To Class or Not to Class

[PeteNutButter]

When creating a system where do you stand on character classes?

Is it better to be a classless system, ala Skyrim, or to have different character classes?

Here are some pros and cons. I'm curious what people think, and what upsides/downsides to both I'm missing.

Classless

Pros:
Ultimate Customization
Easy to add new content
Open ended "fluff"/play whoever however

Cons:
Can be intensely complex
Can be very hard to balance
Lack of included fluff can hurt less creative players, i.e. no identity
Inexperienced/unoptimizing players suffer, potentially making garbage characters

Class System

Pros:
Easier to balance
Potentially more replayability with desired features locked behind other classes
Built in identity and roles

Cons:
Can feel like a prison with role and identity/less freedom

[Man_Over_Game]

I tried my hand at making an original system, but life got in the way before I could make something of it.

One particular thing that I find that works out well is, with classless systems, make them complex to design for the player during their offtime. Maybe you create a "Make Your Own Spell System" that allows you to combine elements, stats, attacks, or whatever. But then make it easy to implement in combat.

For example, say you have the Force attribute that you can tack onto an ability. It can be attached to a base targeting effect, and pushes back an enemy up to your Body stat in distance. In order to gain this attribute, it requires 5 Body.

Also say you have a base targeting effect, like Ray, that simply deals an intellectual modifier's damage on-hit to a single target at range, and requires a minimum of 4 in that modifier. It's Range is 3 + the modifier in question.

Have some other random rules, like maybe you can have one attribute that's applied to all valid effects, or maybe you get one effect on even levels and one attribute on odds, or whatever have you.



Finally, when it comes down to how the character works in combat, with the finished product, the final result would just be:

Force Ray: Range = 3+Int. On hit, deal Int damage and knockback Body in distance.

To further simplify, maybe have your stats be static that only change upon rest or level up, so that players can just fill in the variables on their character sheets with explicitly defined numbers that won't change in combat. (so instead of Range = 3 + Int, it'd just be Range = 6)

Don't worry too much about balance, things can always be tweaked and exceptions made. Sometimes, the way the system is imbalanced creates an ecosystem that is acceptable (such as Saving Throws being more important in late-game DnD than AC).

[Grod_The_Giant]

There's nothing that makes classless inherently more complicated or unbalanced than classes-- something like Fate is simple, classless and balanced, while 3.5 D&D is classed and...well, we all know how that one goes. The core distinction is that a class-based system pushes characters into filling certain archetypes (Powered by the Apocalypse systems are a really pure example of this). That's not inherently a bad thing, but it is a distinctive thing-- is that something you want in your game?

[PairO'Dice Lost]

Classless
Cons:
Lack of included fluff can hurt less creative players, i.e. no identity

Categorizing character identity is a huge issue in RPGs, it's not just a crutch for inexperienced or less creative players.

When you're making a character as part of group, you ideally want to be able to convey that character in shorthand to the rest of the group, so that you can get across their capabilities and figure out what role they play in the party. If we're playing GURPS Fantasy and I tell you I'm playing a character with bonuses to IQ and Perception, improved sight and hearing advantages, this and that resistance, this and that weapon skill, a package of advantages from GURPS Magic in this and that category, and so on, it can take some effort to figure out how that fits into the group and how I can contribute. But if we're playing D&D and I tell you I'm playing a High Elf Evoker, you immediately get a sense of my strengths and weaknesses and what I bring to the party.

Two excellent examples of this phenomenon are Shadowrun and Vampire. In Shadowrun, character building is purely points-based (except that there are existing metatypes, basically races, rather than a build-your-own-race setup) and you can dabble in any subsystem as much as you want. However, soon after release players began categorizing things into archetypes like "rigger" (drone controller), "decker" (dedicated hacker), "face" (social skills guy), "street samurai" (heavily cyber-augmented guy with heavy weapons), "shaman" (character specializing in spirit magic), and so forth. This is partly because investing in magic restricts or penalizes investing in cyberware to varying degrees so it's good to specialize, but more importantly every group needs someone who can deal with magic, someone who can deal with computers, someone who can deal with stealth, someone who can deal with ranged combat, etc. so being able to quickly and easily figure out who does what and build to those strengths is valuable.

In Vampire, meanwhile, you have a sorta kinda class system in that you have to pick clans (kind of like a race/class hybrid), disciplines (basically talent trees or feat chains), and so forth, so character building is closer to class-based "combine these large chunks of character mechanics" building than classless "invest in whatever you want" building even though Vampire is largely points-based. But in Vampire being able to say that you're a Tremere mage or a Giovanni face doesn't really get you anything, because there are really no standard adventures, no standard party roles, or anything like that, which is a pretty big failing of the system (or at least its GMing/adventure design portions).

So even in purely classless systems being able to define what kinds of roles and archetypes are needed in a class-like way is important, and providing classes or class-like bundles of flavor and mechanics is pointless if those mechanics and flavor aren't useful and meaningful in-game.

