What Are Classes?

[GreatWyrmGold]

I've been involved in a few discussions on this forum which touched on, to some degree or another, more or less clearly, the question of what classes are in D&D. I don't mean in a mechanical sense—classes are the basic building block of characters, that much is obvious. The question being touched on is one of what makes classes what they are, why they exist as they do. (Beyond "They did it in AD&D and nobody's bothered to change it since," I mean—we're talking philosophy, not history.)

To explain what I mean, I'm going to point to two ancient classes which lie on opposite ends of the spectrum, Ranger and Fighter.

The Ranger has a very strong theme. The Ranger is Robin Hood, Aragorn, Drizzt. A woodsman who knows the land and knows his prey, who uses cunning and nature lore to defeat great foes. The class is pure archetype, a set of mechanics outlining a familiar fantasy form. But those mechanics are unremarkable. It's unclear what role a ranger is expected to fill in a party; it's less fighty than the fighter and less skillful than the rogue, without any truly distinctive powers of its own.

On the other hand, the Fighter has a very strong mechanical role. (Those mechanics tend to be a bit generic, and in some editions are kinda pathetic—3.5, I'm looking at you—but the intent, at least, was clear.) The Fighter is a warrior, durable enough to weather the storm and strong enough to whip up one of their own. But its flavor is thin; nothing defines a fighter's character beyond the fact that they fight. Theoretically this gives players the freedom to attach a wide variety of character archetypes to the fighter chassis, but in practice most players either don't need to be told they have that freedom or make generic warrior characters when they play Fighter. The Fighter is pure mechanics, with no flavor or archetype to call its own.

The Ranger and Fighter are practically polar opposites from a high-level design perspective. The Ranger exists as a set of mechanics to enable a specific archetype; you can use those mechanics to build a character who doesn't resemble the archetype, of course, but that's not what the class was designed for. The Fighter, on the other hand, exists as a set of mechanics to fill a specific mechanical role; you can bolt almost any archetype on it, but for better or worse it won't be reflected in the mechanics.

Neither of these class design paradigms are inherently bad. Having archetypical classes like the Ranger guarantees that each of your classes will feel distinct; each class will have a set of mechanics designed to evoke a certain type of character, to fulfill that fantasy most thoroughly. Having role classes like the Fighter guarantees that each of your classes will have a place in the game, while giving players with a character type in mind maximum freedom to pick whatever mechanics they want. (In both cases, obviously only if the classes are done right, but the question of if/how it's done right is beyond the intended scope of this discussion.)
Most classes in D&D are somewhere between those two extremes. The monk has a distinct kung fu flavor, and combines unarmed/unarmored combat with a generally mobile combat style. The paladin is the archetypical knight in shining armor, and serves the slightly messy role of a tank/support caster/occasional nova-striker. The rogue is a trickster or crook, and fills in the party's noncombat needs (with a few sneak attacks in combat, of course). But the tension between those two ideas of what a class is can be felt even in the classes with some sort of balance. What's the barbarian doing as a class when fighters are already so good at, well, fighting? What makes a druid conceptually different from a cleric serving a nature deity? Why are the classes divided the way they are?


What are classes? Are they classic character archetypes with mechanics built to match? Are they mechanical roles with appropriate flavor attached? What should they be?

This question can be felt in a wide variety of discussions. Discussions about why class X doesn't have fighting style Y, about how Class Z should best be played, and (of course) about what classes could be removed from the game. But I've never seen it spelled out so clearly and discussed on its own. So I wrote this thread.

So I open the floor to you. What the heck are classes?

[Anonymouswizard]

Fighters honestly can have more flavour, but it's still a bit generic. Fighters are those who have dedicated their lives to mastering weapons to the exclusion of practically anything, until they attain nearly unheard of skill with one or more. This is sadly a flavour D&D seems somewhat unwilling to build into the class, but a Fighter should be Lancelot, or at the very least Boromir (skilled enough with sword and shield to kill a great number of orcs before dying). Every edition of D&D sucks at this flavour except for 4th.

Okay, to me the ideal example of what the Fighter class should be is Guts. He's a man who has pushed his body and skill to the point where he does the inhuman, including taking down 100 men by himself in a brutal skirmish. A man so skilled with two handed swords that he can wield one that weighs as much as him as if it was a hand and a half sword.A man who went three years with practically no sleep because demons attacked him every night.

So yeah, my view is that classes are archetypes instead of skillsets, and the fighter needs an overhaul so that 'ordinary soldier' is not part of it. A Barbarian powers through on rage, and a Monk medidated under a waterfall until they understood the world, but a Fighter trained continually for years or decades until they can leap out of their saddle, catch an arrow, draw their blade, and disarm you all in the same round. And look cool (if not necessar5ily stylish, that's a bard's job) while doing with.

Higher level fighters are doing all of that blindfolded with only a dagger.

[Mechalich]

The Ranger and Fighter are practically polar opposites from a high-level design perspective. The Ranger exists as a set of mechanics to enable a specific archetype; you can use those mechanics to build a character who doesn't resemble the archetype, of course, but that's not what the class was designed for. The Fighter, on the other hand, exists as a set of mechanics to fill a specific mechanical role; you can bolt almost any archetype on it, but for better or worse it won't be reflected in the mechanics.

