates Black as 2 1/2 pawns better. 

7.Qxe5 Qe7 

Again, this move has Jerome's stamp on it, as he played it in several correspondence games, against David Jaeger, in 1880.

I have also faced 7...d6 8.Qg3 Qe7 in perrypawnpusher - frencheng, FICS, 2010 (1-0, 17). 

8.Qf4+ Nf6 9.Nc3 

This is a good move, although it still leaves White about 2 3/4 pawn down.

I have also played 9.d3 in  perrypawnpusher - jonathankochems, blitz, FICS, 2010 (1-0, 39) and perrypawnpusher - frencheng, blitz, FICS, 2010 (1/2 - 1/2, 31), 

9...d6 

I have faced 9...c6 10.d3 d6 in perrypawnpusher - graintrader69, lichess.org, 2022 (1-0, 45). 

10.O-O Kg8 

Prudently castling-by-hand. 

A different idea was 10...Bd4 as in perrypawnpusher - ERICOLSON, blitz, FICS, 2007 (1-0, 33). 

11.d3 Bd7 12.Be3 Bb6 

White has clawed back the equivalent of 3/4 of a pawn, but the question is Now what? What do I have?

I am reminded of the lament: If I had some ham, I'd cook some ham and eggs - if I had some eggs.

[to be continued]

Jerome Gambit: Sudden Changes, Like the Weather

Move-by-move, sometimes, the Jerome Gambit looks like a win one way, then the other. In the following game, Black is winning - until he isn't. Finally, he is checkmated.

angelcamina - i_mark20120818

1 0 bullet, lichess.org, 2025

1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+ 

4....Kxf7 5.Nxe5+ Nxe5 6.Qh5+ Ng6 

7.Qd5+ Ke8 8.Qxc5 Qe7 9.Qc4 Nf6 10.Nc3 d6 11.O-O 

11...Be6 12.Qb4 

Instead, Stockfish 16.1 (30 ply) prefers by 1/4 pawn 12.Qb5+ but angelcamina has had some reverses with that move

12...c6 (12...Bd7 13.Qxb7 Rd8 14.f4 Kf7 15.Qb3+ Be6 16.Qa3 Rhe8 17.f5 Bc4 18.d3 Nxe4 19.fxg6+ Kxg6 20.Nxe4 d5 21.Qxe7 Rxe7 22.Ng3 Ba6 23.Bd2 c5 24.Rae1 c4 25.Rxe7 cxd3 26.cxd3 Bxd3 27.Rff7 Rc8 28.Rxg7+ Kf6 29.Bg5 checkmate, angelcamina - ThePr0FesS0R, 1 0 bullet, lichess.org, 2022) 13.Qa4 (13.Qe2 Rd8 14.b3 Kf7 15.f4 Bc8 16.Ba3 Rhe8 17.Rae1 c5 18.e5 dxe5 19.fxe5 Nxe5 20.Ne4 Kg8 21.Nxf6+ gxf6 22.d3 Ng6 23.Qf2 Qf7 24.Bxc5 f5 25.d4 b6 26.Ba3 Rxe1 27.Rxe1 Bb7 28.c4 Be4 29.d5 Kg7 White resigned, angelcamina - nhv2000, 1 0 bullet, lichess.org, 2022; 13.Qa5 Kf7 14.f4 Bc4 15.d3 Ba6 16.f5 Ne5 17.Bf4 Rhf8 18.Bxe5 dxe5 19.h3 Kg8 20.g4 Nd7 21.Kg2 Qc5 22.Qxc5 Nxc5 23.Rad1 Rad8 24.b4 Nd7 25.a3 b6 26.Ne2 Nf6 27.Kg3 h6 28.h4 Rfe8 29.g5 Nh5+ 30.Kg4 Nf4 31.Nxf4 exf4 32.Rxf4 hxg5 33.hxg5 Kf7 34.Rh1 Rh8 35.Rhf1 Bc8 36.Kg3 Rhe8 37.f6 g6 38.R1f3 Re5 39.c3 Rxg5+ White resigned, angelcamina - DennisMadrigal, 1 0 bullet, lichess.org, 2022) 13... Kf7 14. f4 Bd7 (14... b5 15. Qa5 Bc4 16. d3 Be6 17. f5 Ne5 18. fxe6+ Qxe6 19. Bg5 Rhf8 20. Bxf6 gxf6 21. Qc7+ Qe7 22. Qxe7+ Kxe7 23. Rf5 Rf7 24. Raf1 Raf8 25. h3 Ng6 26. g4 Nh4 27. Kh2 Nxf5 28. exf5 h6 29. Ne4 d5 30. Nc5 Kd6 31. Ne6 Re8 32.Re1 Rfe7 33. d4 a5 34. Kg3 Rg8 35. Re3 h5 36. Rc3 hxg4 37. hxg4 Rh8 White resigned, angelcamina - Chessfists, 1 0 bullet, lichess.org, 2022) 15. e5 dxe5 16.fxe5 Nxe5 17. d4 Ng6 18. Bg5 Rhf8 19. Rae1 Qd8 20. Qb3+ Be6 21. Qxe6 checkmate, angelcamina - sdewqa, 1 0 bullet, lichess.org, 2022) 

