D&D 5E How does your group determine ability scores?

Which method of determining ability scores is the most used in your D&D 5E group?

  • Roll 4d6, drop lowest

    Votes: 43 29.5%
  • Default scores (15, 14, 13, 12, 10, 8)

    Votes: 24 16.4%
  • Customizing ability scores variant (point-buy)

    Votes: 60 41.1%
  • Mix of rolled and default

    Votes: 1 0.7%
  • Mix of rolled and customizing

    Votes: 6 4.1%
  • Mix of default and customizing

    Votes: 8 5.5%
  • Mix of all three

    Votes: 10 6.8%
  • Other (please specify)

    Votes: 22 15.1%

  • Poll closed .
Anecdotes aren't data.
Actually, they are. Often presented as if they were a different kind of data, to support conclusions they don’t, but they are absolutely, incontrovertibly, data.

And in this case, where I was describing my own experience (in contrast to another poster’s anecdotal description of their experience) they are in fact data.

Anyway, have a cookie for mis-applying a false cliché.



That's a bizarre assertion.

Yeah, that would be a first on the forums: two different people using a word differently.

/eyeroll
 

log in or register to remove this ad

Actually, they are. Not always the kind of data I prefer to have to draw the kinds of conclusions that interest me, but they are absolutely, incontrovertibly, data.

And in this case, where I was describing my own experience (in contrast to another poster’s anecdotal description of their experience) they are in fact data.

Anyway, have a cookie for mis-applying a false cliché.
it always bugs me when people say "the plural of antidote isn't data" on these boards... NOBODY HAS BETTER Data...
 

Mannahnin

Scion of Murgen (He/Him)
it always bugs me when people say "the plural of antidote isn't data" on these boards... NOBODY HAS BETTER Data...
(anecdote)
It really depends on the situation and what kind of argument it's being used to support. A single anecdote IS useful data to prove that a thing is possible or sometimes happens. A single anecdote isn't sufficient data to establish a trend or larger pattern. Nor is a few anecdotes.

For SOME things we do have more comprehensive real data, like with WotC surveys about characters in D&D Beyond. They may not be wholly indicative of larger play trends outside DDB, but they are real data about what folks using that program are doing.
 

Mannahnin

Scion of Murgen (He/Him)
PF2 came up with a seemingly clever "ABC" system. Start with a baseline all 10s. Pick you ancestry (formerly race) get a few stat bumps, pick your background get a few more bumps, pick your class and get your final bumps. It all works out to basically 2-4 different arrays and everyone just skips to it that way.
Beyond the Wall and Other Adventures, an OSR game designed to evoke young adult fantasy like Lloyd Alexander's Prydain books and LeGuin's Earthsea takes an approach like this. Each player picks a character pack with an archetype like Wizard's Apprentice, Hunter, or New Guardsman, and gets a baseline set of scores. They then roll on an included set of tables for life events growing up, and get bonuses to their ability scores until they complete the lifepath. That gives you your starting array.

4D6 the way god intended.
(if your god is Gygax) :)
New testament?

Old Testament.
(1979 old!)
Ive heard that like many of the rules Gygax wrote down, he had his own way of doing things despite that. Apparently he liked using 4D6 drop the lowest, but that may have been just for AD&D. shrugs

I don't have my DMG easily to hand, but my understanding is that 4d6-drop-lowest is actually the recommended method in 1st Ed.

2nd Ed had six (IIRC) methods in the PHB, starting with 3d6-in-order and becoming gradually more generous as they went. But the 2nd Ed DMG then had all manner of dire warnings about how 4d6-drop-lowest was a power gamers' paradise and would absolutely devastate any game that dared use it. (I exaggerate, but not by that much...) Of course, 2nd Ed was post-Gygax.
Yes. 4d6 drop the lowest, arrange to taste is the primary recommended method in 1E AD&D, per the 1979 DMG. The PH (which, awkwardly, was published in '78) doesn't actually specify how to roll, but Gygax tells us in it that a character without at least two 15s is probably not good enough and should likely be re-rolled. Which makes sense given the complex AD&D ability bonus tables, which MOSTLY (with a few exceptions) require quite high numbers to get any meaningful bonuses.

In the 1974 OD&D game, and in the Basic/Expert (1981) and BECMI (1983+) lines you roll 3d6 down the line, but a) Bonuses are almost nonexistent/not very important (1974) or easier to get ('81/'83), and b) you've got the ability to trade ability score points, reducing above average stats which are less relevant to your character to increase your Prime Requisite.

