I have to say, there are very few resources out there on the Sarratt variation of the Scotch Gabmit. I find it pretty intriguing so I've been trying to play it and appreciate your posts on the matter as I've come back to it a few times now.
I recently played the computer and got a chance to try it out. Now, I won this game, but REALLY shouldn't have as the computer blundered taking a free rook on move 22.
We were following theory for a little bit (I think from the Morphy game), and I'll admit that it was a lot of fun to continually attack.
However, the computer analysis (my blunder aside) was a bit disheartening. I felt like I was doing better than the analysis showed.
For example, the engine hated move 14 (which I thought was thematic), and wanted c3 played instead. It would have solved my issues with the knight, and if he took it then the analysis swings back in my favor.
After the moves 1.e4 e5 2.Nf3 Nc6 3.d4 ed 4.Bc4 Bc5 5.Ng5 (Sarratt Var./Attack) Black needs to find a way to avoid absolute catastrophe.
In 1818 the game Sarrat-N.N. continued 5...Ne5 (defending f7, but with an already developed piece) 6.Nxf7 Nxf7 7.Bxf7+ Kxf7 8.Qh5+ g6 9.Qxc5 Resigns.
Resigns?!? Not the N.N.s I've seen! Perhaps N.N. was actually a known player, whom Sarratt protected from embarrassment by inserting N.N. instead of his real name. In support of N.N., however, resignation is contemplatable. White is bound to gain the d or c pawn regardless of how Black replies. Meanwhile, the BK needs to find a place to hide and castling is not an option.
Now, let's fast-forward in time to Paris, 1821. The chess gambler John Miles Cochrane faced fellow excellent player Alexandre Deschapelles (who has a Whist/Bridge ploy named after him, called the Deschapelles Coup, by the way.) Both gentleman were probably familiar with the Sarratt Variation, which would appeal to Cochrane's gambling style of chess. So, Deschapelles had a little surprise in store after White played 6.Bxf7+ first, instead of Sarratt's 6.Nxf7.
The game went 1.e4 e5 2.Nf3 Nc6 3.d4 exd4 4.Bc4 Bc5 5.Ng5 Ne5 6.Bxf7+ Nxf7 7.Nxf7 Bb4+.
This is just the sort of move that can cause White to panic and stumble, especially gambling players like Cochrane, who are great attackers, but who may slip when the initiative passes to the opponent. White was all prepared to take a B on c5 with his Q and win a pawn to boot. Instead, the doomed bishop goes "desperado", setting up shop on b4, with check no less.
The game continued 8.c3 dxc3 9.bxc3 Bxc3+ 10.Nxc3
Here is the problem: White has 2 pieces in play and Black none. Plus, the a1-h8 diagonal is easily occupied with strong effect by the Bc1. For this reason, Black might try 9...Qf6 (instead of 9...Bxc3+) to complicate things, but White has a big advantage.
The finish to the game is of no consequence to the discussion, but is given to satisfy the curious: 11.Qd5+ Kf8 12.Ba3+ d6 13.e5 Qg5 14.exd6 Qxd5
15.dxc7+ Kf7 16.Nxd5 Bd7 17.O-O Rc8 18.Bd6 Ke6 19.Bg3 Bc6 20.Rad1 Bxd5 21.Rfe1+ Kf6
22.Rxd5 Nh6 23.Ra5 Nf5 24.Rc5 Nxg3 25.hxg3 Kf7 26.Rd1 Rhe8 27.Rd6 Re7 28.Rf5+ Ke8
29.Rd8+ Rxd8 30.Rf8+ Kxf8 31.cxd8=Q+ 1-0
So, it became apparent that 7...Bb4+ wasn't the answer and the flaw must lie with 5...Ne5. After all, Black was trying to defend by moving one of his pieces twice, which cancelled out White moving the Ng5 twice. So, players started to try 5...Nh6 instead to defend against the coming catastrophe on f7. Then the sequence went 6.Nxf7 Nxf7 7.Bxf7+ Kxf7 8.Qh5 g6 9.Qxc5 [Note: 9.Qd5+ is sometimes played before taking the Bc5] d6 and Black was okay, since the pawn is defended by the Nc6 and White must make yet another Q move.
However, people who play a move like 5.Ng5 are not likely to give up so easily, just because logic tells them they should. So, they began to think: "Sure, our position sucks on the surface; but, we are attack-at-all-cost people! Aha, the Kf7 looks shaky. Let's regroup and attack the palace."
By 1850 a Sarratt fan, Ilia S. Shumov, could defeat a player of reknown, Carl von Jaenisch, with this plan. That game went 1.e4 e5 2.Nf3 Nc6 3.d4 exd4 4.Bc4 Bc5 5.Ng5 Nh6 6.Nxf7 Nxf7 7.Bxf7+ Kxf7 8.Qh5+ g6 9.Qxc5 d6 10.Qb5 Re8 11.O-O Rxe4 12.Qd5+ Re6 13.Bg5 Qe8 14.f4 Kg7
15.f5 Re5 16.f6+ Kh8 17.f7 Kg7 18.fxe8=N+ Rxe8 19.Qf7+ Kh8 20.Bf6+ 1-0
So there it was for all to see: it's okay to move a piece twice in the first half-dozen moves, then top it off with 3 moves by the Q by move 10, if the opponent's K is somewhat exposed.
Then along came a guy from New Orleans named Paul Morphy. By his understanding, you can't move pieces twice in the opening or go flitting about with the queen early just to get a sniff of an attack, without paying a heavy penalty. To prove his point he let himself get drawn into the same "hopeless" position as the game Shumov-Jaenisch against Kennicott through White's 15th move.
From this position Morphy could see he held several trumps: centralized pieces, an extra pawn, and far better chances to get his remaining pieces in play before White could do likewise. The standard response to a wing attack is play in the center, but for tactical reasons the obvious 15...Re5 loses, as Jaenisch found out. So, Morphy eliminates the deadly assassin on f5 and finds a tactical defense using all pieces available, even the Bc8 sitting on its home square.
The game continued 15...gf 16.Qxf5 (16.Rxf5 Re8+ 17.Rf1 Rxf1+ -- 17...Re5 doesn't work, because the Bg5 is taboo after 18.Qf3 -- 18.Kxf1 Be6 and 19.Qf3 Qg6 or 19.Qe4 Qf7+ should win easily) Rg6! (Obvious and necessary, but still very strong. The rook protects f6 against a Q check, discovers an attack against the Q, and "x-rays" the Bg5 and Pg2.) 17.Bf6+ Kg8 18.Qf4 ...
White tries to maintain some semblance of attacking chances by keeping coordinated forces on the k-side. What he will do about his sleeping troops on the q-side remains an open question. At the moment, though, Black's Pd4, which keeps the sleeping Nb1 from active duty on c3, is under assault. Too complicated are plans starting with 18...Rg4, which has more defense than offense about it. Besides, with the better position you wouldn't want complications. Let the opponent complicate, if he can.
Morphy finishes in Morphy-style, by completing his development with a combination to a winning ending.
18...Bh3 19.Bg5 Qe3+ 20.Qxe3 de 21.gh Rxg5+ 22.Kh1 e7 23.Re1 Nd4 24.Na3 Re8 0-1
So, Morphy vindicated opening theory and allowed the good chess folk to once again sleep at night, thinking everything was once again alright in the chess world.
As I pointed out earlier, however, it's hard to change the minds of 5.Ng5 fans, who could not sleep at all after such a drubbing by Morphy. They started working -- some had already been working -- on a new idea: don't take on f7 immediately, but reinforce the attack on it with Qd1-h5.
That's Part III.