[Lacuna Caster]

In Vampire, meanwhile, you have a sorta kinda class system in that you have to pick clans (kind of like a race/class hybrid), disciplines (basically talent trees or feat chains), and so forth, so character building is closer to class-based "combine these large chunks of character mechanics" building than classless "invest in whatever you want" building even though Vampire is largely points-based. But in Vampire being able to say that you're a Tremere mage or a Giovanni face doesn't really get you anything, because there are really no standard adventures, no standard party roles, or anything like that, which is a pretty big failing of the system (or at least its GMing/adventure design portions).
I would just suggest that ensuring synergistic tactical party roles is mainly a priority if the game assumes the PCs will be a party working toward a common strategic objective. My understanding is that Vampire has a background setting and chargen system which actually heavily implies mutual tensions and rivalries among the PCs, but I don't know how far it follows through on that premise mechanically or in terms of scenario design.

In terms of genre content, it's also not obvious that having diverse, specialised experts on a given 'team' is all that ubiquitous. I mean, the Night's Watch are all rangers, right? Ned Stark and Jaime Lannister are both experienced swordsmen, right? It's the personality and values and key decisions that tend to distinguish them.

[Thrawn4]

IMHO classless is more customizable, whereas classes are simpler. Both can be good. But this is just one factor, both ways can be incredible simple or complex.
You can have a classless system with five skills or with fifty. You can have a simple class system with only a few descriptors (warrior,rogue,mage) or with dozens (D&D has classes, but I am told the grapple rules are quite complex, and AFAIK there are dozens of optional rule books).

So yeah, they are just one way to tweak your system. I usually prefer simple classless systems, but I definitely see the appeal of classes (especially for casual games).

I am afraid your other points are wonky. How easily you can add content or balance is based on complexity, which is kinda influenced by classless VS class system. Identity is always up to the players, as someone already pointed out.

[MoleMage]

I would just suggest that ensuring synergistic tactical party roles is mainly a priority if the game assumes the PCs will be a party working toward a common strategic objective. My understanding is that Vampire has a background setting and chargen system which actually heavily implies mutual tensions and rivalries among the PCs, but I don't know how far it follows through on that premise mechanically or in terms of scenario design.

In terms of genre content, it's also not obvious that having diverse, specialised experts on a given 'team' is all that ubiquitous. I mean, the Night's Watch are all rangers, right? Ned Stark and Jaime Lannister are both experienced swordsmen, right? It's the personality and values and key decisions that tend to distinguish them.

I haven't read much of Vampire, but if it's like other World of Darkness splats, the group of PCs are usually pretty well allied. Generally you'd put your group (coven, coterie, motley, pack, what have you) before your secret society, even if your group isn't part of that secret society. My only experience is with nWoD though, and not oWoD or any LARP springing from its pages.

As to the diverse party members argument, it's more a question of scope. Yes, every member of the Night's Watch can handle a sword to some extent, but Jon Snow has his wolf and leadership qualities, Sam Tarly is smarter than the others and understands the political process, etc. In a game about everyone being knights, skill differentiation can still be part of the system, just instead of fighter/mage/thief you have fighter who can track/fighter who has money/fighter who fights dirty/fighter good at dueling/fighter good at fighting multiple foes. If all of those things are made important within gameplay, and there's a trade-off in being good at them, then you have synergistic party roles, just not in the DnD style combat/exploration/social/magic split.

_____

Personally, I like classless systems, but it is important that you know going in (as a person building a character, as a GM, and by extension as a game designer), what are the important parts of play (and how much time are they going to take during play)? If your game is going to have a lot of design built around combat (and consequently, a large portion of game time spend in combat), then it's important that you make sure players who specialize in non-combat functions have something to do during the time that combat is happening at the table. On the flip side, if you decide that combat is not a hugely important feature mechanically (just a couple of rolls and someone loses), then you need to make sure that characters who specialize in combat have something to do during other portions of play that you have decided to make mechanically significant.

FATE does this well in my opinion by having the Create an Advantage system where anyone can contribute to combat by distracting, undermining, or just generally sandbagging the enemy; and by setting up non-violent conflicts in exactly the same way mechanically as combat (duel of wits? same rules, just different skills) meaning tabletime is spent on whatever your players chose to be important.

Shadowrun in my experience handles it poorly. Shadowrun really drilled down on three things mechanically: astral travel, cyberspace, and real-world combat. Since each of these takes up a comparably large amount of tabletime, and since generally only one or two party members is going to specialize in a thing, this means that at any given time around half your table is either doing nothing or taking boring holding actions. If you deliberately choose an archetype that doesn't use any of those three, your essential function is mechanically resolved with just a couple of rolls and some role-playing (which any person can participate in, skills or no skills), rather than a whole system like is used in the big three.

In summary, your assessment of pros and cons is pretty accurate, but if you're going loose-leaf with character specialization, you'd better make sure that your mechanics are still inclusive to the whole table (if you want IC play to only include logical participants, such as the baker and candlestick maker staying out of combat, write rules in place to give those characters' players something to do while Swordy McPlateface is swinging the Sword of Plates). Actually, this advise still applies in class-based games it's just that of those, I've only used DnD derivatives and those include combat functions in every class.

[PeteNutButter]

Long series of replies:


...

I actually have a similar spell customization system already. It's awesome to be able to be more creative and change things around.


There's nothing that makes classless inherently more complicated or unbalanced than classes-- something like Fate is simple, classless and balanced, while 3.5 D&D is classed and...well, we all know how that one goes. The core distinction is that a class-based system pushes characters into filling certain archetypes (Powered by the Apocalypse systems are a really pure example of this). That's not inherently a bad thing, but it is a distinctive thing-- is that something you want in your game?

