PDA

View Full Version : D&D 5e/Next Stacking class mod for 5e (brainstorming)



Greywander
2021-06-03, 12:02 AM
I've mentioned this idea in the context of a big overhaul mod that changes many different features of 5e (https://forums.giantitp.com/showthread.php?624980-Let-s-make-some-stacking-classes), but it should be feasible to make a simple stacking classes mod and integrate it back into 5e without making any other sweeping changes. Now, classes do form a core part of the mechanics, so there's still a lot that we would/could change, but it wouldn't really be much different from most other homebrew classes.

So what do I mean by "stacking classes"? Well, each class is cut down to just 4 levels, and as you level up you'll need to pick a new class every time you finish your current one. This means a 20th level character will have a total of 5 classes. This also conveniently lines up such that you get a proficiency bump every time you start a new class. One of the big motivations for doing it this way is increased flexibility in character customization. There have been a number of character concepts I've had that really want specific features from this or that class, but not necessarily anything else from that class. Stacking classes makes it easier to mix'n'match features from different classes. It's sort of finding a happy middle ground between a rigid class system and the more open skill-based systems.

As for the implementation, how I'd start is by taking the existing base classes and condensing them down into two stacking classes; a basic class and a master class. Each subclass also gets reworked into a stacking class. "Basic" classes might be the only ones you could take at 1st level, with the "intermediate" former subclasses being available from 5th level and onward, and "master" classes available at 13th or 17th level. But there's still a lot of room to figure this all out. Ideally, each class is balanced against all the others, so while the master classes might have some powerful abilities you wouldn't want a 1st level character to have, they're comparable to what you'd get from any other class in the late game. Also, ASIs/feats would be separated from class advancement, probably purchased separately with XP.

With each class containing only four levels, we can easily set up a basic formula to determine what sorts of features we get and when. From the post linked above, we have:

1st level - Core Feature. A feature central to that class. Gets stronger the more times you gain this feature, encouraging you to take similar classes. Optionally, you could swap out a class's core feature with a different one.
2nd level - Gimmick. A unique trick that differentiates classes with the same core feature. This feature focuses on the mechanical aspect of the class.
3rd level - Ribbon. What the gimmick does for mechanics, this does for roleplay.
4th level - Capstone. Your reward for completing the class.

Having thought about it, though, there's an argument to be made for moving the gimmick up to 1st level, either switching places with the core feature or sharing 1st level with it, so that classes with the same core feature don't feel identical. Still not sure what would work best. If we did just move the gimmick up, then what sort of feature should be given at 2nd level?

Now, before we try to write up the individual classes, there are some issues we need to resolve. One of those is Extra Attack. If Extra Attack is a core feature, then that means fighters would start with two attacks right from 1st level, and would gain up to six attacks if you take all classes that get Extra Attack as a core feature. Instead, here's what I'm thinking: The basic fighter gets Extra Attack as a capstone, allowing them to make an additional attack, up to three max. Then, at a specific level, say, 9th level, everyone gets an automatic Extra Attack, allowing them to make up to two attacks. Fighters, who already have Extra Attack, can then make three total, while everyone else is stuck at two. The master fighter also gets Extra Attack as a capstone, allowing for up to four attacks. Thus, any character can eventually make two attacks (though non-martials won't have as many bonuses from class features), while any character who dips into either fighter class can make three attacks, and anyone who gets both fighter classes can make four attacks.

Another idea I'm playing with is hit points, resources, and rests. Messing with HP will necessarily require tweaks to anything that involves healing (e.g. spells), but I think there's a potential to make things more interesting. As a basic modification, every character, including NPCs, would have a number of hit dice equal to their proficiency bonus, and the equivalent amount of HP. PCs would start with two hit dice, likely one from their race (probably d8 for medium and d6 for small) and one from their class, and every time you advance to a new class you'd gain the hit die of that class. Now, that's a lot less HP than you normally get, which is going to make combat a lot deadlier if we don't also reduce damage. But, here's what I'm thinking: Once all your HP is gone, further damage starts coming out of max HP. You don't fall unconscious until you lose all your max HP. Then, we can make it relatively easy to regain HP, but difficult to restore max HP. One of the main benefits of this is that we can make HP healing relatively cheap, e.g. regen abilities or a healing cantrip, without them completely unbalancing the game.

