D&D 5E Point buy vs roll

[Cadence]

For 4d6-L, I think these are the chances of the different possible values of total modifiers for each character, adding across all 6 abilities:

0 or less = 9.3%
1=5.4%; 2=7.2%; 3=8.9%; 4=10.3%;
5=11.09%
6=11.0%, 7=10.2%, 8=8.6%, 9=6.7%
10 or more = 11.3%

Nearly 10% of 4-person parties will have a difference in total modifiers of 12 or more between the characters with the "best overall rolls" and "worst overall rolls" (so an average difference of +2 in terms of modifiers across all abilities). Nearly 70% will have a difference of 6 or more in total modifiers between the highest and lowest. For 6-person parties those jump to nearly 20% and 90% chances of differences of 12+ and 6+.

I've got to hand it to all the players that don't mind when someone else at the table averages a +2 bonus better across every ability. I'm kind of jealous. I think it would start to annoy me.

 

[Oofta]

Compensate is a strong word: Taormina challenges to the party is more like it. Which is always going to he true.

Compensate wasn't my term. But what a small town in Italy has to do with anything is beyond me. Unless "Taormina" is auto correct gone awry or has some other definition.

 

[Parmandur]

Compensate wasn't my term. But what a small town in Italy has to do with anything is beyond me. Unless "Taormina" is auto correct gone awry or has some other definition.

Tailoring the. Fat fingers, Samsung autocorrect...

 

[Parmandur]

For 4d6-L, I think these are the chances of the different possible values of total modifiers for each character, adding across all 6 abilities:

0 or less = 9.3%
1=5.4%; 2=7.2%; 3=8.9%; 4=10.3%;
5=11.09%
6=11.0%, 7=10.2%, 8=8.6%, 9=6.7%
10 or more = 11.3%

Nearly 10% of 4-person parties will have a difference in total modifiers of 12 or more between the characters with the "best overall rolls" and "worst overall rolls" (so an average difference of +2 in terms of modifiers across all abilities). Nearly 70% will have a difference of 6 or more in total modifiers between them. For 6-person parties those jump to nearly 20% and 90% chances of differences of 12+ and 6+.

I've got to hand it to all the players that don't mind when someone else at the table averages a +2 bonus better across every ability. I'm kind of jealous. I think it would start to annoy me.

I don't doubt the math, but it's never come up? Occasional high fives for a cool character?

 

[Cadence]

I don't doubt the math, but it's never come up? Occasional high fives for a cool character?

It just makes me curious now if there is any overlap between folks who think a single +2 ability bonus (+1 modifier bonus) is discouraging in some contexts, and the folks who roll 4d6-L. Here's where a post scanning AI that took verbal commands (or came with a free programmer) would be useful.

 

[Parmandur]

It just makes me curious now if there is any overlap between folks who think a single +2 ability bonus (+1 ASI bonus) is discouraging in some contexts, and the folks who roll 4d6-L. Here's where a post scanning AI that took verbal commands (or came with a free programmer) would be useful.

I think the (casual and not statistically relevant) results of this poll show that WotC made a good decision with the options in Core, as they appeal to different tables at different times.

 

[Maxperson]

For the record,, this ia all thst is needed to be effective in each Class, from the SRD:

Class	Ability Score Minimum
Barbarian	Strength 13
Bard	Charisma 13
Cleric	Wisdom 13
Druid	Wisdom 13
Fighter	Strength 13 or Dexterity 13
Monk	Dexterity 13 and Wisdom 13
Paladin	Strength 13 and Charisma 13
Ranger	Dexterity 13 and Wisdom 13
Rogue	Dexterity 13
Sorcerer	Charisma 13
Warlock	Charisma 13
Wizard	Intelligence 13

In a secondary, support role. As the primary class I maintain that 14(+2) is where the baseline seems to be.

 

[Randomthoughts]

I went with point buy after I came back from 2e (to 4e). I just didn’t want to deal with the hassle of players showing up with a PC that had rolled multiple 18s (sure you did). My rule was rolling in front of each other - but it made for some awkward conversations.

 

[Oofta]

I wrote a program a while back to find how varied individual tables would be if you translated 4d6 drop lowest into point buy. I tweaked it a tiny bit to allow for numbers below 8 and above 15 (the average low and high values).

Spoiler: My slightly modified array

Score	Point Cost
3	-16
4	-10
5	-8
6	-7
7	-2
8	0
9	1
10	2
11	3
12	4
13	5
14	7
15	9
16	11
17	13
18	16

Out of my first run for 100 groups of 6, the biggest variation point buy was:

13, 12, 10, 8, 8, 5​

18, 17, 16, 15, 12, 10​

I think this is a pretty huge difference that is going to have significant lasting effects. Also, coincidentally fairly close to the difference I saw the last time I was in a game that people rolled for ability scores except that the person with the high had at least 2 18s.

I ran it again 10,000 times, just to get the average difference per table of 6. On average there was around a 25 point difference. I picked a sample at random:

20 point buy: 15, 13, 13, 10, 8, 7​

45 point buy: 17, 14, 14, 14, 13, 12​

The 20 point buy PC isn't horrible, but they will always be significantly behind the curve of the guy that got 45 points. I'm sure if I cherry picked at all I could find something worse. Also remember this is just average.

So while this isn't totally official, using my array the average point buy of 4d6 drop lowest was 31 and shows that in the average group the difference isn't just a +/- 1 here and there.

 

[TwoSix]

I went with point buy after I came back from 2e (to 4e). I just didn’t want to deal with the hassle of players showing up with a PC that had rolled multiple 18s (sure you did). My rule was rolling in front of each other - but it made for some awkward conversations.

Yea, rolling on your own is pretty much a non-starter for me. If we're going to roll, we do it at session zero with the whole group there.

For my personal taste, if we're going to play with procedural generation rather than build-to-concept, I'd rather do a stat draft or playbooks. I'd only really do standard rolling in an OSR game where the stats (and the characters, really) matter less.