For this week's edition of "From Opening to Endgame", we will be examining the endgame which results from the 4.Qc2 Nimzo-Indian, after Oleg Romanishin's unusual (but brilliant) innovation 6...Qf5:
1.d4 Nf6 2.c4 e6 3.Nc3 Bb4 4.Qc2 d5 5.cxd5 Qxd5 6.Nf3 Qf5 7.Qxf5 exf5
I don't actually know how to annotate the move 6...Qf5. Alekhine's favorite 5...Qxd5 had almost disappeared from tournament practice because the position after 6.Nf3 seemed very difficult for Black, until GM Oleg Romanishin gave it a new life with his bizarre-looking 6...Qf5. As far as I know, this was introduced in a 1993 game Beliavsky-Romanishin, and the latter won the resulting ending. Probably at the time the move was given "!?" or "!", but now it is simply the only move played.
The evaluation of this ending is critical for the life of the 4.Qc2 Nimzo-Indian. White can avoid the ending - and indeed he often does - by 7.Qd1 or 7.Qb3, but it seems to me that neither of these moves can trouble Black. If Black stands well in this ending, then he has a reliable system to answer 4.Qc2, since the only other way for White to avoid this line is to play 5.a3, but that allows Black some other enticing possibilities.
At first glance, the move 6...Qf5 seems paradoxical and obscure. Black moves his queen again, challenging the queen on c2, and offering a trade which will double his f-pawns and give White two unopposed central pawns. Ironically, however, this trade gives Black additional control of the central squares, while the doubled pawns are not really weak. The d5- and e4-squares fall under Black's control, and only with great difficulty can White break Black's hold there.
Let's look at each side's advantages in this ending.
As I said above, the paradoxical queen exchange strongly increases Black's central control and gives him almost complete dominion over the d5 and e4 squares, which quite often actually do end up being usefully occupied by black pieces. In addition to this, Black also benefits from the centralized position of the king, which often ends up on d7 or e7, where it is quite safe. Another factor in Black's favor is the potential of some pressure on the queenside squares. Often ...a5-a4 gains space and fixes a weakness on b3 after the move a2-a3 has been played.
Let us now see a game in which Black's plusses proved to be more significant. The game in which the variation was introduced should be front and center, and is a perfect illustration of Black's chances.
Everything isn't rosy for Black. While he has gained control of the d5 and e4 squares, the change in the pawn structure also means that Black has lost the opportunity to make certain pawn breaks, such as ...e6-e5. Thus it is hard to challenge the solid white pawn chain f2-e3-d4. The black queenside pawns are also rather inflexible. ...c6 will almost always end up being played, after which the b7- and-c6 pawns will be stuck. White's main plan is the minority attack b4-b5. White hopes that this can allow him to use the half-open c-file; and in the event that Black plays ...c6xb5, White obtains a passed d-pawn. White can also hope to pressure the queenside by bringing a knight to c4 or c5.
In the event of 8.a3 Bd6 9.Nb5, the structure usually ends up being further altered, and White gets the two bishops, which can then become an important factor.
Now let's see a game where White's trumps triumphed.
What is the overall evaluation of this ending? Of course it is hard to talk about an objective evaluation. If a clear truth was known, super grandmasters would not be debating both sides of this position! I certainly don't think Black has an objective advantage, and if White has some edge, it is very small. But the character of the ending is the most important.
There are many draws in this ending. I wouldn't want to play this ending against an opponent who I felt I must beat. I think most competent players could hold the draw as White if that is their goal. There were some games where 2100s drew as White with 2600s. I feel it is likely that those grandmasters regretted their opening choice after the game.
Nevertheless, this is a full ending and Black has his share of the chances. I believe it is important to pick one's opponent for this variation. Against aggressive players, players who are not particularly good technicians, or in situations where a draw is sufficient, then this variation would be fine for Black.
From White's point of view, you have to know yourself. If you are the type of player who likes this sort of thing, then go for it. But I don't think you should go into this ending because of principle, such as trying to prove that 4.Qc2 is the ideal answer to the Nimzo-Indian. You have to be the type of player who likes such endings.
Studying this variation has some benefits for your overall understanding of chess. In particular, I recommend that players under 2000 take a look at this variation. The whole variation might look bizarre at first to non-masters - Black accepts doubled pawns, for what reason? And furthermore, on the next move (in the cases where Black meets 8.a3 with 8...Bd6) he sometimes accepts a further isolated pawn and gives up the two bishops! But there are valid justifications for all of these strange-looking decisions. Most of all, this variation teaches one that nominal "pawn weaknesses" - e.g. doubled or isolated pawns - are not always real weaknesses, while seemingly "healthy" pawns can sometimes actually be weaker.