Dice Probability Explained: Why 7 Wins and How Dice Odds Really Work

If you have ever sat at a board game and wondered why the number 7 keeps showing up, or argued with a friend about whether you are "due" for a good roll, you have already bumped into dice probability. The short answer is that dice are not as random as they feel. Every roll obeys a fixed set of counts, and once you can count, you can predict. This guide walks through the counting, the famous bell shape, and the difference between "exactly" and "at least" odds that trips up so many players.

Two dice, thirty-six outcomes

Start with a single die. A fair six-sided die has six faces, each equally likely, so the chance of any one number is 1 in 6. That part is flat and boring. The interesting math starts the moment you add a second die.

When you roll two six-sided dice, the first die can land six ways and the second can land six ways, so there are 6 × 6 = 36 equally likely outcomes. Not 11, not 12 — thirty-six. The mistake people make is counting the sums (2 through 12, which is 11 totals) and assuming each total is equally likely. It is not. Each of the 36 face pairs is equally likely, but those pairs crowd into the middle sums.

Here is the concrete point. A sum of 7 can be made six different ways: 1+6, 2+5, 3+4, 4+3, 5+2, and 6+1. So the probability of rolling 7 is 6/36, which simplifies to 1/6, about 16.67%. Now look at the extremes. A sum of 2 happens only one way (1+1), and a sum of 12 happens only one way (6+6). Each is just 1/36, about 2.78%. That is why 7 is six times as likely as snake eyes. There is no luck involved — there are simply more roads that lead to 7.

Why 7 is the most likely sum

Every sum has a count: the number of dice combinations that produce it. For two six-sided dice the counts climb to a peak at 7 and fall off symmetrically on both sides.

| Sum | Ways | Probability | |----|------|-------------| | 2 | 1 | 1/36 ≈ 2.8% | | 3 | 2 | 2/36 ≈ 5.6% | | 4 | 3 | 3/36 ≈ 8.3% | | 5 | 4 | 4/36 ≈ 11.1% | | 6 | 5 | 5/36 ≈ 13.9% | | 7 | 6 | 6/36 ≈ 16.7% | | 8 | 5 | 5/36 ≈ 13.9% | | 9 | 4 | 4/36 ≈ 11.1% | | 10 | 3 | 3/36 ≈ 8.3% | | 11 | 2 | 2/36 ≈ 5.6% | | 12 | 1 | 1/36 ≈ 2.8% |

Add up the Ways column and you get 36 — every outcome accounted for. The number 7 sits dead center because it is the only sum reachable by every face of the first die: whatever the first die shows, there is always exactly one value on the second die that completes a 7. No other sum has that property. This is the math behind craps and behind the robber in Settlers of Catan always seeming to camp on the 6 and 8 tiles.

The bell shape, and what happens with more dice

That climb-and-fall pattern is the start of a bell curve. One die is flat. Two dice make a triangle peaking in the middle. Three dice round the triangle off into a gentle mound. By the time you roll ten dice, the sum distribution looks almost exactly like the normal curve from a statistics textbook, centered on the average and tapering to thin tails.

The reason is simple to state: there are many ways to make a middling total and very few ways to make the minimum or maximum. To roll the lowest possible sum on ten six-sided dice you need every single die to show a 1 — one path out of millions. To roll something near the middle, countless combinations work. The middle wins because the middle is crowded.

This matters for game design. Switching a movement roll from 2d6 to 1d12 keeps the same range, roughly, but flattens the curve completely. With 2d6 a player reliably moves around 7 spaces; with 1d12 every distance from 1 to 12 is equally likely and the game feels swingier. Designers reach for multiple dice precisely when they want consistency, and a single die when they want chaos. If you want to see these shapes side by side, the probability distribution visualizer plots them directly.

Exactly, at least, and at most

The single biggest source of confusion in dice odds is the comparator. "What are the odds of a 7?" usually means exactly 7. But "do I beat a difficulty of 7?" means at least 7. These are very different numbers.

Worked example, using two six-sided dice:

Probability of rolling exactly 7: count the ways to make 7 (six of them) and divide by 36. That is 6/36 = 1/6 ≈ 16.67%.
Probability of rolling at least 7: add the ways to make 7, 8, 9, 10, 11, and 12. That is 6 + 5 + 4 + 3 + 2 + 1 = 21 ways out of 36, so 21/36 ≈ 58.3%.

The "at least" answer is almost four times bigger than the "exactly" answer, even though both questions mention the number 7. A saving throw worded "DC 7 or higher" is an at-least question; treating it like an exactly question would badly underestimate your odds.

A second worked example shows a different kind of question. What is the probability of rolling at least one 6 on two dice? It is easier to count the failures: the chance a single die is not a 6 is 5/6, so the chance neither die is a 6 is (5/6) × (5/6) = 25/36. Subtract from 1 and the chance of at least one 6 is 11/36 ≈ 30.6%. The trick of counting the opposite and subtracting saves a lot of arithmetic, and it generalizes to any "at least one" question.

I learned the practical value of this the hard way during a long campaign I was running. I had set a trap's disarm check at "roll 15 or higher on 3d6" and casually told my players it was a coin flip. It is not. Three six-sided dice produce 216 outcomes, and only 20 of them total 15 or more, so the real chance is about 9.3% — closer to one in eleven than one in two. My party walked into that trap believing they had even odds, and my mistake was treating a steep tail of the bell curve as if the distribution were flat. Ever since, I count before I quote.

From counting by hand to letting a tool do it

Counting 36 outcomes by hand is fine. Counting the 216 outcomes of three dice is tedious but doable. Counting the outcomes of ten dice, where the totals run into the millions, is hopeless on paper. That is exactly where a calculator earns its place: it enumerates every combination, reports the true fraction rather than a noisy simulation, and highlights the bar you asked about.

The dice probability calculator does this for any combination from a single d6 up to twenty dice with a hundred faces. You set the number of dice and the number of sides, pick whether you want "sum equals," "at least," or "at most," and it returns the exact odds as both a percentage and a fraction, alongside the expected value and the full distribution chart. Because it counts every outcome instead of rolling thousands of times, 2d6 sum = 7 comes back as precisely 6/36 every run, never 16.6% one time and 16.9% the next.

A few takeaways worth keeping at the table. The expected sum of N dice with S sides is N × (S+1)/2, so two six-sided dice average 7 and three average 10.5. The middle of the range is always the most likely outcome once you use two or more dice. And the comparator you choose — exactly, at least, at most — changes the answer more than the target number does. Get those three ideas straight and most dice arguments end with a real number instead of a shrug.

Made by Toolora · Updated 2026-06-13