Balancing Homebrew Classes
I get a lot of questions about homebrewing so I decided to make a big ole list of things to look at when making a subclass or even a class for D&D 5e. I’ve learned a lot from making my own brews and getting critique from others and reworking my own work, and I think everyone can benefit from these. As you are going through your finished homebrew, check for these:
image credit: Randy Gallegos
Players get several different types of actions each turn: movement, an item interaction, a bonus action, a reaction, and an action. There are several potential problems when it comes to assigning abilities to these different types of actions. On one hand, you could put too many options on one action type. If every feature in a subclass costs one action, the player has potential bonus actions and reactions it isn’t using. Also, if this is a subclass, be aware that you are competing with other features of the base class.
A rogue subclass that has a ton of bonus action features won’t be able to use them if it’s using its Cunning Action every turn. The rogue also has its Uncanny Dodge taking up its reactions. This leaves very few actions to deal with as far as a subclass goes, which is why the book’s subclasses either offer passive features or spells (for the trickster).
The competition for actions can also be a good thing, but try to limit it to two distinct options. Players shouldn’t have to choose between a hundred options each round when you could give them two choices. In the case of the rogue, “do I use my reaction for this attack of opportunity, or wait in case I need Uncanny Dodge later?” Two very different options for one action that can lead to very different outcomes.
Keep an eye on how many ability scores your class cares about. A good class really should only care about two of them. If you introduce a third one, make sure it’s tertiary and does not require as much effort to make it good, yet not breakable if the player ramps up to 20 in that stat. For instance, my Commander fighter subclass during fighter week was really reliant on CHA for a class that needs STR and CON to be good. When this was pointed out to me, I have since reworked the class entirely to not need saving throws, and instead gains a number of uses based on WIS plus a flat number. This means that even a fighter with no WIS still gets to use the abilities, but rewards the player for having a few extra points. Since the features no longer have saves, the player isn’t punished for having a low tertiary score.
Compare Similar Mechanics
The most basic way to check your classes is to compare it to other existing mechanics, but some people forget to do this. There are a few classes I made that essentially gave other classes a ranger’s animal companion. To do that, I made sure to look at the ranger’s companion feature when making those classes.
Compare the damage output against the damage of similar classes. What matters most is damage per round, which you should usually assume takes place over 1 minute. You can use this link to calculate damage per round for fighting classes, as compiled by Kryx. There are some parts of the tables I still don’t understand, but in general this is very helpful.
https://docs.google.com/spreadsheets/d/1d-9xDdath8kX_v7Rpts9JFIJwIG3X0-dDUtfax14NT0/edit#gid=2091322934
It doesn’t contain spellcasters (except for the bladelock) but I have a quick way I judge a caster’s damage output. A spellcaster ideally will be frontloading its most powerful spells with some cantrips between. Assume that the class in question casts its highest level damaging spell once followed by a cantrip the next round, then the next-highest level damaging spell the next turn, then a cantrip, etc. until 1 minute (10 rounds) passes. Use the average damage for each spell and cantrip. Add all those averages together and divide by 10 to get the average damage that the character did over that minute. Overall, spellcasters tend to have a higher average damage than fighting classes, but they become balanced thanks to their limited number of slots each day. Simply because a creature has access to Cone of Cold doesn’t mean it can cast it every single round. Bear that in mind and compare to existing classes.
Honestly, I dislike drawbacks in homebrews unless it’s for a thematic reason. There shouldn’t be a downside to something to gain an upside that is far beyond what another similar class could accomplish. Instead, that upside should simply be balanced appropriately for the situation. Drawbacks can always be negated somehow, so make sure they either can’t be worked around or ensure the effect is fully balanced. For instance, paying health can simply mean spending hit dice during a rest unless you also reduce their hit point maximum.
I was making a class that gave offerings for magical power, so I had it apply effects similar to a sorcerer’s metamagic to their spells in return for a sacrifice of gold each time. However, a player’s wealth very much depends on the DM. I ended up using a chart based on a character’s average wealth as they leveled based on treasure hoards to help choose how much gold it should cost. In the end, I decided that it should probably cost no more than a typical Potion of Healing but the effect applied to the spell shouldn’t be too much more powerful than a sorcerer’s abilities. Even if the character did have limitless wealth to use the effect on every spell, it shouldn’t be enough to break the game.
image credit: Sandro Rybak
Make sure your wording is detailed and specific, and yet short and succinct. Keep an eye out for sentences that could be interpreted in a different way. Ensure the antecedent is clear. I recently noticed a spell I made referred to “creatures on the side of the wall you designate.” I meant for the caster to choose a side of the wall, not choose which creatures get affected, but it could have been interpreted both ways. Careful wording helps prevent players from absolutely busting your homebrew.
