Now that I've created or modified four classes for D&D Next (the alchemist, outlander, sorcerer, and warlock), I want to take a step back and consider what's working, what isn't, and a few new ideas I've had for future classes. It isn't that I want each campaign to have a huge list of classes to choose from; I think 10-12 classes is as many as any one setting needs, and many can do with fewer than that. Fantasy is bad about cleaving to its tropes, though, when some new and wild idea could be right around the bend.
For people who don't super care about reading my personal design postmortem and want to skip straight to the New Ideas, you have my permission to scroll down to the New Ideas heading below.
The first thing I've learned is that it's hard for me to get myself to completely ditch the structural underpinnings of WotC's classes. Given the slightest opportunity, I'll reskin rather than restructure. I have good reasons for this, like a greater assurance that the class will fit in with the advancement paradigms, general feel, and basic rules concepts of other classes... and then some bad reasons as well, like a reluctance to embrace radical change. This is why the alchemist and my revised warlock are reskins of the bard, and the outlander is more or less a ranged monk with guns rather than punches. The sorcerer is more its own thing - and also by far the least successful of my efforts in kitbashed class design.
So what would the sledgehammer solution entail? For starters, it would take a spell list that was comparable to but separate from the cleric and mage lists. This is achievable, but it's a lot of work - a lot more than one post should attempt. I started this kind of effort with the alchemist, offering a paltry 14 formulas, but the post was already pushing 5,000 words. If in the future there is a robust OGL for D&D Next so that I could sell the class as a PDF, that would be great and I would absolutely want to expand the formula list.
The Outlander more or less works as a class, though it could use a second at-will attack - a scattershot attack or something. I know that this is D&DN and WotC doesn't think every class needs two or more at-will attacks that are pretty much always good ideas, but... well, they're wrong. Super wrong. Having two different options won't prevent a fight from being dull, but only having one at-will option (and not all that many daily options) makes it a lot more likely to be boring. When you're staring at the same bonus and damage expression on your character sheet round after round and you know that description is empty fluff... the game's illusion starts to fall apart.
The Outlander could use more flavor and background, emphasizing the role of outlanders in the setting. A whole class of this guy and this other guy and a few others isn't exactly the strongest setting connection. They're kind of the creepy, sexy loner gunslinger right now, and while there's nothing wrong with that, I could stand to provide more nuance and depth by considering what other kinds of outlanders look like. D&D Next offers a handy tool for this: all I have to do is match the class up with each Background option and see what kind of weird new idea that generates. The Minstrel background speaks for itself. On the other hand, I'll have to do a lot more thinking, and maybe adjust aspects of the setting, to match it up with Backgrounds that aren't social outsiders, such as Noble, Priest, and Soldier.
My sorcerer is, to put it briefly, a hot mess. I have some thoughts on how I'd revamp it - possibly swift-action spells that require you to have cast another spell that round. Getting them off of spell points and onto spell slots is certainly useful for getting them to play nice with the multiclassing rules. I was particularly proud of the second sorcerer build that I created, as it fits Aurikesh very well, and thus especially disappointed with my result. The mechanical conception of the class - a middling spellcaster that transitions into a pretty solid melee brute (or some other role, but there's only one example) as it burns through its daily potential - was an instant hit with everyone in my gaming group who saw it. WotC has made it clear that there will be something called a sorcerer, and that there will be a class functionally similar to the thing we like, and that these will be separate things.
My most recent warlock is a little bit more of a departure from WotC's original warlock draft. WotC's version had at-will and per-encounter abilities, but no per-day powers. That isn't a problem in any particular sense, but I watched the later development of other caster classes and came to the conclusion that my first warlock, which stuck with the no-per-day-powers idea, was conceptually losing ground relative to other classes. The Outlander - which again is a very close reskin of the monk - highlighted that the utility of ki powers and Arcane Gifts was substantially outpacing the utility of pact abilities. In actual use, I think we even forgot that Arcane Gifts are written to be per-encounter just like ki powers, a fact I have only just now recalled.
Therefore my most recent draft of the warlock gains pact boons as well as a half-casting progression, comparable with bards, paladins, and rangers - a full spread of at-will, encounter, and daily abilities. I feel like the class would work, and prove to be reasonably competitive, as I've written it, but the flavor I was going for isn't playing all that well with the target audience. The idea was that the warlock's advancement describes an increasing knowledge of forbidden lore, and some of that includes the hidden (i.e., arcane) secrets known to others; because the warlock is a dabbler rather than a true student of any one art, some spells are borrowed from other lists (perhaps leaning on the "beneficence" of the patron as well). What came across instead was the same problem as bards prior to 4e: second-best (or lower) at everything, not shining in any single field.
The thing that I love about outlanders and warlocks is that they emphasize the importance of the classes of supernatural entities that struggle to wrest control of the world away from the gods. This isn't to say that all warlocks or outlanders want these powers to triumph (though the warlock that is in the game would love to see the Fey do so), but it's the primary high-end struggle of the setting.
I'm planning another revision of everything except the alchemist. (Let's be real: the alchemist probably needs revision too. Until someone plays one for a few sessions, though, I won't know what needs to change.)
The sorcerer needs to back way the hell off on the data management. I would be happiest if the sorcerer had something class-directed to pursue in-game, equivalent to a mage's pursuit of spells, but I don't expect that to be as readily manageable without just giving them arcane spells and spellbooks. Since I want them to be more separate from mages than that, I'm not sure what will come out of it. (Suggestions welcome.)
