USR-2026-0020 · CMB

The hemispherical power asymmetry peaks at 2.0σ (p = 0.0250) at LMAX = 64 toward (l,b) = (228°, -24°) — 7° from Lepus — and vanishes (0.2σ) by LMAX = 96.

PROVISIONAL 2.0σ at LMAX = 64; 0.2σ by LMAX = 96 reproduced ≠ discovered re-run & confirmed 2026-07-26 (R0 — founder CI; independence pending)

An optimal estimator scan over LMAX: the asymmetry is a large-scale-only feature. It rises to 2.0σ at the standard LMAX = 64 choice and is gone by LMAX = 96 — both facts seeded. KILL TEST EXECUTED 2026-07-26: survived on COMMANDER, NILC and WMAP ILC with direction coherence to 7.4° and the same die-by-LMAX-96 scale dependence — the anomaly is cleaning- and instrument-robust at its ~2σ level. Cosmic variance still caps it there; LiteBIRD E-modes remain the arbiter.

Falsify-box — how to kill this claim

SURVIVED Re-run the identical optimal quadratic (Hanson-Lewis) estimator of tilt_seam/run_optimal_asymmetry.py on the three cleaned maps it never used - Planck COMMANDER, Planck NILC, WMAP ILC - and extend the LMAX scan to 96 and 128, keeping the 1000-realization LambdaCDM null and the same one-sided p->sigma conversion (norm.isf). DEAD if any one of: (a) LMAX = 64 returns p > 0.16 (below 1.0 sigma, i.e. half the claimed 2.0 sigma) in any of those independent cleanings; (b) LMAX >= 96 returns p < 0.067 (>= 1.5 sigma), which contradicts 'vanishes (0.2 sigma) by LMAX = 96' - 0.2 sigma is p = 0.42, so p < 0.067 is a factor-6 discrepancy in p; (c) the recovered LMAX = 64 modulation-dipole direction moves more than 45 deg from (228 deg, -24 deg), i.e. out of the Lepus hemisphere-pair entirely.

Verification record — every quoted number, re-run 2026-07-26

quantitymeasuredverdict
LMAX = 40 significance 1.5σ MATCH
LMAX = 64 significance 2.0σ MATCH
LMAX = 64 p-value 0.0250 CLOSE
LMAX = 64 direction (l,b) = (228°, -24°), 7° from Lepus MATCH
LMAX = 96 significance 0.2σ (null) CLOSE
CROSS-CLEANING TEST 2026-07-26 (fire_killtest_asymmetry.py, 4000 sims/LMAX) COMMANDER 1.96σ (p=0.0247, dir 7.4° off), NILC 2.06σ (p=0.0197, 2.3° off — sharper than SMICA), WMAP ILC 1.22σ (3.0° off; 1.50σ after deconvolving WMAP's 1° beam) MATCH
direction coherence four cleanings, two instruments, all within 7.4° of (228°, −24°) MATCH
scale-dependence confirmed everywhere peaks at LMAX≈64 and is gone by LMAX=96 in all five variants — the physical fingerprint holds MATCH
faithful-lift check SMICA re-run reproduces the registered 2.0σ (2.05σ, p=0.0202) and direction to 0.4° MATCH

Provenance

script: tilt_seam/run_optimal_asymmetry.py
script status: PRESENT
datasets: Planck SMICA · claim-scripts (downloadable from /data — run it yourself)
re-run: 2026-07-26

Edges

REPRODUCES → The published Planck hemispherical power asymmetry and its scale dependence

Status, honestly

kill test EXECUTED 2026-07-26 and SURVIVED: the asymmetry reproduces on all three independent cleanings it had never been tested against. Status here is computed from evidence — the author cannot set it, and neither can we. Independent reproduction would move it; nothing else will.

§ Figures 1 from tilt_seam/run_optimal_asymmetry.py

What this claim's own script draws. These are the founder's committed outputs — the same plots the script regenerates on any machine that runs it.

optimal_asymmetry.png 44.0 KB download
Share this claim X Bluesky LinkedIn Reddit HN Email