Base Classes and Classplosions

[Username17]

Writing character classes for Dungeons & Dragons is not and never has been terribly difficult. I've done various "one day character class" challenges, and enough of them have turned out alright that I just don't take the claims that character classes need to be rationed or produced slowly terribly seriously. Further, the fact that 4th edition launched with only 8 classes was a serious strike against it. While the edition was hot garbage and got poor word of mouth, an awful lot of people looked at a PHB that lacked Druids and Bards as being cripple-ware, which of course it was. While many people never gave the game a try because word on the street was that it was a crap edition, lots of people waited for those reviews precisely because it seemed like it wasn't even complete (which of course, it was not). There genuinely isn't any excuse in releasing a Player's Handbook that is "incomplete" in terms of not having all the major starting classes.

On the other hand, new classes sell books. They just do. No one fucking wants to read your Open Grave book if it doesn't have a playable Necromancer and Deathknight class in it. The 4e people seemed to understand this fact, and that's why they decided to release the original PHB as crippleware in the first place. This was of course ridiculous, because there is absolutely no shortage of character concepts that you could base character classes on. You don't need to hold back the Druid or the Ranger from the base book, when future books could include Seekers, Scouts, Treesingers, Beastmasters, Wardens, Explorers, Augurs, Mystics, Trappers, Cultivators, Watchers, Herbalists, Witches, or Preservers. Classes which are Ranger-like or Druidish are simply in functionally limitless supply. However many books you care to make for your edition, you simply are not going to run out of class concepts.

The only things you have to do to leave yourself open the possibility of classes is to not write any core classes that monopolize future concepts. So shit like the "Magic User" is a non-starter, because obviously you're going to want to write up more spellcasting classes in the future and having a basic class that does "all forms of magic" is super bullshit for a bunch of reasons.

So long as you create an environment where everyone gets a Prestige Class at a fixed an arbitrary point (like 10th or 11th level) that allows them to compete at the point where armies are marching around, it's genuinely OK if a class concept is something that doesn't inherently scale to adventures beyond the scale of Willow, Dark Crystal, or Disney Princess movies. Classes like Thief Acrobat do not conceptually scale to the point where you are fighting armies of Gnolls or giant God Heads floating in a psychedelic void, but that's totally OK if there's a point where the Thief Acrobat character becomes a Witch Queen and then a Demigod.

-Username17

[Ancient History]

Classplosion is a particular side-effect of 1) efforts to limit multi-classing, and 2) designer efforts generate novel abilities. Combined, that can lead to some good-and-bad things.

The nice thing is, player options increase dramatically when you restrict multiclassing, because suddenly gish classes become exciting. If Fighter/Wizard sucks as an option, you have room for base classes like Hexblade and Duskblade, and prestige classes like Eldritch Knight and Arcane Archer. Prestige classes in particular offer specialized content, for those players that want those options - although that carries with it the drawback that highly specialized content only appeals to a very small slice of the player base, but if you get enough of them out there (and boy did 3e try), you cover almost everybody. Still, if you have enough base pairs, just gish classes will get you dozens of potential mixed classes. Even AD&D understood that with kits.

The bad thing is, the number of interactions increase dramatically the more abilities you add - especially if the abilities are especially novel, like Incarnum or ki points - and you not only have to chronicle the way those new systems interact with all the other existing systems, but they tend to stagnate and die if there is little interest in them - which is why prestige classes that try this tend to suck hard. A thief that enters into a prestige class that grants limited spellcasting for a small spell list might seem very attractive...until the next supplement comes out with a cool spell they'd like to have but can't because whoever wrote that book didn't include their funky little prestige class in the list.

That's a minor-case scenario, the kind of thing you can mindcaulk over and ask Mr. Cavern to overlook, but it gets more complicated when you look at the interaction of divine spellcasting, arcane spellcasting, psionics, incarnum, ki, artificers, invocations (warlock and otherwise), auras, binding, shadow magic, true naming, maneuvers and stance...and whatever other stuff that creative designers throw in, turning daily use abilities like Rage into currency for doing something that isn't raging, etc.

