From the Corner - Smiling

It is hard to mention the chess wise guy Geoff Chandler without thinking about the following game that he posted, years ago, on his "Chandler Cornered" site

Anon - Anon
Edinburgh Congress

1.e4 e5 2.Nf3 Nf6 3.Nxe5 d6 4.Nf3 Nxe4 5.d4 d5 6.Bf4 Be7 7.Nc3 Nf6 8.Nb5 Bb4+

Geoff noted

Black played Bb4+ with the idea of playing Ba5 covering c7.
Whilst White was pondering on his next move Black suddenly stated:
"I thinks It's checkmate!"
And it was!!!
Before the start of the game neither player had noticed that the Black King and Queen were on the wrong squares.

(Okay, that was a Petroff Defense, not a Jerome Gambit. If you feel cheated, don't worry, I have you covered: check out Geoff's legendary "Mars Attacks!") 

Tidying Up - Or Messing Up?

Recently I was looking through long-time friend of this blog IM Gary Lane's 2012 "Trash or Treasure?" column, part of his at "Opening Lanes" efforts at Chess Cafe.

(Actually, I was looking at an old pdf file, stored on my phone - a phrase that would probably have been nearly meaningless when I first started this blog.)

I spotted some apparent confusion related to a Jerome Gambit game, and as I may have had a hand in causing it, I thought I'd try to do some unraveling.

From "Trash or Treasure?" 

...Finally, Mr. Kennedy pointed out a fairly recent game played by Scottish player Geoff Chandler. I have never met him, but I do know that Mr. Chandler has an excellent sense of humour and his old chess blog at Chandler Cornered was zany, thought provoking, and usually very funny. Therefore, the following game looks like a fabrication, but I am happy to be corrected in the future. Here is another Jerome Gambit game that is spectacular as always!

Chandler, Geoff - Dimitrov Todor

Blitz, Edinburgh, 2004

1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+ Kxf7 5.Nxe5+ Nxe5 6.Qh5+

This opening is ideally suited to blitz where you don't care whether you win or lose, but want to play something memorable.  

6...g6 7.Qxe5 d6 8.Qxh8 Qh4 9.O-O Nf6 10.Qd8!

Geoff is a decent club player and could have found this himself if the game was really played.* I still think it was more likely he was following the advice given in the previous Blackburne game, which has been copied up to this point. However, I did look up his old blog and found this comment "I recall about a year ago Todor and me had a dozen or so games playing 4.Bxf7+ at 5 minute chess in Bells." If you think he played a game inside an actual bell, then think again. He is referring to his chess club hosted at a local bar.  

10...Bh3 11.Qxc7+ Kg8

Here IM Lane gives 12.gxh3 and says

Instead 12.Qxb7 is winning, because12...Qg4 can be met by 13.Qb3+! (13.Qxa8+ Kf7 14.Qb7+ Kf8 15.e5? White should keep on checking, but this winning attempt backfires spectacularly upon 15...d5 and it turns out that Black wins.) 13...Kg7 Qxh3 and it is time for Black to put the pieces back into the box.  

Then 12...Qxh3 13.Qxb7 Qg4+ A draw by repetition beckons, but Mr. Kennedy assures me that Geoff went on to win.

 Actually, the game continued 12.Qxb7 Qg4 13.Qb3+ Kg7 14.Qxh3 and according to Chandler, White won.

How did the mixup in the moves of the game occur? I could have jumbled them when I emailed the game to IM Lane - if I actually sent it, as I can't find any record of that amongst our correspondence. (Gary might have made the slip, but is that likely? He's the professional, I'm the amateur.)

Anyhow, the Chandler - Dimitrov game and analysis can get pretty messy, so perhaps that was part of it.    