2E AD&D fumbled the ball on this pretty badly, offering 3d6 arrange to taste as the primary method of character generation, but keeping (somewhat rationalized and simplified) ability bonus tables which require high ability scores to get bonuses, and not giving any point-trading rules. The example of character generation in 2E offers a couple of famously mediocre ways to build the fairly weak PC they demonstrate with. Extremely uninspiring for anyone looking to make a hero.

In practice even 4d6, drop lowest, arrange to taste isn't generous enough to reliably produce PCs with the two 15s minimum Gygax envisioned and the AD&D ability score charts expect, so for most of my 90s-2000s gaming career my main groups used 4d6 drop lowest, roll three sets and pick your favorite, arrange scores to taste.
 
Last edited:

delericho

Legend
2E AD&D fumbled the ball on this pretty badly, offering 3d6 arrange to taste as the primary method of character generation...
It's worse than that. 2nd Ed's default (method 1) was 3d6 in order.

Method 2 was the same, but roll two sets and choose between them. Method 3 was 3d6-and-arrange. Method 4 was roll 3d6 twelve times, and arrange any 6 to taste. Method 5 was the well-known 4d6-drop-lowest-arrange.

Method 6 was an interesting one (also the one we used for all but one of our 2nd Ed campaigns). Each score started at 8, and you rolled 7d6. You could then add the numbers on the dice to the various scores to suit, but you couldn't split a die. So if you were lucky enough to roll 7 sixes, you'd end up with a 14 in every score.

I would dispute the "fumbled the ball" description - they seem to have made the game they intended to make, with the focus being on mere mortals becoming heroes, rather than starting with superhumans as in other editions. That's a valid design choice, albeit not a popular one.

(It's also worth noting that this is one of only two instances where the designers managed to put power-creep in reverse - the other being "The Complete Priest's Handbook" in the same edition.)
 

Maxperson

Morkus from Orkus
I don't have my DMG easily to hand, but my understanding is that 4d6-drop-lowest is actually the recommended method in 1st Ed.
The 1e DMG lists 4 methods for rolling characters, all of which are recommended.

I: 4d6-L
II: 3d6 rolled 12 times and keep the highest 6.
III: roll in stat order 3d6 rolled 6 times for each ability and keep the highest for each stat.
IV: 3d6 rolled in stat order for 12 characters and keep the one set you like.
2nd Ed had six (IIRC) methods in the PHB, starting with 3d6-in-order and becoming gradually more generous as they went. But the 2nd Ed DMG then had all manner of dire warnings about how 4d6-drop-lowest was a power gamers' paradise and would absolutely devastate any game that dared use it. (I exaggerate, but not by that much...) Of course, 2nd Ed was post-Gygax.
2e does give 6 methods, but method 1 is the default. It was 3d6 in order for the 6 stats. Methods II-VI are listed as alternative rolling methods.
 


Maxperson

Morkus from Orkus
I could swear that the 2E example character generation uses arrange to taste, though.
Page 13 of the 2e PHB

"Let’s first see how to generate ability scores for your character, after which definitions of each ability will be given.

The six ability scores are determined randomly by rolling six-sided dice to obtain a score from 3 to 18. There are several methods
for rolling up these scores.

Method I: Roll three six-sided dice (3d6); the total shown on the dice is your character’s Strength ability score. Repeat this for Dexterity, Constitution, Intelligence, Wisdom, and Charisma, in that order. This method gives a range of scores from 3 to 18, with most results in the 9 to 12 range. Only a few characters have high scores (15 and above), so you should treasure these characters."

Then...

"Alternative Dice-Rolling Methods

Method I creates characters whose ability scores are usually between 9 and 12. If you would rather play a character of truly heroic proportions, ask your DM if he allows players to use optional methods for rolling up characters. These optional methods are designed to produce above-average characters."

Method V is 4d6-L and arrange the numbers.

:)
 



Mannahnin

Scion of Murgen (He/Him)
I don't see a sample character in the first 50 pages. Maybe it was in the DMG.
It's not a full sample character with a character sheet. They give an example of ability score generation; I think the example character is named Rath, maybe? They talk through a couple of possible personality types that the (mediocre) ability scores could represent. I think it's at the end of the ability score section/chapter, if memory serves.

I can look it up after work. 🤷‍♂️
 

Maxperson

Morkus from Orkus
It's not a full sample character with a character sheet. They give an example of ability score generation; I think the example character is named Rath, maybe? They talk through a couple of possible personality types that the (mediocre) ability scores could represent. I think it's at the end of the ability score section/chapter, if memory serves.

