D&D 5E Fivethirtyeight Article About D&D Race and Class Combos

[Tony Vargas]

Of the four most popular classes, three are noncasters (meaning they don't have spellcasting as part of the base class, though they might have a caster subclass). Considering there are only four noncaster classes in the game, that's pretty substantial.

They're also the three classes that even have an option of not using magic, at all (Monk has one non-caster sub-class, but it's still using magic, as Ki is explicitly magical in 5e). I guess a further test would be sub-class: Are there a lot of Totem Barbarians, EKs, & ATs, or a lot of Champions, Berserkers, and Assassins?

I'm surprised to see Druid so underplayed. Too weak due to design-level over-reaction to CoDzilla in 3e, maybe?

That was the 4e Druid ("XOMG! That Druid! Let's cut it up into three pieces, so that no one Druid PC gets more than one Cool Thing to do, yeah, that'll learn 'em!"). The pendulum's swung back the other way.
No, it's just never been a popular concept. Too 'tree hugger,' maybe, or too obscure?

I mean, I didn't start playing Druids because I saw it and went "awesome! a class for the ancient-Celtic priest/judge/seer history knows next to nothing about!" No, I read "the Druid is what the ancient Celtic Druids might have become had they survived..." in the PH, got curious, and that grew into a fascination with Cletic mythology. It also turned out to be a pretty amazing class in 1e, in certain ways - some of which are (finally) back in 5e.

Long-term data from our own 1e games shows closer to a 40-25-15-15-5 split Fighter-Cleric-Thief-MU-Other.
Lanefan

Clerics were more obligatory in 1e, and armor a more decided advantage. Could have had something to do with the Cleric ranking.

 

[kenmarable]

This sort of thing always catches my interest...but what I'd really like to know is how much of that data is tainted by multiclassing. Anyone know if there's a version of this data that either strips out multiclass characters or gives a % of each total that comes from multis?

Not sure if someone else mentioned this, but at a rough estimate (if my off the top of my head math is right), there is about a 9% increase in the numbers due to multiclassing. There's about 109,000 per 100,000 characters. So at least [-]81[/-] 91% of characters are not multiclass.

(If there are only 2 class multiclassers, than it's [-]81[/-] 91/9, but with multiclassers with 3+ classes, it shifts even more. However, I'd imagine - completely anecdotally - that 3 classes are less common, and 4+ are very rare.)

The other thing to note - many of the totals are wrong and are off slightly. So these numbers are quite likely rounded (although some are off by 2 which makes less sense of rounding??). So when they say "per 100,000" that does seem accurate and there are likely at least an order of magnitude more PCs in the data and then rounded off here.

Edited: Fixing a brain fart caught by Dausuul. Thanks! *facepalm* Is it the weekend yet?

 

[Leatherhead]

Worth mentioning:

Among races available for free. Characters with multiple classes count once for each class.

I wonder how many of those human fighters were made just so that people could use them to argue about GWM/SS/PM
Partially a jab, but look at that insanely lop-sided the race distribution when compared to every other combination, and remember that variant humans get a feat at level one, which makes them ideal for both low level feat testing and circumventing the main penalty when multi-classing (which is not getting ASI on time)

More interestingly:

The Cleric has been knocked out of the "core 4".

Aasimar are the rarest of all characters, even more rare than birdmen. Though it doesn't specify if this is the DMG's or the Volo's Guide's take of the race.

 

[Caliban]

An article by Gus Wezerek on FiveThirtyEight looks at race and class combination in D&D, using data from D&D Beyond. Wezerek suggests a reason for the popularity of human fighters: "It lets you focus on creating a good story rather than spending time flipping through rulebooks to look up spells."

Or, alternatively - Variant Human Fighters get the greatest number of ASI's and Feats.

 

[iserith]

The second is that I tried to explain that there are those of us who really, genuinely, enjoy having the simple martial character. I use it both for the games I run for kids to introduce them to D&D (simplicity to learn) and the grognard campaigns (because they prefer gameplay to fiddly bits). I think that there are many people that enjoy a simple martial option. That doesn't mean that the Fighter is the best class ever, or the best design of a martial class. But it definitely appeals to people not in spite of, but because of the simplicity. Personally, I prefer the monk when I get to play because it is (very slightly) more complex.

