Capstone Kit · Master Checklist · July 2026

The Master Checklist: Running the Standard, the Four Factors, and the Presumption in Order

Six modules, one sequence. The whole architecture on one page — each phase, the fork it turns on, and where the series answers it in depth.

Each module in this series ended with its own checklist: the standard, each of the four factors, the presumption. Those checklists remain the detail layer. This is the layer above them — the order in which the pieces run, and the decision forks between them. The sequence is not decorative. Most losing bias cases documented in this series lost on order, not on facts: the pattern discovery sought before the threshold showing, the metrics never assembled before the motion, the standard never planted before the evidence was spent.

Work top to bottom. Each phase names the module checklist that carries its detail.

Phase 0 — The free evidence, before anything is served

  • Read the denial report cold against the facial screen. Ten tells, no discovery required — no stated methodology, unsourced principles (“says who?”), invented standards, boilerplate, categorical certainty. Annotate into a numbered deficiency inventory: it is the exhibit every later motion cites. (Detail: the Flawed-Report Checklist.)
  • Pull the insurer’s and the expert’s judicial record. Published opinions criticizing this insurer or this reviewer, by name, are discovery you already have — zero cost, no motion, available before filing. (Detail: the Reasonable-Measures Checklist, move 5.)
  • Fix the posture. First-party bad faith and ERISA run the same architecture through different vehicles: in first-party, the factors feed the duty to investigate and defeat the genuine-dispute defense; in ERISA, they feed the conflict-weighing dial and the burden-shift of Demer v. IBM Corp. LTD Plan. Decide which vocabulary the record is being built in before drafting anything.

Phase 1 — Plant the standard

  • Plead bias as a fair inference, not a corrupt motive. The standard you fail to assert is the one the court picks for you. The inference-of-bias standard — a fair inference of financial conflict from structure, no proof of a guilty mind — is the test everything below feeds. Bagramyan is the cautionary study: relational admissions with no standard and no methodology record went nowhere.
  • Argue the pedigree and the inversion. The neutrality rule is a century old (Tumey), and Commonwealth Coatings inverts the insurer’s safest argument: fewer safeguards around the decider call for more scrutiny, not less. Where the insurer both investigates and adjudicates the claim, the retained expert has the fewest safeguards of all.
Fork — the court resists the standard: do not fall back to “dishonest selection” or actual-bias framing; argue the gradient (appearance → substantial likelihood → intolerable risk) and locate the structure on it. (Detail: Building a Bias Record; the Standards Implementing Kit.)

Phase 2 — Factor 1: the modest showing

  • Get two numbers: compensation and volume. Dollars first, claim counts as backstop; a multi-year window; directed at all four levels — insurer, expert, the expert’s billing entity, and the vendor. Frame every request around the insurer’s affirmative investigation duty, never credibility. (Detail: the Compensation-Discovery Checklist.)
  • Never demand the tax return itself. Invoices, payment records, 1099s — the figure without the privacy objection. Brzezinski is the study in skipping the discipline.
Fork — the numbers are refused: the burden objection to producing the insurer’s own expert-payment records is rarely substantiated, and “we don’t keep track” is not a wall — it is Phase 3’s trigger.

Phase 3 — Invoke the presumption

  • Spend the metrics to shift the burden. The modest showing is what it takes: enough compensation and volume raises the fair inference, and the burden shifts to the insurer to prove its expert was actually neutral — the rule of Demer.
  • Make silence cost. An insurer that could have kept records showing its experts’ neutrality in practice, and didn’t, has missed its opportunity to rebut the inference. Say it early and brief it exactly that way. (Detail: the Presumptions & Burdens Checklist.)
Fork — the insurer answers with a safeguard: match the rebuttal to the conflict. A wall between claims and finance answers the company’s structural conflict; it says nothing about the retained expert’s financial dependence. Keep the factors distinct and make the rebuttal meet the one you raised.

