Fullwaza.com

[Event "5AM blitz game"]
[Site "GameKnot.com"]
[Date "2007.01.05"]
[Round "-"]
[White "mbingham"]
[Black "daj36"]
[Result "0-1"]
[JsCom "startply 0"]
1. e4
{
I thought this was a fantastic blitz game.  It was a new line for me --
one I'll have to remember if I want to keep playing with the fried liver.
}
1... e5 2. Nf3 Nc6 3. Bc4 Nf6 4. Ng5 Bc5
{
d5 is what I usually see here.  Brent Inman and Tom Denton were kind enough to take a minute  to show me that 1.  This isn't the Fried Liver.  Bc5 has negated that.  This is the Wilkes-Barre or Traxler variation. and 2. that it's ugly ugly ugly for white.  GM Nick DeFirmian states that there's no known refutation for it.  Black can ignore the threat to f7 and if white takes f7, he can expect to hurt.
}
5. Nxf7
{
Any computer looks at this position and reagards white as won (unless it uses an openings database by default), yet a positional search on chessgames.com shows black overwhelmingly wins these .
}
5... Bxf2+ 6. Kxf2 Nxe4+ 7. Ke3 Qh4 8. Nxh8 Qf4+ 9. Ke2 Qf2+ 10. Kd3 Nb4+
11. Kxe4 Qf4#
{
No I haven't gone through any new lines, I'm just looking over this with new eyes, trying to let it sink in.  The first time I went through this, I rolled my eyes at my inability to see that the time-waster Nxh8 was fatal, but I think what I have learn about this is why time favors black so well in these lines...  I can state it pretty clearly for the line I played.  Kxf2 allows black to sieze a tempo that brings the queen out of threat and immediately into play.  That, in conjunction with the readily-cooperating knights amounts to a powerful attack.  Add to that the rook which will be locked behind the king and there appears to be compensation more-or-less immediatetly (because that king really shoudl go to g1 -- rooks be damned.
So, yeah.  I'm gonna have to find something besides Ng5 after Nf6.
}
0-1