I think you lose more than you gain by having a Fighter at all. You'd have an easier time selling the Battlemaster as a base class than a subclass.
And I think, at least at this time, far too many D&D fans would refuse to accept a game that lacked a Fighter than you'd gain from such a move. Dramatically, overwhelmingly more, in fact.
We've already seen what people can do just with misinformation.
Actually deleting the Fighter--whether or not that would be a productive idea in game design terms--would be a catastrophe of epic proportions. I really, really wish I didn't have to say that. But I can't see any other possibility in light of what happened with 4e.
I vaguely recall that Warlock and Sorcerer were originally introduced as simpler options to Wizards. :-/ More pew-pew, less "which spells should I prepare and hope are useful today?" But they managed to be complicated in their own ways.
The problem is, by still having a broad spectrum of
potential options, but both (a) getting fewer of them and (b) getting to use those options
less frequently, both the Warlock and the Sorcerer actually cry out for
more optimization than the Wizard, not less. You need to be absolutely sure that every spell and metamagic/invocation you pick is going to be useful, consistently, or else you've been hobbled for an entire level, perhaps more. (And given how stingy DMs are with XP IME and from what I hear from others...)
Much as I like the idea of the à la carte design of the 5e Warlock, I find it really hard to pick from the grab bag, especially with the various prereqs and such, and the choices all seem even more arbtirary than the broad spell catalogue. Which you also have to deal with!
Yeah. The
idea of Warlock seems simpler, but the actual practice is merely shuffling the complication away from day-by-day play and into chargen-play. Running a pregen Warlock would be a breeze relative to a Wizard because you have far fewer choices to make and what choices you do have are, usually, pretty straightforward. By comparison,
levelling up a Wizard is a breeze because you have no class features and just pick two new spells you don't know from the list of highest-level spells you can cast. Sorcerer is kind of in the middle, less build-complicated than Warlock but more build-complicated than Wizard, less in-play complicated than Wizard but more in-play complicated than Warlock.
5e Sorcerers might be the simplest option for a caster, but it's still much more complicated than most martials, at build time and play time.
I have often thought about how to go about building a truly, genuinely
simple spellcaster. Taking leaves from the Battlemaster, actually, though not copying the exact Expertise Dice subsystem. My checklist of requirements is:
- No "memorizing" spells. You just have a set of them available all the time. If you have a set of pick-able magic actions, whatever you've picked, you can always do it so long as you have the juice.
- No lengthy list of flexible options. If there are choices, it's "pick half of these 8 things" or "choose 1 of these 4 things as your core focus." Save flexible alternatives for a subclass.
- Basic "do a magic thing" action, almost like an upgraded merger of various cantrips (mage hand, prestidigitation, minor illusion, etc.)
- "Magic burst" effect that you can use X/SR, probably half proficiency or full proficiency. Charges may be used to power other things, and subclasses might hook into other stuff (e.g. a haemomancer using their own HD to fuel their magic)
That would produce an
actually simple spellcaster.