Also 12.Qa4+ c6 13.d3 Rf8 14.f4 Nd7 15.f5 Nc5 16.Qb4 Bd7 17.fxg6 hxg6 18.Be3 Rxf1+ 19.Rxf1 b6 20.Bxc5 dxc5 21.Qb3 Be6 22.Qa3 Qd6 23.Qa6 Qd4+ 24.Kh1 Qd7 25.a4 Ke7 26.a5 Rf8 27.Rxf8 Kxf8 28.axb6 axb6 29.Qxb6 Kg8 30.Qb8+ Kh7 31.Qg3 Qf7 32.Qf3 Qe7 33.Qg3 c4 34.d4 Bf7 35.Qh3+ Kg8 36.Qc8+ Be8 37.Qh3 Qb4 Black won on time, angelcamina - alwaysdreamer, 1 0 bullet, lichess.org, 2022

12...Rb8 

Or 12...b6 13. d4 Kf7 14. f4 Nh4 15. e5 dxe5 16. Qxe7+ Kxe7 17. fxe5 Nd5 18. Nxd5+ Bxd5 19. b3 Nxg2 20. c4 Nh4 21. cxd5 Rhf8 22. Ba3+ c5 23. dxc6+ Ke6 24. Rxf8 Rxf8 25. Bxf8 Nf3+ 26. Kg2 Nxd4 27. c7 Kd7 28. Rc1 Kc8 29. Bd6 Ne6 30. Rf1 Kb7 31. Rf7 Nxc7 32.Rxc7+ Ka6 33. Rxg7 b5 34. Rxh7 Ka5 35. Rxa7+ Black resigned, angelcamina - pekasmed, 1 0 bullet, lichess.org, 2023. 

Black might have tried 12...c5. 

13.d3 Kd7 14.f4 Ne8 

A slip, costing a piece. Remember, this is a 1-minute game. Better 14...Bf7. 

15.f5 Bxf5 16.exf5 Ne5 17.Bf4 Nc6 18.Qa4 a6 19.Rae1 Qf6 20.Qb3 

White has more material, better development and a safer King.

20...g6 

Instead, 20...Nd4.  

21.fxg6 hxg6 22.Be5 Qh4 23.Bxh8 Qxh8 24.Qe6+ Kd8 25.Rf7 Qg8 26.Qd7 checkmate

Jerome Gambit: Complications Matter

I suspect that one of the main reasons many players gravitate toward the Jerome Gambit is the complicated game that may arise and cause troubles for the opponent - especially in blitz.

The following is a brief example.

Pablo_sko - BolesTal

3 0 blitz, lichess.org, 2024

1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+ 

4...Kxf7 5.Nxe5+ Ke8 

White is ready for 5...Nxe5. Is he ready for this? This is a reasonable question for a 3-minute game - will the surprise factor balance out that the text is weaker than the alternative?

6.Nxc6 bxc6 

Checking with The Database, Black scores 20% with 6...bxc6.

His chances improve to 29%  with 6...dxc6.

They increase further, to 56% after the best move, 6...Qh4 - mostly because in those games, White's responses are not the best. The computer still gives White the edge.

7.Qh5+ g6 8.Qxc5 Qf6 9.e5 d6 

A slip.