You can add me to the camp that likes simple fighters. I always play Champions and turn my nose up at those other subclasses. Having said that, warlord was my favorite class in D&D 4e. So something is clearly wrong with me.

 

[MechaTarrasque]

Broadly, probably not very different, given that this mostly matches with WotC previous shared findings.

But time shall tell.

That is true. I am mostly curious about changes in the lesser-used races and classes after XGtE comes out. If aasimar sorcerers go up after the divine soul gets on Beyond and tiefling paladins go up after the conquest paladin gets on Beyond and that sort of thing. One thing that did catch my eye was genasi monk, especially given how much disdain (much of it justified in my opinion) that the elemental monk gets on the forums (assuming that most genasi monks are elemental). If they decide to go deeper into racial feats, genasi feats that have good synergy with the monk seem appropriate.

 

[lowkey13]

*Deleted by user*

 

[Kobold Stew]

A few observations stand out for me:

1. I'm really impressed with the number of halforc bards and sorcerers. The clearly unoptimized builds for a low-popularity phb race are substantial (the numbers aren't that much lower than the halforc rangers, which is a completely viable build). Cynically, we could say that even they're not being played, that gives a good benchmark of the #s that may be experimenting with builds, or trying to bend the system. But I am heartened by the dwarf wizards and halforc bards.

2. Compare that with the goliath barbarian -- the best represented barbarian race (38% of barbs made, with a non-php race). That's huge, and suggests that a substantial number of users are looking for straightforward optimization.

3. Genasi are popular and spread almost equally across classes. That reflects something of their design -- the apparent equality comes (I'll bet) because players can choose which of four main stats to add a point (though, to be fair, the charisma-classes are not underrepresented in this distribution; a fact that I find encouraging).

4. I'm surprised how well Tieflings do, until I see that over a quarter of them are warlocks (and they are almost a quarter of all warlocks), which skews the numbers.

5. Half of all clerics are humans or dwarves. The only similar clustering among two races is with fighters (also human and dwarf) and wizards (human and elf)

6. Low- or no-spellcasting classes (fighter, barb, rogue) represent 34% of all characters made (less with multiclassing). That's substantial.

 

[kenmarable]

Guess my next character will be an Aasimar Druid.

But, yes, I imagine that this result is because humans are the most powerful race and fighter is the most powerful class, at least at lower levels before options really start opening up for the others.

Unfortunately, there's really no telling why. Given my experience with all the gamers I know, I would presume it's because humans are more "normal" and fits their concept and has zero to do with how powerful they are.

In face-to-face gaming, I have met 1 maybe 2 gamers who made decisions based on what's powerful. Even the level-dippers were just going for a way to mechanically express a prior concept they had for a character. Nothing wrong with picking what you want for any reason at all, of course! I just wouldn't presume one or the other (or a third or a combination or...) is the norm across all players.

There are all sorts of different reasons why certain options will be more popular, and if I had to bet, I'd say there is no ONE reason why. It's likely humans and fighters have more than one reason going for them and it's the alignment of reasons that causes them to rise.

i.e.
Those interested in more power numbers-wise might gravitate towards options A, C, D, and G.
Those who are interested in ease of use might go towards B, C, and F.
Those who like characters similar to ones in books, movies, games, etc. might tend towards A, C, and F.
(etc. for other reasons we could imagine)

So C might really stand out, with A & F in second, and E almost ignored. When people ask why so many like C but not E, all three reasons are, in some way, correct but also miss the points the others make.

 

[TwoSix]

That was the 4e Druid ("XOMG! That Druid! Let's cut it up into three pieces, so that no one Druid PC gets more than one Cool Thing to do, yeah, that'll learn 'em!"). The pendulum's swung back the other way.
No, it's just never been a popular concept. Too 'tree hugger,' maybe, or too obscure?

I imagine it's more implementation, at least for 5e. Druid gets Wild Shape, spellcasting, and not much else. Wildshape is powerful, but can be confusing. It has poor multiclass synergy with almost everything, and has virtually zero potential as a dip class. It only has two subclasses, and they're both pretty boring. (Imagine if they had expanded Land into 6 or 7 regional circles instead, with bonus spells and wildshape bonuses into specific beasts.)

Contrast with WoW, where the druid has a deserved reputation as a powerful, flexible class, and is the 2nd most played class. (I reference WoW primarily because I'm extremely familiar with it, and it provides an enormous data set for contemporary but classic fantasy tropes.)