DZIELAK et al v. WHIRLPOOL CORPORATION et al

Filing 236

OPINION. Signed by Judge Kevin McNulty on 3/17/2017. (JB, )

UNITED STATES DISTRICT COURT
FOR THE DISTRICT OF NEW JERSEY

CHARLENE DZIELAK, et al.,
Civ. No. 2:12-0089

Plaintiffs,

(KM) (JBC)
V.

OPINION
WHIRLPOOL CORPORATION, et al.,

Defendants.

KEVIN MCNULTY. U.S.D.J.:
This putative class action arises from the sale of Maytag washing
machine models that bore ENERGY STAR® labels signifying that they met
federal standards of water and electrical efficiency.’ In May 2012, the U.S.
Department of Energy determined that these particular models did not actually
meet Energy Star efficiency standards, and thus disqualified them from the
program. The plaintiffs, including named plaintiff Charlene Dzielak, are
purchasers of Maytag Centennial models MVWC6ESWWO, MVWC6ESWW1,
2
and MVWC7ESWWO (the “Washers”) They claim that they and a putative
class suffered uniform harm because they all purchased Washers improperly
labeled with the Energy Star logo. The price of the Washers was inflated, they
claim, because consumers will pay a premium for Energy Star-qualified
washing machines, “due to their.

.

.

supposed water usage, energy efficiency,

Familiarity with my earlier opinions (ECF No. 78, 127) is assumed. Herein, I will
render ENERGY STAR® as “Energy Star”.
Models MVWC6ESWWO and MVWC6ESWWI are hereinafter referred to as the
C6 models and MVWC7ESWWO is referred to as the C7 model.

2

1

and ENERGY STAR® qualification.” (Second Amended Consolidated Complaint,
ECF No. 86,

¶

118) Additionally, they claim that each purchaser “paid more

money in additional water and energy costs to operate his or her Mislabeled
Washing Machine than they would have had the appliance actually met the
ENERGY STAR® qualification as represented and promised by Whirlpool.” (Id.)
This opinion addresses four motions in limine—three filed by the
defendants and one by the plaintiffs—seeking to strike proffered expert
testimony concerning class-wide damages. For the most part, the motions are
denied. For the reasons discussed below, I will deny the defendants’ motions to
strike the expert opinions of Dr. Ramamirtham Sukumar and Dr. Michael J.
Dennis (ECF nos. 198, 199). I will grant in part the defendants’ motion to strike
the expert opinion of Mr. Cohn Weir (ECF No. 200), insofar as it is based on
data presented in an anonymous Whirlpool document. Finally, I will deny the
plaintiffs’ cross-motion to strike Dr. Carol A. Scott’s rebuttal and supplemental
expert opinions as a discovery sanction under Federal Rule of Civil Procedure
37 (ECF No. 221).
I.

FACTS & PROCEDURAL HISTORY
The defendants are Whirlpool, which manufactured the plaintiffs’

Washers, and several retailers who sold them: Lowe’s Home Centers, LLC
(“Lowe’s”), Sears Holdings Corporation (“Sears”), The Home Depot, Inc. (“Home
Depot”), Fry’s Electronics, Inc. (“Fry’s”) and Appliance Recycling Centers of
America, Inc. (“ARC”). In my July 31, 2015 opinion on the defendants’ motion
to dismiss the plaintiffs’ second amended complaint, I dismissed plaintiffs’
Magnuson-Moss Warranty Act and unjust enrichment claims against
Whirlpool. What remain are state law claims of breach of express warranty and
the implied warranty of merchantability; unjust enrichment as to the
defendants other than Whirlpool; and violation of consumer protection
statutes, including the New Jersey Consumer Fraud Act (“NJCFA”). (See ECF
No. 127)

2

As relevant here, the plaintiffs have proffered the opinions of three
experts:
(a) Ramamirtham Sukumar, PhD, who means to isolate the portion of the
purchase price of the Washers that is attributable to the Energy Star “price
premium”;
(b) Michael J. Dennis, PhD, who conducted a contingent valuation survey
meant to isolate the portion of the purchase price that is attributable solely to
the Energy Star label; and
(c) Cohn Weir, who proposes a model for calculating class-wide “price
premium” and “energy expense” damages.
Each of these three has submitted a declaration and expert report in
support of the plaintiffs’ motion for class certification. (ECF nos. 171—173). In
response to the defendants’ opposing expert reports, each of the three has
submitted a rebuttal report. (ECF nos. 184—186).3 Pursuant to Daubert v.
Men-eli Dow Pharmaceuticals, 509 U.S. 579 (1993), and Federal Rule of
Evidence 702, Whirlpool, Lowe’s, Sears, and Fry’s (for purposes of this motion,
3
The plaintiffs previously filed with this Court a motion for class certification. By
order dated July 25, 2016 (ECF No. 203), I responded to the parties’ request that that
motion be considered in conjunction with the four motions in limine addressed in this
opinion. I administratively terminated the class certification motion and instructed the
plaintiffs to renew that motion, supplementing as appropriate, following my decision
on the four motions now before me. This is consistent with the Third Circuit’s recent
directive that “a plaintiff cannot rely on challenged expert testimony, when critical to
class certification, to demonstrate conformity with Rule 23 unless the plaintiff also
demonstrates, and the trial court fmds, that the expert testimony satisfies the
standard set out in Daubert.” In re Blood Reagents Antitrust Litig., 783 F.3d 183, 187
(3d Cir. 2015). Thus, although the expert and class certification issues are
intertwined, it seemed best to sort out the Daubert issues first.
In connection with class certification, the plaintiffs intend to show that the
requirements of Fed. R. Civ. P. 23(b)(3) are satisfied. By that they mean that
“questions of law or fact common to class members predominate over any questions
affecting only individual members,” based in part on the opinions of their damages
expert, Mr. Weir. Weir calculates “energy expense damages” and “price premium
damages,” which plaintiffs argue can be calculated on a classwide basis, and thus are
provable by common evidence. (ECF No. 163 (sealed) at 17, 27)

3

the “defendants”) have moved to preclude these three experts from testifying.
(ECF nos. 198—200)
Whirlpool, for its part, has proffered an initial report, as well as a
rebuttal report and a supplemental report, authored by Carol A. Scott, PhD.
Dr. Scott is offered as an opinion witness critical of Dr. Sukumar’s and Dr.
Dennis’s price premium estimates, as well as Mr. Weir’s damages model. The
plaintiffs assert that Scott’s rebuttal and supplemental reports were untimely
filed under Federal Rule of Civil Procedure 26 (“Rule 26”) and this Court’s
January 6, 2015 Scheduling Order (ECF No. 103), as modified by the February
8, 2016 Order Extending Deadlines (ECF No. 174, hereinafter, the “Scheduling
Order”). The plaintiffs have therefore filed a cross-motion (ECF No. 221) to
exclude Dr. Scott’s rebuttal and supplemental reports as a sanction under
Federal. Rule of Civil Procedure 37 (“Rule 37”).
II.

DEFENDANTS’ MOTIONS INLIMLNE
A.

Legal Standards: Admissibility Under Rules 702 and 703

Federal Rule of Evidence 702 (“Rule 702”) states certain prerequisites to
the admission of expert testimony:
A witness who is qualified as an expert by knowledge, skill,
experience, training, or education may testify in the form of
an opinion or otherwise if:
(a) the expert’s scientific, technical, or other
specialized knowledge will help the trier of fact to
understand the evidence or to determine a fact in
issue;
(b) the testimony is based on sufficient facts or data;
(c) the testimony is the product of reliable principles
and methods; and
(d) the expert has reliably applied the principles and
methods to the facts of the case.
Fed. R. Evid. 702.
The case law has extracted from Rule 702 three essential requirements:
“qualification, reliability and fit.” Schneider ex rel. Estate of Schneider v. Fried,
4

320 F.3d 396, 404 (3d Cir. 2003) (citing In re Paoli R.R. Yard PCB Litig., 35 F.3d
717, 74 1-43 (3d Cir. 1994) (citing Daubed v. Merrell Dow Pharmaceuticals, Inc.,
509 U.S. 579 (1993))).

“Qualification refers to the requirement that the witness possess
specialized expertise.” Schneider, 320 F.3d at 404. This is interpreted liberally:
“a broad range of knowledge, skills, and training qualify an expert.” Paoli, 35
F.3d at 741.
“Reliability” requires that the opinion be “based on ‘the methods and
procedures of science’ rather than on ‘subjective belief or unsupported
speculation.” Paoli, 35 F.3d at 742 (citing Daubed, 509 U.S. at 590). Reliability
is a “flexible” test. Kumho Tire Co., Ltd. v. Carmichael, 526 U.S. 137, 141 (1999)
(internal citation omitted). The Third Circuit has explained that factors
“deemed important” for determining reliability include:
(1) whether a method consists of a testable hypothesis;
(2) whether the method has been subject to peer
review; (3) the known or potential rate of error; (4) the
existence and maintenance of standards controlling
the technique’s operation; (5) whether the method is
generally accepted; (6) the relationship of the
technique to methods which have been established to
be reliable; (7) the qualifications of the expert witness
testifying based on the methodology; and (8) the non
judicial uses to which the method has been put.
In re Paoli R.R. Yard PCB Litig., 35 F.3d at 791 n.8. These factors are not

exclusive, nor must they all be applied in every case. Kumho Tire Co., 526 U.S.
at 141; Elcock v. Kmart Coip., 233 F.3d 734, 746 (3d Cir. 2000). Rather,
whether “specific factors are, or are not, reasonable measures of reliability in a
particular case is a matter that the law grants the trial judge broad latitude to
determine.” Kumho Tire Co., 526 U.S. at 153.
“Fit” requires that the expert opinion correspond to the issues in the
case, i.e., that it be relevant and “assist the trier of fact.” Paoli, 35 F.3d at 74243. “Expert testimony which does not relate to any issue in the case is not
relevant and, ergo, non-helpful.” Daubed, 509 U.S. at 591 (quoting 3 Weinstein
5

& Berger

¶

702[02], p. 702—18). Therefore, “even if an expert’s proposed

testimony constitutes scientific knowledge, his or her testimony will be
excluded if it is not scientific knowledge for purposes of the case.” Id. at 743.
The burden of demonstrating admissibility under Rule 702 falls on the
proponent: “The party offering the proposed expert testimony bear[] the burden
of establishing the admissibility of the testimony by a preponderance of the
evidence.” In re Human Tissue Products Liab. Litig., 582 F. Supp. 2d 644, 655
(D.N.J. 2008) (citing Padillas v. Stork-Gamco, Inc., 186 F.3d 412, 417-18 (3d
Cir. 1999); see also Oddi u. Ford Motor Co., 234 F.3d 136, 144 (3d Cir. 2000)

(quoting Daubert, 509 U.S. at 593 n. 10). That inquiry requires the court to
“examine the expert’s conclusions in order to determine whether they could
reliably follow from the facts known to the expert and the methodology used.”
Heller v. Shaw Indus., Inc., 167 F.3d 146, 153 (3d Cir. 1999).

The District Court, exercising its sound discretion, serves as a gatekeeper
to prevent expert testimony that falls short of these requirements from reaching
the jury. Daubert, 509 U.S. at 592-95. Nevertheless, “[t]he Rule’s basic
standard of relevance

.

.

.

is a liberal one.” Id. at 587. Within the principles

outlined above, a judge has “the same broad latitude when it decides how to

determine

reliability as it enjoys

in respect to

its ultimate reliability

determination.” Kumho Tire, 526 U.S. at 142 (internal citation omitted)
(emphasis in original).
Federal Rule of Evidence 703 (“Rule 703”) sets forth the requirements
relating to the underlying facts or data on which an expert may base his or her
opinion. It provides:
An expert may base an opinion on facts or data in the
case that the expert has been made aware of or
personally observed. If experts in the particular field
would reasonably rely on those kinds of facts or data
in forming an opinion on the subject, they need not be
admissible for the opinion to be admitted. But if the
facts or data would otherwise be inadmissible, the
proponent of the opinion may disclose them to the jury
only if their probative value in helping the jury
6

evaluate the opinion substantially outweighs their
prejudicial effect.
Fed. R. Evid. 703.
The Third Circuit announced in In re Paoli that “when a trial judge
analyzes whether an expert’s data is of a type reasonably relied on by experts
in the field, he or she should assess whether there are good grounds to rely on
this data to draw the conclusion reached by the expert.” 35 F.3d at 749. In
making that assessment, Paoli held, a trial judge has considerable latitude. The
judge can “take into account the particular expert’s opinion that experts
reasonably rely on that type of data, as well as the opinions of other experts as
to its reliability, but the judge can also take into account other factors he or
she deems relevant.” Id. at 748.
B.

Defendants’ Motion to Strike the Opinions of Dr. Sukumar

The defendants’ first motion attacks the opinions of the plaintiffs’ “price
premium” expert, Dr. Sukumar, under Daubed’s reliability and fit
requirements. (Defendants do not question Dr. Sukumar’s qualifications.)
The defendants first challenge Dr. Sukumar’s proprietary method of
conjoint analysis, called “ASEMAP”. Second, they challenge Dr. Sukumar’s
conclusions and opinions, which they contend are inconsistent with market
reality. Third, the defendants object to the composition of Dr. Sukumar’s
surveyed population and his exclusion of certain data when arriving at his
conclusions. Finally, the defendants argue that, under Daubed and Comcast
Corp. v. Behrend,

—

U.S.

,

133 S. Ct. 1426, 1433 (2013), Dr. Sukumar’s

opinion must be excluded because his conjoint analysis does not “fit” the
plaintiffs’ theory of liability.
1. Overview of Dr. Sukumar’s Analysis

The plaintiffs asked Dr. Sukumar to “estimate the price premium, if any,
attributable to the Energy Star logo” on the Washers “through the use of a

7

conjoint analysis.” (Sukumar Br. 1 (quoting Sukumar Report 4))4 Conjoint
analysis involves a survey in which respondents are asked “to make a series of
trade-offs between different product features and prices.” The object and result
of conjoint analysis is to “determine the aggregated values customers attach to
different features of a product.” (Sukumar Br. 4 (quoting Sukumar Report 4—
5)); Sukumar Opp. 4)
Dr. Sukumar holds an MBA and PhD in marketing and statistics. He
identifies conjoint analysis, and in particular ASEMAP, as the appropriate tool
to isolate the price premium solely attributable to the Energy Star logo.
Conjoint analysis, he says, “is designed to value a specific product feature,
even in circumstances where the feature’s price is not expressly set by the
manufacturer or retailer.” (Sukumar Opp. 3; see also Sukumar Report 4—5)
Here, the Washers possess many features of potential value to consumers; he
identified 19 such features through a series of consumer focus groups.
(Sukumar Opp. 5) Thus he settled on a specific form of conjoint analysis called
Adaptive Self-Explication of Multi-Attribute Preferences (“ASEMAP”), because it
is “specifically designed to be capable of handling surveys that incorporate a
number of product features.” Other forms of conjoint analysis, he reasoned,
were inferior in this case because they performed less accurately than ASEMAP
The following abbreviations will be used to cite Dr. Sukumar’s submissions and
the briefing on the defendants’ motion to strike his opinion testimony:
•

“Sukumar Report” = Declaration and Expert Report of Dr. R. Sukumar, Dec. 28,
2015, Deckant Deci. Ex. R, ECF No. 2 15-18

•

“Sukumar Rebuttal” = Rebuttal Expert Report of Ramamirtham Sukumar,
Ph.D., Deckant Deci. Ex. 5, ECF No. 2 15-19

•

“Sukumar Br.” = Brief in Support of Motion to Strike the Opinions of Dr.
Sukumar by Defendants Whirlpool Corporation, Lowe’s Home Centers, LLC,
Sears Holdings Corporation, and Fry’s Electronics, Inc., ECF No. 198

•

“Sukumar Opp.” = Plaintiffs’ Opposition to Motion to Strike the Opinions of Dr.
Sukumar by Defendants Whirlpool Corporation, Lowe’s Home Centers, LLC,
Sears Holdings Corporation, and Fry’s Electronics, Inc., ECF No. 214

•

“Sukumar Reply” = Reply Brief in Support of Motion to Strike the Opinions of
Dr. Sukumar by Defendants Whirlpool Corporation, Lowe’s Home Centers, LLC,
Sears Holdings Corporation, and Fry’s Electronics, Inc., ECF No. 228
8

in the literature, had not been subject to peer review, or are better suited to
products with a smaller number of features than those identified here. (Id. 4—5;
Sukumar Report 8-9)
Dr. Sukumar’s analysis drew on an online survey of a random sample of
564 purchasers of washing machines. He compared the values for the Washers
(represented as costing being $300 and $500) with and without the Energy Star
logo to calculate, via ASEMAP, the value attributable to the logo alone. His
analysis concluded that the logo had a value of $180.39, or “44.3% of the
average clothes washing machine price of $406.99.” (Sukumar Opp. 7;
Sukumar Report 14—17). This result represented an average across the
respondent sample and also takes into account differences in how individual
respondents value a dollar. (Sukumar

Opp.

