So, power budget.
Mike Mearls' YouTube videos had some explanation about how they budgeted power for classes when he was describing how much he could do with any given subclass. He didn't get into deep detail, but I recall him using the Eldritch Knight as an example, where he used the spell progression chart to define his budget for another experimental subclass. I don't remember exactly how he figured things, but I do know we can use the DMG's rules for spell creation for the same thing.
The DMG damage budget for a spell at a given level has an interesting correlation with the sorcerer's Font of Magic. The DMG lists a number of d10s for damage against a single target in the order of 2, 3, 5, 6, and 8 dice for levels 1, 2, 3, 4, and 5, respectively. Font of Magic allows you to create a spell slot of a particular level for a cost of 2, 3, 5, 6, and 7 sorcery points, matching the number of d10s for a given spell level for all except 5th level, where it's off by 1.
If we take the number of d10s for a spell slot as a budget metric, we can calculate the total power budget a spell table provides. We can then put the sorcery points available at a given level to gauge how much budget is being attributed to them. I'll also include Arcane Recovery to see how that compares.
| Spells Power | Sorcery Pts | SP Gain | Arcane Rcvy | AR Gain |
---|
1 | 4 | 0 | 0% | 2 | 50% |
2 | 6 | 2 | 33% | 2 | 33% |
3 | 14 | 3 | 21% | 3 | 21% |
4 | 17 | 4 | 24% | 4 | 24% |
5 | 27 | 5 | 19% | 5 | 19% |
6 | 32 | 6 | 19% | 5 | 16% |
7 | 38 | 7 | 18% | 7 | 18% |
8 | 44 | 8 | 18% | 7 | 16% |
9 | 58 | 9 | 16% | 8 | 14% |
10 | 66 | 10 | 15% | 8 | 12% |
11 | 76 | 11 | 14% | 10 | 13% |
12 | 76 | 12 | 16% | 10 | 13% |
13 | 87 | 13 | 15% | 12 | 14% |
14 | 87 | 14 | 16% | 12 | 14% |
15 | 99 | 15 | 15% | 13 | 13% |
16 | 99 | 16 | 16% | 13 | 13% |
17 | 114 | 17 | 15% | 15 | 13% |
18 | 122 | 18 | 15% | 15 | 12% |
19 | 132 | 19 | 14% | 16 | 12% |
20 | 143 | 20 | 14% | 16 | 11% |
Here we can see that, other than the first few levels where rounding bumps things up a bit, sorcery points essentially acts as a 15% increase to the budget of the class relative to the baseline of just being a full caster. Put another way, if you do nothing but spend sorcery points on creating new spell slots, you should get about 15% more total damage generated.
I've also included the calculated benefit budget of Arcane Recovery from the wizard class. This isn't the maximum possible effective value (which would involve taking nothing but 1st level spell slots), but it takes what seems to be reasonable high-level spell slots for best effect. As you can see, it's only slightly below the budget value of Font of Magic.
So what about metamagic? Well, metamagic isn't actually a budget item. It's merely a way to spend your budget. You could spend 2 sorcery points on a 1st level spell slot, or on Quickened Spell. You can do different things, but (assuming the metamagic costs are appropriate) you haven't actually increased your power per day.
We can thus conclude that the power budget for the wizard and the sorcerer at the class level is basically the same, though I won't try to compare the capstones.
We can also see that if you can spend a class resource to gain an effect, that feature is not actually contributing to the power budget of the class. A dragon sorcerer spending sorcery points on Draconic Presence is not adding to the budget of the class; it's spending the existing budget in a different way. Same for a barbarian spending a use of Rage to gain an ability, or a paladin or cleric using Channel Divinity for an effect.
A Battle Master gaining superiority dice
is adding to the budget of the (sub)class, since he's adding a new resource. But once added, the maneuver list is budget neutral.
Bonus Spells
So this now gets to the idea of bonus spells, and indeed spell preparation in general. Aside from being able to select a spell of appropriate level, you do not actually gain more power with a greater number of spells. From a pure budgetary standpoint, 9 spells (one for each spell level) and 90 spells are the same thing, because all spells of a given level "should" have the same effective power. Since your main limitation is how many spells you can cast per day, any variation within that selection is purely arbitrary.
A wizard with the bare minimum number of spells known (44 at level 20 in the PHB version) has the same "power" as one who knows every possible spell.
On the one hand, this means that the PHB sorcerer's 15 spell limit was not a power constraint, but on the other hand it means that granting bonus spells is not actually a power buff.
From this we can conclude that the subclasses with bonus spells cannot be considered as having traded their budget for some other features for the budget for the bonus spells, because the bonus spells are not actually budget items.
The number of spells available to a caster is not built on its budget, but on the intended flexibility of the class, which is an entirely separate metric. Although in the case of the sorcerer, it's generally considered less about flexibility and more about matching the theme of the class. An ice sorcerer with only two ice spells doesn't "feel" like an ice sorcerer. You want additional ice spells, not directly for power, but to fill out the flavor you're aiming for.
Overall, I'd assess bonus spells as a "ribbon" tier feature, as far as class budget is concerned.
UA7
UA7 added a few features which actually affect the power budget.
Innate Sorcery: This increases the strength of spells for one minute. I can only estimate its expected budget value indirectly, via Sorcery Incarnate. If SI judges 2 sorcery points sufficient for an extra use of this ability, I have to consider that it's assumed value per use. Two uses is 4 sorcery points. That's a high budget increase at low levels, and a low budget increase at high levels.
Sorcerous Recovery: This is about a 2.5% bonus to the power budget every time it's used. I don't expect it to be typically used more than twice per day, and more likely only once per day.
Sorcery Incarnate: The ability to gain more uses of Innate Sorcery via sorcery points should be net neutral on the budget. Likewise, being able to use more metamagic on a single spell simply redirects how the budget is being spent. Despite being useful, this does not actually change the power budget of the class.
Overall, we have a very small increase in power, as defined by the class budget, but also some additional flexibility due to Sorcery Incarnate.
Subclasses
We can evaluate the subclasses based on how their features impact the budget.
For Draconic Sorcerer:
- Draconic Resilience: Small budget increase, though it's hard to say how much, exactly.
- Dragon Speech: Flavorful, but ultimately ribbon-tier.
- Elemental Affinity: A budget item.
- Dragon Wings: A budget item.
- Draconic Presence: Budget neutral. It costs an existing resource to use.
For the Wild Magic Sorcerer:
- Wild Magic Surge: A budget item of indeterminate value.
- Tides of Chaos: A budget item.
- Bend Luck: Budget neutral. You have to spend a sorcery point to use it.
- Controlled Chaos: An improvement for rolling on the Wild Magic table, though I'm not sure I'd say it adds to the budget.
- Wild Bombardment: A high budget item with a long cooldown.
Aberrant Mind:
- Psionic Spells: ribbon
- Telepathic Speech: Budget item
- Psionic Sorcery: (Low?) Budget item
- Psychic Defenses: Budget item
- Revelation in Flesh: Budget neutral
- Warping Implosion: Budget item with budget-neutral re-use.
Clockwork Soul:
- Clockwork Magic: ribbon
- Restore Balance: Budget item
- Bastion of Law: Budget neutral
- Trance of Order: Budget item with budget-neutral re-use
- Clockwork Cavalcade: Budget item with budget-neutral re-use
While it varies per subclass, you can kind of get a feel for how much budget is being allowed for across the subclasses. If we recognize that the bonus spells are basically ribbon items, there's not much difference in the budgets of each.
This may also give a better idea of how to approach subclass design.