Representativeness heuristic
Also called representativeness
Judging probability by how closely something resembles a familiar pattern, which leads people to draw firm conclusions from small samples, such as a handful of winning trades.

- Level
- Intermediate
- Reading
- 2 min
- Sources
- 2ResearchRegulator
- Updated
On this page· 5 sections
Four wins in five trades with a new setup feels like a pattern. The small run looks like what a good strategy would produce, so it is taken for one.
That reasoning is the representativeness heuristic: judging how likely something is by how closely it resembles a familiar picture[1].
How it works
In their 1974 paper, Amos Tversky and Daniel Kahneman described three judgment heuristics: representativeness, availability and anchoring. Under representativeness, people judge the likelihood of an uncertain event by how well it represents the process that generates the data[1].
The heuristics are economical and usually effective. They also produce predictable biases[1].
The UK Financial Conduct Authority (FCA) lists a version of this under the name over-extrapolation, also called the law of small numbers. People make predictions from only a few observations, implicitly believing that those observations are representative and show real patterns[2]. As a result, they also underestimate uncertainty[2].
The FCA’s examples sit close to trading. People predict future stock returns from a limited number of past returns, and they rate financial advice as good on the basis of a few successful investments, even though these could reflect pure luck[2].
Example
Say a setup wins half the time in the long run, like a fair coin. A trader tries it five times and wins four.
- The chance of exactly 4 wins in 5 is 5 ÷ 32. The chance of 5 wins is 1 ÷ 32.
- The chance of 4 or more wins is 6 ÷ 32, or 18.75%.
- About one in five traders with a coin-flip setup would see a result that good.
A record of 4 wins in 5 looks like an 80% setup. Five trades are too few to tell it apart from 50%.
Watch out
A run that looks like skill is not evidence of skill by itself. The FCA says a few successful investments could reflect pure luck[2].
The heuristic is not always wrong. Tversky and Kahneman call such heuristics usually effective[1].
Small samples hide uncertainty. Predictions made from only a few observations lead people to underestimate it[2].
Sources
- ResearchUniversity of Pennsylvania (Eugene Garfield library)This Week's Citation Classic: Judgment under uncertainty: heuristics and biases (Tversky & Kahneman, Current Contents, 4 April 1983, on their Science 185:1124-31, 1974)Checked November 9, 2026
- RegulatorFinancial Conduct AuthorityApplying behavioural economics at the Financial Conduct Authority (Occasional Paper No. 1, April 2013)Checked November 9, 2026