On one hand I agree with you. You can have simple classless systems and complex class based ones. But for a player going into a new game in a class system they only have to look at a class list and then class they plan to play. With an equal amount of content that same new player in a classless system would have to browse over a much larger list of those "class" abilities that have been parceled out individually. It is also far more complex in playtesting due to the breadth of options.

As for roles, well there is certainly something to be said for having a role at the table. It helps everyone not be overshadowed.


Categorizing character identity is a huge issue in RPGs, it's not just a crutch for inexperienced or less creative players.

When you're making a character as part of group, you ideally want to be able to convey that character in shorthand to the rest of the group, so that you can get across their capabilities and figure out what role they play in the party. If we're playing GURPS Fantasy and I tell you I'm playing a character with bonuses to IQ and Perception, improved sight and hearing advantages, this and that resistance, this and that weapon skill, a package of advantages from GURPS Magic in this and that category, and so on, it can take some effort to figure out how that fits into the group and how I can contribute. But if we're playing D&D and I tell you I'm playing a High Elf Evoker, you immediately get a sense of my strengths and weaknesses and what I bring to the party.

Two excellent examples of this phenomenon are Shadowrun and Vampire. In Shadowrun, character building is purely points-based (except that there are existing metatypes, basically races, rather than a build-your-own-race setup) and you can dabble in any subsystem as much as you want. However, soon after release players began categorizing things into archetypes like "rigger" (drone controller), "decker" (dedicated hacker), "face" (social skills guy), "street samurai" (heavily cyber-augmented guy with heavy weapons), "shaman" (character specializing in spirit magic), and so forth. This is partly because investing in magic restricts or penalizes investing in cyberware to varying degrees so it's good to specialize, but more importantly every group needs someone who can deal with magic, someone who can deal with computers, someone who can deal with stealth, someone who can deal with ranged combat, etc. so being able to quickly and easily figure out who does what and build to those strengths is valuable.

In Vampire, meanwhile, you have a sorta kinda class system in that you have to pick clans (kind of like a race/class hybrid), disciplines (basically talent trees or feat chains), and so forth, so character building is closer to class-based "combine these large chunks of character mechanics" building than classless "invest in whatever you want" building even though Vampire is largely points-based. But in Vampire being able to say that you're a Tremere mage or a Giovanni face doesn't really get you anything, because there are really no standard adventures, no standard party roles, or anything like that, which is a pretty big failing of the system (or at least its GMing/adventure design portions).

So even in purely classless systems being able to define what kinds of roles and archetypes are needed in a class-like way is important, and providing classes or class-like bundles of flavor and mechanics is pointless if those mechanics and flavor aren't useful and meaningful in-game.

It is nice to have people with guaranteed different roles.


IMHO classless is more customizable, whereas classes are simpler. Both can be good. But this is just one factor, both ways can be incredible simple or complex.
You can have a classless system with five skills or with fifty. You can have a simple class system with only a few descriptors (warrior,rogue,mage) or with dozens (D&D has classes, but I am told the grapple rules are quite complex, and AFAIK there are dozens of optional rule books).

So yeah, they are just one way to tweak your system. I usually prefer simple classless systems, but I definitely see the appeal of classes (especially for casual games).

I am afraid your other points are wonky. How easily you can add content or balance is based on complexity, which is kinda influenced by classless VS class system. Identity is always up to the players, as someone already pointed out.

I certainly agree that identity is up to the players, but classes tend to come with some built in identity. If you are playing a "barbarian" other players will often start with a stereotype of your personality and background which you can either lean into or away from. The same is true for many classes in most games with classes.

On a classless system, it's much more left to the player. They don't have that same springboard.



As to the diverse party members argument, it's more a question of scope. Yes, every member of the Night's Watch can handle a sword to some extent, but Jon Snow has his wolf and leadership qualities, Sam Tarly is smarter than the others and understands the political process, etc. In a game about everyone being knights, skill differentiation can still be part of the system, just instead of fighter/mage/thief you have fighter who can track/fighter who has money/fighter who fights dirty/fighter good at dueling/fighter good at fighting multiple foes. If all of those things are made important within gameplay, and there's a trade-off in being good at them, then you have synergistic party roles, just not in the DnD style combat/exploration/social/magic split.

_____

Personally, I like classless systems, but it is important that you know going in (as a person building a character, as a GM, and by extension as a game designer), what are the important parts of play (and how much time are they going to take during play)? If your game is going to have a lot of design built around combat (and consequently, a large portion of game time spend in combat), then it's important that you make sure players who specialize in non-combat functions have something to do during the time that combat is happening at the table. On the flip side, if you decide that combat is not a hugely important feature mechanically (just a couple of rolls and someone loses), then you need to make sure that characters who specialize in combat have something to do during other portions of play that you have decided to make mechanically significant.

FATE does this well in my opinion by having the Create an Advantage system where anyone can contribute to combat by distracting, undermining, or just generally sandbagging the enemy; and by setting up non-violent conflicts in exactly the same way mechanically as combat (duel of wits? same rules, just different skills) meaning tabletime is spent on whatever your players chose to be important.

