An expert report has two audiences. The first is the reader you want to persuade: opposing counsel evaluating settlement, a mediator, a judge. The second is the reader trying to destroy it: the opposing expert and the attorney preparing to cross-examine its author. A defensible report is written for the second audience, and it turns out that reports written to survive attack are also the ones that persuade.
Here is what separates a defensible electrical engineering report from a vulnerable one.
The opinion follows the evidence, visibly
The core structural requirement is traceability. Every opinion in the report should connect, on the page, to specific evidence: a photograph, a measurement, a document, a standard, a calculation. The reader should be able to start at the conclusion and walk backward through the reasoning to the underlying facts without gaps.
Vulnerable reports invert this. They state conclusions first and cite evidence selectively in support, leaving the reasoning implicit. On cross-examination, implicit reasoning becomes "isn't it true that you assumed" and there is no good answer, because the report does not show the work. A report that shows its work makes the deposition shorter and the expert harder to move.
Competing hypotheses are addressed, not ignored
The most reliable marker of a strong causation report is what it does with the explanations it rejects. A defensible report identifies the plausible alternative causes, states what evidence would be expected if each were true, and shows why the evidence supports one and not the others. This is the substance of a systematic elimination methodology, and it is what distinguishes analysis from advocacy.
A report that considers only the retaining party's preferred theory hands opposing counsel its best exhibit. The first question at deposition becomes a list of the alternatives the expert never mentioned, and each one the expert cannot rule out on the record erodes the stated opinion. Addressing alternatives in the report costs pages; ignoring them costs the opinion.
The methodology is stated and standard
Electrical failure analysis is not improvisation. There are recognized methods: relevant provisions of the National Electrical Code and equipment standards, listing and testing requirements, published failure mode literature, accepted examination and documentation protocols, and where fire is involved, the scientific method framework that fire investigation standards describe. A defensible report names the methodology it applies and applies it consistently.
This matters because admissibility challenges target methodology. The question a court asks is not whether the expert is smart but whether the opinion is the product of reliable principles and methods, reliably applied. A report that can point to its methodology by name, and show each step performed, is built for that challenge. A report that rests on experience alone, without showing how the experience was applied to the facts, is built for trouble.
The report stays inside the expert's lane
Every expert has a boundary of competence, and the defensible report respects it explicitly. An electrical engineer can evaluate the electrical equipment, the failure mode, the code compliance, the installation quality, and whether the electrical evidence supports or contradicts a proposed cause. Where a question belongs to another discipline, such as fire origin determination, metallurgy, or human factors, the strong report says so and stops.
Attorneys sometimes push experts toward broader opinions because broader is more useful. It is not. An opinion outside the expert's qualifications is the single easiest thing to strike, and a struck opinion damages the credibility of every opinion that remains. The expert who declines to overreach in the report is protecting the client's case, not limiting it.
Language discipline: precise, qualified where honest, absolute nowhere
Words get read back. "Always," "never," "impossible," and "certain" are gifts to cross-examination, because engineering rarely supports them and a single counterexample collapses the sentence. The defensible report states opinions with the degree of certainty the evidence actually supports, quantifies where the evidence permits, qualifies where it does not, and never trades accuracy for rhetorical force. Confidence in a report comes from the structure of the evidence, not from adjectives.
The same discipline applies to facts. Dates, model numbers, measurements, and photograph references get verified, because a trivial factual error found by the other side becomes the theme of the deposition: if the expert got this wrong, what else is wrong?
What retaining counsel should look for
These are the tests I apply to my own reports before they go out. Can every opinion be traced to evidence on the page? Are the alternatives addressed and the elimination shown? Is every opinion inside the expert's demonstrated qualifications? A report that passes those tests will not win the case by itself. But it will still be standing after the other side has done its worst, and in expert-driven disputes, the last report standing tends to be the one that decides the outcome.