A Guide To Base Classes, Power, and Identity

[Vadskye]

What Is a Base Class?

A base class is not a concept. Consider the concept of "holy warrior who strikes down his enemies with the power of his deity." Is it a paladin or a cleric? But that's a trick question; that could apply to every single class. Nothing about being a fighter, wizard, or druid prohibits you from fervently following a deity and ascribing your successes to the deity's influence.

A base class is not a profession. You can fight people for money as any class, or enjoy research as any class, or farm as any class. People of some classes are likely to be better at some things, but that's different.

A base class is not a build or character. This is really WotC's fault, when they created excessively specific prestige classes like the arcane archer. Now, these boards are full of classes with excessively specific concepts like "fighter who uses kamas and disarms people". That's not a class

A base class is not a personality. There can be thousands upon thousands of unique characters with the same class, but totally different beliefs, personalities, and goals. You don't even need to be good at fighting to be a fighter. We long ago abandoned ability score prerequisites for classes.

So what is a base class? A base class is two things: a source of power and an identity, with the former being more important. (For now, I'm going to ignore prestige classes and alternate class features that change the definition of a class.)

Base Classes as a Source of Power

Every barbarian draws power from their raw physicality. Not everyone who relies on their body is a barbarian. You don't need to be strong to be a barbarian. But if you don't rely on your body for your power, you're not a barbarian.

Every cleric draws power from their beliefs. Not everyone who has fervent beliefs, religious or otherwise, is a cleric. You don't need to be a diehard zealot to be a cleric. But if you don't rely on your beliefs for your power, you're not a cleric.

Every druid and ranger draws power from nature. Not everyone who revers nature is a druid. You don't need to be a tree-hugging hippy to be a druid. But if you don't draw power from nature itself, you're not a druid or ranger.

Every monk draws power from their ki, which comes from inner perfection. Not everyone who strives for inner perfection is a monk. You don't need to be a Zen master to be a monk. But if you don't strive for inner perfection, you're not a monk.

Every paladin draws power from their righteousness. Not everyone who does good is a paladin. You don't need to be a Lawful Stupid do-gooder to be a paladin. But if you don't draw power from your righteousness, you're not a paladin.

Every sorcerer draws power from the innate magic in their blood. Not everyone who has magic in their blood is a sorcerer. You don't need to do or be anything special to be a sorcerer. But if you don't have innate magic in your blood, you're not a sorcerer.

Every wizard draws power from their study of magic. Not everyone who studies magic is a wizard. You don't need to spend all day cooped up in a tower to be a wizard. But if you don't study magic, you're not a wizard.

There are three missing classes in this list. Fighters and rogues have no proper unifying source of power: fighters fight, and rogues... skill? These are so vague that it's difficult to understand how they are supposed to work. Fighters in particular are such a bland blank slate that the various rewrites have an incredible variety of goals, mechanics, and identities. Bards are missing for a different reason: they aren't a class. They're a concept. The history of bards in D&D is somewhat tortuous and thoroughly inconsistent, and they don't belong in a class of their own. (Perhaps a prestige class, or just a multiclass with feat support.)

However, not everything that makes up a class can be explained by its source of power. Why do class skills exist? Why do monks use exotic weapons? Why do rangers and druids have the same source of power? These can be explained by considering the class as an identity.

Base Classes as World-Building Identities

In any game world, there exist certain stereotypes. Monks generally live in monastaries and fight with unusual weapons. Rogues are generally good at noticing traps and sneaking around. Rangers are (supposed to be) the best trackers. Barbarians are illiterate. None of these are required elements of the class! Rogues can be clumsy, barbarians can be scholars, and monks can use longswords if they want. However, these stereotypes play a critical role in defining the game world.

Standard tropes and expectations are what define a world. When people have expectations of how the world works, everything about a story and a campaign works better. In a world without expectations, where classes have no consistent identity, everything is equally surprising - which is to say that nothing is surprising. It just doesn't work. Classes have identities to provide a baseline from which you can improvise, either as a player or as a DM.

So What?

