Kumho Tire Co. v. Carmichael

After a fatal blowout, the plaintiffs' case rested on a tire-failure analyst who would testify, from visual and tactile inspection, that a defect rather than abuse caused the failure. The Eleventh Circuit had held Daubert inapplicable because the expert's knowledge was technical experience, not science. The question was whether a judge's gatekeeping obligation stops at the border of 'scientific' testimony — and if it does not, how the Daubert factors apply to experience-based disciplines.

Rule 702 makes no relevant distinction between scientific, technical, and other specialized knowledge: the gatekeeping obligation covers all of it. A trial court may consider the specific Daubert factors where they are reasonable measures of reliability, but the inquiry is flexible — the factors do not constitute a definitive checklist, and the law grants the trial judge broad latitude, reviewed only for abuse of discretion, both in deciding how to test reliability and in the ultimate ruling.

Applying that standard, the Court examined the analyst's actual method: a two-factor visual inspection protocol of his own devising, whose criteria he could not show other analysts used and whose error rate no one had measured. The problem was not that experience-based tire analysis is inherently unreliable — it was that this expert could not show his particular application of it was. The Court asked the question that now frames every challenge: whether the expert employs in the courtroom the same level of intellectual rigor that characterizes the practice of an expert in the relevant field.

The exclusion of the expert was reinstated: the district court had acted within its discretion. Together with Daubert and Joiner, Kumho completed the trilogy that governs expert admissibility — and its extension of gatekeeping to 'technical and other specialized knowledge' is the reason a digital forensics examiner, a damages analyst, or a cybersecurity practitioner can be Daubert-tested despite never running a laboratory experiment.