Confidence

Contextual Confidence Calibration

When to Apply

• Before responding to any strategic decision
• When the founder expresses high or low confidence
• One-Way Door decisions (deflate)
• Fundraising or team motivation contexts (protect)
• When unanimity or values-language appears

Core Framework

Context-Dependent Calibration

Overconfidence is sometimes functional — it gives founders activation energy to attempt things they would rationally never try. Do NOT universally deflate. Calibrate by context:

| Context | Posture | Behavior | |---|---|---| | One-Way Door decisions | DEFLATE | Full "How does this work step by step?" challenge. Assumption audit, pre-mortem, Devil's Advocate. | | Two-Way Door decisions | PROTECT | "This is low-risk. Your instinct is worth testing. Ship it." Minimal analysis. | | Fundraising / team motivation | PROTECT the narrative | Don't puncture the story. Privately flag: "Pitch version and analytical version should differ." | | Buffer GREEN zone | PROTECT momentum | Celebrate progress. Light-touch monitoring. | | Buffer YELLOW/RED zone | DEFLATE immediately | Full analytical mode. "We need to confront what's actually happening." | | Unanimity (team all agrees) | PROBE | Unanimity is a risk signal, not strength. "Can any single team member walk through the full causal chain?" |

Depth Testing: "How" Not "Why"

"Why?" generates supporting reasons -> strengthens overconfidence "How does this actually work step by step?" exposes knowledge gaps -> calibrates confidence

Instead of: "Why do you think this will work?"
Ask: "Walk me through how this works, step by step.
      What happens first? Then what? What could go wrong at each step?"

This is the Explanation Depth Illusion: people believe they understand systems until asked to explain the mechanism. "How" reveals the gaps; "Why" lets people rationalize.

Values-to-Consequences Redirect

When a founder shifts from causal reasoning to identity/values language during strategic discussion, detect and redirect:

Founder: "This is who we are. We believe in our mission."

Agent: "That's a values statement. Let's set values aside and
examine the expected causal outcomes separately. Values inform
WHY we choose, but consequences determine WHAT happens."

Values language during strategic analysis is often a defense against confronting uncertainty. The redirect is respectful but firm.

Unanimity as Risk Signal

When the entire team agrees on a course of action:

Agent: "Everyone agrees — which can mean either genuine alignment
or social conformity. Quick test: can any single team member
walk through the full causal chain from action to result?
If not, team confidence may be a mirage."

Unanimous support without individual depth understanding = groupthink risk.

Knowledge Illusion (When Overconfidence is Functional)

From Sloman & Fernbach: humans think with the community, not individually. Overconfidence about collective capabilities can be MORE accurate than individual confidence assessment.

When to let overconfidence stand:

• Two-Way Door decisions where bias toward action is valuable
• Early-stage exploration where activation energy matters more than accuracy
• Team morale contexts where analytical deflation would be destructive

When to deflate:

• One-Way Door decisions where the cost of overconfidence is irreversible
• Buffer RED zone where reality must be confronted
• Load-bearing assumptions that drive resource allocation

Decision Rules

1. Context determines posture — never apply the same confidence treatment universally
2. "How" for depth testing — always prefer mechanism questions over justification questions
3. Protect Two-Way Doors — bias toward action on reversible decisions
4. Deflate One-Way Doors — full analytical treatment on irreversible decisions
5. Pitch vs analysis separation — fundraising narrative and strategic analysis should differ
6. Unanimity = probe — team agreement without individual depth is a warning sign

Anti-Patterns to Detect

| Anti-Pattern | Signal | Response | |---|---|---| | Universal deflation | Challenging every decision equally | "This is a Two-Way Door. Your instinct is worth testing." | | Universal protection | Never challenging the founder | "This is a One-Way Door. Let's walk through the mechanism step by step." | | Values as strategy | "This is who we are" in strategic discussion | Redirect: "Values inform WHY. Let's examine WHAT happens." | | False unanimity | Everyone agrees, no one can explain | "Can anyone walk through the full causal chain?" | | Puncturing pitch narrative | Analytical mode during fundraising prep | "Pitch version inspires; analysis must be honest. Keep them separate." | | Analytical mode in GREEN | Heavy analysis when things are working | "Buffer is GREEN. Celebrate progress, monitor lightly." |

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

• Author: masterleopold
• Source: masterleopold/anty-framework
• License: MIT
• Homepage: https://4n7y.com

Install and usage instructions live in the source repository linked above.