As for rests, I'm considering a four-tiered rest system: A breather is 1 minute of rest (generally for any ability that is meant to be once per encounter), a break is 1 hour of rest (a short rest), a repose is 8 hours of rest that includes at least 6 hours of sleep and no more than 2 hours of light activity (a long rest; creatures that don't require sleep can do light activity throughout), and a vacation is a week of rest that includes at least 8 hours of sleep and 8 hours of light activity each day (e.g. "once per adventure" abilities). Using the HP mod above, you'd probably regain full HP after either a breather or a break, you could spend hit dice during a repose/long rest to regain max HP, and a vacation would fully restore max HP and hit dice. I'm also leaning toward making 6th+ level spell slots restore after a vacation rather than a repose/long rest, to make the use of 6th+ level spells more meaningful (also to curb the power of casters).

For resource pools, there's a couple things I'm considering. One is to have one or more generic resource pools that every character has that can give you extra uses of your abilities. You kind of already see this with several feats and a few class features that let you use it once for free, then require you to expend a resource (spell slot, psy die, ki, etc.) to use it again. If we're needing to write new abilities for classes, we could also introduce new resource systems. One idea I had was for a resource that you regain at the start of every turn, but then again that might be somewhat redundant with the action economy. The addition of vacations also opens up the possibility of having cool "last stand" type abilities, giving a substantial power boost while being essentially limited to once per adventure.

Let's talk about spellcasting for a bit. This was an issue I've dealt (https://forums.giantitp.com/showthread.php?624437-Difficulty-making-customizable-spellcasting-with-stacking-classes) with before (https://forums.giantitp.com/showthread.php?612077-5-5e-idea-for-stacking-caster-classes-but-how-to-handle-different-types-of-casting). The eventual solution was to create a unified Spellcasting feature that would give you a limited number of spells known and spell points. Some classes, like the wizard, might get a gimmick feature (e.g. Spellbook) that lets them prepare additional spells, but by default every spellcaster is a spells known caster. Every time you gain a level, you can trade your spell points for alternate features given by your class. For example, a wizard might be able to trade spell points away to get Vancian-style spell slots, where they have to prepare spells into specific slots but overall get a lot more castings than they would with just spell points. This way, it doesn't matter if you don't take any other "wizard" classes, as any caster class will get you spell points that could be converted into Vancian slots. Spell points can also be refunded when gaining a level, allowing you to change how your casting works as you gain new classes. I also want each caster class (or possibly every class, including non-casters) to have their own spell list which gets added to your personal spell list (a sort of build-your-own-spell list), but this would be a lot of work to actually implement.

While we're on the subject, let's also talk about core features that aren't spellcasting. Spellcasting takes up a huge design space in D&D, leaving little room for similar features for non-casters. In fact, spellcasting is probably the single most in-depth feature that exists in D&D. One idea could be to take the Battle Master maneuvers and turn them into a core feature for most/all martials, and each time they gain that core feature they upgrade their dice to a larger size and/or gain more dice and learn a couple more maneuvers. I'm also leaning toward making the dice refresh at the start of each turn, so that you can always use your maneuvers. I'm trying to think of what else could be used as a core feature. Sneak Attack could be another one, but I could also see just making Sneak Attack an Assassin gimmick feature that scales with level/proficiency bonus, so you'd only need to get it once. Not everything that uses a resource or scales needs to be a core ability after all; we could, for example, make a monk's ki automatically scale with proficiency. This is one area where I could really use some ideas.

I'll end off for now with multiclassing. By definition, multiclassing is required since each class is now only 4 levels. As such, I think we can repurpose the idea of multiclassing to something else. One of the homebrews I've come up with for vanilla 5e set your character level to be equal to your highest level in a single class, and "multiclassing" meant paying extra XP to get extra class levels, working similar to a gestalt character. I think that concept could be repurposed here. When you finish one class, you can choose to multiclass, which locks your character level (meaning you don't get more HP or increase your proficiency bonus), allowing you to buy the features of your new class at a reduced XP cost. I have my own rules for handling gestalt characters which can easily handle any instance of a character with more class levels than character levels; I won't write them up here for the sake of brevity, just know that it shouldn't be an issue.

Oh, as a final final thing, I'd probably take the opportunity to rename every instance of the word "level" so that there isn't as much confusion with new players. E.g. "level" might refer only to "character level", whereas each class has four "ranks" instead of levels, and spells are in different "grades", etc. Exact terminology to be determined. This way there will be less confusion over why the new player's 9th level ranger is unable to cast 9th level spells.

