Case Comparison Matrix¶
Comparison matrix — instantiates Archetype Overmatching Guardrail
Arrays the case's features against the candidate archetype and its nearest neighbors in one grid, so where the case actually lines up — and where it doesn't — is visible at a glance.
Case Comparison Matrix is the side-by-side artifact of the guardrail: a table with features down the side and columns for the case, the candidate archetype, and its nearest rival patterns. Filling every cell turns "which pattern does this resemble most?" from a feeling into a spatial read — the column whose rows line up best is the better fit, and any row where the favoured archetype diverges from the case is impossible to overlook. Its defining property is that it holds several candidates at once, so the match is chosen against alternatives rather than confirmed in isolation.
Example¶
A city faces a spike in harms from a new stimulant and officials reach for the familiar frame: "this is another opioid epidemic — run the opioid playbook." Before committing, analysts build a comparison matrix. Rows are the discriminating features: route of use, dependence pharmacology, whether a reversal agent exists (as naloxone does for opioids), the supply chain (diverted prescriptions versus illicit synthesis), and which interventions have worked. Columns are the current case, the opioid-crisis archetype, and two neighbours — a stimulant wave and an alcohol-harm pattern. Filled in, the grid shows the case matches the opioid column on overdose visibility but diverges on the load-bearing rows: no reversal agent exists and there is no prescription-diversion on-ramp. Row by row it lines up better with the stimulant-wave column. The matrix reframes the response before the opioid playbook — naloxone distribution, prescriber limits — is aimed at a problem it does not fit.
How it works¶
Multiple candidate archetypes sit in adjacent columns against the same feature rows, so comparison is spatial rather than argued: the eye locates the best-aligned column and immediately sees the rows where the preferred archetype breaks from the case. The rows that carry the weight are the discriminating features — the ones that differ across candidates — because rows every pattern shares tell you nothing about which one fits.
Tuning parameters¶
- Column set (which neighbours) — including the genuinely competing rival archetypes is what makes the matrix discriminating; omit the true best-fit neighbour and the grid merely confirms the first guess.
- Row selection — features that separate the candidates, not features they all share; shared rows are wasted space.
- Cell content — checkmarks (fast to scan) versus short evidence notes (preserve nuance); the trade is legibility against fidelity.
- Ordering — putting the most discriminating rows at the top so the divergences are seen first.
When it helps, and when it misleads¶
It shines when several archetypes compete and the decision hinges on which fits best; the grid is legible and hard to hand-wave past. It misleads when discriminating rows are left out (an all-shared-features matrix flatters every candidate equally[1]), when columns are cherry-picked to exclude the inconvenient best-fit rival, or when a tidy grid lends more apparent rigor than the underlying cell judgements deserve. The guarding discipline is to choose rows for discriminating power and to force in the strongest rival column even when it is the one you would rather not see.
How it implements the components¶
case_evidence_profile— the case column is a compact profile of the case's actual features across the chosen rows.neighbor_comparison— placing the candidate archetype beside its rivals in adjacent columns is the neighbour comparison, made visual and simultaneous.
It does NOT weight or total the alignment into a score — that's Pattern Fit Scoring Rubric; it does not actively generate disconfirming cases — that's Counterexample Search Session; and it does not record the reasoning for the winning column — that's Precedent Distinction Memo.
Related¶
- Instantiates: Archetype Overmatching Guardrail — the matrix is the guardrail's picture: the case set against every pattern it might belong to, at once.
- Sibling mechanisms: Differential Pattern Review · Pattern Fit Scoring Rubric · Pattern Fit Checklist · Anti-Pattern Review · Counterexample Search Session · Precedent Distinction Memo · Red-Team Pattern Match Review · Review Queue · Decision Confidence Label
Editorial Notes¶
Form Classification¶
Form family: Assessment, Review & Assurance
Rationale: The mechanism evaluates one case against adjacent candidate archetypes on shared discriminating features and exposes where the preferred fit holds or breaks, so its operative output is a classification assessment.
Nearest alternative: Analysis, Modeling & Optimization — Spatial comparison is analytical, but the mechanism culminates in a bounded case-fit finding rather than a general model.
Review outcome: Adjudicated after independent review; high confidence.
Origin Attribution¶
Primary origin: Organizational & Management Science
Origin pattern: Cross-disciplinary synthesis
Present-day reach: Multi-domain
Rationale: Decision and review practice established side-by-side matrices that compare one case against competing candidate patterns.
Related originating lineages:
- Operations Research — Multi-criteria analysis supplies discriminating rows and explicit comparison against the strongest rival.
Review resolution: Organizational decision practice is primary because the mechanism externalizes alternatives and criteria in a comparison grid for collective review. Operations research contributes systematic option-by-criterion comparison, while the source's unscored, tension-preserving form is a cross-disciplinary synthesis rather than a canonical optimization matrix.
Encyclopedia synthesis: The exact catalogued form synthesizes established practice rather than reproducing a single standard historical label.
Review outcome: Researched adjudication after independent review; medium confidence.
Sources consulted:
Notes¶
The matrix is the artifact; Differential Pattern Review is the reasoning process that works the same case-versus-neighbours comparison as a discipline. They pair naturally — the matrix is often the worksheet a differential review fills in and argues over.
References¶
[1] Heuer, Richards J., Jr. Psychology of Intelligence Analysis. Center for the Study of Intelligence, Central Intelligence Agency (1999). Explains that evidence consistent with every competing hypothesis has no diagnostic value because it does not distinguish among them. registry ↩