In support of that possibility, and a possible clue, it is worth looking at "Updating the Blackburne Defense (Part 2)" where I reference, among a number of things, Dennis Monokroussos's thoughts from about 7 years earlier about Amateur - Blackburne, London, 1884, after 1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+ Kxf7 5.Nxe5+ Nxe5 6.Qh5+ g6 7.Qxe5 d6 8.Qxh8 Qh4 9.O-O 

Dennis M's Chess Site
February 2, 2005

...But now, here's the puzzle. After 9...Nf6, Black has a substantial lead in development and several well-placed pieces ready to commence a feeding frenzy on the White kingside, yet had White found 10.Qd8, pinning the Black Nf6 to the queen on h4, it would have been Black needing to fight for his life! The following might be best play for both sides: 10.Qd8! Bh3 11.Qxc7+ (11.Qxa8? Qg4 12.g3 Qf3 forces mate) Kf8! (11...Kg8? 12.Qxb7 Qg4 13.Qb3+ and 14.Qxh3) 12.Qxb7 Qg4 13.Qxa8+ Kf7 14.Qb7+ Kf8 14.Qa8+ with a draw by perpetual check.  

When I first saw this game and was told about 10.Qd8, it seemed to me that Black just had to have something, but neither I nor my silicon friends have succeeded in proving a win or even an advantage for Black. Can any of my readers find something better for Black?

I can sympathize with Dennis - how can Black not win against the Jerome Gambit?? In a responding comment on his blog I shared

The line gets some analysis by Geoff Chandler and Todor Dimitrov on the former's hilarious website, Chandler Cornered http://www.chessedinburgh.co.uk/index.htm
It goes like this. (Notes by Chandler.)

10.Qd8 Bh3 Threatening simply Qg4 and Qg2 mate. 11.Qxc7+ Kf8 This is best. [In my Game v Todd he played the natural 11...Kg8 which allows a check on b3 12.Qxb7 Qg4 13.Qb3+ Kg7 14.Qxh3] 12.gxh3 forced [If 12.Qxb7 Qg4 13.Qxa8+ Kf7 (13...Kg7 14.e5 d5 15.exf6+ Kxf6 16.Qxd5) 14.e5 d5 15.e6+ (15.Qb7+ Be7 16.e6+ Kg7 17.Qxe7+ Kh6 18.d4+ Kh5) 15...Kg7 16.Qb7+ Kh6 17.d4+ Kh5 and Black mates on g2] 12...Qxh3 This appears to be the best. It keeps the attack rolling and keeps the draw in hand. Remember we are seeing if 10.Qd8 beats the Blackburne line. 13.Qxb7 Ng4 [Or 13...Qg4+ and ...Qf3+ drawing.] 14.Qxa8+ Kg7 15.Qb7+ Kg8 16.Qc8+ Kg7 17.Qd7+ Kg8 18.Qe8+ Kg7 19.Qe7+ Kg8 Black has to allow the draw else 18.Qe8+ Kg7 19.Qf7+ kh6 10.d4+ wins. So it appears 10.Qd8 draws.

Note in the above that the conclusion is that the game is drawn -- the same conclusion as you came to, although the particular line you give (12.Qxb7 instead of Chandler and Dimitrov's 12.gxh3) seems to tilt toward White.

In a later post Monokroussos added

(2) In my main line, Kennedy, citing analysis by Geoff Chandler and Todor Dimitrov, varies from my 12.Qxb7 with 12.gxh3, showing that it likewise draws after 12...Qxh3 13.Qxb7 Qg4+ 14.Kh1 Qf3+ etc. or 13...Ng4 14.Qxa8+ etc. (Note that Black can't escape the checks with 14...Ke7 15.Qb7+! Kf6?? [15...Kd8/e8/f8=] because of 16.e5+ followed by 17.Qg2.) 

(3) Chandler & Dimitrov also mention 12.Qxb7 and suggest it loses, but the culprit is not 12.Qxb7 but their 14.e5?, after which Black has a forced mate. 

Very interesting and I'm grateful to Kennedy for his comment...but my dream remains unfulfilled - can't Black win after 10.Qd8, somehow?

Readers, is this confusing enough for you? Above, I quote Monokroussos quoting me quoting Chandler...

