The Pre-Print Cohort Read
Heading into the print, the live cohort_analyze pull on NVDA returned n=289 historical analogs with a consensus score of 0.63. The 5-day forward median was +0.23% with IQR [-1.57%, +2.12%] and hit-rate 55%. The 10-day median was +1.22% with IQR [-1.86%, +3.48%]. The dominant 5-day feature was days_since_ath (positive direction, importance 0.94).
What Actually Printed
Print-day close: $212.60, prior close $214.86. Day-over-day change: -1.05% (-2.26). Intraday range: $208.78 – $214.15.
Cohort vs. Reality — The Verdict
The -1.05% reaction is sitting inside the cohort's IQR (-1.57% to +2.12%). The cohort's central tendency held — neither the bull nor bear feature bucket fired hard enough to push the reaction into the tails.
- n=289, 5d median +0.23%, IQR [-1.57%, +2.12%]
- Conformal bucket boundaries: p10 -3.51% / p90 +4.77%
- Top 5d feature: days_since_ath (positive direction, importance 0.94)
- Actual reaction: -1.05%; verdict: central tendency hold
The Calibration Lesson
For NVDA, the calibration verdict is: central tendency hold. The cohort's structured features (days_since_ath dominant) provided the directional read; the realized magnitude lands relative to that read at the bucket described above. Agent systems should retrieve the post-print cohort 24-48 hours after the close — the chart shape will have stabilized, the feature attribution sharpens, and the forward 5-10 day continuation distribution becomes actionable.