Overview
A base rate is the underlying frequency of something across a population — how often it happens in general. The model says to anchor your estimate on that base rate first, then adjust for the specifics, rather than being swept away by a compelling individual story. People routinely ignore base rates in favour of vivid detail, and get probability badly wrong as a result.
The discipline is to ask 'how often does this happen for cases like this?' before 'how special is this case?'
When to use it
Estimating how likely something is — especially when a vivid story tempts you to ignore the odds.
How to apply it
Find the base rate
Ask how frequently this outcome occurs for the general class of cases.
Anchor on it
Start your estimate from that general frequency.
Adjust for specifics
Move off the base rate only for genuinely relevant, reliable details.
Resist the vivid story
Don't let a compelling narrative override the underlying odds.
Common pitfalls
- Ignoring base rates entirely in favour of a vivid, specific narrative.
- Adjusting off the base rate for details that aren't actually predictive.
- Using a base rate from the wrong reference class.
Frequently asked questions
What's 'base rate neglect'?
The common error of ignoring how often something happens in general, and judging only by the specific, vivid details of the case in front of you.
How do I use base rates well?
Anchor on the general frequency for similar cases first, then adjust only for genuinely predictive specifics.