The Pre-Print Cohort Read
Heading into the print, the live cohort_analyze pull on COIN returned n=29 historical analogs with a tightness score of 0.64. The 5-day forward median was -1.09% with IQR [-3.31%, +2.24%] and hit-rate 38%. The 10-day median was -1.20% with IQR [-3.85%, +0.34%]. The dominant 5-day feature was abs_price_return_z (negative direction, importance 0.89).
What Actually Printed
Print-day close: $192.96, prior close $197.96. Day-over-day change: -2.53% (-5.00). Intraday range: $190.32 – $198.15.
Cohort vs. Reality — The Verdict
The -2.53% reaction is sitting inside the cohort's IQR (-3.31% to +2.24%). The cohort's central tendency held — neither the bull nor bear feature bucket fired hard enough to push the reaction into the tails.
- n=29, 5d median -1.09%, IQR [-3.31%, +2.24%]
- Conformal bucket boundaries: p10 -5.35% / p90 +4.62%
- Top 5d feature: abs_price_return_z (negative direction, importance 0.89)
- Actual reaction: -2.53%; verdict: central tendency hold
The Calibration Lesson
For COIN, the calibration verdict is: central tendency hold. The cohort's structured features (abs_price_return_z 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.