Some points relating to the analysis of both MNb and Craig Evans:

In Kaissiber Gutman and Bücker recommend interposing (1.e4 e5 2.Nf3 Nc6 3.Bc4 Bc5 4.0-0 d6) 5.h3, before continuing in "delayed Evans" style with b4 etc (the traditional Scotch Gambit lines that arise after 5.c3 Bg4 6.d4 exd4 are absolutely fine for Black).  They think it is mildly advantageous for White, and I'm yet to see a comeback from Black's perspective.

Re. 1.e4 e5 2.Nf3 Nc6 3.d4 exd4 4.Bc4 Be7 5.c3 dxc3 6.Qd5 Nh6 7.Bxh6 0-0, I'm not convinced that White gets any advantage with 8.Nxc3 gxh6 9.Qh5, but I think the 6.Qd5 line still leads to a nice edge for White after after 8.Bxg7! Kxg7 9.Nxc3, as Black's two bishops are outweighed by the exposed king.  But 6.Nxc3 is indeed a good alternative for White.

I think White has a few improvements over Glek's play in the line 4...d6 5.c3 Ne5 6.Nxe5 dxe5 7.0-0, though Black can avoid ending up in a Göring Gambit type position with 5...Nf6 6.Ng5 (probably best, envisaging transpositions back to a line of the Göring Gambit that arises after 5...dxc3 6.Nxc3 Nf6 7.Ng5!?) 6...d5 7.exd5 Qe7+, which is probably good enough for equality.

Re. the Two Knights with 5.0-0 Nxe4 6.Re1 d5 7.Bxd5 Qxd5 8.Nc3 (which I sometimes play with both colours), I always preferred 8...Qa5 over 8...Qh5 as Black due to the sterility of the 8...Qh5 9.Nxe4 Be6 10.Bg5 Bd6 lines.  But the idea 11.Nxd6+ cxd6 12.Bf4 Qd5 13.c3 Qf5!? looks very interesting- it should be equal, but the equality isn't necessarily any less dynamic than in the line 8...Qa5 9.Nxe4 Be6 10.Neg5 0-0-0 11.Nxe6 fxe6 12.Rxe6 Bd6.  I think in the line 11.c4, John Emms's equalising suggestion 11...h6 gives Black more chances of spicing up the play (e.g. 12.Nxd6+ cxd6 13.Bf4 Qc5 improves on the line 11.Nxd6+ cxd6 12.Bf4 Qc5 as White cannot play c2-c3) than the simple equaliser 11...0-0 12.c5 Be5 13.Nxe5 etc, and of course 11.Bf6 Bxh2+ etc. leads to -/+.

One possible alternative issue is 10.Nxd4 Qxd1 11.Rxd1.  Black appears to get an edge in all lines, but I'm doubtful as to whether it's enough to be able to hope to grind out a win against good opposition.  The main line continues 11...0-0-0 12.Be3 Nxd4 13.Bxd4 Bf5 14.Ng5 Bg6 with Black having the bishop-pair.  14.f3 (which I haven't seen mentioned anywhere) is also possible, when White gets an isolated e-pawn in a rook ending following 14...Bxe4 15.fxe4 Rxd4 16.Rxd4 Bc5 (or Black can consider waiting with 14...b6 15.Bc3 Rxd1+ 16.Rxd1 f6, maintaining the bishop pair).  Black stands better, but again, is it enough to have much hope of grinding out a win?  If the answer here is "yes" in both cases then we have another argument for preferring 8...Qh5 to 8...Qa5.

As White in those lines I am attracted to Stefan Bücker's gambit idea (which works after both 8...Qa5 and 8...Qh5) 9.Nxe4 Be6 10.Neg5 0-0-0 11.Nxe6 fxe6 12.Bg5 with the idea 12...Re8 13.Re4, which appears good enough for dynamic equality regardless of whether Black accepts or declines the pawn offer, though I have yet to get the opportunity to use it in a game.

CraigEvans wrote on 01/02/12 at 14:14:43:

Wasn't 9.Re1+ and only after Be7 10.Bg5 more precise? This avoids 9.Bg5 f6 10.Re1+ Kf7. Or is this good after White after 11.Bh6 as indicated by Heyken and Fette 25 years ago?

CraigEvans wrote on 01/02/12 at 14:14:43:

When playing a combination of d4 and Bc4 you can't avoid this.

CraigEvans wrote on 01/02/12 at 14:14:43:

That's not really possible.

CraigEvans wrote on 01/02/12 at 14:14:43:

There isn't - or you already would have known about it.
5...Be7 is basically a Hungarian with a (too) early ...exd4. In this particular case White can try 6.e5. After 6.O-O Black must transpose as Nxe4?! 7.Nf5 is excellent.
More precise then is 5...d6 and I don't trust 6.Ng5 Ne5 7.Bb3 h6 8.f4 hxg5 9.fxe5 dxe5 10.Bg5 Bc5.

CraigEvans wrote on 01/02/12 at 14:14:43:

There is the Evans Gambit Delayed. but a) it isn't in your repertoire and b) 5...Bxb4 6.c3 Ba5 is considered the weaker version these days.
Another option is 5.c3 and 6.b4 with a Bird Variation Delayed.