6; Sukumar Report 14-16)

1) Objection to ASEMAF
The defendants specifically argue that ASEMAP is not a reliable
methodology capable of satisfying Daubed because it is “a little-known
proprietary method that is rarely used—if at all—outside [Dr. Sukumar’sl own
[consulting] firm.” (Sukumar Br. 11) Additionally, the defendants argue,
ASEMAP has been peer-reviewed in only one article, authored by its inventor.
And even that article never specifically discusses the use of the ASEMAP
methodology for determining “price premiums” (Id. 16). Defendants add that
Dr. Sukumar’s conjoint analysis cannot be replicated because ASEMAP is a
proprietary methodology that employs an undisclosed algorithm. (Id. 14)
The plaintiffs respond generally that methodological criticisms of
consumer surveys go to the weight of the evidence, not to its admissibility.
(Sukumar Opp. 8—9). More specifically, they say that conjoint analysis is a
“widely-accepted survey research method for measuring consumer
preferences,” as numerous courts have acknowledged. (Id. 9—10). The
defendants’ challenge to ASEMAP, they argue, is based only on the
unsupported opinion of the defendants’ expert Dr. Peter E. Rossi. (Id. 11) To

9

exclude Dr. Sukumar’s testimony, then, would usurp the jury’s prerogative to
find one expert more persuasive than another. (Id. 11)
The plaintiffs point to other federal courts’ acceptance of conjoint
analysis, if not ASEMAP specifically, “as a valid method of estimating changes
in market value for purposes of performing damage calculations in litigation.”
(Id. 12 (citing TV Interactive Data Corp. v. Sony Corp., 929 F. Supp. 2d 1006,
1022 (N.D. Cal. 2013) (defendant’s challenge to conjoint surveys as
fundamentally flawed and unreliable go to the weight of the survey rather than
admissibility); Microsoft Corp. v. Motorola Inc., 904 F. Supp. 2d 1109, 1120
(W.D. Wash. 2012) (conjoint analysis critiques went “to issues of methodology,
survey design, reliability, and critique of conclusions, and therefore go to the
weight of the survey rather than admissibility”)).
The plaintiffs seemingly concede that that the only peer-reviewed journal
article evaluating ASEMAP is one written by its inventor, Professor Srinivasan.
That circumstance does not, however, detract from the article’s validity in the
plaintiffs’ view. Professor Srinivasan is said to be “an undisputed expert in the
conjoint analysis technique.” (Sukumar Br. 15) Plaintiffs also seem to
acknowledge that the article does not specifically discuss ASEMAP in relation
to the calculation of price premiums. Nevertheless, they say, the article more
generally demonstrates the ASEMAP method’s external and predictive validity.
Plaintiffs also note that Daubert cites publication as a factor, but does
not require or unduly stress it: “Publication (which is but one element of peer
review) is not a sine qua non of admissibility; it does not necessarily correlate
with reliability,

.

.

.

and in some instances well-grounded but innovative

theories will not have been published.

.

.

.“

509 U.S. at 593—94. Peer review

aside, Dr. Sukumar states that he himself has employed ASEMAP extensively,
both in and outside the context of litigation. Whether or not its accuracy has
been established “exclusively through academic journals,” ASEMAP has
withstood the test of time without revealing significant flaws. (Id. 12)

10

As for replicability, the plaintiffs point out that although ASEMAP is
proprietary, it is available to anyone who purchases a software license.
Therefore, they say, the defendants could have independently verified or
challenged Dr. Sukumar’s opinions, but did not do so. If skeptical of the
software itself, defendants could have moved to compel Dr. Sukumar to
produce ASEMAP’s confidential source code, but again they did not. The
discovery they did request, they got: Dr. Sukumar directed defendants to a
peer-reviewed publication that guides readers through ASEMAP, produced his
complete worksheets and datasheets, and detailed the design and
implementation of his survey methods in his expert report. (Id. 12—13)
I find the plaintiffs’ arguments persuasive. Dr. Sukumar’s reports are
well-reasoned, and there is ample evidence that his methods are sufficiently
testable and accepted in the relevant scientific community, as well as the legal
community. Conjoint analysis has won acceptance from courts and legal
commentators. See, e.g., ConAgra Foods, Inc., 844 F.3d 1121(9th Cir. 2017)
(acknowledging that “marketing researchers have used conjoint analysis since
the early 1970’s to determine the values consumers ascribe to specific
attributes of multi-attribute products and to understand the features driving
product preferences,” and holding that conjoint analysis is sufficiently reliable
to be used to calculate class-wide damages); TV Interactive Data Corp. v. Sony
Corp., 929 F. Supp. 2d 1006, 1022 & n.6 (N.D. Cal. 2013) (in product

mislabeling class action, determining that conjoint analysis is accepted in the
relevant community and acknowledging “a handful of cases to demonstrate
that conjoint analysis is increasingly used in litigation”); Sukumar Opp. 10—11
(collecting cases where experts’ conjoint analysis was accepted).
ASEMAP, the proprietary form of conjoint analysis that Dr. Sukumar
employs, is not specifically discussed in the case law, and according to the
defendants, it boasts just one inventor-authored peer-reviewed article. (Dr.
Sukumar’s rebuttal report adds citations to two peer-reviewed articles in
support of the claim that ASEMAP has “external and predictive validity.”

11

(Sukumar Rebuttal 9))5 In the case of a technique that is a variant of a known
technique, however, “the relationship of the technique to methods which have
been established to be reliable”—here, conjoint analysis—may be considered. In
re Paoli, 35 F.3d at 791 n.8. Daubed itself recognized that “[s]ome propositions
are too particular, too new, or of too limited interest to he published.” 509
U.S. at 593. Dr. Sukumar states, without effective rebuttal, that ASEMAP is
simply a variant or an elaboration of the widely accepted conjoint analysis
methodology. That proposition is corroborated by the fact that ASEMAP’s
inventor is Professor Srinivasan, who would know. Srinivasan has been
recognized by federal courts as “coin[ing] the term ‘conjoint analysis’,” TV
Interactive Data Corp., 929 F. Supp.2d at 1020, and as the “father of conjoint
analysis.” Visteon Glob. Techs., Inc. v. Garmin Int’l, Inc., No. 10-CV-10578, 2016
WL 5956325, at *2 (E.D. Mich. Oct. 14, 2016). According to plaintiffs, he has
“authored 24 peer-reviewed research papers on the topic.” (Sukumar Br. 15)
Dr. Sukumar states that ASEMAP has proven reliable outside of the
context of litigation. It has been used in over 100 commercial studies and he
has personally employed ASEMAP in “several commercial studies over the last
8 years” (Sukumar Report 6, 9). ASEMAP has also been acknowledged as
reliable in legal scholarship, if not in the case law. See David Franklyn & Adam
Kuhn, The Problem of Mop Heads in the Era of Apps: Toward More Rigorous
Standards of Value Apportionment in Contemporary Patent Law, 98 J. Pat. &
Trademark Off. Soc’y 182, 218—19 (2016) (“[A]dvances in research software,
such.

.

.

Asemap, have made it much easier to design reliable indirect elicitation

To the extent the defendants disagree that these articles validate ASEMAP, see,
e.g., Sukumar Br. 17 n. 13 (arguing that one of the articles Dr. Sukumar cites in his
rebuttal report never mentions ASEMAP), they are free to pursue this on crossexamination as a weakness impacting the weight of Dr. Sukumar’s opinion. And, to
the extent the defendants take issue with the fact that Dr. Srinivasan’s peer-reviewed
article does not specifically advocate ASEMAP as “a valid methodology for isolating a
historic price premium,” that too goes to weight, not admissibility. See, e.g., In re Paoli
R.R. Yard PCB Litig., 35 F.3d 717, 742 (3d Cir. 1994) (“Rule 702 does not incorporate
the common law Frye rule in which expert testimony is admissible only insofar as it is
based on a technique generally accepted in the scientific community.”).
5

12

surveys. These programs can present a series of ‘bundled’ features comprising
a product and ask the respondent to choose the one they would be most likely
to purchase.

.

.

.

Reliable, objective data then coalesces around individual

product feature valuations.” (emphases added)) •6
As for the defendants’ ability to replicate Dr. Sukumar’s ASEMAP
analysis, Daubert explains that “a key question to be answered in determining
whether a theory or technique is scientific knowledge that will assist the trier of
fact will be whether it can be (and has been) tested.” 509 U.S. at 593. The Third
Circuit refers to this consideration as requiring that an expert present a
“testable hypothesis.” In re Paoli, 35 F.3d at 791 n.8.
As defendants view this requirement, it seems to hinge on whether the
opposing party has successfully replicated the proffered expert’s results. (See
Sukumar Br. 14—15) Surely that goes too far. Courts have interpreted
Daubert’s testing factor as one of replicability, not replication. See, e.g., Zenith

Elecs. Corp. v. WH-TVBroad. Corp., 395 F.3d 416, 419 (7th Cir. 2005) (“An
expert must offer good reason to think that his approach produces an accurate
estimate using professional methods, and this estimate must be testable.
Someone else using the same data and methods must be able to replicate the
result.”). To the extent defendants are arguing that they hold the keys to
admissibility—i.e., that this evidence cannot be admitted unless they have
reproduced Dr. Sukumar’s testing—I cannot accept their analysis. And at any
rate, the defendants have never purchased a license, sought discovery of the
source code, or shown any inclination to reproduce Dr. Sukumar’s analysis.
The mere possibility that they could do so, or that they believe their efforts to
do so, if undertaken, would have been thwarted, does not detract from the
testability or validity of Dr. Sukumar’s conclusions.

6

Franklin and Kuhn discuss Professor Srinivasan’s conjoint analysis in TVI Data

Corp. at length. Their discussion indicates that Professor Srinivasan in fact used
ASEMAP in that case, although the Court’s opinion does not specifically identify that
ASEMAP was used. See Franklin & Kuhn, supra, at 208 (citing Declaration of Dr. V.
Srinivasan, TVI Data Corp. v. Sony Corp., 2013 WL 496310 (N.D. Cal., Jan. 18, 2013)).

13

I am thus enjoined to consider whether Dr. Sukumar’s methodology
involves testing of a hypothesis that is testable. The Advisory Committee notes
to Federal Rule of Evidence 702, for example, describe Daubert’s testing factor
as asking “whether the expert’s theory can be challenged in some objective
sense, or whether it is instead simply a subjective, conclusory approach that
cannot reasonably be assessed for reliability.” Fed. R. Evid. 702, Advisory
Committee Notes, 2000 Amendments (emphasis added). To vary the
terminology, “Testability” has also been described as “falsifiability.” A
proposition is “falsifiable” if it is “capable of being proved false; defeasible.”’
7
United States v. Mitchell, 365 F.3d 215, 235 (3d Cir. 2004) (citations omitted)).
Of course a competing expert’s performance of the same test under the
same conditions would be highly probative. But we do not have that here.
Rather we have contending results that are the product of contending
methodologies.
At the Daubert stage, I am to assess the scientific nature of the
challenged theory or technique, not to take a side in a debate among experts
about the results. Here, I have already concluded that ASEMAP has a sufficient
pedigree— i.e., that it passes muster under the Daubert considerations of peer
review, standardization, judicial and non-judicial acceptance, and its
relationship to other established reliable techniques (particularly, the conjoint
analysis technique of which it is a part). See Paoli, 35 F.3d at 791 n.8. I think
these considerations, in addition to supporting the reliability of the method,
tend to establish that ASEMAP is testable, in the sense of producing results
susceptible of an objective challenge.
Dr. Sukumar’s conclusion is that consumers attach a value, in the form
of a price premium, to the Energy Star logo on the Washers. (See Sukumar

“Falsifiability,” as a basis to distinguish science from non-science, is a term that
has passed into general usage. It has its origin, so far as I am aware, in Karl Popper’s
philosophy of science. See Stanford Encyclopedia of Philosophy, “Karl Popper” § 7,
https: / / plato. stanford. edu / entries / popper / #ScieKnowHistPred; id. “Scientific Method”
§ 3.3, https: / / plato. stanford. edu / entries / scientific-method / #PopFal.
14

Report 4) Defendants may attack that conclusion, employing ASEMAP or some
alternative form of analysis. My point here is that it is capable of being tested.
Although the point is not critical to my analysis, Judge Brown’s decision
in United States u. Ewell suggests that defendants are making a category
mistake in arguing that everything about Dr. Sukumar’s software, as opposed
to the ASEMAP methodology itself, must pass the Daubert test. 252 F. Supp.
2d 104 (D.N.J. 2003), aff’d sub nom. United States v. Adams, 189 F. App’x 120
(3d Cir. 2006). In Ewell, a criminal defendant challenged the government’s use
of specific materials kits for performing “PCR/ STS typing” DNA analysis to
identify the defendant. In an opinion admitting the evidence, Judge Brown
explained:
The hypothesis of PCR/STR DNA typing is that with
proper procedures an expert can determine the allelic
types of given DNA samples at the thirteen core STR
loci.
The Court is not persuaded that the claimed
lack of validation of the efficacy of the kits has any
effect on whether PCR/STR typing has a testable
hypothesis. Contrary to defendant’s assertions, the.
kits merely provide the materials necessary to perform
the PCR amplification process, and thus, the kits need
not independently meet the Daubert standard of
admissibility. Accordingly, challenges as to the efficacy
and reliability of the materials kits go to the weight of
the evidence and not to admissibility.
.

.

252 F. Supp. 2d at 111.
But the government in Ewell, like the plaintiffs here, had a backup
argument. It also offered a peer reviewed article attesting to the validity of the
challenged kits. Judge Brown dealt with that backup argument as well:
Moreover, even if the kits were independently subject
to Daubert, defendant’s argument is not well-founded.
The Government produced as exhibit 7 a peer reviewed
article reporting a study of various commercially
available amplification kits. The article concludes,
inter alia, that Profiler Plus and COfiler kits “can be
used to amplify and type STR loci successfully from
DNA derived from human biological specimens. The
current study demonstrates that the procedures used
15

to type STR loci using these commercial kits are
robust and valid.”
Id.
Of course, the defendants might have attempted to dissect the ASEMAP
source code as part of an attack on plaintiffs’ case, but they have not done so. I
do not think the results of such a dissection are a prerequisite to my ruling on
admissibility. Dr. Sukumar’s analysis remains falsifiable. Defendants may test
his conclusions by comparing them to the choice-based conjoint method that
their own expert, Dr. Peter Rossi, endorses in his critique of Dr. Sukumar’s
report, or by some other means. (See Rossi Report 34-35) I may even assume
the validity of applying Daubert to the details of the ASEMAP methodology. For
the reasons expressed above, that methodology, like the kits in Ewell, has
sufficient support in scientific literature and practice, and it has not been
effectively challenged.
In short, even assuming ASEMAP is as tethered to Dr. Sukumar’s
consulting business as the defendants allege, the plaintiffs have presented
sufficient indicia of reliability to assure me that ASEMAP—both generally and
as Dr. Sukumar applies it—is not so substantively flawed as to render Dr.
Sukumar’s opinions inadmissible. See Koninkjjke Philips Elecs. N. V. v. Hunt
Control Sys., Inc., No. CV113684SRCCLW, 2016 WL 3545529, at *5 (D.N.J.
June 29, 2016) (“In the context of survey evidence, ‘mere technical flaws’ in
methodology go to ‘the weight accorded a survey, not its admissibility.”’

(quoting Fancaster, 832 F. Supp. 2d at 402)). ASEMAP’s shortcomings,
whatever they may be, are fodder for cross-examination.
2) Criticisms of Survey Methodology, Population, and Results
Defendants’ main contention is that Dr. Sukumar’s conclusions flunk
the reality test. According to the defendants, most washers, whether they
boasted an Energy Star logo or not, cost more than $400 during the relevant
time period. Subtracting Dr. Sukumar’s calculated $180 Energy Star price
premium from the prevailing price of a washing machine, however, yields an
16

unrealistically low price in the range of $120 to $227. (Sukumar Br. 17—21;
Sukumar Reply 6) Thus, say the defendants, Dr. Sukumar’s analysis cannot be
correct, and must be the product of an unreliable methodology. (Id.)
One cause, or perhaps a symptom, of that unreliability is said to be Dr.
Sukumar’s exclusion of more than 70% of his survey respondents (in
)
8
particular, “Price Reversal” respondents from his sample. Were those
responses considered, say defendants, Dr. Sukumar’s estimated price premium
would have been even more absurdly high. They point out that the plaintiffs
have failed to “cite a single case in which a court admitted a survey despite the
expert’s exclusion of the majority of data.” (Sukumar Br. 21—22; Sukumar
Reply 7) The defendants also object to Dr. Sukumar’s choice to survey
consumers who had recently purchased high-efficiency (“HE”) washing
machines, which allegedly cost hundreds of dollars more than the Washers at
issue. (Sukumar Report 22—25)
Responding to the defendants’ claims that the $180 price premium defies
market reality, the plaintiffs explain that conjoint analysis often assigns values
to individual attributes which, if added together, would exceed the value of the
product. Dr. Sukumar’s exclusion of Price Reversal data, they say, is a
technique that the defendants’ own expert, Dr. Rossi, supports. It also results
in a more conservative price premium estimate, draws from a higher quality
respondent pooi, and has been used in prior cases. (Id. 18—19 (citing Sukumar
Rebuttal at 12—13))
For Dr. Sukumar’s selection of his survey population, the plaintiffs offer
a laundry list of explanations: Dr. Sukumar employed a widely-used online
marketing tool; screened respondents to ensure representativeness of the
putative class (i.e., household decision-makers who purchased washing
machines); conducted the type of survey routinely used and accepted in

Dr. Sukumar excluded from his data set the answers of respondents who
seemingly preferred higher prices to lower prices.

