Opposition Brief — Union Carbide Corporation, et al., Petitioners v. Lee Ann Sommerville

Supreme Court briefApr 6, 2026

Ask Donna

What actually matters in this document.

Text

No. 25-919

IN THE

Supreme Court of the United States

————

UNION CARBIDE CORPORATION; COVESTRO LLC,

Petitioners,

LEE ANN SOMMERVILLE, individually,

and on behalf of all others similarly situated,

Respondent.

————

On Petition for a Writ of Certiorari to the

United States Court of Appeals

for the Fourth Circuit

————

BRIEF IN OPPOSITION TO

PETITION FOR CERTIORARI

————

ADAM J. GOMEZ

KELLY L. TUCKER

GRANT & EISENHOFER, P.A.

123 Justison Street

Wilmington, DE 19801

(302) 622-7000

JOSH AUTRY

Counsel of Record

MORGAN & MORGAN, PA

199 Water St

Suite 1500

New York, NY 10022

(859) 899-8785

jautry@forthepeople.com

STEPHEN EDWARDS

MORGAN & MORGAN, PA

2005 Market St

Suite 600

Philadelphia, PA 19103

(215) 446-9793

Counsel for Respondent

April 6, 2026

WILSON-EPES PRINTING CO., INC. – (202) 789-0096 – WASHINGTON, D.C. 20002

QUESTION PRESENTED

Whether this Court should intervene in a fact-bound

dispute over the admissibility of expert testimony,

where the district court’s errors would have been

reversed under the law of any circuit.

(i)

TABLE OF CONTENTS

Page

QUESTION PRESENTED ..................................

TABLE OF AUTHORITIES ................................

INTRODUCTION ................................................

STATEMENT OF THE CASE ............................

Source Parameters....................................

Historical Emissions Data ........................

Meteorological Data ..................................

Background EtO Levels ............................

District Court’s Conclusion and Final

Order .........................................................

REASONS FOR DENYING THE PETITION ....

II.

The District Court Would Have Been

Reversed in Any Circuit. ..........................

A. Unlawful Heightened Scrutiny ...........

B. “Validation” Requirement ...................

C. Improper Credibility Determinations.

D. Unwarranted “Scientific” Findings.....

E. Disregarding

Dr.

Sahu’s

Supplemental Report ..........................

The Fact-Bound Nature of this Case

Renders it a Poor Vehicle. ........................

(iii)

TABLE OF CONTENTS—Continued

Page

III. The Purported Split Is Illusory Because

Every Circuit Allows for Exclusion of

Expert Opinions That Lack Record

Support or Fail to Reliably Connect Data

and Conclusions. .......................................

A. The Fourth Circuit ...............................

B. The First Circuit ..................................

C. Other Circuits ......................................

CONCLUSION ....................................................

TABLE OF AUTHORITIES

CASES

Page(s)

Acad. Bank, N.A. v. AmGuard Ins. Co.,

116 F.4th 768 (8th Cir. 2024), reh’g

denied, No. 23-1375, 2024 WL 4499662

(8th Cir. Oct. 16, 2024) .............................

Adams v. Ameritech Servs., Inc.,

231 F.3d 414 (7th Cir. 2000) .....................

Belville v. Ford Motor Co.,

919 F.3d 224 (4th Cir. 2019) .....................

Bower v. Westinghouse Elec. Corp.,

522 S.E.2d 424 (W. Va. 1999) ...................

Bresler v. Wilmington Trust Co.,

855 F.3d 178 (4th Cir. 2017) .....................

Bricklayers & Trowel Trades International

Pension Fund v. Credit Suisse Securities

(USA),

752 F.3d 82 (1st Cir. 2014) .......................

Carmichael v. Verso Paper, LLC,

679 F. Supp. 2d 109 (D. Me. 2010) ...........

Casas Office Machines, Inc. v. Mita

Copystar Am., Inc.,

42 F.3d 668 (1st Cir. 1994) .......................

City of Pomona v. SQM N. Am. Corp.,

750 F.3d 1036 (9th Cir. 2014) ................... 13, 20

Crowe v. Marchand,

506 F.3d 13 (1st Cir. 2007) .......................

Daubert v. Merrell Dow Pharmaceuticals, Inc.,

509 U.S. 579

(1993) ......... 1, 2, 12-18, 20, 23-25, 27, 28, 30, 33

TABLE OF AUTHORITIES—Continued

Page(s)

Doucette v. Jacobs,

106 F.4th 156 (1st Cir. 2024) ....................

E. Auto Distribs., Inc. v. Peugeot Motors

of Am., Inc.,

795 F.2d 329 (4th Cir. 1986) .....................

EEOC v. Freeman,

778 F.3d 463 (4th Cir. 2015) ..................... 26, 27

Exafer Ltd v. Microsoft Corp.,

-- F.4th --, No. 2024-2296, 2026 WL

627886 (Fed. Cir. Mar. 6, 2026) ...............

Gen. Elec. Co. v. Joiner,

522 U.S. 136 (1997) ............................. 18, 25, 32

Gopalratnam v. Hewlett-Packard Co.,

877 F.3d 771 (7th Cir. 2017) .....................

Gulf S. Insulation v. U.S. Consumer Prod.

Safety Comm’n,

701 F.2d 1137 (5th Cir. 1983) ...................

Henderson v. Lockheed Martin Corp.,

No. 6:21-cv-1363, 2024 WL 6896218

(M.D. Fla. Oct. 23, 2024) .......................... 5, 19

In re Joint E. & S. Dist. Asbestos Litig.,

52 F.3d 1124 (2d Cir. 1995) ...................... 20, 23

In re Lipitor (Atorvastatin Calcium) Mktg.,

Sales Pracs. & Prods. Liab. Litig.

(No. II) MDL 2502,

892 F.3d 624 (4th Cir. 2018) ..................... 25, 27

Irvine v. Murad Skin Research Labs., Inc.,

194 F.3d 313 (1st Cir. 1999) .....................

vii

TABLE OF AUTHORITIES—Continued

Page(s)

Kennedy v. Collagen Corp.,

161 F.3d 1226 (9th Cir. 1998) ...................

Le Doux v. Western Express, Inc.,

126 F.4th 978 (4th Cir. 2025) ...................

Milward v. Acuity Specialty Prods. Grp.,

Inc., 639 F.3d 11 (1st Cir. 2011) ............... 30-32

Primiano v. Cook,

598 F.3d 558 (9th Cir. 2010) .....................

Rappuhn v. Primal Vantage Co.,

No. 23-10050, 2024 WL 2930448

(11th Cir. June 11, 2024) .......................... 12, 20

Rodriguez v. Hospital San Cristobal,

91 F.4th 59 (1st Cir. 2024) ........................

Sardis v. Overhead Door Corporation,

10 F.4th 268 (4th Cir. 2021) ..................... 28, 29

Sierra Club v. Energy Future Holding Corp.,

No. 6:12-cv-00108, 2014 WL 12690021

(W.D. Tex. Jan. 29, 2014) .........................

Summers v. Missouri Pac. R.R. Sys.,

132 F.3d 599 (10th Cir. 1997) ...................

Teradata Corp. v. SAP SE,

124 F.4th 555 (9th Cir. 2024), cert.

denied, 146 S. Ct. 118 (2025) ....................

Tyger Construction Co., Inc. v. Pensacola

Construction Co.,

29 F.3d 137 (4th Cir. 1994) .......................

United States v. Ala. Power Co.,

730 F.3d 1278 (11th Cir. 2013) ................ 4-5, 19

viii

TABLE OF AUTHORITIES—Continued

Page(s)

United States v. Barton,

909 F.3d 1323 (11th Cir. 2018) .................

United States v. La. Generating, LLC,

929 F. Supp. 2d 591 (M.D. La. 2012) .......

United States v. Hudak,

156 F.4th 405 (4th Cir. 2025) ...................

United States v. Markovich,

95 F.4th 1367 (11th Cir. 2024) .................

CONSTITUTION

U.S. Const. art. III ........................................