So as a rule I would say it is better if all the basic mechanics are introduce in base classes, and prestige classes only expand on their abilities rather than try to create new ones. (i.e. if you're a thief that wants to dip into spellcasting, your prestige class of choice gives you access to the Sorcerer spell list, not some funky arbitrary subset thereof which no one will remember to update).

[Foxwarrior]

As far as I can tell, the only really good thing about having a classplosion is that it lets every player character be a super special snowflake like in superhero fiction. Superhero fiction usually remembers to make the characters' superpowers be describable in a way that conveys most of what they can do quickly though... things like Warlock, Soulborn, and Truenamer are just like an arbitrary grab-bag of abilities with some themes and resource management tying them together though, and I won't mention Wizard.

From a setting perspective, people often don't even get around to visualizing what countries would really be like with core classes in them, much less Psions and Wu Jens and Icewrights.

It's actually possible to use any old rules-heavy RPG system to run mysteries. Just presenting people with scenes where they look at/for evidence and try to reason about what happened and how to catch the criminal. But only if it's possible to look through the list of things a criminal could potentially be able to do, and narrow it down. After Sundown is pretty great for that, because while the list of magic powers does cover a lot of conceptual ground, it's also reasonably short and readable and you can skim it while looking for anything that could have done whatever you saw.

[Username17]

I'm not talking about the need for Prestige Classes, which I take as given if you intend to have the potential conflicts scale to the point where "being a Ranger" ain't going to cut it. And since even Lord of the Rings went there and had the Ranger prestige class into Ghost King, I don't find the counterargument that you could keep Fighters viable by just having every suck mud forever to be compelling in the slightest.

The issue is that you're also going to want to produce a great number of additional base classes. You're not going to stop with the Rogue and the Assassin, you're going to have people who want to play Scouts, Masked Avengers, Ninjas, Slayers, Nightblades, and Thieves. And while you could reasonably ask "Why the fuck do you gotta play a Nightblade instead of playing an Assassin like a normal person? It's the same thing!" That is in fact wrong.

First of all, if you have access to enough content that people can put "Nightblade" on their character sheet, then they should do that if that's what they want to do. It doesn't do you any harm for someone's character sheet to say "Ninja" or "Slayer" or whatever the fuck. if it gives them any amount of joy at all, fucking why not?

And secondly, as you've doubtless seen with Wizard spells in every edition except 4th, having a never ending funnel of content producing options for one class is the worst thing for game balance. Like, literally the worst thing. Producing new content for new classes segregates the bullshit. Like, if you produce a new version of It's Cold Outside and there's like fifty million ice spells and they are all things that Wizards can dumpster dive through, then you get shivering touch and ice assassin added to the Wizard list and it's a notable powerup for the casters. On the other hand, if those new spells mostly go to the new Snow Scaper and Frost Witch classes, the powerup of the Sorcerer and Elementalist isn't nearly as big.

And thirdly, the creation of new classes allows you to play with alternate resource mechanics. The Assassin has a deal where he can take extra time to do super moves, but the Scout is just about running around and instead gets to do super moves by disengaging and re-engaging as a skirmisher or some fucking thing. You can make Rogue-like abilities that wouldn't be balanced on the sort of conditional activation shenanigans that Rogue Tricks operate on by having the Thief and Masked Avenger have different criteria for activating their powers.

-Username17

[Lokey]

So far we can use lower-case for all examples, so wondered about thoughts on one of my bugaboos: the organization classes such as Harper X or Sword Coast Pirate. I've never liked, don't see why it can't be a sentence in your character history, allies or carried item section(s) of your character sheet outside a game with narrow lines of what characters to run. Maybe there's some somewhere that aren't just class X with a different save progression and starting feat that I haven't seen of course.

[Ancient History]

The things is, D&D 3.x never formalized what a prestige class was supposed to be - so they could be as specialized as the designer wants to be. In that, they're worse than bloodlines in WOD games, where at least the mechanics are minimized to discipline choices and a weakness. So you get really generic prestige classes (Assassin) which can apply to basically any setting, race, or character class, and you have really specific ones like the Netherese Arcanist which only applies to a specific campaign, race, character type, a fucking specific period in that campaign, and is an Epic level bit of nonsense on top of that.