8

17

litigation; and the defendants’ technical criticisms go only to the weight of Dr.
Sukumar’s testimony. (Sukumar Opp. 18—22)
Here, I am guided by the principle that Daubert is essentially results
agnostic. “[I}n deciding whether an expert’s report meets the reliability factor of
a Daubert and Rule 702 analysis, [this] Court is not to weigh the evidence
relied upon or determine whether it agrees with the conclusions reached
therein.” Walker v. Gordon, 46 F. App’x 691, 695 (3d Cir. 2002). In most
respects, the defendants’ contentions amount to attacks on the expert’s
conclusions. They cannot be refashioned into Daubert attacks by the facile
method of saying that such unlikely results could only have been the product
of unscientific methods. Such arguments go to the weight of the expert’s
testimony, not to its admissibility.
The thrust of the challenge here is that Dr. Sukumar excluded from his
data set the answers of respondents who seemingly preferred higher prices to
lower prices. This is asserted as a challenge to the reliability of ASEMAP. (See
Rossi Report 30 (“Dr. Sukumar excludes more than 70 perfect of his
respondents from his WTP computations. It seems that the ASEMAP procedure
and the survey designed by Dr. Sukumar cannot produce data that accord with
the common sense notion that people prefer lower prices to high prices. In my
opinion, this casts grave doubt on the reliability and usefulness of Dr.
Sukumar’s survey data.”)) Ironically, defendants attack the exclusion of
responses on the grounds that including them would result in a higher price
premium—indeed, one that would be self-evidently too high, and call the whole
enterprise into question.
As defendants see it, this would be a case of what statistical wags have
called “torturing the data until it confesses.” It may alternatively be viewed,
9
Some version of the expression has been credited to economist Ronald Coase,
and to Darrell Huff’s classic How to Lie With Statistics (1954). If one posits a
hypothesis, gathers data, discards all data that does not comply with the hypothesis,
then QED, the hypothesis will be proven. That abusive practice, however, must be
distinguished from the common and valid practice of eliminating junk from the data
set.
9

18

however, as the implementation of a methodological assumption—one that
surely may be challenged on cross-examination, but not an unreasonable one.
As I understand Dr. Sukumar’s selection process, it presupposes that what we
are measuring are consumer choices rationally based on price. If that
assumption is accepted, then his claim (I exaggerate to illustrate a point) is
that he has done the equivalent of throwing out ballots voting for both
candidates, or recording a vote for Mickey Mouse. Looking at it from another
perspective, we might say that scientific methods are being applied to the
expressed preferences of consumers who are not always entirely rational or
analytic about their choices (or their responses to surveys). In short, we are
building with crooked timber, but build we must. Whether the resulting
structure is too rickety to withstand the wrecking ball of cross-examination is a
separate question.’°
I agree with plaintiffs that the defendants’ arguments here largely depend
on disagreement with Dr. Sukumar’s conclusions. Such substantive
disagreement is not the correct focus of a Daubert motion. Jama v. Esmor Corr.
Serus., Inc., No. 97-3093 DRD, 2007 WL 1847385, at *27 (D.N.J. June 25,
2007) (“The Supreme Court in Daubert has stated that the focus of the inquiry
should be solely on principles and methodology, not on the conclusions that
they generate.” (citing Daubert, 509 U.S. at

596)).11

For example, plaintiffs’ statement that conjoint analysis often assigns values to
individual attributes which, if added together, would exceed the value of the product,
seems to be more of a restatement of the defendants’ criticism than an answer to it.
That consumers’ valuations of the product’s features add up to more than 100% may
be a fertile area for cross-examination or rebuttal. But the fact that not all consumers
do the necessary math does not mean that their expressed preferences have no
validity; I take it as a given that their preferences, ill-considered or not, are what we
are measuring when we do conjoint analysis. Thus these are not grounds for Daubert
exclusion, in my view.
10

For the proposition that courts will exclude a survey where the majority of data
has been excluded, defendants cite Brokerage Concepts, Inc. v. U.S. Healthcare, Inc.,
140 F.3d 494, 517 n.14 (3d Cir. 1998). There, however, the Third Circuit merely
discussed in dicta why it believed the district court had been wrong to admit a survey
under Fed. R. Evid. 803(24) (now Fed. R. Evid. 807(a)), which permits hearsay where
there are “circumstantial guarantees of trustworthiness” based on “accordance with
11

19

The defendants remain free to argue that Dr. Sukumar manipulated his
data to produce a more plausible result, inappropriately excluded data from
respondents who “prefer higher prices to lower prices and [therefore] don’t
value the Energy Star logo at all,” or drew from a minority population not
representative of the putative class.’ The issue now, however, is not one of the
2
opinion’s ultimate persuasiveness, but of its admissibility. “The judge should
only exclude evidence if the flaw is large enough that the expert lacks good
grounds for his conclusion.” Jama, 2007 WL 1847385, at *27 (citation and
internal quotation marks omitted). See also Microsoft Corp. v. Motorola, Inc.,

904 F. Supp. 2d 1109, 1120 (W.D. Wash. 2012) (declining to exclude Dr.
Sukumar’s opinion based on claims that his survey was unreliable and used a
non-representative sample because “[tjechnical inadequacies in a survey,
including the format of the questions or the manner in which it was taken,
bear on the weight of the evidence, not its admissibility.” (quoting Fortune
Dynamic v. Victoria’s Secret Stores Brand Management, Inc., 618 F.3d 1025,
1038 (9th Cir.2010)); Hartle v. FirstEnergy Generation Coip., No. CIV.A. 08—

generally accepted.
principles.” This standard is not identical to the Daubert
standard. Regardless, the Brokerage Concepts court explained that the challenged
survey in that case involved calling and posing questions to just 20 people, used
improper framing of questions, and excluded data without any statistical basis for
doing so. The entire survey in that case suffered from significantly more egregious
flaws than those the defendants claim here.
.

.

Defendants cite cases finding surveys inadmissible, but those surveys had
significantly greater, and less debatable, flaws. See Pittsburgh Press Club v. United
States, 579 F.2d 751, 759 (3d Cir. 1978) (survey excluded as hearsay because it “was
not scientifically designed,” respondents were all interested in the litigation and “told
the precise nature of the litigation and the purpose of the survey,” and therefore the
survey “suffer[ed] from a severe dearth of any circumstantial guarantees of
trustworthiness.”); J & J Snack Foods, Corp. v. Earthgrains Co., 220 F. Supp. 2d 358,
372 (D.N.J. 2002) (survey to determine whether trademark was descriptive excluded,
in part, because it was directed to consumers at a shopping mall even though the
product was sold only to commercial distributors in the market); Menasha Corp. v.
News Am. Mktg. In-Store, Inc., 238 F. Supp. 2d 1024, 1030 (N.D. 111. 2003) (survey held
unreliable where, among other flaws, the “universe’ of potential respondents consisted
entirely of individuals and companies that [the expert] knew through his work as a
journalist,” and the expert “did not make any attempt to select a sample to
approximate the relevant characteristics of the target population.”).
12

20

1019, 2014 WL 1317702, at *6 (W.D. Pa. Mar. 31, 2014) (excluding a
“fundamentally flawed” survey that “would not give the jury a proper basis for
determining damages,” but declining to exclude a survey suffering only from
“errors of a technical nature that go to the weight of the survey”).
Dr. Sukumar proffers what he plausibly characterizes as “good grounds”
for his population and sample selection.’ (Sukumar Opp. 18—22; Sukumar
3
Rebuttal Report 7—8, 12—15) To the extent defendants do not accept his
contentions, I direct them to “the traditional and appropriate means of
attacking shaky but admissible evidence”: “Vigorous cross-examination,
presentation of contrary evidence, and careful instruction on the burden of
proof.” Daubert, 509 U.S. at 596. If Dr. Sukumar could be shown to have
selected a misleading sample, to have achieved results that add up to more
than 100%, or to have implied a counterintuitively low price for a non-Energy
Star washer, that might prove harmful to the plaintiffs’ case. Such issues,
however, are for the finder of fact.

3) Whether Dr. Sukumar’s Price Premium Analysis Fits the Facts of the
Case and Plaintiffs’ Theory of Liability
The final error the defendants assign to Dr. Sukumar’s opinions is that
his results do not “fit” the facts of the case or the putative class’s particular
theory of liability. Therefore, the defendants urge, Dr. Sukumar’s opinions are
inadmissible under Daubert and Comcast Corp. v. Behrend, 133 S. Ct. 1426,
1433, 185 L. Ed. 2d 515 (2013). (Sukumar Br. 25—28)
Specifically, the defendants argue that Dr. Sukumar’s price premium
only measures respondents’ subjective “willingness to pay” for the Energy Star
logo, and not “what retailers actually charged putative class members in 2009

See In re Paoli R.R. Yard PCB Litig., 35 F.3d 717, 744 (3d Cir. 1994) (“The
grounds for the expert’s opinion merely have to be good, they do not have to be
perfect. The judge might think that there are good grounds for an expert’s conclusion
even if the judge thinks that there are better grounds for some alternative conclusion,
and even if the judge thinks that a scientist’s methodology has some flaws such that if
they had been corrected, the scientist would have reached a different result.”).
‘3

21

and 2010 for the logo on the Washers in this case.” (Sukumar Br. 26) Framed
differently, the defendants’ argument is that the plaintiffs’ theory of liability
properly rests on “the amount that Defendants actually charged purchasers
when they bought their Washers in 2009 and 2010, while Dr. Sukumar’s ‘price
premium’ focuses on the ‘value’ that respondents placed on the logo in 2015.”
(Sukumar Reply 9) Those values, say the defendants, are not at all the same
thing.
The defendants also protest Dr. Sukumar’s failure to subtract from his
price premium any offsetting benefits, such as rebates or utility savings (even if
those benefits were not as great as a fully Energy Star-qualified washing
machine would have yielded). (Sukumar Br. 29) The defendants point to
deposition testimony in which Dr. Sukumar admitted that the Energy Star logo
is a reflection of efficiency. If Dr. Sukumar’s price premium encompasses both
the value of the logo and the value of the utility savings from greater efficiency,
the defendants argue, then he should have offset those efficiency-based
benefits. Citing In re ConAgra Foods, Inc., 302 F.R.D. 537, 579 (C.D. Cal. 2014),
and In re POM Wonderful LLC, No. ML 10-02 199 DDP (RZx), 2014 WL 1225184,
at *5 & n.7 (C.D. Cal. Mar. 25, 2014), the defendants conclude that the “price

premium” is not merely what the Energy Star logo promised, but rather what
the Energy Star logo promised and did not deliver. Otherwise, they say, the
price premium as calculated by Dr. Sukumar would not meet Daubed’s “fit”
requirement. (Sukumar Reply 11—12)
The plaintiffs respond, first, that Dr. Sukumar’s opinion meets
defendants’ objection by incorporating “supply side” factors (here, retail market
prices) into his analysis. Those supply-side factors include “actual sales data
and advertising circulars from retailers who sold the Maytag Centennial clothes
washing machines to determine the price paid by actual purchasers.”
(Sukumar Opp. 23 (quoting Sukumar Report at 6)). Thus, say plaintiffs,
Sukumar did consider what retailers actually charged putative class members.
(Id.)

22

Second, the plaintiffs respond that Dr. Sukumar did not need to subtract
benefits received from his price premium, because such benefits were baked
into the survey questions. Dr. Sukumar’s survey asked respondents to
compare Maytag washing machines with and without an Energy Star logo, with
all other factors held constant. (Id. 25) Thus the survey was designed to isolate
the value consumers attached solely to the Energy Star logo. (Id.)
The plaintiffs’ explanations are not entirely persuasive, but there is a
more fundamental issue. The parties are talking past each other. Essentially,
the plaintiffs are talking about Daubert “fit”; the defendants, about Comcast
“fit.” At this early stage of the case, Comcast “fit” is a premature consideration.
a) Comcast versus Daubed

In Comcast, the Supreme Court held that a putative class failed to satisfy
Rule 23(b)(3)’s predominance requirement, which requires that the putative
class’s damages can be calculated on a class-wide basis. The Court found that
the model proposed by plaintiffs’ expert “failed to measure damages resulting
from the particular antitrust injury on which petitioners’ liability in this action
is premised.” Comcast Corp., 133 5. Ct. at 1433. That damages model was built
on the expert’s assumption that the defendants’ antitrust violations had
distorted the market and impacted prices in four different ways. By the time
the trial court ruled on certification, however, only one theory of antitrust
impact had survived. Id. at 1434. The district court set that aside: it “saw no
need for respondents to ‘tie each theory of antitrust impact to a calculation of
damages,” id. at 1433 (internal quotation marks omitted). The Court of
Appeals, affirming, “concluded that respondents ‘provided a method to measure
and quantify damages on a classwide basis,’ finding it unnecessary to decide
‘whether the methodology [was] a just and reasonable inference or
speculative.”’ Id.
The Supreme Court reversed, explaining that “[t]he first step in a
damages study is the translation of the legal theory of the harmful event into an
analysis of the economic impact of that event.” Id. at 1435 (quoting Federal
23

Judicial Center, Reference Manual on Scientific Evidence 432 (3d ed. 2011)
(emphasis added)). The Supreme Court held that the District Court and the
Court of Appeals had bypassed that mandatory first step. Id.
Daubert’s “fit” test for admissibility, however, is not so stringent, and it
precedes the “first step” [sic] discussed in Comcast. Daubed instructs that for
expert testimony to “fit” the case, it simply must bear enough of a relation to
the facts and issues of the case to be of aid to the jury in resolving a dispute of
fact. Daubed v. Merrell Dow Pharm., Inc., 509 U.S. 579, 591, 113 S. Ct. 2786,
2795—96 (1993). As the Third Circuit has explained:
In assessing whether an expert’s proposed testimony
“fits,” we are asking “ ‘whether [the] expert testimony
is sufficiently tied to the facts of the case
proffered
that it will aid the jury in resolving a factual dispute.’”
Id. (quoting United States v. Downing, 753 F.2d 1224,
1242 (3d Cir.1985)). Put another way, this is a
question of relevance, and “Rule 702, which governs
the admissibility of expert testimony, has a liberal
policy of admissibility” if it has the “potential for
assisting the trier of fact.” Kannankeril v. Terminix Int’l,
Inc., 128 F.3d 802, 806 (3d Cir.1997) (citing Holbrook
v. Lykes Bros. S.S. Co., 80 F.3d 777, 780 (3d
Cir. 1996)); see also In re TMILitig., 193 F.3d 613, 670
(3d Cir. 1999) (“expert evidence which does not relate
to an issue in the case is not helpful”). The “standard
is not that high,” but “is higher than bare relevance.”
In re Paoli R.R. Yard PCB Litig., 35 F.3d 717, 745 (3d
Cir. 1994).
...