I have put the moves to Chandler - Dimitrov, cited by Chandler, above, in italics. The move 12.gxh3, which IM Lane gives as part of the game, is actually part of Chandler's analysis after 11...Kf8, not 11...Kg8, as played in the game - although Chandler says in his note that the move 12.gxh3 is "forced" which may have made it look like it was played.

I muddied things even more by referring, in my comment to Monokroussos, to "Chandler and Dimitrov's 12.gxh3" - the move was from their analysis, as presented by Chandler, above, not their game; andy by referring to 12.Qxb7, the actual move in the game, as "the particular line you give". Monokroussos seems to catch this, as indicated in his (2) note in the later post.

By the way, Monokroussos is right in note (3) in correcting Chandler's analysis (which I had provided) that after 10.Qd8 Bh3 11.Qxc7+ Kf8 12.Qxb7 White does not lose - after 12...Qg4 13.Qxa8+ Kf7 the move 14.e5 is "the culprit... after which Black has a forced mate". Instead, 14.Qb7+ Kf8 15.Qa8+ draws by repetition - as Monokroussos mentioned in his first post, after "...Now here's the puzzle." 

Still, Monokroussos doesn't escape completely. The later post, note (2), above, gives the sideline 12.gxh3 Qxh3 13.Qxb7 Ng4 [instead of 13...Qg4+, drawing] for Black, suggesting that after 14.Qxa8+ etc. the game is drawn as well - but White has, instead of grabbing the Rook, the forced Queen exchange after 14.Qb3+ (how un-Jerome-ish) 14...Qxb3 15.axb3 which leaves him a Rook and 3 pawns better.

Ah, yes, now everything now is as clear as... trash. 

(*- Chandler commented in Chandler Cornered about 10.Qd8 "This is my over the board improvement that I have since learnt was first suggested in 1951." I had told Chandler that P. Wenman mentioned the move in his Master Chess Play (1951). I later learned that the move had been played in Harris, S - Quayle, E., correspondence, 1944, although, of course, the move had been first suggested in the August 1885 issue of the Brooklyn Chess Chronicle.)

Jerome Gambit: No Mercy

In some amateur sports leagues a game will be ended under the "mercy rule" if one team is significantly ahead of its opponent (i.e. unlikely to lose under normal circumstances). Such is not the case in chess - you can have a better position on the board for 18 moves, say, and resign after the 19th. As in the following game.

Wall, Bill - Guest757810
PlayChess.com, 2017

1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+ 

4...Kxf7 5.Nxe5+ Nxe5 6.d4 Bxd4 7.Qxd4 d6 8.Nc3 Nf6 9.Bd2

There are 4 previous games in The Database with this move. White scored 0%. To be fair, all of the Jerome Gambiteers were playing higher-rated opponents.

9...c5

Also seen:

9...Be6 10.O-O-O Qe7 11.f4 Nc6 12.Qf2 Bg4 13.Rde1 Rhe8 14.h3 Bd7 15.g4 Kg8 16.e5 dxe5 17.fxe5 Nxg4 18.hxg4 Bxg4 19.Rhg1 (19.e6!? =) 19...Qe6 20.Re4 Bf5 21.Rh4 Bg6 22.Bf4 Nxe5 23.Bxe5 Qxe5 24.Rhg4 Qe3+ 25.Qxe3 Rxe3 26.Nd5 Re2 27.Nxc7? Rxc2+ 28.Kd1 Rd8+ 29.Ke1 Rxc7 30.a3 Rc1+ 31.Kf2 Rxg1 32.Rxg1 Rd2+ 33.Ke3 Rxb2 34.Rg4 Rb3+ 35.Kd2 Rxa3 36.Rb4 b6 White resigned, HauntedKnight - Molepolole, FICS, 2014;

