The Repeatability Test
Separate durable skill from luck by asking what you could reproduce
- Difficulty
- Easy
- Time to result
- ~days to results
- Steps
- 5
- Confidence
- 94%
The Repeatability Test evaluates success by separating deliberate capabilities from circumstances outside your control. Start with a meaningful outcome and identify the decisions, skills, relationships, timing, and environmental conditions that produced it. Then imagine restarting without money, reputation, or established context and ask which parts you could reproduce. The reproducible elements are better evidence of skill; the irreproducible elements indicate luck, timing, or inherited advantage. This does not diminish achievement. Instead, it improves calibration, discourages overconfidence, and directs future plans toward strengths that remain useful under changed conditions. The output is a more honest account of why an outcome occurred and which advantages can support the next attempt.
Origin
Extracted from Masters in Business during Morgan Housel's discussion of luck, timing, Bill Gross, Warren Buffett, and the distinction between accidental advantages and repeatable ability.
Core principles
- 01Luck lies outside intentional control
- 02Hard work and skill are repeatable
- 03Outcomes usually combine ability and circumstance
- 04Future plans should rely on reproducible advantages
How to run it
- 1
Choose an outcome
Select a success or failure important enough to influence your future decisions. Define the outcome without first explaining why it happened.
Pro tip Use a specific outcome rather than your entire career.
Watch out Do not begin with a flattering story about your own ability.
- 2
Inventory causal factors
List the skills, effort, relationships, resources, timing, and external events that contributed to the result.
Pro tip Include birthplace, era, market conditions, and introductions that you did not choose.
Watch out Invisible advantages are easy to omit.
- 3
Apply the restart question
Imagine beginning tomorrow with no money, name recognition, or existing context. Ask which actions and results you could credibly reproduce.
Pro tip Focus on processes you can demonstrate repeatedly.
Watch out Confidence alone is not evidence of repeatability.
- 4
Classify the evidence
Treat reproducible factors as skill and non-reproducible factors as luck or context. Preserve uncertainty where the classification is mixed.
Pro tip A factor can contain both skill and luck.
Watch out Do not use luck as an excuse to ignore disciplined execution.
- 5
Design around durable advantages
Base the next plan on repeatable capabilities while maintaining safeguards against unfavorable circumstances.
Pro tip Pair confidence in skill with humility about outcomes.
Watch out Do not assume the environment that rewarded you once will remain unchanged.
In the wild
A fund manager attributes exceptional returns to security selection. Applying the test, she finds that disciplined research and position sizing are repeatable, while falling interest rates and unusually favorable sector exposure are not. Her next strategy retains the research process but lowers leverage and tests performance across different regimes.
→ The manager preserves genuine strengths without betting that the same market tailwind will return.
A founder separates the launch team's repeatable messaging and distribution work from an unexpected celebrity mention. The company documents the repeatable campaign process but forecasts future launches without assuming another unsolicited endorsement.
→ Future plans become more realistic and less dependent on accidental virality.
Common mistakes
Renaming hard work as luck
If intentional effort increased the probability of success, that portion is skill or hard work rather than luck. Preserve the distinction so useful behavior can be repeated.
Treating luck as an insult
Acknowledging uncontrollable advantages does not erase intelligence or effort. It produces better calibration about how reliably an outcome can recur.
Is it for you?
Best for
It is best for investors, founders, and professionals reviewing unusually strong outcomes.
Not ideal for
It is not ideal for precisely quantifying luck or predicting whether a specific chance event will recur.
From the transcript
“I think the phrase to use is what is repeatable? What have you done in the past that you can do again?”
“if you started over tomorrow with no money, no name, recognition, no context, could you repeat what you've done?”
“If you can go out of your way to increase it, it is not luck. It's hard work.”
From the episode
The Art of Spending Money with Bestselling Author Morgan Housel