It's basically an NPC prestige class...and we should probably talk about the problems involved with NPC classes in classplosion as some point.

[DrPraetor]

And secondly, as you've doubtless seen with Wizard spells in every edition except 4th, having a never ending funnel of content producing options for one class is the worst thing for game balance. Like, literally the worst thing.

On the other hand, fighting types will want to be able to write "tactic: snow ambush!" on their character sheet, while wizards will indeed want to write "spear of the rimtursar", so you don't have to create a new party to use the content in It's cold outside: Tokyo Drift.

D&D 3rd edition catastrophically failed to solve this problem with feats, but if you have fighter abilities that essentially work like wizard spells - you can write as many as you want on your sheet as a function of training journeys and other narration in play, but for each adventure you have to choose some finite number to prepare - the issue is solved.

At the same time, character specialization by something akin to feat selection (in other thread I suggest calling them "kits") is also desirable, so you can have a fighter who has the "Ambush Fighter" kit and thus is better at ambush tactics and remains so even if the other fighter prepares some ambush-type fighting moves on a particular adventure.

You'd then want resource management schemes on top of that, so wizards could have magic points (or they could exhaust prepared spells), while moves that assassins prepare have triggers.

I don't think it's desirable to rigorously restrict people from dipping into multiple resource management schemes, not because it isn't an avenue for min/maxing but because the game can tolerate significant variation in character competence before people will notice, let alone complain. So if Ninja/Runesmith is the best because your ninja abilities trigger to let you make attacks and then your smith abilities trigger when you make attack rolls using your superior equipment - as long as you aren't enough better to break the game for the guy just taking Paladin levels, it isn't an issue.

That doesn't mean that free multi-classing is necessary or desirable, though.

[Username17]

So far we can use lower-case for all examples, so wondered about thoughts on one of my bugaboos: the organization classes such as Harper X or Sword Coast Pirate. I've never liked, don't see why it can't be a sentence in your character history, allies or carried item section(s) of your character sheet outside a game with narrow lines of what characters to run. Maybe there's some somewhere that aren't just class X with a different save progression and starting feat that I haven't seen of course.

Base Classes in generic books should be generic. It is entirely reasonable to have base classes that are setting-specific in setting-specific books. No reason that Defilers can't have their own base class in Dark Sun or that there can't be a base class Crab Clan Berserkers in Rokugan.

Prestige Classes (or Paragon Classes as they were called in 4th edition) should exist to provide character concepts that have access to level appropriate abilities when things have scaled up to the point where any of the base classes are no longer able to conceptually keep up. Those concepts could be generic like "Merchant Prince" or they could be semi-setting dependent like "Daimyo," or they could be all the way setting dependent like "Red Wizard of Thay." The setting specificity or lack thereof of a Prestige or Epic Class isn't important, the only thing that's important is that your class gives answers to a range of expected problems at the levels you are expected to face them.

My thought would be that Prestige Classes become mandatory when armies are marching around and you need abilities that affect Kingdom Management and mass battles, and that Epic Classes become mandatory when you start facing opponents that are massive or space-based or in some other way completely immune to swording even if those swords are magical and wielded by heroes or very numerous and wielded by hundreds of disciplined Hobgoblins at a time.

You need to have a Prestige Class at this point.

You need to have an Epic Class at this point.

Now it's important to note that there are some character concepts that could scale as-is to both of those is "Sorcerer." A Sorcerer is within his character concept doing cantrips and dispelling curses and using legerdemain to hide babies and pigs - like Willow. Such a character could easily be at home in a low level adventure and be obviously a Sorcerer in concept. But with bigger sorceries, the same character could still be in the same character concept and just turn an enemy army into pigs or shoot a black death bolt that ends the unnatural life of the evil ice mountain. That could still be conceptually a Sorcerer. However, what needs to happen to make the Prestige and Epic classes have traction is for the Sorcerer to be forced to take them as well. That the base class Sorcerers spells simply don't go to 11, and you have to become a Witch Queen or an Arch Mage or something if you want to get the larger spells that can deal with the larger scale problems.