9...Bg4 10.f3 Be6 11.O-O-O a6 12.g4?! Nxf3 13.Qf2 Bxg4 14.h3 Bh5 15.Be3 Re8 16.Rd5?! Nxd5 17.Nxd5 Rxe4 18.h4 Rxh4 19.Rxh4 Qxh4 20.Qf1 Qe1+ 21.Qxe1 Nxe1 22.Nxc7 Rc8 23.Nd5 Re8 24.b3 Bf3 25.c4 Bxd5 26.Kd2 Nf3+ 27.Kd3 Bxc4+ 28.bxc4 Ne5+ 29.Kd4 Nxc4 30.Bc1 b5 31.Kd5 h5 32.Kc6 h4 33.Kb7 h3 34.Bf4 Re4 35.Bg3 g5 36.Kxa6 Re3 37.Bh2 Ra3+ 38.Kxb5 Rxa2 39.Kxc4 Rxh2 40.Kd5 Rg2 41.Kxd6 h2 42.Ke5 h1=Q 43.Kd4 Qe1 44.Kd3 Rd2+ 45.Kc3 Qc1+ 46.Kb3 Rb2+ White resigned, HauntedKnight - BalintM, FICS, 2014;

9...Re8 10.O-O-O Nc6 11.Qe3 Bf5 12.f3 Na5? 13.Qg5 Bxe4 14.Qxa5 Bc6 15.Qb4 a5 16.Qc4+ d5 17.Qh4 b5 18.Bg5 b4 19.Nb1?! (19.Ne4!?) 19...h6 20.Bxf6 Qxf6 21.Qh5+ Qg6 22.Qh3 Re2 23.Nd2 Qxg2? 24.Qf5+ Kg8 25. Rhg1 Qf2 26.Qf4? (26.Qg6!?) 26...Rf8?! (26...Rae8!?) 27.Qxh6 Qd4 (27...Qxg1 28.Rxg1 Re1+ 29.Rxe1 gxh6 =) 28.Nb3 (28.Qxc6 Rxd2 29.Qe6+ Kh8 30.Qh3+ Kg8 31.Rxg7+ Kxg7 32.Qg3+ Kh7 33.Rxd2) 28...Qf6 29.Qxf6 Rxf6 30.Nd4 Rxh2 31.Nxc6 Rxc6 32.Rxd5? Rcxc2+ 33.Kd1 Rxb2 34.Rxa5 Rb1 checkmate, HauntedKnight - makeclean, FICS, 2013; and

9...Rf8 10.O-O-O Kg8 11.f4 Nc6 12.Qf2 Ng4 13.Qf1 Qf6 14.h3 Nh6 15.g4 Nf7 16.Nd5 Qd8 17.Bc3 Be6 18.Ne3?! Ng5!? 19.Qd3 Rxf4 20.a3 Nxe4 21.Rdf1 Rxf1+ 22.Rxf1 Nxc3 23.Qxc3 Qg5 24.Kb1 Qc5 25.Qd2 Qd4 26.Qf2 Re8 27.c3 Qd3+ 28.Ka1 Na5 29.Nf5 Nb3+ Black resigned, Darthnik - rsnoob, FICS, 2011. 

10.Qe3 Rf8 11.O-O-O Nc4 12.Qd3 Nxd2 13.Rxd2 Kg8

Black has castled-by-hand and has a piece for a pawn. If he keeps on making good moves, he will collect the full point.

14.e5 Ng4 15.exd6 Nxf2 16.Qd5+ Kh8 17.Re1 Bd7

Black has to keep his eye on White's pawn at d6, but with an extra piece for a pawn, he is doing well. Still, it is a complicated position. How should he continue?

18.Re7 Rf5 

Protecting the pawn at c5 and attacking the enemy Queen.

19.Qxb7 Black resigned

Wow! Black's Bishop is threatened, and the only square that even looks "safe" allows White's pawn to advance, 19...Be8 20.d7 Bf7 21.Qxa8!? Qxa8 22.d8/Q+ Qxd8 23.Rxd8+ Bg8 24.Ree8 - not so safe for the Bishop after all.

Best for Black seems to be giving up the Bishop for some play, 19...Qf8 20.Qxd7 Ng4, but after 21.Qe6 White will have centralized pieces and a couple of extra pawns, especially the dangerous on at d6.

Jerome Gambit: Choosing Better or Worse?