I disagree. Fighter and Ranger both enable archetypes, it is just that the conceptual space encompassed by 'fighter' is extremely broad and the conceptual space encompassed by 'ranger' is much narrower. Certain other classes, notably paladin and druid, are narrower still. Mechanically the Fighter and Ranger have the same role in D&D - they're both DPS classes (the ranger's utility options have become more important over time, but their initial function was extremely modest).

Generally D&D classes match poorly to 'mechanical role' because D&D has minimal role protection. It really only has three supported roles: DPS, healer, and skill monkey, with the latter being extremely situational - if you pay any isometric D&D video game you know you need a thief, but mostly the thief functions as a DPS until its time to check for traps or open a lock.

A game can use classes as a means of sorting character concepts into roles. This works well for video games that tightly constrain inputs and outputs but is less effective for tabletop, especially since the natural division between combat/non-combat that exists in real life isn't something you can tolerate at tabletop because it leads to players sitting around the table with nothing to do for huge stretches of the game.

Consequently tabletop classes are primarily used to enable archetypes, this makes sense when the gameplay space is appropriately tailored so the game needs to only support archetypes that make sense within the space. Shadowrun, for example, is a game about going on Shadowruns, and the game only needs to support characters who would plausibly be undertaking that activity, you don't need to represent everyone else on the planet with PC classes. D&D is supposed to be about adventurers who go on dungeon crawls (yes sometimes the 'dungeon' is a wilderness or a slum or whatever) and so the archetypes supported should be concepts tailored to the dungeon crawling experience. Unfortunately D&D hasn't exactly held itself to this very well and as a result there are D&D classes that are all over the place, but that's mostly a 'D&D is messy' thing rather than any sort of design principle.

[Lord Raziere]

What are classes? Are they classic character archetypes with mechanics built to match? Are they mechanical roles with appropriate flavor attached? What should they be?

This question can be felt in a wide variety of discussions. Discussions about why class X doesn't have fighting style Y, about how Class Z should best be played, and (of course) about what classes could be removed from the game. But I've never seen it spelled out so clearly and discussed on its own. So I wrote this thread.

So I open the floor to you. What the heck are classes?

Honest Answer? They are Inconsistent. because they are incoherently designed.

some classes fall into "Clear Archetype" version, and some into "Generic Mechanics". To use DnD 5e:

Clear Archetype: Paladin (knight in shining armor), Ranger (Robin hood), Monk (kung fu flicks), Druid (magic nature person), Wizard (Merlin), Warlock (Deal with the Devil), Bard (Spoony Bard), Barbarian (Hulk/berserker trope) Sorcerer (younger more reckless mages)

Generic Mechanics: Fighter, Rogue, Cleric

why did I put Cleric on the generic side? because honestly there is no specific fantasy archetype that it embodies. you can argue that its a white mage, but the current Dnd vision of it is that what your actual focus is depends on the deity you worship and thus the magic they can do and thus what archetype they are varies from cleric to cleric. while rogue has this criminal flavor its actually tremendously flexible in terms of what it can allow to emulate. sure thieves cant is pretty unavoidable, but you can pass that off as the character just being very scholarly and able to decipher their language, despite not being a thief yourself. the rest is just a matter picking the appropriate skills.

But overall I'd say that DnD is more on the archetype emulation side than the generic mechanics side. fighter and rogues are just in the unenviable position of having to be normal people among a bunch of classes that all have some clear supernatural flavor.

but this is general roleplaying. perhaps we should look at another system other than DnD to see how they did classes? like....the fantasy flight Wh40k games. in them, all the classes are real in-setting positions and terms with clear places in society, to use Dark Heresy:
Adept (Nonmagical scholars, with lots of knowledge and skills and also investigators and researchers)
Arbitrator (Enforces law, are basically police)
Assassin (kills people stealthily for hire. exactly what it says on the tin)
Cleric (no magic. inspires, oratory, general spiritual person keeps up morale mainly)
Guardsman (is a soldier, basically, shoots and fights)
Imperial Psyker (handles the magic of the setting so no one else has to)
Scum (is a criminal and does underhanded things of a scoundrel nature)
Tech Priest (handles the technology of the setting)

all of these have clear methods of progression that make sense in setting, with a bunch of titles for this and that class for how much they leveled in their career. but how they advance is half skill-based, and their skillsets are wider than they seem. often they have secondary skills to round out their primary ones in the things they can advance. you may wonder why Assassin is a separate class from Guardsman or Scum but rest assured an Assassin is someone specifically on the path to become a member of the Officio Assassinorum a very important and powerful organization in Wh40k, which the other two are not. how they all fit together as a group is pretty obvious, as they are all investigators for the Inquisition to kill threats to mankind, and each class has some use for that and an established place within the setting. not all of them are equally good at combat- Adept is more useful for the knowledge they have than any lasgun they fire-but they can all contribute in some manner, and all they all have a distinct identity. the difference between playing a fighter in DnD and playing a Guardsman is that the guardsman is probably from the Imperial Guard which has certain assumptions tied into it.

so it could be that DnD classes are incoherently designed, because its setting isn't defined enough. Wh40k for its all grimdarkness is able to establish clear identities for all its archetypes and detail all their jobs and lives in great detail so that you know what each person does IC and OOC. there is some slight wiggle room but in the end the path you choose is pretty clear on what you'll do, where you'll end up, what skills you have and what your place is. if your that thing, your that thing in setting in Dark Heresy with no room for doubt.

for DnD it can be a bit harder to figure out what exactly is a classes place within a setting or if they even have a distinct word for them. some places are clearer than others but the classes even the defined ones are pretty floaty and unbound in terms of what they're supposed to be in relation to them. thus its not always clear what leads to a life with the skills and abilities of that class, or what they're supposed to do with their lives.

