The Bias-Back Charter
Recruit trusted people to expose the biases you cannot see yourself
- Difficulty
- Moderate
- Time to result
- ~weeks to results
- Steps
- 5
- Confidence
- 97%
The Bias-Back Charter creates a reciprocal agreement among trusted peers to detect motivated reasoning. Because people readily notice selective evidence, spinning, and defensiveness in others while missing those patterns in themselves, each member authorizes the others to intervene. The charter specifies useful behaviors: present contradictory information, identify omitted evidence, call out defensive reactions, and focus criticism on reasoning rather than personal worth. Participants promise to receive challenges with as little defensiveness as possible and to provide the same service in return. This does not eliminate cognitive bias, but it catches errors earlier, creates accountability, and makes more accurate belief updating socially supported rather than dependent on individual willpower.
Origin
Annie Duke describes formalizing the candid accountability she received from poker mentors such as Erik Seidel into a general method for collaborative truth seeking.
Core principles
- 01People detect bias more easily in others than in themselves
- 02Intelligence can strengthen motivated reasoning
- 03Disagreement is valuable when accuracy is the shared goal
- 04Feedback requires explicit permission and reciprocal accountability
- 05Defensiveness is itself useful evidence
How to run it
- 1
Select truth-seeking partners
Choose a small group of people who are informed, trustworthy, and willing to prioritize accuracy over agreement.
Pro tip Include people with different perspectives and incentives.
Watch out Avoid members who use candor as permission for hostility.
- 2
Create the charter
Agree explicitly that members may flag spinning, missing evidence, motivated reasoning, and defensiveness.
Pro tip Define acceptable language for interrupting bias before a high-stakes discussion occurs.
Watch out Vague permission may disappear when criticism becomes uncomfortable.
- 3
Share reasoning, not just conclusions
Expose assumptions, confidence levels, alternatives, and evidence so partners can inspect the process.
Pro tip Ask what evidence or explanation you may have omitted.
Watch out A polished conclusion alone hides the reasoning that needs examination.
- 4
Challenge and receive
Partners provide contradictory evidence and identify biased reasoning; the decision-maker listens and asks clarifying questions.
Pro tip Treat an urge to defend yourself as a cue to pause.
Watch out Do not punish people socially for giving feedback you requested.
- 5
Update and reciprocate
Revise the belief when warranted, explain what changed, and perform the same bias-checking role for other members.
Pro tip Track repeated bias patterns across decisions.
Watch out The group can become an echo chamber if disagreement is not genuinely welcomed.
In the wild
An investor presents only evidence supporting a favored company. A charter partner points out an omitted study and notes that the investor is applying a harsher standard to negative evidence. Because this intervention was authorized in advance, the group can inspect the discrepancy without treating it as a personal attack.
→ The investment thesis is recalibrated before capital is committed.
Annie Duke complained to Erik Seidel about terrible poker luck. He refused to reinforce the emotional narrative and asked whether she had an actual question, redirecting attention toward anything that could be learned.
→ The exchange became a defining lesson in accountable, learning-focused mentorship.
Common mistakes
Recruiting only agreeable people
A group that shares the same assumptions may validate bias rather than expose it.
Treating feedback as a verdict
Challenges should supply evidence and test reasoning, not transfer decision ownership to the group.
Retaliating against candor
Defensive punishment teaches partners to withhold the very information the charter was designed to surface.
Is it for you?
Best for
It is best for decision-makers with trusted peers who can challenge one another candidly and constructively.
Not ideal for
It is not ideal where participants lack trust, psychological safety, or commitment to truth over status.
From the transcript
“bring other people in essentially to watch our bias back.”
“if you have information that disagrees with me, please offer it to me.”
“Then I'll do the same for you”
From the episode
Annie Duke Discusses Decision-Making in Poker and Life
Annie Duke