The Basis of an Opinion, and Its Weak Points
Arguing with an expert about their conclusion is the least productive thing an opponent can do, because the conclusion is the part they have prepared most carefully. The weak points sit further back, in what was examined, what was assumed and what was done with it.
The rule in short
An opinion rests on material, assumptions and a method, and each of the three is a point at which it can fail. Material may be incomplete or chosen by somebody with an interest in the result, assumptions may be contested facts the decision-maker has yet to resolve, and a method may be sound in general but wrong for this particular question. Attacking the conclusion directly almost never works.
Cross-examination that opens by disputing an expert's conclusion produces the expert repeating their conclusion, at greater length and with more authority. The work that changes anything happens at the base, where the opinion is standing on things that can be moved.
Material: what was and was not examined
Incompleteness is the first question. What the writer saw defines what the opinion can reach, and material that existed and was not provided is the strongest single point available.
Who selected it matters. Where the instructing party chose what to supply, the selection is part of the analysis, and an opponent is entitled to ask what the criteria were.
Representativeness in sampling. Where a conclusion rests on a sample, how that sample was drawn determines whether the conclusion extends to anything beyond it.
Copies and their limits. Some questions cannot be answered from a copy at all, which matters where originals existed and were not examined.
Material examined but not discussed. Items in the list that never reappear in the analysis are worth asking about, because the omission is either an oversight or a choice.
Timing. Material generated after the analysis, or examined at a different point in a changing situation, may support a different conclusion than the one reached.
Assumptions: the dependencies
Every assumption is a conditional. An opinion given on the basis that an account is accurate is worth exactly as much as that account, and no more.
Contested facts are the vulnerable ones. Where a decision-maker may resolve a fact differently, an opinion resting on one version of it is at risk from a finding that has nothing to do with expertise.
Unstated assumptions are worse. A dependency the report does not identify is a defect that becomes visible under examination, and the discovery does more damage than the disclosure would have.
Ask what changes the answer. The single most useful question to an expert is which of their assumptions, if wrong, would alter the conclusion, and by how much.
Alternative-assumption opinions resist this. A report addressing both versions of a disputed fact survives whichever way it is resolved, which is why careful writers prepare them and why the practice is described in what an expert report must contain.
Assumptions arrive with the instructions. Most of them come from the party who commissioned the work, which is why the instruction letter is worth reading before the analysis is.
| Point of failure | What to ask | Typical answer |
|---|---|---|
| Material | What else existed | Was not provided |
| Selection | Who chose it | The instructing party |
| Assumptions | What if it is wrong | The conclusion changes |
| Method | Why this technique | Standard in the field |
| Verification | Was it checked | Not independently |
Method: right technique, wrong question
A method can be valid and inapposite. Techniques are developed for particular questions, and applying one outside the conditions it was built for produces results that look rigorous and are not.
Standards and departures from them. Where a field has published procedures, departures should be explained, and an unexplained departure is a straightforward line of enquiry.
Error rates and confidence. A method with a known rate of error supports conclusions expressed within it, and conclusions stated more firmly than the method allows overreach.
Software and its assumptions. Analysis produced by a tool inherits that tool's assumptions, and a writer who cannot describe them is relying on something they have not examined.
Verification. Whether the result was checked by any independent route, which is the difference between a result and a result somebody has confidence in.
Exclusion is the extreme case. Where a method fails badly enough, the opinion may not be received at all, and the tests applied are set out in when an opinion is excluded.
Asking an expert which of their assumptions, if wrong, would change the conclusion produces an honest answer in almost every case, because refusing to answer it is worse than answering. It converts an unassailable opinion into a conditional one in a single exchange.
Independence and its appearance
A legitimate line, used carelessly. Questions about payment and repeat instruction are proper, and they fail when they are pressed as accusations rather than as enquiries.
Proportion of work matters. A witness whose income comes predominantly from litigation, and predominantly from one side of it, invites a question that a balanced practice does not.
Contact with the instructing party. Drafts exchanged, comments received and changes made between versions are frequently discoverable and occasionally revealing.
Independence is not the main event. It bears on weight, and a decision-maker who finds an opinion well reasoned will accept it from a witness who is regularly instructed.
Do not lead with it. An independence attack that fails leaves the expert looking wronged, and everything reasoned that follows is heard more sympathetically.
What the arrangement buys. Fee structures and their effect on the standing of an opinion are dealt with in paying for an opinion and what it buys.