STATUTES

42 U.S.C. § 7661c(a) .....................................

Emergency Planning and Community

Right-to-Know Act, 42 U.S.C. § 11001,

et seq. .........................................................

RULES

Fed. R. Evid. 104(a) ...................................... 14, 17

Fed. R. Evid. 702 .......... 1, 2, 4, 5, 7, 11, 12, 14-20,

23-25, 28, 29, 32, 33

Fed. R. Evid. 702(a) ......................................

Fed. R. Evid. 702(b) ......... 1, 11, 15, 20, 24, 25, 29,

32, 33

Fed. R. Evid. 702(c) ...................................... 11, 32

Fed. R. Evid. 702(d) ...................................... 11, 32

TABLE OF AUTHORITIES—Continued

Page(s)

Fed. R. Evid. 702 advisory committee’s

notes to 2000 amendments .......................

Fed. R. Evid. 702 advisory committee’s

notes to 2023 amendments ....................... 16, 17

OTHER AUTHORITIES

29 CHARLES ALAN WRIGHT & ARTHUR R.

MILLER,

FEDERAL

PRACTICE

AND

PROCEDURE (2d ed. 2025 update) .............

Emissions Inventory, W. VA. DEP’T OF ENV’T

PROT., https://dep.wv.gov/daq/planning/

inventory/Pages/default.aspx (last visited

March 12, 2026) ........................................

Matthew W. Swinehart, Note, Remedying

Daubert’s Inadequacy in Evaluating the

Admissibility of Scientific Models Used in

Environmental Tort Litigation, 86 TEX.

L. REV. 1281 (2008) ...................................

INTRODUCTION

The First and Fourth Circuits are not islands of

lawless resistance to the 2023 amendments to Federal

Rule of Evidence 702. Applying settled law, including

cases decided after the 2023 amendments, these

courts regularly examine the factual bases of expert

testimony as part of their reliability analysis under

Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S.

579 (1993). They allow testimony from experts whose

opinions are based on sufficient facts or data and

exclude those who don’t. And where, as happened

here, lower courts overstep the bounds of their

“gatekeeping” role and invade the province of the jury,

the First and Fourth circuits reverse.

Thus, there is no fissure between the circuits in their

application of Rule 702. Petitioners, however, posit

that in the First and Fourth circuits, the judicial

gatekeepers are asleep at their posts, leaving junk

science to run amok. What is worse, they suggest that

the appellate judges presiding over this dereliction of

duty have snubbed the enlightened counsel of the

Rules Committee, stubbornly clinging to a rule that

defies reason and precedent. According to Petitioners,

this Court’s intervention is the only hope for bringing

these rogue jurisdictions into alignment with the law.

This is simply untrue. The First and Fourth circuits

each recognize that expert opinion can be excluded

because it “lacks sufficient support in the record or . . .

there is simply too great an analytical gap between

[the expert’s] data and the opinion proffered.” App’x

27a n.7 (opinion below). These principles fully implement Rule 702(b)’s “sufficient facts or data” requirement.

They also coexist peacefully with the related rule that

questions regarding the “factual underpinnings” of an

expert’s testimony typically go to weight, rather than

admissibility. Neither circuit enforces this latter rule

in the categorical fashion claimed by Petitioners, i.e.,

as an absolute bar to any consideration of the factual

basis of expert opinion. Instead, the rule merely

codifies the commonsense notion that, after an expert

has provided cogent reasons for his choice of data or

other factual inputs, the trier of fact should usually

resolve disputes about the validity of those inputs,

rather than the district judge in Daubert proceedings.

Here, the Fourth Circuit honored this well-established principle by reversing an overreaching judge

who grossly exceeded his gatekeeping role. Among

other errors, the trial court imposed a judge-crafted,

heightened standard of admissibility that is completely

untethered from the text of Rule 702. Inspired by a

series of law review articles critiquing the use of air

dispersion modeling in environmental litigation, the

court declared that experts using such a model must

employ “[a]n in-depth data investigation, a searching

historical analysis, an excruciating attention to detail,

and a methodology designed to wring error out of the

process . . . .” Id. 62a–63a. This exacting benchmark

far exceeds Rule 702’s requirement that expert

testimony be based on “sufficient facts or data.” On top

of this glaring error, the trial court also resolved

conflicts in the evidence, made credibility determinations between the parties’ experts, and hazarded

amateur findings on “the science of wind” that even the

dissenting Fourth Circuit judge acknowledged were

improper.

Plagued by these errors, the district court’s opinion

would have been reversed under the law of any circuit.

Thus, even if there were a circuit split concerning how

to apply Rule 702’s “sufficient facts or data” requirement (which, to be clear, there is not), this case would

be a poor vehicle to address it because this Court’s

intervention would not change the outcome below.

Finally, the granularity of the district court’s

analysis makes this a thoroughly fact-bound case.

Contrary to Petitioners’ narrative, this is not a case

where the appeals court rubber-stamped a patently

defective expert opinion by mechanically applying an

inappropriately deferential rule of law. Rather, because

the district court’s analysis was subtle and fastidious,

the Fourth Circuit reversed on equally painstaking

grounds. The appeals court traced five nuanced evidentiary errors committed by the district court, which

ranged from the lower court’s improper resolution of a

debate about the accuracy of toxic airborne exposure

data from state and federal regulators, to the propriety

of synthesizing weather data from multiple locations

when modeling emissions of air pollution, to an

invented requirement that an expert independently

“validate” emissions data recorded by the opposing

party. If this Court grants the Petition, it will find itself

mired in these questions instead of deciding a clean

legal issue. For this additional reason, this case is a

poor vehicle for this Court’s review.

STATEMENT OF THE CASE

Since 1978, Petitioners have operated a facility in

South Charleston, West Virginia (the “Facility”) that

manufactures ethylene oxide (“EtO”). EtO is an

odorless, colorless gas with industrial applications

that range from sterilizing medical devices to creating

synthetic fibers like polyester. It is also a potent

carcinogen, a fact that government bodies in the

United States have acknowledged since at least 1977.

App’x 91a.

Despite these risks to human health, Petitioners

released significant quantities of EtO into the

community surrounding the Facility over a period of

decades.1 This stream of toxic, airborne emissions put

the inhabitants of this community at an increased risk

of contracting numerous cancers, including leukemia

and breast cancer.

Respondent Lee Ann Sommerville is one such

affected community member. As a resident of the Class

Area she will need to undergo medical testing, for the

rest of her life, designed to detect cancers she is at an

increased risk of developing because of Petitioners’

EtO emissions.

To redress this harm, Respondent filed suit in the

District Court for the Southern District of West

Virginia on behalf of herself and a class of similarly

situated persons. She asserted a medical monitoring

claim under West Virginia law, which allows plaintiffs

to recover the cost of “medical testing” when their

“exposure to toxic chemicals creates an enhanced risk

of disease . . . .” Bower v. Westinghouse Elec. Corp., 522

S.E.2d 424, 431 (W. Va. 1999) (citation omitted).

Among other proof, Respondent supported her claim

with the expert opinion of Dr. Ranajit Sahu. Dr. Sahu

is a Ph.D.-level mechanical engineer with over thirty

years of experience in environmental and pollutionrelated projects, including emissions modeling. App’x

45a. Courts across the country—including the Fifth

and Eleventh Circuits—have approved his emissions

modeling work as reliable under Federal Rule of

Evidence 702. See United States v. Ala. Power Co., 730

Respondent’s claims concern the period from 1978 to 2019

(the “Class Period”). App’x 2a. The area surrounding the Facility

is known as the Kanawha Valley (the “Class Area”). Id. 80a.

F.3d 1278, 1288 (11th Cir. 2013); United States v. La.

Generating, LLC, 929 F. Supp. 2d 591, 596 (M.D. La.

