Fairness

Fairness describes a game or draw in which every possible outcome receives exactly the probability that the rules announce, without any hidden mechanism favouring one result over another. A fair device is said to be "unbiased": its long-term behaviour faithfully matches the theoretical model.

The intuition is simple. If a drawing object physically favours no position, then all the outcomes it can produce are equally likely. A coin whose two sides are symmetrical, a die whose faces are identical, a wheel whose sectors are the same size: nothing in their construction pushes toward a particular answer. Fairness is precisely this absence of leaning.

It is verified by comparing observed frequencies with expected probabilities. For a fair six-sided die, each face has a probability of 1/6, or about 0.1667. Over a very large number of rolls, the share of each face should approach this value; a lasting and marked deviation would betray a bias. On the scale of a single session, however, apparent deviations are normal and prove nothing.

A common misconception is the belief that a fair draw should "spread out" its results evenly, as if it avoided repetitions. This is not so: fairness concerns the probability of each outcome, not the order in which they appear. Three heads in a row with a balanced coin are perfectly compatible with fairness, because each toss remains independent of the previous ones.

On this site, fairness is the design goal of every drawing tool: to produce results you can rely on because they favour no one. Understanding this notion also helps you reason about chance responsibly, distinguishing what an honest device actually guarantees from what is wrongly attributed to it.