Sorcerer Subclass Design

Debuting with D&D 3rd edition back in 2000, the sorcerer was originally designed as an alternative to the wizard. Where wizards could choose their spells each day, the sorcerer had a narrower, fixed set of spells. The sorcerer also gained more spells slots than a wizard but gained higher level ones later than the wizard. Finally, sorcerers also had a casting mechanic like the 5e core rules. Wizards had to prepare specific spells for each slot. Sorcerers picked which spell they wanted to use when they expended a slot.

Those differences made for an interesting change of pace, especially at low levels where the sorcerer’s greater number of slots let them sling more sleep spells or magic missiles.

With 5th edition adopting the sorcerer’s core flexibility, the class has lost some of its unique flavor. Sorcery points and metamagic, another mechanic lifted from 3rd edition, try to emphasize the sorcerer’s core flavor, but they don’t quite do enough to make the class feel distinct.

Subclass design gives us the chance to find ways to make the sorcerer stand out. In addition to showing the basic pattern of design for this class, I am also going to talk about my approach to it as an example.

Sorcerous Subclasses

The sorcerer gains subclass features at 1st, 6th, 14th, and 18th level. Keep in mind that the 2024 revision to the rules is going to move all subclasses to 3rd level, so we’ll eventually have to think about what that means for that first subclass feature.

Since the sorcerer is a full caster, we don’t have a lot of power to work with in crafting features. However, since the sorcerer is such a blank slate – its class features are largely the scaffolding for its spellcasting and metamagic – we have a lot of room for fun, thematic pieces.

At 1st level, the subclass feature should establish the sorcerer’s core identity. The draconic sorcerer gains more hit points and a better AC, allowing them to mix it up closer to their enemies compared to wizards or other sorcerers. The feature at this level should be relatively power neutral unless it requires a resource. Avoid flat bonuses to damage without tying them to a resource spend.

At 6th level, you have the room to include a strict power up. The draconic sorcerer adds Charisma bonus to a damage roll. That’s a nice comparison to use for your own design. Remember that the bonus applies to any spell, including cantrips, giving you space to add a limited effect. A feature usable three times per long rest could grant 3d6 additional damage per use.

The 14th level subclass feature grants at-will utility, flight in the case of the draconic sorcerer. Consider matching that with an effect equal to a 3rd-level spell (or lower) that the sorcerer can use at will.

Finally, at 18th level the sorcerer gains a powerful, limited effect, equal to about a single use of a 5th-level spell.

As you can see, the sorcerer doesn’t have a massive amount of power to work with, but there are some opportunities to add in some flavorful elements.

The Sorcerer’s Identity

I really love the idea of the sorcerer as an innate caster, someone who has access to a font of power that wells up from within them. It provides a nice worldbuilding alternative to the wizard and warlock. It also suggests some interesting stories and options for building your character’s background.

A good sorcerer subclass speaks to that origin and gives a distinct flavor to the sorcerer, one that makes them feel dramatically different from other casters.

I like the idea of the subclass granting a list of bonus spells known. That list of spells expands the sorcerer’s options while also giving you a mechanical hook to tie your features into. You can design class features that activate when the sorcerer casts those specific, bonus spells, giving you the chance to create a signature mechanic that the sorcerer can access on a consistent basis.

Moreover, this approach appeals to me because it carries echoes of the sorcerer’s metamagic abilities. When the sorcerer casts a spell, they can add an additional flourish to it based on their ability to weave magic with an intuitive, creative edge compared to other casters.

Mechanically, the class feature that grants the spell needs to give them a clear label. Other mechanics can then refer to that label and specify what happens when the sorcerer casts one of those spells.

This approach requires a subclass feature at 1st level that is stronger than those typically found in this class. If you scale the mechanic based on the spell slot used for casting, you can ensure that the benefit is small and balanced at low levels but scales upward to remain useful at high ones.

Sorcerer Spells

In working on 5th edition, I think it’s interesting that we created distinct spells for the warlock but didn’t make it a priority for the sorcerer. I think that underscores the sorcerer’s lack of identity. It was easy to think of flavor for the warlock. You simply need to imagine the entity that might form a pact with a warlock, then create a spell that reflects its nature.

For the sorcerer, the path is far cloudier. The game already has dragon-ish spells in burning hands or acid arrow. The other PHB option for the sorcerer, the wild mage, is largely defined by how it casts wizard spells, rather than by distinct spells it uses.

