In the 6th game of the 1909 Capablanca-Marshall match, Black developed his queen bishop prematurely by 1.e4 e5 2.Nf3 Nc6 3.Bb5 d6 4.c3 Bg4.
By pinning the f3-knight, …Bc8-g4 can help Black pressure a white pawn at d4, but White hadn’t committed the d-pawn yet, and 5.d3 left little promise for the g4-bishop. The bishop then came under attack by Nb1-d2-f1-e3, after which g2-g4 plus Ne3-f5 made a strong outpost for the rest of the game.
In general, the “knights before bishops” chestnut is because early on, we’re unsure of the bishops’ best squares, while the knights and their limited moves should start toward the other side before the long-range pieces do.
The earlier we develop bishops, the riskier the bishop move tends to be. For instance:
1. d4 Nf6 2. Bg5 invites an immediate skirmish 2…Ne4.
If White holds the bishop back one more move, Black’s second move could suggest an effective bishop development. C. Torre’s 1. d4 Nf6 2. Nf3 e6 3. Bg5 often leads to Bg5xf6 plus e2-e4, where constructing a favorable pawn center costs bishop-for-knight. If White restrains the bishop until move 4: 1. d4 d5 2. c4 e6 3. Nc3 Nf6 4. Bg5, the pin indirectly adds to the pressure on d5 (along with the direct plays c2-c4 and Nb1-c3.
White delays the bishop development for one move in each case, but in both cases, the positional logic increasingly indicates Bc1-g5 as a good move.