Well put! And this is actually a big problem in literary theory, with which I had a few sparring sessions in the process of my PhD. For instance, the distinction you're making between two different senses of Realism implies that what is real in the real world and what's real in the fiction can be separate. In contrast, Dr. Stacie Friend argues that we subconsciously adopt what she calls The Reality Assumption, which means that "everything that is (really) true is fictionally the case, unless excluded by the work." If we take this to be what's behind the notion of realism in fiction, then two senses of realism that you just described melt into one: Realism does mean "looking and behaving like the real world", except when the work explicitly adds different assumptions, in which case realism means "being well-grounded in established rules that do not change for light and transient reasons".
The problem with the Reality Assumption is that I hate it and how sturdy it is. It made my PhD thesis's work way harder than I thought it would be, and although I feel like there are obvious cases where we do not just assume everything in real life translates to all works of fiction unless stated otherwise, it's a really solid theory that is really difficult to disprove. And as long as it is difficult to disprove, it looks like the two senses of realism must be blended to a certain extent. Which might also be why I value verisimilutude over realism when it comes to fiction — verisimilutude explicitly interests itself with a feeling of things being plausible instead of conforming to reality.
I mean, the single biggest problem with the Reality Assumption is that it WILL lead you astray. Not even in an accidental sense; authors will
exploit it against readers for shock value, subverted expectations, or plot twists.
For D&D specifically, the major issue isn't so much the Reality Assumption, but rather several
unreality assumptions, coupled with some
carte blanche attitudes regarding the realism of specific things.
1. The vast majority of D&D players have extremely unrealistic, and indeed completely counterfactual, understandings of what is physically possible for human beings in the real world to achieve. This is less a "reality" assumption and more a
projection-of-self assumption. D&D characters are often expected to be limited, not by the potential limits of actual human beings, but by what
the player personally can achieve. This is of course patently ridiculous, because the player isn't anywhere near an Olympic athlete, which is a much better standard for what reality
actually permits in terms of human physical fitness.
2. A significant number of D&D players (not quite as many as the previous, since some of them are actual enthusiasts of medieval history) have completely backwards concepts of several verifiable historical or physical things. E.g., how much weapons and armor should weigh, or whether it is appropriate for there to be gunpowder weapons in a D&D context, since most people don't know that matchlock guns appeared in the 1400s, meaning
essentially exactly the same time as what we call "full plate armor" today.
3. People are really, really bad at physics and statistics. The vast majority of people,
even those who have taken introductory physics courses, will default to an Aristotelian model of how the world works. E.g., heavier objects fall measurably faster than lighter objects (they do not; they experience exactly the same amount of acceleration if dropped from the same height), objects naturally slow down until they stop (they do not; they only do so in the presence of friction forces), etc. Likewise, humans have very finely-tuned intuitions only for specific
kinds of probability, and those intuitions produce garbage results outside of those specific questions (this is why simply changing
how a probability question is asked can increase the rate at which people answer it correctly, even though the information content is unchanged.) As a result, they will presume many
entirely unreal things, and then get confused, even angry, when those unreal things are proved unreal (consider the Monty Hall problem).
4. The aforementioned
carte blanche comes in the form of magic. What can magic, or magical things/creatures,
do? Who knows! There are essentially no limits prescribed on what "magic" can do in D&D, just call it "magic" and most people instantly accept whatever you describe without comment. This punctures the Reality Assumption argument from the opposite direction: works often need to do damn near
nothing to (so-called) justify entities or behaviors that aren't grounded in real-world things. Dragons, for example, are so deeply embedded in
human culture (seriously, there's something dragon-like in nearly all cultures on Earth), that their presence in fiction (D&D or otherwise) often doesn't even get any explicit justification
at all, they're just
present and merely
assumed to work because we see them doing the things dragons are supposed to do (fly, breathe fire, usually live a long time, sometimes talk, sometimes use magic, maintain stable populations despite completely insane mating practices and lifespans, etc.)
It's
these things that are the real problem with calls for "realism" in D&D. They reflect that there IS a gap between "semblance to our physical Earth" and "groundedness in a set of cognizable rules." It's just sometimes they'll
create a gap where there shouldn't be one, and other times they literally need nothing more than two words--"it's magic"--to cross a gap as wide as Valles Marinaris!
"Most recent IQ test you have taken"
How many IQ tests did people used to take?
I've taken a handful in my life. If I'm allowed to round, technically speaking, my most recent (which was...a long time ago) would give me a whopping 19 INT. (I'm pretty sure that score was a fluke. I'm smart, but I am fairly sure I am not THAT smart.)
Okay.
