TLDR: WotC said the quite part out loud, caster rule and noncasters drool
Okay, so there are a series of interviews (I think each has it's own thread) and part of one of them Chris Perkins says "a lot of high level spells could just be renamed shenanigan's" (I may not have exact wording right) but then goes on to say that is what makes D&D what it is, when the players pull some shenanigans and change everything...
...
High level spellcasters have TOO MANY ups over high level martial characters. Heck middle level (7-10) full casters have more game changing abilities then most epic level martial characters. and WotC just admitted it.
This seems like strange framing. That D&D has a problem with martial-caster agency imbalance at upper levels (to some, that being anything over ~level 6) is not some hush-hush secret that no one acknowledges, it's been one of the primary points of discussion about the game for several decades. Beyond that, they aren't admitting something (as in what one does with a transgression), but acknowledging it. I'm sure if WotC were really saying the quiet parts loud, it would instead be something like,
'we tried taking baby steps towards fixing this problem, both in late 3e (see weaboo fahtan whatever meme) and in 4e, and people (to a lessor or greater degree, depending upon whom you ask) rejected it.'
Regardless, yes, this is indeed a problem, and has been since the beginning. High levels were seen as some faraway place wherein people probably wouldn't much want to play, and when they were, the fighter having either an army or a golf bag full of magic items (many of which were fighter-only) was seen as a reasonable offset to the high level wizard (who, admittedly, really were a pain in the rear to play). The game has slowly wiped away the fighter=general and fighter=better loot and wizard=highly restricted bits and never really addressed (excepting bo9s and 4e) the underlying imbalance.
Even beyond class differences, D&D has a fundamental framing where magic can just solve a lot of things while mundane solutions are constrained or muted or vague. Even 3e and 4e, which had the most complex skill and world-task-resolution systems had systems which were rigorous, definitive, and weak (excepting the rare exception like 3e Diplomancer, which great makes one semi-mundane build work well but does nothing for any other style of play). Part of that has to do with the game shuffling complex things off into spells (early Rangers getting spells to emulate Aragorn's herb lore because there wasn't yet a skill system), but also because the D&D audience tends to interpret complex systems in-game as magic (Bo9S abilities are 'spell like' pretty much because the audience looks at them and says 'that's a spell by another name' even for the ones that have no obvious supernatural effect).
Yeah I dunno either but I imagine WotC has some idea. Given that 40+ players are only 13% of the market, and the bulk of the market has basically been raised on superheroes and quasi-supeheroes, I suspect they'd be more open to it than us coffin-dodgers.
5e was designed without knowing ahead of time that it would be a smash hit with the new-to-gaming market (and supposedly with an eye towards recapturing part of the market that left in the/at the end of the TSR era). It will be interesting to see what design elements are on the table now that they know this. Perhaps some more, 4e-level breaks with tradition will be deemed acceptable now.
I tend to agree re: pulling casters down. I think we could elevate martials a bit, just with some more interesting ability design (without necessarily going "full superhero", just like keeping it in the literary fantasy/action hero zone), but the top end stuff with casters, and really anything above level 5 spells gets so dominant that it can very quickly become the focus of play even in 5E (especially given the sheer number of spells they can cast at that point). So I think slower progression of spell levels and capping spells at like L6 or L7, and just reconsidering every spell above that level. Really direct combat spells weirdly are not the main issue, it's all the rest.
There were some very specific decisions that could have not happened and had differing outcomes. Some of them being combat: Keeping all the numbers closer together and weaker creatures a threat longer makes summoning/minionmancy more powerful than it otherwise would be. If you shapechange/are shapechanged and hit 0 hp you revert to your normal form instead of just dropping/dying like you might in AD&D. The save math has been changed to make you fail your saves more often, which is good since the consequences of failure are in theory lessoned and thus no more save-or-dies, but that hasn't held up universally. Force Wall and Force cage (and Leodmund's Hut) are immune to damage when they just as easily could have AC and HP like any other obstacle (how many genre fictions have force fields get overloaded?). Others, yes, it is frustrating how much you can spend on mundanes skills and preparations only to have a druid with goodberry and pass without trace or a wizard with knock and passwall solve all the problems. Spells per day are a limit, but I think it seems really common that they just aren't enough of one to actively select a mundane character build (it would be different if you were choosing magical or non-magical tools when planning a given adventure, but when you have to have immutable facets of your character dedicated to the role, the frequency of ('we'll be in a dungeon with more locks than you can cast knocks') is simply insufficient.
