I intended to have this post closer to the original one on the Warlock Problem, but I really wanted to get that post uploaded on the correct day. Such are the hazards of linear time.
It’s not a secret that I prefer having freedom in my character design possibilities. I’m of the opinion that a system should allow a player to design and run a character of their choice (within the limitations of the system and the narrative—if you’re trying to play a space smuggler in D&D, that might be interesting but you can’t be surprised when it doesn’t work). This role-playing game should allow players to decide their own role, not begrudgingly adopt one of the very few roles the game’s authors wanted the players to want. That means players should have some sort of options, so they can select the traits or skills or powers closest to their own vision of their character.
D&D doesn’t have a freeform character creation system where you can slap on any combination of traits as long as the math checks out. It uses classes. That’s not an issue, as long as the classes still allow some freedom of choice. If I’m going to be a fighter, I don’t just want to be the same fighter everybody else is. I want to put my own spin on it. It’s fine if all fighters get certain core abilities, even if they don’t fit the exact character I have in mind, but the fighter needs to gives me some sort of wiggle room so I can make it work for me instead of the other way around.
I’ve been using the term “points of divergence” for this concept. A point of divergence is a place within the character design process (or really, the design process of anything, but we’re talking about characters here) that allows a player to make a choice about what the character is. They are the points at which a character is allow to diverge from the One True Build, points where a player can put their own stamp on a character instead of just following along a single, hard-coded path laid out for them by the authors. The actual design process is full of points of divergence; players can pick their character’s name, race, class, skills, feats, height, sex, and so on. But specific steps of this process don’t necessarily allow the same freedom, intentionally.
Consider background. A player can choose from several backgrounds for her character. If she picks a sage, that sage will always gain proficiency in Arcana and History, gain the Researcher ability, and add a letter from a dead colleague to her starting equipment. She can, however, select two languages of her choice and select from lists with several scholarly specialties and personality characteristics (or more, if she is allowed to pick something not in the book). A background is a point of divergence, because the players has options. The background’s skill proficiencies are not a point of divergence, the languages are, and the equipment is not.
For the purposes of D&D, a point of divergence must have a mechanical effect. A character’s name is, technically, a point of divergence because the player can make it whatever she wants. But the system doesn’t care about that. The option to choose your own name isn’t a deliberate freedom intended by the designers, it’s an outright expectation. The same goes for height, sex, clothing, even weapons. This is why reskinning doesn’t factor into points of divergence. For example, a monk has proficiency with short swords. If a dwarven monk decides to reskin a short sword as a light war pick, that’s not actually a freedom granted by the system. That’s a freedom granted by the DM, and the system doesn’t get credit for it.
We should also note that not all points of divergence have the same weight. Some points simply give players more options, or more important options mechanically, than others. A ranger can choose the form of his animal companion, but that’s not as important as the choice to have an animal companion in the first place. A paladin can choose her spells, but they don’t have the same impact as when a wizard chooses hers. When we consider a point of divergence, we have to see how much freedom it actually gives.
Sometimes “freedom” is a nebulous concept as it applies to class design. All clerics gain the spellcasting feature. There is absolutely no mechanical variation in this, and a cleric cannot avoid or modify it in any way. But because a cleric can change her prepared spells every day, this enables a great deal of mechanical variation during play. An individual cleric can choose to prepare nothing but support and healing spells, or nothing but spells that effect enemies, or anything but cure wounds, or only spells that contain the letter S. This non-optional feature enables further character options, so even though it’s hard-coded, we consider it a major point of divergence because it enables so many characters. Essentially, we have to consider how much flexibility a feature gives us, not necessarily in round-by-round actions but in what characters it lets us make.
With this in mind, let’s consider some of the base classes to see what points of divergence they have:
Paladin
No divergence: divine sense, lay on hands, divine smite, divine health, extra attack, aura of protection, aura of courage, improved divine smite, cleansing touch
Minor divergence: fighting style (there are a few options, but this choice doesn’t mechanically limit the character in any way and only rarely provides a new benefit, so it’s less a character feature and more a declaration of mechanical intent), spells (the spell list is small and the number of available slots is low)
Medium divergence: sacred oath (each oath hard-codes every spell and feature, but they do meaningfully impact the character’s options and impact)
Major divergence: none
Cleric
No divergence: turn undead, destroy undead, divine intervention
Minor divergence: none
Medium divergence: domain (seven options, more than anybody except the wizard, and each domain differs significantly from the others)
Major divergence: spells (a big spell list with plenty of variation)
Fighter.
No divergence: second wind, action surge, extra track, indomitable
Minor divergence: fighting style (more options than the paladin, but it still doesn’t change the character that much), ability score improvement (an honorable mention here because the fighter gets more improvements than other classes, and they can optionally be used for feats)
Medium divergence: martial archetype
Major divergence: none
Note that we’re not getting into whether the subclasses themselves offer points of divergence. That’s a valuable consideration, but watching me document them isn’t a terribly interesting read.
There’s a running theme here: classes with good spellcasting diverge greatly, and classes without spellcasting tend to have no features with remotely that much divergence. All else equal, a spellcasting class just gives players more options than a non-spellcasting class. But most spellcasting classes diverge primarily because of their spellcasting, leaving few options in other features. I interpret this as a pretty significant problem with the class design of D&D, that uniqueness only occurs as a result of magical ability. And speaking of which:
Warlock
No divergence: eldritch master (this is the capstone, L20 feature, and it’s the first time the warlock doesn’t get to make a decision)
Minor divergence: pact magic (very narrow spell list, even with bonus spells from a pact), mystic arcanum (even narrower spell list)
Medium divergence: otherworldly patron (additional variation in a limited spell list, new features, and access to invocations), pact boon (like fighting style, this is more about declaring a mechanical intent than making a real choice, but it gets bumped up because it also gives the player access to invocations)
Major divergence: eldritch invocations (they do so many different things and vary so much based on other class features, it’s not an exaggeration to say they’re the most important part of the class)
And this is why I gush about the warlock. Almost every class feature is a point of divergence. More so, everything feeds off of each other; a warlock picks a pact boon, the pact boon synergies with the patron, the patron opens up certain invocations, the invocations add to the spell list, the spells affect how the warlock uses the pact boon. A cleric can choose to make herself interesting by artificially restricting her spell list, but for a warlock restrictions like that are built in, and they inform every other decision the player makes. This is the sort of class I want: plenty of divergence that meaningfully impacts the character build, manifests in the character’s round-by-round actions, and produces characters so varied that it’s not always clear what their original class even is.
Acknowledging the value of points of divergence had been a valuable tool for me in describing what I like and don’t like about classes in 5E. And because I’m physically incapable of leaving well enough alone, it’s also valuable as I consider the existing classes and feel the call to write a new one.