So I think classes if they are to exist, best work when they are tied into the setting and have a very well defined place in them.

[Anonymouswizard]

Generally D&D classes match poorly to 'mechanical role' because D&D has minimal role protection. It really only has three supported roles: DPS, healer, and skill monkey, with the latter being extremely situational - if you pay any isometric D&D video game you know you need a thief, but mostly the thief functions as a DPS until its time to check for traps or open a lock.

Sadly, the one time D&D included specific (combat) roles for classes it got a very negative reaction. And I think the same happened with Pathfinder 2e sliding casters out of blasting and tanking and solidifying them as battlefield control. Reduced skill selection has also helped let players define more clear outside of combat roles, but it's also lead to a skill tax problem *everybody wants Perception, and a good number of low skill classes also want Athletics).

Honestly D&D could do much worse than going back to the Defender/Leader/Striker/Controller(/Artillery) setup, although they might want to leave it out of the books and keep it as just a design process. It would give each class a mechanical purpose, and then they could go through the big list of fantasy archetypes and assign one that fits the class. (Also bring back Racial Powers, they did better at differentiating races than stat boosts.)

4e was very good when it came to assigning classes mechanical roles, and somewhat decent at giving them archetypal ones. There were of course problems, but each class had a decent place and it gave the poor Fighter a bit more of an identity.

[Cluedrew]

What are classes? [...] So I open the floor to you.

Assorted cool ideas picked at random.

People talk about them being archetypes but really... are they? "The Knight in Shining Armour" might have been an archetype but the holy powers thing doesn't seem to be common until the Paladin came along. The Cleric isn't really either and was created out of a need to have a healer in combat, with a hint of vampire thrown in apparently (maybe that's where turn undead came from?).

There was no centralized design philosophy, and it there was it wasn't updated as the editions went on (except for 4th).

[Saint-Just]

but this is general roleplaying. perhaps we should look at another system other than DnD to see how they did classes? like....the fantasy flight Wh40k games. in them, all the classes are real in-setting positions and terms with clear places in society, to use Dark Heresy:
Adept (Nonmagical scholars, with lots of knowledge and skills and also investigators and researchers)
Arbitrator (Enforces law, are basically police)
Assassin (kills people stealthily for hire. exactly what it says on the tin)
Cleric (no magic. inspires, oratory, general spiritual person keeps up morale mainly)
Guardsman (is a soldier, basically, shoots and fights)
Imperial Psyker (handles the magic of the setting so no one else has to)
Scum (is a criminal and does underhanded things of a scoundrel nature)
Tech Priest (handles the technology of the setting)

all of these have clear methods of progression that make sense in setting, with a bunch of titles for this and that class for how much they leveled in their career. but how they advance is half skill-based, and their skillsets are wider than they seem. often they have secondary skills to round out their primary ones in the things they can advance. you may wonder why Assassin is a separate class from Guardsman or Scum but rest assured an Assassin is someone specifically on the path to become a member of the Officio Assassinorum a very important and powerful organization in Wh40k, which the other two are not. how they all fit together as a group is pretty obvious, as they are all investigators for the Inquisition to kill threats to mankind, and each class has some use for that and an established place within the setting. not all of them are equally good at combat- Adept is more useful for the knowledge they have than any lasgun they fire-but they can all contribute in some manner, and all they all have a distinct identity. the difference between playing a fighter in DnD and playing a Guardsman is that the guardsman is probably from the Imperial Guard which has certain assumptions tied into it.

so it could be that DnD classes are incoherently designed, because its setting isn't defined enough. Wh40k for its all grimdarkness is able to establish clear identities for all its archetypes and detail all their jobs and lives in great detail so that you know what each person does IC and OOC. there is some slight wiggle room but in the end the path you choose is pretty clear on what you'll do, where you'll end up, what skills you have and what your place is. if your that thing, your that thing in setting in Dark Heresy with no room for doubt.


for DnD it can be a bit harder to figure out what exactly is a classes place within a setting or if they even have a distinct word for them. some places are clearer than others but the classes even the defined ones are pretty floaty and unbound in terms of what they're supposed to be in relation to them. thus its not always clear what leads to a life with the skills and abilities of that class, or what they're supposed to do with their lives.

So I think classes if they are to exist, best work when they are tied into the setting and have a very well defined place in them.

Surprisingly 2e D&D (oldest that I read, though never played) talks about clerics as being modeled on Templars and the like, so not your average priest (that's why the heavy armor, martial weapons and so on). Which makes it almost redundant on the conceptual level with Paladin (and even though by 3e it went somewhat separate ways you still see the suggestions of replacing the Paladinb with ACF'd Cleric or converting it to the prestige class - including in the Unearthed Arcana).

