The center of gravity for expert-evidence law remains the Supreme Court’s trilogy—Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579 (1993); General Electric Co. v. Joiner, 522 U.S. 136 (1997); and Kumho Tire Co. v. Carmichael, 526 U.S. 137 (1999). What has changed, and now deserves pride of place in any modern analysis, is the 2023 amendment to Federal Rule of Evidence 702. Effective December 1, 2023, Rule 702 makes explicit that expert testimony is admissible only if the proponent demonstrates—by a preponderance of the evidence—that the opinion satisfies every part of the rule, including that “the expert’s opinion reflects a reliable application of the principles and methods to the facts of the case.” See Fed. R. Evid. 702 (as amended 2023); Fed. R. Evid. 104(a). The Supreme Court’s transmittal order further provides that the amendment governs proceedings commenced after December 1, 2023 and, where just and practicable, pending matters.
Florida now walks in step with that federal baseline. In 2019, the Florida Supreme Court exercised its rulemaking authority to adopt the Legislature’s Daubertamendments to sections 90.702 and 90.704, replacing Frye and foreclosing the old “pure opinion” pathway. In re Amendments to the Florida Evidence Code, 278 So. 3d 551 (Fla. 2019). The Court’s action makes clear that Florida courts must apply Daubert’s reliability-and-fit analysis to all expert testimony, not merely “new or novel” science. Florida’s appellate decisions fill in the practical contours. Perez v. BellSouth Telecommunications, Inc., 138 So. 3d 492, 496–98 (Fla. 3d DCA 2014), explains that the Legislature intended conformity with Daubert, Joiner, and Kumho and that “pure opinion” testimony is prohibited. Kemp v. State, 280 So. 3d 81, 91–94 (Fla. 4th DCA 2019), then applies § 90.702 with teeth, reversing a vehicular-homicide conviction where an “expert” braking opinion rested on nothing more than a visual read of crush damage; the opinion lacked sufficient facts or data and a reliable method applied to the case.
The federal appellate framework—particularly in the Eleventh Circuit—shows how these principles decide cases. General Electric Co. v. Joiner established two pillars that practitioners ignore at their peril. First, abuse of discretion governs appellate review of Rule 702 rulings; the trial judge has a “range of choice.” 522 U.S. at 141–46. Second, a court may find “too great an analytical gap” between data and opinion and exclude testimony connected to the evidence only by the expert’s ipse dixit. Id. at 146–47. The Eleventh Circuit applies those pillars with rigor. In United States v. Frazier, 387 F.3d 1244, 1259–61 (11th Cir. 2004) (en banc), the court reiterated that Joiner’s standard “requires that we not reverse … unless the ruling is manifestly erroneous,” and that district judges enjoy “considerable leeway” in gatekeeping. That deference is not theoretical; it sustains both exclusions and admissions that are well reasoned.
The practical import for reliability and “fit” comes through sharply in toxic-tort and medical-causation cases. In Rink v. Cheminova, Inc., 400 F.3d 1286, 1288–91 (11th Cir. 2005), causation theories about pesticide degradation failed Daubertscrutiny; once the experts were excluded, summary judgment followed because plaintiffs lacked admissible proof of a necessary element. Rink is routinely cited for the proposition that Daubert guards juries from “speculative, unreliable expert testimony.” Id. at 1291. Allison v. McGhan Medical Corp., 184 F.3d 1300, 1304–06 (11th Cir. 1999), provides the companion blueprint: after a multi-day Dauberthearing in silicone-implant litigation, exclusion of causation experts—and summary judgment—were affirmed, with the court emphasizing the proponent’s preponderance burden and the “range of choice” afforded to trial judges.