If a sorcerer subclass grants additional spells, it’s easy to think of populating that list with one or more new spells. That still gives us the key question: What is a sorcerer spell?

For my work here, I hit upon the idea of stealing a concept from Magic: the Gathering called modal spell. A modal spell allows you to pick its effect from a short list when you cast it. They grant flexibility in play, letting you cover a few situations with a single card or accounting for an edge case your deck might face, like battling an artifact-heavy deck, without necessarily sacrificing core utility. A card with the option to damage a creature or destroy an artifact is a great, flexible tool in an aggressive burn deck.

Since sorcerers are natural casting talents, modal spells fit right in with them. I think of them as masters of improvisation, able to shift their spells to fit the challenge at hand. A modal spell does add complexity, so consider offering at most three options and preferably two. Beyond that, the spell likely becomes too complex for most players to make easy use of them in play.

Here’s an example of a modal spell, designed for a sorcerer who derives their power from radiant energy. While this spell offers three options, I think they are far enough apart from each other that they offer a relatively easy choice for a player.

As an added note, while healing is not normally associated with arcane casters I prefer to focus that restriction on the wizard. Spells are balanced by level, rather than the class that has access to them (exception, a few iconic spells like fireball that are placed in a separate tier). Giving a sorcerer healing is another way to make them distinct from warlocks and wizards.

Mystic Lantern

2nd-level conjuration

Casting Time: 1 action

Range: 60 feet

Components: V, S

Duration: Concentration, up to 1 minute

A 1-foot diameter lantern of radiance appears in an unoccupied space of your choice within range. The lantern can be positioned in mid-air. It sheds bright light in a 20-foot radius and dim light for an additional 20 feet. Once during your turn, if you can see the lantern, you can command it to fly up to 30 feet. It can stop in midair, cannot be affected by physical or magical effects, and does not block movement or provide cover.

Once on each of your turns, including the turn you cast this spell, you can pick one of the following effects:

Baleful Glow. Choose a creature within the lantern’s area of bright or dim light. It must make a Constitution saving throw. The creature takes 2d8 radiant damage on a failed save, or half as much on a successful one.

Revelatory Light. Choose a creature within the lantern’s area of bright or dim light. Until the start of your next turn, attacks against it have advantage and the creature gains no benefits from cover, lightly obscured areas, or heavily obscured areas.

Soothing Radiance. Choose a creature within the lantern’s area of bright or dim light. It has advantage on all saving throws until the start of your next turn.

At Higher Levels. When you cast this spell using a spell slot of 3rd level or higher, the damage for Baleful Glow increases by 1d8 for each slot above 2nd.

Bonus Topic: Spell Slots

I’m not sure I’ve met a game designer who thinks that spell slots are a good mechanic. Most video games use a mana bar, a simple measure of the fuel for your abilities. Spell slots introduce nine separate little fuel tanks for your spells, each of which has their own cost. Some spells also allow you to use higher level slots for a more powerful effect.

After the release of Baldur’s Gate 3, I saw several articles about how BG 3 was the first time that even veteran D&D players understood the game’s magic system. That’s not a great look.

So, why does D&D stick with spell slots?

I think the answer lies in the game’s focus on low levels. With most campaigns ending at around 5th level, the mechanic isn’t quite so burdensome. At that level, a player has three levels of slots to pick from. The power difference between each level is vast. Compare fireball to web or magic missile.

At this level, a player manages the difference between a 3rd level spell and a 1st level one. They bust out the big spells in the face of danger and use their lower-level ones more freely. In a pure mana point-based system, the player needs to navigate the tradeoff between using lots of weaker spells, or a fewer number of stronger ones.

I think that choice is much harder to navigate than we veteran players estimate. It requires a player to understand the future ramifications of a choice, a situation that results in analysis paralysis in my experience. With slots, a player can comfortably hoard their highest-level slots for a boss fight.

With all that said, these benefits evaporate at higher levels. The delta between spells becomes much muddier, the variety of slots overwhelming, and the clarity of which spell is your best much less clear.

Were I designing the game from scratch, I’d avoid an abstract spell slot or point system entirely. I’d have spells each recharge on their own schedule. A caster might have 10 to 15 spells, with stronger ones requiring a long rest to regain and weaker ones at-will or returning with a short rest. While that approach lacks flexibility, I think it would be more playable for non-D&D veterans.