I do agree that Dark Heresy's (I am somewhat less acquainted with other product lines) advancement is less arbitrary because it's grounded in the institutions of the Imperium. I think the piecemeal leveling also helps very much (to those who hadn't read it - it's a restricted point-buy, you can learn a new skill or talent or increase one characteristic for 100 XP well before you achieve 2nd rank, and when you achieve that rank you don't get any automatic increase, you get more and stronger options to purchase for more XP, so you don't have sudden jumps in power and also there is no dead weight - if the option (like increased HP, or more skill with melee weapons, or ability to track) is not helpful to the character then the player does not take it). But even if you notice that "Guardsman" is not necessary from Imperial Guard, nor Adept is not necessary from Administratum it still doesn't allow you to play Imperium:the RPG. Dark Heresy's classes work better than D&D because the authors had chosen a simpler task and did it well. Unless you treat D&D as primarily a tactical game where character options doesn't need to emulate anything besides themselves then even exploring the dungeons will likely involve more diverse people not necessary reducible to the core classes. And then it gets much worse when you try to emulate other fantasy plots - courtly intrigue, monster-hunting with a focus on hunting etc. Right spell can solve anything but that just creates another pitfall.

In the end I do agree that strict classes work better in a narrow situation - one setting, and probably a particular tone of game also.To properly cover different times or different tasks you'd need a new set of classes (that's kind of what FFG did with Wh40k RPGs - they made a separate one for at least five different "taskforces" - and yes, it can be a desire to extract more cash from gamers, yet I do not think it was solely motivated by that). Alternatively you need a classless system, or at least potentially classless one - where your starting packages are limited but potentially you can learn anything regardless of starting class. Those can have their own issues, of course.

[Mechalich]

Sadly, the one time D&D included specific (combat) roles for classes it got a very negative reaction. And I think the same happened with Pathfinder 2e sliding casters out of blasting and tanking and solidifying them as battlefield control. Reduced skill selection has also helped let players define more clear outside of combat roles, but it's also lead to a skill tax problem *everybody wants Perception, and a good number of low skill classes also want Athletics).

Dedicated combat roles tend to feel very 'gamey,' meaning they often demand rather extreme bits of suspension of disbelief that can be tolerated in video games (often because they are obscured mechanically), but become rather ridiculous when employed at tabletop. Tanking, in particular, tends to rely on very arbitrary mechanics - 'I shout really loud and now all the enemies who attack anyone other than me suffer a damage penalty' - that don't mesh with any concept of immersive roleplay.

There's also the issue that, if you want highly engaging MMO-style tactical combat you should just play an MMO. Advances in technology have made computer games simply better at offering certain types of experiences than tabletop can ever be. This is a natural progression in entertainment mediums. In the late 19th and early 20th centuries live theater was dominated by epic melodramas like Ben Hur and The Count of Monte Cristo, but theaters no long put on such pieces because they simply cannot match the kind of energy and spectacle of the film versions.

The generalized trend towards rules-lite and narrative TTRPG systems is indicative of this, since they focus on the parts of the tabletop experience that works better around a live table rather than on a game server (even D&D 5e represents a significant simplification in many ways). Of course, such games have less need for a class system because the greatest utility of classes is in simplifying mechanical complexity by pre-selecting options along a progression.

[Lord Raziere]

Assorted cool ideas picked at random.

Which really is my entire point. classes again work best when they are well-defined parts of the setting. when they are chosen not for random coolness but for their functionality not in an abstract mechanical sense but in an in-setting sense. the Wh40k classes/career paths are full of things chosen for functionality. I've already detailed Dark heresy, so lets how we can makes classes to be functional in an adventuring DnD like world, discarding the normal archetypes.

What do we need?
-to fight things
-to get rid of traps since that seems to be a common problem, much like Indiana Jones
-to deal with magical problems
-investigate things to learn more about them, whether it be in old ruins or otherwise
-to explore and traverse dangerous environments
-to negotiate with people you've never met before as your constantly traveling

so the classes might be something like:
-Archeologist (a knowledgeable person specifically designed for delving into dungeons, getting rid of traps and whatnot)
-Explorer (a person who knows how to navigate wilderness, chart out paths and some fighting stuff along the way)
-Merchant (a social face who is also a common traveler and goes looking for stuff in ruins to sell)
-Bodyguard/Mercenary (a bodyguard who protects the other guys with their ability to fight however that is)
-Field Mage (an archetype of a magic user explicitly less about academic stuff from books and more about getting hands on knowledge in the field)
-Missionary (a divine person if we want to replace cleric who travels to other lands spreading their faith, with social skills to match)

now, Archeologist could probably be rolled into the Explorer, but you see what I mean. these are all more clear about the classes purposes and what they're supposed to be. "rogue" or "fighter" is vague but "Explorer" or "Bodyguard" has explicitly connotations about exploring wild untamed lands or protecting someone else. Field Mage focuses on the wizardly archetype into someone who goes for magic that will be useful for adventuring, while missionary focuses divine magic into a reason they are traveling around. its IC functional, and thus can easily be OOC functional as well. there is no technical need for adventurers to involve themselves in often society-based underworlds of crime, so rogue is out and you get explorer which makes much more sense for what DnD adventurers do. now there is the argument that these archetypes are less general and more focused and thus cut off other campaigns, but I'll wait for people to respond with the kind of campaigns they'd want that this list would cut off before thinking any further on it.