2012); Sierra Club v. Energy Future Holding Corp., No.

6:12-cv-00108, 2014 WL 12690021, at *8 (W.D. Tex.

Jan. 29, 2014); Henderson v. Lockheed Martin Corp.,

No. 6:21-cv-1363, 2024 WL 6896218, at *2–3 (M.D. Fla.

Oct. 23, 2024).

II.

After analyzing Petitioners’ operations at the

Facility, Dr. Sahu prepared an air dispersion model

designed to reconstruct the flow of EtO emissions and

measure the levels of exposure inflicted upon the

surrounding community. Dr. Sahu built his model by

selecting relevant data points and then using a

computer algorithm to project the effect of those forces

over time. App’x 19a n.4 (explaining AERMOD

modeling program). These inputs included “source

parameters,” or descriptions of pollutant-emitting

activities at the Facility, emissions data submitted

to regulators, and meteorological data. Id. 19a. As

the Fourth Circuit later acknowledged, Dr. Sahu

presented detailed reasoning regarding how he chose

these inputs. Id. 20a. After applying this method, Dr.

Sahu concluded Petitioners exposed the surrounding

community to levels of EtO far in excess of normal,

background levels.

Before trial, the district court granted Petitioners’

motion to exclude Dr. Sahu’s testimony. Id. 42a. The

court based its ruling on its disapproval of the data

utilized by Dr. Sahu, and his explanations therefor, in

four key areas, discussed in turn below. Overarching

these more particular findings, however, was the

district court’s conclusion that the uncertainties

inherent in air dispersion modeling necessitated a

heightened standard of reliability under Rule 702. In

the court’s view, such a model is only admissible if it

applies “an in-depth data investigation, a searching

historical analysis, an excruciating attention to detail,

and a methodology designed to wring error out of the

process.” Id. 62a–63a, 87a (cleaned up). The court

ultimately concluded that Dr. Sahu failed to meet this

heightened standard, and erroneously excluded his

opinions on that basis.

A. Source Parameters

Regarding source parameters, the district court

faulted Dr. Sahu for aggregating emissions data from

two different facilities operated by Petitioners: the

Facility and another plant located in nearby Institute,

West Virginia. Dr. Sahu explained that the model allows

the user to isolate data for any particular emission

source at either facility, thereby alleviating this

“disaggregat[ion]” concern. Id. 62a. Notwithstanding

this proof, the district court, without citation to

contrary fact, inexplicably held that the model “cannot

determine the reliability of the estimated exposure

from the Defendants’ operations based on the

information provided.” Id.

The district court also criticized Dr. Sahu’s decisions

to use “modeling conducted in 2019 by the West

Virginia Department of Environmental Protection”

(“WVDEP”) to model source parameters for the entire

Class Period and to model all Facility emissions as

“fugitive” during the years 1985–89. Id. 63a–65a.2 Dr.

Sahu, however, explained that these assumptions were

Fugitive emissions are “releases to air that don’t occur

through a confined air stream[,] such as via equipment leaks.”

App’x 19a n.3 (cleaned up) (citation omitted). By contrast,

“point source” emissions “occur through identifiable confined air

streams, such as stacks, ducts or pipes.” Id.

necessary given the lack of specificity in Petitioners’

emissions records. Id. 65a (“Such sources may have

been capable of being modeled with greater precision

had [Petitioners] taken efforts to describe these

sources more precisely in the ordinary course of

business—but they did not.”).

B. Historical Emissions Data

Next, the district court took issue with certain

assumptions Dr. Sahu used to fill gaps in historical

emissions data. For the year 1984, the court faulted Dr.

Sahu for relying on data submitted by Petitioners to

the WVDEP as part of its “1984 Emission Inventory.”

Id. 68a. Despite being prepared by the operators of the

Facility for use by regulators, the court concluded this

data was “patently unreliable” because after submitting it, Petitioners sent a letter to the WVDEP

expressing reservations about its validity. Id. 68a–69a.

Dr. Sahu explained that he thought these reservations

were unfounded, and that Petitioners had given him

no real reason to doubt the 1984 data. Id. 68a. He also

made the reasonable assumption that “the company

would not provide misleading data to its regulator.” Id.

69a. But to the district court, this was not enough: Dr.

Sahu was obligated to “validate” the 1984 data himself.

Id. 69a–70a. The court did not explain, however, what

this “validation” would look like, nor did it cite Rule

702 or any other source of law for this requirement.

For the years 1985–89, the court criticized Dr. Sahu

for relying on another type of submission to the

WVDEP: Petitioners’ “Toxic Air Pollutant Facility

Registration Summary Sheets.” Id. 71a. The court

cited Petitioners’ expert’s view that these sheets reflect

maximum possible emissions, rather than actual

emissions. Id. But Dr. Sahu interpreted these records

differently: because the emissions figures on the

summary sheets varied substantially over time, they

would have had to represent actual emissions, as a

true emissions ceiling for the Facility would not

fluctuate without emissions point or infrastructure

changes. Id. 72a. Dr. Sahu explained there was no

evidence of substantial changes in operations at the

Facility from 1985 to 1989. Id. (“There were no process

changes [Dr. Sahu] could identify in the record

between 1989 and the gap years . . . .”). This, again,

could not satisfy the district court: it found Petitioners’

expert more persuasive. Id.

For the period from 1990–2019, the court found it

unacceptable that Dr. Sahu utilized Petitioners’ own

emissions data submitted under the EPA’s Toxics

Release Inventory (“TRI”) and WVDEP’s annual

emissions inventory.3 Parroting Petitioners’ expert, the

court held that because this data was not compiled for

the express purpose of “exposure reconstruction,”

Dr. Sahu was obligated to “validate” it himself before

using it in his model. Id. 74a–75a. Again, the court did

not explain what this additional step of “validation”

would consist of.

Lastly, the court faulted Dr. Sahu for modeling

Petitioner Covestro’s emissions at a constant rate

throughout the Class Period despite the availability of

site-specific data showing otherwise. Id. 67a. But, as

The TRI is a database maintained by the EPA pursuant to

the Emergency Planning and Community Right-to-Know Act, 42

U.S.C. § 11001, et seq.

WVDEP maintains an annual emissions inventory pursuant to

its responsibilities under Title V of the Clean Air Act. 42 U.S.C.

§ 7661c(a) (requiring emissions reporting by permit holders); see

also Emissions Inventory, W. VA. DEP’T OF ENV’T PROT., https://

dep.wv.gov/daq/planning/inventory/Pages/default.aspx (last visited

March 12, 2026).

the Fourth Circuit later noted, Dr. Sahu corrected this

issue in a supplemental report. Id. 27a. Without

explanation, the district court ignored this submission.

Id. 67a.

C. Meteorological Data

The district court also quarreled with Dr. Sahu’s

choice of meteorological data inputs. Dr. Sahu relied on

data from multiple locations to model weather

conditions around the Facility over the Class Period.

His primary reason for doing so was the lack of on-site

data: Petitioners could only provide “on-site wind

speed and wind direction data collected between” 1985

and 1986. Id. 77a. Not only was this data limited in

time: it did not address “other required meteorological

parameters” that impact EtO air dispersion, including

“atmospheric pressure, relative humidity, cloud cover,

and precipitation.” Id.

Dr. Sahu also needed data from higher elevations to

reconstruct emissions from “elevated point source[s]”

at the Facility. Id. 25a. And he inputted weather data

from outside the Kanawha Valley because he sought to

model the flow of EtO “in the wider area.” Id. 81a.