Shadowrun in my experience handles it poorly. Shadowrun really drilled down on three things mechanically: astral travel, cyberspace, and real-world combat. Since each of these takes up a comparably large amount of tabletime, and since generally only one or two party members is going to specialize in a thing, this means that at any given time around half your table is either doing nothing or taking boring holding actions. If you deliberately choose an archetype that doesn't use any of those three, your essential function is mechanically resolved with just a couple of rolls and some role-playing (which any person can participate in, skills or no skills), rather than a whole system like is used in the big three.

In summary, your assessment of pros and cons is pretty accurate, but if you're going loose-leaf with character specialization, you'd better make sure that your mechanics are still inclusive to the whole table (if you want IC play to only include logical participants, such as the baker and candlestick maker staying out of combat, write rules in place to give those characters' players something to do while Swordy McPlateface is swinging the Sword of Plates). Actually, this advise still applies in class-based games it's just that of those, I've only used DnD derivatives and those include combat functions in every class.

My system which basically has two drafts (One with classes and another where I pulled the abilities from the classes) has some workarounds for this. The classless design separates noncombat abilities and combat abilities into separate point pools. That means everyone has a role both in and out of combat. Even if you want to play the "big dumb fighter" you'll still be good at something aside from swinging a blade, even if that's just keeping said blade maintained.

[Anonymouswizard]

@OP: pretty much the main thing to consider is how balanced you want the game, and how much testing you're willing to do. A classed system will inevitably lead to less combinations to test than a classless system, which is pretty much the entirity of why they're 'easier to balance'.

I used to have a massive bias towards classless systems, although I'm starting to prefer loosely classed systems that let you pick your party role and then build around that. What I mean is you might be encouraged to pick your combat role (front line, skirmisher, sniper, artillery, buffer, ect) or profession (warrior, alchemist, minstrel, priest, etc) but also get a lot of ability to decide exactly where you want to take it and the ability to add in stuff not covered by the role. It's a sort of 'worst of both worlds' setting, but does provide help with archetyping.


Categorizing character identity is a huge issue in RPGs, it's not just a crutch for inexperienced or less creative players.

When you're making a character as part of group, you ideally want to be able to convey that character in shorthand to the rest of the group, so that you can get across their capabilities and figure out what role they play in the party. If we're playing GURPS Fantasy and I tell you I'm playing a character with bonuses to IQ and Perception, improved sight and hearing advantages, this and that resistance, this and that weapon skill, a package of advantages from GURPS Magic in this and that category, and so on, it can take some effort to figure out how that fits into the group and how I can contribute. But if we're playing D&D and I tell you I'm playing a High Elf Evoker, you immediately get a sense of my strengths and weaknesses and what I bring to the party.

...

So even in purely classless systems being able to define what kinds of roles and archetypes are needed in a class-like way is important, and providing classes or class-like bundles of flavor and mechanics is pointless if those mechanics and flavor aren't useful and meaningful in-game.

The problem is as much one of setting as being classed or classless. For example in The Dark Eye if I tell you I'm going to be playing a Thorwalian Skald, an Ore Dwarf Warrior, a Glade Elf Spellweaver, a Middenrealmian Mage, or a Hill Dwarf Blessed One of Boron you know exactly what I'm playing and what part of the game's combat/wilderness/social/knowledge/magic/etc roles I fulfill. It's the same with Shadowrun, the setting and default PC party puts forward X areas of expertise a team might need, and each area of expertise has one or more archetypes that players have come to recognise. A settingless system will do much better with classes, if we're playing Fantasy AGE and I say my character is a Beastfolk Mage focused on the Illusion and Psychic Arcana* you know that I'm not intending to take a primary combat role, and as such a Warrior might be useful.

* A massive advantage to having spell categories actually. It helps pin down exactly which of the class's roles your character is filling.

[Grod_The_Giant]

Long series of replies:



I actually have a similar spell customization system already. It's awesome to be able to be more creative and change things around.



On one hand I agree with you. You can have simple classless systems and complex class based ones. But for a player going into a new game in a class system they only have to look at a class list and then class they plan to play. With an equal amount of content that same new player in a classless system would have to browse over a much larger list of those "class" abilities that have been parceled out individually. It is also far more complex in playtesting due to the breadth of options.

As for roles, well there is certainly something to be said for having a role at the table. It helps everyone not be overshadowed.



It is nice to have people with guaranteed different roles.



I certainly agree that identity is up to the players, but classes tend to come with some built in identity. If you are playing a "barbarian" other players will often start with a stereotype of your personality and background which you can either lean into or away from. The same is true for many classes in most games with classes.

On a classless system, it's much more left to the player. They don't have that same springboard.
It sounds like you're strongly leaning towards being at least generally class-based.


The classless design separates noncombat abilities and combat abilities into separate point pools. That means everyone has a role both in and out of combat. Even if you want to play the "big dumb fighter" you'll still be good at something aside from swinging a blade, even if that's just keeping said blade maintained.
That's a pretty neat mechanic-- I like it, and I'd certainly encourage you to keep some variation of it in a class system. (Perhaps a gestalt thing, so you'd be a Wizard//Conman or a Fighter//Smith?)

