The great Swedish player Ulf Andersson is well known as a master of the endgame, and created quite a few masterpieces in this area of chess. Naturally, he was happy to go straight from the opening to the endgame whenever possible. One of his specialties was the following queenless middlegame, which arises after only seven moves, when White tries to avoid the Grünfeld Defense by playing 1.Nf3 Nf6 2.c4 g6 3.Nc3 d5 4.cxd5 Nxd5 5.e4 Nxc3 6.dxc3 Qxd1+ 7.Kxd1
I have even heard a tale somewhere that Andersson was happy to play blitz, but wanted all games to start from this endgame! Here we have chess, but with the queens (and a pair of knights) removed. In a way it makes it simpler, but in another way it makes it more complex. It is not always easy to perceive the small mistakes which lead to defeat or victory from this position.
The importance of this endgame is that by this move order White is trying to avoid the Grünfeld Defense proper by delaying d2-d4 and then recapturing on c3 with the d-pawn. How can we evaluate the starting position of this endgame?
We can see that White already has slightly the freer play. The white pawn has reached e4, while the black e-pawn cannot yet do that, and to manage to play ...e5 Black will have to play the slightly awkward moves ...f6 or ...Nd7. The white bishops can easily develop to active posts such as c4 and f4, e3, or g5. Black in turn has played the move ...g6, but it is far from clear that the bishop will be well placed on g7, where it faces a sturdy pawn on c3. It is likely that the bishop will be happier on c5.
The White king has been "disturbed", but this is not really a relevant factor. After one move - Kc2 - the king will stand perfectly well. It is more likely that the black king will face problems. However, one important point worth noting, when the white king is on c2, is that f2 can be vulnerable.
Overall, it is clear that White has some advantages. That's not really in doubt. The question, however, is whether he can make any use of them to obtain a long-term advantage. If not, and if Black can complete his development and equalize space in the center (carry out ...f6 and ...e5), then the game will be fully equal. Basically I think it is a simpler mirror of the initial position in chess - Black needs to be somewhat more careful than White at the beginning of the game, but then he can look forward to a full game of chess.
First, let's see two classic examples of Ulf Andersson winning as White in this ending.
Now to see the other side of the story, let's observe how a super-GM quickly equalizes and then smoothly outplays his opponent with the black pieces.
How to summarize this endgame? Two very important positional motifs stand out. First of all, the exchange of bishops. In general White wants to exchange light-squared bishops, and Black wants to exchange dark-squared bishops. Second, the battle for space on the queenside is important. Nepomniachtchi in the above example gained a lot of space on the queenside by ...a5, ...b5 and ...a4 and this was the key to his victory. In the game Andersson-Franco Ocampos, Ulf Andersson managed to play b4-b5, a4-a5, etc. and gained an initiative on the queenside (in addition to effecting the ideal exchange of minor pieces).
Overall I believe that it is possible to play for a win on either side of this endgame. Basically White has the only advantages as the endgame begins, but they are temporary. White is better when nobody knows anything, but by now the correct methods of play for Black have been worked out, and thus he can most likely equalize, after which a game of chess begins, albeit in a simplified version.
View more games with this Grünfeld Endgame position;
Read Ulf Andersson's entry in Chessopedia;
Watch GM Dejan Bojkov's video General Strategy: When To Exchange Pieces;
Reach GM-elect Bryan Smith's article Space in the Ending.