United States v. Schiff 602 F.3d 152, 173 (3d Cir. 2010).
The defendants would have me conclude that Comcast adds a layer of
complexity to the Daubed admissibility determination. 14 Neither the Third
Circuit nor courts within this district have read Comcast that way. Rather, they
have distinguished the manner in which the court admits opinion evidence
under Daubed and the manner in which the court weighs such evidence when
See Sukumar Br. 26-27 & n.20 (arguing that Dr. Sukumar’s price premium
cannot establish “an absolute valuation to be awarded as damages” because it does
not incorporate market supply and that this argument is “relevant to both the Rule 23
inquiry and to whether Dr. Sukumar’s testimony meets Daubert’s “fit” requirement).
24

assessing predominance under Comcast. See, e.g., In re Urethane Antitrust
Litig., 166 F. Supp. 3d 501, 510 (D.N.J. 2016) (“Comcast is of dubious
relevance because it involves predominance issues arising in the Rule 23
context, whereas the instant matter concerns whether expert testimony is
reliable and relevant under Rule 702.”).15
In In re Blood Reagents Antitrust Litigation, the Third Circuit joined other
Courts of Appeals in holding that “a plaintiff cannot rely on challenged expert
testimony, when critical to class certification, to demonstrate conformity with
Rule 23 unless the plaintiff also demonstrates, and the trial court finds, that
the expert testimony satisfies the standard set out in Daubert.” 783 F.3d 183,
187 (3d Cir. 2015). That language necessarily implies that the Daubert issue is
prior, and that class certification is separate and distinct. See id. at 184
(presenting as separate issues on appeal (1) whether Rule 23 requires that
challenged expert testimony be scrutinized under Daubert and (2) whether
class certification was proper in light of Comcast). Of course, if the expert
opinion is inadmissible under Daubert, it will not be considered in the Comcast
analysis. See id. at 187 (“Expert testimony that is insufficiently reliable to
satisfy the Daubert standard cannot ‘prove’ that the Rule 23(a) prerequisites
have been met ‘in fact,’ nor can it establish ‘through evidentiary proof’ that
Rule 23(b) is satisfied.”).’ The clear takeaway from In re Blood Reagents is that
6

In fact, The Third Circuit has explained that Comcast turned on “a
straightforward application of class-certification principles,” Neale v. Volvo Cars of N.
Am., LLC, 794 F.3d 353, 374 (3d Cir. 2015) (quoting Comcast, 133 S.Ct. at 1433), and
does not change the “well nigh universal” recognition “that individual damages
calculations do not preclude class certification under Rule 23(b)(3).” Id. (quoting
Corncast, 133 S.Ct. at 1437 (Ginsburg, J. & Breyer, J., dissenting)) (collecting Court of
Appeals cases reading Comcast similarly). See also City of Sterling Heights Gen.
Employees’ Ret. Sys. v. Prudential Fin., Inc., No. CIV.A. 12-5275, 2015 WL 5097883, at
*7 (D.N.J. Aug. 31, 2015) (“[C]lass treatment would
still be appropriate here even if
damages were required to be calculated on an individual basis.”). The Third Circuit
also explicitly cautioned that Comcast’s “predominance analysis was specific to the
antitrust claim at issue.” Neale, 794 F.3d at 374.
Indeed, In re Blood Reagents remanded for the trial court to consider the
Supreme Court’s then-recent decision in Comcast.
16

25

Comcast presents questions for the class certification stage that are distinct
from, and logically posterior to, the Daubert analysis.
b) Defendants’ reliance on Comcast is premature, if not entirely
misplaced
On this motion, I am still at the Daubert gatekeeping step. I am deciding
what opinion evidence is admissible for purposes of, inter alia, the class
certification inquiry. It follows that questions of “fit” under Comcast are
premature. Now, I must consider the reliability of an expert’s opinion or
whether it sufficiently relates to the facts and issues of the case (Daubed “fit”).
Only later do I consider whether the expert’s damages model comports with the
plaintiffs’ theory of liability, or with the type of damages legally available (i.e.,
Comcast “fit,” assuming it applies at all outside of the antitrust context).
That two-step approach is illustrated by the first of the two cases upon
which defendants chiefly rely, In re ConAgra Foods, Inc., 90 F. Supp. 3d 919
(C.D. Cal. 2015), aff’d, 844 F.3d 1121 (9th Cir. 2017), and No. 15-55727, 2017
WL 53421 (9th Cir. Jan. 3, 2017). There, the United States District Court for
the Central District of California considered a price premium damages
methodology proposed by the plaintiffs’ expert (Cohn B. Weir, as it happens).
The court had little trouble finding Mr. Weir’s methodology admissible under
Daubed: “Admissibility turns on whether Weir’s methodology is sufficiently
reliable,” the court wrote, and it found that the price premium methodology
passed that test. Id. at 946. But, the court cautioned, “whether [the price
premium methodology] satisfies Comcast and shows that a class should be
certified is another question altogether—one which the court will address infra

See also City of Sterling Heights Gen. Employees’ Ret. Sys. v. Pnidential Fin.,

Inc., No. CIV.A. 12-5275, 2015 WL 5097883, at *3 (D.N.J. Aug. 31, 2015) (“Courts are
also frequently called upon to consider expert opinion offered to support or oppose
class certification. Where an expert opinion is critical to class certification and a party
challenges the reliability of that opinion, the reviewing court must engage in a twostep analysis before analyzing whether Rule 23’s requirements have been met: (1)
whether the party’s challenges bear upon ‘those aspects of [the] expert testimony
offered to satisfy Rule 23’ and (2) if so, whether the opinion is admissible as to those
aspects under Federal Rule of Evidence 702 and Daubed.” (citations omitted)).
26

in conducting a Rule 23(b)(3) predominance analysis.” Id. Later in its opinion,
the court performed that Comcast “fit” analysis. It held that Weir’s methodology
was not sufficient, on its own, to define classwide damages under Rule 23(b)(3)
because it did not “isolate[] and quantif[y damages associated with plaintiffs’
specific theory of liability.” 90 F. Supp. 3d at 1024. What ConAgra did not hold
is that an expert opinion should be held to flunk Daubert because it flunks
Comcast—quite the opposite, in fact.

The second of the two cases upon which defendants chiefly rely is In re
POM Wonderful LLC, No. ML 10-02199 DDPRZX, 2014 WL 1225184, at*5
(C.D. Cal. Mar. 25, 2014). There, the court applied Comcast to find that an
expert’s price premium model could not satisfy Rule 23(b)(3), explaining:
“Rather than draw any link between Porn’s actions and the price difference
between the four-juice average benchmark price and average Porn prices, the
Price Premium model simply calculates what the price difference was. This
damages ‘model’ does not comport with Comcast’s requirement that class-wide
damages be tied to a legal theory.

.

.

.“

Again, however, the essence of the

holding is not a Daubert admissibility analysis, but a Comcast “fit” analysis.’
7

It must be said, however, that POM Wonderful did not so clearly distinguish
between the standards and purposes of Daubert and Comcast. It analyzed and rejected
class certification, but also stated, in a footnote: “For these and related reasons, [the
expert’sl report and testimony are not admissible under Daubert.
2014 WL
1225184, at *6 (emphasis added). The opinion rested on the complete disconnect
between the proffered damages model and the facts of the case. See id. at *5 (noting
that the damages expert presented no “survey or other evidence of what consumers’
behavior might otherwise have been,” and “made no attempt, let alone an attempt
based upon a sound methodology, to explain how Defendant’s alleged
misrepresentations caused any amount of damages.”). In such a clear case, it was
perhaps less critical to distinguish between the two, accounting for the throwaway
Daubed analysis in a footnote.
17

..

.“

See also In re NJOY, Inc. Consumer Class Action Litig., 120 F. Supp. 3d 1050,
1119—1122 (C.D. Cal. 2015) (on class certification motion, expert’s analysis did not
satisfy Comcast and Rule 23(b)(3) because they “provide[d] only a model for testing
what a consumer is willing to pay, without considering other factors in a functioning
marketplace,” whereas only restitution damages (the difference between the product as
labeled and the product as received) were recoverable under statute); In re NJOY, Inc.
Consumer Class Action Litig., No. CV 14-428-JFW (JEMX), 2016 WL 787415, at *9
(C.D. Cal. Feb. 2, 2016) (declining to strike, under Daubed, plaintiffs’ expert’s opinions

27

Here, whether the plaintiffs’ proposed damages model and its underlying
methodologies (such as Dr. Sukumar’s price premium) fit the plaintiffs’ theory
of liability will depend on as-yet-undetermined issues. Among these are likely
to be a determination of the type of damages recoverable for breach of express
warranty and the implied warranty of merchantability, unjust enrichment, and
violation of consumer protection statutes under the laws of various states. For
these and other reasons, a final Comcast determination of “fit” for purposes of
class certification would be premature. What I have decided here is that Dr.
Sukumar’s opinion is admissible under Daubed and therefore will go into the
mix when I do decide the class certification issues.
To summarize, Rule 702 permits a wide range of testimony as long as the
expert is qualified and the testimony is reliable and relevant. Dr. Sukumar’s
opinion testimony, and its supporting methodologies, are sufficiently reliable
and relevant to be admitted into evidence for consideration at the next stage of
this case—a motion for class certification under Rule 23. Thereafter, the
defendants will have the opportunity to address their concerns with Dr.
Sukumar’s opinions through traditional methods of presentation of their case,
including cross-examination and rebuttal evidence. Accordingly, the
defendants’ motion to exclude Dr. Sukumar’s opinions will be denied.
C.

Defendants’ Motion to Strike the Opinions of Dr. Dennis

The defendants argue that the plaintiffs’ other price premium expert, Dr.
Dennis, also used methods that flunk the Daubed test. The defendants raise
three broad arguments and several sub-arguments as to why I should strike
Dr. Dennis’s opinions for lack of “fit.” I first describe Dr. Dennis’s method of
analysis, and then address defendants’ arguments.
on the basis of alleged flaws in conjoint analysis and regression analysis, but
concluding, under Comcast that Rule 23(b)(3) could not be satisfied because “Plaintiffs
have not proffered a model capable of calculating damages on a classwide basis.
Apple, Inc. v. SamsungElecs. Co., No. 11—CV-01846-LHK, 2014 WL 976898, at *13
(N.D. Cal. Mar. 6, 2014) (“Here, of course, the Court is not addressing a Daubert
challenge.”).
.

28

.

1) Dr. Dennis’s Analysis
Dr. Dennis surveyed 1,122 people who had purchased top-loading
washing machines in the United States since 2009. (Dennis Opp. 1; see
generally Dennis Report.)’ The method employed by Dr. Dennis is called
8
contingent valuation, an “approach based on the direct elicitation of

[1

values

from individuals through the use of carefully designed and administered
sample surveys.” (Id. (quoting Kenneth Arrow, et al., Report of the NOAA Panel
on Contingent Valuation (Jan. 11, 1993) (“NOAA Report”), Deckant Dccl. Ex. I,
ECF No. 215-9)). Dr. Dennis’s surveys purportedly measured “respondents’
valuations of products that are identical in all respects except for the presence
of absence of the Energy Star® logo,” and “controlled for all other variables to
isolate the price premium solely attributable to the Energy Star® logo.” (Dennis
Opp. 1) Dr. Dennis’s initial report discusses his contingent valuation and its
results. His initial report also summarizes the results of Whirlpool’s own
consumer research on the Energy Star logo. (See Dennis Report

¶J

44—49) Dr.

Dennis also submitted a rebuttal report defending his contingent valuation

The following abbreviations will be used to cite Dr. Dennis’s submissions and
the briefing on the defendants’ motion to strike his opinion testimony:
18

•

“Dennis Report” = Declaration and Expert Report of J. Michael Dennis, Ph.D,
Deckant Decl. Ex. F, ECF No. 2 15-6

•

“Dennis Rebuttal” = Rebuttal Expert Report of J. Michael Dennis, Ph.D.,
Deckant Dccl. P, ECF No. 2 15-16

•

“Dennis Br.” = Brief in Support of Motion to Strike the Opinions of Dr. J.
Michael Dennis, Ph.D. by Defendants Whirlpool Corporation, Lowe’s Home
Centers, LLC, Sears Holdings Corporation, and Fry’s Electronics, Inc., ECF No.
199

•

“Dennis Opp.” = Plaintiffs’ Opposition to Motion to Strike the Opinions of Dr. J.
Michael Dennis, Ph.D. by Defendants Whirlpool Corporation, Lowe’s Home
Centers, LLC, Sears Holdings Corporation, and Fry’s Electronics, Inc., ECF No.
213

•

“Dennis Reply” = Reply in Support of Motion to Strike the Opinions of Dr. J.
Michael Dennis, Ph.D. by Defendants Whirlpool Corporation, Lowe’s Home
Centers, LLC, Sears Holdings Corporation, and Fry’s Electronics, Inc., ECF No.
229
29

methodology against the criticisms of defense expert Dr. Carol A. Scott. (See
Dennis Rebuttal.)
The defendants do not dispute that Dr. Dennis is appropriately qualified
and credentialed. He is currently a Senior Vice President of the National
Opinion Research Center at the University of Chicago. He has “been personally
involved in the design and conduct of over a hundred statistical surveys using
the internet mode of data collection,” including for several United States federal
agencies and in at least six prior litigations. (Dennis Opp. 3—4 (citing Dennis
Report

¶J 4, 6—7))

2) Objection to selection of contingent valuation method
First, the defendants argue that Dr. Dennis’s opinions should be
excluded because he should not have opted for this contingent valuation
methodology at all. Contingent valuation, say the defendants, is a second—best
method that should be used only when objective indicia of market value, such
as prices actually paid for “market goods,” are not available. Dr. Dennis
himself, they say, generally agrees with this principle. (Dennis Reply 3)
Here, the defendants say, objective market data are available: The “C500”
model Maytag washing machine (which retailed at a higher price, on average,
than the Washers) was a comparable washing machine that did not bear the
Energy Star logo. (Dennis Br. 10—12; Dennis Reply 6). Therefore, in lieu of
theorizing, Dr. Dennis should have simply analyzed the comparable C500 sales
data, which show what consumers actually paid during the relevant time
period for a similar washing machine without the Energy Star label.
The plaintiffs respond that the C500 machine is not comparable. In their
brief, they say that there are many differences between the Washers and the
C500, and that those differences would confound the price premium analysis.
(Dennis Opp. 10) All very well, say defendants, but Dr. Dennis did not himself
perform such an analysis, did not determine whether the non-Energy Star
washer was a comparable “market good,” and therefore did not choose the

30

contingent valuation methodology “based on his reasoned determination that
the [C500] was not sufficiently similar.” (Reply 5)
I recognize the origin and purposes of the contingent valuation method. It
was developed as a means of assigning a dollar value to what are obviously
non-market goods, i.e., things that are not bought and sold, such as the
preservation of a wilderness area. In lieu of consumers’ “revealed” preferences
(i.e., what they actually pay for a market good), contingent valuation relies on
consumers’ “stated” preferences (i.e., what they say a good is worth to them).
Saying is not paying, and that suggests some of the possible shortcomings of
the stated-preference approach. It costs a survey respondent nothing to say $0
or $1 billion, to ignore tradeoffs or income constraints, to register a protest
(“bankruptcy is too good for any company that put the wrong label on my
purchase”), and so on. The potential pitfalls, and techniques for minimizing or
avoiding them, are discussed more formally in the seminal NOAA Report,
Sections II & III. (A copy is attached to defendants’ papers. (ECF no. 215-9))
My task here, however, is not to identify the best methodology and
exclude all others, but rather to determine whether contingent valuation is a
reasonable, reliable methodology. See Lentz v. Mason, 32 F. Supp. 2d 733, 746
(D.N.J. 1999). The facts and case law before me establish that it is.
Dr. Dennis’s report bespeaks a thorough familiarity with the contingent
valuation method and its limitations. Asked if contingent valuation would be an
appropriate approach even if he knew that the C500 model was comparable,
Dr. Dennis opined that the “contingent valuation approach would still be a
reliable tool

.

.

.

.because

.

.

.

.

[it]

.

.

.

has a powerful ability to isolate the value

that consumers and purchasers place on particular attributes.

.

.

.“

(Dennis

Dep. 67:1268:3)19 Contingent valuation is a tool long used by the federal

Defendants dispute that the distinctions between the C500 and the Washers
would bear on the price comparison in any relevant way. (Id.) For the reasons
expressed in text, this is not critical to the Daubert analysis, but I do find the
defendants’ attempt to minimize the distinctions between the C500 and the Washers
to be unclear and undeveloped.
19

31

government. (See Dennis

Opp.

