In the

    United States Court of Appeals
                 For the Seventh Circuit
                     ____________________
No. 17-1810
S.V. GOPALRATNAM, et al.,
                                                Plaintiffs-Appellants,

                                 v.

HEWLETT-PACKARD COMPANY, et al.,
                                               Defendants-Appellees.
                     ____________________

         Appeal from the United States District Court for the
                   Eastern District of Wisconsin.
             No. 2:13-cv-618 — Pamela Pepper, Judge.
                     ____________________

  ARGUED DECEMBER 1, 2017 — DECIDED DECEMBER 15, 2017
                     ____________________

   Before BAUER, FLAUM, and ROVNER, Circuit Judges.
    FLAUM, Circuit Judge. Plaintiffs’ son tragically perished in
a fire at plaintiffs’ home in June 2010. Believing that the fire
was caused by a defective lithium ion battery cell from their
son’s laptop, plaintiffs filed a products liability suit against
separate manufacturers of the laptop, battery pack, and indi-
vidual battery cells. Plaintiffs supported their causation the-
ory solely through testimony from two expert witnesses,
whom defendants later moved to exclude under Federal Rule
2                                                  No. 17-1810

of Evidence 702 and Daubert v. Merrell Dow Pharmaceuticals,
Inc., 

(1993). The district court granted defend-
ants’ motions to exclude, and therefore entered summary
judgment in their favor. Plaintiffs now appeal the district
court’s ruling. We affirm.
                          I. Background
    A. Factual Background
    On June 17, 2009, Arun Gopalratnam, a twenty-three year-
old college student at the University of Wisconsin-Milwau-
kee, purchased a laptop computer manufactured by defend-
ant Hewlett-Packard Company (“HP”). The laptop contained
a battery pack manufactured by defendant DynaPack Tech-
nology Corporation (“DynaPack”), which in turn held three
cylindrical-shaped lithium ion battery cells manufactured by
defendant Samsung SDI Company, Limited (“Samsung”).
    Approximately one year later, on June 4, 2010, the Me-
nomonee Falls, Wisconsin Fire Department responded to a
major fire in a basement bedroom of the home of Arun’s par-
ents, plaintiffs S.V. and Hemalatha Gopalratnam. After the
fire was extinguished, firefighters discovered Arun deceased
on the floor of the room. A later autopsy classified smoke in-
halation as the cause of death. The medical examiner discov-
ered no evidence of pre-fire injury or disease, and a toxicology
screen evidenced no drugs or alcohol in Arun’s system.
   Due to Arun’s death, Special Agent Antonio H. Martinez
of Wisconsin’s Department of Criminal Investigation was as-
signed to conduct a fire investigation. Special Agent Martinez
concluded that the fire originated in the basement bedroom
where Arun’s body was located. Although Special Agent
No. 17-1810                                                     3

Martinez excluded multiple potential sources of the blaze (in-
cluding the home’s electrical and gas meters, electrical distri-
bution panels, and gas-fueled furnaces, as well as the electri-
cal plugs, light switch, and ceiling light fixture in the bed-
room), he could not ascertain the fire’s ultimate cause. His in-
vestigation did not eliminate, however, “a possible fire within
the mattress area” of the bedroom “that extended into the ceil-
ing area.”
    During the fire investigation, investigators collected burnt
debris found on the remnants of the bedroom mattress. The
remains of Arun’s HP laptop and Nokia cell phone, including
two of the three laptop battery cells, were found amongst this
debris. Investigators then shoveled the remaining bedroom
debris out the bedroom window and into random piles in
plaintiffs’ yard. The third laptop battery cell was later found
in one of these piles.
   B. Procedural Background
    On June 4, 2013, plaintiffs filed suit in the Eastern District
of Wisconsin against HP and its insurer, defendant ABC In-
surance Company, alleging negligence, strict products liabil-
ity, and breach of warranty. Plaintiffs claimed that a defective
lithium ion battery cell in Arun’s laptop caused the fire that
led to their son’s death. On July 16, 2013, HP filed a third-
party complaint against DynaPack and Samsung. On October
31, 2014, plaintiffs amended their complaint to include Dyna-
Pack and Samsung as defendants.
   Plaintiffs supported their causation theory with two ex-
pert witnesses. First, plaintiffs retained Dr. Daniel H.
Doughty, who holds a Ph.D in inorganic chemistry, as an ex-
4                                                    No. 17-1810