[Yerok LliGcam]

Classless

Pros:
Ultimate Customization
Easy to add new content
Open-ended "fluff"/play whoever however

Cons:
Can be intensely complex
Can be very hard to balance
Lack of included fluff can hurt less creative players, i.e. no identity
Inexperienced/unoptimizing players suffer, potentially making garbage characters



liked your list, here's my thoughts to it:

I'm a huge fan of the 7d system developed by DMG on youtube. i'm one of his supporters, and he created a classless system.

the interesting thing he did to try to remove the complexity was this:

1. no classes (obviously)
2. no levels.
- this one's interesting. you don't have a level, and the way he balanced it is quite cool and well balanced (IMO) which allowed players to keep play simple, no growing modifiers that make it hard to balance a mixed party, and the formula for dice checks is pretty straightforward too.
3. no hit points. instead, there's an injury system. if something takes 7 injuries it dies/falls unconscious.
4. no gold, there's economic level, as well as other simplifying systems in place like that.


so here's the breakdown more or less, if you wanna learn more you can check out his website or youtube channels. just google 7Dsystem.

okay:

1. there's 7 levels of difficulty.

each one is associated with a die. (d2, d4, d6, d8, d10, d12, d20)

type = difficulty, number of dice you roll = your skill in that thing.

what things can you be good at?

there's 8 basic abilities, and you can branch out into feats, then expertise. and you can eventually make 1 expertise a specialty.

physical, mental, education, senses, emotion, creative, constitution, social

these all branch out to customize your character, so for assassin types, physical -> dexterity -> sneak, sleight of hand. and education->coup de grasse -> assassination.

you get more skilled in batches of 2 ranks. basic abilities = 2 ranks, feats = 2 ranks, expertise = 2 ranks. then specialty = 5 ranks.


EVERY check, combat, non-combat, roleplaying etc.

follows this formula:

Difficulty = up to DM / calculated by a combat score/protection value (determines the type of dice you roll)

how many you roll =

1 base die + ranks in base score, + ranks in applicable feat, + ranks in applicable expertise + ranks in applicable specialty.

so you want to intimidate someone.

that's a social -> intimidate -> coerce -> coerce specialty check.

you count up your ranks and roll the type of die.

ALL ROLES - to succeed you need to role a 1.

and that's it.

i've had remarkable experiences with brand new players to TTRPG, as well as seasoned veterans, all picking up this system within the first 5 minutes of a game.

it's a lot of fun. i recommend checking it out.

[Yerok LliGcam]

There's nothing that makes classless inherently more complicated or unbalanced than classes-- something like Fate is simple, classless and balanced, while 3.5 D&D is classed and...well, we all know how that one goes. The core distinction is that a class-based system pushes characters into filling certain archetypes (Powered by the Apocalypse systems are a really pure example of this). That's not inherently a bad thing, but it is a distinctive thing-- is that something you want in your game?

amen! (10 char limit for response? really?)

[Composer99]

@OP: I would say that if you go classless, if you think players would appreciate being able to build to strong archetypes, if that's what they want, you could include a few sample builds:

"Here's a high-fantasy wizard! [Build...]"
"Here's a classic "paladin"-style holy warrior! [Build...]"
"Here's a necromancer! [Build...]"

And so on. Maybe have the builds only partly fleshed with a few core features, leaving the rest for players to fill out to make their characters their own.

Those examples are obviously presuming a certain style of game, but you could do the same with cyberpunk, modern, swords-and-sorcery, space opera - most any genre, really.

[GreatWyrmGold]

As a rule, I prefer classless. However, there are two times when character classes make sense:

1. When character classes represent something specific in the game—say, distinct disciplines of magic or superheroic ideals. It can be fine to have one or a few "generic" classes for people who don't fall into those categories (self-taught mages, superheroes without a distinct ideology, etc), but if you have more than one, they need some kind of meaningful in-world distinction between them. Same goes for the other classes; if multiple classes have very similar flavor, you should find some meaningful way to distinguish them. I can't think of any examples of this off the top of my head.

2. When character classes are there to facilitate character creation. I'm not personally a fan of restrictions class-based systems impose by their very nature, but one man's restrictions are another man's guidelines. When it comes to making characters quickly, just about the only thing faster than a class-based system is a premade character. Classes not only help accelerate character creation, but character design (ie, helping you figure out what character you want to play); instead of needing to consider all the aspects you think of, you can restrict it to the broad archetypes presented by the rulebook and then work within the confines of those archetypes.
This is yet another way that D&D makes a good introductory TRPG. It's simple to pick up a rulebook, roll up a fighter/cleric/monk/whatever, and get playing. (I'd also like to mention that 5e's "archetype" system, essentially classes within classes, is also well-designed. You usually get a couple of levels to get used to your general character archetype before having to specialize.)

Personally, I'm a fan of GURPS's system. They have one of the freest-form classless systems I've seen, but they also churn out "character templates" which have just about all the benefits of a class system...but since they're built with the classless point-based system, if you want to modify it freely, you can. I'd argue that they're even easier to pick up and play than D&D classes (though few templates have much to offer in the way of character advancement, so D&D still offers more overall benefits for introductory games than even lite versions of GURPS).

[Knaight]

@OP: pretty much the main thing to consider is how balanced you want the game, and how much testing you're willing to do. A classed system will inevitably lead to less combinations to test than a classless system, which is pretty much the entirity of why they're 'easier to balance'.
That said, when they have balance issues they're often a lot more intractable - unbalance tends to hit large chunks of content at once, and you end up redoing entire classes instead of having to tweak the occasional skill or power.