2—3) Courts have accepted it as a sufficiently

reliable basis to estimate the economic impact of alleged product defects or
labeling misrepresentations in consumer class actions. See, e.g., Miller v. Fuhu
Inc., No. 2:14-CV-06119-CAS-AS, 2015 WL7776794, at *21 (C.D. Cal. Dec. 1,
2015) (stating “numerous courts, including this one, have accepted both
[Choice-Based Conjoint Analysis] and [Contingent Valuation Method] as
reliable methodologies for calculating price premiums on a classwide basis in
consumer class actions,” and collecting cases); In re Toyota Motor Corp. Hybrid
Brake Mktg., Sales Practices & Prod. Liab. Litig., No. MDL 10-02 172-CJC, 2012
WL 4904412, at *4 (C.D. Cal. Sept. 20, 2012) (finding contingent valuation
admissible to estimate damages arising from an anti-lock braking system
defect, and explaining that the “proposed methods of analysis, i.e. hedonic
regression, contingent valuation, and discrete choice, are generally accepted,
have been tested, and are part of peer-reviewed studies.”) 20

For example, citing no case law or expert testimony, the defendants assert that
whether a washing machine features an “end of a cycle signal” is a “distinction[]
without a difference.” (Dennis Reply 5) They also argue that certain distinguishing
features the plaintiffs identify “would have resulted in the C500 costing less than the
labeled version of the Washer, not more, meaning that the market prices charged for
the two machines should have shown an even larger price premium than Dr. Dennis’s
methodology predicted.” (Dennis Reply 5) (emphasis in original) But as an example of
this argument, they state that “the larger capacity of the C500 is a desirable feature
that should have resulted in a higher price.” (Id.) (emphasis added).
20
The defendants argue that in the few cases in which courts have accepted
contingent valuation, the product in question contained an undisclosed defect. On
that basis, the proponent of contingent valuation argued that the subject of the survey
was actually a “nonmarket good” in the sense that no market existed in which
consumers had purchased the product without the defect. (Dennis Br. 12-13). The
plaintiffs’ argument here is more debatable; they must acknowledge that there is a
market for non-Energy Star washers, but they contend that none of them are truly
comparable to the Washers at issue. (Dennis Opp. 10—11) Defendants’
counterargument is based on the opinion of their own expert, Dr. Carol Scott. (See,
e.g., Expert Report of Carol A. Scott, Ph.D., April 26, 2016, Deckant Deci. Ex. L ¶{ 53,
71—74) The dispute raises a factual question going to weight, not admissibility. See
Hartle, 2014 WL 1317702, at *7 (W.D. Pa. Mar. 31, 2014) (“A factual disagreement
between experts is a matter for the jury to resolve.” (citing Lansford—Coaldale Joint
WaterAuth. v. Tonolli Corp., 4 F.3d 1209, 1216 (3d Cir. 1993)).
32

Granted, some courts have rejected contingent valuation studies where
they cannot be used to reliably measure damages, or where the study proposed
is too vague. In Miller v. Fuhu Inc., supra, for example, a contingent valuation
study (Dr. Dennis’s, as it happened) proposed for the purpose of quantifying
the economic impact of handheld tablets’ defective charging systems, was
excluded under both Daubert and Rule 23. There, the court found, Dr. Dennis
had done nothing but sketch out a plan for conducting a contingent valuation
survey to measure the hypothetical sale price of a product if it had not been
defective. 2015 WL 7776794, at *21. But the Miller court did not rule out a
properly conducted contingent valuation survey. Rather, it stated that, on a
renewed motion to certify, the “proposed methodology could, in theory, provide
a legally justifiable measurement for classwide damages,” if the plaintiffs
“provided more concrete details regarding [Dr. Dennis’s] proposed survey.” Id.
at 22.
Here, in contrast, Dr. Dennis is not merely proposing to do a vaguely
described survey. His opinions are based on a fully completed contingent
valuation survey and data analysis, both of which are described in detail in his
expert report. The concerns raised in Miller are not present in our case.
As with Dr. Sukumar, the defendants have stated a disagreement with
the results of Dr. Dennis’s analysis, and tried to work back from there to a
Daubert-based exclusion of his opinion. As I see it, their argument relies on
disputed facts and differences of opinion bearing primarily on the persuasive
weight a jury might assign to Dr. Dennis’s analysis. See, e.g., Leonard v.
Stemtech Int’l mc, 834 F.3d 376, 391 (3d Cir. 2016) appeal docketed, No. 16928 (U.S. Jan. 26, 2017) (“disagreement with the calculation methodology and
the underlying assumptions.

.

.

goes to the weight given to [an expert’s]

testimony, rather than admissibility.”); Breidor v. Sears, Roebuck & Co., 722
F.2d 1134, 1138—39 (3d Cir. 1983) (“Where there is a logical basis for an
expert’s opinion testimony, the credibility and weight of that testimony is to be
determined by the jury, not the trial judge.”); Lentz, 32 F. Supp. 2d at 746 (for

33

“testimony to be reliable, and, thus, admissible under Daubert, [the expert]
need not have used the best method available, only a reasonable one.
Moreover, the Defendants are free to challenge [the expert’sl methodology on
cross-examination and in their closing arguments.”).
Contingent valuation is a reasonable methodology. I decline to exclude
Dr. Dennis’s opinions on the basis of his having used it.
3) Criticisms of Survey Methodology, Population, and Results
The defendants also assign a number of errors to Dr. Dennis’s
application of the contingent valuation method, and to his analysis of data.
They aver that Dr. Dennis’s contingent valuation measures only consumers’
stated “willingness to pay” a premium for the Energy Star logo—a concept not
equivalent to what the putative class members actually paid. Relatedly, they
argue that Dr. Dennis failed to subtract offsetting actual benefits or to account
for supply-side factors that affected what putative class members actually paid
for their Washers. The defendants add that Dr. Dennis surveyed consumers in
2015, but the putative class members purchased their Washers under the
different economic and competitive conditions that prevailed in 2009 and 2010.
The plaintiffs deny that what Dr. Dennis has offered is merely a
“willingness-to-pay” calculation. Willingness to pay, they say, is a theoretical
concept untethered to marketplace data. A survey based on this concept might
simply consist of asking respondents, “How much are you willing to pay for X?”
Such an informal, qualitative, and unanchored survey might raise reliability
concerns. But here, the plaintiffs emphasize, Dr. Dennis’s survey presented
respondents with the actual prevailing $400 market price for the Washers, a
real-world price that necessarily incorporated supply and demand factors.
(Dennis

op.

7, 12—13) The plaintiffs add that Dr. Dennis “worked closely with

an expert economist” to ensure that his survey design took into account both
supply-side and demand-side factors. (Dennis Opp. 1)
The defendants reply, in effect, that Dennis is attempting the impossible:
a one-sided comparison. A valid contingent valuation, they say, would have to
34

“account for the impact of supply-side factors

.

.

.

.

on the price of the Washer

relative to the price of the unlabeled version of the Washer, and [Dr. Dennis]
[would] need[] an accurate measure of the price of both to identify any alleged
price premium.” (Dennis Reply 8) (emphasis in original). In reality, they say,
“intense market competition for similarly-priced Energy Star top-loaders drove
down prices for that category of machines and eliminated any potential price
premium for the Washers as compared to the unlabeled C500 version.” (Id.)
As for the timing issue, the plaintiffs point out that Dr. Dennis testified
that his 2015 survey is a reliable and valid measurement for washing machine
purchases for the entire 2005—20 15 time frame. The Energy Star label, he
stated, has had “such a remarkable hold on the consumer mind—and it’s been
that way for a significant period of time

.

.

.

.“

(Id. (quoting Dennis Dep.,

Deckant Decl. Ex. 0, 123:13—25)) This is thin stuff, say the defendants; there is
“too great an analytical gap” between Dr. Dennis’s 2015 survey results and the
conclusions he draws about 2009—20 10 purchases. (Id. (quoting General Elec.
Co. v. Joiner, 522 U.S. 136, 146, 118 S. Ct. 512, 519 (1997))).
For the reasons discussed Section III.A.3., supra, with respect to the
defendants’ similar criticisms of Dr. Sukumar’s analysis, I find that Dr.
Dennis’s methodology is reasonable and reliable overall. Defendants’ challenges
to the “willingness to pay” and survey population components go to weight, not
admissibility. The defendants’ criticisms are far from frivolous. Those
criticisms, however, do not so undermine the reliability of Dr. Dennis’s
methods or their application to the case as to warrant exclusion.
4) Whether Dr. Dennis’s Price Premium Analysis Fits the Facts of the Case
and Plaintiffs’ Theory of Liability
Finally, the defendants argue that Dr. Dennis’s opinions are misaligned
with the facts of the case and the plaintiffs’ theory of the case. (Dennis Br. 18)
First, the defendants attribute to Dr. Dennis the incorrect “binary”
assumption that the Washers either (a) were Energy Star qualified, and
therefore delivered a host of efficiency and rebate benefits, or else (b) were not
Energy Star qualified, and therefore delivered no benefits. In reality, the
35

defendants claim, the purchasers of the Washers received tax rebates and
incentives, and also enjoyed water and energy efficiencies which, even if they
did not meet Energy Star standards, nevertheless exceeded those of other, nonEnergy Star washers. (Id. 19—21) The defendants argue that Dr. Dennis failed
to account for these benefits when he treated the Energy Star logo in binary,
black-or-white fashion. His opinions therefore would not align with the
plaintiffs’ “real theory of liability,” which the defendants frame as follows:
“consumers who purchased a Washer in 2009 and 2010 paid a price premium
in exchange for improved water and energy efficiency.” (Dennis Reply 10—11
(emphasis in original))
The plaintiffs confirm that Dr. Dennis considers the Energy Star
“promise” to be a binary one. But this, they say, is not an error. Rather, they
explain, the promise is binary because “Plaintiffs’ theory of liability is not based
on the price premium for the relative efficiency of the washers at issue, but
rather the price premium solely attributable to the presence or absence of the
Energy Star® logo.” (Id. at 14) Plaintiffs’ damages expert, Mr. Weir, accounted
for energy expense losses in his “energy expense damages model.” (see
Declaration of Cohn B. Weir, Dec. 28, 2015, Deckarit Deci. Ex. C, ¶J 9—23))
Weir’s energy expense model, however, is distinct from his “price premium
model,” which comes at the same issue from the perspective of consumer
preferences, and is based on Dr. Dennis’s results. (Id. at 15) Dr. Dennis did not
need to consider rebates, the plaintiffs say, because “price premium damages
occurred at the time of purchase and rebates, if any, were only paid out
later.” (Id.) Even if he did need to account for rebates, they say, Dr. Dennis has
stated that his survey respondents surely “were thinking about [publicly
known] rebate programs” (Id. (quoting Dennis Dep., Deckant Deci. Ex. 0,
165:18—21)) when they responded. Therefore, the impact of rebates was
“prebaked into the survey results.” (Id.) Moreover, the plaintiffs explain, the
defendants’ own expert did not adjust for rebates. (Id.)

36

This first debate revolves primarily around the Comcast question of
whether Dr. Dennis’s model fits the plaintiffs’ theory of the case. For the
reasons stated above, consideration of that Comcast issue is premature at this,
the Daubert stage. The defendants also state that “the fact that Dr. Dennis
failed to account for the substantial evidence showing that most, if not all, of
the benefits of Energy Star were delivered

[1

renders his opinions inadmissible

under Rule 702.” (Dennis Br. 20) But this alleged deficiency depends on
questions of fact that go to weight of the evidence (and also directly to the
merits, it would seem), not its admissibility. These flaws, if that is what they
are, does not rise to the level of requiring me to exclude Dr. Dennis’s opinion
from evidence.
I note, by the way, that the plaintiffs’ endorsement of the “binary”
approach here is seemingly at odds with their position in relation to alleged
“double counting” in Mr. Weir’s opinion, discussed in Section II.D.4, infra.
There, they wriggle free of defendants’ criticism by acknowledging that the
value of the Energy Star logo and the value of the promised energy savings may
well overlap. Here, however, they place consumer preferences in a black box
and accept that consumers want what they want. That, too, however, is an
issue that may be refined in the course of class certification or the calculation
of damages, and may ultimately be placed before the jury for decision.
Second, the defendants claim that Dr. Dennis’s contingent valuation
model cannot reliably be applied in this case because “real world” market data
shows that non-Energy Star washing machines sold for just as much as Energy
Star washing machines. On that view, Dr. Dennis’s price premium would fail to
correspond to market reality.
The plaintiffs respond that defendants’ preferred “actual market data”
approach, whatever its validity, actually proves plaintiffs’ case. Data from the
time that the Washers at issue and the non-Energy Star C500 model were both
were on the market, they say, shows that the C500 washers were 48% cheaper
than the C6 Washers. (Id. 16 (citing Rebuttal Declaration of Cohn B. Weir,

37

Deckant Deci. Ex. E,

¶

8)) That 48% price differential, they say, is roughly

consistent with Dr. Dennis’ 48.5% price premium and with Dr. Sukumar’s
44.3% price premium (see Sukumar Report ¶ 14—17), as well as the view
expressed by “Whirlpool’s own executives.” (Id.)
Third, the defendants say Dr. Dennis’s contingent valuation study does
not fit the facts of the case because his survey focused on retail prices and
retail consumers, ignoring the fact that the C7ES model of the Washers was
sold primarily to rental companies. (Dennis Br. 22) The plaintiffs dismiss this
third argument as lacking evidentiary or case law support. (Dennis Opp. 16—
17).
The defendants’ second and third points do not raise significant Daubert
problems. They merely take issue with Dr. Dennis’s interpretation of facts, with
specific methodological choices he made in his survey design, and with his
conclusions. As noted above, the focus of Rule 702 “must be solely on
principles and methodology, not on the conclusions that they generate,”
Daubert, 509 U.S. at 595.
In sum, the plaintiffs have demonstrated that Dr. Dennis employed
reasonable methodologies and that his conclusions rest on valid grounds. As
before, defendants have identified significant vulnerabilities, which they no
doubt will exploit at trial. But viewed as a question of Daubed admissibility,
defendants’ motion to strike Dr. Dennis’s opinions must be denied.
D.

The Defendants’ Motion to Exclude the Opinions of Mr. Weir
Finally, the defendants move, pursuant to Daubed and Federal Rules of

Evidence 702 and 703, to strike the opinions of the plaintiffs’ class-wide
damages expert, Cohn B. Weir. As discussed below, the defendants contend
that Mr. Weir’s model is generally unreliable; is based on unreliable data, false
assumptions, and miscalculations; and is inconsistent with the facts of the
case and with the plaintiffs’ theory of liability.

38

1) Mr. Weir’s Damages Model
Mr. Weir holds an MBA from Northeastern University, a Bachelor of Arts
in Business Economics from The College of Wooster, has testified in numerous
cases and government hearings on damages calculations, and has co-authored
several articles in the field of economics. Although the defendants describe Mr.
Weir as “a professional plaintiffs’ expert who invariably opines that damages
are capable of measurement on a classwide basis,” they do not formally
challenge his qualifications as an expert opinion witness. I find that Mr. Weir is
sufficiently qualified; it appears that his education and experience have given
him a specialized knowledge in damages calculations that may help a jury
understand the evidence or make a determination. See Fed. R. Evid. 702.
Mr. Weir proposes a hybrid damages model addressing two categories of
damages: (1) “price premium” damages representing the portion of the retail
price attributable solely to the Energy Star label; and (2) “energy expense”
damages stemming from utility bills the putative class members paid, which
were allegedly higher than they would have been if the Washers truly had met
Energy Star standards. (See generally Weir Report)
’
2
For the price premium component of his damages model, Mr. Weir relies
(a) on the price premiums calculated by Drs. Sukumar and Dennis, and (b) on
21
The following abbreviations will be used to cite Mr. Weir’s submissions and the
briefing on the defendants’ motion to strike his opinion testimony:

•

“Weir Report” = Declaration of Cohn B. Weir, Dec. 28, 2015, Deckant Decl.
Ex. C, ECF No. 2 15-3

•

“Weir Rebuttal” = Rebuttal Declaration of Colin B. Weir, June 10, 2016,
Deckant Deci. E, ECF No. 2 15-5

•

“Weir Br.” = Brief in Support of Motion to Strike the Opinions of Mr. Cohn Weir,
Ph.D by Defendants Whirlpool Corporation, Lowe’s Home Centers, LLC, Sears
Holdings Corporation, and Fry’s Electronics, Inc., ECF No. 200

•

“Weir Opp.” = Plaintiffs’ Opposition to Motion to Strike the Opinions of Mr. Cohn
Weir by Defendants Whirlpool Corporation, Lowe’s Home Centers, LLC, Sears
Holdings Corporation, and Fry’s Electronics, Inc., ECF No. 212

•

“Weir Reply” = Reply Brief in Support of Motion to Strike the Opinions of Mr.
Cohn Weir by Defendants Whirlpool Corporation, Lowe’s Home Centers, LLC,
Sears Holdings Corporation, and Fry’s Electronics, Inc., ECF No. 230
39

a 2006 Whirlpool document comparing sales data between Energy Star certified
and non-certified washing machines. (Id.

¶J

32—33)

For the energy expense component of his damages model, Mr. Weir
compares the energy and water consumption allegedly promised by Energy
Star-qualified washing machines to the results of laboratory audit tests that
the DOE performed on the Washers in 2009 and 2010. (Weir Br. 4—6; see Weir
Report

¶J

9—15)

2) Mr. Weir’s Price Premium Analysis
a) Mr. Weir’s conception of price premium damages
The defendants first argue that Mr. Weir’s calculations of price premium
damages “conflict[] with Plaintiffs’ allegations and do not fit the evidence in this
case.” (Weir Br. 7) In his deposition, Mr. Weir testified that consumers paid a
55.7% price premium for the Energy Star logo on the Washers, but declined to
say “what consumers would have paid had the logo not been there.” (Weir Br. 8
(quoting Weir Dep. 193:13—194:6, 195:9—18)). That, from defendants’ point of
view, constitutes an admission that Weir’s damages model is not “tied to the
presence or the absence of the Energy Star logo on the Washers,” and therefore
“is not designed to measure damages that flow from Plaintiffs’ theory of liability
in the case.” (Weir Br. 9).
The plaintiffs answer, and I agree, that the defendants’ argument is
based on a selective quotation of Mr. Weir’s deposition testimony and on
22
distinguishable or otherwise inapt Rule 23-focused case law. (See Weir Br. 9;
While precedent is of course relevant, I do not agree that rejection of an
particular expert’s opinion in a prior case forever taints that expert’s future opinions.
Each opinion must be measured against the requirements of Rules 702 and 703.
22

The defendants cite In re ConAgra Foods Inc., 302 F.R.D. 537, 552 (C.D. Cal.
2014), where a court struck Mr. Weir’s declaration for failure to satisfy Rule 702.
There, Mr. Weir testified that classwide “price premium” damages could be calculated
for consumers of mislabeled cooking oil provided necessary data is obtained. He
generally described the method of analysis he proposed using in his expert reports and
testified that the necessary data could be easily obtained from business records or
market research data, but did not identify any variables he intended to build into his
models or identify data in his possession that could be applied to models. Therefore,
40

Weir

Opp.

