My young friend and former student Theo Slade quoted this to me a while ago.
"To take is a mistake"
In some circumstances, it's easy to see why:
A capture by either side gives up the d-file and invites Rd7/Rd2, so both sides need to find something else to do.
There is a YouTube lecture by Smirnov on this subject, which has been made into a Lichess study, which makes the point thoroughly: https://lichess.org/study/MJ9ZFfKI
Another quote, from a current student:
"I'm a bit nervous about tension."
OK, I get that. If you respond to a threat by defending the attacked piece, the threat remains in the air, and if you overlook it later, it can bite you. This is Smirnov's first example:
So 1.Nxg4 Bxg4=, but 1.Nf3+/-
But of course, if control of h2 remains disputed, the N cannot move away from f3. But 1.Nf3! is still the best move and should be played!
A slight digression: The Ruy Lopez
"why is the Ruy Lopez ... so difficult for Black to combat? (Nunn and Griffiths)
Good question. Let's have a look:
Michael Stean comments here:
"The first and most primitive idea behind 3. Bb5 is to lay siege to Black's e-Pawn, which will subsequently be liquidated by d4, thereby opening the floodgates for the White Pawn centre to scatter the enemy forces with the allied pieces following up to rout the broken army -- no prisoners taken. Black cannot however be forced to surrender the centre. By keeping his own e-Pawn firmly defended Black can thwart all White's aspirations of conquest in the centre, but this will lead to a rather cramped position. So the offspring of White's plan is an advantage in space due to the necessity for his opponent to maintain a firm central barricade. This spatial plus however is not very big and the only way to maintain it is by avoiding any exchange of pieces. Black's position (Diagram 73) has sufficient 'capacity' for three minor pieces, but not clearly enough for all four of them. This explains why White is willing to invest so much time early in the game (Bb5, Ba4, Bb3, Bc2, h3) purely to avoid exchanges. If his spatial strategy is to succeed he must leave Black with four minor pieces. These closed Ruy Lopez positions are some of the most subtle and complex in the whole opening repertoire. It is probably no coincidence that nearly all the great players of recent times have been deadly exponents of the Ruy Lopez. In the hands of Fischer or Karpov 3. Bb5 sometimes appears to win by force. For example, Karpov-Westerinen, Nice 1974. 1. e4 e5 2. Nf3 Nc6 3. Bb5 ...
The clarity and popularity of Stean's book has led to this being a leading interpretation of White's chances in the Closed Morphy Defence. But there is more to it than that. Let us continue with the whole of that quote from Nunn and Griffiths:
"But already this opening has certain puzzling aspects. Why should White consistently maintain his initiative, both in the variations we have looked at and in those to come? Why should Black's position be so awkward? The question really boils down to a more basic one: why is the Ruy Lopez (which this opening has virtually become) so difficult for Black to combat? Look at the present position: Black's pieces are sensibly developed; he has as much space as White; his pawns are strong. Yet he has problems. "The answer seems to be that in this type of Ruy Lopez position Black can easily get caught in a situation where his game cannot unfold. Here for instance, White has his plans of Ng3-f5 and later d4, but it is less easy for Black to find something profitable to do without weakening himself or making some serious concession. His pieces may look reasonably placed, but they cannot readily achieve anything constructive or relevant. I should make it clear that this does not have to happen in a Lopez; it is far from being a bad opening for him. But in practice one error (6...Qe7) can leave him in misery. And so, if a player seems to have a respectable game (in a Ruy Lopez or any other opening for that matter), yet still loses, his misfortune may often be traced back to this lack of life in his position." -- NUNN AND GRIFFITHS
Trying to go beyond the Good Doctor's commentary is surely lese-majeste, but let me have a go anyway. Euwe and Kramer call this type of structure the Tension Form, and there we have another secret of the position. There is unresolved tension. The central pawns, White and Black, can take each other, be captured, recapture in different ways, or White can shut down the tension by d4-d5. Now, exactly where you want your pieces may be a different, depending on the central structure. If White takes on e5, the outpost on d5 may become tempting for a Knight (Stean is very good on this too), and so Black might want to keep a wary eye on that possibility, and be able to respond in the best way. If White plays d4-d5, that possibility vanishes, but other ideas appear. Operating under this tension and flexibility and uncertainty is difficult for Black, who has that constraint on their manoeuvres presented by the d6 pawn.
So the next question arises: why on earth would Black want to subject themselves to this sort of unpleasantness? Perhaps it is for lack of alternatives, but there is something else... Operating under this tension and flexibility and uncertainty is difficult for White too! Perhaps a little less so than for Black, but the positions are actually tricky for both sides, and not easy to simplify. So Black might willingly take on a slightly squashed position -- in the Lopez, in the Hedgehog, or in the Hippopotamus -- in the hope that the opponent with more space will still face difficulties in finding the right piece placements and the right plans in a tense, changeable position. As Suba pointed out, once you have your pieces arranged prettily after the opening, you still have to play chess... So, learn to love tension!
[Event "EuKra0211"]
[Site "Baden-Baden"]
[Date "2022.05.27"]
[Round "?"]
[White "Tarrasch S"]
[Black "Alekhine A"]
[Result "0-1"]
[ECO "C53"]
[PlyCount "58"]
[EventDate "2022.??.??"]
1. e4 (1. c4 c5
2. Nf3 Nf6 3. g3 b6 4. Bg2 Bb7 5. O-O e6 6. Nc3 Be7 7. d4 cxd4 8. Qxd4 d6 {
Hedgehog}) 1... e5 (1... g6 2. d4 Bg7 3. Nc3 d6 4. Be3 a6 5. a4 b6 6. Nf3 Bb7
7. Bd3 Nd7 8. O-O e6 9. Qd2 h6 10. h3 Ne7 {Hippopotamus}) 2. Nf3 Nc6 (2... d6 {
Philidor's Defence} 3. d4 Nf6 4. Nc3 Nbd7 5. Bc4 Be7 6. O-O O-O 7. Re1 c6 {
[%csl Gb5][%cal Gb7b5]} 8. a4 b6 {[%csl Gb7][%cal Gb7b6,Gc8b7]}) 3. Bc4 Bc5 4.
c3 Bb6 5. d4 Qe7 6. O-O Nf6 $1 7. Re1 d6 8. a4 a6 9.
h3 O-O 10. Bg5 h6 11. Be3 {[%csl Gd4][%cal Ge1e8]} Qd8 $1 {[%csl Ge8][%cal
Gf8e8,Ge8e1,Gd6d5,Ge5d4] "This paradoxical move -- the most difficult in the
game -- is very effecÂtive. The double idea is to prepare an eventual action in
the centre startÂing with ... exd4 followed by ... d5 and, at the same time,
free the e-file for the rook."} 12. Bd3 (12. d5 Ne7 (12... Bxe3 13. Rxe3 Ne7 {
[%csl Gc8,Gc4]})) 12... Re8 13. Nbd2 Ba7 14. Qc2 exd4 $1 15. Nxd4 (15. cxd4 Nb4
$1 {[%csl Rd3,Rc2]}) 15... Ne5 16. Bf1 d5 $1 17. Rad1 (17. exd5 Nxd5 {[%cal
Gc7c6]}) 17... c5 18. N4b3 Qc7 19. Bf4 Nf3+ 20. Nxf3 Qxf4 21. exd5 Bf5 22. Bd3
Bxh3 23. gxh3 Qxf3 24. Rxe8+ Rxe8 25. Bf1 Re5 26. c4 Rg5+ 27. Kh2 Ng4+ 28. hxg4
Rxg4 29. Bh3 Rh4 {0-1 Black wins.} 0-1