When you see your homebrew through this lens, you can understand what makes a class "tick" better. If you consider a base class to be a source of power, the differences between each class become much easier to define. That can guide you to think of new abilities that might be worth adding. When you're designing a new class, and you notice it has multiple inconstent sources of power, you're probably making a mistake somewhere. Likewise, if your class doesn't have any real source of power, like the fighter, it needs some more definition. Finally, notice that the sources of power are very broad, and easily applicable to characters of many different concepts and personalities. If your class has an extremely specific or boring source of power that defines all of their abilities, like "using weapons" or "ability to manipulate the color plaid", you're not probably not going to get a class that contributes much to the game, even if it works perfectly mechanically.

You should also consider your class's identity separately from its raw source of power. This is particularly true when rewriting classes to suit your own campaign settings. If you consistently apply simple changes to the fundamental chassis of a few core classes, you can generate very different worlds. Class skills, and other identity mechanics, are a way to make the fluff of the world deeply integrated into the mechanics, which makes it more believable and immersive. If you want to create a world full of deception and intrigue, you could try giving everyone Bluff, Hide, and Move Silently as class skills, and watch as it develops naturally. Of course, that kind of change harms the uniqueness of the classes which originally had those as class skills; it's always a balancing act.

I provide this guide in the hopes that it may be useful to anyone who plays with base class design, in the hopes of getting feedback to show me the error of my ways, and in the hopes of figuring out what the heck to do with monks.

Changelog:
4/1: Fixed incorrect argument about fighter and rogue rewrites. Thanks, Just to Browse!

[LordErebus12]

this makes a lot of sense to me. good work on this, Vadskye

[SiuiS]

What about rangers being the chosen of humanity, imbued by the gods with powers to fight against humanity's enemies? They don't have nature powers, they have divine powers and also live in nature.


I think this is an accurate portrayal but not the accurate portrayal. I've seen barbarians whose rage was Psionic, their weapons flowing around them in an energy field and obeying their mental whims a-la ultimate form from kingdom hearts II. It may be worth noting that even if done poorly, base class was supposed to be concept and profession. That's why you have class skill lists; to allow the illusion of freedom while still reinforcing your profession and concept.

Well done all around. :)

[Vadskye]

What about rangers being the chosen of humanity, imbued by the gods with powers to fight against humanity's enemies? They don't have nature powers, they have divine powers and also live in nature.
That's one theme you could give rangers in a campaign or setting. But it's not consistent with the core description, which mentions nothing about humanity and says they "gain divine spells from the power of nature."


I think this is an accurate portrayal but not the accurate portrayal. I've seen barbarians whose rage was Psionic, their weapons flowing around them in an energy field and obeying their mental whims a-la ultimate form from kingdom hearts II. It may be worth noting that even if done poorly, base class was supposed to be concept and profession. That's why you have class skill lists; to allow the illusion of freedom while still reinforcing your profession and concept.
Psionic, KHII style barbarians are super cool. It would also fall pretty thoroughly under the realm of alternate class features that change the power source of a class. I really should have gotten around to that at some point in the original post.

Essentially, just like you can change the identity of a class to suit a particular setting without changing the power source, you can also change the power source of the class without changing the identity. In general, if you change both, it's hard to tell that you're still talking about the same class, which isn't good. But if you change only one or the other, you can generate fantastic things.


Well done all around. :)

this makes a lot of sense to me. good work on this, Vadskye
Thank you both! :smallsmile:

[Drako_Beoulve]

You helped me to do better Fixes, Thank you so much!!

[Just to Browse]

This is very much an opinion piece and not something one should design all their classes around. Overly-specific base classes are still base classes, just like healers, fire mages, and gravity warriors are all base classes. Specificity does not disqualify those from their base class status, nor does it make them lower-quality.

Base classes do not need to be a "source of power" either, and lacking power sources are not a reason for a class being bad. People don't regularly rewrite the swashbuckler (which has the same power source as the fighter and rogue), and lots of people rewrite the wizard/sorc/cleric/druid (which have a power source). If a class is just a set of abilities without flavor attached, it can still be a good base class. Barbarians can be fueled by inner magic or the elements, and druids can cast magic by getting very angry.

The only thing I agree with is that classes are world-building tools, but that's because any kind of diversification implies different niches. As soon as you write w class with superior stealth abilities, it logically follows that people who steal things will favor that class.

Now you can write classes as a source of power, and you can also make your classes generic, and I encourage people to think of where their class fits in the fantasy world threy make it for, but those aren't required. A base class is just a class that you can take at level 1, and is not special in any way other than its stretch of abilities.

