Selective Admission Band Protocol¶
Procedure — instantiates Selectivity-Window Calibration
Uses an allowable strictness band to admit targets while minimizing qualified exclusions and non-target admissions.
A single hard cutoff forces false certainty on the exact cases that deserve it least — the ones sitting right on the line. Selective Admission Band Protocol is the decision-time procedure that replaces the line with a band: cases clearly above the band are auto-admitted, cases clearly below are auto-declined, and the uncertain middle is routed to human review or a defined fallback rather than being pushed through a coin-flip boundary. Its defining feature is that it executes a band on live cases and owns the route-to-review path for the ambiguous zone — it is the mechanism that applies the rule at the moment of decision, as distinct from the document that authored the rule. The band exists precisely so that a qualified target is not excluded for missing a threshold by a hair, and a marginal non-target is not admitted for clearing it by one.
Example¶
A university reworks undergraduate admissions after noticing that a single composite-index cutoff was both rejecting strong applicants who fell a point short and admitting weak ones who scraped a point over. The protocol installs a band. Applicants with a composite index above 92 are auto-admitted; those below 70 are auto-declined; everyone from 70 to 92 lands in the review band and goes to a holistic committee — the fallback for cases the number alone cannot resolve. A first-generation applicant with a 74 and an extraordinary essay is no longer discarded by arithmetic, and a 71 with nothing else going for them is no longer waved in by it. Each decision is logged so the downstream drift monitor can see whether the band is still doing its job.
How it works¶
- Score the case and compare to band edges. Apply the specification's auto-admit and auto-decline edges to each incoming case.
- Auto-decide the clear cases. Admit those clearly above and decline those clearly below without human handling, reserving attention for genuine ambiguity.
- Route the middle to fallback. Send cases inside the review band to human judgment, abstention, or a request for more information — the explicit non-binary path.
- Log outcomes for monitoring. Record admits, declines, and routed cases so false-exclusion and non-target-admission rates stay visible over time.
Tuning parameters¶
- Band width — how wide the review zone is. A wider band makes fewer hard errors at the edge but sends more cases to costly human review.
- Edge placement — where the auto-admit and auto-decline thresholds sit. Shifting them trades qualified exclusions against non-target admissions directly.
- Fallback destination — review, abstain, or request-more-information. Richer fallbacks handle ambiguity better but cost time and capacity.
- Capacity coupling — whether band edges flex with throughput demand. Flexing preserves capacity but is the exact knob through which a band quietly creeps.
When it helps, and when it misleads¶
Its strength is replacing false certainty at the boundary with a routed judgment — cutting both the qualified applicant wrongly excluded and the marginal one wrongly admitted, by refusing to pretend the edge cases are clear.
Its failure mode is window creep: under capacity or throughput pressure the band silently widens or shifts, sweeping in non-targets or excluding qualified people while the stated rule looks unchanged, an effect studied in personnel selection as banding and its adverse-impact consequences.[n1] The classic misuse is collapsing the review band back into a hard cutoff to clear a backlog, discarding the whole point of the mechanism. The guarding discipline is to monitor the false-exclusion and non-target-admission rates the band is meant to hold down, and to protect the review and appeal path from being quietly optimized away.
How it implements the components¶
operating_window_rule— it executes the admit / review / decline bands on live cases, applying the rule at the point of decision.fallback_or_nonselective_mode_guard— it routes ambiguous, in-band cases to human review or abstention instead of forcing them through a binary gate.
It applies the band but does not author or version it — writing the stratified_window_table and window_version_register is Operating Band Specification's job — and it does not track whether the band is drifting, which is Window Drift Control Chart's selectivity_drift_recalibration_loop.
Related¶
- Instantiates: Selectivity-Window Calibration — this procedure is where the window is applied to real cases at decision time.
- Consumes: Operating Band Specification supplies the band edges and actions this protocol executes.
- Sibling mechanisms: Selectivity Curve Sweep · Operating Band Specification · Challenge-Panel Cross-Reactivity Test · ROC or Precision–Recall Surface Review · Bycatch Audit · Window Drift Control Chart
Editorial Notes¶
Form Classification¶
Form family: Control, Automation & Runtime
Rationale: Selective Admission Band Protocol operates as a live operational control that automatically routes, enforces, adapts, or responds during execution because it uses an allowable strictness band to admit targets while minimizing qualified exclusions and non-target admissions.
Independent corroboration: The frozen evidence defines Selective Admission Band Protocol as 'Uses an allowable strictness band to admit targets while minimizing qualified exclusions and non-target admissions', so its operative form is Control, Automation & Runtime.
Nearest alternative: Decision, Gate & Allocation — Selective Admission Band Protocol includes features of a case-specific gate, selection, routing, prioritization, or resource disposition, but its defining operation is a live operational control that automatically routes, enforces, adapts, or responds during execution.
Review outcome: Independent reviewer agreement; medium confidence.
Origin Attribution¶
Primary origin: Statistics & Experimental Design
Origin pattern: Cross-disciplinary synthesis
Present-day reach: Multi-domain
Rationale: Balancing qualified exclusions against non-target admissions across a threshold band is statistical classification and decision-threshold design.
Related originating lineages:
- Data Science & Analytics — Data science, analytics, and operational monitoring supplies a parallel or contributing lineage for the mechanism's defining operation: uses an allowable strictness band to admit targets while minimizing qualified exclusions and non-target admissions.
- Mathematics — Mathematical modeling, proof, and abstract-structure practice supplies a parallel or contributing lineage for the mechanism's defining operation: uses an allowable strictness band to admit targets while minimizing qualified exclusions and non-target admissions.
- Medicine & Healthcare — Diagnostic admission rules formalize sensitivity-specificity tradeoffs and indeterminate bands.
- Operations Research — Constrained optimization frames the operating point as a tradeoff among competing error costs.
- Public Administration & Policy — Eligibility systems use bounded discretion around cutoffs to control both exclusion and leakage.
Review resolution: The blind reviewers agree that statistics_experimental_design is the primary origin and differ only on alternate origin disagreement, origin mode disagreement. I preserve every independently explained alternate from both records rather than imposing a numeric cap. I retain cross_disciplinary_synthesis because the combined record shows material contributions from several lineages. The broader reach of multi_domain records portability separately from historical provenance, and encyclopedia_synthesis=true preserves the affirmative synthesis judgment where either reviewer identified one.
Encyclopedia synthesis: The exact catalogued form synthesizes established practice rather than reproducing a single standard historical label.
Review outcome: Reconciled after independent review; medium confidence.
Notes¶
[n1] In personnel selection, banding treats scores within a defined range as effectively equivalent rather than rank-ordering them by a single cut score; the practice is analyzed alongside adverse impact (disproportionate exclusion of a protected group), the standing risk when a band's edges shift under pressure. ↩