Stockfish gives +6.6 to the following position.

enter image description here

The recommended move by stockfish is Rc1 which maintains +6.6 advantage but if I exchange the queens right away (Qxd6), stockfish gives me +5. What's the reason behind the difference (in stockfish points) between two positions?

I can understand by Rc1 we put pressure on c7 but black can exchange the queens and later can defend the pawn too. Also, as a rule of thumb, I don't usually exchange queens off the board if I can take advantage of the position. But here it seems the queens are going to be exchanged anyway.

You can find the full game here.

  • 16
    +6.6 and +5 isn't a huge difference, they're both completely winning. Commented May 28, 2018 at 9:11
  • 1
    @RemcoGerlich Yes, they are both winning but it is a huge difference in positional terms. My question is why one position is better than the other Commented May 28, 2018 at 9:21
  • 10
    @LaschetJain The difference between +6.6 and +5 is not necessarily the same as the difference between +0 and +1.6, at least not in human terms. The less pieces there are on the board, the more blurry engine evaluations tend to get.
    – Annatar
    Commented May 28, 2018 at 9:46
  • 6
    What @Annatar means is that 5 is winning. 6.5 is winning. The difference between them doesn't matter much. However, 0 is even, and 1.5 is a slight advantage. In the latter case, one would desire the path that gave the extra 1.5. This is a pretty standard chess engine interpretation.
    – Tony Ennis
    Commented May 28, 2018 at 14:37
  • 1
    @LaschetJain The value there is the pawn advantage. Playing a king and pawn endgame with 5 pawns or 6 pawns against a pawn does not really matter. You'll win anyhow. But playing the same endgame with a pawn or two pawns matters because in the first case, your opponent has drawing chances.
    – padawan
    Commented May 28, 2018 at 15:44

8 Answers 8


I think it is because 21.Rfc1 wins the c7 pawn. If black responds with a queen trade, she can't defend the pawn on c7 because white's bishop can attack a defending rook on c8 or d7.

[White "NN"]
[Black "NN"]
[FEN "3r3r/2p1kppp/B2qp3/2Qp4/6b1/PP2P3/3N1PPP/R4RK1 w - - 0 1"]

1. Rfc1 Qxc5 2. Rxc5 Kd7 3. Rac1 {and the pawn falls}

That said, the position is completely winning either way and exchanging off the queens is a practical decision that reduces risks and keeps everything under control. White has an extra piece and a passed pawn.

  • 2
    I would exchange queens for pragmatism.
    – Tony Ennis
    Commented May 28, 2018 at 14:27
  • 9
    Black's queen is pinned, so exchanging this move is quite trivially not necessary, making the better move easier to see.
    – Spork
    Commented May 28, 2018 at 14:50
  • Seems to me in can be delayed 22. ... Kd6 23. Rac1 Rd6 24. Bb5 Re7 Commented May 28, 2018 at 23:04
  • @Acccumulation: Rc6+ chasing the king away and Rxc7 works anyway, but at the end of your line the black king should also be very wary of mating nets (something like f4 and Rc6+, maybe Ne4+ somewhere). I don't think there's an immediate win but the king should be very careful. Commented May 29, 2018 at 7:13

It is true that Rfc1 wins the c7 pawn, but even if that where not the case, Rfc1 is significantly better technique than initiating the exchange yourself.

The black queen is pinned against the king, so there is no risk of black evading the queen exchange. Stepping out of the pin will just waste a tempo, because then you'll still be able to exchange.

If you exchange, you improve position of the black king, which enables him to play c5 next, improving his pawn structure.

If he exchanges, he improves the position of your rook and the c7 pawn stays backward. That is positionally much preferable.

All this doesn't matter much, because you are a piece up anyway. But without the extra piece it is exactly this type of precision leading to slightly better positions that wins games.

  • 5
    This is typical of what any strong player would feel. Whites position is so superior and so easy, that there is no reason to snatch at a simple move when the better move is equally simple. Perhaps it does not matter that you "throw away" 1.5 points when you are so far ahead anyway, but if you have that attitude in a position that is not quite so overwhelming you may throw away another 1.5 points next move, and again the next move after that. Laziness is never rewarded
    – Philip Roe
    Commented May 28, 2018 at 22:21

The reason why a computer makes this type of suggestion i a completely winning position, where a human would choose another completely winning continuation, is because the computer is not human, and does not have a general intuitive understanding of the concept of completely winning.

It bases its evaluation of lines on concrete calculations, and then it notes its evaluation of the resulting position in the form of a numerical value which is often related to the material balance at the end of the line. When evaluating a line, the computer makes no practical considerations whatsoever and it doesn't care at all how difficult a position is to play when presenting a continuation as optimal.