It could probably use some work, but a more IC functionality and defined place for the classes would probably do wonders.

[OldTrees1]

At their root classes are a delivery mechanism for mechanics used by a class based game as a tool to help the player instantiate their character concepts.

So, if that is the goal then how are they designed?
1) Players have character concepts.
A class with characterization (strong theme) will have mechanics that synergize to reinforce that characterization and enable the characterization to continue or evolve to meet the new demands of higher and higher levels. This gives new players an idea of what the mechanics will do for them and an archetype for them to consider as an example when crafting the characterization of their character concept.

A class without characterization is going to be a blank slate. That provides no help to a new player working on the characterization of their character, nor on finding which mechanics help instantiate their character concept. On the other hand a blank slate does not get in the way if someone has a better grasp of their characterization and what mechanics they are looking for.

So generally a class tries to strike some sort of idea balance where they are not too specific (Waterdeep Geology Professor) nor too general (Adventurer). There might even be a mixture where some are more specific (Ranger) and others less specific (Fighter) so there are specific classes and an alternative generic class for when the specific classes don't fit the character concept.

2) Classes are an ordered bundle of features
The Barbarian class knows the 14th level Barbarian feature will arrive at 14th level or later and will come after Barbarian 1-13. This lets the design plan ahead with some idea of context. It can design how the bundle of mechanics will evolve as the character goes to higher and higher levels.

Multiclass design is just an advanced topic / course of this design. If you allow multiclassing in your design, then you want reasonable multiclasses to also evolve to meet the expectations and demands of higher levels.



Now which concepts are turned into classes is a subjective decision on the part of the designer.

[MoiMagnus]

The main goal of classes is to guide players toward creating characters that make D&D what it is.
You can take the same rule system, but if the classes were "Noble, Architect, Merchant, etc", the game would be vastly different.

Sure all the customisation options and subclasses published in latter books mean that experienced players will be able to go outside of those tropes reasonably easily, but the point of the classes is to ensure that those tropes remain tropes.

Compare the cleric class to the wizard class. One of them emphases high level of knowledge, and does not wear any armour or battle protection, while the other is likely in battle armour and is more competent at being a field medic than at organising a religious office.
Sure, you can make a battlemage in heavy armour, and you can make a priest in robe. But those are not what the class system will push the players toward, and those kind of design choices shape what D&D is.

The goal of classes is to guide the player. In some sense, it's not a problem by itself that the classes are inconsistent, as different classes might target different kind of players.

Together with that first goal, the set of classes is also here to guide players toward creating a varied but compatible team.

All D&D classes heavily favour combat, and that's a good thing that they are homogenous on that.
It's much more a problem when one class push forward a specific kind of gameplay for the campaign, but the other classes either do not cover at all this gameplay, or at the contrary push toward ignoring it.

[Anonymouswizard]

Dedicated combat roles tend to feel very 'gamey,' meaning they often demand rather extreme bits of suspension of disbelief that can be tolerated in video games (often because they are obscured mechanically), but become rather ridiculous when employed at tabletop. Tanking, in particular, tends to rely on very arbitrary mechanics - 'I shout really loud and now all the enemies who attack anyone other than me suffer a damage penalty' - that don't mesh with any concept of immersive roleplay.

Eh, 4e tanking worked about as well as could be expected because it punished via dealing damage. But yes, there's a reason you use the 13th Age model of not giving explicit tags in the book and instead talk about the intended play style. Although I think 13th Age mainly got rid of Defenders in exchange for having them be more durable strikers.

There's also the issue that, if you want highly engaging MMO-style tactical combat you should just play an MMO. Advances in technology have made computer games simply better at offering certain types of experiences than tabletop can ever be. This is a natural progression in entertainment mediums. In the late 19th and early 20th centuries live theater was dominated by epic melodramas like Ben Hur and The Count of Monte Cristo, but theaters no long put on such pieces because they simply cannot match the kind of energy and spectacle of the film versions.

Having played tactical miniatures combat, it's still fun and enjoyable even though computer games do it 'better' (and actually they don't do so definitively).

If computer games did tactical combat better, would Games Workshop still be selling Warhammer? There's obviously something about the game people can't find on a computer.

Oh, and I've seen epic productions on stage (and heard them on radio). They can be very good, because they can't rely on the impressive visuals films use and so have to get more artistic.

The generalized trend towards rules-lite and narrative TTRPG systems is indicative of this, since they focus on the parts of the tabletop experience that works better around a live table rather than on a game server (even D&D 5e represents a significant simplification in many ways). Of course, such games have less need for a class system because the greatest utility of classes is in simplifying mechanical complexity by pre-selecting options along a progression.

Whether or not this is true is debatable, there is still a market out there for more rules-heavy games. It is the current trend of the industry, but that doesn't mean it's going to remain the trend forever.

Especially because crunchy games like Wrath & Glory and The Dark Eye still sell. So somebody still thinks crunch is fine at the game table, even if it's no longer the majority of players.

[Cluedrew]

Having played tactical miniatures combat, it's still fun and enjoyable even though computer games do it 'better' (and actually they don't do so definitively).