Ignoring these explanations, the district court held

that Dr. Sahu’s meteorological inputs were not

“representative” enough. Id. 79–80a. Opining from the

bench on “the science of wind,” the court deemed it

improper to utilize weather data from multiple

locations because “a single location . . . cannot have

multiple wind speeds coming from various directions

at one specific time.” Id. 24a, 80a.4

In the Fourth Circuit, even the dissenting judge recognized

the district court’s overreach on this point. App’x 39a n.9

The court also claimed Dr. Sahu “ignor[ed] the

topographical characteristics of the region” by

utilizing data from other locations. Id. 80a. In so doing,

the court ignored Dr. Sahu’s explanations of how data

from multiple elevations would actually best represent

“South Charleston’s mountainous geography,” particularly in light of the limitations of localized data. Id. 25a.

Finally, the court faulted Dr. Sahu for using the

1985–86 wind speed and direction data as a constant

for the entire Class Period (1984–2019). Id. 80a. But

again, Dr. Sahu explained his use of this data: there

was no better data available outside 1985–86, and Dr.

Sahu combined this data with additional inputs

regarding other meteorological factors across a longer

timeframe to reach his conclusions. Id. 80a–81a.

D. Background EtO Levels

Lastly, the district court held that Dr. Sahu chose the

wrong figures to represent background levels of EtO in

the Class Area. “‘Background’ refers to exposure levels

experienced by the general population as opposed to

the levels [Respondent] alleges she was exposed to due

to [Petitioners’] negligence.” Id.

Dr. Sahu used the national annual average for EtO

exposure contained in the EPA’s National Air Toxics

Assessment (“NATA”) as his background figure. Id.

82a. The district court found this figure too detached

from localized conditions around the Facility: in its

view, the preferred data source was a single local

monitoring report issued by the WVDEP based on

“four . . . 24-hour sampling events.” Id. 26a.

(“It strikes me as unwise for courts to opine on ‘the science of

wind’—whatever that entails—unless absolutely necessary.”).

Dr. Sahu explained he did not use the WVDEP’s

monitoring data because “[f]our days of data cannot be

used to calculate risk over a 70-year period.” Id. 26a.

Thus, despite there being two imperfect data sources,

without explanation as to why it was superior or met

its heightened reliability standard, the district court

decided to credit Petitioners’ choice of data inputs over

Respondent’s. Id. 88a. In so doing, the court held that

Dr. Sahu failed to meet its specific, heightened

standard for air modeling experts. Id. 87a.

E. District Court’s Conclusion and Final

Order

Ultimately, the district court excluded Dr. Sahu’s

testimony in its entirety because it was allegedly “not

based upon sufficient facts or data.” Id. 88a. In the

alternative, the court also held that Dr. Sahu’s

testimony was not the product of “reliable principles

and methods,” and that he had not “reliably applied”

his methods to the facts of the case. Id. The court thus

invoked every Rule 702 subpart aside from relevancy.

See Fed. R. Evid. 702(b)–(d).

In a later ruling, the court granted summary

judgment to Petitioners on Article III standing

grounds that are not the subject of the current

Petition. App’x 89a.

III.

A divided panel of the Fourth Circuit reversed the

district court’s exclusion of Dr. Sahu’s testimony.5 The

appeals court held that the district court abused its

discretion in arriving at each of its “supposedly

The panel also reversed the district court’s ruling on

Respondent’s standing to sue. App’x 6a–17a. Petitioners do not

seek review of that decision.

independently-sufficient reasons” for excluding Dr.

Sahu. Id. 20a.

Regarding source parameters, the court held that

the district court wrongly favored Petitioners’ expert

on disputed data-choice issues. Id. (“The district court

. . . simply ignored or discounted Dr. Sahu’s proffered

explanations in favor of those that [Petitioners’]

expert, Dr. Ranjit Machado, offered.”). This, the court

observed, “was not a true critique of Dr. Sahu’s

‘methodology,’ but a veiled credibility determination

based on Dr. Sahu’s choice of which data to input into

his model.” Id. Given the presence of conflicting expert

testimony and multiple sets of available data, the

questions raised by the district court “affect[ed] the

weight and credibility of [Dr. Sahu’s] assessment,

not its admissibility.” Id. 21a. To bolster this point,

the appeals court also cited an Eleventh Circuit

case decided after the 2023 amendments to Rule

702 explaining that “[c]rediting one expert over

another . . . misapplies Daubert and intrudes on the

province of the jury.” Id. (citing Rappuhn v. Primal

Vantage Co., No. 23-10050, 2024 WL 2930448, at *4

(11th Cir. June 11, 2024)).

The appeals court then rebuked the district court for

imposing the “extratextual” requirement that Dr. Sahu

“validate” data before using it in his model. Id. As the

court put it, “[n]othing in Rule 702 requires an expert

witness to ‘validate’ data.” Id. What is more, the

district court never “explain[ed], in practical terms,

how Dr. Sahu was supposed to undertake ‘validation.’”

Id. (emphasis in original).

Regarding historical emissions data, the appeals

court again found the district court abused its discretion

by crediting Petitioners’ expert over Respondent’s in

fact disputes regarding the reasons an expert might

choose one data set over another. It concluded the

district court erred by crediting Petitioners’ reservations about the data they reported to WVDEP, and by

disregarding “Dr. Sahu’s testimony justifying his . . .

interpretation of the” Toxic Air Pollutant Facility

Registration Summary Sheets. Id. 22a–23a. On this

latter point, the appeals court relied on case law across

various circuits, including the Eighth and Second, and

specifically quoted the Ninth Circuit in finding that

“[s]uch a ‘factual dispute is best settled by a battle of

the experts before the fact finder, not by judicial fiat.’”

Id. 23a (quoting City of Pomona v. SQM N. Am. Corp.,

750 F.3d 1036, 1048–49 (9th Cir. 2014)).

Moving on to meteorological data, the appeals court

again criticized the district court for “ignor[ing] Dr.

Sahu’s testimony” explaining his choice of data,

specifically as to “why using wind data from various

altitudes was proper . . . and why [he] chose to use

onsite data from 1985–86 to model 1984–2019.” Id.

25a. Per Daubert, the appeals court concluded that to

the extent any of these explanations were unsatisfying

to Petitioners, their recourse lay in “cross-examination, presentation of contrary evidence, and careful

instruction on the burden of proof ” at trial. Id. (quoting

Daubert, 509 U.S. 579, 596 (1993)). The appeals court

also rebuked the district court fashioning itself an

amateur scientist and reaching unsupported findings

regarding “the science of wind,” and crediting its own

amateur conclusions over Dr. Sahu’s testimony

regarding wind data. Id. 24a–25a.

Finally, regarding background EtO levels, the

appeals court held that the debate between the parties’

experts regarding whether to use the NATA baseline

figures or the more local but time-limited data from

WVDEP presented a “credibility determination[]” the

district court was not entitled to make. Id. 26a.

To conclude, the appeals court held that “in

accordance with Federal Rules of Evidence 104(a) and

702, Sommerville has established by a preponderance

of evidence that Dr. Sahu’s testimony is admissible.”

Id. 27a.

In dissent, Chief Judge Diaz cited a host of Fourth

Circuit cases affirming the exclusion of experts whose

opinions were based on faulty data or lacked a

sufficient basis in record evidence. See id. 34a–37a.

In response to Judge Diaz’s criticisms, the majority

reaffirmed that district courts can always “decide that

an expert’s opinion lacks sufficient support in the

record or that there is simply too great an analytical

gap between [an expert’s] data and the opinion

proffered.” Id. 27a n.7 (cleaned up). But when a district

court excludes an expert “based on (1) its mere disagreement with an expert’s choice of data or (2) its own

assessment of the correctness of an expert’s opinions,”

it exceeds its gatekeeping role under Daubert. Id.

REASONS FOR DENYING THE PETITION

I. The District Court Would Have Been

Reversed in Any Circuit.

The district court committed a series of blatant,

fundamental errors that would have earned it reversal

in any circuit. In turn, the appeals court analyzed

these errors in detail—it did not simply rubber-stamp

Dr. Sahu’s testimony because Petitioners challenged

its factual basis. This creates vehicle concerns for this

Court because there are clear, alternative grounds on

which to affirm. It also highlights that the Fourth

Circuit’s rule regarding the role of an expert’s data in

the Rule 702 reliability analysis is not the categorical

bar that Petitioners claim.