However, for humans, the difficulty of playing a position is an important consideration to make. Therefore, in the position you provided, you made a move which no human would ever even think of criticizing you for making while Stockfish noted it as a "mistake".

While you may note that white is completely winning and the conversion to a victory should be a simple task after trading queens and simplifying the position, Stockfish notes that there is a way to win even more material by playing 21.Rfc1 and will therefore criticize your move, even though the win is not made any more difficult by playing your move.


Lots of people are pointing out that Rfc1 wins you c7, but this is proof of more general advice

Employ Zwischenzug where possible

Zwischenzug mean "intermediate move"; it's the idea of placing your opponent in more pressure before making an obvious move.

The most clear examples involve being in a position to take an opposing piece, lets say a rook. Instead of taking it right away, you might be able to put your opponent in check first, developing a piece. Then, after your opponent has answered the more pressing concern of check, now you can go ahead and take their rook.

In this example, you've pinned black's queen. Nice work. Now, there's no immediate move black has to get out of it. You could go ahead and take black's queen, black captures back, then you play on like usual.

Instead, what Stockfish is proposing is that you place black in a more uncomfortable position before making the trade. There are only a few responses to Rfc7; none of them are very good for black. So why not go ahead and force one of those "not very good moves", especially if it lets you develop a rook? The queen exchange isn't going anywhere, and you can use it to leverage an even larger advantage.

"To take is a mistake"

Now, that's one of those over-generalized "rules" which only applies in certain cases, but this is one of them!

The idea is that you never want to sacrifice the momentum or positioning gained in an exchange if you don't have to, and often, the person initiating an exchange loses both of those.

I'm obviously not saying that taking a piece is always (or even usually) bad; if black's queen weren't pinned, you should obviously capture it. But right now, you don't have to take it. It's not going anywhere. Consider these outcomes:

You take the queen and black recaptures with their rook, threatening white's bishop. Now white can move their bishop to d3, which encourages black to play g6, or they can move their bishop elsewhere, encouraging Rhd8. So now black's either developed a rook and strengthened their king-side pawns, or developed a battery on a semi-open file. In exchange, white's lost development on their bishop.

White should still win on material, but they didn't have to give black that positional advantage. If white moves Rfc7, then black captures the queen, white recaptures with their rook. The queens are off the board and white developed a rook; advantage white.

Of course, if black doesn't make the trade you've still got the advantage, and if black removes the pin on their queen, you can just grab it then.

Qxd6 says "I'm trading queens because I'm up on material", but Rfc7 says to black "You can initiate the trade and I'll develop a rook for it, or you can squirm a little first and I'll do it when I'm ready."


I don't have enough points to comment, but please note that if Stockfish actually thought it was a bad move, you would have an annotation like "?!" and a comment like "Inaccuracy" or "Mistake". The fact that there is no such annotation means that, although the positional evaluation is lower, it is not considered an error.


Playing Rfc1 essentially gives White two extra tempi. Consider the following lines:

A) 21.Qxd6 Kxd6 22.Rfc1

B) 21.Rfc1 Qxc5 22.Rxc5 Kd6 23.Rac1

The only difference between the lines is that, essentially, the a1-Rook in line A) has been transferred to c5 in line B).

These two extra moves allow White to win the c7-pawn. Continuing line B):

23...Rd7 24.Rc6+ Ke7 25.Rxc7

Whereas if we continue line A):

22...c5! and White isn't winning the c-pawn immediately.

As a final note, Black doesn't have to exchange queens in line B), and can protect the c7-pawn by playing 21...Rd7. However, then after 22.Bb5 Rd8 23.Qxc7 White wins the pawn anyway.


Rc1 is preferable over an intimidate queen trade because white's queen is active, and black's queen is passive.

Moving the white queen to a7, a5 or b7 to defend the bishop loses the activity. Therefore, by playing Rfc1, white can activate one more piece. In case black goes for the trade, now queens are off the board, and white has a very active rook on c5, ready to support the passed pawn on a-file.

But as others mentioned, both moves are winning. Only difference is that Rfc1 has a slightly more active game.


It is NOT bad. Possibility of a better move, doesn't make the worse move a bad move.

Usually, when you're the attacking one, you don't want to exchange pieces. That's a general rule.

On the other hand, if you have material advantage, you should rather try to exchange as many pieces as you can.

Given position is not worth analyzing, white is a piece up, it's game over already. HOWEVER, if you are less experienced player (I assume from the question) I think that exchanging queens is even BETTER. Why? Because it will make it harder for black to come back to the game. Queens are dangerous!

Pragmatism for the win! (for example, look at the game of Magnus Carlsen vs Fabiano Caruana: https://www.chessbomb.com/arena/2018-altibox-norway-chess/01-Carlsen_Magnus-Caruana_Fabiano - look how red it is! Did he played that bad? No, he played safe (in most cases)).

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.