Probability guide

Why random results look uneven in small samples

Fair randomness describes the chance on each spin. It does not promise alternating results, equal short-run counts, or a correction after a streak.

Last updated July 27, 2026

Fair does not mean tidy

On a fair Yes or No Wheel, each spin has a 50% chance of YES and a 50% chance of NO. The wheel does not remember that YES appeared three times and make NO more likely next. If spins are independent, the next probabilities remain 50/50.

Streaks are part of randomness

The probability of four particular YES results in a row is 0.5 × 0.5 × 0.5 × 0.5, which equals 6.25%. That is uncommon but far from impossible. When people perform many sets of spins, some sets naturally contain noticeable streaks.

A system that carefully alternated YES and NO would look balanced, but it would not be independent randomness.

Small samples have wide variation

Ten fair spins do not have to produce five YES and five NO. Results such as six and four, seven and three, or even eight and two can occur without any change to the underlying probability. As the number of spins grows, the observed proportion often moves closer to the expected probability, but it can continue to fluctuate.

Weighted options behave the same way

An option with a 10% chance is expected to appear about ten times in one hundred spins over many repeated experiments. It can still appear zero times, twice in a row, or more than ten times in one particular set. “Expected” is a long-run average, not a quota.

The gambler’s fallacy

After several NO results, it may feel as if YES is “due.” That belief is the gambler’s fallacy. For an independent wheel, earlier outcomes do not create a debt that the next spin must repay. The probability changes only when the active options or weights change.

How to check the rule instead of the pattern

  • Confirm the active entries and weights before the test.
  • Use a predetermined number of spins rather than stopping at an interesting streak.
  • Record every outcome, not only surprising ones.
  • Compare results over many runs and expect ordinary variation.
  • For scientific, financial, or regulated work, use appropriate statistical tests and independently validated tools.

What a wheel can demonstrate

A browser wheel is useful for seeing probability in action and discussing why intuition often expects patterns that randomness does not provide. It is not evidence that a small observed sample proves or disproves the fairness of a system.

Explore the probabilities

Try equal outcomes with the Yes or No Wheel or compare visible percentages with the Weighted Wheel.