Deposing an Expert With Exclusion in Mind

Two depositions, one witness

The same seven hours can be spent on two incompatible examinations, and choosing between them is a strategic fork rather than a matter of emphasis.

A discovery deposition is open. The goal is to understand the opinion and find out what the case is facing. Questions are broad, favorable answers are followed up and developed, an unclear basis is clarified by asking the expert to explain, and teaching the expert about the weaknesses is an unavoidable side effect.

An exclusion-oriented deposition is closed. The goal is a record for a Rule 702 motion. Questions are precise and carry one fact each. A favorable answer is a stopping point, not a starting point — the follow-up question is what lets the expert explain and repair. An unclear basis is not clarified; the absence is pinned: “you did not do X, correct?” Teaching is minimized, because every signalled weakness is one the expert can cure before trial.

The tension is unavoidable: the questions that build a Rule 702 record are the same questions that tell the opponent what the motion will say. Counsel taking the exclusion-oriented deposition generally accepts that cost, because these motions are decided on the record and a motion unsupported by deposition admissions is decided against the movant. Counsel intending only to cross-examine at trial has the opposite incentive and may deliberately leave the best material undeveloped.

A second consequence: the expert's own counsel is taking notes throughout. A deposition that exposes a gap and leaves it uncured hands the proponent months to supplement, retain a second expert, or reframe the opinion. Taking it close to the dispositive-motion cut-off limits that window.

What the rules give the examiner before the first question

FRCP 26(b)(4)(A) is two sentences and does two things: “A party may depose any person who has been identified as an expert whose opinions may be presented at trial. If Rule 26(a)(2)(B) requires a report from the expert, the deposition may be conducted only after the report is provided.” No leave of court, no showing of need — unlike the non-testifying consultant provision at Rule 26(b)(4)(D), which requires exceptional circumstances. The trigger is identification, so the rule also reaches FRCP 26(a)(2)(C) witnesses who never write a report; the sequencing sentence is conditional on a report being required. That sequence is a floor, and it is what makes the report rather than the deposition the operative disclosure: the deposition tests a document already fixed.

On duration, FRCP 30(d)(1) limits a deposition to one day of seven hours “[u]nless otherwise stipulated or ordered by the court,” and obliges the court to allow additional time “if needed to fairly examine the deponent.” There is no expert carve-out in the text — but the 2000 Advisory Committee note anticipated the problem: “[W]ith regard to expert witnesses, there may more often be a need for additional time—even after the submission of the report required by Rule 26(a)(2)—for full exploration of the theories upon which the witness relies.” Seven hours means record time, not time in the building, and in technical cases the operative number is usually in the scheduling order rather than in Rule 30.

Under FRCP 26(b)(4)(E)(i) the party seeking the discovery pays the expert a reasonable fee for time spent responding, absent manifest injustice — a budget item, and one reason exclusion-oriented depositions are planned rather than exploratory.

Closing the opinion set before anything else

The first structural objective has nothing to do with methodology. It is to convert the report from a document into a closed set: are those all of the opinions the expert intends to offer at trial; is there any other basis for that opinion not already described; has any analysis been performed that is not reflected in the report.

The answers matter twice. They fix the target, so the proponent cannot later recast the challenge as aimed at an opinion the expert was not offering. And every later addition runs into FRCP 26(e)'s supplementation limits and the FRCP 37(c)(1) exclusion sanction.

Distinct from that, and just as valuable, are the boundary answers: that the expert offers no opinion on causation, did not evaluate damages, expresses no view on the standard of care, or assumed rather than tested a key fact. Boundaries are what keeps an expert inside their lane at trial, and they are the answers most likely to be attacked later by errata.

Basis, opinion by opinion

This is the FRE 702(b) examination, done one opinion at a time rather than globally. For each stated opinion: every document, data set, test, observation, calculation and assumption supporting it, enumerated — then the closing question establishing that the enumeration is complete.

The value of the completeness question is what it forecloses. Without it, an opinion whose basis looks thin can be defended by pointing to material the expert reviewed but did not mention. With it, the thin basis is the whole basis, and the court can evaluate sufficiency without a hearing.

What was not considered

The single most productive line. FRCP 26(a)(2)(B)(ii) requires the report to disclose the facts or data “considered,” which means the report supplies the list of what was reviewed; the deposition establishes what is missing from it.

The recurring categories: materials in the record never reviewed; depositions not read; the opposing expert's report not read; the site never visited; the product never examined; the underlying data never obtained, or obtained only in summary form. Each is a fact, verifiable against the file, answerable only yes or no.

This line has to be worked carefully, because it is where an expert most often tries to recover by explaining why the omitted material was unnecessary. The technique is to establish the omission and move on, leaving the justification to surface for the first time in a brief written by counsel, where it is visibly counsel's.

Alternative explanations not ruled out

This is the FRE 702(d) examination and it is the one that most often produces a usable exclusion argument in a causation case. What else could have produced the outcome, and what did the expert do to exclude it?

Where the method is differential diagnosis or differential etiology, the ruling-out step is the method, so an expert who cannot describe how an obvious alternative was excluded has not applied the methodology they claim to have used. Tamraz v. Lincoln Electric Co., 620 F.3d 665 (6th Cir. 2010), is the standard citation: the Sixth Circuit reversed the admission of a treating neurologist's opinion that welding fumes caused the plaintiff's parkinsonism, holding that the causal chain rested on speculation rather than on a reliably performed differential etiology.

The productive form is not “did you consider other causes,” which invites a yes. It is to name the specific alternative, establish that the expert agrees it can produce the same outcome, and then ask what was done to exclude it in this case.

