538 reports on frequency of PC Class/Race combos

[alwaysbebatman]

https://fivethirtyeight.com/features...haracter-rare/

An interesting article from the mainstream data journalism site fivethirtyeight. Somebody may want to also post this in the roleplay forum.

But I thought it might be an interesting conversation starter here, as well.

If you compare the most common combinations we see here with the "PC-type" characters in the comic... Not too many surprises really, as to which characters are more playing to type (human fighter, elven wizard, dwarf cleric) and which play against type (halfling ranger)...

The Scribble is more of a straightforward generic party: halfling rogue, dwarf barbarian, etc.

I thought for a second that Girard was their playing-against-type character, but no: sorcerer is one of the more common classes for dragonborn.

Anyway, just something I thought people might find interesting...

[Kish]

Looking at the article in question, it appears, though the edition isn't stated, that it's aimed at 5th edition, for whatever effect that has on the applicability of the results. (Notably, the existence of the "dragonborn" race, which means a lizardfolk-mammal hybrid Rich has made fun of, not a human with a touch of dragon blood.)

[alwaysbebatman]

Oh! That's not the same as what Girard is? I didn't realize.

Yeah, this is presumably 5th ed related. Some sort of online player sheet they're using now.

I suppose something similar being used by Pathfinder players would be more relevant to OOTS.

[Kish]

That's what the lizardfolk with breast implants meant by saying she was trying to stay current; 4ed had just come out and introduced the new "dragonborn" core race.

[Synesthesy]

Dnd is built giving races the right bonus to make some classes better.... an elf wizard is better then an elf barbarian, just to say.... I remember that in my first campain, the enemy was an evil elven barbarian tribe....
I plaied against type in Skyrim, when I used a Kajit Wizard to pass level 80: that game is built that if you play against type, you are more balanced in the early game.

Looking at the article in question, it appears, though the edition isn't stated, that it's aimed at 5th edition, for whatever effect that has on the applicability of the results. (Notably, the existence of the "dragonborn" race, which means a lizardfolk-mammal hybrid Rich has made fun of, not a human with a touch of dragon blood.)

Thank you, you taught me something

(the dragonborn as a race and the simple past of to teach )

[D.One]

I thought for a second that Girard was their playing-against-type character, but no: sorcerer is one of the more common classes for dragonborn.

I've never quite seem Girard as Dragonborn, as the race shown in late editions, but solely as the explanation for his sorcerer powers. In 3.5, if I recall correctly, sorcerers' spellcasting powers derive from innate magical abilities due to their lineage, with them being usually distant descendants from beings of great magical prowess, as outsiders, fey or dragons. There's even a prestige class, Dragon Disciple, that further increases that bloodline power.

In fact, in the Class and Level Geekery thread, we put him as a black-dragon blooded human.

[Potatomade]

Assuming it's 5e only, I wonder how many people chose "variant human" instead of regular ol' human. Unless I missed it, the article doesn't talk about that, and just mentions humans as having +1 to all ability scores (which is actually weak as hell compared to more tailor-made races or vumans). Does D&D Beyond even have variant humans?

[JustAnotherSoul]

That is interesting. It amuses me that elf barbarian is more than twice as popular as half-elf barbarians. In fact, that's the lowest race/class combo until you get down to Half-orc.

It also looks like there were about 9,000 multi-class data points (the numbers adds up to about 109,200 and they said this was out of 100,000 and multiclassed characters count as one for each class). I'd be interested to see if certain races are multi-classed more than others etc etc. I'm also curious how this data looks not based on characters but levels in each class (obviously it would be skewed towards L1 but still). I might see if I can find out how they got this data and see if I can play around with it myself.

It strikes me now that this is also probably off topic in comics -> The order of the stick section of the forum...

[alwaysbebatman]

Yeah, it's a little off-topic, mainly because it's 5e and therefore not really as relevant to analysis of the comic as if it were 3.5 or pfs.

But since you mentioned other data that would have been interesting (specific multi-classes), another that would have been extremely interesting would be a whole other axis for alignment...

[wumpus]

Yeah, it's a little off-topic, mainly because it's 5e and therefore not really as relevant to analysis of the comic as if it were 3.5 or pfs.

But since you mentioned other data that would have been interesting (specific multi-classes), another that would have been extremely interesting would be a whole other axis for alignment...

Any guesses that if you checked "all characters that were played multiple sessions" that fighter wouldn't be so high? Of course, 5e is a bit of a throwback to AD&D (which recommended even more fighters than that), but I still expect that many of those fighters were made to be "beginner friendly".

Human tends to be a "safe bet" for D&D, and as the player's race has certain role playing advantages. While I expect that clerics, wizards and rogues should start to equal fighters, I really don't expect the other races to catch up to humans (unless there are sufficient advantages: see early DDO, which was said to stand for "dwarves and drow online").

[137beth]

Looking at the article in question, it appears, though the edition isn't stated, that it's aimed at 5th edition, for whatever effect that has on the applicability of the results. (Notably, the existence of the "dragonborn" race, which means a lizardfolk-mammal hybrid Rich has made fun of, not a human with a touch of dragon blood.)

Actually the edition is specified, albeit indirectly. They say in the article they are basing their statistics on D&D Beyond, which, among other things, contains WotC's online character generation tool for 5e. D&DB only supports 5e, so that is what the data in the article is based on.