A distribution is a complete list
A probability distribution is simply a full account of what can happen and how likely each result is. It pairs every possible value of some random quantity with a probability, and the probabilities add up to 1.
A discrete distribution covers values you can list: die faces, the number of heads in some coin flips, the rarity tier of a loot drop. (Quantities that vary smoothly, like a measured height, need a continuous version, such as the normal distribution.)
The word “distribution” can sound technical. Think of it as a table with two columns: outcome and chance.
The uniform distribution
If every outcome is equally likely, the distribution is uniform. A fair six-sided die is the standard example:
| Face | 1 | 2 | 3 | 4 | 5 | 6 |
|---|---|---|---|---|---|---|
| Probability | 1/6 | 1/6 | 1/6 | 1/6 | 1/6 | 1/6 |
Six probabilities of 1/6 sum to 1, as required. Any fair die with s sides is uniform with probability 1/s per face, a fact that extends to d4 up to d20 polyhedral dice. A shuffled deck is uniform for a single card too: each of the 52 cards has probability 1/52.
Uniform is the baseline most people assume, and it is the right model only when the device is genuinely fair. A die that is physically biased, or a game that deliberately weights outcomes, has a different distribution.
A non-uniform example from dice
Sum two dice and the totals are no longer equally likely. Count the ordered pairs (out of 36) that make each total:
| Total | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 |
|---|---|---|---|---|---|---|---|---|---|---|---|
| Ways | 1 | 2 | 3 | 4 | 5 | 6 | 5 | 4 | 3 | 2 | 1 |
| Probability | 1/36 | 2/36 | 3/36 | 4/36 | 5/36 | 6/36 | 5/36 | 4/36 | 3/36 | 2/36 | 1/36 |
The ways add to 36, so the probabilities add to 1. The shape rises to a peak at 7 and falls symmetrically. The inputs (each die) are uniform, but their sum is not. This is one of the most useful lessons about distributions: combining uniform pieces tends to produce a peak in the middle, which is part of what game designers consider when they choose dice distributions for a rule system.
Weighted outcomes
A distribution can be built to order. Imagine a loot table (an illustration, not a real game’s figures):
| Tier | Common | Rare | Epic |
|---|---|---|---|
| Probability | 0.70 | 0.25 | 0.05 |
The three add to 1.00. This is a weighted random table: each tier has its own chance, and the designer chooses the weights. Checking that the probabilities total 1 is the first thing to do with any such table, since an error there will quietly distort every result.
From a distribution you can compute averages. If the tiers were worth 1, 5 and 20 points, the expected value would be 0.70 × 1 + 0.25 × 5 + 0.05 × 20 = 0.70 + 1.25 + 1.00 = 2.95 points per drop.
The binomial distribution
Many game questions have the form: repeat the same independent trial n times, each with success chance p, and count the successes. The distribution of that count is binomial.
Illustration: flip a fair coin 3 times and count heads.
| Heads | 0 | 1 | 2 | 3 |
|---|---|---|---|---|
| Ways (of 8) | 1 | 3 | 3 | 1 |
| Probability | 1/8 | 3/8 | 3/8 | 1/8 |
The probabilities sum to 8/8 = 1. Notice the shape again: the middle values are the most likely, and the extremes are rare. That bunching toward the centre becomes more pronounced as n increases, and for large n the outline approaches the bell shape of the normal distribution. The full formula and more examples are in the dedicated article on the binomial distribution.
A binomial model needs four things: a fixed number of trials, only two results per trial (success or failure), the same success chance each time, and independence. Drawing cards without putting them back violates the last two, and a different model applies.
What a distribution lets you read off
Once you have a distribution you can answer many questions:
- The chance of a single value, by reading it from the table.
- The chance of a range, by adding the relevant rows. For two dice, the chance of a total of 10 or more is 3/36 + 2/36 + 1/36 = 6/36 = 1/6.
- The expected value, by multiplying each value by its probability and adding.
- The spread, using variance or its square root, the standard deviation.
Two distributions can have the same average and look nothing alike. A game with outcomes concentrated near the average feels steady, while one with the same average but many extreme results feels swingy. That contrast is the basis for discussions of game volatility.
Distribution versus what you see
A distribution describes long-run chance, not a promise for a short stretch. Rolling a fair die 12 times will rarely give each face exactly twice, even though the distribution is uniform. Over very many rolls the observed proportions settle closer to 1/6, which is the law of large numbers, but individual runs stay unpredictable. Expecting short sequences to mirror the distribution is the root of the gambler’s fallacy.
Using distributions responsibly
Distributions are descriptive tools. They help explain why some results are common and others rare, and how a game’s rules shape what players experience. They do not provide a way to beat a game with a built-in house edge. Outcomes are uncertain, and gambling can be harmful; if it is affecting you or someone you know, local support services are available.