The trouble with this trap (as with many others played *as* traps) is that it relies for its validity on a mistake by the opponent.

In this case, of course, White refutes the trap by simple positional means. After 9. ...cxd4?! 10. exd4, White has achieved one of his positional objectives (creating a favorable imbalance with the use of the half-open e-file to exert pressure on that line and discourage the break ...e6-e5) at no cost to himself. (This is similar to White exchanging on d5 early in the Orthodox QGD, which allows immediate near-equality.)

Thematically, a lesson one has to learn, sooner or later, to become a stronger player, is *not* to set traps unless they are incidental to the sound and harmonious development of one's forces.

One of my regular opponents in years past was a highly unconventional player named Steve Sullivan, whose USCF rating never got out of B range despite his actual playing strength being c. 2050. Steve was a fine tactician, but one thing that held him back was his stubborn insistence on assuming that even strong opponents would fall into the multitude of traps he went out of his way to set.

In themselves, the traps were sound: If you unwarily took what he offered, he'd have a surprise waiting for you like the sting in a scorpion's tail. But take a moment of pause, and you could find the flaw: There was always a way to decline the trap and get a positional advantage out of it.

Actually, after 8.....Nxd2 9. Qxd2 cxd4 10. exd4 Bd6, it's just a normal game. These sort or structures come up in the QGD as you say, or in the Exchange variation of the Caro-Kann. Play might continue 11. Bg3 Qc7 12. Bd3 0-0. At some point black might get ....b5 and ...b4 in but it does seem it's just a tad slow. I suppose white has a small advantage. With careful play, black draws anyway.

I do agree in retrospect that 8..... Nxd2 9. Qxd2 Bd7 is more flexible. As Capa did in a famous game, maybe at some point you can organize ....Bb5 for black, killing white's attack before it gets started.