10.Qxc6+ Black resigned

Jerome Gambit: Played Too Quickly

 

The following game reminds me of the Roadrunner, Wile E. Coyote, and a cliff... If they were playing blitz.

Pablo_sko - BetterCallBatu

3 0 blitz, lichess.org, 2024

1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+ 

4...Kxf7 5.Nxe5+ Nxe5 6.d4 d5 

The move ...d5 is almost the Swiss Army Knife of Jerome Gambit defenses.

The Database shows White winning 9 games with this position, versus 6 games by Black, but this is misleading. Stockfish 16.1, at 44 ply, rates Black as 2 1/2 pawns better.

7.dxc5 

Take the Bishop or take the Kight? Surprisingly, Stockfish 16.1 at 44 ply, prefers taking the Knight, 7.dxe5. 

7...dxe4 

Whoa, whoa, whoa!

The players are moving quickly with only 3 minutes apiece on the clock. Alas, this move was played too quickly.

8.Qxd8 Black resigned

(For the record, both players were rated above 1900.)

Jerome Gambit: Both Knights

The following blitz game moves quickly to the endgame - one which is both educational and a bit humorous.

Pablo_sko - Dominator696969

3 0 blitz, lichess.org, 2025

1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+ 

4...Kxf7 5.Nxe5+ Nxe5 6.d4 

At first 6.d4 was Alonzo Wheeler Jerome's choice, although later he switched to 6.Qh5+.

6...Bxd4 7.Qxd4 Qf6 

With the small trap, if now 8.Nc3 or 8.O-O, then 8...Nf3+ would lead to the win of the White Queen. This idea goes back to the correspondence game Jerome - Norton, 1876 (16 moves, unfinished).

8.c3 

White says Yes, I see it. I would have preferred the developing 8.Be3.

8...Ne7 9.O-O N7c6 10.Qd5+ Qe6 11.f4 

Allowing the exchange of Queens, which is usually fovorable for Black in the Jerome Gambit, as it blunts White's attack.

However, that is not the case here.

11...Qxd5 12.exd5 

White misses the intermezzo 12.fxe5+ when 12...Kg6 13.exd5 Nxe5 would produce a position where material would be equal, but in which the first player's King would be safer. But, remember, this is a 3-minute game; things happen.

12...Nb4

Both of Black's Knights are under attack, but he knows that White can capture only one of them. Can Black get this Knight to c2, to threaten the Rook at a1?

No. It was better to try 12...Nd3 13.dxc6 bxc6 with an edge.

13.fxe5+ Black resigned

The second Knight will be lost as well.

Jerome Gambit: Too Fast For Me

I continue to be amazed at the interesting chess created by regular contributor angelcamina - played at one minute a game.

That's too fast for me. My heart raced as I played over the following game - and I was simply sitting at my computer, composing.

angelcamina - dv_66

1 0 bullet, lichess.org, 2025

1.e4 e5 2.Nf3 Nf6 3.Bc4 Nc6 4.Nc3 Nxe4 5.Bxf7+ 

Jerome Gambit's cousin, the Noa Gambit. (See "Further Explorations 2, 3, 4, 5 & 6") 

5...Kxf7 6.Nxe4 d5 7.Neg5+

Giving Black the opportunity to choose the wrong retreat. Remember, this is a 1-minute game, and if Black is not familiar with the opening...

He has also tried 7.Nc3 e4 8.Ng1 Bc5 9.d4 Bxd4 10.Nge2 Bb6 11.O-O Rf8 12.Nxd5 Kg8 13.Nxb6 axb6 14.Qe1 Ne5 15.Qc3 Qf6 16.Bf4 Ng4 17.Qxf6 Nxf6 18.Bxc7 Bg4 19.Ng3 Rac8 20.Bxb6 Rxc2 21.Bd4 Rd8 22.Bc3 b5 23.a3 Rd3 24.h3 Be6 25.Rad1 Bb3 26.Rxd3 exd3 27.Rd1 Bc4 28.Ne4 Nd5 29.Nd2 Nxc3 30.Nxc4 Rxb2 31.Rxd3 Nd1 32.Kh2 Rc2 33.Rd8+ Kf7 34.Ne3 Nxe3 35.fxe3 Rf2 36.Rd7+ Kg6 37.Ra7 Black lost on time, angelcamina - florianschreib, 1 0 bullet, lichess.org, 2019. 

