The Only Three Questions That Count
Challenge beliefs, find unseen possibilities, and audit your own brain
- Difficulty
- Advanced
- Time to result
- ~ongoing to results
- Steps
- 5
- Confidence
- 99%
Ken Fisher's three-question method begins by asking which of your beliefs is actually false, because conventional wisdom is widely known and therefore already priced into markets. The second question asks what you can fathom that others are not fathoming, shifting attention toward possibilities outside the consensus. The third asks what your brain is doing to mislead you, recognizing that human instincts evolved for survival rather than capital-market decisions. Fisher makes the first question operational: state your beliefs, find relevant historical data, and test causal claims with correlation where possible. A belief that lacks the expected relationship should be discarded. Together, the questions turn contrarianism from reflexive disagreement into a repeatable process of self-audit, evidence testing, and search for overlooked information.
Origin
Ken Fisher says the questions emerged from a lifetime of examining what he got wrong and trying to improve. He discussed them in his book The Only Three Questions That Count and explained the method on Masters in Business.
Core principles
- 01Widely known information is already reflected in market prices
- 02A belief should survive historical and quantitative testing
- 03Useful foresight comes from considering what others have not
- 04Human instincts can misdirect attention and action
How to run it
- 1
State your beliefs
Write down the claims guiding the decision. Fisher says that if you cannot identify what you believe, you should not be operating in markets.
Pro tip Phrase causal beliefs as an X-causes-Y claim when possible.
Watch out An unstated belief cannot be tested clearly.
- 2
Search for false assumptions
Ask which belief may be false rather than seeking support for every belief. Treat conventional wisdom as suspect because markets already discount widely known information.
Pro tip Start with beliefs repeated most confidently in public discussion.
Watch out Do not rationalize a belief merely because changing your mind is uncomfortable.
- 3
Test against history
Find historical instances and relevant data, then measure whether the expected relationship appears. Fisher recommends correlation coefficients for data-rich causal claims.
Pro tip Measure outcomes across more than one time horizon.
Watch out Not every claim can be tested with correlation, and a first-ever event remains possible.
- 4
Fathom what others have not
Look beyond what the crowd is discussing and identify plausible developments it has not priced. This narrows the search toward information with potential forecasting value.
Pro tip Use consensus expectations to define what is already known.
Watch out Being different is not enough; the alternative still needs a mechanism.
- 5
Audit your brain
Ask how instinct, emotion, or a desire to act may be drawing attention toward the obvious threat. Slow the decision down before acting.
Pro tip Explicitly ask how you would measure your emotional interpretation.
Watch out The urge to act quickly can override an insight you understand intellectually.
In the wild
Fisher proposes finding prior initial interest-rate hikes, counting them, and measuring market results after one week, one month, three months, one year, and five years. He says this historical test contradicts the public discussion that treats the first hike as a reliable market signal.
→ The investor replaces a popular fear with measured historical evidence.
Fisher compares market attention to people around an ancient campfire turning toward a loud noise while the greater danger may be behind them. In markets, he says widely discussed worries are already being handled by everyone else and are likely priced, so the investor should investigate what is not receiving attention.
→ Attention shifts from the obvious consensus worry to potentially unpriced risks.
Common mistakes
Rationalizing instead of testing
People spend time defending a wrong belief instead of changing it. The method requires a genuine willingness to stop believing a claim that fails the evidence test.
Confusing consensus with insight
A heavily discussed idea may feel important while carrying little new pricing power. Search beyond what is already widely known.
Acting on emotional urgency
Human attention is naturally captured by vivid threats. Pause long enough to assess whether the reaction is measurable and whether the event is already priced.
Is it for you?
Best for
It is best for investors and forecasters making decisions amid strong consensus and emotional noise.
Not ideal for
It is not ideal for claims with no usable history, data, or meaningful way to test them.
From the transcript
“what do you believe it's actually false?”
“what can I fathom that other people aren't fathoming”
“What is my brain doing to mislead me?”
From the episode
Interview With Ken Fisher: Masters in Business (Audio)