Base Rate Fallacy

Category: Probability & Belief

If presented with related base rate information (i.e. generic, general information) and specific information (information only pertaining to a certain case), the mind tends to ignore the former and focus on the latter.

How it works

The Base Rate Fallacy is what happens when a vivid specific detail bulldozes a boring background statistic. The base rate is the underlying frequency of something in a population, how common librarians are, how rare a disease is, how often a test gives false alarms. When you're handed both this general number and a colorful particular description, your mind clamps onto the description and quietly discards the number that actually does most of the predictive work.

The culprit is the representativeness heuristic: you judge probability by how well something matches a stereotype rather than by how often it actually occurs. A quiet, bookish man feels like a librarian, so you call him one, never mind that salespeople outnumber librarians by something like fifty to one, which makes a bookish salesperson far more likely in raw counts than a bookish librarian.

The stakes get serious in medicine. Suppose a disease afflicts 1 in 1,000 people and a screening test is 99% accurate. You test positive. It feels like a 99% chance you're sick. But because the disease is so rare, the test's small false-positive rate is applied to a huge healthy population, generating far more false alarms than true cases. Your real odds of being sick are closer to 9%. Ignore the base rate, and you'll misjudge it by a factor of ten.

Where you'll see it

  • A doctor tells a patient with a positive result for a rare cancer that they 'almost certainly' have it, forgetting that false positives across the vast healthy population dwarf the handful of real cases.
  • An investor pours money into a friend's startup because the founder 'has that visionary spark,' ignoring that the vast majority of startups fail regardless of founder charisma.
  • A hiring manager rejects a soft-spoken candidate for a sales role, picturing salespeople as loud extroverts, while overlooking that most successful reps in the company are actually low-key.

Where it comes from

The fallacy was crystallized in Daniel Kahneman and Amos Tversky's research in the early 1970s, most memorably through the 'Tom W.' and 'cab' problems. In one classic study, participants were told that a city's cabs are 85% green and 15% blue, that a witness identified a cab in an accident as blue, and that the witness is 80% reliable, yet most people guessed the cab was blue, all but ignoring the 85% green base rate that should have dominated the calculation. The work became a cornerstone of behavioral economics and Kahneman's later book 'Thinking, Fast and Slow.'

How to counter it

Always anchor on the base rate first, before the story gets to you. Ask: forgetting everything specific about this case, how common is this thing in the general population? That number is your starting point, and the vivid details only nudge it, they don't replace it.

When you face a 'positive test' of any kind, a medical screen, a fraud alert, a personality match, run the numbers in raw counts instead of percentages. Imagine 1,000 people, sort them into the true cases and the false alarms, and look at how big each pile actually is. Concrete counts make the false-positive trap obvious in a way that lonely percentages never do.

And train yourself to be suspicious of any judgment that feels too easy because the description was so fitting. Representativeness is a feeling, not a probability. The more perfectly someone matches a stereotype, the more deliberately you should ask how rare that category is to begin with.

The tell

You're doing it when a detailed, fitting description convinces you something is likely while you've never once asked how common that thing actually is.

Related biases

Featured in