pert on “battery safety.” In addition, plaintiffs retained Mi-
chael F. Hill, Sr., a retired Certified Fire Investigator with the
Illinois Chapter of the International Association of Arson In-
vestigators, to opine about the “cause and origin” of the fire.
Both Doughty and Hill issued expert reports and were de-
posed by defendants during the course of expert discovery.
       1. Doughty’s Expert Report
    Doughty physically examined the cell phone battery and
three laptop battery cells recovered from the fire. From this
examination, Doughty noted that the two laptop cells found
on the bedroom mattress (which Doughty labeled as “Cell B”
and “Cell C”) had retained their cylindrical dimensions and
internal contents throughout the fire (as did the cell phone
battery). In contrast, Doughty observed that the third laptop
cell found in the debris pile in plaintiffs’ yard (which Doughty
labeled as “Cell A”) had ejected its contents and warped into
an elliptical shape.
    Doughty then set out to explain the difference between
Cell A and Cells B and C. Doughty opined that the appear-
ance of Cell A was typical for a cell that had experienced se-
vere “thermal runaway,” which he defined as “the condition
when the rate of heat generation inside the battery cell … is
greater than [the] rate of heat dissipation.” According to
Doughty, “[a] battery cell is an energy storage device.” If bat-
tery energy “is released in a controlled manner (i.e., normal
discharge), the device is safe.” If the energy is released “in a
rapid, uncontrolled manner,” however, “[a]dditional heat
and gas are produced” that can trigger thermal runaway. This
produces “very high temperature internal to the cell” which
can cause the cell to “vent[] violently or explode[].” Doughty
No. 17-1810                                                    5

further stated that the hot “ejecta” from an exploding cell
“provides a ready source of ignition of flammable materials.”
    Doughty outlined several potential causes of thermal run-
away, including: (1) “electrical abusive conditions” (such as
an external short circuit, overcharge, or overdischarge);
(2) “mechanical abusive conditions” (such as shock, vibra-
tion, or penetration); (3) “high temperature abusive condi-
tions,” including heat from an external fire; and (4) “flaws
from within the cell” that cause an internal short circuit.
    Doughty excluded electrical abuse because there was no
evidence of an external short circuit—Arun’s laptop was not
plugged in at the time of the fire, and overcharge is usually a
benign event. He further excluded mechanical abuse based
upon his inspection of the laptop and the fact that it had been
tested against existing safety standards, including tests for vi-
bration, shock, and impact.
    Regarding the possibility of overheating from an external
fire, Doughty stated that multiple design elements within the
laptop, battery pack, and battery cells were engineered to pro-
tect against excessive external temperatures. Importantly,
Doughty further opined that, based upon existing literature,
the exposure of lithium ion cells to external fire causes “pred-
icable results.” He noted that the design of Arun’s laptop bat-
tery pack placed the cells close together and in a straight line.
Thus, Doughty reasoned, all of the cells would have been ex-
posed to the same external heat conditions from the fire.
Therefore, Doughty concluded that if external fire was the
cause, one would expect to see a uniform temperature re-
sponse from all three cells. This, however, was not the case, as
deformation in Cell A significantly differed from Cells B and
6                                                    No. 17-1810

C. To Doughty, this suggested a “different failure mecha-
nism.”
    Doughty concluded that the rate of release of the stored
energy in Cell A was “much more rapid” than the other cells,
and that the “[v]ery rapid gas generation” that created such a
substantial pressure rise was “only consistent with an internal
short circuit” that led to thermal runaway. Doughty went on
to state that internal short circuits are known to be caused by
“[f]laws within the cell,” such as “debris, foreign objects, con-
taminants, wrinkles in the electrode, etc.” These flaws can ex-
ist even in “approved cell designs” and “in cells that have
passed safety tests.” Doughty thus concluded that the internal
fault was caused by “either a manufacturing defect in the cell
that … caus[ed] an internal short circuit, or a failure of the
computer’s control/safety circuity to function as designed …
which in turn caused an internal short circuit in the cell.”
       2. Hill’s Expert Report
     Hill, the Fire Investigator, conducted site studies of plain-
tiffs’ home on two occasions in the weeks after the fire. He
also reviewed physical evidence retrieved from the home, as
well as reports, photographs, x-rays, diagrams, and blue-
prints compiled by the Menomonee Falls Fire and Police De-
partments. Hill concluded that: (1) the fire originated on the
top of the bed in the basement bedroom; (2) the “most proba-
ble” ignition source was the laptop battery; and (3) the fire
was accidental.
       3. Proceedings Below
    On June 24, 2016, defendants separately moved to exclude
the testimony of both Doughty and Hill under Federal Rule
of Evidence 702 and Daubert. Defendants contended that Hill
No. 17-1810                                                        7

