You know what, though? It doesn't matter. As far as the D&D franchise is concerned Linear Warriors/Quadratic Wizards still beats the pants off of Linear Warriors/Linear Wizards. I mean it. And I don't mean it in an Ars Magica or Captain Hobo kind of way where people find metagame workarounds, I mean that the status quo of only having a third of the classes continue to function with an average-case DM with an experienced party after level 9 or so is superior to 4E and 5E D&D's solution. I mean this even though 3E D&D's LWQW is an objective problem while LWLW is a preference issue. Why do I say this? Well...
- A low-powered fantasy setting where most of the plot power exists in the hands of the DM is literally the easiest kind of game to design. Between modern, historical, hard or soft sci-fi, horror, superhero, or even off-the-wall TTRPG concepts like Toon, the salient fact is that low-powered fantasy games are significantly easier to design than the second-easiest game genre. Now, just because such games are easy to design don't make them bad. But that does means that most completed TTRPG projects will be low-powered fantasy. Go to DriveThruRPG.com, why don't you? Even accounting for the fact that D&D is the prototype for TTRPGs, low-powered fantasy is way, waaaay overrepresented even among non-d20 games.
Why? Because the biggest purpose of any setting for a traditional game is to provide a framework for characters to do things in. And it's just a brute mathematical fact that it's easier to brainstorm things for Conan to do than it is for Goku. Conan is thwarted by shit like broken bridges and smallpox epidemics and running out of trail rations; Goku is not. Yet most anything Goku is threatened or inconvenienced by also threatens or inconveniences Conan.
Making things even easier is how resilient low-powered fantasy settings are to fuck-ups, elisions, and author laziness compared to other settings. For example, smartphones have so upended horror fiction to the point where many writers just throw up their hands in defeat and set things before the late 90s. If you're writing a WWI alternate history game and get the political climate of the Triple Alliance wrong, people will nail your ass to the wall. But a Game of Thrones or Conan the Barbarian ripoff? Bitch, you don't have to worry about geography, politics, economics, or any of that shit. Any inconsistency or wonkiness you can't explain away with 'it's a totally different world from ours' can be explained with 'sociopolitics are so primitive that you can't prove that bread shouldn't cost 10 silver coins' and/or 'The Empire of Gond's laws are fucked up, they fall in the next scene anyway!'. Anything you can't explain away with 'sociopolitics are so primitive that blah de blah' can be explained away with 'it's magic, I ain't gotta explain shit'. Even better: if the players try to get some of that plot (-hole spackle) magic you can just snip their balls off like naughty poodles and say that the zombie apocalypse ritual or the forging of Masamune was a one-time thing, you fucking munchkins.
3E D&D at the end of the day is sitting in relatively uncharted territory while 4E and 5E D&D heads down the beaten path. Even before we talk about the actual rules, 3E D&D already has a structural advantage when it comes to satisfying customers. 3E D&D is just plain competing against fewer games within its niche than 4E and 5E D&D. 4E and 5E D&D don't just have to make a case as to why we should play their implementation of low-powered fantasy over 3E D&D's -- it also has to make the same case against Fantasycraft and GURPs and Earthdawn and FATE Core and Burning Wheel/Torchbearer and etc. If I want a game that has high-powered fantasy (and I mean meaningful high-powered, see below) I and players like me pretty much have 1E-3E D&D, Exalted, and Ars Magica. - LW/QW, and by extension 3E D&D, gives more stuff for players to do. This means that 3E D&D characters can tell more stories; this causes people to get bored of the game less quickly and encourages more off-game investment. True, this caveat only applies to the cleric, wizard, sorcerer, and druid but that's still more than the successor editions -- none of the character classes can look forward to doing anything meaningfully different. Oh, sure, 4E and 5E D&D characters grow but they don't evolve. They get more number penises and more combat maneuvers, but that's fucking it. Sometimes these editions will attempt to humor you by giving you bullshit like Hurl Through Hell or weak-ass summons, but at the end of the day you can't really do anything different with them. They're like the high-level abilities on a World of Warcraft character; you're still just shooting numbers at mobs. 3E D&D characters can replace iron industries, build castles out of thin air, put together meaningful undead and demon armies, construct interdimensional trade empires, construct businesses from scratch if you're using the DMG2 or Pathfinder crap, cultivate huge sections of land, etc. if they really try. Discussions of how PC spellcasters, using only abilities on their character sheet, can directly influence politics and the economy are some of the most popular threads on this forum.
And you know what? The 4E/5E D&D refrain of having to blow the DM if you want to make a castle or found a nation is fucking horseshit. Don't get me wrong, 3E D&D's ability to do so without tonguing the DM's asshole is extremely haphazard and is more of a happy accident than anything, but the fact that it exists at all is a huge selling point. I have spent dozens of hours pouring through the PHB and mentally designing castles and trying to figure how to get the most bang for my buck with Leadership and Plant Growth and Decanters of Endless Water; this kind of effort is a priori pointless in 4E D&D and 5E D&D. The kind of wizard's tower I get and who attends and where fixtures are and even where it's located doesn't really matter because in the end I get whatever the DM says that I get. Its effect on gameplay is whatever the DM says it is. And it doesn't even really matter if I give the DM a long list of features and details about what I want to do or build in the world and they say 'yes' to every single one -- it's still not mine. The DM has veto power on whether I can staff the stables with goblins or dwarves and how skilled my librarians are. Yes, fighters and rogues and other noncasters in 3E D&D are in the same boat as their counterparts in 4E and 5E D&D; again I'll say that it's better that only a handful of classes get to do anything cool than no one getting to do anything cool.
PS: This is not supposed to be comparing the merits of LW/LW vs. LW/QW vs. QW/QW in abstract. LW/QW is obviously an implementation problem no matter what system you're using unless you're playing something odd like Ars Magica. And my fix for LW/QW wouldn't look anything like what Pathfinder or 4E/5E D&D did. That a game with an objective problem is more appealing than a revision that attempts to fix the problem by patching in a system that's ultimately a preference issue says more about the game designers than the underlying paradigm.