A dice roller example shows the exact output a virtual die produces — the face that landed on top, the running total across multiple dice, and a short rolling history that lets you review recent rolls at a glance. A typical sample might display a d20 landing on 14, a 3d6 result of 11 spread across three faces such as 3, 4, and 4, or a custom d100 reading 73. Each individual number appears next to its die, the combined total sits beneath the row, and the recent history stacks up below the totals so you can confirm what was actually rolled at any time. Because every roll is generated client-side using crypto.getRandomValues(), each face of the die has a statistically equal chance of appearing, and the visible history reflects what was rolled rather than a re-run of a biased script. The sample outputs below walk through what a single-die roll, a multi-die roll, and a custom-sided roll actually look like inside the Dice Roller, and how the on-page history plus session statistics behave once you start repeating rolls to build up a real distribution.
What a Dice Roller Result Looks Like
When you tap a die type and press the Roll button, the tool returns three pieces of information for every example run. First, the individual face values appear in the order the dice were rolled, so a five-die sample might read 6, 2, 5, 1, 4. Second, the running total is shown underneath the row — in that case 18. Third, the roll is appended to a short history panel so you can scroll back through the last several rolls and compare the numbers at a glance.
On top of the per-roll output, the tool keeps a slim session statistics panel that updates as you go. You typically see the number of rolls in the current session and the average total across those rolls. None of those numbers are stored on a server — they live in the page memory only — so closing the tab clears them. This is the entire shape of a dice roller example: per-roll numbers, an immediate total, a small history list, and a running summary that builds up as you keep rolling.
A Sample d20 Run in Practice
The d20 is the most common die for tabletop role-playing games, so it is also the easiest format to use as a worked example. With the d20 button selected and a single die in the queue, a few sample reads in a row might look like 14, 7, 19, 11, 4, 20, 8, 13. Each of those is a single number from 1 to 20, equally likely because the source is a uniform CSPRNG.
If you raise the count to two dice and roll 2d20 ten times, a sample run could read like this: (12, 7) total 19, (3, 20) total 23, (16, 16) total 32, (9, 1) total 10, (14, 11) total 25, (20, 18) total 38, (5, 9) total 14, (17, 6) total 23, (11, 13) total 24, (8, 15) total 23. The two individual faces are displayed, the total sits below, and the history now shows ten rows you can scroll back through. Once you reach a few dozen rolls of this kind, the session average typically settles near the expected sum of about 21 for two 20-sided dice, which is twice the mean of 10.5 for a single uniform die from 1 to 20.
Notation Patterns Worth Recognising
The shorthand used to describe dice rolls shows up everywhere dice are mentioned, so a quick way to read an example is to recognise the notation. The format "XdY" means "roll X dice, each with Y sides". So 2d6 means two six-sided dice, 3d8 means three eight-sided dice, and 1d20 means one twenty-sided die. A "mod" or modifier is sometimes added or subtracted, as in 2d6+3, which means the sum of two d6 plus three. Recognising those few patterns turns almost any dice roller example into something readable at a glance.
| Notation | Reading | Typical use |
|---|---|---|
| 1d20 | One twenty-sided die | D&D 5e attack rolls, saving throws |
| 1d6 | One six-sided die | Board games, small damage rolls |
| 2d6 | Two six-sided dice | Movement in many board games, craps |
| 3d6 | Three six-sided dice | Classic stat generation in old-school RPGs |
| 4d6 | Four six-sided dice | D&D 5e ability score generation (drop lowest) |
| 1d100 | One hundred-sided die | Percentile checks, random tables |
| 1dN | Custom number of sides | Exotic shapes (d3, d24, anything from 2 to 1000) |
The Custom option in the Dice Roller covers the "1dN" case for any number of sides from 2 to 1000, so unusual shapes such as d3, d24, or d100 all sit behind the same control. For a broader notation reference, the dice roller cheat sheet for notation and quick odds covers the same patterns in more depth.
How to Roll an Example Set
If you want to produce an example roll of your own, the on-screen flow is the same every time. The controls are all visible on the page, and the tool runs entirely in your browser, so nothing is uploaded and no account is required.
