Rodi covers this in "Modernised Queen Gambit Declined" probably his NIC articles but maybe a few updates. After 4Nf3 Nb-d7 5pxp pxp black can play as in Gukesh v Caruana Candidates with 6Bg5 c6 7e3 h6 8Bh4 g5 9Bg5 Nh5 etc. So there is something like a complete rep with nb-d7 in that book

After 1. d4 d5 2. c4 e6 3. Nc3 Nf6 4. Nf3, I think that 4...Nbd7!? retains it's status as often recommended side-line. It's not suitable for players who want to play the Tartakover variation, or who want to defer ...Nbd7 in the exchange variation. However, that would not bother some players. If White wants to avoid transposition to the Manhattan, Cambridge Springs, Orthodox, or Exchange variations, he can play
5. Bf4!
5...dxc has been recommended, and more recently 5...Bb4. My note on sources:

Davies (Kindle Location 3022) for 5...dxc.
Max Illingworth, ChessPub, for 5...Bb4.
More recently:
GM Arjun (Modern-Chess) and Krishnater (Chessable) for 5...Bb4.
Rodi (Forward chess) and NIC 119, 121, 122, for 5...dxc4.

5...Bb4 analysis by Arjun and Krishnater diverge in at least one crucial line. I think they both demonstrate that Black hangs in there in complex play but the analysis sometimes runs deep.

The older 5...dxc4 provides interesting, perhaps simpler play. After,
1. d4 d5 2. c4 e6 3. Nc3 Nf6 4. Nf3 Nbd7 5. Bf4 dc4 6. e3 b5!? (6...Nb6, 6...Nd5, 6...a6) 7. Nb5 Bb4 (7...Nd5) 8. Nc3 (8Nd2) Nd5 9. a3, 9...Nxc3 and 9...Bxc3 have been played or analyzed. 9...Ba5!? seems to be new and might be ok.


Here is some extended analysis with stockfish, by way of a sample line:
10. Bc4 (10Rc1) Nc3 11. Qd2 Nb6!? 12. b4!? Nc4 13. Qc3 Ba6!? 14. ba5 c5 15. dc5 f6 16. O-O e5 17. Rfd1 Qe7


Stockfish still prefers White a little, but the position looks ok to me.

4. ...Nbd7 5.cd5 ed5 and Black lost some options with Bc8 in the exchange line.

Nigel Davies' book on the Queen's Gambit Declined includes Bf4 lines preceded by either ...Be7 or ...Nbd7.

He provides seven annotated games after 1. d4 d5 2. c4 e6 3. Nc3 Nf6 4. Nf3 Nbd7: four games with 5Bf4 dxc and three with 5cxd exd 6Bf4. ChessPublishing has also covered these lines. It seems that Black is doing reasonably well in both theory and practice.

The ...Nbd7 Bf4 lines are especially relevant to anyone considering the Cambridge Springs variation as black.

Methodchess wrote on 02/29/16 at 02:38:38:


I continued on your on-screen board. The final position looks so much like the Exchange variation that I thought it would transpose. It's not so simple. In some lines of the exchange variation white prefers Bf4 to the exchange of bishops. The main variation has black's knight on f8 instead of d7. I found one line with an early Nf3 (not considered best) where black is one tempo ahead. So your sequence puts black one tempo behind in a variation that is generally avoided by white. So it turns out that comparing lines is not so easy!

Methodchess wrote on 02/29/16 at 02:38:38:

Strictly speaking not me, but I'd still like to point out 5.cxd5 exd5 6.Bf4.

In another thread I created regarding a query about the name of the 7...Nbd7 variation of the QGD, I mentioned in passing the line in the title of this thread, I'm interested to discuss this in more depth with any others that are knowledgeable of this line, or are interested in looking into it.

I like this following set-up used by Grachev and Zvjaginsev in particular but game I used below was from Jianu-Marin, 2014: 4...Nbd7 5.Bf4 dxc4 6.e3 Nb6 7.Bxc4 Nxc4 8.Qa4+ c6 9.Qxc4 Nd5 now there are several options but will just show most testing: 10.Bg3 Qb6 11.Qe2 Bb4 12.0-0 Nxc3 13.bxc3 Bxc3 14.Rab1 Qd8= there is a game that reached this position Jianu-Marin 2014 that ended in a Black win.

The alternatives on move 10 for White seem to be 10.Rc1, 10.0-0 and 10.Be5, all of which lead to nothing as well as far as I can tell. If anyone knows a path to +/= for White in this line I would be interested to see it. If not, this could force White players into abandoning 5.Bf4 from the 4...Nbd7 move-order and going back to 5.Bg5 which is more interesting to play imo.

Further to the above I also think the following line is worth considering as an alternative to 5...dxc4, instead play 5...c6 and we have the 5.Bg5 Semi-Slav position except the bishop is on f4 instead of g5 for White, and Black has the added move Nbd7. I'm not sure if this favours either side but I think this is an interesting line for further research as both Kramnik and Carlsen tried it out once each, both were blitz games played in 2014. Carlsen drew his against Grischuk in WCh blitz in Dubai, 2014. Kramnik lost against Aronian in Norway Blitz, 2014. I prefer the way Carlsen handled the position, although I prefer a different continuation to what Carlsen played in his game. I have included very brief analysis of 5...dxc4 and 5...c6 in pgn games below:





I would be interested to see if anyone can come up with a path to +/= for White somewhere after 5.Bf4. If not, hopefully we may see a return to 5.Bg5 in the future. I would rather play either of the two aforementioned lines as Black, than memorise the theory needed to play 4...Be7 5.Bf4. I also think both lines pass the "Carlsen test", by this I mean: I don't think Carlsen would accept a draw in either of these lines for Black.