Early access — cascade metrics are real (derived from canonical token telemetry); the operator field is a curated seed. Learn more about the data

Operator Separability

0Concept

Whether different operators produce measurably different metric profiles. If all operators look the same, the measurement has no discriminative power.

Last updated: 2026-09-01Spec: SigRank Standard v1.0 (proposed)

Definition

Whether different operators produce measurably different metric profiles. If all operators look the same, the measurement has no discriminative power.

Inputs

Metric profiles for multiple operators.

Derived variables

Between-operator variance, within-operator variance, F-ratio.

Claim

If the measurement is valid, different operators should produce significantly different metric profiles.

Test

Compare between-operator variance to within-operator variance. Compute F-ratio and significance.

Observable

F-ratio, p-value, effect size.

Falsifier

Between-operator variance is not significantly different from within-operator variance.

Evidence

Field data shows operator-level differences in yield, SNR, leverage, velocity.

Limitations

Operators using the same tools and workflows may have similar profiles. Separability depends on task diversity.

Lineage

Psychometric validation methodology, SigRank field data. Architecture: mos2es.com/architecture.

Cross-references