On first sight, 1.e4 c6 2.Ne2 d5 3.Ng3 dxe4 4.Nxe4 looks a bit harmless. True, there are other lines where White takes back with the "wrong" knight on e4, e.g. Nf3 (dxe4) Nf3-g5xe4, which can become a gambit, if Black defends the extra pawn with f5. In the French Defence, 1.e4 e6 2.d4 d5 3.Nc3 Bb4 4.Nge2 dxe4 5.a3 Be7 6.Nxe4 Nf6 7.N2g3 is a respectable line (Speelman - Foisor, 1-0; Maróczy - Szigeti, 1-0). 

Ben's proposal to provoke f7-f5 looks interesting though. I'd be tempted, by the way, not to play 4.Nxe4 directly. 

(a) Maybe 4.b3 raises the chances that Black plays 4...f5. Then 5.Bb2 Nf6 6.f3 exf3 7.Qxf3 may be possible. 

(b) 4.Nc3 f5 5.d3!?, is there something wrong with the good old Tarrasch rule that three developing moves are worth a pawn? If 4....Nf6, the reply 5.Ncxe4 h5 comes into consideration (at least this move order excludes 5...Bf5). 

(c) More solid is 4 Bc4(!), followed by Qe2. After an exchange on e4, the queen might travel to h4. In the French Defence we have 1.e4 e6 2.d4 d5 3.Nc3 Bb4 4.Qd3, so why not... It can be seen as an Englund Gambit where you don't have to fear Nc6-d4 (because of the pawn c6).

Ben_Hague wrote on 02/18/10 at 09:34:26:


Black also gets to make moves, by the way. And he can use them quite effectively to develop the queenside and leave White with a rather passive position.

In the end I agree with BPaulsen - why put effort into continually trying to making 3.Ng3 work when you could spend the same amount of time learning a proper answer to the Caro-Kann, such as the Panov-Botvinnik Attack.

I would think that after, say, 8...e6, there is nothing special and Black has a better-than-usual Caro position.

Ben_Hague wrote on 02/18/10 at 22:06:40:


Apparently immaturity is linked with not being illogical enough to apply gross generalizations to entire openings when there's a specific variation under discussion.

It's amazing enough that someone honestly thinks 3. Ng3 constitutes an actual try for an advantage. Then again, I wouldn't be surprised if you think white has something after 4...Bf5 at this point. Who needs an extra tempo on the main lines?

Ben_Hague wrote on 02/17/10 at 20:20:53:


Speaking about openings in general terms is pointless.

Even more so since you're trying to push for a white edge in something specific.

And the rest of your logic is inane so I snipped it. Just because white can play something doesn't indicate in leads to some kind of "normal white advantage". The d-pawn specials are living proof of this.

This entire line is irrelevant after 4...Bf5 anyway where white (yet again) has absolutely nothing. At best he ends up a tempo down on the main lines, because the extra option of d3 is not challenging in any way.

1.e4 c6 2.Ne2 d5 3.Ng3 de4 4.Ne4 e5 is equal. White can't play d4, White's knight will be kicked back with ...f5, Black has a central space advantage, and what does White have in return?

Playing ...e5 without first inserting ...Nxe4 looks like a good attempt to tweak this for Black.

BPaulsen wrote on 02/13/10 at 13:09:27:


To clarify what I mean, I would consider a normal opening to be one that has commonly has grandmasters playing either side of it. I would further consider that all these opening have fairly similar evaluations, i.e. White has decent chances to get an advantage and Black has decent chances to equalise. E.g. the Spanish and QGD I would consider to both be better for White, regardless of whether theory states that Black can equalise or White has an edge because White is always the one pushing for something. So this week it may be += or = but you can consider a sort of average to be +==.

I then use this a rule of thumb when evaluating unusual  positions, e.g. this looks like opening x, but where Black has done y, which he wouldn't normally do in opening x, therefore White is better. In this case I would consider that both players have done something that they wouldn't normally do, so as a first approximation they can be considered to balance and so the evaluation reverts to the default. 

Of course this can be wrong, but most opening positions are so rich (unless someone has done something badly wrong) that proving an evaluation can take many games over many years and even then  it can still be overturned.

If you look at an opening like the Latvian it looks iffy on general principles, but it's taken a vast amount of effort to reach the current point where it's almost dead (and it's still not nailed down). Would you really want to have to go through that process every time you came across a position that looked a bit Latvian-like? Or would you prefer to say it looks like a Latvian so I'll assume it's good for White until proved wrong.

Anyway the sense in which I use normal White advantage simply means worth playing for White.