Trompowsky: why does 1.d4 Nf6 2.Bg5 d5 3.e3 e6 score so highly for white in master games?

Question

I'm trying to find a suitable response to the Trompowsky using LiChess. I saw the following line, which looks like it might transpose into more familiar QGD territory.

[fen ""]
1.d4 Nf6 2.Bg5 d5 3.e3 e6

On LiChess, in master games, white scores 52% wins, 33% draws, and 15% losses. It seems black's moves are faulty for some reason. I'm not sure why though---it looks like white has a sharp line.

For contrast, after 3...c5 white scores 24% wins, 48% draws, and 29% losses. So 3...e6 scores significantly worse.

Question: Why does 1.d4 Nf6 2.Bg5 d5 3.e3 e6 score so highly for white in master games?

Answer 1

3..e6 closes the way for your light-squared bishop, but I don't think that's too good of a reason to reach such scores (I checked on ChessTempo and got similar numbers)

Please be careful about possible bias due to player strength differences. A move may score worse simply because it is often used as a "drawish weapon" against a stronger player. This is the case of the French exchange variation, where Black scores better in a symmetrical position! But there could be other reasons why an opening system is more common among weaker players

I would tend towards an explanation of this kind, since there does not seem to be a clear "refutation" in the position you show (4.c4, 4.Nd2, 4.c3 and 4.Nf3 all score similarly.

Please specify your own skill level for further advice on how to build a repertoire around these lines to be possible

Answer 2

Those statistics are surprising but I think it has to do more with player strength than the quality of the opening moves.

Of course Black has closed his Bc8 in behind his pawns, but this bishop is often a problem in queen pawn openings. And in fact the game could transpose into some main line opening. So is 4.e6 bad? No, it is just this line of play is not something strong players do. And if strong players don't do this that means the Black players are weak and thus the statistic.

Why wouldn't a strong player make these moves, they seem reasonable:

1.d4 Nf6 2.Bg5 d5 (2...Ne4 is more common but 2.d5 is another main move)

Black says, "go ahead and double my pawns. If 3.Bxf6 exf6 opens my Bf8 with gain of time, and I get the Bishop pair."

3.e3 e6

Now Black is saying, "I am just going to pin my Knight and block my c8 Bishop, but my d-pawn is solid and I can just develop my Bf8 next." [Weak!]

Compared to 3.e3 "How can I make that Bg5 move look stupid... 3...c5 opens my queen to possibly attack the weakened b2 pawn or even get a check on a5. Plus I am attacking the center and trading my c-pawn for his d-pawn is desirable."

Answer 3

White's position is more flexible.

3...e6 is not an horrible move, of course. But it restricts Black's actve possibilities (Bc8 is locked in, Nf6 is pinned) without putting pressure on White as 3...c5 or 3...Ne4 would.

As a consequence, White has a wide choise how to continue. 4.c4 would reach a QGD without exploiting the move order, while 4.Nf3 or 4.Bd3 leads either to a Torre attack (with c3) or a delayed Queen's Gambit (with c4), and Black has to be careful to prepare for both plans.

But the main trouble is that White is allowed to enter a Stonewall with the bad bishop out of the pawn structure, with 4.c3, 5.f4, 6.Bd3, 7.Nf3, 8.0-0. This system cannot be reached usually because of some ...Ne4 or ...Bf5 or ...c5 disturbing White's construction early on, but here they will enjoy good spots for all their minor pieces, a firm grip on e5 and excellent prospects for a kingside attack should Black cancel on this side. This line is already quite playable with White bishop on c1, but having it active on g5 is a huge asset.

If you really aim for a QGD as Black, I would rather suggest 3...Nbd7!? when 4.c4 e6 and 4.f4?! Ne4 are fine for you and 4.Nf3 lets you choose between ...e6, ...Ne4 or ...c5.