D&D General Normal Distribution Ability Scores

Try random stats where if they don't all add up to the same number you roll other dice to add or subtract until they do.

For instance, let's say you want to use 4d6 drop lowest. If the mean result of that causes abilities to add up 73 (quick search showed that--I have not verified), then you set 73 as the total number that all players ability scores will add up at the end.

So everyone rolsl their stats. Then you check their numbers. If they are above or below 73, you start rolling to get them there.

Roll a d6 to randomly target a stat. Then roll a d4 and add/subtract it to that stat. If it would take it over/under, only add/subtract enough to hit the right number. If you still aren't there, roll d6 to target a random stat and do it again until you are.

This gives you all the interesting results of having particularly low or high scores while eliminating the imbalance between characters.

You could even allow players to choose to take point buy instead if they didn't like their results. Or you could allow players to trade their stat arrays.

I hate with a burning passion PC ability score inequality, but there are ways to spice it up with randomness in situations where everyone doesn't already have a defined character concept without causing that issue.
 

log in or register to remove this ad

Hriston

Dungeon Master of Middle-earth (He/him)
(15d2) - 12 comes very close to a normal distribution. 10.5 average / standard deviation 1.94.

Or you could roll three "averaging dice", d6's where 1's are counted as 3's and 6's are counted as 4's. The range is 6-15 with an average of 10.5 / standard deviation 1.66.
 



Xeviat

Dungeon Mistress, she/her
@Xaviet are you looking for a normal distribution or do you want it skewed for higher scores like most methods?
Both! 5d4-2 gets you a good normal distribution for NPC generation, while 7d4(keep 5 highest)-2 would be for PCs.
 


Xeviat

Dungeon Mistress, she/her
So, you really want the skewed for PCs. I thought you were looking for a normal distribution for everyone...
Nah, PCs are special. This was more an exercise in seeing what a more normal distribution could be for, and how to switch rolling to d4s
 



Bacon Bits

Legend
Also, rolled stats can help you get a feel for "regular folk".

I don't see why I'd want that. And I say that not to poo-poo your idea. I say it because over those 30 years we have strongly favored rolling dice for characters for D&D.

I can't imagine one time in the past 30 years across all the editions that I've played that I've ever wanted to play "regular folk" in D&D. I'm playing D&D because I'm not playing a regular folk. Even in B/X and 2e, I'm not playing a regular folk. It's just that the monsters are really nasty.

If I'm playing a game where I am playing a regular folk, I'm extremely unlikely to be playing any fantasy genre at all. I'm almost always playing a horror genre of some kind where some of the themes are about denying agency to the players. CoC, Twilight: 2000, All Flesh Must Be Eaten, etc. Even sword & sorcery or sword & sandal settings don't really feel like they're geared towards the PCs being regular folks in quite the same way that almost anything contemporary or modern does.

What we want is:
  1. A chance at greatness
  2. A chance at a flaw
  3. A good overall baseline
  4. Sometimes having to make hard choices
  5. Sometimes not having optimal stats
  6. Not being 100% equal to everyone else, but not being 50% of anyone, either. Nobody likes having a +5 primary when others are stuck with a +2.
  7. Not always getting a perfect point buy array for a Wizard means you can't play a Wizard in every campaign over and over and over again. It doesn't enforce variety, but it can encourage it.
So we're not just looking for equitable uncertainty. We're looking for a just fate. But dice are random, not just, and not fate. Dice are terrible storytellers, and we like them not because they can see the future, but because they have no memory.

In the past, we've done things like:
  1. Roll whatever, but you can always choose 15, 14, 13, 12, 10, 8 if you don't like your roll. So there's a higher floor.
  2. You're spotted a 15. Then roll 5 stats.
  3. 2d6+6 rolled 7 times. Keep 6 of them. If the score you throw out isn't the lowest score you have or tied for the lowest score you have, you get a bonus of ~200gp or so (I don't remember the specifics anymore).
  4. As above, but 1d10+8.
  5. Taking a half dozen or so pre-made arrays that are within an acceptable range, putting them on a d6 chart and rolling for them. Honestly, I wish this were presented in the DMG.
  6. Using the "cards" method. Like you take 12 cards: 4,4,5,5,6,6,7,7,8,8,9,9 (or whatever power level is appropriate for your game... a total of 72-78 total face score works well). Shuffle them, and deal them into six piles of two. Each pile added together is one of your stats. Optionally, one time you can exchange one card from one pile with one card from another pile.
The big issue with point buy is that it's actually, "pick one of 3-5 arrays." But it hides that reality behind this illusion of choice that the complexity brings with it. Why not just give us the list of 6 or so arrays that we can choose from? Then if we want to we can "roll" and generate the array we use randomly. It also means that those arrays can be balance individually for the game, not balanced around some nebulous "ability cost" system that certainly doesn't map to the reality of how stats work in the game for all classes. Heck, you could tie the arrays to class, even! Wow, then you can get rid of this annoying +2/+1 or +1/+1/+1 thing we keep shuffling around for historic reasons.
 

Split the Hoard


Split the Hoard
Negotiate, demand, or steal the loot you desire!

A competitive card game for 2-5 players
Remove ads

Top