“Fit” is not a slogan; it is a filter. McDowell v. Brown, 392 F.3d 1283, 1299–1301 (11th Cir. 2004), affirmed exclusion where an expert’s case-specific causation opinion leapt from a general medical proposition (earlier intervention is better) to a specific causal conclusion about the plaintiff’s outcome without research tying the time interval to outcome for the condition at issue. The 2023 amendment to Rule 702 dovetails with McDowell: not only must a method be reliable in the abstract; the expert must show a reliable application to the case facts, and the proponent must establish that by a preponderance under Rule 104(a).
Florida cases now mirror these federal demands. Perez rejects correlation and anecdote as a substitute for method, makes clear that Daubert applies across the board (not only to novelty), and underscores legislative intent to prohibit “pure opinion.” 138 So. 3d at 496–98. In Kemp, the Fourth District catalogued § 90.702’s text and concluded the braking opinion “was not based upon legally sufficient facts or data” and failed the statute’s reliability requirements; because the State leaned on that opinion in closing, the error was not harmless. 280 So. 3d at 91–94.
Experience-based experts receive the same scrutiny. The Eleventh Circuit’s en banc decision in Frazier warns that jurors may attach “talismanic significance” to expert credentials; gatekeeping prevents unearned authority from masking unreliable reasoning. 387 F.3d at 1259–61. Florida’s Kemp is the in-state analogue: an officer’s “I see a dip, so there was braking” opinion was inadmissible without testable, articulated principles reliably applied to measurable facts.
Differential diagnosis requires both candor and rigor. Kilpatrick v. Breg, Inc., 613 F.3d 1329, 1336–44 (11th Cir. 2010), affirmed exclusion where the specific-causation opinion rested on temporal sequence and an incomplete rule-in/rule-out analysis; the court stressed that a differential diagnosis presumes general causation and must be applied in a scientifically disciplined way to be reliable.Kilpatrick also illustrates a recurring Eleventh Circuit theme: background risk and unknown causes matter, and ignoring them undermines reliability. Id. at 1341–43.
With the doctrine framed, the methodology chain that courts expect can be stated plainly. General causation asks whether the agent is capable of causing the disease in the relevant population. Exposure assessment asks whether the plaintiff was exposed at a dose, by a route, and for a duration consistent with the agent’s known effects. Specific causation asks whether, given competing causes and background risk, the agent more likely than not caused this plaintiff’s disease, supported by a disciplined and transparent differential diagnosis. When any link in that chain is missing—or when there is an unexplained leap from underlying studies to the case facts—Joiner allows trial judges to say “stop.”
Florida practice also imposes trial-management guardrails that pair with Daubert. Section 90.704 restricts disclosure of inadmissible “basis” evidence unless its probative value substantially outweighs the risk of prejudice. That provision channels how experts may present their reasoning to juries after the gatekeeping decision and reduces the risk that inadmissible hearsay or undisciplined materials will backdoor their way to the fact-finder. Section 90.705 similarly matters in trial practice: while an expert may be examined regarding the underlying facts or data, the court retains discretion to require that such basis information be disclosed on voir dire outside the jury’s presence when necessary to avoid prejudice and confusion, and counsel should plan examinations accordingly.
Procedure and Strategy for Daubert Motions: Timing, Record-Building, Hearing Mechanics, and Preservation. The strategic question in both Florida and federal court is not whether to file a Daubert motion, but when and how to win one. Timing is dictated by disclosure mechanics. In federal court, the most effective challenges follow Rule 26(a)(2) reports and a focused deposition that exposes gaps between the expert’s stated methods and the application to the case facts.
In Florida practice, the same cadence applies once the expert’s report, affidavit, or Rule 1.280 disclosures supply the proponent’s methodology. A record built at deposition is often the lifeblood of the hearing: it should pin down precisely what the expert did to move from each study or measurement to each conclusion, and what the known limitations, error rates, and confounders are. The Eleventh Circuit’s decisions reveal why this sequencing matters.