A settingless system will do much better with classes, if we're playing Fantasy AGE and I say my character is a Beastfolk Mage focused on the Illusion and Psychic Arcana* you know that I'm not intending to take a primary combat role, and as such a Warrior might be useful..
I'd actually go exactly the other way with this - almost every advantage classes have is in the advancement of specific settings, and a class based generic system is almost certainly a terrible idea. The big part is that classes are setting; every class made says something about the setting they're in. Some classes at least say fairly little about their setting, with a Fighter implying the existence of professional combatants and specific implementations usually pinning down some setting details around expected technology and the expected role of these warriors in society; some classes say a great deal about their setting, with a Glitterboy creating on the spot an order of futuristic knights in fancy powered armor with massive railguns and implying a great deal about the manufacturing capacity of the society they're from or a Motumancer just quietly inserting anarchist occultists that gain power by disrupting social groups and use it to sow chaos into a setting.

When you actually have a particular setting, this is great stuff. RIFTS has a lot of problems, that they made a Glitterboy class isn't one of them*, Unknown Armies has a lot of really cool classes in it that don't actually use the term class or classes, of which the Motumancer is one of many, picked basically at random because I was reading that particular page recently. Even early D&D has a lot of this, with some classes having a lot of identity built in. Fighters and thieves are one thing, but the mere presence of a Magic User says a fair amount of the setting, as does the magic it comes with. The cleric meanwhile is a distinctive chunk of setting identity, unnoticed now only because it's become ubiquitous from imitation. That whole arcane-divine magic split is distinctive, and it's there from just putting in the classes they put in. Classes inherently populate a mileu.

For a generic or semigeneric (e.g. d6 Space, which is for a wide range of science fiction, space opera, space fantasy, etc.) systems though? Every class you put in it hacks away at how generic it is, and cuts off settings that could otherwise be there. The ways around this mostly involve adding lots and lots of mechanics, until you have so many classes that a small and appropriate subset can be picked while still leaving enough to support the setting, and any advantages class systems might have in terms of ease of character creation and the like are right out the window by then.

*As a rule RIFTS problems aren't that they made something, but how. It's a game with a lot of cool ideas in it, and its implementation is awful. I'd call it uniformly awful, but it manages to be awful in so many different ways simultaneously.

[PeteNutButter]

Thanks everyone for the replies. I think I'm pretty much set with how I'm going to do this. I'm going to go with a "class" system but enough customization to keep it interesting and not so pigeon-holey. For me that means mixing 3 layers of classing.

Layer 1: Your base class or Chassis. This determines your combat role.
Layer 2: Your specialization. These can be applied to any chassis and significantly alter game play with some built in fluff/out of combat abilities. These will be most similar to the D&D type classes people are familiar with.
Layer 3: Your profession. Primarily a fluff/out of combat perk set, but with a couple combat benefits thrown in. (Like if 5e's background system also gave some combat abilities and different things as you leveled up.)

With the 4 chassis, 10 specializations, and 10 professions I have to start that makes 400 different combinations. From there I can sprinkle on "feats" for further customization.

[nineGardens]

So... I suspect I am late to the party and you've already decided all the important stuff but....
One question, one comment.

Question:
What makes you feel like Classless systems are easier to add new content?

In a class system, I can throw in a new class, and BOOM, new content. A new player can come and play that class, no problemo.

In a classless system, in theory, all the skills of the game already exist. You can't add "Alchemy" as a skill halfway through a campaign, because then how do you decide how good your current players are at it.
If you have a "Everything is feats" style system, then retroactively adding feats can lead to buyers remorse for those players who might have wanted to take those feats at level 2 before the feat existed.

If you add a class however, with a bunch of classlocked features, if I'm playing a sweet paladin or whatever and you add pirates to the game, I'm unlikely to sit there and go "Gee, I wish I had that +2 bonus with firearms" - because that is part of the class, its part of a package deal, and I have my own package.

Perhaps I've missed something- what is it about classless systems that makes you think it is easier to add new content later, as opposed to classed systems?



And Comment:
By the looks of your last post, it seems that you are aiming to build a multiclass system (everyone has multiple classes by default).
Awesome!
Good luck!

I spent the last few years building a multiclass system, (see signature) and I can tell you, they are an absolute ball to play.
If you are sold on the idea of having classes as "Chassis, Specialty, profession", then by all means go for it, but I can tell you that that level of structure is by no means needed; as far as game design is concerned you could also just have a single list of classes and tell your players "Pick any three".
There are advantages and disadvantages to both "Layered" and "free for all" multiclass systems. If you are working on this system and ever want to talk to someone who has gone a similar route, flick me a PM or whatever, and I'll be able to tell you what has and has not worked for me, or just bounce ideas around.

[PeteNutButter]

So... I suspect I am late to the party and you've already decided all the important stuff but....
One question, one comment.

Question:
What makes you feel like Classless systems are easier to add new content?

In a class system, I can throw in a new class, and BOOM, new content. A new player can come and play that class, no problemo.

In a classless system, in theory, all the skills of the game already exist. You can't add "Alchemy" as a skill halfway through a campaign, because then how do you decide how good your current players are at it.
If you have a "Everything is feats" style system, then retroactively adding feats can lead to buyers remorse for those players who might have wanted to take those feats at level 2 before the feat existed.

If you add a class however, with a bunch of classlocked features, if I'm playing a sweet paladin or whatever and you add pirates to the game, I'm unlikely to sit there and go "Gee, I wish I had that +2 bonus with firearms" - because that is part of the class, its part of a package deal, and I have my own package.

