lichess.org
Donate

Catalan move order

I've been studying some games from the Candidates tournament and failed to understand a move order subtlety of the Catalan.

So my question is very specific.

Some players wanted to avoid Karjakin's queens indian defense and therefore played 3. g3 after 1. d4 Nf6 2. c4 e6.

Looking at this particular move order, I noticed that most white players, in this position continued by delaying for one more move the developpement of the g1 knight, with 4. Bg2.

So I'd like to understand the differences between the positions arising after 4. Bg2



And 4. Nf3 (or 3.Nf3 4.g3 which transposes)



The only explanation I could come up with is that some very rare lines are no longer available to black.

After 4...dxc4 5. Qa4+ (which isn't the most testing move, but considered to be a solid line for white) some players tried 5...Qd7 with the idea of Qc6 :


which is obviously not as attractive after 4. Bg2:



This 4. Bg2 move also seems to prevent a serious option for black after 4...Bg4+ 5.Nd2 5...dxc4



is no longer available:

Delaying Nf3 keeps the long diagonal open and prevents an early b5 in the open variation (4...dxc4). Playing Nf3 early lets White play Qc2 (or Qb3) quickly since the d4-pawn doesn't need to be defended by the queen anymore. If Black plays the closed variation (no dxc4 before move 8) or one of the semi-open variation (4...Bb4+ or 4...Be7 before 6...dxc4), there is no difference between 4.Nf3 and 4.g3.
Guess you are lucky that there is an IM willing to answer you, and I did play Catalan myself, including this move order.

My answer to you is that: actually Bg2 is much more natural than Nf3. Because you will have the diagonal for your bishop anyway, whereas you do not need the Knight on f3 right away. The issue is less about why do people do play 3.Bg2 but much rather why do they not? And there there is an answer: there are some lines (Bb4+ Bd2 Bd6 e.g.) where you actually do not play Bg2 even though you have played g3 (but rather Qc2,e4,Bd3, maybe 0-0-0)

This topic has been archived and can no longer be replied to.