D&D 5E A different way to roll stats for 5E

[Zelc]

TLDR: try using 4x(d4+1) drop lowest, add +1 to the three highest scores below 15.

Ability score increases are a significant way in which characters progress in 5E. As a result, the maximum score you can achieve through the default array and point buy is 15 unmodified, or 17 after racial adjustments. Rolling 4d6 drop lowest can generate abilities as high as 18 unmodified, or 20 after racial adjustments. This might take away some of the power increase the player feels through gaining levels.

A simple change is to roll 4x(d4+1) drop lowest instead. This generates ability scores between 6 and 15. Below is a table of the summary statistics between the default array, 4d6, and 4x(d4+1):

	Default	4d6	4x(d4+1)
Mean (Total)	72	73.5	69.7
Var (Total)	NA	21.4	48.6
Pr(<=9)	16.7%	17.5%	14.1%
Pr(>=15)	16.7%	23.1%	5.1%

The mean and probability of getting a 15 are lower for 4x(d4+1) than for the default array. Adding +1 to the three highest ability scores below 15 solves this problem. The mean becomes 72.7 and the probability of a 15 is roughly 16.8%.

This method also has much lower variance than the 4d6 method. This reduces the likelihood of a character being much more powerful or much weaker than the others at the table. For some, this might be a downside. I'd consider this an advantage for the 4x(d4+1) method.

If you want to try this:

http://anydice.com/program/5ec1

Test out a few rolls of 6 (try not to draw conclusions from less than 20 trials ). Remember to add +1 to the three highest numbers below 15.

 

[doctorhook]

Nice work! I'll play around with your method.

FWIW, I developed another rolling method a few months back, which has definite similarities to yours (including using AnyDice.com while designing it): roll 3d4, drop the lowest roll, then add 7 to the sum of the two highest rolls in order to generate a score--repeat this twice so that you've generated three scores; next, roll 3d4, drop the lowest roll, then add 6 to the sum of the two highest rolls to generate another score--repeat this twice as well so that you've generated a total of six scores. The lowest possible array ends up being 9 9 9 8 8 8, while the highest is 15 15 15 14 14 14. I designed it to mimic the range of scores possible using point buy. (I don't know how to describe my method in nice notation, btw.)

Having used it to generate all the characters in my newest campaign, I learned a couple things about my method: 1) You get a lot of blandly well-rounded characters. Almost all of them have at least 12s in every score. 2) Several characters ended up "too weak", having no 15s at all. I ended up letting everyone replace one score with a 15, which also removed their lowest scores, making them even more well-roundedly bland. 3) Even though we were rolling, it didn't feel like there were much stakes. Overall, I think I'll try a different approach next time, since I think my idea worked a lot better in theory than in practice.

 

[CapnZapp]

In the end, this mostly adds complexity.

I really don't see how this can be worth not simply taking the standard array. And for those that absolutely can't stand having the same numbers as the next guy, there is point buy.

There really is no way you can have random rolls that are also controlled and easy. Choose any two, but not all three.

 

[Sacrosanct]

If I want a rolling method that ends up with pretty much the array, I'll just use the array and skip the math. No offense intended to your work, but just saying what I'd probably do.

Also, I know 5e gives you the opportunity to increase your stats at every 4 levels (approximately), but that's never been tied to how I feel my character advances when hitting a level. I certainly don't feel like I'm losing out on a power increase. IIRC, that was a 3e thing (stat increases at level gain). I've always felt things like increased hp, better spells, etc all did that well enough. Then again, I've never believed in "dead levels" anyway. HP increases at every level, which is pretty darn important. YMMV of course (and often it does).

 

[the Jester]

What are you trying to accomplish here? If you want less variability in stats, I'd suggest using an array or point buy. Most people who want to roll want the swinginess, including the chance of low stats.

 

[GMMichael]

In the end, this mostly adds complexity.
I really don't see how this can be worth not simply taking the standard array. And for those that absolutely can't stand having the same numbers as the next guy, there is point buy.

4d4+3 drop lowest is complex? This looks to me like a really good system for reigning in errant ability scores. It uses some variability, trims off the outlying scores, and most importantly it includes a "drop lowest" feature so that players can get their optimization fix at the beginning of the game.

There really is no way you can have random rolls that are also controlled and easy. Choose any two, but not all three.

All three: "3d6, go cry to yo momma."

 

[Hereticus]

You can just use d6s, or you can use d4s+1, or you can use d8s-1 (stats range from 0 to 21).

Back in my 3.5 days, I gave the option to use 4d6 (drop lowest) or 2d6+1d8 (no drop).

 

[SubDude]

What is the concern with just using the default 4d6 from the PHB? Is it more the chance of extreme scores, or the variability from PC to PC?

 

[Jeff Albertson]

2d8, drop the lower result, add 7, six times, arrange as desired.

 

[doctorhook]

What is the concern with just using the default 4d6 from the PHB? Is it more the chance of extreme scores, or the variability from PC to PC?

For me it was the fact that 5E's hard cap on ability scores meant that characters who rolled 16 or better at level 1 didn't have much room left to grow. At 4th level, our dwarf fighter found some Gauntlets of Ogre Strength in Wave Echo Cave, and he had no use for them because he was already stronger than an ogre! Overall, it's not the worst problem, but it helped me realize the advantage of using the default array.