Phase 4 — Factors 2 and 3: fortify the inference

  • Get the files, not the people. Other insureds’ claim files involving this expert — redacted, participant-keyed, purpose-named — are where the pattern lives. Colonial Life settled that the files are reachable; Tilem shows the mismatch that loses them; Shirley shows what asking for names instead costs. Delay any request for PII until absolutely needed and fully defensible. (Detail: the OICF and PII Checklists.)
  • Run the sources demand. For every principle the report applies: identify the source, produce it, test it against the claimant’s facts. An unsourced premise is presumptively invented until the expert identifies its authority — and under Sargon, an opinion whose reasons are unsupported is not evidence at all. (Detail: the Flawed-Report Checklist; the Factor 3 Kit.)
  • Reconstruct the process timeline. Referral question → file sent → drafts → iterations → final report → denial. The irregularities live between the steps.
Fork — a single claim’s irregularities plus metrics still feels thin: it may be. The winning structure is multiplicative — metrics × irregularities × pattern — and the pattern comes from the cross-claim reports (Phase 4, first box) or from the insurer’s missing measures (Phase 5).

Phase 5 — Factor 4: the system the insurer never built

  • Argue the omission, not just the commission. What did the insurer build — before this claim — to keep expert evaluations neutral? Selection vetting, outcome tracking, rotation, remediation. Under Glenn, real measures are structural and continuous; their absence leaves the Phase 3 inference unrebutted.
  • Treat the paper as a claim, not a fact. A “principles” document is an interested party’s factual assertion. The response is a records request: show each claimed measure operating — audits, evaluations, any expert ever disciplined or dropped under it. Fessenden is the study in what happens when nobody asks.
  • Don’t let the vendor be the wall. “The vendor pays the doctor” relocates the dependence one step; ask for the vendor relationship itself. (Detail: the Reasonable-Measures Checklist.)

Phase 6 — Meet the endgame

  • Know the summary-judgment mechanics. The insurer moving on genuine-dispute grounds must make its own initial showing before any burden shifts; you need a triable issue, not trial-quality proof. A dispute manufactured by an unreasonable or dishonestly selected expert is not genuine.
  • Watch the recurring deflections. “Prove actual bias” (the inference is the standard); “one declaration answers everything” (match rebuttal to conflict); the supplemental-briefing ambush (a safeguards document produced only after your motion is on file — demand the custodian’s deposition; if withheld from served discovery, move to exclude). (Detail: the Presumptions & Burdens Checklist and Kit.)

Phase 7 — Protect the record

  • Get rulings on objections to your key declarations. Centerpiece evidence left in limbo is evidence the court is free to pass over.
  • Preserve every factor argument in the opening brief. Points raised late are forfeited. The architecture only works whole; make the record show the whole of it.
Where this checklist stops. The phases above are the sequence. The instrument for what comes after discovery — the four factors converted into declaration paragraphs, computable metrics, and coding rules your AI can run — is the companion SKILLS File (free) and, as its stage-by-stage operating manual, the Evaluation Workflow (paid).

Get the Implementing Kits →   Open the SKILLS File →   Read the series recap →

Related

The series this checklist sequences: The Paradigm, Operationalized. The record-building chart it supersedes for sequencing: Building a Bias Record. The evaluation instrument that runs on the record this sequence produces: The SKILLS File.

Assembled from the six published module checklists and their kits. Seminal and case-study authority only — Demer v. IBM Corp. LTD Plan, 835 F.3d 893 (9th Cir. 2016); Metropolitan Life Ins. Co. v. Glenn, 554 U.S. 105 (2008); Sargon Enterprises, Inc. v. University of Southern California, 55 Cal.4th 747 (2012); Tumey v. Ohio, 273 U.S. 510 (1927); Commonwealth Coatings Corp. v. Continental Casualty Co., 393 U.S. 145 (1968). The full authority behind each phase is carried in the module Implementing Kits. Educational and informational only; not legal advice.