Because if the Ranger is being forced to become a Ghost King, the Sorcerer has to be forced to become a Witch Queen at the same time. Not because it's conceptually required for the Sorcerer concept, but because it's social dynamically required to not have to tell the players that one player's character concept isn't good enough to keep playing and the other player's concept is.

On the other hand, fighting types will want to be able to write "tactic: snow ambush!" on their character sheet, while wizards will indeed want to write "spear of the rimtursar", so you don't have to create a new party to use the content in It's cold outside: Tokyo Drift.

This is true. You're going to want to have some of the advancement options be available to a character even though they have already settled on a Class to Prestige Class progression from other books. So if someone is playing an Enchanter who intends to go into Archmage, they should still be able to use some of the stuff in your new version of It's Cold Outside.

But you definitely want to avoid the dynamic of previous editions where the Cleric kept getting more usable options in every book and the Ranger got a rock.

D&D 3rd edition catastrophically failed to solve this problem with feats

Yes. You might be able to right that particular ship by giving people a feat every single adventure. Like, more feats than they have character levels. The primary reason that feats like Constant Guardian and Cautious Attack are worthless is that they are in fact extremely terrible and arguably not even worth having if they were free. But things like Deceptive Illumination and Imperious Command are actually kinda good - you just still never take them because you're only ever going to see 3 feats in the likely length of the campaign and there's no fucking way you get deep enough into your bucket list that you are going to get there.

If you got feats raining down on you, shit like Snatch Weapon would sometimes come up. And then writing weird niche feats like Sea Legs and Sand Runner would be things people would actually consider writing on their character sheet. And that would make those sections of books like It's Hot Outside and It's Raining Outside have actual value to players other than the primary casters.

-Username17

[hyzmarca]

I find the classssplosion to be incredibly annoying, personally. As someone trying to pick a class, when you have to buy about 50 to get all my options, and read those 50 books, that's sort of daunting. 3.x multiclassing makes it worse by forcing players to cross-reference all those books looking for synergies.

The classsplosion sort of defeats one of the main points of having classes at all, providing players with a ready-made package of abilities so that they don't require system mastery to be effective. At the point where you have 50 books full of classes with unique abilities and slightly different progressions, you're probably better off just going classless and just letting people pick whatever level-appropriate ability they want each level.

[Mord]

Yes. You might be able to right that particular ship by giving people a feat every single adventure. Like, more feats than they have character levels. The primary reason that feats like Constant Guardian and Cautious Attack are worthless is that they are in fact extremely terrible and arguably not even worth having if they were free. But things like Deceptive Illumination and Imperious Command are actually kinda good - you just still never take them because you're only ever going to see 3 feats in the likely length of the campaign and there's no fucking way you get deep enough into your bucket list that you are going to get there.

How does a player manage the cognitive load of having that many fiddly shit buttons to press on their character sheet, on top of their basic class features?

[Username17]

I find the classssplosion to be incredibly annoying, personally. As someone trying to pick a class, when you have to buy about 50 to get all my options, and read those 50 books, that's sort of daunting. 3.x multiclassing makes it worse by forcing players to cross-reference all those books looking for synergies.

The classsplosion sort of defeats one of the main points of having classes at all, providing players with a ready-made package of abilities so that they don't require system mastery to be effective. At the point where you have 50 books full of classes with unique abilities and slightly different progressions, you're probably better off just going classless and just letting people pick whatever level-appropriate ability they want each level.

The question of how much content you should produce is certainly an open one. The days of there being a severe bottleneck at the production end of RPG materials are completely over. Between ubiquitous word processors and online distribution, even second tier companies can produce more text than you're willing to read. In January of 2006, White Wolf released a Vampire book (Bloodlines: Legendary), a Mage book (Legacies: Sublime), a Werewolf book (Blasphemies), and a generic WoD book called Armory. That works out to a whole book of player facing options for every week of the month. That's clearly too much. And yet, that's not even the most ridiculous month of White Wolf, that's just the first month I happened to look up from White Wolf's shovelware phase.

On the other hand, producing no player facing content doesn't go over well either. WotC's current shtick of releasing one sad-sack book every six months is a fucking joke. Clearly some compromise has to be reached. You have to produce some amount of player-facing content that is between 1/6th of a book per month and 4 books per month. Fortunately, there are a lot of numbers in between those two extremes. You could produce monthly books, or even books every other month.

