Use reference classes to ground personal estimates in base rates
Before estimating how your situation will unfold, find similar past situations and check what happened.
Key takeaways
- What it is: Before estimating how your situation will unfold, find similar past situations and check what happened.
- Why it works: People generate predictions primarily from inside view — their own model of the specific case — which is systematically optimistic because it ignores how often similar plans fail in general. Reference class forecasting corrects this by grounding the prior in outside view: the statistical distribution of outcomes for similar cases. The two views are then combined rather than the inside view dominating.
- Evidence: Backed by observational / correlational evidence.
- Avoid: Choosing the reference class that makes your project look most distinctive (therefore most likely to be exceptional) rather than the class that most honestly resembles it.
Why it works
People generate predictions primarily from inside view — their own model of the specific case — which is systematically optimistic because it ignores how often similar plans fail in general. Reference class forecasting corrects this by grounding the prior in outside view: the statistical distribution of outcomes for similar cases. The two views are then combined rather than the inside view dominating.
How to do it
- 1Define your situation type: "This is a home renovation project / a software product launch / a behavior change attempt."
- 2Find data on how that type of project typically goes: timeline, cost overruns, success rates.
- 3Use that base rate as your starting probability, then adjust for features that make your case better or worse than average.
- 4Do not let the inside view increase your probability more than the specific distinguishing features actually justify.
What the evidence says
ObservationalReference class forecasting was developed by Daniel Kahneman and Amos Tversky and later formalized by Bent Flyvbjerg in infrastructure project planning. Studies show it reliably reduces planning fallacy — especially cost and time overruns — compared to inside-view estimates.
Honest caveat: Reference class forecasting works best when good historical base rate data exist; for novel situations, the reference class itself is uncertain.
- — Flyvbjerg (2006), from Nobel Prize to project management: getting risks right, Project Management Journal
Common mistake
Choosing the reference class that makes your project look most distinctive (therefore most likely to be exceptional) rather than the class that most honestly resembles it.
IX Coach prompts you to identify a reference class for any goal you set and surfaces what the realistic success rate looks like for that category before you commit to your plan.
Practice this with IX Coach →