In Rink, exclusion of expert testimony demolished the causation element and made summary judgment inevitable; the appellate court affirmed both because the record showed how the opinion failed the reliability test and why the gap could not be bridged without reinventing the case. 400 F.3d at 1288–91. The same pattern appears in Allison, where a multi-day evidentiary hearing produced a detailed reliability analysis that the Eleventh Circuit sustained on abuse-of-discretion review. 184 F.3d at 1304–06.
A winning motion is short on rhetoric and long on the method-to-conclusion chain. Judges repeatedly emphasize that credentials are not a substitute for methodology. Quiet Technology DC–8, Inc. v. Hurel–Dubois UK Ltd., 326 F.3d 1333, 1335 (11th Cir. 2003), affirmed a district court that engaged with the mechanics of a computational fluid dynamics model rather than the expert’s résumé and praised the “admirable” execution of the gatekeeping function notwithstanding scientific complexity.
The hearing should therefore proceed in the same granular way: identify what was measured, explain how it was measured, show how each analytical step was validated, and tie the result to the disputed facts. Where the proponent’s application of a method drifts from the precepts that make that method reliable, the 2023 amendment to Rule 702 supplies the doctrinal hinge because it requires a preponderance showing that the expert has reliably applied reliable principles and methods to sufficient facts and data. The Eleventh Circuit’s “range of choice” deference then does the rest for a carefully reasoned ruling. Frazier, 387 F.3d at 1259–61.
Fit is a distinct, independent requirement and must be proved on the same preponderance standard. McDowell, 392 F.3d at 1299–1301, shows how a court tests fit. There, the plaintiff’s medical causation theory made an inferential jump from the general proposition that earlier intervention is better to a case-specific conclusion about the effect of a particular delay without research tying the alleged delay interval to outcome in the precise condition. The Eleventh Circuit affirmed exclusion and, with it, summary judgment. The 2023 text of Rule 702 complements McDowell by reminding courts that reliable application to the facts is not optional; it is an admissibility prerequisite, not a weight question.
The Florida analogue proceeds identically. After adopting the Daubertamendments as procedural rules in 2019, Florida courts apply § 90.702 across the spectrum of expert testimony. Perez, 138 So. 3d at 496–98, is explicit that the Legislature intended Florida practice to conform to Daubert, Joiner, and Kumho, and that “pure opinion” is not a back door to admissibility. Kemp, 280 So. 3d at 91–94, puts bite behind that principle by reversing where an officer’s “braking” opinion rested on a subjective visual impression of crush damage unsupported by testing, error-rate analysis, or literature; the Fourth District held there was “too great an analytical gap” between the observation and the conclusion and found the error harmful.
A procedural roadmap emerges. The moving party should file a focused motion that begins with a short field primer explaining what “good method” looks like for the discipline in question and then tracks each step the expert took against that benchmark. The record should include the critical report excerpts, the deposition passages where the expert commits to specific steps, and any literature the expert claims as validation. At hearing, the proponent must carry a preponderance burden on each element of Rule 702 or § 90.702. The opponent should be prepared to walk the court through the inferential gaps and misapplications, cross-referencing the hearing exhibits. Where genuine complexity risks confusion, both sides benefit from urging the court to make explicit findings of fact and conclusions of law keyed to reliability and fit. Appellate review under Joiner is abuse of discretion, and careful, anchored findings travel well on appeal. 522 U.S. at 141–47.
Experience-based testimony deserves its own treatment because it appears reliable to jurors yet often rests on opaque reasoning. The Eleventh Circuit has never suggested that experience is disqualifying; it insists, however, that the reasoning process must be articulated and shown to be reliable in application. Frazier captures this point, affirming a district court that allowed some testimony but excluded other opinions where the proponent failed to explain how experience produced a trustworthy conclusion and why that conclusion fit the case. 387 F.3d at 1259–61. Florida’s Kemp supplies the in-state example: an officer’s “I see a dip, therefore braking” inference was not a method and could not be verified, so it failed § 90.702. 280 So. 3d at 90–94.