and Doughty were unqualified to render expert opinions and
that their respective opinions were unreliable. Defendants
further moved for summary judgment on the theory that,
without their experts, plaintiffs could not prove causation in
any of their causes of action.
    The district court granted defendants’ motions on March
21, 2017. Although the court found both Hill and Doughty
sufficiently qualified, it nonetheless deemed their opinions
unreliable. Because the court excluded their expert testimony,
it further found that plaintiffs could not support their claim
that a defective lithium ion battery cell led to their son’s death.
As a result, the court granted summary judgment in favor of
defendants. This appeal followed.
                            II. Discussion
   A. Principles of Law
       1. The Admissibility of Expert Testimony
    “Any assessment of the admissibility of expert witness tes-
timony begins with Federal Rule of Evidence 702 and the Su-
preme Court’s opinion in Daubert, as together they govern the
admissibility of expert witness testimony.” Krik v. Exxon Mobil
Corp., 

, 673 (7th Cir. 2017). This is true even when,
as here, “our jurisdiction rests on diversity.” C.W. ex rel. Wood
v. Textron, Inc., 

, 834 (7th Cir. 2015); see also Wallace
v. McGlothan, 

, 419 (7th Cir. 2010) (noting that
“standards for admitting expert evidence” are “matters that
fall on the procedural side of the Erie divide” and are thus
governed by federal law). Rule 702 states:
8                                                             No. 17-1810

        A witness who is qualified as an expert by
        knowledge, skill, experience, training, or educa-
        tion may testify in the form of an opinion or oth-
        erwise if:
            (a) the expert’s scientific, technical, or other
            specialized knowledge will help the trier of
            fact to understand the evidence or to deter-
            mine 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 prin-
            ciples and methods to the facts of the case.
Fed. R. Evid. 702.
    In Daubert, the Supreme Court interpreted Rule 702 to re-
quire “the district court to act as an evidentiary gatekeeper,
ensuring that an expert’s testimony rests on a reliable founda-
tion and is relevant to the task at hand.” 

(citing 

). 1 This is due to the fact that

    1 Technically, Daubert interpreted a prior version of Rule 702, which
only stated:
        If scientific, technical, or other specialized knowledge will
        assist the trier of fact to understand the evidence or to de-
        termine a fact in issue, a witness qualified as an expert by
        knowledge, skill, experience, training, or education, may
        testify thereto in the form of an opinion or 

(quoting the prior rule). Rule 702 was substantially revised
in 2000 “to ‘affirm[] the trial court’s role as gatekeeper and provide[] some
general standards that the trial court must use to assess the reliability and
No. 17-1810                                                                   9

“[e]xpert evidence can be both powerful and quite misleading
because of the difficulty in evaluating it.” 

(quoting Jack B. Weinstein, Rule 702 of the Federal Rules of
Evidence Is Sound; It Should Not Be Amended, 

, 632
(1991)). This is particularly true in cases such as this involving
expert testimony “on the ultimate issue of fact.” See United
States v. Navarro, 

, 1260 n.14 (7th Cir. 1996) (quot-
ing United States v. Boyd, 

, 672 (D.C. Cir. 1995)).
   The district court’s “‘gatekeeping’ obligation … applies
not only to testimony based on ‘scientific’ knowledge, but also
to testimony based on ‘technical’ and ‘other specialized’
knowledge.” Kumho Tire Co., Ltd. v. Carmichael, 