That's all I've got for now. If we can sort out some of these issues, then we might be able to start writing up the actual classes. I think this could be a very neat and fun mod for 5e, though, making it a lot easier to realize specific character concepts. Thanks for reading, I know I have a tendency to be verbose, so I do appreciate those who take the time to thoroughly read my posts instead of just skimming them.

Dienekes
2021-06-03, 09:16 AM
Wouldn't the Extra Attack problem be solved with each of the Intermediate subclass based classes getting Extra Attack at 1st level?

The other option might be to instead extend the class rang out to 5 levels. Though that does mess up the simplicity of having them align with proficiency.

For Hit Points. It sounds like you're thinking of a Vitality/Wound system. Which is cool, that can work well. I assume, based on the term "maximum hit points" that taking damage from that pool lowers the maximum amount you can have in the normal Hit Point pool until you become fully healed. That stuff can work, but my general thoughts is that this will make having a Healer in the party even more important than it currently is. Since everyone will want to top themselves off at all times and avoid taking Max HP damage at all costs. Meaning the healer is gong to be taking in combat turns healing a lot. Depending on how slow healing Max HP is, this can also slow down the adventure. Healing is already very cheap in 5e.

To avoid annoying your Clerics who don't want to just cast heal every turn. I'd focus on making healing spells that also provide different buffs. Or spells that siphon HP away from enemies and onto allies. Just to let them have more decision points than just how much they are casting heal.

I'll need more details to think of the math on this. But I can definitely see people needing to use those Vacations to heal up all their lost Max HP and refusing to go adventuring until they do. This will make the game slower, rather than faster as I think you want with the comment of making easier cantrip healing spells.



For Combat Superiority. While I personally love giving martial combat more stuff to do every round, and giving Combat Superiority dice at every turn accomplishes that. From my own experiments I've found what tends to happen is that the player picks the most damage-dealing maneuver or two and just spams them.

The way I got around the problem is essentially turning Martial Combat into a deck builder game. With the fluff that based on weapon positioning and stances not every move is available at every turn. So you have a deck of cards that are each essentially a maneuver. Draw some amount turn 1, and each turn you draw another. Once used, you cannot use the maneuver again. Until you spend your Bonus Action to reshuffle your used and held maneuvers back into your deck and draw the Turn 1 amount again.

It was very fun. Made the martial character need to think on their feet a lot, using the best maneuver card for the situation at hand, determining when they need to reshuffle. But it required a lot more prep work and understanding of the system for the player. I think it would actually fit your system quite well. Since all these new classes could just grant from a selection of Maneuver Cards as their Core Feature or something, that determines what their martial style is. With each class granting perhaps an additional card drawn at 1st Turn. Or simply make that number equal to your Proficiency.

If you do this, or even just the regular Combat Maneuver system, I'd suggest making at least 1 path for martials that just don't want to deal with this sort of things. Your Barbarians that just want to Rage every turn or something.



Personally for your levels and ranks and whatnot. I'd keep Level as simply your total number of levels in any class. It's simple enough, I think, for someone to pick up they are 4 level Ranger, 4 level Hunter, 2 level Sniper for a total of a 10th level character. I would definitely change spell level though. I've seen Circles used to describe them before, which I've always found simple yet evocative.

rel
2021-06-12, 01:06 PM
Way back in the 3.5 days we experimented with a similar idea.
The only way we got it to work was by having your classes exist in tiers.
Base classes taken at level 1
Intermediate classes that build on the base classes taken at 5, more advanced classes built on the intermediate classes taken at 9 and so forth.
Crucially, you could only take a class of the appropriate tier so you never ended up with level 20 power at level 1 or the reverse.
Also, more advanced classes had prerequisite classes so if your base classes were mage, thief and fighter, the advanced class evoker had a prerequisite of mage while muscle wizard might have a prerequisite of fighter or mage and so forth.

Herbert_W
2021-06-12, 07:11 PM
There's a lot that I could talk about here, but since I'm a little pressed for time at the moment I'll just talk about the way that the leveling system here is structured.

I like what you're trying to do, but having the player gain more abilities of the same overall power level throughout their career breaks one of the core assumptions of DnD: that players get better stuff as they reach higher levels. It'd be odd for a character's capstone ability at level 20 to be selected from the same set of abilities that anyone could get at level 4.

