Dice Probability Explained: Odds, Distributions & the 2D6 Bell Curve
From the uniform distribution of a single die to the bell curve of two D6s — the math behind every roll, explained plainly.
Single Die: The Uniform Distribution
A fair six-sided die produces each face with equal probability of 1/6 (approximately 16.7%). This is called a uniform distribution — every outcome is equally likely. There is no memory in dice; a sequence of six sixes doesn't make a seven any more probable on the next roll. Each roll is an independent event.
For any fair N-sided die, the probability of rolling any specific face is 1/N. A D4 gives you a 25% chance per face. A D20 gives you a 5% chance per face. A D100 gives you a 1% chance per face. The expected value of a single die roll — the long-run average — is (N+1)/2. For a D6, that's 3.5. For a D20, that's 10.5.
Two Dice: The Bell Curve Emerges
When you roll two D6s and sum them, the distribution stops being uniform and becomes a triangle distribution (approximating a bell curve). There is only one way to roll a 2 (1+1) and one way to roll a 12 (6+6), but there are six ways to roll a 7 (1+6, 2+5, 3+4, 4+3, 5+2, 6+1). The 2D6 sum distribution across all 36 possible outcomes is:
| Sum | Combinations | Probability |
|---|---|---|
| 2 | 1 | 2.8% |
| 3 | 2 | 5.6% |
| 4 | 3 | 8.3% |
| 5 | 4 | 11.1% |
| 6 | 5 | 13.9% |
| 7 | 6 | 16.7% |
| 8 | 5 | 13.9% |
| 9 | 4 | 11.1% |
| 10 | 3 | 8.3% |
| 11 | 2 | 5.6% |
| 12 | 1 | 2.8% |
This is why 7 is the most important number in many board games — including Catan, where the robber triggers on a 7. Designing games around 2D6 sums exploits this natural probability distribution to create meaningful variation without requiring complex systems.
Expected Value
The expected value is the theoretical average result over a large number of rolls. For a single D6, it's (1+2+3+4+5+6)/6 = 3.5. For 2D6, it's 7. For a D20, it's (1+20)/2 = 10.5. Expected values are additive — rolling 3D6 has an expected value of 10.5 (three times 3.5). This principle underlies most game balance decisions involving dice.
Variance and Why It Matters
Variance measures how spread out the outcomes are from the expected value. A D20 has high variance — you'll frequently get results far from 10.5. Rolling 10D6 has nearly the same expected value (35) but far lower variance — you'll almost always land somewhere between 25 and 45. More dice rolled together create more consistent average outcomes, even if the total range is wider.
This is why tabletop game designers use multiple dice for predictable damage ranges and single dice for binary or high-variance outcomes. A 1D12 sword (range 1-12, high variance) feels wilder and less reliable than a 2D6 sword (range 2-12, average 7, lower variance). Same theoretical max, very different playing feel.
Advantage & Disadvantage in D&D
Dungeons & Dragons 5e introduced the advantage/disadvantage mechanic — rolling two D20s and keeping the higher (advantage) or lower (disadvantage) result. Rolling with advantage is mathematically equivalent to roughly a +4 or +5 bonus on most of the probability curve. The expected value of the highest of two D20s is approximately 13.82, compared to 10.5 for a single D20. For a detailed breakdown with probability tables, see the Advantage & Disadvantage guide.
The Law of Large Numbers
The law of large numbers states that as you roll more and more times, your observed frequency of each outcome will converge on the theoretical probability. Rolling a D6 ten times might produce no 4s at all — that's not unusual. Rolling it 10,000 times will produce close to 1,667 fours. This is why randomness feels "unfair" in short game sessions but averages out over long play.
For practical gaming: don't read into short streaks. A run of five bad rolls doesn't mean the next roll is due to be good — each roll is independent. Understanding this is the first step to interpreting your dice results accurately rather than emotionally.