,
141 (1999); see also Lees v. Carthage Coll., 

, 521 (7th
Cir. 2013) (“[T]he Daubert analysis applies to all expert testi-
mony under Rule 702, not just scientific testimony.”).
    In performing its gatekeeper role under Rule 702 and
Daubert, “the district court must engage in a three-step analy-
sis before admitting expert testimony. It must determine
whether the witness is qualified; whether the expert’s meth-
odology is scientifically reliable; and whether the testimony


helpfulness of proffered expert testimony.’” Dhillon v. Crown Controls
Corp., 

, 869 (7th Cir. 2001) (alteration in original) (quoting Fed.
R. Evid. 702 advisory committee’s note to 2000 amendment). These “gen-
eral standards” derive from Daubert and its progeny. See Fed. R. Evid. 702
advisory committee’s note to 2000 amendment. Thus, even though “[a]t
this point, Rule 702 has superseded Daubert, … the standard of review that
was established for Daubert challenges is still appropriate.” United States
v. Parra, 

, 758 (7th Cir. 2005); see also Manpower, Inc. v. Ins. Co.
of Pa., 

, 806 (7th Cir. 2013) (“Daubert interpreted an earlier ver-
sion of Rule 702, but it remains the gold standard for evaluating the relia-
bility of expert testimony and is essentially codified in the current version
of Rule 702.”).
10                                                     No. 17-1810

will ‘assist the trier of fact to understand the evidence or to
determine a fact in issue.’” Myers v. Ill. Cent. R.R. Co., 

, 644 (7th Cir. 2010) (quoting Ervin v. Johnson & Johnson,
Inc., 

, 904 (7th Cir. 2007)); see also Bielskis v. Louis-
ville Ladder, Inc., 

, 893–94 (7th Cir. 2011). In other
words, the district court must evaluate: (1) the proffered ex-
pert’s qualifications; (2) the reliability of the expert’s methodol-
ogy; and (3) the relevance of the expert’s testimony. Steps one
and three are not at issue here; the district court found both
Doughty and Hill to be qualified, and their testimony not only
helpful, but necessary to prove plaintiffs’ theory of liability.
See Gopalratnam v. Hewlett-Packard Co., No. 13-cv-618, 

, at *3 (E.D. Wis. Mar. 21, 2017) (“The court finds that
the plaintiffs need expert testimony to prove their claim that
a defective cell in the battery pack in Arun’s laptop caused the
fire.”). Regarding step two, however, the court concluded that
both expert opinions were unreliable.
    According to our circuit’s precedent, courts should evalu-
ate the reliability of a qualified expert’s testimony by consid-
ering, amongst other factors: “(1) whether the proffered the-
ory can be and has been tested; (2) whether the theory has
been subjected to peer review; (3) whether the theory has been
evaluated in light of potential rates of error; and (4) whether
the theory has been accepted in the relevant scientific commu-
nity.” 

(quoting Baugh v. Cuprum S.A. de
C.V., 

, 844 (7th Cir. 2017)). In addition, the Rule
702 advisory committee's note to the 2000 amendment out-
lines other benchmarks relevant in assessing an expert’s reli-
ability:
       (5) whether “maintenance standards and con-
       trols” exist; (6) whether the testimony relates to
No. 17-1810                                                   11

       “matters growing naturally and directly out of
       research they have conducted independent of
       the litigation,” or developed “expressly for pur-
       poses of testifying”; (7) “[w]hether the expert
       has unjustifiably extrapolated from an accepted
       premise to an unfounded conclusion”;
       (8) “[w]hether the expert has adequately ac-
       counted for obvious alternative explanations”;
       (9) “[w]hether the expert is being as careful as
       he would be in his regular professional work
       outside his paid litigation consulting”; and
       (10) “[w]hether the field of expertise claimed by
       the expert is known to reach reliable results for
       the type of opinion the expert would give.”
Fuesting v. Zimmer, Inc., 

, 534–35 (7th Cir. 2005),
opinion vacated in part on reh’g, 

(7th Cir. 2006)
(quoting Fed. R. Evid. 702 advisory committee’s note to 2000
amendment).
     “Importantly, this list is neither exhaustive nor manda-
tory.” 