The district court committed five glaring, overarching errors by 1) erecting a unique, heightened reliability

standard for experts engaged in air dispersion

modeling; 2) requiring Dr. Sahu to independently

“validate” data Petitioners submitted to regulators; 3)

making credibility determinations between competing

experts on issues of the appropriate choice of data; 4)

issuing unwarranted, amateur scientific findings from

the bench; and 5) ignoring corrections introduced by

Dr. Sahu’s supplemental report. These errors go the

heart of trial courts’ gatekeeping role under Daubert

and Rule 702—correcting them does not depend on

any specific formulation of Rule 702(b)’s “sufficient

facts or data” requirement.6

A. Unlawful Heightened Scrutiny

Based on its own hostility toward the science of air

dispersion modeling, the district court concocted a

standalone, heightened standard for experts in this

field: in the court’s view, such experts must employ

“[a]n in-depth data investigation, a searching historical

analysis, an excruciating attention to detail, and a

methodology designed to wring error out of the

process . . . .” App’x at 62–63a. This exacting

benchmark far exceeds Rule 702’s requirement that

the proponent of expert testimony show it is more

likely than not that the expert’s opinion is “based on

sufficient facts or data . . . ,” Fed. R. Evid. 702(b), and

In any event, the Fourth Circuit expressly held that

Respondent “established by a preponderance of the evidence that

Dr. Sahu’s testimony is admissible.” App’x 27a.

is not supported by law in any circuit as shown in

Section C(iii), infra.

In keeping with the flexibility of the Daubert inquiry,

there is not a hard and fast definition of what

constitutes “sufficient facts or data” to support an

expert opinion. See Daubert, 509 U.S. at 594–95

(describing reliability analysis as “flexible” and “focused

on principles and methodology, not . . . the conclusions

that they generate”). However, the Advisory Committee’s

Notes indicate the inquiry is “quantitative rather than

qualitative . . . .” Fed. R. Evid. 702 advisory committee’s

notes to 2000 amendments. Thus, “[t]he question is

whether the expert considered enough information to

make the proffered opinion reliable.” 29 CHARLES ALAN

WRIGHT & ARTHUR R. MILLER, FEDERAL PRACTICE AND

PROCEDURE § 6268 (2d ed. 2025 update) (emphasis added).

Rule 702’s text also provides guidance, as “[t]he word

‘sufficient’ signifies that the expert may properly base

her opinion on something less than all the pertinent

facts or data.” Id. (emphasis added). “Thus, sufficiency

is not a matter of whether the judge believes in the

facts or data on which the expert relies.” Id.; see also

Fed. R. Evid. 702 advisory committee’s notes to 2000

amendments (“The emphasis in the amendment on

‘sufficient facts or data’ is not intended to authorize a

trial court to exclude an expert’s testimony on the

ground that the court believes one version of the facts

and not the other.”).

“[O]nce the court has found it more likely than not

that the [sufficient facts or data] requirement has been

met, any attack by the opponent will go only to the

weight of the evidence.” Fed. R. Evid. 702 advisory

committee’s notes to 2023 amendments. Crucially, this

framework does not “require[] the court to nitpick an

expert’s opinion in order to reach a perfect expression

of what the basis and methodology can support.” Id.

And where experts disagree on the proper basis for

scientific opinion, the Advisory Committee Notes

express a preference for adversarial testing, not

outright exclusion:

It will often occur that experts come to

different conclusions based on contested sets

of facts. Where that is so, the Rule 104(a)

standard does not necessarily require exclusion of either side’s experts. Rather, by

deciding the disputed facts, the jury can

decide which side’s experts to credit.

Proponents do not have to demonstrate to the

judge by a preponderance of the evidence that

the assessments of their experts are correct,

they only have to demonstrate by a preponderance of evidence that their opinions are

reliable . . . The evidentiary requirement of

reliability is lower than the merits standard

of correctness.

Id. (cleaned up) (citation omitted).

Here, the district court’s approach deviated wildly

from these principles. Rather than analyze whether

it’s more likely than not that Dr. Sahu considered

“enough information to make [his] opinion reliable,”

the court raised the much higher bar of whether Dr.

Sahu employed “[a]n in-depth data investigation, a

searching historical analysis, an excruciating attention to

detail, and a methodology designed to wring error out

of the process . . . .” App’x 62a–63a. This is not an

application of Rule 702 or Daubert standards; instead,

it is a standalone rule without support in the law.

Viewing the case through this improper lens, the

district court discredited Dr. Sahu’s data extrapolations, even though such extrapolations are inherent in

air dispersion modeling. Cf. Gen. Elec. Co. v. Joiner, 522

U.S. 136, 146 (1997) (“Trained experts commonly

extrapolate from existing data.”); see also Primiano v.

Cook, 598 F.3d 558, 565 (9th Cir. 2010) (“Lack of

certainty is not, for a qualified expert, the same thing

as guesswork.”).

Stated another way, far afield from the text of Rule

702, the district court essentially crafted a “precision”

prerequisite for expert testimony. To justify this

requirement, the court reached all the way back to a

pre-Daubert decision by the Fifth Circuit—that did not

even consider Rule 702—for the proposition that “[t]o

make precise estimates, precise data [is] required.”

App’x 67a (citing Gulf S. Insulation v. U.S. Consumer

Prod. Safety Comm’n, 701 F.2d 1137, 1146 (5th Cir.

1983)). The district court thus erred by straying

outside the framework of Rule 702, Daubert, and its

progeny.

It is telling that the district court’s opinion is replete

with quotations from law review notes and articles

that advocate abandoning the Daubert admissibility

inquiry when courts evaluate environmental modeling

experts. See, e.g., Matthew W. Swinehart, Note,

Remedying Daubert’s Inadequacy in Evaluating the

Admissibility of Scientific Models Used in EnvironmentalTort Litigation, 86 TEX. L. REV. 1281, 1283 (2008)

(“Daubert is not the appropriate way to evaluate model

reliability, or at least it requires significant amendment to provide a credible analytic process.”); see also

App’x 54a, 58a, 59a, 87a (relying on Swinehart article).

Heavily influenced by these articles, the district court

left Daubert behind in pursuit of an academic quest

to rectify the so-called “troubled marriage between

science and law” purportedly reflected in litigants’ use

of dispersion modeling. App’x 58a. Any circuit would

have reversed this blatant legal error. This case is thus

not about whether the Fourth Circuit should follow the

dictates of Rule 702 (it does); this case is about

whether a district court can amend Rule 702 based

upon its personal views of a particular field of science

(it cannot).

B. “Validation” Requirement

The district court took the remarkable position that

Dr. Sahu’s opinion was unreliable because he failed to

independently “validate” emissions data submitted by

Petitioners to regulators. See App’x 70a, 74a–76.

Stated differently, the court held that an expert was

not entitled to rely on pollution data created by the

party seeking to exclude the expert and reported

to regulators who have the authority to impose fines

for inaccurate or incomplete emissions reporting. This

“glaring abuse of discretion,” App’x 21a, would have

been reversed in any circuit. See, e.g., Ala. Power, 730

F.3d at 1287 (“[T]here was a sufficient evidentiary

basis under Rule 702—including Alabama Power’s

own records and predictions—to support [the expert’s]

assertions . . . .”); Adams v. Ameritech Servs., Inc.,

231 F.3d 414, 427 (7th Cir. 2000) (“The underlying

information . . . came from the defendants ultimately,

and as such we see no problem in [expert’s] decision to

rely on it.”); Henderson, 2024 WL 6896218 at *2 (“If the

actual emissions are lower than what Lockheed’s

records say they are, Lockheed’s experts can explain to

the jury why—but Sahu using Lockheed’s own

numbers as his source does not make his methodology

unreliable.”).