[Grinner]

To present a slightly different point of view, I'd like to present this StackExchage thread (http://rpg.stackexchange.com/questions/11800/guidelines-for-creating-homebrew-classes-base-prcs).

[Vadskye]

This is very much an opinion piece and not something one should design all their classes around. Overly-specific base classes are still base classes, just like healers, fire mages, and gravity warriors are all base classes. Specificity does not disqualify those from their base class status, nor does it make them lower-quality.
This is definitely an opinion piece; mechanically, virtually anything presented on these boards could be a base class. I'm trying to define what should be a base class, which will always be somewhat subjective. But I think it's justifiable.

I very much disagree that healers and fire mages should be base classes, because they already are base classes: clerics, sorcerers, and wizards. Those are just specific builds of existing classes that have some mechanics attached. That's what makes the existing base classes so enduring (and I would argue that it is a sign of good design).

But I agree with you on the gravity warrior, because it is something impossible to replicate using existing classes, both mechanically and conceptually. It has a unique source of power and the potential to have a unique identity. If you treat this assessment as a "test" that classes must pass, a gravity warrior would pass the test. (I think it's more specific than most campaigns would find useful, but that is entirely campaign and context dependent.)


Base classes do not need to be a "source of power" either, and lacking power sources are not a reason for a class being bad. People don't regularly rewrite the swashbuckler (which has the same power source as the fighter and rogue), and lots of people rewrite the wizard/sorc/cleric/druid (which have a power source). If a class is just a set of abilities without flavor attached, it can still be a good base class. Barbarians can be fueled by inner magic or the elements, and druids can cast magic by getting very angry.
Are you saying that the swashbuckler is an example of good base class design? I would argue that people don't rewrite the swashbuckler precisely because it's so similar to the fighter and rogue. It's a mashup of the two most generic classes, and it has terrible mechanics. In any event, you're right that I shouldn't use rewrites as a metric the way I did. That was a lazy comparison, and I'll edit it to be more clear.

Like I said to SiuiS, there's nothing wrong with changing a class's power source while keeping its identity. That doesn't even disagree with the point of this post, because the core elements of the class are still its (new) power source and its identity in the world. But if you just throw a bunch of abilities together without a consistent theme for those abilities, I doubt you're going to end up with a good class.


Now you can write classes as a source of power, and you can also make your classes generic, and I encourage people to think of where their class fits in the fantasy world threy make it for, but those aren't required. A base class is just a class that you can take at level 1, and is not special in any way other than its stretch of abilities.
Sure, it's not required - but good design isn't "required" either. There can always be exceptions that don't follow these guidelines, because good design is about knowing when to break the rules. But if this way of thinking about classes helps people design better, I'm happy.

[Just to Browse]

This is definitely an opinion piece; mechanically, virtually anything presented on these boards could be a base class. I'm trying to define what should be a base class, which will always be somewhat subjective. But I think it's justifiable.I took that the wrong way, then. That's definitely a more justifiable position, but I do think your points are still wrong.


I very much disagree that healers and fire mages should be base classes, because they already are base classes: clerics, sorcerers, and wizards. Those are just specific builds of existing classes that have some mechanics attached. That's what makes the existing base classes so enduring (and I would argue that it is a sign of good design).There is no sorcerer or wizard build that gives level-appropriate area fire attacks at-will. There is no cleric build that provides universal healing and temp buffs so efficiently that he wants to use them every round in combat. Those classes totally fail at doing what is required for basic fire mage and healers to function, and their construction actively inhibits these builds because Vancian is strongly opposed to at-will. You couldn't even write a class called "At-Will Hero" and have it cover both of these concepts, because one of these focuses on attacks while the other focuses on full action economy manipulation, and fusing those fosters abuse.


Are you saying that the swashbuckler is an example of good base class design? I would argue that people don't rewrite the swashbuckler precisely because it's so similar to the fighter and rogue. It's a mashup of the two most generic classes, and it has terrible mechanics. In any event, you're right that I shouldn't use rewrites as a metric the way I did. That was a lazy comparison, and I'll edit it to be more clear.That wasn't my point. You said "X, Y, and Z classes are bad. You can tell because they are fixed on these boards a lot." I countered by saying "V and W classes suck, and are basically never fixed." Correlation is not causation.