Until we get holographic projectors I don't think computers will be able to replicate the feeling of setting up and realizing your opponent's army is made almost entirely out of extra-large miniatures. But still designing "board games on the computer" is a thing people do, that is design computer games using the lessons learned (or restrictions forced on you) by board games.

I will say though for all the flack 4th edition gets it was the only system where combat interested in at all. I do want to try Lancer sometime which is built off that formula, or at least has the same feel.

[Morty]

Assorted cool ideas picked at random.

People talk about them being archetypes but really... are they? "The Knight in Shining Armour" might have been an archetype but the holy powers thing doesn't seem to be common until the Paladin came along. The Cleric isn't really either and was created out of a need to have a healer in combat, with a hint of vampire thrown in apparently (maybe that's where turn undead came from?).

There was no centralized design philosophy, and it there was it wasn't updated as the editions went on (except for 4th).

Correct. D&D class design is a work of accretion. Ideas were slapped on as needed, but without anything resembling a unified or coherent design direction. And after 3E, it seems no class can ever be removed.

[Anonymouswizard]

Until we get holographic projectors I don't think computers will be able to replicate the feeling of setting up and realizing your opponent's army is made almost entirely out of extra-large miniatures. But still designing "board games on the computer" is a thing people do, that is design computer games using the lessons learned (or restrictions forced on you) by board games.

Oh, I do love computerised board games, but there's something great about handling actual physical miniatures (or LEGO, depending on the game and who's running). And there's nothing wrong with using lessons learned in one medium when creating something in another, in fact it's a very good thing.

I will say though for all the flack 4th edition gets it was the only system where combat interested in at all. I do want to try Lancer sometime which is built off that formula, or at least has the same feel.

4e has a lot of good ideas that 3e threw out to appeal to the hatebase. Healing Surges, abilities that worked on a per-eno****er basis (instead of a per three encounters basis), separating combat and outside of combat magic, stronger roles for classes, attacker always rolls, three/four defences each dependent on the better of two stats, and other bits that skip my mind now. It was just combined with a (seeming) radical shift in priorities, a flawed marketing campaign, and an apparent reduction in class variation.

Oh, and Racial Powers. Those were an excellent idea.

I'm really liking 13th Age the more I read it and really want to run it. It's got most of the good parts of 4e, but with some changes (Recoveries are randomised unlike Healing Surges, defences are based on the middle of three stats, combining the three tiers into one...), while bring in more 3.X style class variation. Plus more freeform rituals, all you need to do is justify one of the effects being related to one of your spells.

Also having been annoyed by how little my 5e Battlemaster used their maneuvers, I'm really liking Flexible Attacks. It looks like it makes playing a Fighter fun, instead of making me worry if I'll need any of my three SD later in the battle.

[Quertus]

Well, it's complicated.

I have my own ideas on the topic, but those who have few sanity points to spare should probably skip on past my mad rambling.

Example 1 - 1 modular class

There is only 1 class, with modular options. Each level, you choose which options to take.

Maybe this level you took "wear armor", "cast in armor", and "cantrips of Tzeentch with the cost of potential sanity loss".

Or maybe this level you took "Toughness of Boromir" and "voice of an angel".

Or maybe this level you took "10' step", "parry lasers" and "Lay on Hands at the risk of Warp phenomenon".

Or maybe this level you took, "recover Hero Points by drinking blood" and "skill monkey".

Example 2 - 2 teams

Green.

Purple.

(There is no difference between these classes - they merely represent or determine which side of the conflict you are on.)

Example 3 - 3 pillars

There's 3 pillars in D&D: social, exploration, and combat.

A) what if those were your 3 classes? Your class tells you which part of the game you play.

B) what if your class was a Tristalt, where you had to choose a class to specify *how* you contribute to each pillar?

Example 4 - 4 elements

What if your class determined your acumen with each *approach* - an Earth heavy class might struggle with stealth or deception compared to a Water or Air class, for example.

Example 5 - 5 character choices

Suppose I create a new game, with the following classes:

Quertus, powerful tactically inept academia mage.

Thristen, skulking vampire skill monkey.

Bob, Beholder warrior.

Evan, the Last Son of Magnus, devout follower of Tzeentch.

Sentient Potted Plant, considers abilities like "holds items", "pushes buttons", and "moves under own power" to be super powers beyond its reach.

-----

So, what *can* classes do?

Well, *something* needs to tell you what, mechanically, you can do.

Something needs to provide a rough measure of what you're capable of.

Something needs to be used to represent how you grow over time / as you gain experience.

Something needs to indicate which "team" you're on.

Something needs to give you a role.

Something needs to give you a personality.

Something needs to indicate what you are and have the opportunity to become familiar with.

Something needs to show your knowledge and skills.

And Classes *could* (and do) cover any or all of those.

-----

How *should* classes be built?

I think that the "Quertus" class is too narrow of a focus. But I think that choosing between social / exploration / combat, or the 6 d20 stats, is… wrong on several levels, but also too vague.

Saying that you are a "Ranger" or a "Dread Pirate" is evocative. (And that, I think, is the true value of a class).

But there *also* needs to be enough generic space - classes like "Fighter" to fill in all the gaps.

So, if I want to be a Dread Pirate in 6e, I should be able to take a generic class, mix and match kits / ACFs / feats / whatever resources I have, and make something playable that doesn't get in the way of my concept.