Perhaps I've missed something- what is it about classless systems that makes you think it is easier to add new content later, as opposed to classed systems?

I get your point. What I mean to say in a classless system it's pretty easy to come up with content later piecemeal. In a class system I need a series of abilities to go with a new class. Also in a class system if I suddenly think up a new ability that fits an existing class, I have to either alter the current class or risk over-tuning it by adding the new feature to the class. A third option would be to add variant features Ala 3.5 D&D.

As for buyer's remorse, well I'm still very early in the playtesting phase (we've done games with pregenerated characters), and my playgroup is understanding of that.

I'd still say it's generally easier to come up with more individual abilities than an entire new class (which should likely come packaged with multiple abilities). Sometimes if it's thematic it would be easier to add the class, ala your gunslinger example.



And Comment:
By the looks of your last post, it seems that you are aiming to build a multiclass system (everyone has multiple classes by default).
Awesome!
Good luck!

I spent the last few years building a multiclass system, (see signature) and I can tell you, they are an absolute ball to play.
If you are sold on the idea of having classes as "Chassis, Specialty, profession", then by all means go for it, but I can tell you that that level of structure is by no means needed; as far as game design is concerned you could also just have a single list of classes and tell your players "Pick any three".
There are advantages and disadvantages to both "Layered" and "free for all" multiclass systems. If you are working on this system and ever want to talk to someone who has gone a similar route, flick me a PM or whatever, and I'll be able to tell you what has and has not worked for me, or just bounce ideas around.

Will do!

[Lacuna Caster]

If you add a class however, with a bunch of classlocked features, if I'm playing a sweet paladin or whatever and you add pirates to the game, I'm unlikely to sit there and go "Gee, I wish I had that +2 bonus with firearms" - because that is part of the class, its part of a package deal, and I have my own package.
Technically, yeah, but the players can still ask reasonable questions like "where were these pirates all this time?" or "how did firearms suddenly get invented when we never saw them on the battlefield?" Plus, if you allow multiclassing, there's still the same potential for buyer's remorse.

[GreatWyrmGold]

Question: What makes you feel like Classless systems are easier to add new content?
In a class system, I can throw in a new class, and BOOM, new content. A new player can come and play that class, no problemo.
In a classless system, in theory, all the skills of the game already exist. You can't add "Alchemy" as a skill halfway through a campaign, because then how do you decide how good your current players are at it.
My first thought is how GURPS would have done it. I don't have a clue what skill system you have in mind, but I'm pretty sure most systems (GURPS included) would have you be as good as if you hadn't invested any character points/XP/whatever into the skill.


If you have a "Everything is feats" style system, then retroactively adding feats can lead to buyers remorse for those players who might have wanted to take those feats at level 2 before the feat existed.
If you add a class however, with a bunch of classlocked features, if I'm playing a sweet paladin or whatever and you add pirates to the game, I'm unlikely to sit there and go "Gee, I wish I had that +2 bonus with firearms" - because that is part of the class, its part of a package deal, and I have my own package.
...But you don't think people might want to play the pirate class instead? Especially if they were playing just about anything but a paladin? (e.g, a rogue or swashbuckler or ranger)


Perhaps I've missed something- what is it about classless systems that makes you think it is easier to add new content later, as opposed to classed systems?
Probably the fact that you don't have to design a whole new class at once. If it was a GURPS-ish system, you could design just one or a couple of advantages and then playtest/refine those, for instance.

[Anonymouswizard]

I'd actually go exactly the other way with this - almost every advantage classes have is in the advancement of specific settings, and a class based generic system is almost certainly a terrible idea. The big part is that classes are setting; every class made says something about the setting they're in. Some classes at least say fairly little about their setting, with a Fighter implying the existence of professional combatants and specific implementations usually pinning down some setting details around expected technology and the expected role of these warriors in society; some classes say a great deal about their setting, with a Glitterboy creating on the spot an order of futuristic knights in fancy powered armor with massive railguns and implying a great deal about the manufacturing capacity of the society they're from or a Motumancer just quietly inserting anarchist occultists that gain power by disrupting social groups and use it to sow chaos into a setting.

Eh, I'd say that both classed and classless benefit from settings, but that until you get to 'anything goes' classes make the structure of the game much more apparent.

I mean let's say we have a game. If all the classes are based on your combat role then we can say that the game is combat focused, while if the classes are focused on the weird magic your character can cast we know that magic is an important part of the game.


When you actually have a particular setting, this is great stuff. RIFTS has a lot of problems, that they made a Glitterboy class isn't one of them*, Unknown Armies has a lot of really cool classes in it that don't actually use the term class or classes, of which the Motumancer is one of many, picked basically at random because I was reading that particular page recently. Even early D&D has a lot of this, with some classes having a lot of identity built in. Fighters and thieves are one thing, but the mere presence of a Magic User says a fair amount of the setting, as does the magic it comes with. The cleric meanwhile is a distinctive chunk of setting identity, unnoticed now only because it's become ubiquitous from imitation. That whole arcane-divine magic split is distinctive, and it's there from just putting in the classes they put in. Classes inherently populate a mileu.

*As a rule RIFTS problems aren't that they made something, but how. It's a game with a lot of cool ideas in it, and its implementation is awful. I'd call it uniformly awful, but it manages to be awful in so many different ways simultaneously.