Walking the analytical gap

This line serves 702(b) and 702(d) together and tracks the language of General Electric Co. v. Joiner, 522 U.S. 136, 146 (1997), where the Court said that nothing requires a district court to admit opinion evidence connected to existing data only by the “ipse dixit of the expert,” and that a court may conclude “there is simply too great an analytical gap between the data and the opinion proffered.”

The examination is built in two stages. First, establish what each cited study, data set or document actually shows — on its own terms, in the expert's own words. Then establish the distance between that and the opinion: different dose, different population, different route of exposure, different product, different time period, association rather than causation.

The expert who concedes each intermediate proposition and then asserts the conclusion anyway has supplied the ipse dixit on the record. That is a transcript a court can rule on. Arguing in a brief that the gap exists asks the court to take counsel's word for what a discipline permits.

What counsel supplied, and what the expert did personally

FRCP 26(b)(4)(C) protects communications between counsel and a report-writing expert as work product, and FRCP 26(b)(4)(B) protects drafts of the report. Three exceptions are written into (C), and they map exactly onto three of the most productive deposition lines: communications that relate to compensation, that identify facts or data the attorney provided and the expert considered, and that identify assumptions the attorney provided and the expert relied on.

So the questions are squarely permitted: who selected the documents; did the expert ask for anything not provided; did the expert ask for the materials cutting the other way; were counsel's assumptions identified as assumptions in the report. The protection is also narrower than commonly assumed — work product rather than privilege, defeasible on a showing of substantial need, and by its terms running only to experts required to provide a Rule 26(a)(2)(B) report, not to non-reporting FRCP 26(a)(2)(C) witnesses.

A related line is who performed the analysis. Where the work was done by staff and the named expert cannot explain it, the opinion is exposed under 702(b) and (d), and under Rule 26(a)(2)(B)'s requirement that the report be prepared and signed by the witness.

The two lines that are routinely mixed up

Compensation is bias material, not exclusion material. FRCP 26(a)(2)(B)(vi) requires a statement of the compensation for the study and testimony, and Rule 26(b)(4)(C)(i) removes work-product protection from communications relating to it, so the examination is available: rates by task, total billed to date, total billed for this firm or party across all matters, engagements per year, the plaintiff-defense ratio, the share of professional income from litigation work. All of it is legitimate cross-examination on credibility; almost none of it supports exclusion under Rule 702. Conflating the two is the most common error in writing on this subject, and a motion that leads with fee figures announces a weak methodology argument.

Qualification is narrower than it looks. The question is not whether the expert is qualified — an expert is always qualified in the abstract — but whether the qualification reaches this opinion: the orthopedist opining on device warnings, the economist opining on medical causation, the reconstructionist opining on human factors. The Tenth Circuit's test, from Compton v. Subaru of America, Inc., 82 F.3d 1513 (10th Cir. 1996) and applied in Ralston v. Smith & Nephew Richards, Inc., 275 F.3d 965 (10th Cir. 2001), asks whether the issue is “within the reasonable confines” of the expert's subject area. The same authority cuts the other way: a mere lack of specialization goes to weight so long as the expert stays inside those confines. The mismatch argument wins when the opinion is outside the subject area, not when the opponent's expert is more specialized.

Prior testimony, the record, and errata

FRCP 26(a)(2)(B)(v) requires a list of all cases in which the witness testified at trial or by deposition in the previous four years, and that list is the roadmap. The productive use is not “you have testified before” but “you took the opposite position on the same methodological question in X.” It is also the route to prior exclusions on the same methodology — powerful cross-examination, though not preclusive.

Record the deposition on video. FRCP 30(b)(3)(A) requires only that the notice state the method; no leave and no consent are needed, and a second party may add video to a stenographic notice at its own expense under (b)(3)(B). Video is the format in which hesitation, evasion and reliance on counsel are visible; a transcript flattens all of it. Retained experts are also frequently beyond subpoena reach, and FRCP 32(a)(4)(B) makes the deposition of a witness more than 100 miles from the place of trial usable for any purpose — in which case the video is the trial testimony.

Understand what errata can and cannot do. FRCP 30(e)(1) allows the deponent 30 days after notice that the transcript is available to review it and, if there are changes in form or substance, “to sign a statement listing the changes and the reasons for making them” — but only if review was requested before the deposition was completed. That is what the closing “read or waive” exchange is. Reasons are mandatory, and a reason-less errata sheet is defective on the face of the rule. Rule 30(e)(2) requires the changes to be attached, so the original answers stay in the record and both versions are available at trial.

Whether substantive changes are permitted at all is a genuine three-way split. The permissive view takes the rule at its word — Podell v. Citicorp Diners Club, Inc., 112 F.3d 98 (2d Cir. 1997) — and leaves the credibility problem to the jury. The restrictive view, from Greenway v. International Paper Co., 144 F.R.D. 322 (W.D. La. 1992), holds that errata correct transcription errors, reasoning that “[a] deposition is not a take home examination”; the Tenth Circuit adopted it in Garcia v. Pueblo Country Club, 299 F.3d 1233, 1242 n.5 (10th Cir. 2002). A middle group, including the First, Third, Seventh and Ninth Circuits, permits the change under the text but will not let a contradictory one manufacture a summary-judgment dispute without adequate explanation. For an expert, an errata sheet converting a concession of not having considered something into a claim of having considered and rejected it is an attempt to repair the reliability record after the fact, and which rule applies decides what happens to it.

Deposition rules discussed here: Federal Rule of Civil Procedure 30 and Rule 26.

Frequently Asked Questions

Published 2026-08-12