Base Rate Fallacy
Type: Cognitive Bias
Local HTML: base_rate_fallacy.html
Definition
Ignoring background statistics in favor of specific, individuating information. The base rate is the boring background fact; we chase the vivid details.
Classic example: A test is 95% accurate. The condition affects 1 in 1000 people. You test positive. What’s the chance you have it? Most say 95%. Correct answer: ~2%.
Why It Matters
Medicine: Patients (and doctors) ignore base rates, overreact to test results. Security: Profiling ignores base rates — most people are not terrorists. Hiring: “They went to Harvard!” — ignoring base rate of success from all schools. Investing: Chasing hot stocks, ignoring market base rates.
The Math
Bayes’ Theorem in plain English:
- Start with base rate (prior probability)
- Update with new evidence
- But don’t forget where you started
Most people skip step 1.
Examples
- Cab problem: Witness says blue cab (80% accurate). But 85% of cabs are green. Most ignore the 85%.
- Medical testing: Rare disease + accurate test ≠ likely positive
- Stereotypes: “He’s shy, must be a librarian” — ignoring base rate of shy people vs librarians
Fighting It
- Always start with base rate — What’s the background frequency?
- Use natural frequencies — “1 in 1000” not “0.1%”
- Draw tree diagrams — Visualize the probabilities
- Ask: “What if there was no specific info?”
Related Biases
- [[Conjunction Fallacy** — Adding details seems more likely
- [[Representativeness Heuristic** — Matching to stereotypes
- [[Confusion of the Inverse** — P(A|B) ≠ P(B|A)
Audio
Podcast episode: Base Rate Fallacy
Part of the Cognitive Bias Reference