- Pick a dice type by tapping d4, d6, d8, d10, d12, or d20 — or choose Custom and enter any number of sides from 2 to 1000.
- Set the number of dice using the − and + stepper. The page accepts up to 12 dice per roll.
- Press the Roll button to generate a fresh sample. Each individual face value appears in order, with the total directly underneath.
- Read the roll history below the totals to check prior rolls in the current session.
- Check the session statistics to see how many rolls you have made and the running average of totals.
A short test that demonstrates the full flow is: tap d6, set the count to three, press Roll four or five times in a row, then read the history and average. Within a few rolls you should see totals clustered around the midpoint, which matches the expected behaviour of summed dice.
Sample Rolls for Common Game Settings
The shape of an example roll changes with the game you are playing, because each game reads dice totals differently. A board game player only needs a few fair d6 results; an RPG player is interested in a d20 attack roll plus a damage roll of a different size; a teacher wants enough rolls to show how a distribution looks. The tool supports all of these without changing screens, because Custom lets you swap face counts and the stepper lets you swap the count.
| Setting | Dice typically used | How the example output helps |
|---|---|---|
| Classic board games | d6 | Replaces lost dice mid-game with a fair face-by-face read |
| D&D 5e and Pathfinder | d4, d6, d8, d10, d12, d20 | Handles attack rolls, saves, and damage of every size |
| Percentile systems | d100 | Generates a uniform 1–100 outcome for random tables |
| Probability lessons | Any dice type | History doubles as a live distribution demo |
| Two-option decisions | Custom 2 sides (coin flip) | Provides a clean random yes or no |
For RPG combat the typical example flow is to roll 1d20 for the attack, then a smaller die for damage. The Dice Roller handles both back-to-back on the same page, so you can read the attack roll, swap die types in a couple of taps, and roll the damage as a separate sample without leaving the screen.
Reading the History and Session Averages
The history list is a running log of the most recent rolls in your current session. It does not look like a graph, but it acts like one: as you stack up rolls, the rows of numbers reveal the distribution by eye. Once the list has a dozen or so entries, you can spot clustering around the centre for summed dice and a roughly even spread for a single die.
The session statistics panel summarises the same information. The most useful fields are the rolling count and the mean of all totals you have produced. If you keep rolling 1d6 with a fresh sample each time, the rolling mean should drift toward 3.5, which is the expected average of a uniform distribution from 1 to 6. That convergence is itself a small example of the law of large numbers — and the panel shows it in real time.
Why Fair Randomness Matters in Any Example
An example only carries weight if it is reproducible — and reproducible only means anything if the source is unbiased. The Dice Roller uses the browser's built-in crypto.getRandomValues() function, which is a cryptographically secure pseudo-random number generator (CSPRNG) designed to produce statistically uniform output. Compare that to a plain Math.random() call: it is fast, but is not guaranteed to be uniform across engines and is not designed to be unpredictable, which is why quick scripts are not what you want when a game or lesson depends on a fair die.
The CSPRNG claim is testable in a few minutes of rolling. Two six-sided dice have 6 × 6 = 36 equally likely combinations when rolled together, and six of those combinations sum to 7 — namely 1+6, 2+5, 3+4, 4+3, 5+2, and 6+1. That means 6 ÷ 36 = 1/6 ≈ 16.67% of all 2d6 rolls should land on 7 if the dice are fair, which is why 7 is the most likely total in craps and is the canonical "middle of the bell curve" example. Inside the tool, rolling 2d6 several hundred times will see 7 come up close to 1/6 of the time, with totals near 2 and 12 appearing roughly 1/36 of the time each. The same idea extends to every other die type, which is why the history panel is also a quiet probability lesson.
Putting all of this into a single example, a sample session for a teacher who wants to show convergence might look like 50 rolls of 2d6 logged in sequence, with the average drift visible after every batch. The samples you produce by clicking Roll are the raw data; the history and session statistics do the rest of the explanation. As long as the source is fair — and the CSPRNG source keeps it that way — the example you show is also a fair representation of what a real roll would look like. Open the Dice Roller and run a few sample sets of your own to see the pattern.