![]()
We’re all probably familiar with the alternate dice-rolling methods for character generation that first appeared in the 1st Edition Dungeon Master’s Guide and then later in the 2nd Edition Player’s Handbook. I’ve never been much of a fan of these methods. They tend to emphasize the “roll” rather than the “role” in RPGs. They often felt like a race to “all 18s,” couched in the language of “these are just suggestions.”
While it may simply be a matter of language, I would prefer that all methods of character generation be presented and codified as viable methods. Below is a list of methods that I find most compelling, presented for use with the One-Sheet rules.
Natural Selection
Roll 3d6 in order for each attribute. Die rolls of 12, 9, 4, 5, 15, and 15 resulting in Might 12, Magic 9, Piety 4, Agility 5, Stamina 15, and Charm 15. A bit clumsy, but this could easily be a Hero or Dwarf with commanding leadership skills.
Nurture vs. Nature
Roll 3d6 six times and arrange by preferences. The same die rolls above could give us a much more effective Hero: Might 15, Magic 4, Piety 5, Agility 12, Stamina 15, Charm 9.
Balanced Party
The GM rolls 3d6 six times and each player arranges the same results according to preference. Continuing with the die rolls above, one player might create the same Hero while another player might create a Wizard for the party: Might 4, Magic 15, Piety 5, Agility 15, Stamina 9, Charm 12.
A Little Extra Luck
Roll 4d6 in order for each attribute, dropping the lowest die.
A Lot of Extra Luck
Roll 4d6 six times, dropping the lowest die. Arrange in order of preference.
Point Distribution
The middle of the bell curve for a 3d6 die roll is 10.5, multiple this by 6 attributes for a total of 63. Distribute these points to each of the attributes. You could max out your Hero (Might 18, Magic 4, Piety 4, Agility 16, Stamina 16, Charm 5), but only by sacrificing wit and charm.

I personally have always been a big fan of what you term natural selection. You sometimes get characters who just suck. And guess what? They usually die fairly early on! Like would happen if I decided to go out and fight a bunch of orcs. And I had some great characters that were great “characters” because they had strengths and some glaring weaknesses, and I worked within those bounds to structure a personality that might logically have stemmed from those attributes. Oh sure, when I was 12 I just kept rolling until I got all 18s. But those aren’t the characters or the adventures I remember.
I’ve had much the same experience, but I realize there are other forms of play out there. Some people only want to play heroes in the sense that they are on the top half of the bell curve. I think the bell curve is there for a reason. Of course, I’m the guy who had never played a Paladin in AD&D.
I lean toward the Nuture vs. Nature. Roll your stats but assign them to the attribute you want. I think this gives people the chance to play the character they want to play. This isn’t life, so why limit the players to playing a hero because they rolled a bad magic attribute.
I find the idea of playing a wizard with a low magic attribute pretty compelling. He may not survive for very long, but if he does, the advancement in level will eventually even out his humble beginnings.
I generally do 3D6 down the line, and then switch the scores of two attributes (one single switch, that is). In simpler games where most classes/races are only really dependent on one single attribute, this means you get a lot of flexibility while still maintaining randomness.
It all depends, though. Some variants (notably AD&D) where abilities might only give real bonuses at scores of 16+ really don’t support this method that well. In the end, I usually confer with the players and we decide upon a method together.
I hate challenge and detest losing. I have too much of both in RL, so I’d favor “A lot of extra luck.” Failing that, I think the most interesting proposal is balanced party. That way at least no one I’m playing with is statistically better than I am. If you let me distribute points, I will nuke my charisma and will power so I can have an IQ of 21 and a dex of 20.
The modifier distribution is an issue worth addressing that this exercise has caused me to examine. In OD&D the bonuses were almost entirely related to XP and the prime requisite. By 1st Ed AD&D the ablility tables were extremely cumbersome, particularly all the various strength modifiers.
A more equitable distribution might be:
3 = -4
4 = -3
5-6 = -2
7-9 = -1
10-11 = 0
12-14 = +1
15-16 = +2
17 = +3
18 = +4
It would create more value to ability scores that deviate from the mean values of 10-11. This is all good fodder for unit testing.
I’m wondering about a Traveller style mechanic where characters can die during creation, e.g., a Wizard with a stamina of 4 rolls 1d6 for HP and gets a 2. Apply the -3 modifier and the Wizard is no more. This would weed out the least hardy adventurers.
The downside of the Traveller method was old starting players.
HarnMaster was similar if you remember.
I mean old characters.
Given the amount of time to get through character creation in Traveller, I think old players is accurate as well.
You wouldn’t have old starting characters. They would die pretty quickly. Roll abilities. Roll HP. Continue or rinse and repeat.
I’ve been thinking of a different tactic. Characters could start with a base value of HP, say 6, plus the die roll. So the Wizard with the -3 modifier would start with 6 – 3 + 1d6 for a range of 4 to 9.