The two questions every dice roll answers
When you throw dice you are really asking two different things, and games blur them together so often that the difference gets lost. The first question is what number came up — the raw result you read off the faces. The second is how likely that number was — the probability that produced it. A tool like the Dice Roller hands you the first instantly; this guide is about the second, because understanding the odds is what turns a pile of plastic into a fair, predictable system you can reason about.
A single fair die is the simplest case. A six-sided die (a d6) has six faces, each equally likely, so every number from 1 to 6 comes up one time in six — about 16.7 percent each. A twenty-sided die (a d20) spreads that same certainty across twenty faces, so each number is a 5 percent shot. This is called a uniform or flat distribution: a bar chart of the odds is a row of equal-height columns. Nothing about the die favours a 1 over a 20. That flatness is exactly why a single die is such a good impartial picker — it is the same logic behind a coin flip or a decision wheel, just with more sides.
Why more dice changes everything
The interesting behaviour starts when you roll more than one die and add them together. Roll two d6 and the total can range from 2 to 12 — but those totals are not equally likely. There is only one way to roll a 2 (a 1 and a 1) and only one way to roll a 12 (a 6 and a 6), while there are six different ways to roll a 7 (1+6, 2+5, 3+4, 4+3, 5+2, 6+1). So a 7 is six times more likely than a 2. The flat distribution of a single die has become a triangle peaking in the middle.
Add still more dice and that triangle rounds into the familiar bell curve. Three d6, four d6, a fistful of dice — the more you sum, the more sharply the results cluster around the average and the rarer the extremes become. This is the central limit theorem doing its quiet work, and it has a very practical consequence for game design: a system built on many small dice (like rolling several d6 and adding them) produces consistent, predictable results, while a system built on one big die (like a single d20) produces swingy, dramatic results. Neither is better; they feel different. If you want to see this for yourself, open the Dice Roller, set it to roll 20d6, and watch how the total almost always lands near 70 — the average — even though the possible range is 20 to 120.
What a modifier actually does
Most tabletop systems let you add a flat number to a roll: 1d20 + 5, or 2d6 − 1. It is tempting to think a modifier “improves your odds,” but that is not quite what happens. A modifier shifts the entire distribution by a fixed amount without changing its shape. If 1d20 gives you an even 5 percent chance of each number from 1 to 20, then 1d20 + 5 gives you that same flat 5 percent chance of each number from 6 to 25. Every outcome moved up by five; none became more or less likely relative to the others.
Where a modifier does change your chances is against a fixed target. If you need to roll 15 or higher on a d20, you have a 30 percent chance (results 15 through 20, six faces out of twenty). Add a +5 and now you succeed on a raw 10 or higher — eleven faces out of twenty, or 55 percent. The modifier did not bend the die; it lowered the bar. This is the crucial mental model: modifiers matter relative to a threshold, not in the abstract. The Dice Roller shows the modifier separately from the dice total and reports the possible range, so you can always see both the raw roll and where the bonus landed you.
Reading a multi-die roll correctly
A roll written as 2d6 + 1d20 + 1 looks like notation, but it is just a recipe: roll two six-sided dice, roll one twenty-sided die, add them all together, then add one. The convention is always [number]d[sides], so 3d8 means three eight-sided dice and 1d100 means one percentile die. When you combine different dice types in a single throw, each group keeps its own odds — the d20 is still swingy, the pair of d6 is still clustered — and the grand total inherits a blended shape. Good rolling tools, including ours, show each individual die face and each group’s subtotal rather than only the final number, so you can verify the arithmetic or read a specific die when a rule calls for it.
The theoretical minimum and maximum are worth knowing too. For any combination, the minimum is every die showing 1 plus the modifier, and the maximum is every die showing its highest face plus the modifier. For 2d6 + 1d20 + 1 that is a minimum of 4 and a maximum of 33. Knowing the range tells you instantly whether a target is even reachable — no modifier in the world makes a 2d6 roll hit 15.
Common misconceptions
The biggest is the gambler’s fallacy: the belief that after several low rolls a high one is “due.” Dice have no memory. Each roll is independent, so a d20 that just produced three 1s in a row is exactly as likely to roll a 1 again as it ever was. Streaks feel meaningful but are simply what randomness looks like up close.
The second is confusing average with typical. The average of 1d20 is 10.5, but you will rarely roll exactly the average — every face is equally likely, so 1 and 20 are just as common as 10 or 11. Averages only become reliable predictions when you sum many dice, which is precisely why bell-curve systems feel steadier.
The third is assuming a digital roller is somehow less random than physical dice. In practice the opposite is often true: real dice are subtly weighted by manufacturing, wear, and how they are thrown, while a software roller draws uniformly from a clean random source on every face. The selection logic in the Dice Roller is a small, pure function, so the odds are exactly even by construction.
Putting it to use
Once you understand that one die is flat, many dice cluster, and a modifier slides the range, you can choose the right dice for the job. Want a fair, dramatic single pick? Roll one die. Want a steady, predictable total? Roll several and add them. Need a bonus to clear a target? Apply a modifier and check it against the range. For a hands-on feel, try a few combinations in the Dice Roller and watch the result card, and if you are after a different flavour of fair randomness — picking from a named list rather than a number — the companion decision wheel guide covers when a spin beats a roll.