Like I said to SiuiS, there's nothing wrong with changing a class's power source while keeping its identity. That doesn't even disagree with the point of this post, because the core elements of the class are still its (new) power source and its identity in the world. But if you just throw a bunch of abilities together without a consistent theme for those abilities, I doubt you're going to end up with a good class.Doubt is not proof. Power sources are flavor, and flavor is mutable. Thus classes do not need to be "power sources". They don't even need to have pre-written tactical themes, because one of the big things in game design is that humans recognize patterns. A good example of that is Abaddon (http://dota2.gamepedia.com/Abaddon) from DotA 2, who has a long range attack spell, an exploding shield that requires close range, a basic attack that slows people, and a self-heal that relies on taking damage (i.e. shields are bad). That's a kit that just screams "WTF AM I DOING", but people selectively focus on certain abilities to become strong supports or assault specialists.


Sure, it's not required - but good design isn't "required" either. There can always be exceptions that don't follow these guidelines, because good design is about knowing when to break the rules. But if this way of thinking about classes helps people design better, I'm happy.All right, it's not very useful. Working on class design with either or both of these in mind shuts out a lot of beginner and intermediate design tactics that I think people need in order to improve as designers. You should discourage the things that make legitimately bad classes, not the things that aren't aesthetically pleasing.

[PairO'Dice Lost]

I very much disagree that healers and fire mages should be base classes, because they already are base classes: clerics, sorcerers, and wizards. Those are just specific builds of existing classes that have some mechanics attached. That's what makes the existing base classes so enduring (and I would argue that it is a sign of good design).

On the contrary: if you have base classes whose power sources are "studying magic," "having faith," and "channeling nature," then every other magic-using class is going to be just a build of those classes. If the gravity warrior is a good class because it has a unique power source and identity, then the fire mage and healer should be viewed as well-designed classes because they have their own power sources, their own ways of accessing said power sources, and their own mythological draws, while the wizard, cleric, and druid are badly-designed because they each mash together a bunch of archetypes into a single bland base class.

If uniqueness is to be valued, then why is a fire mage (who meditates on the cleansing power of flame and studies the chemistry of combustion until they can burn anything and everything and has plenty of traction in the form of ever-popular pyromancers) bad while a cleric (who is supposed to represent a priest of any deity in a polytheistic religion but is really a ripoff of a monotheistic Judeo-Christian priest, and doesn't even really fit that perfectly) is good?

The wizard and cleric are just like the fighter and the rogue in that they're more of an umbrella class for a lot of archetypes, but the former two are considered good classes while the latter two are considered bad classes because you can use a wizard or cleric to competently duplicate many different magic-user concepts from fiction while trying to duplicate a lot of heroic concepts from fiction using a fighter or rogue either isn't doable or leaves you with a character that sucks. Splitting those broad classes up into more defined and interesting classes leads to more balance and variety for all of them, because it limits the casters' breadth thematically and gives the noncasters an excuse to get Nice Things.

[Vadskye]

There is no sorcerer or wizard build that gives level-appropriate area fire attacks at-will. There is no cleric build that provides universal healing and temp buffs so efficiently that he wants to use them every round in combat. Those classes totally fail at doing what is required for basic fire mage and healers to function, and their construction actively inhibits these builds because Vancian is strongly opposed to at-will. You couldn't even write a class called "At-Will Hero" and have it cover both of these concepts, because one of these focuses on attacks while the other focuses on full action economy manipulation, and fusing those fosters abuse.
sigh. When I saw healer and fire mage, I jumped to the implementations I've seen of the classes, which have largely been Vancian and unimpressive. That was stupid. You're completely right that there's all sorts of interesting mechanical design space for healers and fire mages that the core classes completely fail at covering. They are just as valid as the gravity warrior.

I've tried to figure out what exactly I don't like about those classes. They're just... too specific. The mechanical potential isn't enough to say that "fire mage" should be a base class. There are many ways you could provide the mechanics for a fire mage:
Feats which grant at-will abilities that scale appropriately
Prestige class
Change spell mechanics in general
Alternate class features
New base class which gives level-appropriate at-will single-target and area attacks of malleable fluff/power source.

