Concept
Fit to the Case
Proper subject matter for expert testimony
Fit to the Case
Rule 702(a) admits expert knowledge only if it “will help the trier of fact to understand the evidence or to determine a fact in issue.”1 The closing words tie the testimony to the dispute being tried, and courts call that tie fit. The earlier sections asked whether the witness has knowledge of the right kind and whether jurors need it. Fit asks a third question: does that knowledge bear on the specific question the jury must answer? Daubert v. Merrell Dow Pharmaceuticals, Inc. stressed the point, saying that Rule 702’s helpfulness standard “requires a valid scientific connection to the pertinent inquiry as a precondition to admissibility.”2
Fit is a form of relevance. Expert testimony that is unrelated to the facts at issue does not help, however sound its science.3 The trial court decides the question under Rule 104(a), alongside the witness’s qualifications and the reliability of the method as applied.4 A court applying Daubert on remand called this the second prong of the analysis, the one the Supreme Court named “fit,” and framed it as whether the testimony “logically advances a material aspect of the proposing party’s case.”5
Fit earns its own inquiry because credentials can mislead. A qualified witness speaking from valid science carries authority into the courtroom, and jurors cannot easily check whether that science reaches the question in dispute. Requiring the link before the jury hears the testimony keeps the expert’s prestige from doing work the knowledge cannot do.
Daubert’s moon illustration holds the knowledge constant and changes the question. The study of the moon’s phases yields valid knowledge about whether a certain night was dark, and when darkness is a fact in issue, that knowledge helps the jury.6 Offered to show that a person was unusually likely to act irrationally on a night when the moon was full, the same knowledge helps with nothing, because absent credible grounds linking lunar phase to behavior, the astronomy says nothing about human conduct.7 The astronomer’s qualifications and accuracy do not change between the two settings; only the fit does. Valid science for one purpose does not carry over to another, and the proponent must supply the link.
Later cases describe fit a second way, as protection against an expert who stretches an accepted premise to an unfounded conclusion.8 The Advisory Committee’s note to amended Rule 702 quotes General Electric Co. v. Joiner for the point that a court may find “too great an analytical gap between the data and the opinion proffered.”9 On that description fit overlaps with reliability, a boundary the later section on helpfulness versus reliability takes up. For now, the working question stays simple: what does this knowledge actually establish, and is that what the jury must decide?
A second kind of misfit appears when an opinion rests on factual assumptions the evidence does not support. In a wrongful-death case, an economist’s projection of lost future income is irrelevant and of no help if it assumes past earnings that have no basis in fact.10 The same is true when the reasoning is so illogical that it cannot affect the probability of any fact in issue.11 An opinion connects to the case only through its premises, and a premise missing from the record ties the testimony to a life the jury will never find.
Fit also measures the testimony against the legal standard the proponent must meet. On remand in Daubert, the court looked to California tort law, which required plaintiffs to show not merely that Bendectin increased the likelihood of injury, but that it more likely than not caused their injuries.12 For epidemiological proof, that standard demands a relative risk above two: the studies must show that children whose mothers took Bendectin were more than twice as likely to develop limb reduction birth defects as children whose mothers did not.13
The arithmetic explains the threshold. A relative risk of two means that exposure doubles the chance of the defect, so half the defects among exposed children would have occurred anyway, and no child’s injury would be more probable than not the drug’s. Only a risk above two tips the balance. The plaintiffs’ epidemiologists asserted a statistically significant relationship between Bendectin and birth defects, but none stated that the relative risk exceeded two.14 Their studies therefore addressed whether the drug increases the likelihood of injury, which the standard says is not enough, and not whether it more likely than not caused these injuries, which the standard requires. Sound science that answers a nearby question still does not fit. A study reporting a relative risk above two would speak directly to the standard, which is why fit looks to the legal threshold and does not stop at asking whether the science is good.
Eyewitness research applies the same test one identification at a time. In the Stevens appeal, the trial court admitted a psychologist’s testimony on cross-racial identification, weapon focus, and stress because each was tied to the facts: the victims were white and the assailant was black, the assailant brandished a gun, and both victims were under stress.15 The inquiry follows the approach of United States v. Downing: research about memory in general does not fit unless the proponent ties it to features of the identification actually at issue.16
The same case shows the principle from the other side. The trial court excluded testimony about the suggestiveness of the wanted board, the “relation-back” question, and the correlation between confidence and accuracy, finding no fit.17 On appeal, the confidence testimony fared differently. Unlike the wanted-board and relation-back proffers, the psychologist’s explanation of the confidence and accuracy studies could help the jury assess the identifications, because it is fairly counterintuitive that witnesses so often profess considerable confidence in erroneous ones.18 The testimony touched a feature of the identifications on trial, and it corrected an assumption jurors might otherwise bring to it. Whether eyewitness research helps jurors at all is a broader question, taken up in the later discussion of contested subjects.
The fourth step does the real work. It is where testimony that looks sound separates from testimony that helps: the lunar science and the studies showing an increased risk were both valid, and each answered a question one step away from the one the jury faced. The third and fifth steps catch the other two failures, an opinion floating above the record and general research that no one has tied to the identification being tried.