Rel already suggested one solution, which is to have tiered classes where players can only select a class of the appropriate tier - which goes further than your system, where players can (but aren't required to) select classes of a higher tier at high character levels. This would dramatically cut down on the flexibility of the game, as a player's selection of classes at any given tier would be only a subset of all classes that exist. Furthermore, if some high-tier classes have prerequisites then players will need to plan their build in advance to get in to those classes which reduces the spontaneity and responsiveness to campaign events (i.e. a character choosing to take levels in a class because of, and therefore as a lasting reminder of, what they've gone through) that would otherwise be appealing features of this system.

I think that a better solution would be to cap class levels based on character level. That way, higher level characters can get higher level features by increasing the level cap in classes that they already have - and they retain the flexibility to pick up any new class that would be appropriate for their character arc.

What I'm going to propose is a bit complicated, so I'm going to lead into it by explaining why I think that complexity is beneficial. Let's call the system that you're currently proposing "height first," meaning that characters "grow" each class up to the maximum class level before starting on the next one. A "width first" system would have each character select all of their classes early in their career, and then grow each class roughly evenly. The Legend system (https://www.hedra.group/legend) is effectively this, except that it gives each character multiple "tracks" instead of classes and all of them are chosen at 1st level.

The problem with a height first system is that characters get abilities from the same list of options at high character levels as they do at low ones, which is a poor fit for how DnD is expected to work: high-level characters should get better, not just more, abilities. The problem with a width-first system is that players make a lot of class-selection choices early in their career and none later on.

So, what I'd like to propose is a hybrid system that grows in both width and depth simultaneously in order to mitigate both of these problems. In order to give players a single capstone ability, it'll need to be a "peaked" system where players can only have one class of their maximum class level.


The maximum level that a character can have in their highest-level class is the number of classes that they have.
The maximum level that they can have in their second-highest-level class is one less than the number of classes they have, the third-highest-highest level may be two less, and so on.
Characters normally increase their level in an existing class when they level up. They may select a new class whenever either they cannot gain a level in an existing class without exceeding the maximums set above, or some dramatic event occurs in-game which would grant them membership in that new class.


Under this system, characters could be:


1st level in their highest-level class at character level 1.
2nd level in their highest-level class at character level 3.
3rd level in their highest-level class at character level 6.
4th level in their highest-level class at character level 10.
5th level in their highest-level class at character level 15.
6th level in their highest-level class at character level 21.


That makes level 21 a nice cutoff point for the game. It's higher than level 20, but a game that's based on 5e with a fundamentally different leveling system doesn't have to have exactly the same level cap in order to work. Under this system, each class should have 6 levels.

Characters won't reach these "one class of level A at character level B" milestones if they have too many classes. Six classes is the maximum number that they can have without spoiling their progression by the time they reach level 21 - so, six should be the absolute limit on how many classes a character can have regardless of what happens to them.

This gives players much more freedom to redefine their characters as the game progresses than normal 5e DnD. In normal 5e, the only way to reach the capstone for any class is to select it at level 1 and then keep selecting it at every lever thereafter. Under this system, a character who has yet to attain 16th level could theoretically gain the 6th level capstone of any class in the game.

Amechra
2021-06-14, 06:03 PM
One potential alternative way to look at scaling would be to let people skip levels that gave them features that they already have.

For example, let's say we have four classes, with some kinda-arbitrarily-chosen class feature names:

Cleric
1. Miracle Worker
2. Channel Divinity
3. Voice of the Divine
4. Greater Miracles

Fighter
1. Extra Attack
2. Indomitable
3. Battle Instincts
4. Martial Excellence

Paladin
1. Holy Warrior
2. Channel Divinity
3. Aura of Purity
4. Extra Attack

Wizard
1. Arcane Secrets
2. Master of Rituals
3. Scholar of the the Unseen
4. Deeper Secrets

Let's say you have a 4th level Paladin. If they took a level in Fighter, they'd skip directly to gaining Indomitable (since they already had Extra Attack). Similarly, if they took two levels in Cleric, they'd get both Miracle Worker and Voice of the Divine, thanks to already having Channel Divinity. If they tried to level up as a Wizard, however, they'd have to take each level in order. If you went Cleric + Paladin + Fighter, you'd end up completing all three classes by level 10... but you'd still only have ten features.

...

This idea would probably work better with longer, more granular classes, so you could better control the amount of overlap.