Furthermore, as the Fourth Circuit correctly noted,

the district court’s validation requirement was wholly

“extratextual,” as “[n]othing in Rule 702 requires an

expert witness to ‘validate’ data.” App’x 21a. It was

also unworkably vague, prompting even the dissenting

judge to guess at the lower court’s meaning. Id. 37a.

Rule 702 requires “sufficient” data to form a reliable

opinion. Fed. R. Evid. 702(b). Validation is an

additional step stemming not from Daubert’s concept

of reliability, but from the district court’s idiosyncratic

views on the science of air dispersion modeling. In any

circuit, a district court manufacturing an extratextual

admissibility requirement would result in reversal.

C. Improper Credibility Determinations

Every circuit agrees that district courts cannot

make credibility determinations or resolve fact

disputes in the course of the Rule 702 inquiry. See, e.g.,

City of Pomona, 750 F.3d at 1044; United States v.

Barton, 909 F.3d 1323, 1333 (11th Cir. 2018) (“[I]t

would have gone beyond the gatekeeping function of

the trial court to exclude Zuleger’s testimony on the

basis of a credibility determination favoring [the

competing expert].”); In re Joint E. & S. Dist. Asbestos

Litig., 52 F.3d 1124, 1133 (2d Cir. 1995) (“Trial courts

should not arrogate the jury’s role in evaluating the

evidence and the credibility of expert witnesses by

simply choosing sides in the battle of the experts.”)

(cleaned up); App’x 23a. This remains true post-2023

amendments to Rule 702. See Rappuhn, 2024 WL

2930448, at *4 (“[C]rediting one expert over another . . .

misapplies Daubert and intrudes on the province of

the jury.”). The district court flouted these principles

by consistently “ignor[ing] or discount[ing] Dr. Sahu’s

proffered explanations in favor of those that

[Petitioners’] expert, Dr. Ranjit Machado, offered.”

App’x 20a. The court made at least four of these

improper credibility calls:



Meaning of “Summary Sheets”: The parties’

experts offered competing interpretations

of historical “Toxic Air Pollutant Facility

Registration Summary Sheets” submitted

by Petitioners to the WVDEP. Petitioners’

expert believed these sheets reflect maximum rather than actual emissions at the

Facility. Dr. Sahu disagreed, contending

that the change in the supposed “maximum”

figures over time showed they represent

actual emissions. The district court improperly resolved this conflict by “ignor[ing] Dr.

Sahu’s testimony justifying his contrary

interpretation of the summary sheets.” Id.

23a.

Crediting UCC’s Reservations About 1984

Emissions Data: To reconstruct “EtO

emissions . . . for 1984,” Dr. Sahu relied on

data Petitioners submitted to the West

Virginia Air Pollution Control Commission.

Id. 68a. When they submitted this data,

Petitioners included a letter stating certain

reservations about its validity. Dr. Sahu

believed these “conclusionary” reservations

provided no real basis to doubt the accuracy

of the 1984 figures. Id. He also explained he

reasonably assumed “the company would

not provide misleading data to its regulator[.]” Id. 69a. Again, the district court sided

with Petitioners, choosing to credit the

claims in their letter over Dr. Sahu’s “reasons

for why he did not believe [Petitioners’]

claim that [their] 1984 data were

inaccurate.” Id. 22a.



Choice of Meteorological Inputs: To model

weather conditions over the Class Period,

Dr. Sahu utilized a combination of

meteorological data from the Facility and

other locations with similar geography. As

with many of his other inputs, Dr. Sahu

relied on data from other locations because

reliable on-site data was lacking. Id. 25a.

He also explained how data from multiple

elevations would accurately represent

“South Charleston’s mountainous geography.”

Id. The district court ignored these

explanations, ultimately finding Dr. Sahu’s

reasoning “unpersuasive.” Id. 81a.

Use of NATA Figures to Set Background

EtO Levels: Dr. Sahu used the NATA

national average for EtO exposure to

represent the “background” exposure level

in the Class Area. Petitioners’ expert

opined that local monitoring data from the

WVDEP would more accurately represent

conditions around the Facility, but Dr. Sahu

explained he decided against using the

local data because it was based on limited

sampling: just four days of testing, which

was insufficient to create an annual

average. Id. 26a. As the appeals court

recognized, by resolving this evidentiary

dispute in favor of Petitioners’ expert, the

district court “waded into credibility

determinations.” Id.

D. Unwarranted “Scientific” Findings

The district court also excluded Dr. Sahu’s testimony

because his opinion conflicted with the court’s own

amateur understanding of “the science of wind.” Id.

24a. Without citing to any source, the court declared

that “a single location . . . cannot have multiple wind

speeds coming from various directions at one specific

time.” Id. 80a. The entire appellate panel agreed that

this conclusion was both scientifically suspect and

legally unnecessary. Id. 24a, 39a n.9.

This kind of overreach would earn reversal under

the law of any circuit. See, e.g., In re Joint E. & S. Dist.

Asbestos Litig., 52 F.3d at 1137 (reversing directed

verdict based on exclusion of expert testimony because

“the district court impermissibly made a number of

independent scientific conclusions . . . in a manner not

authorized by Daubert”); Summers v. Missouri Pac.

R.R. Sys., 132 F.3d 599, 604 (10th Cir. 1997) (holding

that “district courts must be careful not to don the

amateur scientist’s cap in ruling on scientific validity”)

(citation omitted); see also Daubert, 509 U.S. at 601

(Rehnquist, J., concurring in part and dissenting in

part) (observing that Rule 702 does not confer on

district courts “either the obligation or the authority to

become amateur scientists in order to perform” their

gatekeeping role).

E. Disregarding Dr. Sahu’s Supplemental

Report

Lastly, the district court inexplicably excluded Dr.

Sahu’s opinions based on conclusions that he amended

in a supplemental report. App’x 27a. Ignoring record

evidence that supports the reliability of an expert’s

opinion is an obvious error that would have been

reversed in any circuit.

II. The Fact-Bound Nature of this Case

Renders it a Poor Vehicle.

If this Court grants the Petition, it will find itself

mired in the minute factual questions undergirding

the district court’s opinion. The Fourth Circuit’s decision

bears this out: there, the appeals court painstakingly

analyzed each of the district court’s factual errors and

evidentiary missteps. If there were a need for new

jurisprudence regarding Rule 702(b)’s “sufficient facts

or data” requirement (which, to be clear, there is not),

this case would not be the vehicle for it.

Petitioners attempt to paper over the fact-bound

nature of this case by oversimplifying the disputes

that were central to the courts below. See Petition at

10–11 (reducing panel majority’s opinion to a single

legal principle). But as the foregoing discussion shows,

the district court’s over-exacting approach produced

evidence-intensive rulings that do not depend on any

specific formulation of Rule 702’s “sufficient facts or

data” requirement. If this Court grants review, its time

will be expended on these matters rather than

developing the law of Daubert on a clean record.

III. The Purported Split Is Illusory Because

Every Circuit Allows for Exclusion of

Expert Opinions That Lack Record

Support or Fail to Reliably Connect Data

and Conclusions.

Petitioners’ summary of the First and Fourth

Circuit’s approach to Rule 702(b)’s “sufficient facts or

data” requirement is a baseless caricature. Contrary to

their argument, those circuits do not follow a

categorical rule forbidding any and all scrutiny of the

factual bases of expert opinion. Rather, like every

other federal court of appeals, the First and Fourth

circuits regularly sanction the exclusion of expert

testimony that either lacks sufficient support in the

record or rests on an untenable “analytical gap

between the data and the opinion proffered.” Joiner,

522 U.S. at 146. These principles, which the Fourth

Circuit affirmed in this case, see App’x 27a n.7, fully

accommodate Rule 702(b)’s requirement that an

expert’s opinion rest on “sufficient facts or data.”

