Skip to main content

A registered result in its source context

Original ledger classification: Numerical [N] (statistical null-model rejection; Python-only)

Recorded statement: NUMEROLOGY NULL TEST (look-elsewhere corrected; executes the legacy 'numerology grammar test' TODO from the CMB paper): (A) EXACT CENSUS - for each of the 13 scored frozen-registry observables the FULL declared grammar class (one-term…

Statement, assumptions and sources

Paper context: TFPT — Introduction and reading guide

The exact identifier, original classification and full claim are available below.

The source wording and classification below are preserved as recorded. A paper link, shared script or authored topic relation does not upgrade the claim or report a new proof run. Read later reviewed findings separately.

Read the exact registered claim and its dependencies

AUDIT.NULLMC.01

Original ledger classification: Numerical [N] (statistical null-model rejection; Python-only)

Original claim text

NUMEROLOGY NULL TEST (look-elsewhere corrected; executes the legacy 'numerology grammar test' TODO from the CMB paper): (A) EXACT CENSUS - for each of the 13 scored frozen-registry observables the FULL declared grammar class (one-term (p/q) phi0^a lamC^c (1+lamC)^d pi^b e^(u/w) with p,q<=130, |a|<=2, |c|<=3, |d|,|b|<=1, |u|<=6, w<=6; two-term sums with coefficients<=12; machine-checked fairness: the class contains every TFPT formula it scores) is enumerated exactly against conservative data windows (max of achieved deviation, documented sigma, quote resolution, 5% quark scheme spread - tension observables enter with their honest LARGE windows and contribute almost nothing): joint formula-fishing probability prod p_i = 1e-25.8, robust to <2 dex under coefficient bound 80..200 and plausibility window 5x..20x; (B) MONTE CARLO 2x200k deterministic pseudo-theories (random rational dials in TFPT's own formula skeletons, seed phi0 fixed resp. drawn log-uniformly) reach at most 5/13 resp. 4/13 hits vs TFPT 13/13; (C) POWER controls: perturbing phi0 by 1%/10%/30% collapses TFPT's own scorecard to 9/6/4 hits, shuffling the data<->formula assignment to mean 1.2; (D) ALPHA CLASS: all 94500 complexity-matched F_U1 variants (coefficients, M, transport exponent, resolvent power) root-solved, exactly ONE - the TFPT instance (2, 4/5, M=41, e^-2a, -5/4) - lands inside the achieved 4e-8 CODATA window => p_alpha <= 1.1e-5. TOTAL: P(random complexity-matched theory reproduces the scorecard incl. alpha) <= 1e-30.7 (~102 bits), CONDITIONAL on the declared grammar - a look-elsewhere-corrected null-model rejection, never 'certainty'. Complementary to the freeze: v84 prevents FUTURE look-elsewhere drift, v100 quantifies the RETROSPECTIVE look-elsewhere burden

Original classification detail

Numerical [N]: joint null probability <= 1e-30.7 conditional on the declared grammar

Dependencies as recorded

REG.FREEZE.01 (scored predictions = frozen registry); DATA.SCORECARD.01 (documented data centrals)

Open the original ledger source, row 183

Location recorded in the ledger

These document locations are recorded directly on the ledger row. They are document level navigation, rather than a newly inferred section citation.

Navigation through a shared verification script

These sections cite a script associated with the claim. This is a source navigation join, rather than an explicit citation of the claim or a theorem equivalence.

All 23 linked sections

Registered verification modules

  • v100_numerology_null_mc.pyRegistered verification module · registry
    Read the description recorded in the module registry

    Numerology null test (look-elsewhere corrected) [N]: exact census of the FULL declared formula grammar (provably containing every scored TFPT formula) over the 13 scored frozen-registry observables with conservative data windows ⇒ joint formula-fishing probability ∏pᵢ = 10⁻²⁵·⁸, robust to budget/window variation; 2×200k Monte-Carlo pseudo-theories reach at most 5/13 hits vs TFPT 13/13; power controls: φ₀ ±1%/10% collapses TFPT's own scorecard to 9/6, data shuffle to mean 1.2; all 94 500 F_U(1) variants root-solved — exactly ONE hits the 4×10⁻⁸ CODATA window ⇒ total ≤ 10⁻³⁰·⁷ (~102 bits), CONDITIONAL on the declared grammar (a null-model rejection, never 'certainty'); complementary to the v84 freeze

A registered module is a source reference. Its presence does not assert that the module was executed in this snapshot.

Search later reviewed research for this identifierRead the coverage boundaries