Yes, you could create a "fire mage" base class. But that's not a robust solution, because it scales horribly. That makes an entire class devoted to a single spell descriptor. Now someone wants to play an "ice mage", and what do you do? You can't just refluff the class, because cold and fire do different things, thematically and mechanically. Then you look at "electricity mage", "gravity mage", "fire warrior"... The classes are overspecialized to the point of uselessness, even if the mechanics of the class are perfect for its original conception. Unless you want to create a new base class for every character, classes need to be more malleable than "fire mage" or "healer".


That wasn't my point. You said "X, Y, and Z classes are bad. You can tell because they are fixed on these boards a lot." I countered by saying "V and W classes suck, and are basically never fixed." Correlation is not causation.

Fair enough. It's been edited to remove the offending claim; I think the rest of the argument stands without it.


Doubt is not proof. Power sources are flavor, and flavor is mutable. Thus classes do not need to be "power sources". They don't even need to have pre-written tactical themes, because one of the big things in game design is that humans recognize patterns. A good example of that is Abaddon (http://dota2.gamepedia.com/Abaddon) from DotA 2, who has a long range attack spell, an exploding shield that requires close range, a basic attack that slows people, and a self-heal that relies on taking damage (i.e. shields are bad). That's a kit that just screams "WTF AM I DOING", but people selectively focus on certain abilities to become strong supports or assault specialists.
Flavor is mutable, but if it was never consistent in the first place, you can't change it consistently. It's easy to see the theme and fluff behind a barbarian whose power stems from innate magic because it's very clear what is changing. If you take a class that provides ambiguous or inconsistent explanations for how it does things (bard), it's much more difficult to change the fluff.

As for Abaddon, the message I'm getting is that classes don't need a power source and don't need consistent mechanics. So is there anything that separates a good class from a bad class other than mechanical power/utility/balance?

All right, it's not very useful. Working on class design with either or both of these in mind shuts out a lot of beginner and intermediate design tactics that I think people need in order to improve as designers. You should discourage the things that make legitimately bad classes, not the things that aren't aesthetically pleasing.
There is plenty of advice to give on mechanics - and that would be more useful as "Class Design 101". But that wasn't my goal here. That would be a different post entirely (though one I might write someday).


On the contrary: if you have base classes whose power sources are "studying magic," "having faith," and "channeling nature," then every other magic-using class is going to be just a build of those classes. If the gravity warrior is a good class because it has a unique power source and identity, then the fire mage and healer should be viewed as well-designed classes because they have their own power sources, their own ways of accessing said power sources, and their own mythological draws, while the wizard, cleric, and druid are badly-designed because they each mash together a bunch of archetypes into a single bland base class.
If uniqueness is to be valued, then why is a fire mage (who meditates on the cleansing power of flame and studies the chemistry of combustion until they can burn anything and everything and has plenty of traction in the form of ever-popular pyromancers) bad while a cleric (who is supposed to represent a priest of any deity in a polytheistic religion but is really a ripoff of a monotheistic Judeo-Christian priest, and doesn't even really fit that perfectly) is good?
The wizard and cleric are just like the fighter and the rogue in that they're more of an umbrella class for a lot of archetypes, but the former two are considered good classes while the latter two are considered bad classes because you can use a wizard or cleric to competently duplicate many different magic-user concepts from fiction while trying to duplicate a lot of heroic concepts from fiction using a fighter or rogue either isn't doable or leaves you with a character that sucks. Splitting those broad classes up into more defined and interesting classes leads to more balance and variety for all of them, because it limits the casters' breadth thematically and gives the noncasters an excuse to get Nice Things.
I described my core problem with this above. If "fire mage" is an appropriate standard for a base class, you'll need to make a staggering number of 20-level base classes to adequately capture the flexibility provided by a single class in the existing system. Either that, or you'll have to make a "fire mage" base class which is generic enough to suit a wide variety of mechanical purposes and roles. If you do that, I'm happy, but it isn't a "fire mage" anymore, it's a "attack mage" or perhaps a "mage".

Note that I'm not saying that the cleric/wizard/sorcerer classes are well designed mechanically. Heaven knows they aren't! Individual casters need to be more different from each other (spontaneous casting would do wonders), the classes need to be beat with a nerf bat, they need class features, and some built-in at-will magical ability wouldn't be bad either. But that's not what I'm trying to argue here. I'm saying that the core design of the class - the unique power source, and the identity - is good. It provides a solid foundation for mechanics (though the current mechanics fall short).