Sometimes a computer chess engine will recommend a line of play which is "objectively" better than an alternative that a human might choose. Yet there might be good reasons to choose the "worse" line of play, if it is tricky and has some traps. See the note to Black's 7th move.

On the other hand, sometimes "human" moves fall short, as well. See the note to Black's 9th move.

Wall, Bill - Guest902091
PlayChess.com, 2017

1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+ 

4...Kxf7 5.Nxe5+ Nxe5 6.d4 Nf6 

Black has so many ways to return some of the sacrificed material!

7.dxe5 Bxf2+ 

This move appears in 10 previous games appearing in The Database, with White scoring 70%. However, that is not the whole story.

Stockfish 8 recommends, instead 7...Qe7!?, which does not show up in any of the games in The Database. White does best to grab a second piece with 8.exf6, but then is faced with how to answer 8...Qxe4+.

Probably best is 9.Kf1 which we will come back to in a moment, after first examining the messy 9.Be3.

Black can respond to this latter move with 9...Bxe3 10.fxe3 Qxe3+ 11.Qe2 when the exchange of Queens would leave him a pawn up. Or he could go for 11...Qc1+ when 12.Kf2 is the offer of a Rook, i.e. 12...Qxh1, although it does not turn out well for the second player: 13.Qe7+ Kg6 14.Qxg7+ Kf5 15.Nc3 Qxh2 (taking the other Rook with 15...Qxa1 leads to mate) 16.Rf1 and Black's position is miserable as he has only 16...Qf4+ to help him extend the game, but 17.Kg1 Qxf1+ 18.Kxf1 still is hopeless. After visiting c1, Black's Queen should return to f4 with check, and eventually wind up a pawn ahead, again. 

Going back to the computer-recommended 9.Kf1, Black responds with 9...Qf5 and after 10.Qf3 Qxf3 11.gxf3 d5!? he is a bit better (better pawn structure, two Bishops, equally insecure Kings) even if White finds 12.h4!? to prevent ...Bh3+, forcing his King into the center where the enemy Rooks can roam.

By the way, Bill dispatched 7...Nxe4 quickly in a couple of games: Wall,B - NN, lichess.org, 2016 (1-0, 16) and Wall,B - TenAndOnly10, lichess.org, 2016 (1-0, 21)

8.Kxf2 Nxe4+ 9.Kg1 Rf8 

Preparing to castle-by-hand, a standard defensive plan in the Jerome Gambit; but overlooking the Queen check at d5 - odd, in that his Bishop sacrifice seemed aimed against that same eventuality.

Instead, 9...d5 would be about even, although Bill has a couple of wins against the move:  Wall,B - MyDrunkAccount, lichess.org, 2016 (1-0, 20) and Wall,B - Szachowski, playok.com, 2017 (1-0, 20). 

10.Qd5+ Ke8 11.Qxe4 Rf7 12.Nc3 d6 

This hurries Black's demise.

13.exd6+ Kf8 14.Qxh7 Qxd6 15.Qh8+ Ke7 16.Bg5+ Kd7 17.Rd1 Black resigned

Jerome Gambit: Paths Off of Paths

The Jerome Gambit is an opening that is clearly "off the beaten path". Yet, in pursuing it, players often find themselves further afield - and it is good to know about some of those side paths as well.

It is clear in the following game that the defender was not prepared to have the game "Jerome-ized".

Wall, Bill - Guest761989
PlayChess.com, 2017

1.e4 e5 2.Nf3 Nc6 3.Bc4 Na5 4.Bxf7+ 

With his third move Black imagines a quiet win of the "minor exchange" with 4.Bb3 Nxb3. He is too hopeful, as White has, instead, 4.Nxe5, and after 4...Nxc4 5.Nxc4 d5 6.exd5 Qxd5 7.Ne3 White is a pawn up, and Black has little to show for it.

However, players familiar with the Jerome Gambit will be tempted to play the Bishop sacrifice on move 4, giving Black a very un-quiet game.

4...Kxf7 5.Nxe5+ Ke6 6.d4 d6

White has two pawns for his sacrificed piece, but his play against Black's King gives him adequate compensation.