9—14). Having reviewed the deposition, I am not convinced overall

that there is no “connection between [Dr. Weir’s]

.

.

.

test result to be presented

and particular disputed factual issues in the case.” In re Paoli R.R. Yard PCB
Litig., 35 F.3d at 743. To the contrary, Mr. Weir’s price premium damages
analysis aims to “calculate the portion of the retail price attributable to the
Energy Star label, and/or to estimate the portion of the revenues and profits
gleaned by Whirlpool,” which fits several of the plaintiffs’ theories of harm.
(Weir Report

¶J

6, 24; see also, e.g., Compi. ¶160 (alleging plaintiffs were

harmed by defendants’ breach of the implied warranty of merchantability by
paying a price premium for the Energy Star logo); id.

¶

167 (alleging defendants

were unjustly enriched and plaintiffs were injured by paying a price premium
for the logo)) That is a close enough “fit” to satisIr Daubert.
To the extent the defendants seek to challenge the damages component
of Mr. Weir’s model as inconsistent with the types of damages legally available
by statute, they raise Comcast-style Rule 23 arguments, to be decided at the
class certification stage.

the court concluded that Mr. Weir had provided “no damages model at all,” and thus
his declaration was “so incomplete as to be inadmissible as irrelevant.” 302 F.R.D. at
552 (internal quotation marks omitted). This is not the case here, where Mr. Weir
possesses what he considers the necessary data and uses that data to calculate actual
proposed damages figures.
The plaintiffs point to a case in which Mr. Weir’s opinions fared better—Dei
Rossi v. Whirlpool, 2015 WL 1932484 (E.D. Cal. Apr. 28, 2015) (see Weir Opp. 1—3)—
but it, too, is distinguishable. Dei Rossi is another consumer class action based on
allegations of mislabeling. There, the court granted class certification and denied
Whirlpool’s motion to strike a declaration by Mr. Weir that calculated proposed price
premium and energy expense damages. Id. at *1 n. 1; (Weir Opp. 1) The court
concluded that Mr. Weir’s proposed price premium and energy savings methods were
sufficient at the class certification stage of the case because even after Comcast, Ninth
Circuit precedent directs that “plaintiffs need only propose a valid method for
calculating class wide damages, not an actual calculation of damages.” Dei Rossi,
2015 WL 1932484 at *1 n.1 (citing Leyva v. Medline Indus., Inc., 716 F.3d 510, 514
(9th Cir. 2013)). But the Dei Rossi court did not evaluate Mr. Weir’s methodology for
admissibility under Rule 702 and Daubert.
41

b) Hamish v. Widener
As to Mr. Weir’s price premium opinions, both parties cite the Third
Circuit’s decision in Hamish v. Widener University School of Law, 833 F.3d 298
(3d Cir. 2016). There, law school graduates claimed they had been induced to
pay inflated tuition based on the school’s overstatement of post-graduation
employment statistics. Id. at 303. Those plaintiffs sued solely under the New
Jersey Consumer Fraud Act (“NJCFA”) and Delaware Fraud Claims Act
(“DCFA”). Neither state’s Supreme Court, however, had recognized priceinflation as a theory of damages under those two statutes. The Third Circuit
therefore held that the plaintiffs had failed to propose a theory of damages
supportable by class-wide evidence and affirmed the district court’s decision to
deny class certification under Fed. R. Civ. P. 23(b)(3). Id. at 309—3 13.
The defendants argue that, under Harnish, I should strike Mr. Weir’s
price premium damages model because the plaintiffs’ fifth count in its
amended complaint in this case is a NJCFA claim. (Weir Reply 8-9; See ECF
No. 86) But Harnish concerns Rule 23(b)(3)’s predominance requirement, not
Daubert admissibility. It is possible that Hamish will be a hurdle for Mr. Weir’s
price-premium damages model with respect to the plaintiffs’ NJCFA claims at
the class certification stage. I will not decide class certification (or even a basis
to dismiss a count on the merits), however, by the back-door method of
deciding a Daubert motion.
To analyze defendants’ contentions here, I must be in a position to
“formulate some prediction as to how specific issues will play out” on the
merits. Hamish, 833 F.3d at 304 (quoting In re Hydrogen Peroxide Antitrust
Litig., 552 F.3d 305, 311 (3d Cir. 2008)). And those “specific issues” will, of
course, include the plaintiffs’ several non-NJCFA claims. It is premature to
determine whether a price premium damages model must be excluded
pursuant to Fed. R. Civ. P. 23 or Comcast. Suffice it to say that Hamish does
not preclude Mr. Weir’s price premium damages model at the Daubert stage.

42

c) Mr. Weir’s reliance on the Whirlpool document
The Defendants next attack Mr. Weir’s reliance on a Whirlpool document,
author unknown (the “Whirlpool document”) to calculate a price premium. (A
copy of the Whirlpool document is at Bursor Decl. Ex. 1, ECF No. 164-1.) This
Whirlpool document includes a chart comparing pricing data for Energy Star
versus non-Energy Star washing machines. The chart does not indicate what
models or brands of washing machines are represented, and it appears to
depict data from 2006. (Id.) From the data in the chart, Mr. Weir calculates a
55.7% price premium, which he attributes to Whirlpool itself as a kind of
admission. (Weir Report ¶j 26, 32—33 (“Whirlpool determined that there was a
price premium of 55.7% attributable to the Energy Star.”); Id. 3; see also Weir
Br. 10)
Defendants contend that Mr. Weir’s deposition testimony establishes that
he knows nothing about the date, author, content, basis, or reliability of the
Whirlpool document. (Weir Br. 10 (citing Weir deposition testimony, ECF No.
198-3) Defendants point also to the document’s date of 2006, which precedes
by three years the relevant 2009-2010 time frame (by which point, the
defendants claim, competition drove down the price differential). (Id. 11—12)
As further support for their position, the defendants have submitted a
declaration by Whirlpool’s Global Sustainability Director, Ronald L. Voglewede.
(See Decl. of Ronald L. Voglewede (“Voglewede Decl.”), ECF No. 178-9) Mr.
Voglewede states that the chart in the Whirlpool document “does not show the
existence of a 55.7% price premium attributable to the Energy Star logo for the
Maytag Centennial washers at issue here or any other washers. Rather, the
charts on page WDZ9999209 appear to show a comparison of the prices of
Whirlpool washers that happen to be Energy Star versus prices for washers that
happen to be non-Energy Star.” (Id.

¶

5) Whirlpool’s Energy Star-certified

washers, he says, “were far more likely to have several premium features that
contributed to higher prices” in the 2006-2007 time frame. (Id.

¶

6) By

contrast, “[t]he 2009-20 10 timeframe saw particularly intense competition and

43

downward pressure on the prices of Energy Star top-loading washers.” (Id.

¶

7)

The plaintiffs respond first that criticisms of an expert’s assumptions or
the incompleteness of the expert’s analysis, absent a severe methodological
flaw, go to the weight of testimony only. (Weir Opp. 15). Even setting that aside,
they say, “Mr. Weir did not analyze [the Whirlpool document] in a vacuum,” but
also considered “a plethora of additional internal documents” and
corroborating admissions from Whirlpool executives. (Id. 16) For example,
Weir’s rebuttal report cites a presentation given by Whirlpool to its investors
indicating “an expected up front price premium of $200 for clothes washers,”
as well as more general statements by Whirlpool indicating that Energy Star
qualified products sell at higher prices. (Weir Rebuttal

67) Further, the

¶

plaintiffs continue, Mr. Weir concluded from internal documents that the price
differential reported in the 2006 Whirlpool document was applicable to the
2009—2010 time frame. (Id. 16—17) And that differential, they say, survives a
reality check: the 55.7% price differential in the Whirlpool document “is
approximately within the range of Dr. Dennis’s survey results (48.5%) and Dr.
Sukumar’s survey results (44.3%).” (Id. 16)
On reply, the defendants protest that the corroborating documents cited
are no more specific than the Whirlpool document itself. (Weir Reply 4) For
example, Mr. Voglewede states that the “2008 presentation given at an equity
conference..

.

.

shows a supposed $200 initial price premium for an Energy

Star washer, but it is unclear from the graph what ‘price premium’ should
mean in this context, whether the price analysis involved comparable washers,
and what year or years this data is supposed to cover.

.

.

.“

(Voglewede Decl.

¶

8) He provides similar explanations for the other documents Mr. Weir cites.
Here is how I see it. The Whirlpool document is a Bates-stamped
document produced by Whirlpool in discovery. It can fairly be read to support
the generalized notion that unspecified (perhaps all?) Whirlpool Energy Star
qualified washing machines yielded higher revenues than non-qualified
washing machines. It is not tied specifically to the C6 and C7 Washers at issue
here. From the face of the document, one could reasonably glean—as Mr. Weir
44

does—that the data depicted in the document comes from a “Whirlpool
Government Relations analysis of Energy Star qualified product sales versus
non-qualified products.” (Whirlpool document at WDZ00002O8) Likewise, it
would not be unreasonable to read the “plethora” of corroborating Whirlpoolproduced documents reviewed by Mr. Weir relies as stating that Whirlpool
executives placed some independent economic value on the Energy Star
qualification. (See, e.g., Weir
Weir Report at

¶J

Opp.

24 (citing documents relied on and cited in

24—31)) Therefore, I would not go so far as to say, as the

defendants suggest, that Mr. Weir “knows nothing about the document or the
data it summarizes.” (Weir Br. 10) The results of Mr. Weir’s independent
inquiry seem to be far from definitive, but neither was that inquiry so
completely lacking as those in the cases defendants cite.
23
Still, to extract a 55.7% Energy Star price premium for the Washers
based on this document, without any independent investigation into the data
depicted in the document, is just too much. Here, Mr. Weir makes the kind of
logical leap that other district courts in the Third Circuit have deemed
unreliable and grounds for exclusion. “[E]xperts who use data in their reports
24

E.g., Legendary Art, LLC v. Godard, No. CIV.A. 11-0674, 2012 WL 3550040, at
*4 (E.D. Pa. Aug. 17, 2012) (“There is no evidence that [the plaintiffs expert] did
anything to veriIr [the plaintiffs own] profit and loss projections. Indeed, his report
states several times that his analysis assumed the validity of the projections.
Moreover, there is no evidence that Mr. Slocumb had any familiarity with the methods
or reasoning used by [the plaintiff] to arrive at the projections. Mr. Slocumb’s reliance
on projections supplied by [the plaintiff], without independent verification, renders his
analysis unreliable.” (internal quotation marks and citations omitted)).
23

24

See, e.g., Bruno v. Bozzuto’s, Inc., 311 F.R.D. 124, 138 (M.D. Pa. 2015)

(“Plaintiffs’ experts were wholly unfamiliar with the methodology or purposes
underlying the [internal documents]. They were not involved in data gathering or
composition.
Plaintiffs’ experts
; they merely took what Defendant gave them.
sole connection with the composition of the [defendants’ sales projection documents]
came either from what Plaintiffs or their counsel chose to inform them of or what they
remembered from certain discovery documents they were given to review—conduits
that the evidence reveals were inadequate to furnish the experts with the requisite
background information. The experts’ memories as to whether a market study had
been conducted, how many [sales projection documents] were generated, and which.
was in fact the final one on which Defendant relied were largely incomplete and often
contradicted by Defendant’s own testimony and internal documentation.”); see also
.

.

.

45

.

.

without independently verifying the accuracy or reliability of those figures fail
to satisfy this Circuit’s reliability requirement.” Bruno v. Bozzuto’s, Inc., 311
F.R.D. 124, 138 (M.D. Pa. 2015).
In short, it is not enough to simply seize on the figure because it appears
in a document obtained from Whirlpool’s files. And the “corroborating”
documents, while a step in the right direction, do not sufficiently validate the
Whirlpool document or pin down its meaning. I am confirmed in that
conclusion by Mr. Voglewede’s declaration. There are simply too many
unknown variables that would affect the data in the anonymous Whirlpool
document. Without more serious investigation of the basis for the figures, they
cannot furnish a reasonable basis for an Energy Star logo price premium
estimate. Cf Magistrini v. One Hour Martinizing Dry Cleaning, 180 F. Supp. 2d
584, 604 (D.N.J. 2002) (excluding expert’s opinion as unreliable, noting “there
is simply too great an analytical gap between the data purportedly relied on.
and the opinion.

.

.

proffered,” in part because the expert “did not adequately

explain his methods for assessing the internal validity of the studies that he
relied upon in rendering his opinion.” (internal quotation marks omitted)),
aff’d, 68 F. App’x 356 (3d Cir. 2003).
Therefore, I will grant the defendants’ motion to exclude Mr. Weir’s
opinions to the extent they rely on the data presented in the Whirlpool
document as representing Whirlpool’s own “price premium” projection for these
Washers. I will strike the 55.7% price premium Mr. Weir derives from the
anonymous Whirlpool document and the projected price premium damages of
Montgomery Cty. v. Microvote Corp., 320 F.3d 440, 449 (3d Cir. 2003) (district court
did not abuse its discretion in excluding expert’s testimony where expert had not
independently investigated unreliable data underlying a party document and did not
know, inter alia, how the document was created).
The cases that the plaintiffs cite for the proposition that an expert’s failure to
conduct further investigation is an issue of weight rather than admissibility do not
concern an expert’s reliance on unverified, unvalidated documents or data. Rather,
those cases address criticisms of an expert’s chosen methodology or reliance on
another expert’s sufficiently reliable data. (See Weir Opp. 15 (collecting cases))

46

$97,498,148 that he projects based on that figure. (See, e.g., Weir Report

¶J

33—34, 53 (“Whirlpool Study” row in Table 4))
d) Whether Mr. Weir should have considered conflicting market data
from 2009—2010 or subtracted benefits received
The defendants next make arguments similar to those made against Drs.
Sukumar and Dennis: first, that Mr. Weir did not examine actual,
contradictory marketplace data from 2009 and 2010, and second, that Mr.
Weir failed to subtract from his price premium damages the corresponding
benefits that purchasers may have received. (Weir Br. 13_16).25 The defendants
themselves frame their second criticism as a Comcast problem (see Weir Br.
16—17), and on reviewing it, I agree. For the reasons expressed above, I do not
26
address it on this Daubed motion. See Section III.A.4, supra.

To the first criticism, the plaintiffs respond that consistency with other
market data is a matter of weight, not admissibility. (Weir Opp. 17) Setting that
aside, they point to available 2009 and 2010 market data concerning the C500
The defendants also argue that Mr. Weir’s price premium damages opinions
must be excluded because they rely, in part, on the “unreliable and inadmissible”
price premium opinions of Drs. Sukumar and Dennis. (Weir Br. 17—18) Because I find
the Doctors’ opinions admissible under Rule 702, there is no question that Mr. Weir is
entitled to rely on those opinions without rendering his own opinions unreliable. See
Fed. R. Evid. 703; see also, e.g., Summit 6, LLC v. Samsung Elecs. Co., 802 F.3d 1283,
1298 (Fed. Cir. 2015) (explaining that an expert “need not be a survey expert to testify
about the information compiled by third-party surveys, so long as the information is of
a type reasonably relied upon by experts in the field to form opinions upon the
subject.”).
26
Even if the defendants had framed their concern as a Daubert reliability or fit
issue, the plaintiffs present good—if not overwhelmingly convincing—grounds as to
why Mr. Weir chose not to subtract benefits received from his damages calculations:
(1) the defendants’ own expert, Dr. Scott, excluded rebates from her definition of
market prices and made no adjustments for rebates or tax refunds in her analyses (Id.
22); and (2) realization of rebates and refunds is uncertain and occurs, if at all, after
the time that the Washers were purchased, without any tie to the conduct of the
defendants, whereas the plaintiffs’ theory of price premium damages assumes they
permitted to base his opinion on a
occur at the time of purchase. (Id.) “An expert is
particular version of disputed facts and the weight to be accorded to that opinion is for
a proper subject for cross-examination.” Walker v. Gordon, 46 F.
the jury. It is also.
Appx 691, 695—96 (3d Cir. 2002) (affirming district court’s decision to admit expert
opinion over charge that expert had failed “to rely on all of the evidence in the case”).

25

.

.

.

47

.

.

