"Ng3+ 19.Kg1 Ng3" I don't know what that second Ng3 is, but at any rate 19... Nde2+ is equal according to Fritz.
Traxler discussion

Oops, I typed that wrong
12.Qd2 Bh4 13.Nc3 b5 14.b4 Qb6 15.Nxe5 bxc4 16.Nxc4 Qe6 17.h3 Nh5 18.Kg1 Ng3+ 19.Kh2 Nxh1 20.Kxh1 Bxh3 21.Bb2 Bf5 22.Qf4

I used to play this a lot back in the late 1980s. I had a theoretical OTB discussion going on with someone. I owned the book "The Real American Wilkes-Barre", a book by Ken Williams of the Wilkes-Barre Chess Club! I wish I hadn't sold it now since that book, of about 50 pages, is now selling for $60! However, I did write down a few lines out of it in my large analysis book. Here they are...
7.Ke3!? Qh4 8.g3 Ng3 9.hg Qd4 10.Kf3 d5 11.Be2! and was listed as unclear. Apparently I recall reading that Williams had talked with several people about the move 11.Be2! and no one could find anything wrong with it.
Ok, I decided to check the above on Rybka which seems to like 11.Kg2 with the possibility of 12.Be2 or 12.Rh4. Rybka lists 11.Be2! as its third choice.
Also 7.Ke3!? Qe7 8.c3! (8.Nh8 d5 9.Qd5 g6 10.Ng6 Qc5 11.Ke2 Qf2 12.Kd1 Nf6 13.Rf1! Qf1 14.Bf1 Nh5) Nd4! (8...Qc5?! 9.d4 ed 10.cd Qe7 11.Kf3 +/-) 9.Rf1! is now listed as unclear as well as the try 9.Qh5!?
In line above "(8...Qc5?! 9.d4 ed 10.cd Qe7 11.Kf3 +/-)" it seems 11.Nh8 can be played instead. After 11...Nc3 12.Kd2 maybe the pin 13.Re1 was overlooked.
I also have some other analysis with a few other lines but I don't have them with me, but they are not in the 7.Ke3.

Ng3 is not a check, another notation mistake. One last try at this...
12.Qd2 Bh4 13.Nc3 b5 14.b4 Qb6 15.Nxe5 bxc4 16.Nxc4 Qe6 17.h3 Nh5 18.Kg1 Ng3 19.Kh2 Nxh1 20.Kxh1 Bxh3 21.Bb2 Bf5 22.Qf4

I have never seen 11.Be2!? before in the Traxler with 7.Ke3. My knowledge of theory says that the main move is 11.Rh4
OK, now it makes sense ;) 18... Nf4 19. Kh2 Nxg2 20. Kxg2 Bf3+ 21. Kh2 Bxh1 22. Qe3 Bc6 is the best I can do right now, with Black a pawn down and only the Bishop pair for compensation. I'm really not satisfied with that, but I think I'm going to stop analyzing for now.
Yeah, I was having this problem too. A group would be a good idea- we could also play games against each other in this line and have topics for critical lines. Is it possible to have vote chess inside a group (team 1 has players 1, 3, 5, etc.)? I'm not a member of any group so I don't know, but we could even have a vote chess game in the Traxler.

I think more people need to help us with these lines, or we should just focus on a few lines in particular, analyse them out, then move to the next variation. I think we have worked out the 10.Nf7 Qc5 lines and there is only one way for white to have a solid advantage.
1.e4 e5 2.Nf3 Nc6 3.Bc4 Nf6 4.Ng5 Bc5 5.Nxf7 Bxf2+ 6.Kf1 Qe7 7.Nxh8 d5 8.exd5 Nd4 9.d6 Qxd6 10.Nf7 Qc5 11.d3 Bg4 12.Qd2 Bh4 13.Nc3 b5 14.b4 Qb6 15.Nxe5 bxc4 16.Nxc4 Qe6 17.h3 Nh5 18.Kg1 Nf4 19.Kh2 Bxh1 20.Qe3 Bc6 21.Qxd4
This is the critical line for the entire Nxf7 variation.
Can anyone in the community who can analyse that variation to find any improvements?
In the meantime, lets just focus on the variation 10.Nf7 Qe7 as that seems to be the last crucial variation. We can avoid information overload that way.

