Base Rate

Definition

A base rate is how common something is in the whole relevant population before you consider any specific evidence. Ignoring it leads to overconfident judgments: a highly accurate test for a rare condition can still produce mostly false alarms.

In plain terms

When a vivid detail arrives, such as a positive test or a convincing story, people tend to forget how rare the thing was to begin with. The base rate is the reminder: how often does this happen at all?

Evidence should move your belief starting from the base rate, not replace it.

Example

The numbers below are synthetic. Suppose a condition affects 1% of people, a test catches 90% of true cases, and it wrongly flags 9% of people who do not have it. Take 10,000 people:

Of the 981 positive results, only 90 are real. A positive test therefore means about a 9% chance of having the condition, not 90%. In symbols:

$P(\text{condition} \mid +) = \frac{0.90 \times 0.01}{0.90 \times 0.01 + 0.09 \times 0.99} \approx 0.092$

Common misunderstandings

• “An accurate test means an accurate result.” Accuracy per case and reliability per positive result are different things when the condition is rare.
• “The base rate is just my opinion.” It is a measured frequency in a chosen reference group. The hard part is picking the right group.
• “Base rates are fixed.” They change over time and between populations, so check that yours still applies.