4. o-o is my usual choice, where a trade would be nice (into the muzio, but without sacrificing the piece!), but I never get one... just a long, painful, obnoxious pin that stalls my attack. All the while, black recovers the development -- 4. ... Nc6 or some such.

I'm in this mess in a couple of correspondence games online where I'm trying to kill that line with massive sacrifices. Like 5. Bxf7+ Kxf7 6. Ng5+ Qxg5 7. Rxf4+ etc. I really don't have high hopes for it though.

My preferred 5. o-o is in there, as is Nc3, which I simply do not understand. It doesn't deal with the problem at all. 5. Nc3 Bxf6 6. Qxf6 Qh4+ and white's kingside is in a complete mess. Obviously there's SOMETHING wrong for black in that line, because it hasn't been played in any of the grandmaster games in the database... but what?

With 5. o-o, the two games there aren't very enligntening. In M Zimmermann vs Schierling Eckhard, 1994 black succeeds in quieting the game down, then (move 17) bizarrely sacrifices a knight for totally inadequate compensation, at which point white simplifies and wins the endgame.

In the other, M Vann vs R Pullen, 1992, black simply plays like a class z player, blundering the exchange, then a piece, before resigning. The eventual Qd3 from white seems to work post castling, and white kept the initative, but only, I think, because black wasted time with 5. ... Nh6

Just for the record, by the way... 5. Bxf7+ Kxf7 6. Ng5+ Qxg5 7. Rxf4+ does indeed fail. Miserably. to 7. ... Nf6. Which, for some reason, I completely failed to calculate when I went on my little orgy of sacrifice. Ooops.

The problem with 5. d3, I think, is that it intices his queen to f6, where it seriously hinders queenside development, guards the f file, and is immune from harassment. I really don't like facing ... Qf6 anywhere in this opening, and encouraging it is never good.

Here's one possibility that makes me a little worried, for example. Black gives the pawn back to murder white's kingside pawn structure.

1. e4 e5 2. f4 exf4 3. Nf3 d6 4. Bc4 Bg4 5. d3 Qf6 6. Nc3 d5 7. exd5 Qe7+ 8. Ne2 Bxf3 9. gxf3 g5

now, granted, in that variation, black is less developed, and is weak in the center, but is only a move from castling queenside, whereas it's going to take a bunch of moves to get white's king anywhere safe. Then black can fianchetto the dark-squared bishop, and I really don't like white's position.

Hmm... I guess white can save that by 7. Bxd5 Bb4 8. o-o but then black picks up an extra pawn with the exchange on c3, and if white defends that with 8. Bd2, it also removes the threat to recapture the f pawn, functionally allowing black to not worry about it anymore and complete development by 8. ... Ne7 (although I think actually capturing the bishop would be a blunder because of 9. o-o Nxd5 10. Nxd5 and black loses the bishop on b4) then o-o, and it looks like black is quite happy with the position.

Who do you think stands better, refutor, after this line I've just kicked around, with best play, as far as I can determine for white? It looks ok for black, developmentally. White's center is stronger, but black has that extra pawn and looks to be able to hold onto it...

1. e4 e5 2. f4 Qh4+ 3. g3 Qe7 4. d4?!

I like 4. d4 as a defense to the pawn disordering of the early check because it defends the kingside pawn chain as well as threatens to get a good stomp going in the center... what do folks think? It only seems to appear in one game in the database...

1. e4 e5 2. f4 exf4 3. Nf3 g5 4. h4 g4 5. Ne5 d6 6. Nxg4 Be7.

Should you play d4 or d3? I don't see any advantage for White either way. d4 prevents a later Qc5 (after Qg5) but weakens the d-pawn. Gallagher sees good chances for White after

7. d3 Bxh4 8. Qf3 or Qe2 Bg3