Maytag model. Mr. Weir explained in his rebuttal report, however, that his
rejection of the C500 data had a reasonable basis: (a) the C500, which had
different features from the Washers, was not comparable; and (b) nearly all of
the C500 models had been sold by the time the C6 models, “designed and
introduced as the successor to the C500,” hit the market. (Weir Opp. 17—19
(citing Weir Rebuttal

¶J

5—31)) In short, the two did not compete for the

consumer dollar. Even so, Mr. Weir calculated for purposes of argument that
the C6 model carried a 48% price premium over the C500 model.
In response to Mr. Weir’s 48% figure, the defendants’ expert Dr. Scott
submitted a rebuttal report (discussed further in Part III, below). Dr. Scott
concluded that the average weighted price of the C6 Washers was actually
lower than the non-Energy Star C500 models. (Id. 20—21 (citing Rebuttal
Expert Report of Carol A. Scott, Ph.D., July 1, 2016, Deckant Decl. Ex. G, ECF
No. 215-7) The plaintiffs rejoin that Dr. Scott cherry-picked data from
inappropriate time periods. (Id.)
Enough. What is presented here is a classic battle of the experts over
disputed facts, to be settled by the finder of fact; it does not affect admissibility.
See, e.g., In re Gabapentin Patent Litig., No. CIV.A. 00-293 1, 2011 WL
12516763, at *10 (D.N.J. Apr. 8, 2011) (concluding that defendants’ critiques
of plaintiffs’ experts’ methodology and inconsistent conclusions presented “a
battle of the experts, and both sides will be permitted to present expert
testimony on these issues and to cross-examine the other side’s expert
witnesses.”); Walker, 46 F. App’x at 695—96, supra 19; Hartle, 2014 WL
1317702, at *7, supran.18.
3) Mr. Weir’s Energy Expense Damages Model
The defendants next turn to the “energy expense damages” component of
Mr. Weir’s model. This component, they claim, should be excluded because it
rests on false assumptions and miscalculations. (Weir Br. 18) The defendants
especially take issue with the model’s assumptions that: (a) the Washers
consumed more energy than Energy Star regulations permit; and (b) the United
48

States Department of Energy (“DOE”) tests for measuring energy consumption
reliably represent real-world use. They argue that the Washers only failed to
meet the DOE’s Energy Star standards because the DOE changed its procedure
for measuring washer capacity. The Washers themselves remained unchanged
from the time when they did meet Energy Star standards. Because Mr. Weir did
not account for this, the defendants say, his “energy expense calculations lack
any factual foundation, and his methodology is designed to maximize the
appearance of damages where there are none.” (Weir Br. 19) Additionally, the
defendants argue, Mr. Weir failed to account for how the putative class
members actually used their washers, as opposed to the way the DOE tested
the Washers in the laboratory. (Id. 21)
The defendants also submit that when Dr. Weir calculated the electrical
use of the Washers, he “used a made-up drying energy input when he should
have used the actual drying energy from the DOE’s audit tests

.

.

.

.“

(Id. 23)

The defendants claim Dr. Weir’s use of an unexplained, made-up number
resulted in a $15 million error in the plaintiffs’ favor.

Such a basic

mathematical miscalculation, the say, renders Dr. Weir’s analysis unreliable
and highlights Dr. Weir’s misunderstanding of the DOE tests on which he
relies. (Id. 24)
The plaintiffs respond that the defendants’ arguments concerning the
DOE’s results go to the merits of the case and have no bearing on whether Mr.
Weir has presented an adequate damages model. (Weir Opp. 25) And, they say,
Mr. Weir chose to rely on DOE data because he reasonably understands the
data to be representative of general consumer behavior—that is, it is relevant in
practice, not just theory. (Id. 26—27 (citing Weir deposition testimony, Deckant
Decl. Ex. D, ECF No. 2 15-4)) As for the miscalculations, the plaintiffs stress
that a potentially incorrect input, the problem identified by the defendants’
experts, does not render Mr. Weir’s methodology unreliable. (Id. 27)
I agree with the plaintiffs that the defendants’ criticisms go to weight, not
admissibility. First, I do not think Mr. Weir’s reliance on DOE results and

49

figures raises the same concerns as, say, his reliance on the Whirlpool
document. Mr. Weir has researched and understands the source, context, and
meaning of the DOE’s data here, and the DOE’s calculations have a direct
factual connection to this case. Indeed, they are at the root of the plaintiffs’
allegation that the Energy Star label was inappropriate.
As an opinion witness, Weir is entitled to rely on facts or data of the type
reasonably relied upon by expert in the field. Daubert, 509 U.S. 579; Fed. R.
Evid. 703. Here, I think Mr. Weir has good grounds to rely on the DOE data for
purposes of calculating potential energy expense damages, and I see no reason
that experts in the field of damages would not reasonably rely on data of this
type. See In re Paoli R.R. Yard PCB Litig., 35 F.3d 717, 749 (3d Cir. 1994)
(“[Wihen a trial judge analyzes whether an expert’s data is of a type reasonably
relied on by experts in the field, he or she should assess whether there are
good grounds to rely on this data to draw the conclusion reached by the expert.
Whether experts in the field rely on this type of data will simply continue to be
a part of the judge’s analysis.) If the DOE data is truly flawed or inapt, the
defendants should have no problem making that clear on cross examination.
Additionally, if it is the case that the Washers are just as efficient as
Energy Star qualified washing machines, or as efficient as they are labeled—a
theory directed to the merits of plaintiffs’ allegations—then the defendants will
have the opportunity to establish this by opposing evidence, as opposed to
exclusion of the plaintiffs’ evidence.
Finally, mathematical errors and flawed data can of course be significant
in the Daubert calculus. See, e.g., Dart v. Kitchens Bros. Mfg. Co., 253 F. App’5c

395, 398 (5th Cir. 2007) (mathematical errors in another expert’s underlying
calculations on which expert relied was one of several flaws contributing to the
unreliability of expert’s methodology). Here, however, I am not convinced that
Mr. Weir’s alleged miscalculations render his energy damages model unreliable
and inadmissible. In fact, the “miscalculation” the defendants allege is not
really a matter of mathematics, as implied; rather, it is a factual dispute among
experts regarding the value and source of an input that is fully disclosed and
50

can be disputed factually. (See Weir Report Ex. 3 (“Drying Factor” input)) The
27
principle of GIGO (garbage in, garbage out) can be grasped by a jury.
Having reviewed Mr. Weir’s initial and rebuttal reports, I agree that he
could have provided a clearer explanation as to how he chose a drying energy
input. (See Weir Report

¶ 10, Ex. 3; Weir Rebuttal ¶ 129) But the defendants’

own expert, Dr. Fessler, achieves the opposite of the defendants’ intended
purpose on this Daubert motion; he figures out that Mr. Weir derived his drying
energy input by “back-calculating” from figures displayed on the Washers’
EnergyGuide label. (See Fessler Report
28

fi 56—62) Mr. Weir’s back-calculation

from official DOE test results is a sufficiently reliable methodology which rests
on a sufficient foundation for Daubed purposes. It can be tested on cross29

The defendants’ arguments are inspired by the opinions of their own experts,
Dr. M. Laurentius Marais (see Expert Report of M. Laurentius Marais, Ph.D., April 26,
2016, ECF No. 177-4 (“Marais Report”)) and Dr. John R. Fessler (see Expert Report of
John R. Fessler, Ph.D., P.E., ECF No. 177-6 (“Fessler Report”)), who attack various
aspects of Mr. Weir’s methodology. Mr. Weir, in turn, rebuts the opinions of Drs.
Marais and Fessler and presents reciprocal quibbles with Drs. Marais’s and Fessler’s
methodologies. (See Weir Rebuttal ¶J 89—133)
28
According to the government-sponsored Energy Star website:
27

All major home appliances must meet the Appliance
Standards Program set by the US Department of Energy
(DOE). Manufacturers must use standard test procedures
developed by DOE to prove the energy use and efficiency of
their products. Test results are printed on faj yellow
EnergyGuide label, which manufacturers are required to
display on many appliances. This label estimates how much
energy the appliance uses, compares energy use of similar
products, and lists approximate annual operating costs.
Your exact costs will depend on local utility rates and the
type and source of your energy.
https: / / www. energystar. gov/ index. cfm?c=appliances. pr_energy_guide (emphasis
added).
As the DOE’s fmdings that the Washers do not meet Energy Star standards lie
at the heart of plaintiffs’ claims, I also find that there is a sufficient factual connection
between the DOE’s results and this case for purposes of Daubert’s fit requirement.

29

I reach these same conclusions with respect to the defendants’ related
argument, made in a footnote, that Mr. Weir made “several other significant
calculation errors.” (Weir Br. 23 n. 13) Mr. Weir’s reports identify reliable sources for
51

examination, for example by plugging in alternative, arguably better-chosen,
inputs. (See Weir Rebuttal ¶j 129—131 (explaining that Drs. Marais and Fessler
only take issue with his choice of inputs, not his basic framework, and
accepting that other inputs can be tested in his model); Weir Report Ex. 3
(indicating that the drying energy input is adjustable))
Once again, I conclude that we are faced, not with an inadmissible expert
opinion, but with a battle of the experts, suitable for sorting out in subsequent
motion practice or at trial.
4) Whether Dr. Weir Double-Counts Damages
Finally, the defendants argue that Dr. Weir’s model double-counts
damages: that is, it assumes invalidly that the putative class should be
awarded both “price premium” and “energy expense” damages. This
assumption, the defendants say, threatens to reimburse the plaintiffs for the
Energy Star “price premium” (the price differential between an Energy Star and
non-Energy Star washer, which is rationally based on the expectation of
efficiency-based savings) and also to compensate them for not having received
those very savings. (Weir Br. 23—24) In other words, “it would be as though
Plaintiffs had bought a conventional washer.
efficiency benefits of the Energy Star washer

.

but realized the energy

.

.

.

.

.“

(Id. 25)

The plaintiffs say that the defendants have simply misunderstood: “Mr.
Weir is not proposing that his two damages models must be used in
combination.” (Weir Opp. 28) Although Mr. Weir testified that he believes both
methods could be used concurrently, he also states that the models are
independent. (See Weir Rebuttal

¶ 90 (“I believe the determination of which

damage methodologies are selected in this litigation will be made by the Court
or jury, and I offer no opinion as to whether Plaintiffs are legally entitled to one
of both of these damage recoveries. I have simply offered multiple damage
his input data—mainly DOE results and Energy Star specifications set forth in the
Code of Federal Regulations—and his input values are disclosed in his initial report
and thus susceptible to challenge at trial.
52

methodologies that I believe comport with Plaintiffs’ stated theories of liability
.“))

Thus, the plaintiffs take the position that either type of damages may be
awarded, depending on the merits of the case and the court’s determination as
to which method or methods are appropriate for calculating class wide
damages through the use of common evidence, as Fed. R. Civ. P. 23 requires.
(Id. 28—29) Here, they seem to accept, at least in principle, that the two may

overlap. (But see Section II.C.4, pp. 35—38, supra.)
For now, I will permit the plaintiffs to explore what are potentially
alternative measures of damages. The plaintiffs’ position is appropriate at this
stage of the case and raises no Daubert issue. To the extent the plaintiffs do
seek double recovery, the defendants may raise their argument at a later, more
appropriate stage of the case. See, e.g., Dewey v. Volkswagen of Am., 728 F.

Supp. 2d 546, 596 (D.N.J. 2010) (rejecting proposed double counting of
damages on motion to approve settlement), rev’d and remanded on other
grounds sub nom. Dewey v. Volkswagen Aktiengesellschaft, 681 F.3d 170 (3d

Cir. 2012).
In sum, the defendants’ motion to exclude the opinions of Mr. Weir is
granted as to the opinions of Mr. Weir’s opinions that rely on the data
presented in the anonymous Whirlpool document as Whirlpool’s own “price
premium” projection. The defendants’ motion is otherwise denied.

53

III.

PLAINTIFFS’ MOTION TO STRIKE DR. SCOTT’S REPORTS

The plaintiffs’ only motion in limine (ECF No. 221) seeks to strike the July
1, 2016 rebuttal and supplemental expert reports of Dr. Carol A. Scott, a
°
3
damages expert for the Whirlpool defendants. The plaintiffs raise no arguments
under Rule 702 or 703, Fed. R. Evid. Rather, they say the rebuttal and
supplemental reports should be stricken as a sanction under Rule 37, Fed. R.
Civ. P., because they were submitted too late and without justification under
this Court’s Scheduling Order. (See Scott Br. 9—12) That Order sets April 26,
2016 as the deadline for “Defendants’ class based expert reports in support of
their opposition to class certification,” and does not provide for submissions of
additional expert reports. (See Scheduling Order).
Legal Standard: Exclusion of Expert Testimony Pursuant to Rule 37(c)(fl

A.

Rule 26(a)(2) provides that an expert witness retained to testify in a case
must submit a report. Rule 26(e)(1) provides:
A party who has made a disclosure under Rule 26(a)—or who
has responded to an interrogatory, request for production, or
request for admission—must supplement or correct its
The following citations will be used for Dr. Scott’s reports and the briefmg on
the plaintiffs’ motion to strike them:
30

•

“Scott Report” = Expert Report of Carol A. Scott, Ph.D., April 26, 2016, Deckant
Decl. Ex. M, ECF No. 2 15-13

•

“Scott Rebuttal” = Rebuttal Expert Report of Carol A. Scott, Ph.D., July 1, 2016,
Deckant Decl. Ex. G, ECF No. 2 15-7

•

“Scott Supplemental” = Supplemental Expert Report of Carol A. Scott, Ph.D.,
July 1, 2016, Deckarit Ex. H, ECF No. 2 15-8

•

“Scott Br.” = Memorandum of Law in Support of Plaintiffs Motion to Strike Dr.
Carol A. Scott’s July 1, 2016 Rebuttal Expert Report and Supplemental Expert
Report, ECF No. 224

•

“Scott Opp.” = Opposition to Plaintiffs’ Motion to Strike Dr. Carol A. Scott’s July
1, 2016 Rebuttal Expert Report and Supplemental Expert Report By Defendants
Whirlpool Corporation, Lowe’s Home Center, LLC, Sears Holdings Corporation,
and Fry’s Electronics, Inc., ECF No. 231

•

Scott Reply” = Reply Memorandum of Law in Further Support of Plaintiffs
Motion to Strike Dr. Carol A. Scott’s July 1, 2016 Rebuttal Expert Report and
Supplemental Expert Report, ECF No, 232
54

disclosure or response:
(A) in a timely manner if the party learns that in some
material respect the disclosure or response is
incomplete or incorrect, and if the additional or
corrective information has not otherwise been made
known to the other parties during the discovery
process or in writing; or
(B) as ordered by the court.
Fed. R. Evid. 26(e)(1).
Under Rule 37(c)(1), “[ijf a party fails to provide information or identify a
witness as required by Rule 26(a) or (e), the party is not allowed to use that
information or witness to supply evidence on a motion, at a hearing, or at a
trial, unless the failure was substantially justified or is harmless.” Fed. R. Civ.
P. 37(c)(1). It follows that a court has the discretion to bar expert testimony
that is not proffered in accordance with Rule 26, unless the noncompliance is
excused. See Geis v. Tricam Indus., Inc., No. CIV.A. 09-1396 MLC, 2010 WL
8591142, at *2 (D.N.J. Oct. 6, 2010). A sister court within this Circuit has
usefully summarized the standards under Rule 37(c)(1) for excusing
noncompliance as “substantially justified” or “harmless”:
The party against whom the sanction is sought has the
“burden of proving substantial justification or that its
failure to produce was harmless.” Tolerico v. Home
Depot, 205 F.R.D. 169, 175 (M.D. Pa. 2002).
“Substantial justification’ for the failure to make a
required disclosure has been regarded as ‘justification
to a degree that could satisfy a reasonable person that
parties could differ as to whether the party was
required to comply with the disclosure request.’ “ Id.
(quoting United States v. Dentsply Intern., Inc., No. Civ.
A. 99—5, 2000 WL 654378, at *7 (D. Del. May 10,
2000). “The test of substantial justification is satisfied
if ‘there exists a genuine dispute concerning
compliance.”’ Id. at 175—76 (quoting Henrietta D. v.
Giuliani, No. 95—CV—0641, 2001 WL 1602114, at *5
(E.D.N.Y. Dec. 11, 2001)).
Vaskas v. Kenworth Truck Co., No. 3:10-CV-1024, 2013 WL 1207963, at *3
(M.D. Pa. Mar. 25, 2013).
55

The exclusion of critical evidence, however, “is an ‘extreme’ sanction, not
normally to be imposed absent a showing of willful deception or ‘flagrant
disregard’ of a court order by the proponent of the evidence.” Meyers v.
Penriypack Woods Home Ownershz, Ass’n., 559 F.2d 894, 904—905 (3d Cir.
1977) (citation omitted), overruled on other grounds, Goodman v. Lukens Steel,
777 F.2d 113 (3d Cir.1985), affd 482 U.S. 656, 107 S. Ct. 2617 (1987). The
Third Circuit has set forth factors (the “Pennypack factors”), to be considered
when determining whether “exclusion of evidence is an appropriate sanction for
failure to comply with discovery duties”:
(1) the prejudice or surprise of the party against whom
the excluded evidence would have been admitted; (2)
the ability of the party to cure that prejudice; (3) the
extent to which allowing the evidence would disrupt
the orderly and efficient trial of the case or other cases
in the court; and (4) bad faith or willfulness in failing
to comply with a court order or discovery obligation.
Nicholas v. Pennsylvania State Univ., 227 F.3d 133, 148 (3d Cir. 2000); see also
Pennypack Woods, 559 F.2d at 904—05. The Third Circuit has supplemented
that Pennypack list, also considering (5) “the importance of the excluded
testimony” and (6) the party’s explanation for failing to disclose.
Konstaritopoulos v. Westvaco Corp., 112 F.3d 710, 719 (3d Cir. 1997) (quoting
Pennypack, 559 F.2d at 905); Vaskas v. Kenworth Truck Co., No. 3: 10-CV1024, 2013 WL 1207963, at *3 (M.D. Pa. Mar. 25, 2013) 31
B.

