Thank you, between this and Snarf's spoilered rundown, it is a very good summation. I'll try to look up the financial things referenced for a numeric analysis, but otherwise I consider this as answering my questions to my satisfaction.
I disagree. I think we can get much more specific, while still remaining 100% objective. For example, D&D is a cooperative game. (You can choose to play it competitively, but that's never been what it was designed for.) Being a cooperative multiplayer roleplaying game induces a variety of expectations and limitations that are significantly more specific than "is it literally at all possible for someone to use it" and "is it literally at all possible for someone to enjoy it."
Of course it does. However, we don't all agree on what is necessary for a cooperative game, nor the relative importance of these things in the game's quality. Therefore, a quality measure based on these expectations and limitations will be debatable, ending in differing people having differing ratings for the games based on which qualities of quality they consider relevant or most important.
When we make the only standard of quality something that is literally impossible to fail, the conversation becomes completely pointless. That's why it's unacceptable. You have reduced the conversation about "quality" to a triviality; this not only accomplishes nothing, it is actively caustic to actually productive, interesting discussion.
Calling someone else's position caustic does not meet my civility threshold. We don't need to continue if to do so would be in this manner. Regardless, I have not done this at all. I have drawn a line demarcating where I think universally-agreed-upon measures stop, nothing more. Honestly, I don't know why this is a problem, nor why it would intersect negatively with productive, interesting discussion. There's much more to be discussed when things aren't clear-cut than when they are (undoubtedly a component to why there is so much more discussion surrounding movies and art and the best hamburger than there are about jet engine performance or bridge structural soundness, and the like).
Again, I completely disagree. There are several functions beyond the "cooperative TTRPG" example I gave above that are useful for honing in on the design purpose of D&D specifically. Among them: "roleplaying" is clearly a factor, and that tells us things about what the rules are supposed to do; the "three pillars" (combat, exploration, socialization) are explicit design purposes, literally the designers saying what D&D is about (whether or not players use them is their prerogative, but the designers have been very clear that that's what they made 5e "for," and I have no reason to think this is not true of any other WotC edition); the need to be open to homebrewing, and yet also somewhat standardized so people can do things like "organized play" and "discuss the game on forums"; the overall thematic focus of the game being fantasy as opposed to sci-fi, horror, romance, or other themes; etc.
These are great examples, and they are worthy of discussion. Highly worthy. At no point have I implied that the components of a game are not important facets of discussion, nor that they do not contribute to a games' quality. Individual facets of a game are the places where things come closest to universal-consensus measures and judgements occurring. For example, 5e's stealth and vision rules -- these certainly rise to the point where one could find near universal consensus that they are not only bad, but diminish the quality of the game --although almost immediately the question of 'how much' becomes contentious. You can even make with/without comparisons like (see any number of write-ups about how Monopoly as-written is better than Monopoly with the common 'cash on free parking' variant).
It is once you start combining those facets into a cohesive measure of game quality that things get subjective, arguable, and potentially contentious. One person can say
Bunions & Baggins is a better game because it accounts for hobbit foot damage while another person says that's a pointless mechanic modelling parts of the game for which no one really want a codified mechanic and clearly
Hinfolk & Heroics is a better game for focusing on the action aspects, and neither can point to a specific unequivocal law declaring them right.
For that reason (and tying this back to the OP's thread-premising question), I think quality measures look less like ordinal numbers or the like and more like movie reviews -- potentially a numeric or 'thumbs up/down' score, but really being inseparable from a paragraph- to thesis-sized argument for the position, including some caveats and declared assumptions* with which the reader may or may not agree.
*example: I think it was Roger Ebert who was a little more agreeable to the notion that you didn't go to 70s-90s horror movies if you weren't expecting some hokey acting and unconvincing special effects and thus didn't hold those against such a movie as much as Siskel. This could be a parallel to a 'Well, gp=xp is a bad mechanic, because it only incentivizes treasure-centric dungeon crawls' 'But that's an expectation of the game.' - style disagreements.
The thread premise was how do we measure quality, and my answer is (roughly) 'with much complexity, and the final measurement output (coming from significant discussion and disagreement) still ending up looking like an argument or position synopsis rather than a simple number.
<also redacted tangent>