The headline signal — how confidence moved from the previous snapshot, and why.
vs 2.1.30.00 —
Held at 2.1.3 levels (0.82 ± 0.04). No new CPI / PPI / PCE / wage data in the 12.5-hour window. Tuesday-morning 10Y holds ~4.60% (intraday range 4.56-4.63%) - essentially unchanged from Monday's 4.601% close, even through the overnight White House rejection of Iran's proposal that flipped the diplomatic-track signal. Global yields (German bund, JGB) still at multi-year highs. The macro mean that drives this thesis has not moved. Beta(33, 7.2) parameterization continues to encode ~40 effective independent observations. April PCE on May 28 is the next primary move, then June 16-17 FOMC.
The thesis
The claim and where confidence stands now.
Beta(33, 7.2) · 95% CI [0.74–0.89]
April CPI on May 12 fired the energy-passthrough binary tell directly: headline 3.8% YoY (highest since May 2023), core 2.8% YoY. April PPI on May 13 ran hotter than CPI - headline +1.4% m/m / +6.0% YoY, core +1.0% m/m / +5.2% YoY (highest in three years). April import prices on May 14 added +1.9% m/m / +4.2% YoY (largest YoY since October 2022) with imported fuel +16.3% m/m. Friday May 15 delivered the sharpest single-session rate-path repricing of the cycle: oil +4.5% (WTI to $106, weekly +11%), the 10Y to 4.59% (up ~13bps in one day, fresh ~1-year high, biggest weekly yield jump since April 2025), and CME FedWatch hike-by-December odds to ~56% from ~36% Thursday and ~16% a week earlier - a 20pp single-day move. Monday May 18 cash session closed 10Y at 4.601% (vs 4.59% Friday) on the strike-cancellation tactical pullback; Tuesday morning 10Y holds ~4.60% (range 4.56-4.63%) through the Axios rejection-of-Iran-proposal news. Global yields still at multi-year highs (German bund 15-year high, JGB 10Y 29-year high). Calendar: April 29 FOMC minutes Wednesday May 20 2 PM ET, Walmart Q1 FY2027 Thursday May 21 pre-open with tariff-passthrough lens, April PCE the formal invalidation indicator on May 28.
Drivers
The underlying macro forces this thesis expresses - the loading mean is how much each force drives the thesis, the stddev our confidence in the mapping.
Supporting evidence
Typed, citation-backed observations across time, grouped by strength. Hover a point for the claim.
What would invalidate this
The machine-evaluable conditions that would falsify the thesis.