7.Qg4+ Ke7 8.Qg5+ Ke8 

Better is 8...Nf6, although after 9.Nf3 Nc6 10.e5 dxe5 11.dxe5 Kf7 12.exf6 Qxf6 White would have a small advantage.

9.Qxd8+ 

It is important to note that in the 4 earlier games in The Database, White had tried 9.Qh5+. Best play would continue 9...g6 10.Nxg6 Nf6 11.Qh4 hxg6 12.Qxh8 Nxe4 When White would have a Rook and two pawns against a Bishop and a Knight. This is a small edge for White, and I can see why Bill chose a different path: His active Knights will go after the slumbering Rooks in the corners!

9...Kxd8 10.Nf7+ Ke8 11.Nxh8 g6 12.Nc3 Bg7 

As indicated.

13.Nd5 c6 14.Nc7+ Black resigned

Black will lose his other Rook, and while he will gather in the Knight at h8, he will have a very difficult time catching the other one, leaving him down a couple of exchanges, plus a pawn or two.

The Delayed Pie-in-the-Face

The difference between a very strong counter in the Jerome Gambit at move 6 and an ineffective defensive line (the same move one tempo later) shows the problem with delaying an active riposte. 

Wall, Bill - Guest6766281
PlayChess.com, 2017

1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+

4...Kxf7 5.Nxe5+ Nxe5 6.d4 Bb6

Black can decide which piece to save, the Bishop or the Knight. The text is a reasonable line.

Of course, Black's strongest response is to ignore both of the pieces and play 6...Qh4!?, what I have referred to as a pie-in-the-face variation.

7.dxe5 Qh4

A delayed pie-in-the-face? Tossing pies is very much a part of slapstick humor - and timing is always critical in comedy.

This line has been faced by Jerome Gambiteers chessmanjeff, frizerkaHR, GOH, HauntedKnight, jfhumphrey, ndrwgn, stretto, Wall and yorgos. As a group, they have not been impressed.

8.Qf3+ Ke7

A slightly better retreat is 8...Ke8, e.g. 9.Nc3 (9.O-O Ne7 10.Nc3 Rf8 11.Qe2 g5 12.Be3 Ng6 13.Nd5 Kd8 14.Qd2 h6 15.Bxb6 axb6 16.Nxc7 Kxc7 17.Qd6+ Kd8 18.Qxg6 Re8 19.Qxb6+ Ke7 20.Qf6 checkmate, Wall,B - Guest2616286, PlayChess.com, 2017) 9...Ne7 (9...Bxf2+ 10.Qxf2 Qxf2+ 11.Kxf2 Nh6 12.Nd5 Ng4+ 13.Kg3 Kd8 14.Bg5+ Nf6 15.exf6 h6 16.fxg7+ Ke8 17.gxh8=Q+ Kf7 18.Rhf1+ Kg6 19.Qxh6 checkmate, Wall,B - Shillam, lichess.org, 2016) 10.g3 Qh3 11.Be3 d6 12.Bxb6 cxb6 13.exd6 Ng6 14.Nd5 Qd7 15.Nc7+ Kd8 16.Nxa8 Qxd6 17.Rd1 Ne5 18.Rxd6+ Ke7 19.Qc3 Kxd6 20.Qd4+ Ke6 21.Nc7+ Kf7 22.Qxe5 Rd8 23.Nb5 Bg4 24.Qf4+ Kg8 25.Qxg4 a6 26.Qe6+ Kh8 27.Nd6 Rb8 28.Nf7+ Kg8 29.Nh6+ Kh8 30.Qg8+ Rxg8 31.Nf7 checkmate, Wall,B - Itboss, lichess.org, 2016.

9.Nc3

Or 9.O-O Nh6 10.Nc3 Rf8 11.Nd5+ Kd8 12.Qxf8 checkmate, Wall,B - Guest2293428, PlayChess.com, 2017.

9...Nh6 10.Nd5+ Kd8

Once again, e8 is the better square for the King.