The outlander is okay, but clarifying a second at-will (or, even better, giving three options and letting the player choose two) really ought to happen. I want to make sure the idea generalizes well and thematically to non-starlock (that is, flintlock) ranged weapons. Beyond that, I'm not sure what the class needs; the third Outlander has just been rolled up as a character, so between those three players I may see some feedback.
The warlock will shift away from mage-like spell slots, and toward a deeper examination of pact boons. I do think they need some one other Thing that is only indirectly part of the pact and boons - maybe a push-your-luck mechanic allowing them to delve into forbidden lore to produce arcane spell effects. It could get to be a bit much on the data tracking, but it could also be super cool.
Also, I won't be at all satisfied with the warlock until I have written rules to support at least one pact for each of Fey, Infernal, the Dead, and the Abominations. More would be better.
I don't know yet whether these will fit Aurikesh, much less any other setting, but I was thinking about reskins and came up with some ideas that are a bit more off the beaten path of fantasy, or emphasize different things about existing concepts.
I talked a long while back about liking the idea that clerics channel the power of the gods, and that such forces are more raw and direct than the current spellcasting mechanics support. The barbarian's Rage mechanic represents a state change and some loss of control, something that could be reworked to be appropriate for spellcasters. I still remember the "Rage Mage" prestige class from Dragon Magazine, very shortly after 3.0's launch; I don't think I'm going too far to call it a regrettable conception. Still, channeling the Powers That Be is a pretty common element of fantasy, to say nothing of some real-world religions.
It shouldn't be terribly hard to shift advantage on all Strength checks and attack rolls (from Reckless Attack) over to advantage on all Wisdom or Charisma checks (whatever the casting stat is) and imposing disadvantage on all of the opponent's saving throws. What's that you say? That sounds overpowered? Take it up with the barbarian class, my friend. Anyway, reskinning the barbarian's totems should likewise be a straightforward exercise. I'd also take a step or two back from the d12 hit die. As for the limitations on spellcasting, that's a little trickier, but once I work out the aforementioned push-your-luck mechanic for warlocks, introducing it to the channeler should be a natural fit.
This one is not a deep reskin, and people who care about having as few classes in the game as possible may find that it doesn't justify its existence all that well. Still, the paladin class is pretty dedicated to its heavy armor, while the ranger (appropriately) emphasizes the wilderness. A class focused on combining clerical magic with cloak-and-dagger spy work has a certain appeal, though the obvious choice is to build the concept through a cleric-rogue multiclass rather than developing a new class. Still, if there were new mechanics that gave the Inquisitor an edge in social encounters (now that WotC is actually designing social mechanics, rather than halfassing and assuming all will be well) and a custom spell list that was as different from the cleric as the paladin is, but in its own direction? Sure, that might justify its existence. (The low-hanging-fruit solution is probably to introduce cleric spells that interact with the social mechanics instead.)
This is less about Aurikesh and a bit closer to Dust to Dust, but a class dedicated to grafting on monster parts? Yeah, that could be pretty awesome. It's the kind of thing that 3.x would handle as a prestige class and 4e as a paragon path, but why not let something like that work from the start of play? The monk is one good option here, as it goes long on customization within the class: just replace the Way of the Elements stuff with monster abilities (adjusting numbers as needed); some fleshcrafters go deep on a single graft and get all of the different dragon-breath abilities or whatever, while others get one ability from each of several different things. The barbarian totem warrior path would work too - just make the abilities gained from each beast a bit more literal and visceral, rather than "because it's my spirit animal," and add a lot of new options for monsters rather than woodland creatures, and we're basically there.
Assuming one is willing to go through more of the Bestiary to support this class, the fleshcrafter has an endless justification for going out into the world and looking for new things. Horrible one-off monstrosities? Yes please! Legendary beasts imprisoned since the world was young? Sign me up! Much like the homunculi of DtD, the class operates on and pursues a category of treasure that no one else even notices. The downside, of course, is that the monster-of-the-week gameplay that this class encourages isn't right for every campaign, and is far less desirable for some other classes.
As I talked about with the outlander, the fleshcrafter doesn't fit into polite society all that well. It's kind of a single concept, rather than something broad and open to a ton of different interpretations. I think it would be fun to play, though. Given its internal customization options, no two fleshcrafters might be quite alike in powers. On the other hand, beyond that, they might be too similar in characterization - not that this has been a problem with DtD's Krudrunis homunculi. (Of course, legally speaking, there are no Krudrunis in Marath Suvla.) That's a case where restricting the class to a single race strengthens its theme and story.
I don't feel too bad about being on my third draft of these classes. It's not like any of WotC's classes are still on their first draft, right? And while I don't encourage the extreme proliferation of classes in games, I think that some campaigns could set themselves apart in the players' minds and communicate a ton of theme by cherry-picking an unusual arrangement of classes and banning the rest. For example, what if the fleshcrafter was the only melee brute class in the game? It would say some unusual things about the setting, or at least the story that the DM cared to tell within that setting. If you added some healing options to the mage spell list, you could tell some very Dust to Dust-appropriate stories with only the fleshcrafter and mage classes - while other things exist in the world, the DM is saying that those things aren't what this campaign is about. (What's that you say? It sounds a lot like a Mage/Promethean mashup? Don't mind if I do.)