The Framework Detected Structural Severity Years Before Each Failure

The historical backtest applies the Four Frequencies framework to six documented failures across six sectors, using federal data (SEC filings, FDIC call reports, NTSB investigations, congressional testimony) to measure structural severity over time. The backtest tests whether the framework can detect escalating structural conditions before each crisis event. In all six cases, the framework identified elevated severity years before the failure occurred.

Between 80% and 97% of the data in each backtest case comes from federal sources: SEC 10-K and S-1 filings, FDIC call reports, NTSB investigation reports, FAA Office of Inspector General audits, FRA enforcement databases, Congressional hearing testimony, CISA alerts, and bankruptcy court filings. These are numbers organizations filed under legal obligation. The remaining data comes from official company communications and industry analysis, classified as DOCUMENTED-VERIFIED.

Temporal lead varied by case: 4.8 years for SVB, approximately 15 years for the drug shortage crisis, 5 years for Boeing 737 MAX, 6 years for East Palestine, 5 years for CrowdStrike, and 7 years for WeWork. In every case, the composite severity score crossed into elevated territory well before the crisis event, using only publicly available federal data.

The sensitivity analysis tested multiple parameter combinations per case by varying normalization ranges and composite weights across a range of configurations. The key findings (temporal lead, trajectory shape, severity band at crisis) held at 100% stability across all combinations. The absolute severity numbers shift with different parameters, but the structural reading does not change.

The framework identified six structurally distinct failure architectures: Connected Crisis (SVB), Chronic Erosion (Drug Shortage), Cascading Crisis (Boeing), Chronic plus Acute Revelation (East Palestine), Instantaneous Cascade (CrowdStrike), and Narrative Implosion (WeWork). The fact that the framework assigns different keystones, different weights, different temporal signatures, and different failure modes to each case confirms it is reading the structure rather than projecting a template.

The keystone frequency is the structural condition whose removal from the architecture would prevent the failure mode from existing at all. It is identified through a counterfactual test: if this frequency were at baseline, would the failure still have occurred? The keystone is not necessarily the highest-scoring frequency at the crisis point. It is the frequency that determines whether the failure mode is structurally available.

Phase 1 is explicitly retrospective validation. It claims structural legibility, not prediction: the ability to read the conditions that precede failure from publicly available federal data. Prospective application is the Phase 2 objective, where the framework will read live sector conditions using the same data sources. The backtests earn the right to make that attempt by first demonstrating correct readings of known failures.

Each case includes explicit falsification conditions that would disprove the structural reading. These are specific, testable statements about the data relationships the framework relies on. If those relationships do not hold, the reading fails. The existence of falsification conditions means the framework exposes itself to failure rather than insulating itself.