Regardless, once you've committed to making a book, you have to set aside some amount of content that is accessible in different contexts. Some of it can be new advancement options to aspire to, but some of it is necessarily going to be new options for new characters. And this being D&D we're talking about, that means Races and Classes. It can also mean Backgrounds and 1st level options like new armors and regional weapons and shit, but mostly it's going to be new Races and new Classes. Indeed, you're going to want to introduce new Classes in pretty much every fucking book, and new Races in most of them.

Yes, there's very real option paralysis that comes with having too many available classes. You shouldn't introduce classes thirty at a time (although obviously you could). You should let them trickle in so that players get used to the options already on the table before you introduce more. But introducing like 4 new classes per book is both reasonable and achievable.

-Username17

[Foxwarrior]

So... your ideal is to make the game excessively bloated (by hyzmarca's standards) after about five years. I'm still inclined to believe that many other good things will be sacrificed long before the setting hits fifty books, but okay.

After all, all these books will end up online in one way or another, so it won't go like: campaign goes into the snow -> that group buys the Snow book -> players grab some Snow powers. It's more likely to go like: campaign goes into the snow -> one person sees something interesting in the Desert book, another sees something excessive in the Illithid book -> players grab some powers that have nothing to do with the campaign world so far except that those powers are useful to the characters they're playing.

[OgreBattle]

How does a game like Shadowrun where all character concepts revolve around the same universal mechanics for things like magic and sneaking sell new books

[Koumei]

How does a game like Shadowrun where all character concepts revolve around the same universal mechanics for things like magic and sneaking sell new books

Pretty sure they flesh out parts of the setting with the advanced story line, and then sell issues of Gun Porn Monthly.

[Hadanelith]

I *wish* it was Gun Porn Monthly. That would be both a reasonable thing, and presumably include a decent quality level. Instead, Shadowrun releases are...miserable bullshit lately, really.

[Ancient History]

Monthly content for RPGs used to be stuff like Dragon and Dungeon Magazines. Stuff that wasn't playtested much, varied in quality, but varied also in content. Lots of good material, lots of crap.

[mlangsdorf]

How does a game like Shadowrun where all character concepts revolve around the same universal mechanics for things like magic and sneaking sell new books

The GURPS Dungeon Fantasy line is a points-based, universal mechanic game with theoretically free builds, and sells expansions with:

* GM advice
* monster catalogs
* treasure catalogs
* new sample builds (templates) and variant powers
* pet catalogs
* alternate rules systems (magic, so far)

it's a mostly PDF line, so the releases are short - 20-60 pages. But there's been 25 of them, so that's enough for 6 books at 128 pages each if SJ Games decided to consolidate them.

I'd say they're running out of steam, but they just did their first traps book and can probably do 2-3 more, and the expansion books for the basic templates have proved popular and there's another 5-6 of those they could do easily.

[Mask_De_H]

So... your ideal is to make the game excessively bloated (by hyzmarca's standards) after about five years. I'm still inclined to believe that many other good things will be sacrificed long before the setting hits fifty books, but okay.

After all, all these books will end up online in one way or another, so it won't go like: campaign goes into the snow -> that group buys the Snow book -> players grab some Snow powers. It's more likely to go like: campaign goes into the snow -> one person sees something interesting in the Desert book, another sees something excessive in the Illithid book -> players grab some powers that have nothing to do with the campaign world so far except that those powers are useful to the characters they're playing.

One man's excessively bloated is another man's wealth of options. The needle being threaded is releasing enough content slowly enough that the player base gets acclimated to it instead of being overwhelmed by options. Because shit man, there are people who think 5e is bloated with its fuck-all amount of options. Everyone's threshold is different, but aiming low has failed in elfgames way more than aiming high has.

And having players pick stuff from different books isn't a bad thing; as long as the MC can use stuff from It's Cold Outside to run their snow sessions and there's something in the book PCs can use in general, the book's done it's job.

[Username17]

Monthly content for RPGs used to be stuff like Dragon and Dungeon Magazines. Stuff that wasn't playtested much, varied in quality, but varied also in content. Lots of good material, lots of crap.