-----

Classes serve two opposed purposes: to facilitate the instantiation of character concepts, and to provide a framework of what valid concepts look like in the first place. To feed inspiration, and to ground it. It is not surprising, then, that two opposed class types would be required to facilitate both of these goals.

Anybody got enough sanity left to comment on whether or not that made any sense?

[MoiMagnus]

Anybody got enough sanity left to comment on whether or not that made any sense?

It does. I'd add that a game probably benefits from having more than one class system.
E.g 5e has classes and subclasses that are at different level of precision.

Taking one of your example, I'd probably not expect "Dread Pirate" to be a class because of how specific it is IMO, but that would be a perfect subclass.

And "Example 4 - 4 elements" made me think that D&D's "elements" are actually ability scores.
The use of Cha/Wis/Int for spellcasting does not have a major gameplay significance, but it shapes what kind of ability is expected and the approach a character of this class is expected to take. Con spellcasting, when it exists, also communicate something very different about the character.
Similarly, Dex vs Str for martial is probably what you would expect from having martial themed by two opposite elements.

[Luccan]

Correct. D&D class design is a work of accretion. Ideas were slapped on as needed, but without anything resembling a unified or coherent design direction. And after 3E, it seems no class can ever be removed.

3e had dozens of base classes that haven't made a reappearance in 5e and 4e has at least a few that have also not seen a return. D&D just hasn't removed any "classics", because those have made their way into existing D&D settings as actual in-fiction categories of people.

Edit: should probably be pointed out, D&D has never really dropped any class that appeared in an initial Player's Handbook. It's not a new thing that Sorcerers stuck around after 3e. It's actually a bit of a break in the pattern that Warlord didn't get updated to 5e

[Lord Raziere]

Edit: should probably be pointed out, D&D has never really dropped any class that appeared in an initial Player's Handbook. It's not a new thing that Sorcerers stuck around after 3e. It's actually a bit of a break in the pattern that Warlord didn't get updated to 5e

Technically if you count Banneret and Battlemaster some of the features did make it into 5e fighter. though they thought that the warlord was somehow not a good enough archetype on its own but thought the fighter was a better class to go with, who knows. we could've had some much needed utility and a good reason why the PC class isn't just some random fighter dude: they're not just a soldier but someone who can lead others. if the person wanted to just hit things anyways, they could've done that anyways with a warlord and have common useful social skills out of combat if they ever feel like it, like persuasion and insight or perception that would help in more investigative parts of a campaign.

[Luccan]

Technically if you count Banneret and Battlemaster some of the features did make it into 5e fighter. though they thought that the warlord was somehow not a good enough archetype on its own but thought the fighter was a better class to go with, who knows. we could've had some much needed utility and a good reason why the PC class isn't just some random fighter dude: they're not just a soldier but someone who can lead others. if the person wanted to just hit things anyways, they could've done that anyways with a warlord and have common useful social skills out of combat if they ever feel like it, like persuasion and insight or perception that would help in more investigative parts of a campaign.

Were I to replace Fighter with something (I'm hesitant to do so, I think there's legitimate value in a less rigidly archetypal warrior class), it would be with something like Warlord. I'd honestly be fine combining the two, but that might overcomplicate things.

[HumanFighter]

3e had dozens of base classes that haven't made a reappearance in 5e and 4e has at least a few that have also not seen a return. D&D just hasn't removed any "classics", because those have made their way into existing D&D settings as actual in-fiction categories of people.

Edit: should probably be pointed out, D&D has never really dropped any class that appeared in an initial Player's Handbook. It's not a new thing that Sorcerers stuck around after 3e. It's actually a bit of a break in the pattern that Warlord didn't get updated to 5e

I thought the Battlemaster subclass for Fighter was kind of the Warlord for 5E?

Anyways, if we're talking numbers of different classes in a Tabletop RPG game, what do you guys think is the "optimal" amount? I think I prefer 6 or 7 different classes in total.
Not enough class choices, and you feel you just don't have enough options as a player, and it hurts replayability.
Too many class choices is overwhelming, so it is good to find a right amount, a happy medium if you will.

[Mechalich]

I thought the Battlemaster subclass for Fighter was kind of the Warlord for 5E?

Anyways, if we're talking numbers of different classes in a Tabletop RPG game, what do you guys think is the "optimal" amount? I think I prefer 6 or 7 different classes in total.
Not enough class choices, and you feel you just don't have enough options as a player, and it hurts replayability.
Too many class choices is overwhelming, so it is good to find a right amount, a happy medium if you will.

In general classes are more useful as a game design element if they are narrowly tailored, and also easier to balance. As such, a larger number of narrow classes is more useful than a small number of broad but largely meaningless classes. Star Wars Saga, for example, has only five base classes: Jedi, Noble, Scoundrel, Scout, and Soldier and in terms of matching character concept to mechanics they are largely useless with even the Jedi class being almost endlessly malleable.

At the same time, having too many classes is likely to lead to balance issues because one particular class, or more likely one specific class/race/build combo, always has the chance to become the 'one true build' and thereby drastically lowering the utility of all other character options. This is particularly common in tabletop games with significant publication schedules, since it quickly becomes impossible to test all options against each other.

