@smmenen Actual design is messy. It's hit-and-miss. We have a restricted list because of this. We have cards that see no play (in part) because of this. It's really strange to say "Well, you couldn't know 100% for certain that a card would lead to more meaningful non-obvious decisions, so we better not take that into account when designing cards in any way". That an insanely higher bar than any other design principle has to pass.
Although I disagree with your position and many of your ideas, it is difficult to avoid nitpicking when you make claims of such staggering breadth. The last sentence above is a case in point. It's impossible to know whether you actually mean something less hyperbolic and the error is in formulation or whether your ideas are just badly conceived. I usually take sloppy formulation as a sign of sloppy thinking.
Case in point: I can think of a dozen design principles (top down design, for example) that do not require such information at all. That's because most design principles don't require knowledge of specific Vintage outcomes. I already pointed out that Wizards, by their own admission, do not test cards for Vintage play. At all.
The problem with your counterpoint here isn't that Wizards can't know with 100% certainty whether one card design or an alternative, both aiming at the same target, would both preserve the same degree of meaningful choice and also contribute to the same match win %. The problem is that there is no way of knowing with even 50% certainty, or perhaps any confidence at all.
The problem of the multiplicity of formats underscores, most severely, the problem with your ideas for design. A card might actually maximize meaningful choice in one format, but be destructive to it in another format. What if the card maximizes meaningful choice in Vintage, but is utterly destructive of it in Modern or Standard? How would you resolve that problem if you were a designer? How would you, as a designer, even assign weights to that, let alone to match win %ages across formats?
This is, of course, aside form the theoretical impossibility of what you ask (the most advanced magic theory suggests that win%ages and choices are correlated and explain each other). The practicalities are worse.
Your design ideas aren't operationalizable.