11.O-O

White also has the sneaky 11.h3!? threatening g2-g3, winning the Queen. Black could escape by returning a piece for two pawns, i.e. 11...Bxf2+ 12. Qxf2 Qxe4+ but White would be better.

For the record, 11.g3?! directly could be answered by 11...Qg4 and White does not gain much by exchanging Queens, while he would lose time if he movee his Queen to avoid the swap.

11...Bd4

Looking to grab the pawn at e5, but he is overlooking a few things. He could have maintained an even game with 11...c6 12.Nxb6 axb6.

12.g3 Qh3

Why was e8 the better square for Black's King to retreat to (as suggested in the note to move 10)?

Why was the Black Bishop's trip to d4 problematic (as mentioned in the note to move 11)?

Why couldn't Black retreat his Queen to g4 (as in the note to move 11), instead of h3?

13.Bg5+ Ke8 14.Nxc7 checkmate

Jerome Gambit: Skeptical Defense

Wall, Bill - Guest673244
PlayChess.com, 2017

1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.Bxf7+

4...Kxf7 5.Nxe5+ Nxe5 6.Qh5+

This is interesting. Bill usually plays 6.d4.

6...Ng6 7.Qd5+ Kf8 8.Qxc5+ d6 9.Qe3 Nf6 10.Nc3 Be6

Black develops a piece, and remains slightly better.

Despite the fact that his King is uncastled - and, therefore, he might feel more comfortable not opening up the  position - Black could have played 10...d5, and after 11.exd5 Kf7 12.0-0 Re8 13.Qd4 c6 he would have recovered his sacrificed pawn and his prospects would have been brighter than in the game.  

11.b3

Feeling adventurous, Bill bypasses the thematic 11.f4 and the reliable 11.d4 for a chance to put his Bishop on the long diagonal. We have seen this strategy for White before: why hurry? Let Black figure things out on his own.

11...Kf7 12.Bb2 Re8 13.O-O-O Kg8

Black has played steadily, developing his pieces and castling-by-hand.

This is just the kind of position that could feature opposite wing pawn storms, and White wastes no time in getting his started.

14.h4 Ng4

Black is not convinced. He could have tried 14...d5 or 14...h5 instead.

15.Qg3 Nf6 16.h5 Ne5

17.f4 Neg4 18.f5 Bf7 19.h6

This is what Bill is looking for: to open the h-file for his Rook, the g-file for his Queen, and the a1-h8 diagonal for his Bishop.

Black can stop White's plan, but he will have to give back some material.

19...gxh6

Instead, 19...d5 20.Rh4!? Qd6 21.Rxg4 Nxg4 22.Qxg4 Qxh6 23.Nxd5 would have helped on defense, although White would have two pawns for the exchange, and the initiative.

 20.Rxh6 Kf8 21.Qf4

21...Re5

Black remains skeptical, and it is easy to see why. Take, for example, his best defense, instead of the text: 21...Nxh6 22.Qxh6+
Ke7 His King appears to be escaping trouble, and he has an extra Rook (for two pawns). However, if White finds 23.Nb5!?, uncovering the Bishop, then after 23...Nh5 24.g4 Kd7 25.Qxh7 Re7 26.gxh5 he will have 3 pawns for the exchange - and the pawns will become more dangerous with each step forward.

22.d4 Re8

Come and get me, says Black.

23.e5

As you wish, says White.

23...Ng8

Tougher was 23...Qe7, when 24.Rxf6 Nxf6 25.Qh6+ Kg8 again leaves White a Rook down, but 26.Rh1 (better than 26.Rd3) Nh5 27.Ne4!? Qf8 28.Rxh5 Qxh6+ 29.Rxh6 again puts the first player in the better position, with two pawns for the exhange - and his pawns are ready to cause trouble.

24.Rxh7 Black resigned

I think Black was worn out and no longer skeptical. His Knight on g4 is hanging, his Bishop is threatened with e5-e6, and his King is not going to be able to escape, i.e. 24...Ke7 25.e6, etc.

Stockfish 8 gives as best 24...N4f6, when 25.exf6 Qxf6 26.Nb5!? would be too much for Black to deal with.