Discussion

The plaintiffs argue that the six Pennypack factors weigh in favor of
striking Dr. Scott’s reports, for the following reasons: (1) the plaintiffs were
prejudiced because Dr. Scott’s supplemental and rebuttal reports were
submitted just hours before the close of expert discovery, leaving the plaintiffs
no time to depose Dr. Scott or submit a responsive expert report; (2) the
plaintiffs cannot cure the prejudice, unless granted leave to file an expert

For simplicity, I will refer to all six as the “Pennypack factors”.
56

report in response to Dr. Scott’s additional reports, because expert discovery
32
has closed; (4) disclosure of the reports on the very day the plaintiffs were
required to submit their reply brief in support of class certification indicates
bad faith; (5) Dr. Scott’s additional reports are important to the damages
analysis because Dr. Scott opines that C6 washers do not carry a price
premium over the C500 washers; and (6) the defendants cannot argue that Dr.
Scott’s additional reports were not untimely by virtue of their compliance with
Rule 26.
The defendants respond that I should deny the plaintiffs’ motion for three
reasons.
1) The Plaintffs’ Failure to Follow Local Rule 37.1

First, the defendants say the plaintiffs themselves have failed to comply
with this District’s Local Civil Rule 37.1(a) and the initial scheduling order in
this case. The Rule instructs parties to confer and attempt to resolve discovery
disputes among themselves before seeking the court’s intervention. (Scott Opp.
9—i 1) The initial scheduling order provides that “[nb discovery.
motions for
sanctions for failure to provide discovery shall be made without prior leave of
.

.

Court,” and directs the parties to bring discovery disputes to the court that
cannot be informally resolved within 14 days. (ECF No. 103

¶

8) By defendants’

reckoning, the motion was brought tardily and without the required prior
consultation.
The plaintiffs do not deny that they never sought to informally resolve the
Scott-related issues with the defendants. They contend that they adequately
disclosed their intentions, however, when they filed a letter with this Court
requesting a unified briefing schedule for both sides’ Daubert motions. (Scott
Reply 10) The plaintiffs’ motion to strike pursuant to Rule 37(c)(1), however, is
not a Daubert motion. It is a discovery motion that should have been discussed
if not resolved by the parties, with or without the help of the Magistrate Judge,

32

The plaintiffs do not make any argument as to the third factor.
57

months ago. Finding some procedural fault on both sides, however, I will
excuse the plaintiffs’ noncompliance and consider the merits of their motion.
2) Whether the Reports Comply with Rule 26
The defendants attack the very premise of the plaintiffs’ motion. Dr.
Scott’s Rebuttal and Supplemental Reports, they say, were timely submitted
under Rule 26.
a) Dr. Scott’s Rebuttal Report
The defendants argue that Dr. Scott’s rebuttal expert report was
intended solely to contradict or rebut a new opinion stated in Mr. Weir’s
rebuttal expert report. The Scheduling Order did not specifically address the
timing of defendants’ rebuttal reports. Therefore, they say, we fall back upon
the default deadline in the Federal Rules: the defendants’ rebuttal reports
needed to be disclosed within 30 days after the plaintiffs’ reports to which they
responded. Fed. R. Civ. P. 26(a)(2)(D)(ii). (Id. 12—13). Mr. Weir’s rebuttal report
was dated June 10, 2016; Dr. Scott’s Rebuttal Report in response was dated
July 1, 2016, within the Rule’s 30-day deadline.
Rule 26(a)(2)(D)(ii) provides: “Absent a stipulation or a court order, the
disclosures must be made:

.

.

.

(ii) if the evidence is intended solely to

contradict or rebut evidence on the same subject matter identified by another
party under Rule 26(a)(2)(B) or (C), within 30 days after the other party’s
disclosure.” Fed. R. Civ. P. 26(a)(2)(D)(ii) (emphasis added). The highlighted
language raises the question of whether the Scheduling Order in this case is a
“court order” that overrides the deadline in the Rule. The Scheduling Order
does, of course, set a deadline of April 26, 2016 for “Defendants’ class based
expert reports in support of their opposition to class certification” and a
deadline of June 10, 2016 for “Plaintiffs’ class based rebuttal expert reports.”
33

See Scheduling Order at 2 (also noting that “no further extensions will be
granted).
58

The Scheduling Order, however, is “silent as to the submission by
Defendants of rebuttal reports in the event that Plaintiffs’ experts offer new
opinions.” (Scott Opp. 12) Several cases hold that where a scheduling order
exists, but is silent on a particular point, then it does not constitute an
overriding “court order” within the meaning of Rule 26(a)(2)(D)(ii). I agree.
34
Under the circumstances, I apply the default Rule 26 deadline of 30 days.
Because Dr. Scott’s rebuttal report was submitted within 30 days, it was
timely.
To say that defendants met the deadline for a rebuttal report, however,
begs the question of whether Dr. Scott’s report is properly regarded as a
rebuttal report at all. Plaintiffs, citing Rule 26(a)(2)(D)(ii), say no, for two
reasons. First, they say, Mr. Weir’s rebuttal report did not contain any “new
opinion” requiring additional rebuttal, but only a brief concession arguertdo
that fWeir were to compare the C500 and C6 models (a comparison which he
maintains is inappropriate), then he would find a positive price premium
discount “of as much as 48% or $196.81.” Second, they say that Dr. Scott’s
rebuttal adds nothing to the conclusions of her initial report. (Scott Reply 6)
I disagree. Mr. Weir’s calculation of a price premium between the C500
and C6 models is new; it does not appear in his earlier report. (See Weir
Rebuttal

¶

24) That it is narrow, brief, or stated arguendo does not make it any

less new. There is no room in the Rules for plausible deniability; plaintiffs are
either offering Weir’s new opinion for the fact finder’s consideration, or they are
not. If they are not, it should not be in the report; if they are, it is subject to

See, e.g., Fulton Fin. Advisors v. Natcity fnvs., Inc., Civil Action No. 09-4855,
2016 WL 5461897, at *2 (E.D. Pa. Sept. 28, 2016) (“Courts have held that where a
[case management order] is silent as to rebuttal expert reports, the terms of Rule
26(a)(2)(D)(ii) permits them if they are timely disclosed.”); Teledyne Instruments, Inc. v.
Cairns, Civ. A. No. 12-854, 2013 WL 5781274, at * 17 (M.D. Fla. 2013) (“The
[scheduling order’s] failure to set a deadline for the disclosure of rebuttal expert
witness reports does not mean that rebuttal expert witness reports are not permitted.
It simply means that rebuttal expert witness reports must be submitted within the
period set forth in Rule 26(a)(2)(D)(ii).” (also collecting cases and noting this is the
majority view)).

59

rebuttal. Such rebuttal, of course, must be confined to the scope of the new
35
opinion. Dr. Scott’s rebuttal report, I find, is confined to rebutting Mr. Weir’s
new “48% or $196.81” calculation. It does nothing more than describe Mr.
Weir’s new calculation (Scott Rebuttal Report
the basis for that calculation (id.

¶J

¶

5); analyze the data he cited as

7—8); and conclude that Mr. Weir did not

provide sufficient information to permit verification of it (id.

¶J

3—4, 9).

Dr. Scott’s report is properly considered a rebuttal report. As such, for
the reasons expressed above, it is timely. (See Scott Reply 4—5)
b) Dr. Scott’s Supplemental Report
As for Dr. Scott’s Supplemental Report, the defendants argue that
disclosure was timely pursuant to Fed. R. Civ. p. 26(e). Rule 26(e) requires an
expert to supplement or correct his or her disclosure “in a timely manner if the
party learns that in some material respect the disclosure or response is
incomplete or incorrect, and if the additional or corrective information has not
otherwise been macic known to the other parties during the discovery process
or in writing.” Fed. R. Civ. P. 26(e)(1)(A). The deadline for such a supplemental
expert report is “the time the party’s pretrial disclosures

.

.

.

are due.” Fed. R.

Civ. P. 26(e)(2).
36
Dr. Scott’s Supplemental Report meets the Rule 26(e) requirements, say
defendants, because it “solely concerns data that Dr. Scott did not have at the
See, e.g., In re Asbestos Prod. Liab. Litig. (No. VI), No. 09-CV-74351X, 2012 WL
661673, at *1 (E.D. Pa. Feb. 8, 2012), report and recommendation adopted sub nom. In

reAsbestos Prod. Liab. Litig. (No. IV), No. 09-74410, 2012 WL 661660 (E.D. Pa. Feb.
29, 2012) (“It is well settled that evidence which properly belongs in the case-in-chief
but is first introduced in rebuttal may be rejected so as to avoid prejudice to the
defendant and to ensure the orderly presentation of proof.” (quoting Emerick v. U.S.
Suzuki Motor Corp., 750 F.2d 19, 22 (3d Cir. 1984)); Hans v. Tharaldson, No. 3:05-CV115, 2011 WL 6937598, at *10 (D.N.D. Dec. 23, 2011) (“The function of rebuttal
evidence is ‘to explain, repel, counteract or disprove evidence of the adverse party.’
Experts are typically allowed to introduce new methods of analysis in a rebuttal report
if they are offered to contradict or rebut another party’s expert; however, a rebuttal
report should not extend beyond the scope of the other party’s expert reports.”
(quoting United States v. Luschen, 614 F.2d 1164, 1170 (8th Cir. 1980) (additional
citation omitted))).
36

The deadline for pretrial disclosures has not yet been scheduled in this case.
60

time of her original report.” Specifically, it is based on sales data that
defendant Home Depot produced for the first time on June 9, 2016. The
Supplemental Report was therefore timely produced 21 days later, on July 1,
2016. (Scott Opp. 14—15) The plaintiffs reply that Rule 26(e)(1)(A) does not
apply because the late-arriving data was not produced by them, but by one of
Whirlpool’s co-defendants, Home Depot. Data in the hands of Home Depot,
they say, should have been obtained by Dr. Scott in time for her original report.
(Scott Reply 2—3, 6)7
Rule 26(a)( 1)(A) is not confined to information withheld by an adversary.
Rather, it authorizes (indeed, requires) correction or supplementation where
the person learns that an earlier response was “incomplete or incorrect,” unless
the corrected or supplemented facts have been disclosed by other means.
The defendants have not explained in detail why Dr. Scott did not
possess Home Depot’s data. (See Scott Reply 7) For these purposes, however, I
think the explanation given—that Home Depot had not produced it—is
sufficient. Absent any indication that defendants gained a procedural
advantage by stringing out Dr. Scott’s analysis, I accept that the information
simply arrived late. Indeed, blank columns in the exhibits to Dr. Scott’s initial
expert report corroborate that impression: some data from Home Depot is
analyzed, but the rest is missing. (See Scott Report Exs. 5A—5B (leaving blank
columns for certain Fry’s and Home Depot data)) This, I think, weighs against
any inference that Dr. Scott is seeking to use Rule 26(e) as an “avenue to
correct ‘failures of omission because [she] did an inadequate or incomplete
preparation.”’ In re Asbestos Prod. Liab. Litig. (No. VI), 289 F.R.D. at 425
(quoting (AkevaL.L.C. v. Mizuno Corp., 212 F.R.D. 306, 310 (M.D.N.C. 2002)).

For what it is worth, defendants do not stand united as to this issue. Dr. Scott’s
reports are submitted on behalf of Whirlpool only. (See Deckant Decl. Ex. 0 ¶ 2; Id.
Ex. H ¶ 1 (supplementing original report); id. Ex. M ¶ 7 (ECF No. 2 15-7, 8, 13).
Whirlpool, Lowe’s, Sears, and Fry’s oppose the plaintiffs’ motion to strike Dr. Scott’s
reports. (See Scott Opp.). Home Depot takes no part.
37

61

Accordingly, I will accept Dr. Scott’s Supplemental Report as a proper,
timely supplement under Rule 26(e).
3) Substantial Justification and the Pennypack Factors
Even if Dr. Scott’s rebuttal and supplemental reports were not timely
under Rule 26, I would be reluctant to exclude critical evidence, an extreme
sanction that should be avoided if the untimely disclosure was “substantially
justified or is harmless.” Fed. R. Civ. P. 37(c)(i). Under the Pennypack factors,
which inform this standard, I would find the tardiness of Dr. Scott’s additional
reports both substantially justified and harmless.
First, the plaintiffs were neither surprised nor prejudiced. Dr. Scott’s
rebuttal report presents no new opinions, but is confined to disputing one new
point that Mr. Weir made in his rebuttal declaration. Her supplemental report
merely analyzes new Home Depot data, missing from the original report, in a
way that confirms and elaborates on the initial report. (See Scott Opp. 16—17)
Were the plaintiffs truly at risk of being prejudiced, they surely would have
sought some sort of relief from the Magistrate Judge, rather than just bringing
a retaliatory cross-motion.
Second, because I find no prejudice, there is no pressing need to
consider whether prejudice can be cured. In an abundance of caution,
however, I have afforded plaintiffs the option of submitting a 5-page
supplemental report discussing the Home Depot data.
Third, at this very early, pre-class-certification stage, consideration of Dr.
Scott’s additional reports does not threaten to disrupt the orderly and efficient
trial of this case.
Fourth, because I find that the defendants submitted Dr. Scott’s
additional reports under color of Rule 26(a)(2)(D) and 26(a)(1)(A), and at worst
were reasonably mistaken in their interpretation of the Rule, I cannot find bad
38
faith.
38
The plaintiffs stress that the defendants submitted Dr. Scott’s additional
reports “literally hours before the close of expert discovery, and the deadline for

62

Fifth, because Dr. Scott’s additional reports respond to a new opinion of
Weir and fill a prior hole in Dr. Scott’s data analysis, I think the information is
“important” to the defendants’ case.
Sixth, for the reasons already expressed, I find the defendants have
provided adequate explanations for failing to turn over Dr. Scott’s additional
reports sooner. The reports came within a few weeks after disclosure of the
information to which they responded; there are no indications that defendants
held back the reports for tactical advantage.
Taking these factors into consideration, I find that exclusion under Rule
37 is not warranted. Even assuming arguendo that Dr. Scott’s additional
reports did not comply with the procedures of Rule 26, I would still find that
the defendants have shown “a genuine dispute concerning compliance,” and
could “satisfy a reasonable person that parties could differ as to whether the
[defendants were] required to comply with” the deadlines in the Scheduling
Order. Vaskas, 2013 WL 1207963, at *3 (internal quotation marks and
citations omitted). See also Rojas v. Acuity Brands Lighting, Inc., No. CIV.A. 122220 DMC, 2014 WL 794364, at *1 (D.N.J. Feb. 27, 2014).
I find the defendants had substantial justification for acting as they did,
and that the plaintiffs suffered no harm as a result. Accordingly, the plaintiffs’
motion to strike Dr. Scott’s Rebuttal and Supplemental Reports is denied.
4) Plaintiffs’ Request for Leave to Submit a Supplemental Expert Report
Plaintiffs request that in the alternative, I grant leave for Mr. Weir to
submit a supplemental expert report in response to Dr. Scott’s additional
reports. (Scott Reply 11) This request should have been made pursuant to Rule
37.1 at the time Dr. Scott’s Rebuttal and Supplemental reports were first

Plaintiffs to submit their reply brief in further support of class certification.” (Scott
Reply 9) I think the defendants’ eleventh-hour filing bespeaks a rush to meet the
discovery deadline, rather than a desire to interfere with plaintiffs’ (also last-minute)
drafting. If the plaintiffs truly felt the need to address the additional reports,
presumably they would have sought additional briefing or leave to file supplemental
expert responses of their own. They did not.
63

disclosed. Made now, in the plaintiffs’ briefs in support of their belated motion,
it seems more like tit-for-tat than a serious request. Someone’s report has to be
last, and any issues left hanging in Dr. Scott’s reports appear to be minor.
I will, however, permit the following, and only the following. Within 14
days, the plaintiffs may—if they truly believe it necessary—submit a
supplemental expert report, not to exceed five pages, confined to a discussion
of the Home Depot data in Dr. Scott’s Supplemental Report. A further response
to Dr. Scott’s Rebuttal Report is not authorized.

IV.

CONCLUSION
For the reasons stated above, the defendants’ motions to strike Dr.

Sukumar’s and Dr. Dennis’s opinions are denied. The defendants’ motion to
strike Mr. Weir’s opinions is granted as to the portions of Mr. Weir’s opinions
that rely on the anonymous Whirlpool document as representative of
Whirlpool’s own price premium, and denied as to the remainder of Mr. Weir’s
opinions. Finally, the plaintiffs’ motion to strike the rebuttal and supplemental
reports of Dr. Scott is denied. Plaintiffs may, at their option, submit a
supplemental report, not to exceed 5 pages, confined to the issue of the
additional Home Depot data discussed in Dr. Scott’s Supplemental Report.
A separate order will issue.
Dated: March 17, 2017

KE INMCNULTY
United States District Jud

64