Sorcerers, Warlocks, and Bards

Over the past few weeks, Mike Mearls has been giving teasing previews of a few classes, revealing how they’re being updated to Dungeons & Dragons 5. He’s previewed the sorcerer, the warlock, and the bard. The first two were first previously updated in the August 2012 playtest package and the bard made its debut in the last playtest package.

It’s a bit unfortunate that the playtest has ended and WotC feels it cannot release these classes, even in a playtest document just containing those three or a preview article giving a look at the actual mechanics.


The sorcerer is a tricky class as it’s not really a class but a background, an origin for how an arcane spellcaster gained their magic – through birth opposed to study. The origin of the sorcerer as a class is equally bland. During the creation of 3rd Edition it was noted that a large percentage of the book was taken up by spells usable only by a single class: the wizard. The cleric shared spells with the druid and paladin, but only the bard could cast wizard spells, and those were limited. So the sorcerer was born, a wizard with a slightly different method of casting spells (and no bonus feats). In 4e the sorcerer was equally unremarkable, being the arcane striker it was the wizard that did damage.

What made the sorcerer different in earlier editions doesn’t work in D&D5 as the default spellcasting is already close to 3e sorcerous spellcasting. In addition, given the edition is planned to be modular there will likely be several variants of spellcasting, so the DM (or player) can opt to use spell points or classic Vancian magic if they wish. So defining the sorcerer by how they cast spells seems silly when how they cast spells will not remain constant.

This was a big reason the sorcerer and wizard were planned to be folded together into the “mage” class, which was rejected by the fans, mostly because WotC never released a sorcery variant to show how it would work (and because WotC also tried to squish the warlock into the mage, which would have been harder to pull off).

D&D5’s first attempt at a sorcerer was “meh”. It was designed as a gish, a warrior-mage who mixed weapons with magic. But the execution was flawed: the class was more than a little overpowered and it didn’t allow existing characters to be updated into D&D5; established sorcerers in both fiction and games didn’t fit the paradigm of the new martial sorcerer.

It doesn’t help that D&D5’s wizards haven’t had many unique mechanics or abilities that evoke a scholarly, academic spellcaster. The most recent version of the class just gets ability increases/feats and bonuses from its tradition (which are few and far between). Each individual wizardry tradition might as well be its own class for the amount the “wizard” influences a character; there’s very little “wizard” in the “mage” class, and very few uniquely wizardly powers. Which will make it harder to design a sorcerer that can do unique things without being more powerful than the mage. (Ironically, this is the opposite of 3e, where wizards got bonus feats and sorcerers gained nothing but new spells.)

Looking at what they’ve designed, the sorcerer casts spells like the wizard but has an additional “sorcery point” system that can be used to cast additional spells or augment magic. The sorcerer has a number of points it can spend to augment spells increasing the range and damage of spells, or gaining additional spells. And there is another variant that merges the wild mage with sorcerer, like in 4e. Which sounds cool and the wild mage very much fits the tone of the sorcerer, but I hope it still uses the same sorcery points or there’ll be very little unifying features to the sorcerer; the sorcerer’s subclasses shouldn’t make it feel like two different classes that just use the same spellcasting system.

The idea of modifying spells with points is a good idea but more suited to psionics than the sorcerer. By introducing it to the sorcerer now it will be that much harder to make a great psion class. But they need something for the sorcerer now, so making the psion is future-WotC’s problem

The wizard is supposedly balanced through “flexibility” but I hope the wizard has additional class features added. This implies an approach similar to 3e, where wizards know more spells stored in their spellbook while sorcerers will know fewer numbers of spells. Which a balance for the spellcasting system but not sorcery points. And from experience, wizards tend to settle on a handful of go-to spells and only change their spellcasting when the situation or environment changes dramatically, so it’s not *much* of a difference. And, by default, wizards can only learn a certain number of spells each level, and have to expend a painful amount of treasure to learn more, which is also dependant on the number of other wizards in the world.


Like the sorcerer, the warlock is less of a class and more of an origin. It’s very possible to imagine an otherworldly pact being used to gain the abilities of a wizard – or even druidic magic from a pact with nature spirits. It’s even possible for a campaign setting to attribute all magic and wizardry to extraplanar deals. But the warlock also has some unique mechanics that are very different from the wizard, it has a legacy of being able to cast spells at-will rather than having daily resources, as well as some unique spells. However, the mechanics of the warlock don’t really fit the background of the class, they neither mesh with the portrayal of warlocks elsewhere in fantasy fiction nor emulate witches & warlocks in literature or folklore.

This makes the warlock less of a generic class and more a D&D-specific class, something unique to its campaign settings and less a typical fantasy archetype.

The first draft of the D&D5 warlock didn’t seem too bad, it wasn’t nearly as off base as some of the other first attempts at classes. I imagine the reason it vanished was because of the “make it a mage” plan, or because there was only a single option previewed, which was a non-standard version of the class. Now they’re reworking the warlock again.

It’s really, really, really hard to get a feel for the warlock based on Mearl’s article. The warlock gets fewer spells but always casts them at their maximum level, which is fine but might make it hard to adapt to use spellpoints or other variant spellcasting systems. And apparently the warlock also regains spells each short rest.

