Quick Answer: Random sequences naturally contain runs, clusters, and apparent patterns — this is mathematically expected, not a sign of bias. In a truly random sequence of 200 coin flips, you will almost certainly see at least one run of 7 or more consecutive identical outcomes.
What Makes a Sequence "Random"?
A random sequence is one where each element is generated independently, with no information about previous elements influencing the next. Importantly: a random sequence will contain runs, clusters, and apparent patterns. In fact, a sequence that appears perfectly alternating (H-T-H-T-H-T) is statistically suspicious — true randomness produces clumping.
Human Perception vs Mathematical Reality
When researchers ask people to write out a "random" sequence of H and T without a coin, they consistently produce sequences with too many alternations and too few runs. The human-generated sequences look "more random" to human eyes but are actually statistically distinguishable from true randomness by their low number of consecutive streaks.
Sequence Probability Examples
| Sequence | Probability | Notes |
|---|---|---|
| HHHHH (5H) | 3.125% | Independent of history |
| HTHTH (alternating) | 3.125% | Exactly same probability as above |
| HHTTH (any specific sequence of 5) | 3.125% | All specific 5-flip sequences are equally likely |
| At least one run of 5+ in 100 flips | >97% | Runs are very common in long sequences |
The Clustering Illusion
The clustering illusion is our tendency to see patterns in small random samples that are simply due to chance. Basketball fans see a player who scores several consecutive baskets as 'hot' — but statistical analysis of professional basketball data shows that consecutive scoring is consistent with independent random probability. The same applies to coin flips and every other genuinely random process.