Both roll from 1 to 12 — sort of — but one gives you a flat spread and the other a bell curve. If you design encounters or houserule damage, that difference is the whole game.
Open any RPG forum and someone is arguing that a greatsword (2d6) and a greataxe (1d12) "do about the same damage." On the average, they're not wrong — but average is the least interesting number a die produces. Roll 2d6 a hundred times and roll 1d12 a hundred times and you'll get two visibly different piles of results. One clusters. The other doesn't. If you design encounters, balance weapons, or just want to know why your barbarian feels swingy, the shape of the distribution matters far more than the mean.
Here's what's actually happening when those dice hit the table — and why "1 to 12" hides two completely different games.
1. One die is flat. Two dice are a hill.
A single twelve-sided die is a uniform distribution: every face from 1 to 12 is equally likely, each at exactly 1/12 (about 8.3%). A 1 is as probable as a 7 is as probable as a 12. The die has no opinion.
Two six-sided dice are a different animal. You're not rolling a number from 2 to 12 — you're rolling two numbers and adding them, and there are far more ways to make a 7 than a 12. Six combinations sum to 7 (1+6, 2+5, 3+4, and their mirrors); only one combination makes 12 (6+6). So a 7 lands 6/36 of the time (16.7%) while a 12 shows up 1/36 (2.8%). Plot it and you get a triangle peaking at 7 — the classic bell-ish curve every craps player has internalized.
2. The averages lie to you
2d6 averages 7. 1d12 averages 6.5. Half a point apart — which is exactly why people call them equivalent. But that half-point is the only thing the two have in common, and even it points somewhere interesting: 2d6 is the higher average despite the d12 being able to roll a clean 12 just as often as anything else.
The reason is that the d12 wastes a lot of its range on low rolls. One in six d12 rolls is a 1 or a 2. On 2d6, the odds of a total that low (a 2) are 1 in 36 — you almost never whiff that hard. The two dice "protect" each other: for both to come up tiny is rare. That's the practical meaning of a bell curve. Extreme outcomes get squeezed out toward the tails.
3. Consistency vs. swing — pick your feeling
This is the part that actually matters at the table. Variance is the spread of results around the mean, and the two dice could hardly be more different. The standard deviation of 1d12 is about 3.45; for 2d6 it's about 2.42. Lower spread means 2d6 hugs its average — most rolls land in the 5–9 band (that's about 67% of them), and you rarely see the extremes.
1d12 is pure swing. A third of your rolls are either 1–4 or 9–12. For a player, that's a slot machine: thrilling, streaky, occasionally devastating. For a designer, it's a budgeting nightmare, because the same weapon can chip for 1 or crater for 12 with equal ease. If you want a weapon that feels reliable, 2d6 is your die. If you want one that feels dangerous, reach for the d12.
4. Why designers care: the encounter-math problem
When you set a monster's hit points or a save DC, you're implicitly betting on a distribution. Say a creature has 8 HP and you want a single hit to "usually" drop it. With 2d6 you'll deal 8+ damage about 72% of the time — dependable. With 1d12 you clear 8 only 5/12 of the time (about 42%), because all those low faces drag you down. Same "1–12 weapon," wildly different kill reliability.
This is also why adding dice is the standard designer trick for taming randomness. 3d6 (range 3–18, mean 10.5) is tighter still — GURPS built its entire resolution system on it precisely because the bell curve makes skill matter more than luck. The more dice you sum, the more the result converges toward the center, an effect that's really just the central-limit intuition showing up on a tabletop.
"The probability distribution of the sum of two dice is not uniform... the most likely outcome, 7, is six times as likely as the least likely outcomes, 2 and 12."
— Wikipedia, "Dice" (Probability) (CC BY-SA 4.0)
5. When the flat die is the right call
None of this makes 1d12 a bad die — it makes it a different tool. A uniform roll is exactly what you want for a random table: 12 equally weighted treasure results, wandering monsters, or a chaos-magic surge where every outcome should be equally on the menu. The moment you want "all options equally likely," the bell curve is your enemy and the single die is your friend.
Flat dice also reward all-or-nothing builds. A great-weapon barbarian who triggers extra effects on a maximum roll wants a die that hits its ceiling often — and 1d12 maxes out 8.3% of the time versus 2d6's 2.8%. Mechanically, that's a meaningfully better crit-fishing platform, which is part of why later editions handed the d12 to exactly those characters.
The one-line takeaway
2d6 and 1d12 share a range and almost share a mean, and that's where the resemblance stops. 2d6 is consistency — a tight hill around 7, made for things that should feel dependable. 1d12 is variance — a flat field from 1 to 12, made for things that should feel risky or fair-by-equal-chance. Choosing between them isn't a rounding decision; it's a decision about how you want the moment to feel.
If you want to watch the shapes for yourself, our dice roller handles arbitrary dice notation (roll 2d6 and 1d12 side by side a few dozen times — the clustering is obvious fast). When you're tuning a weapon, the damage calculator works out averages and DPR across hit chances, and if you're balancing how fast those hits translate into levels, the XP curve calculator is the other half of the encounter-design picture.
← All articles