Metric Gaslighting Detector

Dashboards rarely contain false numbers. They contain true numbers arranged to create false beliefs. This skill audits the arrangement — the eleven standard distortions through which honest data becomes dishonest narrative.

Required Inputs

• The metrics artifact — the dashboard description, KPI table, chart, or the numbers with their labels exactly as presented. Include axis ranges, time windows, and any annotations; the lie usually lives there.
• The claim being made with it (if any) — "churn is under control", "the launch worked". The audit tests the claim-data connection, not the data alone.

The Eleven Distortions

1. Denominator games — the base changed ("of active users" quietly became "of weekly active")
2. Survivorship framing — measuring only what remained (retention of cohorts that didn't churn early)
3. Y-axis crimes — truncated baselines, dual axes, log scales without labels
4. The cherry window — the date range that starts at the trough or ends before the drop
5. Mix-shift laundering — the aggregate improved because composition changed, not performance
6. Ratio without magnitude — "+40%!" concealing 5→7
7. The vanity proxy — measuring what moves instead of what matters (signups for activation)
8. Goodhart's ghost — the metric improved because it became a target, and the gamed behaviour is visible elsewhere
9. Smoothing to silence — rolling averages wide enough to bury the event being asked about
10. The missing counterfactual — "up 20% since launch" with no baseline trend (it was up 25% before)
11. Significance theatre — differences within noise presented as movement ("ticked up to 4.6 from 4.5, n=41")

Output Format

1. The audit table — metric | distortion(s) detected | severity (🔴 changes the conclusion / 🟡 shades it / 🟢 clean) | the honest version of that number's sentence.