The phrase on which Petitioners seize—that “questions

regarding the factual underpinnings of the [expert

witness’] opinion affect the weight and credibility of

the witness’ assessment, not its admissibility”—does

not operate as a “get-out-of-Daubert-free card.” In re

Lipitor (Atorvastatin Calcium) Mktg., Sales Pracs. &

Prods. Liab. Litig. (No. II) MDL 2502, 892 F.3d 624, 641

(4th Cir. 2018). Instead, it merely underscores the

widely recognized notion that picking apart an

expert’s supporting data, literature, or other source

material is typically the province of cross-examination

and other forms of adversarial rebuttal. See

Gopalratnam v. Hewlett-Packard Co., 877 F.3d 771, 781

(7th Cir. 2017) (“The district court usurps the role of

the jury . . . if it unduly scrutinizes the quality of the

expert’s data and conclusions rather than the

reliability of the methodology the expert employed.”).

A review of the jurisprudence of the First and

Fourth circuits shows that these courts take seriously

their obligation to “gatekeep” expert testimony that

rests on an unsound factual basis. Further, none of the

cases cited by Petitioners involve the type of legal error

they claim: judges categorically refusing to consider

the factual bases of an expert’s opinion as part of the

Rule 702 reliability analysis. Instead, each case

faithfully performs the screening required by Rule

702(b) while honoring Daubert’s preference for adver-

sarial testing of disputed or doubtful expert opinion.

Petitioners’ claims notwithstanding, there are no

rogue jurisdictions in need of correction by this Court.

A. The Fourth Circuit

The Fourth Circuit has a robust body of law

affirming the exclusion of expert opinion that lacks a

sufficient factual basis. For example, in Tyger

Construction Co., Inc. v. Pensacola Construction Co., 29

F.3d 137, 144 (4th Cir. 1994), the appeals court

affirmed the district court’s exclusion of an expert

whose opinion was “based on assumptions which

[found] no support in the record.” The expert made a

host of assumptions that were either contradicted by

the record or simply made up out of whole cloth. 29

F.3d at 143–44. The appellate panel reversed the

district court’s decision to admit this testimony after

scrutinizing each faulty assumption in detail. Id. at

142–45.

Crucially, Tyger repudiated the “trial judge’s

belief . . . that the question of whether an expert’s

opinion had an adequate basis in fact should be

handled by opposing counsel through cross

examination and in jury argument.” Id. (emphasis

added). The court held that admitting an unreliable

expert opinion on this basis “was an abuse of

discretion” because district courts “may not abdicate

[their] responsibility to ensure that only properly

admitted evidence is considered by the jury.” Id. at 143.

Tyger is no outlier. In EEOC v. Freeman, 778 F.3d

463 (4th Cir. 2015), the appeals court affirmed the

exclusion of an expert who relied on a database that

inexplicably omitted critical data about the defendant’s

hiring practices. The court held that the “sheer

number of mistakes and omissions in [the expert’s]

analysis,” including the unexplained absence of

“hundreds of [job] applicants,” rendered it unreliable.

Id. at 467. Notably, in concurrence, Judge Agee

specifically rejected the EEOC’s argument that

“[p]urported flaws in [the expert’s] analyses concerned

data . . . and therefore concerned weight/credibility

issues for trial, not admissibility.” Id. at 472 (cleaned

up). Judge Agee wrote that “no court has accepted the

agency’s argument” because it “ignores Daubert’s

instruction that the district court must act as a

gatekeeper.” Id.

Similarly, in Le Doux v. Western Express, Inc., 126

F.4th 978, 984 (4th Cir. 2025) the appeals court

affirmed the exclusion of an accident reconstructionist

who, in tandem with a weather modeling expert,

sought to depict the conditions of a highway

accident during heavy rain. The court deemed the

reconstructionist’s method unreliable “because he

used inaccurate time stamps” that made it impossible

to accurately predict when the rain began, which was

a key issue in the case. Id. at 985. Thus, because the

expert relied on spurious data, the district court

properly excluded his testimony.

The Fourth Circuit has also excluded expert

testimony that “‘cherry-picks’ relevant data . . . .”

Lipitor, 892 F.3d at 634 (citation omitted). And its preDaubert jurisprudence upholds the same core

principle of screening expert testimony without a

proper factual foundation. See E. Auto Distribs., Inc. v.

Peugeot Motors of Am., Inc., 795 F.2d 329, 338 (4th Cir.

1986) (affirming exclusion of expert “[i]n light of the

unsupported and speculative assumptions underlying

[his] calculations”).

Just as consistently, the Fourth Circuit affirms the

exclusion of expert testimony based on there being “too

great an analytical gap” between the expert’s

conclusions and the available data. See United States

v. Hudak, 156 F.4th 405, 409–10 (4th Cir. 2025)

(affirming exclusion based on “striking incongruence”

between expert’s conclusion that mental health issues

prompted defendant’s aggressive behavior and “her

report, which included almost no facts about the period

of time in which the conduct occurred”); Belville v. Ford

Motor Co., 919 F.3d 224, 234 (4th Cir. 2019) (affirming

exclusion based on “considerable gap between [expert’s]

theory and any evidentiary proof of causation”).

These precedents are not undermined by the

statement from Bresler v. Wilmington Trust Co., 855

F.3d 178, 195 (4th Cir. 2017), that “questions regarding

the factual underpinnings of the expert witness’

opinion affect the weight and credibility of the witness’

assessment, not its admissibility.” (cleaned up). As the

Fourth Circuit recognized here, that notion does not

vitiate district courts’ obligation to exclude expert

testimony that is without a sufficient factual basis or

suffers from an “analytical gap” problem. App’x 27a

n.7. Instead, Bresler’s rule means that “a district court

may not exclude expert testimony based on (1) its mere

disagreement with an expert’s choice of data or (2) its

own assessment of the correctness of an expert’s

opinions.” Id. This is fully consonant with Daubert’s

principle of reliability, under which “[t]he focus . . .

must be solely on principles and methodology, not on

the conclusions that they generate.” 509 U.S. at 595.

Lastly, Petitioners’ depiction of the Fourth Circuit as

a rogue jurisdiction that fails to heed the 2023

amendments to Rule 702 runs headlong into Sardis v.

Overhead Door Corporation, 10 F.4th 268 (4th Cir.

2021), where the court praised the proposal that

preceded the amendments. The Sardis Court

extolled the Advisory Committee on Evidence Rules’

effort to correct the “pervasive problem” of judges “not

apply[ing] the preponderance standard of admissibility to [Rule 702’s] requirements of sufficiency of basis

and reliable application of principles and methods,

instead holding that such issues were ones of weight

for the jury.” Id. at 283–84. Sardis squarely refutes

Petitioners’ position that Fourth Circuit precedent is

“irreconcilable with . . . the clear direction provided by

the 2023 amendments to Rule 702.” Petition at 18.

B. The First Circuit7

The First Circuit also recognizes that expert

testimony should be excluded if it lacks an evidentiary

basis or is plagued by too many “analytical gaps.” For

example, in Bricklayers & Trowel Trades International

Pension Fund v. Credit Suisse Securities (USA), 752

F.3d 82, 91 (1st Cir. 2014), the appeals court affirmed

the district court’s exclusion of an economist who

arbitrarily chose “event dates” that drove his

calculation of losses on certain stock trades. The court

rejected the plaintiffs’ argument that the district court

excluded their expert based on a “credibility

determination,” finding instead that that the expert’s

chosen dates were “unrelated to the [plaintiffs’]

allegations and therefore [did] not ‘help the trier of fact

to understand the evidence or to determine a fact in

issue.’” Id. at 92 (quoting Fed. R. Evid. 702(a)).

Bricklayers is of a piece with numerous other cases

holding that district courts “may exclude expert

testimony . . . that . . . has no foundation or rests on

Because the Fourth Circuit’s jurisprudence complies with

Rule 702(b), this Court should not use this case as a vehicle to

render an advisory opinion on any independent flaws it identifies

with the First Circuit’s approach.

obviously incorrect assumptions or speculative evidence.”

