D&D 5E (2014) The Philosophy Behind Randomized and Standardized Ability Scores

[Charlaquin]

I agree, but honestly IME people like to roll want to get better than average scores. This is why I see so many alternate rolling rules that prevent "bad" results. My group does this, with 4 different DMs and 4 different options. When I DM the best I was able to get people to agree to is if you don't like your results you can default down to the standard array. I technically offer point buy (which is superior to the array, due to versatility), but no one every takes it.

Well yeah, of course they do. Everyone wants good stats. But part of the point of rolling is the risk vs. reward prospect.

I get it, it sucks to get terrible scores. That’s why I allow folks to roll a new set if they don’t think they’ll have fun with the ones they got. But, I do encourage them to keep bad sets if they think they can live with them. If you aren’t prepared for the possibility of getting low rolls and willing to take them as a challenge, I encourage taking the standard array/point buy instead.

In fact, my advice is, if you already have a character in mind, you should take point buy and make the best version of that character you can with the tools available to you. Rolling for stats when you already know what character you want to play only introduces a chance that you won’t get stats that suit that character. If you want to roll, you should go in with no preconceived ideas about the character, roll in order, and then create a character based on whatever stats you got. Even if the stats you roll are total garbage, the character will likely die pretty quickly anyway and then you can roll up someone new.

 

[Krachek]

But that's not what I've ever experienced. It's not "dying valiantly" it's "my PC is an idiot who commits suicide by doing really, really stupid things". Pretty much everyone knew what they did and why.

Then we go back to the OP, scores generation is tied directly to play style expectation.
If you play with players that just want bigger score, offer them this chance with a safety belt mechanic, it will be better than dummy suicide!

 

[Charlaquin]

As a player, if I’m given the option of point buy/array or rolling stats, I’ll decide based on what kind of campaign the DM is running. If they’ve pitched a plot-based campaign, where a tight-knit group of plucky heroes go on an epic quest to stop some terrible evil, I’ll come up with a character that suits that sort of story and use point buy to hand-craft that character’s stats as I envision them. If they pitch a wide-open sandbox where the characters explore wilderness, hunt for treasure, and generally pursue their own goals, I’ll roll stats, in order, and make a character that suits the stats I roll. Those are basically my two modes. I don’t really see much point in rolling if you aren’t going to roll in order.

 

[MarkB]

I agree that fairness across the group is a major factor in choosing ability score generation methods. I'm not sure I agree that the concept's been given short shrift in previous discussions. My impression was that it was one of the primary motivators for introducing point buy / standard array.

 

[Krachek]

I think of a small survey.

will you be happy to
Play a character with better stats than the others?
Play a character with lower stats than the others?
Play a character with one level higher than the others?
Play a character with one level less than the others?
Play a character with better magic items than the others?
Play a character with lesser magic items than the others?

Compare the yes/no answers. That will give you a good take on fairness tolerance or paradox!

 

[Cadence]

One addition I like beyond standard point buy/array is where PF have several levels of them based on how the campaign wanted to have the PCs relative to the world. (So some games might have a much higher than standard array start for everyone).

 

[Lyxen]

I used to love the principle of point-buy, but in these days of "builds", I think it makes for characters that are too standardised, so now it's controlled randomness, much better in the end.

 

[Oofta]

One addition I like beyond standard point buy/array is where PF have several levels of them based on how the campaign wanted to have the PCs relative to the world. (So some games might have a much higher than standard array start for everyone).

They also did that with 3.5. Funny thing is though, at least for me, is that we tried the "more powerful" variant and didn't repeat it for the next campaign. We did a 32 point buy with an option to buy up to 16 but it just felt like the PCs were almost too competent because of bounded accuracy.

YMMV of course.

 

[tetrasodium]

random rolls worked well back in 1e&2e, but not since 3.5 and there are reasons related to the historical system mechanics for that. Without those differences they cause problems. It worked till 2e because of system differences & failed different in 3.x again because of differences. 5e is different from both & is unable to make use of the ways it could work or kinda work that those earlier editions enjoyed

Prior to 3.x (1e/2e) A +1 started around a 15 in an ability while -1 started around a six in an ability giving a nine point dead zone of +/-0 that was likely to contain nearly every ability score roll. The different abilities would add a small bonus or penalty to specific things even within that dead zone, but in general it was a very minor thing. Rolling ability scores itroduced some random near ribbon flavoring to characters as a result of this dead zone that no longer exists.

One of the big improvements that 3.5* made was to standardize the ability score mods to the -1/+1 at 8 & 12 with another +1/-1 every 2 points in an ability that we all know today. You could still roll stats but the gm needed to be more careful when awarding treasure & strongly consider everyoe's rolls against the magic items they dish out making it a bit of a mess that wound up feeling lacking compared to just starting everyone with a few levels to kick them off a seasoned.


5e has the same +/-1 every 2 points from 10 as 3.x did but also assumes no feats & no magic items so the gm doesn't even have the option of massaging the trasure rewards like in 3.x & certainly doesn't have the kind of deadzone present in 1e/2e

*Maybe 3.0 too I dunno

 

[Blue Orange]

I kind of feel like with the shift from 'D&D as game of skill' to 'D&D as collaborative narrative' people want to have more fine control over their character, as BookTenTiger said. You want to make a half-orc who rebelled against the stereotypes to be a wizard and specialized in enchantment, you've got to pick your scores carefully. You just want to make a wizard, you need a high INT.