CraigEvans wrote on 01/02/12 at 14:14:43:

The only way to counter this point is to include some stuff you might not like. But there is the Scotch Gambit:
1.e4 e5 2.Nf3 Nc6 3.d4 exd4 4.Bc4
a) 4...Nf6 5.O-O and you are happy.
b1) 4...Bc5 5.O-O Bg4!
b2) 5.c3 Nf6 6.O-O and you regret having played c2-c3 iso e4-e5.
b3) 5.b4!? has been played by Capablanca (informal game), Marshall and Minic and might be worth a try.

Another advantage is that you avoid the Closed Hungarian (3.Bc4 Be7 not ...exd4). 3.d4 exd4 4.Bc4 d6/Be7 5.Nxd4 or 5.O-O and 6.Nxd4.
There is some gambit stuff here, but it's not clear if it's playable:
a) 4...d6 5.c3 Ne5 (dxc3 6.Nxc3 is the Göring Gambit) 6.Nxe5 dxe5 7.O-O;
b) 4...Nf6 5.O-O see above;
c) 4...Be7 5.c3 dxc3 6.Qd5 Nh6 7.Bxh6 O-O! and Black is quite OK with the Pair of Bishops. But 6.Nxc3 d6 transposes to a line of the Göring Gambit that is thought fine for White.

CraigEvans wrote on 01/02/12 at 14:14:43:

That's somewhat a pity. Now the smart transposition 2.d4 exd4 3.Nf3 is not available because of Bb4+. And 3.Bc4 d6 4.Nf3 Nf6 5.O-O Be7 (but there won't be a Larsen as ...g6 is met with Ng5) 6.Re1 is not the Qxd4 line.
Reason why I bring this up is of course 2.d4 exd4 3.Bc4 Nf6 4.Nf3 when Nf6 just transposes and Nxe4 is the interesting Urussov Gambit. 4.Nf3 Bc5 5.O-O Nf6 6.e5 and you are happy, but 5...d6 6.c3 is like the Scotch Gambit Declined (Bg4!?). At the other hand 5.e5 d5 is similar to your beloved ML.
So perhaps this is still worth considering.

SWJediknight wrote on 10/23/11 at 10:24:01:


Just two points here - 1) Avoid Pinski's recommendation of (10.Bg5) Bd6 11.Nxd6+ cxd6 12.Bf4 Qc5?! 13.c3 dxc3 14.Bxd6 Qa5? due to the improvement 15.b4! which he misses entirely. White is clearly better here.

Secondly, after 12...Qd5 13.c3, the move to really surprise your opponent is 13...Qf5!? - I've done quite a bit of home analysis on this and I think it is black's best chance of mixing things up. White tends to opt between 14.cxd4 Qxf4 15.d5 and 14.Bxd6!?! O-O-O, both giving sharper endgames than the sterile positions after 13...Rc8, where white might get a niggling edge. If you really want to play for a win as black, some work here might be fruitful (but I won't give away my secrets... at any rate, with correct play white is equal here too).

I also agree that 10...Bb4 gives white a chance of a small niggle in the endgame... though again it is probably nothing.

Downward,
      Marin put out an update that covered the 1.e4 e5 2.Bc4 move order. it can be found on the download section of the Quality Chess website I think.

Markovich wrote on 10/23/11 at 13:42:54:


Your advice is quite sound. However white can also go for the Urusov gambit 1.e4 e5 2.Bc4 Nf6 3.d4 exd4 4.Nf3.

See further discussion of these issues:
http://www.rci.rutgers.edu/~goeller/urusov/bishops/index.html
and
http://www.rci.rutgers.edu/~goeller/urusov/gambit/index.html

Edit: But I suppose, as Mnb pointed out earlier, that of course the solution here is to play 4...Nc6 and tranpose to Two Knights Defence 

Markovich wrote on 10/23/11 at 02:05:32:

I was going to leave this alone, but it just kept gnawing at me.

In this day of database information and complete game scores from U12 tournaments around the world, yes. I can say quite confidently that not every +2700 player today played the Two Knights as juniors. You don't need to yield to my extensive knowledge, you could look up the juvenalia of today's young players yourself.

From about 1950-1975(ish), your advice would have seemed orthodox. Today's juniors study with computers and come up with their own lines. There's actually quite a wide literature on how computers have revolutionised learning for students.

Again, I agree with Markovich that the Two Knights is probably preferable to the Italian game. MNb's, Willempie's,  and others' comments have pretty much reified that position.

Markovich wrote on 10/22/11 at 20:34:07:

I dont disagree, but this also goes for the Giuoco. I would even go as far as stating that you can't play the TKD properly without having some schooling in the Giuoco and vice versa. Any schooling should include both openings as white and black imo.

My personal opinion is that the Giouco is a better choice theorywise, but also has less chance for the full point.

In the Giuoco I could suggest the following lines (I've played them myself and on occasion still do):
-Evans Gambit. Just accept it with Ba5 and keep in mind the Lasker defense ideas.
-d3 and Nc3 (Canal), Play Na5 at an appropriate time and you're better than equal
-d3 and c3, get a double espresso and imagine you're playing a closed (Moller) Ruy. The main plus you have is that that bishop is on d4 so if you can push d5 you gain a tempo.
-4.c3 systems. This is the real meat. The lines with Bd2 are equal, but you need to know how to  play these positions. The lines with Nc3 are amongst the most fun in chess, play the Lasker defense. Few moves to remember and it annoys white to no end.

In the TKD I agree with what others wrote, though my personeal preference is the Traxler (it is never too late to correct a mistake ) . Studying that line will do nothing but good for your tactical sight.