See, e.g., Casas Office Machines, Inc. v. Mita Copystar

Am., Inc., 42 F.3d 668, 681 (1st Cir. 1994); Irvine v.

Murad Skin Research Labs., Inc., 194 F.3d 313, 321

(1st Cir. 1999) (“Absent adequate factual data to

support the expert’s conclusions his testimony was

unreliable.”).

As in the Fourth Circuit, these principles easily

harmonize with the rule that “[w]hen the factual

underpinning of an expert’s opinion is weak, it is a

matter affecting the weight and credibility of the

testimony—a question to be resolved by the jury.”

Milward v. Acuity Specialty Prods. Grp., Inc., 639 F.3d

11, 22 (1st Cir. 2011) (citation omitted). This principle

does not completely bar district court consideration of

the factual bases of an expert’s testimony. See

Carmichael v. Verso Paper, LLC, 679 F. Supp. 2d 109,

119 (D. Me. 2010) (“The adequacy of an evidentiary

foundation for the admissibility of an expert opinion

can be properly challenged with a Daubert motion.”).

Rather, it merely reinforces Daubert’s express preference

for “vigorous cross-examination over exclusion.” Id.;

see also Crowe v. Marchand, 506 F.3d 13, 18 (1st Cir.

2007) (“Objections . . . which question the factual

underpinnings of an expert’s investigation[] often go

to the weight of the proffered testimony, not to its

admissibility.”) (emphasis added).

Milward bears this out. There, the appeals court

reversed the district court’s exclusion of a highly

qualified expert8 who used the “weight of the evidence”

The expert was “acknowledged as a leading expert on the

study of the toxic effects of chemicals and drugs on the human

body, with particular emphasis on the mechanisms by which

benzene and its metabolites cause damage to both cells and the

human organism as a whole.” 639 F.3d at 15.

methodology to draw a causal connection between a

rare type of leukemia and the plaintiff ’s “workplace

exposure to benzene-containing products . . . .” 639 F.3d

at 13, 16. In a nutshell, the expert’s methodology

involved a cumulative finding inferred from a range of

sources, none of which was independently sufficient to

support the ultimate conclusion. Id. at 23.

The district court took aim at the studies relied upon

by the expert, raising doubts as to whether each one,

standing alone, supported a causal inference between

benzene exposure and leukemia. The lower court also

cited countervailing studies that, in its view, disproved

causation. Id. at 20–21. The First Circuit found two

errors in this approach: first, it “repeatedly challenged

the factual underpinnings of [the expert’s] opinion,

and took sides on questions that are currently the

focus of extensive scientific research and debate—and

on which reasonable scientists can clearly disagree.”

Id. at 22. Second, by demanding that each study relied

on by the expert independently prove causation, the

district court misunderstood the cumulative nature of

the “weight of the evidence” method. Id. at 23.

Thus, Milward did not simply wave the expert’s

opinion on to trial, as Petitioners contend. Petition at

20–21. Instead, it reversed an overzealous district

court that waded into disputed scientific questions

that should have been resolved at trial. Milward thus

stands for the uncontroversial proposition that when a

district court exhaustively interrogates the expert’s

factual basis rather than screening it for reliability, it

“cross[es] the boundary between gatekeeper and trier

of fact.” 639 F.3d at 22.9

Milward also reiterated that “expert testimony may be

excluded if there is too great an analytical gap between the data

Petitioners cite Doucette v. Jacobs, 106 F.4th 156 (1st

Cir. 2024) and Rodriguez v. Hospital San Cristobal, 91

F.4th 59 (1st Cir. 2024), as evidence that the First

Circuit continues to apply an errant rule. But Doucette

shows the opposite. There, the appeals court affirmed

the exclusion of a physician’s testimony based on her

“failure to ground her conclusions in the specifics of

the record – or even to consider key aspects of the

record . . . .” 106 F.4th at 170. These failings “meant

that [her] report fell short of Rule 702’s requirements

that her ‘testimony [be] based on sufficient facts or

data’ and that she ‘reliabl[y] appl[y] [her] principles

and methods to the facts of the case.’” Id. (quoting Fed.

R. Evid. 702(b), (d)).

Crucially, Doucette cited both Milward’s “factual

underpinnings” rule and Joiner’s rule regarding

“analytical gaps.” Id. at 169. And the case ended with

the exclusion of an unreliable expert. Doucette is thus

living proof that these rules complement each other

and do not, as Petitioners contend, render the First

Circuit a rogue jurisdiction where courts blink at every

challenge to the factual basis of an expert’s testimony.10

C. Other Circuits

Since the 2023 amendments to Rule 702, other

circuits, too, have cautioned district courts against

and the opinion proffered.” 639 F.3d at 15. It simply concluded

that, on the facts before it, “the gap was of the district court’s

making.” Id. at 22 (quoting Kennedy v. Collagen Corp., 161 F.3d

1226, 1230 (9th Cir. 1998)).

Rodriguez also affirmed the exclusion of unreliable expert

testimony, but not on “sufficient facts or data” grounds. 91 F.4th

at 72 (affirming district court’s conclusion that expert’s opinion

was not “the product of reliable principles and methods”) (quoting

Fed. R. Evid. 702(c)).

delving too deeply into experts’ factual bases when

conducting Daubert’s reliability analysis. See, e.g.,

Exafer Ltd v. Microsoft Corp., -- F.4th --, No. 2024-2296,

2026 WL 627886, at *3 (Fed. Cir. Mar. 6, 2026)

(reversing exclusion in part because Microsoft’s own

documents supported expert opinion); Teradata Corp.

v. SAP SE, 124 F.4th 555, 571 (9th Cir. 2024), cert.

denied, 146 S. Ct. 118 (2025) (“As with Asker’s other

conclusions, a trier of fact might disagree. But at this

stage, it is not our role to determine the veracity of the

expert’s conclusions. Asker’s assumption that runtime

HANA provides analytical functionality is sufficiently

plausible to constitute a competing version[ ] of the

evidence.”) (citation omitted and cleaned up); United

States v. Markovich, 95 F.4th 1367, 1377 (11th Cir.

2024) (“The Markoviches’ objections to the inadequacies of Clark’s study attack the weight of the evidence,

not its admissibility.”) (citation omitted and cleaned

up); Acad. Bank, N.A. v. AmGuard Ins. Co., 116 F.4th

768, 791 (8th Cir. 2024), reh’g denied, No. 23-1375,

2024 WL 4499662 (8th Cir. Oct. 16, 2024) (“[A]ny

failure to take account of windchill, pressure, and the

amount of water merely provided material for crossexamination or opposing expert testimony.”).

These cases further demonstrate that the First and

Fourth circuits are not outliers – they are in good

company when it comes to ensuring that Rule 702(b)’s

“sufficient facts or data” requirement does not replace

the adversary process. Thus, there is no “deep fissure,”

Petition at 30, or even a split among circuits regarding

application of Rule 702, making this Court’s intervention wholly unnecessary.

CONCLUSION

For the reasons given, the Petition should be denied.

Respectfully submitted,

ADAM J. GOMEZ

KELLY L. TUCKER

GRANT & EISENHOFER, P.A.

123 Justison Street

Wilmington, DE 19801

(302) 622-7000

JOSH AUTRY

Counsel of Record

MORGAN & MORGAN, PA

199 Water St

Suite 1500

New York, NY 10022

(859) 899-8785

jautry@forthepeople.com

STEPHEN EDWARDS

MORGAN & MORGAN, PA

2005 Market St

Suite 600

Philadelphia, PA 19103

(215) 446-9793

Counsel for Respondent

April 6, 2026

This is a copy of a public record, reproduced as it was published. It is not legal advice, and it may not be the version a court would rely on. Check the official source before you cite it.