Suppose you have 1000 applicants and 900 slots. 900 of the applicants are white, 100 are black.

In traditional racism you give all 900 slots to the 900 white applicants.

Using a woke quota system you would have to arrange for it to be exactly proportional, i.e. 810 white and 90 black, even if that required taking race directly into account in order to force the numbers.

Using some kind of merit-based evaluation, you might have 830 slots go to white applicants and 70 to black applicants.

The first system is clearly morally wrong, but it's also misaligned with the optimal allocation. The second system has the same issue as the first; it's basing the decision directly on race even when it doesn't align with merit.

The biggest criticism of the third system is that maybe your metric for evaluating merit is imperfect, but it might still be pretty good. And other than that it's clearly the one with the moral high ground. So where is the moral question? The moral system and the optimal system are the same one. All that's left is the technical question of how to implement a more accurate measurement of merit, if possible.