Being concise also helps overly wordy things become easier to understand. I guarantee that few players will actually read through your entire homebrew if it’s too wordy and doesn’t get the point across quick. In my future edits of my classes, I have since removed a lot of the different options that were available for things like the Xammux wizard from my wizard week, which wanted to be a hundred different classes complete with evil flesh grafting rules. Distill your class down to its essence.
What also doesn’t hurt is summarizing a feature with a sentence before defining the rules, cluing players into why the rules are that way. “A bright streak flashes from your pointing finger to a point you choose within range and then blossoms with a low roar into an explosion of flame.” Okay so it deals an area of fire damage. Got it.
Try to use similar terminology that Wizards uses in their classes. I still sometimes use terminology from 3rd edition because that’s what I grew up with. For instance I will sometimes say Handle Animal instead of Animal Handling when referring to the specific skill. Just as well, I often refer to skill checks when they are technically ability checks that just happen to use a skill. I should be saying Strength (Athletics) check, not just an Athletics check. Even though it’s easier to say out loud or in-game, and on Tumblr posts I will often do so to make things easier for myself, it’s not technically correct. Look at similar existing spells and class features to figure out your wording.
image credit: James Ryman
It is difficult to always run numbers and playtest if you are working mostly by yourself (like me), but there are some intense examples you can test your class against to see if it can be easily integrated into the game. Imagine each of the following for your class:
Imagine the whole party was composed of this class. Would its features still be balanced or is there a way to break them? Has the damage become absurd or awful?
Imagine the class in the ideal party, with a cleric, fighter, rogue, and wizard. Have your class replace any one of them and see if the party can still function. If it can’t, the class might still need balancing to be useful.
Imagine someone is multiclassing into this class after 5 levels of Fighter, or some other class that grants an Extra Attack by then. Basically, does your class become unbalanced if you give them an extra attack? Keep in mind that 5 levels is still an investment, so it usually won’t be a problem, but if the class has an extremely powerful ability that gets even MORE powerful if you can attack more than once in a turn, it might need balancing.
Imagine another class multiclasses by dipping a few levels into your class. While you usually want the class’s earliest ability to be their “schtick” that they use very often, you have to make sure it’s not so powerful that other classes can become overpowered by taking only 3 or so levels in your class. For instance, during Sorcerer Week I made a swarm hivemind subclass that essentially became a swarm, granting it resistance to physical damage. It made sense flavorfully, but someone pointed out that if a Fighter or other combat class dipped into it, they would suddenly have the resilience of a barbarian for almost no cost, in addition to all those spells. I tried fixing it and left it up but in the book compilation I’m making I actually ended up removing it to replace it with a new class.
Try thinking of other ways to test the limits of the homebrew. Chances are someone out there will want to be able to min-max your class so you have to be ready for anything.
While it’s unfeasible to playtest every single thing you make in large focus groups with detailed data collection, especially if you are by yourself and make a bunch of it like I do, you can always run through a self-playtest. All you have to do is play the class or mechanic in question in turn order against enemies of an appropriate challenge rating. Use turn order and run through the mechanic precisely as you wrote it. Check what happens if a check is failed OR succeeded to cover all your bases.
For the spells that I make, I will imagine the entire process of the spell from casting to effect (usually more for spells with a duration). Use props for minis if it’s easier. This helps the distinctions for “start of turn” and “end of turn” effects become more clear. For instance, if I have a disabling spell that requires an initial save and then give the creature a new save at the start of each of their turns, the creature might fail their initial save but then pass their start-of-turn save, meaning they will never have a single round inhibited by the spell! I might get rid of the initial save or change the recurring saves to happen at the end of the creature’s turns, so the creature will at least lose one turn on a failed save.
prguitarman
ellievsbear
snowghoul
stormborns
titsay
jbbartram-illu
yousharknotpass
