Clustering Illusions
How genuine randomness produces clumps, gaps, and streaks, and why we so often read them as meaningful or as “not random.”
Random Streaks
In Part 2, you tried to imitate randomness. Now let chance do it. Before we flip 50 fair coins, predict how long you think the longest uninterrupted streak will be.
Part 3: What Does a Fair Coin Actually Do?
In 50 fair coin flips, how long do you think the longest streak will be?
Your prediction:
Sequence continues left-to-right across rows.
· ·
Highlight runs of:
Longest streak in 1,000 random sequences
| Longest streak | Observed | Exact probability |
|---|
Threshold:
Streaks are not the opposite of randomness. They're part of it. In only 50 fair coin flips, five in a row happens about 82% of the time, and six in a row happens more often than not. A streak can feel like evidence that something changed, but first we need to ask how often a streak that large would show up by chance alone.
Randomness makes streaks. The question isn't whether a streak appeared; it's whether the streak is unusual under the right chance model.
Important caveats
A streak does not prove the odds changed. Fair coins produce streaks on their own just fine. But a streak does not prove nothing changed either: a process whose odds really do shift over time can also produce streaks. The real question is always whether the data fit one explanation better than another, not whether a streak happened at all.
This exhibit models 50 independent fair-coin flips with the odds fixed at 50% throughout. It doesn't try to judge whether any specific real-world streak (a winning streak in a game, a market swing, a run of good outcomes in a treatment) reflects a real change in what's driving it. That takes a model and evidence specific to that situation. Later statistical work also found a subtle flaw in how some influential early studies measured real-world "hot hand" streaks in sports, so this exhibit doesn't claim every apparent hot streak is imaginary.
Research behind this
The math is exact. The longest-run probabilities above come from a calculation that checks every possible 50-flip sequence, not a simulation. The experimental findings are about human intuition. People tend to expect shorter runs and more switching than a real coin actually produces, and the third repeat in a row is usually where a sequence starts feeling like a "streak" to an observer. The wider interpretation stays deliberately narrow here: this exhibit shows how streaks behave under one clear-cut random process, and it doesn't weigh in on any specific real-world claim of a hot streak or a change in the underlying odds.
- Carlson, K. A., & Shu, S. B. (2007). The rule of three: How the third event signals the emergence of a streak. Organizational Behavior and Human Decision Processes, 104(1), 113–121. doi:10.1016/j.obhdp.2007.03.004
- Bar-Hillel, M., & Wagenaar, W. A. (1991). The perception of randomness. Advances in Applied Mathematics, 12(4), 428–454. doi:10.1016/0196-8858(91)90029-I
- Falk, R., & Konold, C. (1997). Making sense of randomness: Implicit encoding as a basis for judgment. Psychological Review, 104(2), 301–318. PDF
- Nickerson, R. S. (2002). The production and perception of randomness. Psychological Review, 109(2), 330–357. doi:10.1037/0033-295X.109.2.330
- Schilling, M. F. (1990). The longest run of heads. The College Mathematics Journal, 21(3), 196–207. doi:10.1080/07468342.1990.11973306
- Miller, J. B., & Sanjurjo, A. (2018). Surprised by the hot hand fallacy? A truth in the law of small numbers. Econometrica, 86(6), 2019–2047. doi:10.3982/ECTA14943
Further reading: The Gambler's Fallacy on Wikipedia.