Can someone explain to me the logic behind the following popular line in the Caro-Kann:

1. e4 c6
2. d4 d5
3. Nc3 dxe4
4. Nxe4 Bf5
5. Ng3 Bg6
6. h4

Why h4? Why waste the pawn move?
Why not just play Nf3 and then Bd3?

Thanks in advance for anyone who can shed some light on this.

I was in a little hurry when I wrote that message, I want to elaborate further on what I don't understand regarding that move:

1) It damages the pawn structure.
2) If you'll decide to castle on the king side later, it seems problematic.
3) The move doesn't seem to have much merit besides threatening to attack the bishop, which is probably going to be exchanged without much trouble a few moves later (even without playing that move).

So what do you think?

The critical move is h4. And the pawn move isn't wasted.

The Nf3-Bd3 plan you mention is common as well. It's fine but according to current theory you should be playing h4. There is a threat to this move, quite simply to trap the bishop which is very often a target in the Caro-Kann. The main line continues 6. ... h6 7. Nf3 Nd7 8. h5 Bh7 9. Bd3 Bxd3 10. Qxd3 after which white gets a tidy position where black is still restricted. As to your 2nd point, it doesn't matter much, if after the said continuation 10. ... Ngf6 11. Bf4 e6 white castles long.
The pawn chain is a bit compromised true enough, but black won't be taking that anytime soon. I've often played into the main lines here and I've never had to worry about the pawn on h5 as most play revolves around the center and queenside (black wants to castle there as well).

You know, with all the discussion of this line in this thread and in my thread about the CK, I think I've learned more about this particular line than what I looked up about the exchange variation. Maybe I'll try this against the Caro Kann instead of the exchange if it comes up at my tournament this weekend.

--Fromper

Yes this is a good idea but in case you dont already know you should be aware that black might decide to play the following line: 1.e4 c6 2.d4 d5 3.Nc3 dxe4 4.Nxe4 and now instead of Bf5 black might play 4...Nd7 if so then this is the line 5.Ng5 now it is tempting to play h6 but it's a mistake because after h6 then 6.Ne6(if fxe6 then 7.Qh5+ g6 8.Qxg6#) 6...Qb6 7.Nxf8 Nxf8 8.Nf3 with the advantage getting back to the right line 5...Ngf6 6.Bd3 e6 (if 6...h6 then 7.Ne6 fxe6 and 8.Bg6#) 7.N1f3 Bd6 8.Qe2 h6 9.Ne4 Nxe4 10.Qxe4 Nf6 11.Qe2 b6 12.Bd2 Bb7 13.0-0-0 Qc7 if 13...o-o then g4! opening lines on the enemy king
hope that helps
-seguin99

I frown upon early Ng5 maneuvers in the Two Knights, the same in the Nimzowitsch system of the Caro-Kann. I know it's a main line and I also know us mortals shouldn't bother with such unprincipled chess. The trap is well-known however, no Nd7-CK advocate will fall for it after having seen it once, making the knight move rather wasted in the final analysis and.

Do trust your CK opponent to be able to defend himself.

Ng5 is a very good line. It's the best try for an advantage there.

Also, 10...Nf6 is very bad. It gains a tempo, but basically hands over the e5 square for white. After this it's very hard to avoid being squeezed. I believe the main move is 10...Qc7 11 Qg4 Kf8. The idea is that eventually black will be able to break free with either ...e5 or ...c5.

One of my friends trashed an FM who played 10...Nf6 recently, it's just not a good move.

Good for Grandmasters I should say.

I think 1800+ players can play it. The problem is if you don't play it, black has a fairly easy time equalizing. The key to the move is to keep black restricted.

Of course you can play it, all I'm saying is that it isn't best for amateurs; there are better ways to keep black restricted. I've lost my fair share of games in the 4. ... Nd7 CK, but only one in one of the Ng5 lines, to an IM and it wasn't the opening; that says something to me.