Chance Has No Memory

Past results don't change the odds of an independent event.

By the Math Says Yes editorial team

Human-reviewed under our source and correction standards.

Separate chance events shown as unconnected spins and dots.

Independence resets the trial

An independent event is not influenced by the outcomes before it. A fair coin does not know it just landed heads five times. A die does not remember which side has been missing. The next trial uses the same probabilities as any other trial. Past outcomes can be useful for checking whether the process is fair, but if the process really is independent, they do not create a debt that future outcomes must repay.

The gambler's fallacy

The gambler's fallacy is the belief that a streak makes the opposite result more likely. After several reds at roulette, black feels due. After many missed shots, a make feels owed. This mistake comes from applying long-run balance to a short-run sequence. Long-run frequencies can settle near their expected rates without any single future spin, flip, or shot being pushed in the opposite direction.

Long-run balance is not short-run correction

If you flip a fair coin many times, the proportion of heads tends to move toward 50 percent. That does not happen because tails becomes more likely after heads. It happens because new independent flips keep adding ordinary random variation, and early streaks become a smaller part of the total. The absolute difference between heads and tails can even grow while the proportion gets closer to half.

Check whether independence really holds

The lesson is not that history never matters. It matters when outcomes are not independent. A tired athlete, a changing machine, a biased shuffle, or a learning algorithm can make past results informative. The key is to ask whether the mechanism carries memory. If the process is physically or behaviorally connected across trials, update your belief. If it is genuinely independent, do not treat a streak as a promise of reversal.

FAQ

Does a long streak make the opposite result more likely?

Not if the trials are independent. A fair coin, die, or roulette spin keeps the same probabilities regardless of the previous streak.

When should past outcomes change my prediction?

Past outcomes matter when the process has memory, fatigue, learning, mechanical change, or bias. They should not change the odds of a truly independent trial.

Core Concepts

Independence

Independent events do not influence each other: the probability of the next outcome is the same no matter what came before. A coin, a die, and a roulette wheel have no memory of their past results.

See it in action

A roulette wheel has no memory

Independent events don't balance themselves out in the short run. A streak changes how the wheel feels, not the odds it actually pays.

Try the idea

Spin the wheel

Bet the colour that feels “due” and watch what the streak is really worth.

—

Place a bet and spin to start a streak.

—

“due” colour won

—

longest streak

—

total spins

Red

Black

After a long run of one colour, the other feels overdue — but the wheel has no memory. Over many spins the “due” colour wins about 48.6% of the time, exactly its baseline. The streak never moved the odds.

Quick Check

A fair European roulette wheel just landed on red ten times. What is the chance the next spin is red?

Higher — red is on a hot streak

About 48.6% — the same as every spin

Much lower — black is overdue

Exactly 50%