No. 232 · Evidence · Statistics and Probability

Distributions

A single number hides the shape of the spread, and the shape often matters more.


A distribution describes the full spread of possible outcomes: where values cluster, how wide they range, and how often extremes appear. Two situations with the same average can have very different shapes, and the shape is often what matters for decisions.

To use it, ask about the range, not just the typical case. Plan your time, money and designs around the realistic spread, including the bad days. Gather several observations before judging, and remember that past patterns may not cover every future surprise.


In the Almanac

“Quetelet’s soldiers”, 1835–1846 · “ThirdLove”, 2013

Both stories in full, three ways to use it and where it misleads.

Get the Almanac

Listen · Watch

Distributions

Questions

What is Distributions?

A distribution describes the full spread of possible outcomes: where values cluster, how wide they range, and how often extremes appear. Two situations with the same average can have very different shapes, and the shape is often what matters for decisions.

How do you use Distributions?

To use it, ask about the range, not just the typical case. Plan your time, money and designs around the realistic spread, including the bad days. Gather several observations before judging, and remember that past patterns may not cover every future surprise.

When does Distributions matter?

Distributions is one of the 68 mental models for evidence, the situations where you ask: Is this really true? Within that, it belongs to “Reading the numbers”, alongside Base Rate Fallacy, Bayes’ Theorem and Central Limit Theorem.

What is an example of Distributions?

Two stories show it at work: Quetelet’s soldiers (1835–1846) and ThirdLove (2013). Both are told in full in The Mental Models Daily Almanac, which has two stories for each of its 365 models.

Which mental models are related to Distributions?

Fat-Tailed Distributions, Power Law and The Map is not the Territory. Distributions itself comes from statistics and probability.