Can any Traxler player check the above variation? Post if you find any novelties.
Everyone's imput is appreciated.
The more people we have posting, the more conclusive this thread becomes.
I'm not sure making groups and having vote matches would really facilitate the work. The information overload (that I fully acknowledge) is probably not so much caused by the structure of our discussion, but rather by the overwhelming amount of, well, information. The only thing I can think of that might really help is posting a summary. Maybe I'll do that tomorrow (I have saved most the work done so far on my computer).
I've been looking at Jerz's analysis of 10.Nf7 Qe7 and it looks very promising and exceedingly difficult. Here's one line he looks at, supplemented by some lines had Fritz examine:
11.d3 Bg4 12.Qd2 Bh4 13.Nc3 b5!! (an amazing move, I think):
a) 14.Nxb5 Ne4! 15.dxe4 Qf6+ 16.Kg1 Ne2+ 17.Qxe2 Bxe2 18.Nxc7+ Kd7 19.Be3 Bxc4! 20.Nxa8 Qxf7 and black has a very strong initiative.
b) 14.Nxe5! 0-0-0! (what a position. There are many possibilities, I'll show three, I'm not sure which is best):
b1) 15.Nxg4 Rf8 16.Ne5 Ne4+ 17.Qf4 Rxf4+ 18.Bxf4 Nd2+ 19.Bxd2 Qxe5 20.g3 bxc4 21.Kg2 Bf6 and black's initiative seems sufficient for the slight material loss.
b2) 15.Qe3 Nxc2 16.Qxa7 Qxe5 17.Qa6+ Kd7 18.Qxb5+ Qxb5 19.Bxb5+ c6 20.Ba4 Rf8 21.Bxc2 Nh5+ 22.Kg1 Bf2+ with perpetual check.
b3) 15.Bxb5 Qxe5 16.Qe3 Qf5+ 17.Qf4 Qe6 18.Bd2? Nh5 19.Bc4 Be2+ 20.Kg1 Qb6 21.Be6+ Kb8! 22.Qe3 Bg5! and wins.
Edit: I missed your last post there, Gonnosuke. You may be right that dividing the work is a useful idea. Also,as I said, summarizing our conclusions is useful, for one thing it allows people to take issue with them.
Agreed with Gonnosuke, and about to take a look at mathijs' lines.
Wow, this is some deep stuff. Fritz likes 15.Nf3 bxc4 16.Nxh4 Rf8 17.Nf3 Nxf3 18.gxf3 Bxf3 19. Qe3 Qxe3 20. Bxe3. White's up a pawn in an endgame R+N+B with the Bishops on opposite colors. A key move seems to be Black's 15th, there may be something else there.
Yes, you're quite right Gonnosuke, it's agood idea to split up the work. I wasn't really thinking straight anymore (wonder why). I suggest that we have seperate discussions on the following:
- 5.Nxf7,6.Kxf2, 7.Kg1/other moves
- 5.Nxf7,6.Kxf2, 7.Ke3
- 5.Nxf7,6.Kf1, white doesn't play d6
- 5.Nxf7,6.Kf1, 9.d6 cxd6
- 5.Nxf7,6,Kf1, 9.d6 Qxd6 10.other moves than Nf7
- 5.Nxf7,6,Kf1, 9.d6 Qxd6 10.Nf7
- 5.Bxf7+, 6.Bb3
- 5.Bxf7+, 6.Bd5
- 5.d4
Does that seem like a reasonable division.
Sure enough, I tried Fritz's second favorite on move 15 and it led to a fairly even position: 15... Nxf3 16.Qe3 Qxe3 17.Bxe3 bxc4 18.h3 Re8. Interesting is 19.hxg4 Rxe3 20.gxf3 Rxf3+ 21.Ke2 Rf2+ where despite being an exchange down, Fritz evaluates it as equal because one of White's Rooks is stuck behind the King in the corner and Black's kingside passers give massive play.
Mister Jerz prefers 10.Nf7 Qe7!, a line he analyses extensively. I haven't looked at it yet, I'm tired too. Have a nice vacation SirDavid.