7...Kg8 

Got it right, this time.

White also faced 7...Ke8 8.d4 h6 9.Nh3 e4 10.Ne5 Nxe5 11.Qh5+ Nf7 12.Nf4 Qf613.Nxd5 Qd6 14.Bf4 Qc6 15.Nxc7+ Ke7 16.Nd5+ Ke8 17.Nc7+ Ke7 18.Nxa8 Qxc2 19.O-O Be6 20.Rac1 Qxb2 21.Rc7+ Ke8 22.Rxf7 Bxf7 23.Qe5+ Be7 24.Nc7+ Kf8 25.d5 Qb6 26.d6  Qxd6 27.Qxd6 Bxd6 28.Bxd6+ Kg8 29.Rc1 Bxa2 30.Nb5  a6 31.Nd4 Kh7 32.Rc7 Rd8 33.Be5 b5 34.Rxg7+ Kh8 35.Rd7+ Black resigned, angelcamina - Farbic, 1 0 bullet, lichess.org, 2024; and 

7...Kf6 8.d4 Bg4 9.dxe5+ Nxe5 10.Qd4 Bxf3 11.Nxf3 Qe7 12.Bg5+ Black resigned, angelcamina - GrainBowl, 1 0 bullet, lichess.org, 2023.

8.d4 e4 

Kicking one Knight, but he would have done better to kick the other one with 8...h6. 

9.Ne5 Nxe5 10.dxe5 h6 

Also seen 10...c6 11.Nxe4 Bf5 12.Ng3 Bg6 13.O-O h6 14.f4 Be7 15.f5 Bf7 16.Qg4 Kh7 17.Nh5 Bxh5 18.Qxh5 Qb6+ 19.Kh1 Rhf8 20.Qg6+ Kh8 21.f6 gxf6 22.exf6 c5 23.Qg7 checkmate angelcamina - Lelzwashere, 1 0 bullet, lichess.org, 2024) 

11.Nxe4

Because he can. 

11...Bg4 

Or 11...Bb4+ 12.c3 Be7 13.Ng3 c6 14.O-O Be6 15.f4 Kh7 16.f5 Black resigned angelcamina - Otschin_Karascho, 1 0 bullet, lichess.org, 2024) 

12.Qxg4 dxe4 13.Qxe4 

Black has no compensation for his two pawns (unless it is on the clock).

13...Bc5 

Must be the clock.

14.Bd2 

Missing 14.Qc4+ and 15.Qxc5. Tick Tock, Tick Tock.

14...Qe7 

Covering the Bishop, and missing the tricky 14...Bxf2+.

15.O-O-O a5 16.f4 b5 17.f5 

White's pawns are more dangerous than Black's.

17...Kf7 

I am not sure what was behind this provocation - getting a Rook to f8? At any rate, this is risky business.

18.e6+ Kf6 

Retreating was essential.

19.Bc3+ Kg5 20.h4+ Kh5 21.g4 checkmate

Sweet.

BSJG: Hunting the Uncooperative King

The Following Blackburn Shilling Jerome Gambit game features a King hunt, and a very uncooperative King.

White eventually corralled the monarch, but not without effort.

anilone - mohammadkhanbeyg

10 0 rapid, lichess.org, 2024

1.e4 e5 2.Nf3 Nc6 3.Bc4 Nd4 

The Blackburn Shilling Gambit.

4.Bxf7+ 

The Blackburn Shilling Jerome Gambit.

4...Kxf7 5.Nxe5+ Ke6 

The King can also retreat with 5...Ke8, as in Nater - Rogers, Ballarat, 1975 (0-1, 46).

6.c3

This is the right idea, and it tempts Black to pursue material. 