[PairO'Dice Lost]

When I saw healer and fire mage, I jumped to the implementations I've seen of the classes, which have largely been Vancian and unimpressive. That was stupid. You're completely right that there's all sorts of interesting mechanical design space for healers and fire mages that the core classes completely fail at covering. They are just as valid as the gravity warrior.
[...]
I described my core problem with this above. If "fire mage" is an appropriate standard for a base class, you'll need to make a staggering number of 20-level base classes to adequately capture the flexibility provided by a single class in the existing system. Either that, or you'll have to make a "fire mage" base class which is generic enough to suit a wide variety of mechanical purposes and roles. If you do that, I'm happy, but it isn't a "fire mage" anymore, it's a "attack mage" or perhaps a "mage".

How is "fire mage" not enough to cover multiple roles? A fire mage does't just shoot fireballs at people, it should be able to create and control fire (and thus light, heat, steam, smoke, etc. as byproducts), heat objects or draw heat away from them, summon and control fire creatures (or turn into fire creatures themselves), and become resistant and eventually immune to fire, just to start with. They might also be able to do things like boil blood and heal diseases if you let them control creatures' body temperature, perform limited divinations if you take pyromancy literally, manipulate the earth and sky to a limited extent if you extend "fire" to encompass thematically-similar things like magma and the sun, gain flight if they can use fire jets for propulsion, take on various creature qualities if you give them some red dragon/salamander/phoenix/elemental/etc. themed abilities, and more.

So just the basic fire manipulation abilities give them direct damage, battlefield control, buffing, debuffs, summoning, and some utility, more than enough to match up to an Evocation, Conjuration, or Transmutation specialist wizard, and they can dabble in healing, terrain shaping, shapeshifting, gishing, and a few other areas depending on the thematic influences to the class. Not so limited anymore, huh?

Like you said, people hear "fire mage" and think "blaster who only uses fire damage," which contributes to the dominance of the core three casters--every other caster concept must be so limited as to easily be folded into one of them, right? But if you really sit down and think about how to make them unique, these more specialized classes can be just as powerful and versatile as the core three within their area of specialty and be much more interesting besides.

[Realms of Chaos]

So just the basic fire manipulation abilities give them direct damage, battlefield control, buffing, debuffs, summoning, and some utility, more than enough to match up to an Evocation, Conjuration, or Transmutation specialist wizard, and they can dabble in healing, terrain shaping, shapeshifting, gishing, and a few other areas depending on the thematic influences to the class. Not so limited anymore, huh?

While this is certainly the case, there is still kind of a problem with your fire mage. Even when you create the perfectly unique and multi-faceted fire mage, you aren't just creating the fire mage. On some level, you are also creating the expectation for there to be an ice mage... and a lightning mage (weather mage?)... and more mages on top of that. If you create a class that does nothing but figments, someone will ask where the phantasm class has gone. Unless you fight the uphill battle to justify why there is only a fire mage (likely granting a power source and identity in the process), this unmet expectation is a little bit annoying/disappointing even if attached to an otherwise wonderful class.

With that said, I personally don't think that there is much of a problem so long as the negative possibility space is eventually filled, which has happened several times over the course of 3rd edition. The relationship between the wizard and sorcerer gave rise to the favored soul class, for example, and the mess that was the warmage paved the way for the more acceptable dread necromancer and beguiler. Likewise, I'd even argue that the wizard's ability to learn spells created the creation space that came to be filled by the archivist and erudite. Also, it's not as big of a deal to fill the negative possibility space on a homebrew forum than in official sources as... well... we can homebrew the rest. As a counterpoint, however, I would comment that making one specialist particularly unique may make it harder to homebrew counterparts as there is far less that can be simply reflavored and a common balance point is probably preferable to all "parallel classes" to avoid a repeat of the warmage.

At the end of the day, I would prefer a single elemental mage over a generic or overly specialized fire mage/ice mage/lightling mage/etc. but if each of those classes was truly well-made and stood out on their own, I'd prefer to have the individual specialist classes. Of course, that is simply a matter of opinion and I wouldn't want to stop newcomers from taking a chance of a healer or mind-reader class.

TLDR: while I hesitate to give "power sources" as much emphasis as the OP, using a single power source out of an obvious set (one of the eight schools of magic, one of the four bodily humors, one of the five energy descriptors/4 elemental descriptors) makes it seem arbitrary that the other members of that set lack classes, which isn't the best thing. I'd recommend either addressing all members of the set or making a unique power source (such as having your fire mage use the "truefire" burning at the start and end of existence instead of plain old fire) so that it no longer seems to belong in the set... but again, that's just opinion.

