See, if I could have game sessions that came within cooee of LotR, they would be absolutely amazing. If they could come within cooee of the better REH Conan stories they would be awesome.
And all those stories are constructed, like all fiction is. They rest on contrivance and authorial "manipulation" of the fiction. Presumably on the long trek from Rivendell to the pass of Caradhras someone got sore feet or a stubbed toe, but JRRT doesn't tell us about that. He does, though, tell us about the things that happen in the pass, when dark forces bring the mountain down upon the Fellowship.
I personally don't feel that the fact that JRRT, and REH, only provide us with dramatically meaningful fiction reduces verisimilitude at all.
To me, these questions have more teeth than concerns about verisimilitude, which - as per my remarks just above - does not seem to me to be threatened one iota by a focus on the "dramatic" or the "meaningful".
I don't have an easy answer with the context of D&D, other than 4e which doesn't really use attrition as its basis and hence does allow a focus on meaningful encounters (yes, it has daily powers and healing surges but I found that 4e works perfectly well with these treated as (i) a type of pacing mechanism and (ii) a means to push the players hard in relation to their resource suite).
Even in my RM play, which also has the potential for attrition as an element of play, my group never focused on attrition as a significant consideration. (And once we worked out that everyone wanted to play a caster, we reworked some of the rules around power points and multiple attacks so as to make sure that a nova-ing caster and a non-magical warrior were roughly comparable in mechanical effectiveness. In that second campaign we had two non-magical warriors as key characters up to and beyond 20th level.)
I don't want to exaggerate what counts as "significant", "dramatic", "meaningful" etc. There's plenty of scope for rising action as well as climax.
Also, I think one element of some clever RPG design is to leave the question of
whether something is rising action, or climax, as open-ended until the resolution itself takes place. Eg in my last Torchbearer session the PCs went to their frenemy's house, as they had agreed, to free his cook from possession by an evil spirit which had escaped from the heart of one of the PCs when she attempted but failed to cast a spell. This looked like it would most likely be a component of the rising action, with the climax being something involving the spirit itself (they hoped to cow it so that it would carry a spellbook into the PC's dream library, so she would then be able to add the spells to those she can memorise).
But the conflict was a spectacular failure for the players, and hence they ended up facilitating the spirit's escape from the cook - who was left a dried and withered husk - and into their frenemy, who is now just an enemy again, powered up by the evil spirit which also took the spell book into
his dream library, so now he has a better range of spells with which to foil the PCs. Prior to the conflict, the random weather roll had indicated rain, and in my narration of the failure - which was also the end of the session - I decided to unleash my climactic cliches and described a lightning bolt striking the house and blasting it asunder, with the PCs out on the street with the charred timber of the front of the house while their shadow-empowered enemy stood above them gloating.
This all makes sense. I'm sorry that I don't think I've got much that is useful to say within a D&D attrition paradigm. Maybe
@hawkeyefan has some thoughts?
I've seen that thread. I can see the overlap, but I think the two issues
can be orthogonal to one another - as in, it is possible to design a RPG which has resource management but separates it from the issue of "meaningful" encounters. I think Torchbearer is an example of this, because it doesn't rely on "filler"/"slog"-type events or encounters to generate resource attrition. It just uses an ever-ticking clock (which the game calls "the Grind").