6...Kxe5 

Of equal strength is 6...Nc6, e.g. 7.Qg4+ Kxe5 8.d4+ (8.Qf5+ Kd6 9.d4 Nce7? 10.Bf4+ Black resigned, anilone - butilicapetrica, lichess.org, 2024) 8...Kd6 9.e5+ Nxe5 10.dxe5+ Kxe5 11.Bg5 Nf6 12.Qf4+ Ke6 13.O-O d5 14.c4 c6 15.Nc3 Bd6 16.cxd5+ Kf7 17.Qf3 Rf8 18.dxc6 bxc6 19.Qxc6 Be6 20.Rad1 Ke7 21.Ne4 Qd7 22.Rxd6 Qxc6 23.Rxc6 h6 24.Bxf6+ gxf6 25.Re1 Kf7 26.Nd6+ Kg6 27.Rxe6 Rab8 28.b3 a5 29.Ne4 a4 30.Nxf6 axb3 31.Nd7+ Kg7 32.Rg6+ Kf7 33.axb3 Rfd8 34.Ne5+ Kf8 35.Rcf6+ Ke7 36.Rf7+ Ke8 37.Rg8 checkmate, anilone - fhat31, lichess.org, 2024. 

7.cxd4+ 

White has also experimented with 

7.Qh5+ Ke6 8.Qd5+ (8.cxd4 Nf6 9.Qd5+? White resigned, anilone - Togtamyrat, lichess.org, 2024) 8...Ke7 9.Qxd4 d6 10.d3 c5 11.Qe3 Kf7 12.O-O Be7 13.f4 h6 14.e5 d5 15.f5 Bg5 16.e6+ Kf6 17.Qf3 Bxc1 18.Rxc1 Qd6 19.d4 c4 20.Re1 Ke7 21.Re5 Qb6 22.Qg3 Qxb2 23.Qxg7+ Kd6 24.Re1 Qxa1 25.e7 Nxe7 26.Qxe7+ Kc6 27.Qc5+ Kd7 28.Re7+ Kd8 29.Qc7 Black resigned, anilone - PPNRegKal, lichess.org, 2024; and 

7.f4+ Kd6 8.e5+ Kc6 9.cxd4 d5 10.O-O b6 11.Qa4+ Kb7 12.Nc3 Bd7 13.Qb3 Bc6 14.a4 Rb8 15.a5 Ka8 16.axb6 cxb6 17.Nb5 Bxb5 18.Qxb5 Ne7 19.d3 Qe8 20.Qa6 Nc6 21.Be3 Qd7 22.Rfc1 Nb4 23.Qa3 Nc2 24.Qxa7+ Qxa7 25.Rxa7+ Kxa7 26.Rxc2 Be7 27.Rc6 Rb7 28.f5 Rf8 29.g4 Ka6 30.Kg2 Kb5 31.Re6 h6 32.Kg3 Kb4 33.Bd2+ Kb3 34.Bc3 Bb4 35.Bxb4 Kxb4 36.Rd6 Kb3 37.Rxd5 Kxb2 38.e6 Kc3 39.Kf4 Kxd3 40.Ke5 Kc4 41.Rd7 Rxd7 42.exd7 Rd8 43.Ke6 b5 44.h4 b4 45.g5 hxg5 46.hxg5 Kxd4 47.f6 gxf6 48.gxf6 b3 49.f7 b2 50.Ke7 Rxd7+ 51.Kxd7 b1=Q 52.f8=Q Qb5+ 53.Ke6 Qd5+ 54.Kf6 Qf3+ 55.Kg7 Qg4+ 56.Kf7 Qf5+ 57.Kg8 Qe6+ 58.Qf7 Qg4+ 59.Qg7+ Qxg7+ 60.Kxg7 Drawn, anilone - dgverma93, lichess.org, 2024 

7...Kxd4 

Taking the bait. In 212 games in The Database, White now scored 65%. 

Taking the e-pawn was no solution, either, e.g. 7...Kxe4 8.Qh5 Nf6 9.Qe5+ (9.d3+ Kxd4 10.Qb5+ Kxd4 11.Be3+ Ke4 12.Nc3 checkmate) Kd3 10.Qe3+ (10.O-O!?) Kc4 11.Na3+ (11.b3+ Kb5 12.Nc3+ Kb6 13.d5+ Bc5 14.Na4+ Ka6 15.Nxc5+ Kb5 16.a4+ Black resigned, anilone - STEAUA1980, lichess.org, 2024) 11...Kd5 12.Qe5+ Kc6 13.Qb5+ Kd6 14.O-O Ke7 15.Re1+ Kf7 16.Nc4 d6 17.Qd5+ Kg6 18.Ne5+ Kh5 19.Qf3+ Bg4 20.Nxg4 Nxg4 21.h3 Qh4 22.Qf5+ Qg5 23.hxg4+ Black resigned anilone - Sanmisongz. lichess.org. 2024. 