I can look it up after work. 🤷‍♂️
So looking for Rath, his name comes up a few times in the individual stat sections, but with no word on how they arrived at his number. At the end of the ability score section it gives his stats as S:8 D:14 C:13 I:13 W:7 CH:6, so it doesn't seem like he rolled 4d6-L, but rather the 3d6 in order.
 

James Gasik

Legend
Supporter
It's worse than that. 2nd Ed's default (method 1) was 3d6 in order.

Method 2 was the same, but roll two sets and choose between them. Method 3 was 3d6-and-arrange. Method 4 was roll 3d6 twelve times, and arrange any 6 to taste. Method 5 was the well-known 4d6-drop-lowest-arrange.

Method 6 was an interesting one (also the one we used for all but one of our 2nd Ed campaigns). Each score started at 8, and you rolled 7d6. You could then add the numbers on the dice to the various scores to suit, but you couldn't split a die. So if you were lucky enough to roll 7 sixes, you'd end up with a 14 in every score.

I would dispute the "fumbled the ball" description - they seem to have made the game they intended to make, with the focus being on mere mortals becoming heroes, rather than starting with superhumans as in other editions. That's a valid design choice, albeit not a popular one.

(It's also worth noting that this is one of only two instances where the designers managed to put power-creep in reverse - the other being "The Complete Priest's Handbook" in the same edition.)
Yeah but they quickly reversed that trend once Legends and Lore and Demihuman Deities, making powerful specialty priesthoods. And once Forgotten Realms dropped their specialty priesthoods, well, Priests being arguably the best class in the game was pretty much a done deal.
 

delericho

Legend
The discussion has stirred up a memory...

The more control you have, the more scope there is for optimization. That is, although the 4d6-drop-lowest on average produces better results than point buy, the fact that you can't determine the exact scores undercuts that at least somewhat.

Back when we were coming to the end of our 3e play, we therefore offered each player a choice:
  • 4d6-drop-lowest, with a reroll for "hopeless" characters,
  • A standard array of 16, 15, 13, 12, 10, 8,
  • 28-point buy.
Per the 3e point buy scheme, the average for that rolled method was, in theory (and IIRC), around 30.5 points, while the array was 29.

In 5e terms, what that mostly means is that the "rolled" method is actually, on average, a good bit "better" than point buy - but at the cost of loss of control. (This method also has the scope for throwing out a 16+ score...) Note that 5e as written doesn't have the reroll for "hopeless" characters, though I'd expect most DMs apply one.

And 5e's standard array is actually slightly worse than point buy - although the points can generate exactly that array, for almost all characters there is a more optimised option out there.

That being the case, I'd be tempted to suggest offering a choice as above: 4d6 with a "safety net", or a slightly enhanced fixed array (possibly including a 16 - indeed, possibly the same one given above), or the normal point buy as given.

At this point, though, I'm mostly just churning a bunch of ideas. :)
 


DND_Reborn

Legend
In practice even 4d6, drop lowest, arrange to taste isn't generous enough to reliably produce PCs with the two 15s minimum Gygax envisioned and the AD&D ability score charts expect, so for most of my 90s-2000s gaming career my main groups used 4d6 drop lowest, roll three sets and pick your favorite, arrange scores to taste.
FWIW, the expected "standard array" for 4d6-L is 16, 14, 13, 12, 10, 9; which in AD&D would have barely given you anything for just 2 scores...

(bold added)

We also did three sets back in the 80's and 90's, but rolled in order -- OR --- one set, arrange to taste.
 

payn

Legend
It makes sense, but I don’t like it in practice as it slows down character creation and makes it too convoluted for the benefit it provides. I’d prefer Point Buy and letting the player decide if and how the character’s background ties into the ability scores.
I agree, most groups dispensed it for the 2-4 arrays that everyone eventually ends up with any way. 🤷‍♂️
 

PF2 came up with a seemingly clever "ABC" system. Start with a baseline all 10s. Pick you ancestry (formerly race) get a few stat bumps, pick your background get a few more bumps, pick your class and get your final bumps. It all works out to basically 2-4 different arrays and everyone just skips to it that way.
I thought that would be clever, a life path type system. I've always had a great appreciation for Traveller.

You start with X in each, and you choose which attribute will be "prime". This might be the one you want to be highest, or one that is appropriate for the family. Then you have a coming of age event, and two choices, things go well or things go poorly. You gain +2 / +1d6 regardless of the choice, but the attribute is different. Your background is from a list derived from the prime attribute. You might come from a family of farmers or masons if Strength is prime, Charisma if travelling entertainers. Ancestral mods to one or two attributes, and then maybe +1d6 to each to give some randomness / variety.

Haven't had the time to mess with it yet.
 

Dungeon Delver's Guide

An Advertisement

Advertisement4

Top