Using the weak points
Concede the expertise. Accepting that the witness is qualified costs nothing and clears the ground for the arguments that can actually succeed.
Work backwards from the conclusion. Identify each step the conclusion depends on and test whether the material and the assumptions support it, rather than disputing the destination.
Get the concessions in short questions. An expert will readily agree that an opinion depends on an assumption, and that agreement is more useful than any disagreement about substance.
Use a second opinion for the method. Method arguments generally need somebody qualified to make them, which is the subject of two experts who disagree.
Know when to leave it. Where the base holds, further attack strengthens the opinion, and the better course is to accept it and argue about what it means.
The base of an opinion is where challenges succeed, and the conclusion is where they are usually aimed. The gap between those two facts accounts for a great deal of wasted cross-examination.
Material is the most productive area because it is factual. What the writer saw, who chose it and what existed that they did not see are questions with checkable answers and no room for expert judgment.
Assumptions are the most efficient area because an expert will concede them. Every opinion resting on a contested fact is conditional, and getting that on the record takes two questions.
Method arguments are the hardest to run without a second expert, because they require somebody qualified to say that the technique was wrong for this question. Where one is available they can be decisive.
The discipline throughout is to concede what cannot be won. Accepting the qualification and the good faith of a witness makes everything that follows more credible, and it costs nothing that was ever available.
Points to carry away
- Material, assumptions and method are the three points of failure.
- An opinion falls with an assumption that is not made out.
- A sound method applied to the wrong question produces a wrong answer.
- Material the expert did not see is a stronger point than material they misread.
- Arguing with the conclusion produces a restatement of it.
Questions readers ask
Why is attacking the conclusion ineffective?
Because the expert will simply restate it, with more supporting detail than the original report contained, and a decision-maker hears an authoritative witness explaining their reasoning at length. The conclusion is the part of the opinion the witness has thought about most and is best prepared to defend. The material, the assumptions and the selection of method are questions of fact and choice, where answers are checkable and where a witness has far less room to elaborate their way out of difficulty.
What is the strongest single point against most expert reports?
Material that existed, was relevant, and was not examined. It requires no expertise to make, it is checkable, and it raises the question of who decided what the writer would see. If the instructing party made that selection, the analysis is shaped by a choice the expert did not make and cannot defend. Reports are far more often vulnerable for what they did not look at than for what they got wrong in what they did.
Is it worth challenging an expert's independence?
Sometimes, and rarely first. Questions about the proportion of a witness's income from litigation, or about repeat instruction by the same firm, are legitimate and occasionally productive. But an independence attack that does not land leaves the witness looking unfairly treated, and every reasoned answer afterwards is received more sympathetically. The usual sequence is to work through material, assumptions and method first, and to raise independence only if something in those answers supports it.
Sources
- Federal Rules of Evidence — Rule 703, Bases of an Expert's Opinion Testimonylaw.cornell.edu
- Federal Rules of Evidence — Rule 705, Disclosing the Facts Underlying an Expert's Opinionlaw.cornell.edu
- Federal Rules of Evidence — Rule 702, Testimony by Expert Witnesseslaw.cornell.edu
- Federal Rules of Civil Procedure — Rule 26, Duty to Discloselaw.cornell.edu
- Legal Information Institute — Daubert Standardlaw.cornell.edu
- Federal Rules of Evidence — Rule 611, Mode and Order of Examining Witnesseslaw.cornell.edu
True Justice Record is a publication, not a law firm. This article states general rules and cites its sources; it is not advice about any particular case, and the law differs by state and changes over time.
More in Opinion That Counts as Evidence
Who Qualifies as an Expert
A witness may be qualified by knowledge, skill, experience, training or education, and any one of them suffices. The enquiry is whether this person knows something useful about this question, not whether they hold a particular qualification. A separate and prior question is whether the claimed field is one in which expertise genuinely exists.
Paying for an Opinion, and What It Buys
An expert is paid for time and analysis, and that arrangement is normal and disclosed. Fee structures tied to the outcome are treated differently and are prohibited in some fields. The questions that carry weight concern contingency, the proportion of a witness's work from one source, and the extent of instructing-party involvement in drafting.
What an Expert Report Must Contain
A usable expert report identifies the question it answers, lists the material examined, separates assumptions from findings, describes the method, shows the reasoning, states the qualifications relied on and carries the declarations the forum requires. Omissions are treated as weaknesses rather than as tidiness problems.