Generally a game wants a class package for every major concept that could conceivably see play, but if the number of major concepts is over a dozen, then classes are probably not the way to go.

[MoiMagnus]

I thought the Battlemaster subclass for Fighter was kind of the Warlord for 5E?

Thematically, somewhat. But if I remember correctly one of the point of 5E Warlord is that it was roughly able to compensate for the lack of a Cleric in term of support capability, potentially including healing if you chose the adequate options. (While Battlemaster can only provide temporary hit points to allies, and is not very good at it).

Battlemaster is more "what if the 4e Warlord could only buff themself?"

[Luccan]

I thought the Battlemaster subclass for Fighter was kind of the Warlord for 5E?

Anyways, if we're talking numbers of different classes in a Tabletop RPG game, what do you guys think is the "optimal" amount? I think I prefer 6 or 7 different classes in total.
Not enough class choices, and you feel you just don't have enough options as a player, and it hurts replayability.
Too many class choices is overwhelming, so it is good to find a right amount, a happy medium if you will.

Battlemaster, from the arguments I've seen, doesn't quite work for what people are looking to get out of Warlord. Frankly, its tactical maneuvers are a bit underwhelming in comparison to just using personal maneuvers. There's also a lot of "use up an Ally's reaction" going on and a number of them will only benefit front-liners most of the time. It's just not designed to be a replacement Warlord. Also, apparently people really liked the Warlord's capacity for non-magical healing, which they shoved onto a different, worse subclass.

I think you could easily cut D&D down to about 7 or 8 classes, though I don't think it's necessary. Taking 5e, Barbarian, Paladin, and Ranger can be rolled under Fighter as subclasses. For this hypothetical, Pally and Ranger keep their current casting mechanics and Eldritch Knight gets a boost to match. Some people want to put Monk there, but I think Monks have enough differences (all armor vs no armor, all weapons vs specific weapons, monks have a lot of special abilities). So that's already down from 5e's current 13 to 10.

Pick two of the closely related full-casters (Cleric/Druid, Cleric/Warlock, or any two arcanists but Bard) and roll them into one class. I favor getting rid of Warlock and giving Sorcerers Invocations, but take your pick. That's 9.

Turn Bard back into the Jack of All Trades it used to be and make it a Rogue subclass, with casting comparable to 5e's half casters (this replaces Arcana Trickster). That's 8.

If you want it down to 7, there's arguments for Artificer being a Wizard subclass, but I'm actually going to say it should be a Rogue subclass instead. They've always had ties to the skill-monkey, traps-and-locks roll, might as well stick them in the class that already gets Thieves Tools. Get rid of the spell-casting and refocus them on infusions and getting a boost to item crafting.

Again, due to how many unique mechanics they get, I think Monk is better off as its own class, but if you had to make it a subclass I'd make it one for Fighter or Rogue.

[Duff]

It's no surprise the the original 4 classes (Fighter, rogue, cleric and wizard)* are also the least defined by Achetypes and most defined by an in-game role and set of mechanics, given the history of the game


* Yes, some of the names have changed but I think it's pretty clear these are the current names for the classes which are basically the same linages

[Pauly]

Well you can have ‘classless’ games. In these systems either all the players are the same class (eg Superheroes) or it’s a reality based system (eg everyone is a regular human).
In the classless games players choose their skills and progress them as they see fit. A character’s attributes will impact on their ability to use their chosen skills, but ultimately the player has freeform to choose whatever they want. Sometimes they will yave backgrounds/careers that allow access to specific skills.
The problem a lot of players have with these systems is that the menu of choices is too big and complex. Unless the player has an expert level in the system it is very hard to build effective characters.

What character classes do is to pre-package skills and abilities. This allows the player to only have to master one of the skill sets and know one of the skill trees. If I’m playing a wizard I don’t need to fill my head with the nuances of being a Rogue or Cleric.
The problem with this approach is that it seems forced and arbitrary.

[Morty]

3e had dozens of base classes that haven't made a reappearance in 5e and 4e has at least a few that have also not seen a return. D&D just hasn't removed any "classics", because those have made their way into existing D&D settings as actual in-fiction categories of people.

But the cure class list was rolled back to 3E's + warlock after the modest changes 4E tried to make.

[Yora]

Does anyone know of a game that has classes but not levels?

[Batcathat]

Does anyone know of a game that has classes but not levels?

I can think of one, but it might not be of much use for most of you. It's Swedish game called Drakar och Demoner ("Dragons and Demons", I wonder what name they were inspired by? ). Like the more familiar game with the same initials it has a bunch of different editions and I think some of them have levels (it was something of a controversy) but the one I played as a teenager did not, while it did have classes.

[Cluedrew]

Does anyone know of a game that has classes but not levels?

Many Powered by the Apocalypse systems including Apocalypse World itself don't have levels but still have playbooks which act as classes. They do have level ups where you earn XP and then upgrade, but its basically a point-by system where all upgrades have the same cost. Now Apocalypse World divides things up into tiers where you need a certain number of upgrades (at least level X) to get them.

I think Blades in the Dark has no leveling at all. Outside of a couple abilities that can be repurchased and improved and the max stat ceiling going up (from 2 to very situationally 5) there is no ordering on any upgrades.

And possibly every Story-Teller system. There is some ordering required in them and the classes are pretty soft but I would probably count them.