The Pre-Print Cohort Read
Heading into the print, the live cohort_analyze pull on TTD returned n=289 historical analogs with a tightness score of 0.00. The 5-day forward median was +0.80% with IQR [-4.07%, +5.12%] and hit-rate 53%. The 10-day median was +0.21% with IQR [-6.18%, +7.88%]. The dominant 5-day feature was yield_curve_state=inverted (negative direction, importance 1.22).
What Actually Printed
Print-day close: $23.49, prior close $24.01. Day-over-day change: -2.17% (-0.52). Intraday range: $23.37 – $24.66.
Cohort vs. Reality — The Verdict
The -2.17% reaction is sitting inside the cohort's IQR (-4.07% to +5.12%). 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 +0.80%, IQR [-4.07%, +5.12%]
- Conformal bucket boundaries: p10 -9.23% / p90 +11.31%
- Top 5d feature: yield_curve_state=inverted (negative direction, importance 1.22)
- Actual reaction: -2.17%; verdict: central tendency hold
The Calibration Lesson
For TTD, the calibration verdict is: central tendency hold. The cohort's structured features (yield_curve_state=inverted 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.