[PairO'Dice Lost]

While this is certainly the case, there is still kind of a problem with your fire mage. Even when you create the perfectly unique and multi-faceted fire mage, you aren't just creating the fire mage. On some level, you are also creating the expectation for there to be an ice mage... and a lightning mage (weather mage?)... and more mages on top of that. If you create a class that does nothing but figments, someone will ask where the phantasm class has gone. Unless you fight the uphill battle to justify why there is only a fire mage (likely granting a power source and identity in the process), this unmet expectation is a little bit annoying/disappointing even if attached to an otherwise wonderful class.

Well, 3e gave us a very fiendish warlock, which claimed to be able to draw power from many sources but really only gets fey/chaos/celestial/etc. stuff with reflavoring; it gave us the dragonfire adept, but not a hellfire adept or a faeriefire adept; it gave us the beguiler, dread necromancer, and warmage, but not the shifter, oracle, summoner, or warder. It's not really unprecedented for the game to have a certain subset of a themed group and then say "fill in the rest yourself."

And even if you grant that we should have an elementalist class rather than a fire mage class--and I'm fine with that, I was just using the fire mage since that's the example that was already being used and since I wanted to show that even something seemingly narrow like a fire mage can be unique and interesting--the same point remains: "elementalist" is still more narrow than "wizard" and doesn't have a unique power source, and would be subsumed by the wizard under the class = power source paradigm, but it's still the elementalist that makes a more unique and interesting class and the wizard that is overly broad and needs to be broken up.

[Vadskye]

How is "fire mage" not enough to cover multiple roles? A fire mage does't just shoot fireballs at people, it should be able to create and control fire (and thus light, heat, steam, smoke, etc. as byproducts), heat objects or draw heat away from them, summon and control fire creatures (or turn into fire creatures themselves), and become resistant and eventually immune to fire, just to start with. They might also be able to do things like boil blood and heal diseases if you let them control creatures' body temperature, perform limited divinations if you take pyromancy literally, manipulate the earth and sky to a limited extent if you extend "fire" to encompass thematically-similar things like magma and the sun, gain flight if they can use fire jets for propulsion, take on various creature qualities if you give them some red dragon/salamander/phoenix/elemental/etc. themed abilities, and more.

So just the basic fire manipulation abilities give them direct damage, battlefield control, buffing, debuffs, summoning, and some utility, more than enough to match up to an Evocation, Conjuration, or Transmutation specialist wizard, and they can dabble in healing, terrain shaping, shapeshifting, gishing, and a few other areas depending on the thematic influences to the class. Not so limited anymore, huh?
That sounds great! It also sounds like a wizard or sorcerer with a specific spell list and a few fire-themed feats. If the fire mage draws power from study, it's a wizard - but more likely, it's going to have an innate mastery of fire, making it a sorcerer. That's what makes the sorc/wiz so useful; they can be so many concepts wrapped into one. If you're being that broad, "fire mage" is pure fluff (which is in no way a mark against its awesomeness), and should be treated as such. Really, that sort of thematic flexibility is what building a character is all about.

This also solves the problem of the "ice mage", the "phantasm mage" every other themed mage: they're all just mages! You just change the spell list, while still keeping it as a subset of the sorc/wiz spell list. Fundamentally, you're building characters, not entirely new base classes. (Again, sorc/wiz need better design, but that's a separate issue.)

With that said, this discussion has made it clear that my definition of "power source" and "identity" was incorrect, or at least incomplete. It doesn't adequately describe how and when to change fluff. It doesn't describe how to design a class which encourages the creation of rethemed "subclasses" or alternate classes, like the fire mage (a subclass, theme, or character built from a sorcerer), the healer (the same for a cleric), and so on. I'll give some thought to how to design that.

[PairO'Dice Lost]

That sounds great! It also sounds like a wizard or sorcerer with a specific spell list and a few fire-themed feats.[...]That's what makes the sorc/wiz so useful; they can be so many concepts wrapped into one.

