6-Sided Dice Probability Calculator

How 6-sided dice probability works

Dice probability is counting. A fair six-sided die has six equally likely outcomes — 1, 2, 3, 4, 5, and 6 — so the probability of any single face is 1/6, about 16.67%. When you roll n dice, every die is independent: no die remembers or influences any other. By the multiplication principle, the number of equally likely ordered outcomes is 6ⁿ — 36 for two dice, 216 for three, 7,776 for five, and 60,466,176 for ten. Order matters here: (1, 6) and (6, 1) are two different outcomes, and that is exactly why 7 is the most common two-dice sum while 2 and 12 are the rarest.

The probability of any event is then a simple fraction:

The hard part is counting the favorable outcomes correctly. For a sum, the counts follow the convolution of uniform distributions, which is why the distribution of sums is always bell-shaped, peaking at the mean. The mean of one die is (1+2+3+4+5+6)/6 = 3.5, so the mean sum of n dice is 3.5n (7 for two dice, 10.5 for three). The variance of one die is 35/12 ≈ 2.9167, and variances of independent dice add, giving 35n/12 — a standard deviation of about 2.42 for two dice and 2.96 for three. For many dice the distribution approaches a normal curve with that mean and standard deviation, which is why extreme sums become vanishingly rare: rolling 50+ on ten dice happens only 174,746 times out of 60,466,176, about 0.29%.

For a specific face (for example, "at least two sixes"), each die is a yes/no trial with success probability 1/6, so the count follows the binomial distribution:

"At least k" just sums the exactly-k, exactly-(k+1), …, exactly-n terms. Odds are the same information in a different format: a probability of 1/6 is "1 in 6" odds, while 763/3888 (≈19.62%) is "1 in ~5.10". One classic trap is the gambler's fallacy: after five rolls without a six, the next roll is still 1/6 — dice have no memory, and "due" numbers do not exist. This calculator counts exact outcomes with integer arithmetic, so its fractions (like 763/3888) are mathematically exact, not Monte-Carlo estimates. The four modes above map directly onto these ideas: the Sum tab counts ordered outcomes for exact, at-least, at-most, and between conditions; the Specific faces tab applies the binomial formula; the Matching tab handles all-same and all-different patterns; and the Distribution tab renders the full bell curve with its exact counts.

Two-dice (2d6) probability table

All 36 ordered outcomes of two six-sided dice, with exact counts and probabilities:

The counts 1-2-3-4-5-6-5-4-3-2-1 form a triangle peaking at 7 — the most likely sum.

Three-dice (3d6) probability table

All 216 ordered outcomes of three six-sided dice:

Three dice have twin peaks: 10 and 11 are equally most likely at 27/216 = 12.50% each.

Worked examples

Practical tips

• Use "at least" for thresholds. Game rules usually say "roll 8 or higher" — that is the at-least mode, not the exact mode.
• Count ordered pairs. When estimating by hand, remember (2,5) and (5,2) are distinct. Forgetting this is the most common dice-probability mistake.
• 7 is king of 2d6, 10/11 rule 3d6. Design or bet around the peaks: 7 at 16.67% for two dice; 10 and 11 at 12.50% each for three dice.
• Convert to "1 in N" for intuition. A 2.78% chance is easier to feel as "1 in 36". The calculator shows odds next to every result.
• Don't chase "due" numbers. Every roll is independent — a six is always 1/6 no matter what came before.
• For many dice, think normal. With 6+ dice the sum distribution is close to a bell curve with mean 3.5n and standard deviation √(35n/12).