No. The point about replacing a test with more subjective criteria is completely invalidated when, instead, multiple tests are replaced with a single test.

EDIT: Instead of completely invalidated, I should say irrelevant to MIT's change. In another context it could be an important point.

I think the top level comment is based on the assumption that a greater quantity of objective data will lead to a greater weighting of objective data, and a lower quantity will lead to a lower weighting, leading to a greater weighting of subjective measures.

Likewise, there are many skills you can't demonstrate on the SAT such as Spanish language proficiency. For whatever reason I took the French subject test but no AP test, perhaps there are other schools that do this.

The math score might say a little about the other sciences, but the blmultilingual students are going to be disadvantaged by this it seems to me.

This is completely incorrect. The SAT is not changing to incorporate the skills that would have previously measured by the subject test. As a result we are simply going to have fewer objective metrics.

When I say "replacing" I mean in terms of the weight they carry in admissions.

Exactly. If you look at the comments on the linked article, or the ones that were on top here when I wrote mine, people are making comments that aren't specific to the SAT subject test at all -- "this is great, because I bombed that test", "this lowers stress", "that test was too easy anyway".

If the subject test was a thing that everyone was informed about, did not cost extra, and was readily available, if not required alongside standard SAT administration, then I would be more inclined to agree. But here knowing that they even exist is half the battle.