Anti Ng5 Two Knights - Surely this must be losing?!

Well, that may be true.  I guess you can send Blatny a challenge and blow him away!  

I tried to find the 3...h6 game, and had no luck.  I know that it is referenced in the Italian Game and Evans Gambit book.  But I did find 4...h6 in the database, with good results for Black:

This is strange... Ive been playing 3 h6 ever since i was little because I didn't know about Be7 really... =P Are you saying its rubbish? xD I don't think its unsound, maybe not very good then compared to Be7... hmm time for a change in openings.

Ive met d4 a ton of times and survived head on. h6 isn't nothing you know. Infact I think ill stick with h6, maybe try Be7, for the sake of experience =)

 Wilkes-Barre...?! Is there a refutation?

I know the refutation actually. I know :D Google search.

http://www.chesscafe.com/text/kaiss23.pdf hehe this guy goes along with Kf1 exd5 etc! :D He also says h3 is ok!!! woot my analysis is so good in just a few minutes.

however he puts forward d6!! as good. Lets see:

Shadow, Conquistador's comment is relevant in terms of the database, but I remember reading Blatny's use of 3...h6?! in the Two Knights.  Maybe you can post a game or two of your handling of it - the main line in context, which would be the most testing obviously, is 4. d4.  I personally prefer the Hungarian 3...Be7, but I am slow to push 3...h6 to the wolves, seeing that a greater than I has tried it. 

To shadowknight

Not sure where your refutation is in the first diagram since that is horribly played by black.  Your move numbering is kinda weird too. 

Main line with 5.Nxf7 is 4...Bc5 5.Nxf7 Bxf2+! 6.Kf1 Qe7 7.Nxh8 d5 8.exd5 Nd4 9.d6 Qxd6 10.Nf7 Qe7!= given by John Jerz as an improvement over 10...Qc5.

There is no known refutation yet I don't think. Black is not playing horribly, I followed some analysis on google search, read the article *huff*. Actually the first diagram was following NM cobbiepot or something (sorry!), and the only move out of "order" was perhaps h3 by myself as white. So black is not playing horribly... dunno where u see that in, its the same as in the 2nd diagram?

That diagram above is NOT a refutation.

Qe7 is better than Qc5? hmm well I have no idea.

Actually there are a number of refutations of sorts, but there have been many improvements made to the theory of the Traxler.

Your first diagram has

1.e4 e5 2.Nf3 Nc6 3.Bc4 Nf6 4.Ng5 Bc5 5.Nxf7 Qe7? which is a known error among beginners.  White follows 6.Nxh8 Bxf2+ 7.Kxf2 Qc5+ 8.d4 Qxc4 9.d5 Nd4 10.c3 Nb5 11.Rf1 Nd6 12.Nd2 Qc5+ 13.Ke1 and black is in quite a bit of trouble with his kingside under threat to be blown apart.

huh so if not Qe7, whats the move?

5...Bxf2+

oh yes your right on the first one sorry. The 2nd one is fine though, Bxf2+ is played first.

Just to note, when Qe7 Nxh8 Bxf2 Kxf2/Kf1 is played, dont you end up in the same position?

eg, 5. Bxf2+ Kxf2 THEN Qe7, whats the difference?

The difference is that 1.e4 e5 2.Nf3 Nc6 3.Bc4 Nf6 4.Ng5 Bc5 5.Nxf7 Qe7? allows white to gain a tempo and enter into a much more favorable variation with best play.  Of course, white could still blunder and transpose back into the main lines. 

But the line 1.e4 e5 2.Nf3 Nc6 3.Bc4 Nf6 4.Ng5 Bc5 5.Nxf7 Bxf2+! is a much better handling of the Traxler as it gains a tempo forcing white to decide his king placement before moving the queen.  This gives black the advantage of having a much more effective queen placement.

Ah shoot I just lost my diagram I was going to post with all the analysis.

I hear you now.

A much better way to play the Traxler as Black is not playing it at all.