; see also Kumho 

(“Daubert makes clear that the factors it mentions do not
constitute a ‘definitive checklist or test.’” (quoting 

)); 

(“Despite the list, we
have repeatedly emphasized that ‘no single factor is either re-
quired in the analysis or dispositive as to its outcome.’” (quot-
ing Smith v. Ford Motor Co., 

, 719 (7th Cir. 2000)));
United States v. Cruz-Velasco, 

, 660 (7th Cir. 2000)
(“Although the Daubert Court identified a number of factors
to be considered when evaluating the admissibility of expert
testimony … these factors do not establish a definitive check-
list.”). Instead, “a trial court may consider one or more of the
12                                                    No. 17-1810

more specific factors that Daubert mentioned when doing so
will help determine that testimony’s reliability.” Kumho 

.
    Ultimately, “there are many different kinds of experts, and
many different kinds of expertise.” 

The test of reli-
ability, therefore, “is ‘flexible,’ and Daubert’s list of specific
factors neither necessarily nor exclusively applies to all ex-
perts or in every case.” 

(quoting 

); see also 

(“Ultimately, reliability is
determined on a case-by-case basis.”). Rather, “[t]he district
court may apply these factors flexibly as the case requires.”

; see also Kumho 

(“[T]he
law grants a district court the same broad latitude when it de-
cides how to determine reliability as it enjoys in respect to its
ultimate reliability determination.”). In the end, “the gate-
keeping inquiry must be ‘tied to the facts’ of a particular
‘case,’” Kumho 

(quoting 

), and “the reliability analysis should be geared toward
the precise sort of testimony at issue and not any fixed evalu-
ative factors.” 

.
    At the same time, this flexibility is not without limit.
“[T]he district court’s role as gatekeeper does not render the
district court the trier of all facts relating to expert testi-
mony. … The jury must still be allowed to play its essential
role as the arbiter of the weight and credibility of expert testi-
mony.” Stollings v. Ryobi Techs., Inc., 

, 765 (7th Cir.
2013) (citations omitted). Rather, “Rule 702’s reliability ele-
ments require the district judge to determine only that the ex-
pert is providing testimony that is based on a correct applica-
tion of a reliable methodology and that the expert considered
No. 17-1810                                                    13

sufficient data to employ the methodology.” 

(em-
phasis added). This examination “does not ordinarily extend
to the reliability of the conclusions those methods produce—
that is, whether the conclusions are unimpeachable.” 

(emphasis added). In other words, “[a]n expert may provide
expert testimony based on a valid and properly applied meth-
odology and still offer a conclusion that is subject to doubt. It
is the role of the jury to weigh these sources of doubt.” 

.
    The focus, therefore, “must be solely on principles and
methodology, not on the conclusions that they generate.”

; see also Ford Motor 

(“[W]e emphasize that the court’s gatekeeping function fo-
cuses on an examination of the expert’s methodology.”). “The
soundness of the factual underpinnings of the expert’s analy-
sis and the correctness of the expert’s conclusions based on
that analysis are factual matters to be determined by the trier
of fact, or where appropriate, on summary judgment.” Ford
Motor 

; see also Manpower, Inc. v. Ins. Co. of
Pa., 

, 806 (7th Cir. 2013) (“Reliability … is primar-
ily a question of the validity of the methodology employed by
an expert, not the quality of the data used in applying the
methodology or the conclusions produced.”). “The district
court usurps the role of the jury, and therefore abuses its dis-
cretion, if it unduly scrutinizes the quality of the expert’s data
and conclusions rather than the reliability of the methodology
the expert employed.” 

.
   “This is not to say that an expert may rely on data that has
no quantitative or qualitative connection to the methodology
employed.” 

Indeed, Rule 702 explicitly requires that
expert testimony be “based on sufficient facts or data.” Fed.
14                                                  No. 17-1810

R. Evid. 702. In the “quantitative” sense, “‘sufficient facts or
data’ means 'that the expert considered sufficient data to em-
ploy the methodology’”; “an opinion about an average gross
sales price,” for example, “could not be reliably supported by
evidence relating to sales to only one customer ‘because a sin-
gle observation does not provide a sufficient basis for calcu-
lating an average.’” 