Other recurring disciplines benefit from concrete markers of reliability. Accident reconstruction is reliable when it uses measurable scene data, replicable calculations, and validated principles; it is unreliable when it converts a visual impression into a conclusion without a bridge of testing or accepted protocols. Kemp makes the latter point unmistakable and demonstrates that even an officer’s training will not salvage an ipse dixit application. Economic damages, life-care planning, and vocational assessments pass Daubert when the expert shows how accepted tools—actuarial work-life tables, utilization rates, published cost benchmarks, standardized functional capacity evaluations—were selected and applied to the plaintiff’s verified medical and employment facts, and when sensitivity analyses demonstrate the opinion’s stability under reasonable variations. The same ethos governs forensic engineering and failure analysis, where adherence to published protocols and reproducible measurements separate principled opinions from speculative leaps. Although these examples are discipline-specific, the doctrinal engine is the same: reliability of method proved by a preponderance, and reliable application to the facts at hand.
Courts have tools for technically dense disputes that help both accuracy and appellate durability. Justice Breyer’s concurrence in Joiner describes the utility of pretrial scientific conferences, targeted Rule 104 hearings, and court-appointed experts under Rule 706, particularly when generalist courts face complex subjects. 522 U.S. at 147–49 (Breyer, J., concurring). These mechanisms are not a detour from gatekeeping; they are ways to make it rigorous and reviewable. In Florida practice, a structured case-management order can serve the same function by sequencing expert disclosures, Daubert motions, and hearings in a way that permits the trial court to make explicit reliability findings well before trial.
The final theme is preservation. Parties seeking exclusion or admission should invite specific findings resolving the pivotal reliability issues, propose narrow rulings that match the record, and, where appropriate, offer limited alternatives such as admitting only those sub-opinions that satisfy the method-and-application test. Appellate courts review these calls for abuse of discretion. Joinerreaffirms that standard and approves exclusion when “analytical gaps” are too large to cross. 522 U.S. at 141–47.
Jurisdictional Portability. Although this article focuses on Florida and the Eleventh Circuit, the analysis tracks the federal baseline and travels nationally. Practitioners should adjust for local rules that sequence expert disclosures, page limits, and hearing practices, but the post-2023 Rule 702 framework—preponderance burden on every element, reliable application under 702(d), and Rule 104(a) adjudication by the court—applies in every federal court and is widely mirrored in state systems that have adopted Daubert.
Conclusion. The controlling standard in both Florida and federal courts asks two questions. Has the proponent shown, by a preponderance, that the opinion is the product of reliable principles and methods applied reliably to sufficient facts and data? And does the opinion “fit” the issues the jury must decide? When the answer to both questions is yes, shaky but admissible testimony may proceed to cross-examination. When the answer to either is no, Daubert keeps speculation from reaching the jury.
- Selected Authorities (first mentions in text provide full citations):
Daubert v. Merrell Dow Pharms., Inc., 509 U.S. 579 (1993); Gen. Elec. Co. v. Joiner, 522 U.S. 136 (1997); Kumho Tire Co. v. Carmichael, 526 U.S. 137 (1999); Quiet Tech. DC–8, Inc. v. Hurel–Dubois UK Ltd., 326 F.3d 1333 (11th Cir. 2003);United States v. Frazier, 387 F.3d 1244 (11th Cir. 2004) (en banc); McDowell v. Brown, 392 F.3d 1283 (11th Cir. 2004); Rink v. Cheminova, Inc., 400 F.3d 1286 (11th Cir. 2005); Allison v. McGhan Med. Corp., 184 F.3d 1300 (11th Cir. 1999); Kilpatrick v. Breg, Inc., 613 F.3d 1329 (11th Cir. 2010); Perez v. BellSouth Telecomms., Inc., 138 So. 3d 492 (Fla. 3d DCA 2014); Kemp v. State, 280 So. 3d 81 (Fla. 4th DCA 2019); In re Amendments to the Fla. Evidence Code, 278 So. 3d 551 (Fla. 2019).