That's the way it is now, but that isn't how it has to be, or even should be. If someone were to have asked you shortly before November of 2004 how to build a character who makes a pact with a demon lord for power, I bet you'd have said "If they study the demon's teachings and gain power from that, just be a wizard with summons, fire spells, and curses, and if they actually worship the demon lord, just be a cleric with the Evil and Destruction domains." If someone then were to ask you how to build a blaster mage with an emphasis on metamagic and a broad spell selection, you'd probably tell them to be a sorcerer, pick a lot of metamagic feats, and go into a PrC that added a bunch of free spells to their list.

And then Complete Arcane came out, giving us the warlock and the warmage. Both of those concepts could have been covered by a core caster with strategic spell selection, but (A) the new classes provided new mechanics to express those concepts that you couldn't get from the core classes--invocations and full-list casting/sudden metamagic, respectively--and (B) the new classes are all in one place so a player who can't or doesn't want to dumpster-dive and reflavor to build a character can have those concepts in one simple package.

If you're going to be that reductive about concepts, we only need the four classic classes of fighter/wizard/rogue/cleric and all other characters are simply reflavored versions of those classes with certain feature selections. Barbarians are fighters with tanky feats, ninjas are rogues with magical talents, sorcerers are wizards with an ACF, druids are clerics of a nature god, as you describe with fire mage and healer being subclasses. Or even further, we just have Fighting Man and Magic User, and everything else is build choices and multiclassing. That's certainly a way to handle things, but it's not really fair to condense casters into do-anything characters like that while leaving martial types split up because there's no one power source that can let the fighter or rogue do anything the way "totem spirits" enables barbarian stuff, "the power of Good" enables paladin stuff, "chi mastery" enables monk stuff, and so forth.

Harry Potter, Harry Dresden, Gandalf, Merlin, Pug, Ged, and Dr. Strange are all described as "wizards," yet they all work their magic very differently. One certainly can approximate all of them by building a 3e wizard with the appropriate spells, but some of them would be better suited as other classes (Merlin as a druid, Gandalf as a bard, Ged as a truenamer, etc.) and all of them would be even better suited to their own sort of wizard class, as the innumerable "Harry Potter in D&D" threads in Homebrew attest. I hope you can agree that a specific Potterverse Wizard class would fit Harry Potter et el. much better than the standard wizard or sorcerer while still allowing room for all the different character concepts we see in the novels, so why shouldn't people make a Fire Mage class to cover all the pyromaniacs in fiction from the Human Torch to Dark Souls pyromancers to Melisandre from ASoIaF and make other specialized classes for whatever else they want to cover that a Vancian wizard with a spellbook and a familiar doesn't quite fit?

[Just to Browse]

Yes, you could create a "fire mage" base class. But that's not a robust solution, because it scales horribly. That makes an entire class devoted to a single spell descriptor. Now someone wants to play an "ice mage", and what do you do? You can't just refluff the class, because cold and fire do different things, thematically and mechanically. Then you look at "electricity mage", "gravity mage", "fire warrior"... The classes are overspecialized to the point of uselessness, even if the mechanics of the class are perfect for its original conception. Unless you want to create a new base class for every character, classes need to be more malleable than "fire mage" or "healer".Wbu? If you want to be a fire mage without the base class, you'll also have to do all those things, except you'll also have to jump through a bunch of hoops to sqeeze Vancian mechanics into an at-will mold. You need to write brand new feats, think up new spells, go back and edit old stuff to add disclaimers about the new mechanics, and so on. You're still doing all that work, but now it's spread out and hidden in different places and players also have to do a lot of work in order to get their build done.


Flavor is mutable, but if it was never consistent in the first place, you can't change it consistently. It's easy to see the theme and fluff behind a barbarian whose power stems from innate magic because it's very clear what is changing. If you take a class that provides ambiguous or inconsistent explanations for how it does things (bard), it's much more difficult to change the fluff.I seriously disagree. The bard is easy to reflavor as a draconic magetype, a fledgling siren, a priest who talks really well, or even an inspiring barbarian chieftan (rage to inspire your friends, rage to use bull's strength). My guess is that you find it too hard because you haven't done it enough.


As for Abaddon, the message I'm getting is that classes don't need a power source and don't need consistent mechanics. So is there anything that separates a good class from a bad class other than mechanical power/utility/balance?I would say if it encourages toxic behavior (Blood in the Water, or kender), or if it fails to do what you want, it's also a bad class.