When WotC shut down Dragon and Dungeon, that was a terrible thing. The reasoning was sound: obviously at some point D&D should convert its magazines to an online format and simply have subscribers get access to truly ridiculous amount of D&D content with Daily Dungeons, genre fiction, experimental character classes, Treasure Tuesdays, and a special adults-only section where the Minotaurs are drawn with penises.

The thing was that WotC was obviously unprepared to produce the amount of content that a real "online magazine" would require. Such a setup needs to have like 64 page equivalents a month, which means that you need like 15-20 thousand words every week split up into 10-20 articles so you have content every day. As it was, only Robert Schwalb was actually keeping up his share of Dragon writing, and no one was doing Dungeon writing at all. That's fucking bullshit.

People need to feel that they could log into Dragon every day and have some new D&D content to read. Every day. If you're going to make a subscription online magazine, it has to be something that people can link to their RSS feeds for updates and shit.

But not only do you need more than one person turning in articles on time, you have to take the idea of doing all the content in-house and laugh it the fuck out of the room. It's D&D, people make content for the game just by playing it. Obviously you should take reader submissions, and you should publish the best reader submission most days.

-Username17

-Username17

[Hadanelith]

By the *gods* I miss Dragon. Going and getting the new issue was often the highlight of my month in middle and high school.

[tussock]

The biggest problem for 3.5 with classplosion seemed to be support.

New books would have 2-4 new classes, but also some options for the pre-existing classes, which inevitably were power-ups. So they have to put out stuff for the expanded world of classes now and then too, so as they can keep up a bit.

Which is all fine, with 13 classes, and OK with 30 where most of them use the same options to buy their powers anyway (feats or spells or items).

But got trickier once there was classes that could supercharge a sub-group of those options (because then that option group is balanced against the supercharging class and useless for everyone else), or had their own thing like the Binder where your content is maybe interesting to the group with the guy playing a Binder, only probably not even then (and also eats page space like crazy, in that case).

--

But if you go all special content containers for each class, Barbarian rages, or Rogue tricks, or Paladin smites, or, well 4th edition D&D ... well, eventually you can't support most of that and every group is unhappy because there's a bunch of content they can't even use at all and the stuff some of them want (Warlocks need new 5th level Daily Power!) isn't even in the new book because even less people would be able to use it.

Which is a bit of the Gnome problem, in that a large proportion of groups have one of those classes that almost no one is playing being played in their group, and want support for it to continue, only out past even 13 classes, it can get tough to support in a way that makes anyone happy.

--

Or you can just have a Type XII Wizard class where the spellcasting is the same list and maybe less evocation and necromancy, but more illusion, but different numbers and recovery rates on the spell slots, and the Fighter VII gets a mix of situational damage bonuses, some bonus feats, and a few spells, that are all super-easy to add to his lists in later books in a way everyone can enjoy (like, as a spell: expand situations where damage bonus applies).

But which ultimately doesn't support all that many classes before they all blend together and the names stop meaning anything, which is a very bad thing.

Class names should definitely carry meaning, that's part of the point. So I think there's real limits to the number of classes you can make and support in a way that keep your customers happy, though your class design need not be particularly shit for support like 4e D&D.

[Grek]

A very easy way to fix that problem is to have distinct (but not setting- or role-based) clusterings of classes that all go in the same books together.

Let's say that, after the base book is out of the way, you decide to put out a Nature themed book. It includes new wilderness monsters, art with lots of trees in it, advice for the DM on how to run adventures set in the middle of nowhere and, relevantly to this thread, four new base classes (Shifter, Shaman, Ranger and Berserker); four new prestige classes (Green Mage, Coven Witch, Chieftain and Fey Blessed) and four new epic classes (Treelord, Faerie Queen, Untamed Avatar and Once and Future King). Later, when you revisit the Nature theme with your Fey Book and your Jungle Book and your Plants book, each one introduces new content for each of the prior classes, plus ONE new class of each level.