Oh, certainly. As I said, I think both benefit from lots of nice setting. SETTING FOR ALL.

Unknown Armies is actually a great example of both. The classes benefit from the setting, in that a Cinemancer or Avatar of the Star tell you exactly the kind of thing you can expect about magic in the setting (my current campaign includes a Weabomancer as a minor character who has powers based on thei misconceptions about Japan).

I still dislike how the arcane-divine split is now so common people think of it as normal. Almost every time I see a game with a magic system I see 'what about divine magic', as if it is something that has to be inherent to every magic system.


For a generic or semigeneric (e.g. d6 Space, which is for a wide range of science fiction, space opera, space fantasy, etc.) systems though? Every class you put in it hacks away at how generic it is, and cuts off settings that could otherwise be there. The ways around this mostly involve adding lots and lots of mechanics, until you have so many classes that a small and appropriate subset can be picked while still leaving enough to support the setting, and any advantages class systems might have in terms of ease of character creation and the like are right out the window by then.

True, but I should have worded my post clearer. The point was that it's not being generic in the way that say GURPS or Fudge are, but that it's settingless with a number of core assumptions. Call it the 'D&D-style fantasy' style, you aren't making an explicit setting (world), just creating a basic 'setting' via the assumptions you make. This is partially because I think a game with explicit assumptions is a better game, and a game with classes is one where they have to present those assumptions.

Although I'll also a note that a classed game that doesn't notice it's assumptions or has different assumptions to you can be incredibly disappointing. For example White Star (https://www.drivethrurpg.com/product/148169/White-Star-White-Box-Science-Fiction-Roleplaying-Swords--Wizardry?term=white+star&test_epoch=0) is a game I have in my pdf library because I saw that it claimed to allow 'classic science fiction adventure'. I thought it sounded cool, a game based on things like Lensman and Foundation but with OSR mechanics sounded like just the thing I needed to run awesome science fiction adventures where engineers, pilots, secret agents, and potentially soldiers explored the galaxy. My disappointment began when I cracked open the pdf and noticed that it was based entirely on Star Wars, down to having a knock-off Jedi as a character option but no engineer (sorry Scotty). Now White Star is still one of my choices when I want to run pulpy space opera because with the Companion it becomes a grab bag of popular SF, but it does point out something I've seen often overlooked in games (communicate your assumptions clearly).

I suppose what I should be saying is that settingless is helped by having some sort of structure until you're used to the system. That structure might be closer to 'X stat points, Y skill points, and Z special abilities' or it might be an attempt at a very generic class system. But I agree that almost any structure will begin adding assumptions and setting, the question is just when does the assumed setting become too much. Really the only way we can go without having any implied setting is to be as toolkity as Fudge, which isn't bad but Fudge is already that and is very good at being a toolkit.

Note that for a variety of reasons I will always choose to go classless over classed. I've just come to like a big more structure to it than the GURPS method.

[Knaight]

Eh, I'd say that both classed and classless benefit from settings, but that until you get to 'anything goes' classes make the structure of the game much more apparent.
...
Oh, certainly. As I said, I think both benefit from lots of nice setting. SETTING FOR ALL.
...
I suppose what I should be saying is that settingless is helped by having some sort of structure until you're used to the system. That structure might be closer to 'X stat points, Y skill points, and Z special abilities' or it might be an attempt at a very generic class system. But I agree that almost any structure will begin adding assumptions and setting, the question is just when does the assumed setting become too much. Really the only way we can go without having any implied setting is to be as toolkity as Fudge, which isn't bad but Fudge is already that and is very good at being a toolkit.
I think I've found our core disconnect - I have settings, I like making settings, and I'm just generally covered on settings. I get games for their mechanics for the most part, and while I'll learn something setting heavy if it's cool enough (e.g. Unknown Armies) I tend to favor systems with at little of that as possible, but with tools that I can steal for structural purposes.

Fudge is also both my favorite system and the one I did a lot of games in and thus probably the one that most shaped my expectations, and while I generally come in less toolkity I definitely favor that general end of the spectrum.

[Anonymouswizard]

I think I've found our core disconnect - I have settings, I like making settings, and I'm just generally covered on settings. I get games for their mechanics for the most part, and while I'll learn something setting heavy if it's cool enough (e.g. Unknown Armies) I tend to favor systems with at little of that as possible, but with tools that I can steal for structural purposes.

Fudge is also both my favorite system and the one I did a lot of games in and thus probably the one that most shaped my expectations, and while I generally come in less toolkity I definitely favor that general end of the spectrum.

It's not that I don't like making settings, it's just that I can rarely make both the setting and the adventure (plus I find that other people tend to be more inventive than I am, so my standard move is to take a high level setting guide and fill in the details myself).

Plus I really do like it when mechanics and setting interact neatly, and tend to find toolkits a little bland. I love Fudge, and have games I plan to run with it, but whenever I use a generic there will always be a setting/system disconnect for me no matter the work I put in.

[GreatWyrmGold]

Plus I really do like it when mechanics and setting interact neatly, and tend to find toolkits a little bland. I love Fudge, and have games I plan to run with it, but whenever I use a generic there will always be a setting/system disconnect for me no matter the work I put in.
I won't say you're wrong, but I will say that I enjoy finding tools in the GURPS toolbox to make its rules fit weird settings.