The warlock’s signature mechanic is its eldritch invocations, which are described as super-cantrips and rituals. This really just sound like warlock-only bonus spells. But this matches the design of the 3e warlock, so it’s updating and reworking a signature mechanic, so it’s harder to fault that design.

The warlock also seems to have two points of customization, compared to most classes’ one. They pick their pact (blade, chain, or tome) and their patron (likely fey, infernal, demonic, star, and elemental). I like this as it’s not forcing the different styles of warlock (hexblade, summoner, or straight magic user) into the choice of patron. It’s not that different from martial classes choosing a weapon specialization in addition to their subclass.

The warlock sounds good on paper, but it’s really hard to judge the execution without seeing it. For example, eldritch blast was a signature spell/power in both 3e and 4e, so it makes sense to make that a class feature or an always prepared spell rather than a spell that can be chosen.


The bard has been pretty inconsistant over the years. In 1e it was a druid/fighter/thief while in 2e and 3e it was a thief that could cast some arcane spells. In all three editions it could also inspire allies through music. In 4e it was a leader that could focus on either ranged or melee spell attacks. It could heal as well as any other leader, but had a handful of music related rituals and small features.

The bard is arguably one of the harder classes. It’s a Jack of All Trades class being proficient in many different things but excelling in none, which doesn’t work well in D&D, which is a game of optimization. It covers a lot of different roles, potentially replacing the fighter as a melee combatant, the rogue as lockpicker or skill monkey, the wizard as a utility magic user or sage, and the cleric as the healer or party face. And during 3e and 4e it was possible to build a bard as a ranged combatant. So the bard could try and fill almost any role (except tank). The bard was often the best fifth-man in the party: because it wasn’t a specialist, it didn’t step on the toes of other characters and could assist everyone else in the party. While the bard had fewer moments to shine and take the spotlight, it was always useful.

The first draft of the D&D5 bard was a little mixed, being equal parts warlord and bard. It also fell apart at higher levels, lacking uniquely bardic abilities and having a lot of filler features. The choice of builds didn’t help as there was the fascinating bard and, rather than something else traditionally bardic, there was the College of Valour bard that was a renamed warlord.

In response to difficulty finding a place for the bard, WotC is making the class a full spellcaster. Magic has always been just one aspect of the bard, with the class previously been portrayed as “picking up a little magic on the road”, so making it a full spellcaster seems odd. Instead of being the Jack of all Trades, this means the bard is instead a spellcaster that can focus in other directions.

Being a full spellcaster also means the bard will have fewer special abilities (read: abilities unique to the bard), as there will be more levels it just gets spells. The wizard has seven dead-ish levels where it only gains a new spell level, while the cleric and druid have six. The bard might be doubling the number of levels it’s only new class feature is “more spells”. This also means it’s existing powers have to be significantly less powerful, to avoid making the bard more potent than either the sorcerer or the wizard, which don’t get a lot of other powers in addition to spells, especially at first level.

Mearls actually summarizes the change in design pretty well:

In our design work on D&D Next, we took a sledgehammer to a few of the bard’s traditional concepts in order to give the class a clear, unique place in the game.

Or, in other words, to make the bard fit the game they’re making it less of a bard. I… can’t agree with this design. Every other class was allowed to be itself. They didn’t make the fighter less fighty or wizard less wizardy to make them conform to the system. Why should the bard have to compromise? This all seems a little needless, as bounded accuracy means the bard’s master-of-none design should work well in this edition. It should have been easier than ever to balance the bard’s spellcasting with melee combat. It also missed the point that some players might want a class that’s not a specialist and more of a generalist, or be happy not being in the spotlight.

Bardic Inspiration seems to be bard’s new signature mechanic, which is replacing the more traditional bardic music. This is to accommodate non-musical bard, such as ones giving speeches. Again we see some warlord creeping into the bard. I’m fine with the warlord finding a place in the bard, but when you’re changing the core assumptions and presentation of the class to make room for a side concept… that makes me less happy. (Especially since the bard is suddenly a primarily a spellcaster, which doesn’t mesh with the warlord presentation.) The ability itself sounds subpar. It’s an impressive ability on paper, adding a d6 to any d20 roll a certain number of times each day, which is pretty much 3e’s Inspire Competence and Inspire Courage rolled into power. And in an edition with flat DCs, the straight bonus is very nice. But it’s only for a single roll, which means it could fail and end up doing nothing. And having this floating bonus a character can apply some time later is very gamist in execution.


The point of the playtest was to gauge the reaction of the fanbase to various classes. It took several attempts to really nail down the execution of a few classes, with the fighter and rogue especially undergoing numerous revisions and changes. There’s no reason to believe creating these three classes will be any smoother than other classes, or that feedback is any less needed. The bard is an especially tricky class, with an imperfect execution in most prior editions, as is the sorcerer that has to be different from the wizard but cannot just cast spells differently.

Even if the previews were designed to elicit feedback, without the surveys it will be harder to separate the opinion of the majority from a few vocal detractors. There wasn’t even polls in the articles.

It would be a shame, after all that work that went into the public playtest, to product some imperfect classes at the end.

I loved the 3e bard. I enjoyed playing this non-offensive support character that buffed her allies without throwing a punch. While a poorly designed bard will hurt D&D5 in my eyes, it’s not going to break the game. It just means I’ll have to multiclass with rogue or something to get a character I want to play, or find something else that draws my attention.