Giving Black an edge was 7...Ke6 8.d3 (or 8.d5+ Ke7 9.d4 d6 10.Bg5+ Nf6 11.e5 dxe5 12.dxe5 Kf7 13.O-O h6 14.Bxf6 gxf6 15.Qh5+ Kg7 16.e6 Qe7 17.Nc3 Rh7 18.f4 Kh8 19.f5 Bg7 20.Ne4 b6 21.Rad1 Bb7 22.d6 cxd6 23.Nxd6 Bc6 24.Nf7+ Kg8 25.Rd6 Rc8 26.Rfd1 Bf8 27.Qg6+ Rg7 28.Nxh6+ Kh8 29.Nf7+ Rxf7 30.exf7 Qe3+ 31.Kh1 Bxd6 32.Qxf6+ Kh7 33.Qxd6 Bxg2+ 34.Kxg2 Rc2+ 35.Kh1 Qf3+ 36.Kg1 Qf2+ 37.Kh1 Qg2 checkmate, anilone - lorountu, lichess.org, 2024) 8...Qf6 9.d5+ Kf7 10.O-O Bc5 11.Be3 d6 12.Nc3 Ne7 13.d4 Bb4 14.f4 Bxc3 15.bxc3 Rf8 16.e5 Qf5 17.c4 Kg8 18.e6 Bxe6 19.dxe6 Qxe6 20.Qd3 Nf5 21.Rae1 Nxe3 22.Rxe3 Qf5 23.Qxf5 Rxf5 24.Re7 Rf7 25.Rfe1 b6 26.g3 c5 27.d5 a6 28.R1e6 Rxe7 29.Rxe7 b5 30.Rd7 bxc4 31.Kf2 c3 32.Ke2 Rb8 33.Kd3 Rb2 34.Kxc3 Rxa2 35.Rxd6 Rxh2 36.Rxa6 Rg2 37.d6 Rxg3+ 38.Kc4 Kf7 39. f5 g6 40. Kxc5 gxf5 41. Ra7+ Ke8 42. Rxh7 f4 43. Kd5 f3 44. Kc6 f2 45.Rh8+ Kf7 46.Rh7+ Kg6 47.Rh1 Rg1 48.Rh8 f1=Q 49.g8+ Kf7 White resigned, anilone - MARQO333, lichess.org, 2024

8.Qh5 

Cutting the enemy King off from returning to his home is a good idea, although it is not the best one. Instead, 8.Nc3 was seen in 13 games in The Database, with White scoring 72%.

Such an exposed King surely must get checkmated!

8...Nf6 9.Qg5 

Keeping the enemy far afield, and an improvement over 9.Qb5 c6 10.Qa4+ Ke5 11.d4+ Kxe4 12.Nc3+ Kf5 13.d5 Qe7+ 14.Be3 cxd5 15.O-O Kg6 16.Bd4 Qb4 17.Qc2+ Kf7 18.Bxf6 gxf6 19.Nxd5 Qc5 20.Qf5 Bg7 21.Rac1 Qa5 22.Rfe1 d6 23.Re7+ Kg8 24.Rxc8+ Rxc8 25.Qe6+ Black resigned, anilone - Wacwacnepo, lichess.org, 2023. 

9...d5 

This is an all-around useful move for the defender.

10.e5 Qe7 

Black pins the e-pawn, but this move neglects the safety of his King. Better was 10...Kc5 11.d4!? Kc6 12.O-O Be6 13.exf6 Qxf6 with continued equality.

11.Qe3+ Kc4 

12.Qc3+ 

Instead, 12.Na3+ would force the win of the enemy Queen with 12...Qxa3 13.bxa3, because 12...Kb4 would allow 13.Qc3+ Ka4 14.b3 checkmate.

12...Kb5 

The King keeps trying to escape. White could now try 13.a4+ Kb6 14.a5+ Ka6 15.O-O Ne8, but Black would have an edge. Frustrating!

13.Na3+ Ka6 

A slip - and a fatal one. Still good for Black was 13...Kb6 14.Qb3+ Qb4 15.Qxb4+ Bxb4 16.exf6 Bf5 17.O-O gxf6.

14.Qd3+ Kb6 15.Qb5 checkmate