Meanwhile, you're still doing the same thing in the other themed clusters. Desert comes in with Pyromancer, Summoner, Thief and Prince, but none of these provide any new options for Book of Trees classes. Ocean has Sea Witch, Cultist, Musketeer and Pirate but makes no mention of either Summoner or Berserker. And so on and so forth. You keep the combinatorials down by doing multiple smaller classplosions in parallel instead of letting everything mix with everything.

[jt]

The Book Of Nine Swords has the start of a good system for dealing with classplosion support. Maneuvers are divided into schools, some strictly martial, some divine magic themed, some explicitly magical. The classes have different ways of recovering maneuvers, but more importantly, they have access to a different subset of the schools.

If you expanded that to daily-tier (like spells), encounter-tier (like maneuvers), and at-will-tier (something new) lists, each divided into schools, that gives you tons of nobs to tweak classes by just fiddling with which lists they have access to and how they recover them. A pure martial class with mostly at-will abilities is a fighter, one with mostly encounter abilities is a warblade, one with mostly daily is a barbarian. And you can fiddle with recovery a lot - you can give someone an encounter-like recovery mechanism for something that's normally at-will as their secondary shtick.

Then if you come up with a new archetype that wants to flesh out the daily martial abilities more, the barbarian is automatically happy.

Casters already almost have this (thus getting more support than everyone else), but they make the mistake of giving each access to all the schools by default, thus making it harder to use this system to carve out any niche. If the default was more like the beguiler (illusion/enchantment only), then you have a lot more room to play with, because you don't need to add whole new schools to make something unique; each combination of existing schools is unique already.

[Ancient History]

If you expanded that to daily-tier (like spells), encounter-tier (like maneuvers), and at-will-tier (something new) lists,

At-will is basically psionics.

The thing with resource-management is that D&D 3.x kept piling on different resource-management schemes, without any sort of comprehensive management plan. That rarely led to any particularly overpowered nonsense where you could get $TEXAS bonuses for <given expenditure>, but it did get really fucking messy and annoying trying to keep track of how many times per day you could do X, and then get Y feats/magic items/class features that let you spend a daily use/whatever for <some effect>.

The best you can say about it is that they deliberately avoided doing a thing where you could swap ki points for power points for spell slots or daily rage usage or things like that. But they still had way too many bullshit little feats/class features that worked to create their own little resource pool *cough* FATESPINNER *cough*.

[Emerald]

Along the "group abilities into disciplines" and "don't give every new class its own fiddly resource system" lines, something along the lines of cleric domains (where you get several new domains in various books, but individual characters can only select a handful of domains) are possibly a good example of how to expand things incrementally without giving a single class all the new options.

You could reverse the current cleric situation, so that clerics primarily drew their spells from domains (which could be expanded so that each is like multiple 3e domains, with several granted powers and multiple spells of each level) and had a relatively small "generic divine caster" shared spell list that was expanded only rarely, rather than a massive shared list with only a handful of differentiating spells and powers in domains, and modify all divine casters to use the same spellcasting setup. Then instead of adding new cleric and druid spells in new books and neglecting paladins, rangers, and non-core divine casters, you'd add new domains that are each available to a subset of divine casting classes--a Sacred Exorcist domain for clerics and paladins, a Holy Warrior domain for paladins, blackguards, and rangers, a Spirits domain for druids and shamans, and so forth.

That would help avoid neglecting later classes, because it's easier to assign new domains to non-core classes as each new book comes out than to comb through a grab-bag of spells to determine which ones deserve to be on the shugenja list or whatever, and you don't overpower core casters because even if you did allow clerics and druids to select from any domain (which you definitely shouldn't) they wouldn't have a constantly-expanding One Spell List to Rule Them All anymore.

And heck, instead of a single shared list for generic divine spells, you could just have a few generic domains every class gets, the same way all three Tome of Battle classes get access to Stone Dragon, so you wouldn't need to go outside the domain structure at all. It also opens up the possibility of very focused domains to flesh out single classes or single patrons, like Initiate feats do in 3e, so a paladin might have a Paladin domain only it gets for archetypal paladin spells and powers, some thematic Holy Warrior/Divine Blade/etc. domains that it and several other classes get for "armored divine caster" spells and powers, and some generic Priest/Divine Servant/etc. domains that all divine casters get for a tiered class group/class/subclass setup.