# Base Rate Fallacy

> Why a positive test for a rare illness is often wrong. Why bad luck seems to be for others. Why each job takes longer than we plan.

**Course:** [Psychology and Cognitive Biases](https://learn.benibix.com/learn/general-psychology)  
**Time:** 12 min

**After this lesson you can:**

- Work out what a positive test means for a rare thing
- Spot optimism bias in a plan
- Use past jobs to estimate a new one

## Why is a positive test for a rare illness often wrong?

The base rate fallacy is ignoring how rare a thing is when you judge one case. Take an illness that 1 in 1,000 people have, and a test that wrongly flags 5 in 100 healthy people. Test 1,000 people, and about 51 come back positive. Only 1 of them is sick. A positive result means about a 2 percent chance.

Most people guess far higher, often 95 percent, because the test sounds good. They focus on the test and forget how few people had the illness to begin with. The rare thing has to beat the false alarms. And there are far more healthy people to set them off.

This is why doctors do not screen everyone for every rare illness. It is also why a second, different test is used to confirm a first positive. The base rate is the starting point, and the test only moves you from there.

Optimism bias is thinking bad luck is for other people. Most people expect to live longer, earn more and stay married more than the numbers say. It lifts the mood. It also means people skip insurance, savings and seat belts.

The planning fallacy is guessing a job will be quick, even when past jobs ran late. In a 1994 study, students guessed they would finish their final project in about 34 days. On average they took about 56. Fewer than a third finished by their own guess.

The cure for the planning fallacy is to use the base rate of your own past. How long did the last three projects like this take? Start there, not with the best case. People who ask how long it usually takes guess far better than those who imagine this time.

All three come from looking at the case in front of you and missing the bigger pattern. The fix is the same question each time. How often does this happen, to everyone or in my past?

![Three panels. Planning fallacy: two bars, a short one for the guess of 34 days and a long one for the real 56 days. Optimism bias: a crowd under a rain cloud, with one person under a small patch of sun saying not me. Base rate fallacy: a tree that starts at 1,000 people tested, leads to 51 who test positive, and splits into 1 who is sick and 50 who are well.](https://learn.benibix.com/diagrams/gen7-odds-and-plans.svg)

### Example: Owen plans a kitchen

Owen plans to redo his kitchen in 3 weekends. His last two home jobs each took about three times as long as he planned.

He tells his family to expect 9 weekends. It takes 8, and for once nobody is cross.

**Words in this lesson:** [Base Rate Fallacy](https://learn.benibix.com/term/general/base-rate-fallacy), [Optimism Bias](https://learn.benibix.com/term/general/optimism-bias), [Planning Fallacy](https://learn.benibix.com/term/general/planning-fallacy)

**Question:** A test for a rare illness comes back positive. What is the first thing to ask?

**Answer:** How rare is the illness to begin with?. If the illness is rare, most positives come from the many healthy people. The base rate tells you how much to trust the result.

---

Source: https://learn.benibix.com/learn/general-psychology/base-rate-fallacy
Author: Behnia Zarei (a developer in Vienna, Austria, who builds and teaches data, AI and business intelligence at BENIBIX)
Publisher: BENIBIX Learn (https://learn.benibix.com/)
Last updated: 2026-11-19
Licence: CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). Quote freely with attribution.
