science & techWikipedia18th century

Bayes' theorem lets you flip a probability around to find the cause from the effect

the gist

Bayes' theorem is a rule for reversing conditional probabilities. It lets you work out the chance of a cause from the effect you observed. A medical example: if a test is positive for a disease, the theorem combines how accurate the test is with how common the disease is, to show how likely the patient really has it. Thomas Bayes developed it in the 18th century, and Pierre-Simon Laplace independently developed the same idea.

how it works

The theorem multiplies the prior chance of a cause by how likely the effect is if that cause is present, then divides by the overall chance of the effect. Starting from how common something is in the population, new evidence updates the estimate step by step.

so what

The theorem underpins modern statistics, spam filters, medical diagnosis and machine learning. It explains why a very accurate test can still give mostly false alarms when a disease is rare.

the bigger pattern

Bayesian reasoning is used to revise beliefs in fields from forensics to weather forecasting. The same base-rate logic explains why rare conditions produce many false positives.

more on this

From Wikipedia, the free encyclopedia · text CC BY-SA 4.0

Illustrative photo · Qniemiec, CC0, via Wikimedia Commons