(quoting Stol-

). To be “qualitatively” adequate, “an ex-
pert must employ ‘those kinds of facts or data’ on which ex-
perts in the field would reasonably rely.” 

(quoting
Fed. R. Evid. 703).
    We have recognized that the line between conclusions and
methodology “is not always an easy line to draw.” 

“[C]onclusions and methodology are not entirely distinct
from one another. Trained experts commonly extrapolate
from existing data.” Gen. Elec. Co. v. Joiner, 

, 146
(1997). Nevertheless, “[t]he critical inquiry is whether there is
a connection between the data employed and the opinion of-
fered; it is the opinion connected to existing data ‘only by the
ipse dixit of the expert’ that is properly excluded under Rule
702.” 

(quoting 

) (first emphasis added). Said another way, there must be
a “rational connection between the data and the opinion.” 

.
       2. Standard of Review
    “It is well established that issues related to expert opinion
testimony are matters of law to be determined by the trial
judge.” Bradley v. Brown, 

, 436 (7th Cir. 1994). This
is “because of the trial judge’s first-hand exposure to the wit-
nesses and the evidence as a whole, and because of the judge’s
familiarity with the case and ability to gauge the impact of the
No. 17-1810                                                    15

evidence in the context of the entire proceeding.” United States
v. Walton, 

, 449 (7th Cir. 2000) (quoting United
States v. Van Dreel, 

, 905 (7th Cir. 1998)). Indeed,
we have colorfully said that “[a]ppellants who challenge evi-
dentiary rulings of the district court are like rich men who
wish to enter the Kingdom: their prospects compare with
those of camels who wish to pass through the eye of the nee-
dle.” 

original) (quoting United States v. Cole-
man, 

, 1061 (7th Cir. 1999)).
     Therefore, “[w]hether the district court applied the Daub-
ert framework properly is a question we review de novo,” but
we review the ultimate decision “to exclude or admit the ex-
pert witness testimony for an abuse of discretion only.” 

. In other words, “[i]f the district court properly
applied the Daubert analysis, then ‘we will not disturb the dis-
trict court’s findings unless they are manifestly erroneous.’”
United States v. Adame, 

, 645 (7th Cir. 2016) (quot-
ing Lapsley v. Xtek, Inc., 

, 809 (7th Cir. 2012)). But
“if the district court failed to conduct a Daubert analysis, then
we review de novo whether the expert’s testimony was admis-
sible under Federal Rule of Evidence 702.” 

is an abuse of discretion only if ‘no reasonable
person would agree with the decision made by the trial
court.’” Smith v. Hunt, 

, 807–08 (7th Cir. 2013)
(quoting United States v. Thomas, 

, 845 (7th Cir.
2006)). This can occur when a court “commits ‘a serious error
of judgment, such as reliance on a forbidden factor or failure
to consider an essential factor,’” Ford Motor 

(quoting Powell v. AT&T Comm., Inc., 

, 825 (7th
Cir. 1991)), when a court treats a single Daubert factor as dis-
16                                                    No. 17-1810

positive, see 

or when “the record contains no ev-
idence upon which the trial judge rationally could have based
his decision.” United States v. Savage, 

, 760 (7th Cir.
2007).
    Notably, “[t]he party seeking to introduce the expert wit-
ness testimony bears the burden of demonstrating that the ex-
pert witness testimony satisfies the [Daubert] standard by a
preponderance of the evidence.” 

; see also
Lewis v. CITGO Petroleum Corp., 

, 705 (7th Cir.
2009); Fed. R. Evid. 702 advisory committee’s note to 2000
amendment (“[T]he admissibility of all expert testimony is
governed by the principles of Rule 104(a). Under that Rule,
the proponent has the burden of establishing that the perti-
nent admissibility requirements are met by a preponderance
of the evidence.”).
     B. The district court applied the proper legal standard
        when it barred the testimony of plaintiffs’ experts
    To apply the proper legal standard, “judges merely need
to follow Daubert in making a Rule 702 determination.” Naeem
v. McKesson Drug Co., 

, 608 (7th Cir. 2006). “While
the Daubert standard does not have to be recited mechani-
cally, ‘it is nonetheless crucial that a Daubert analysis of some
form in fact be performed.’” 

421 F.3d at
535).
   “[T]he court must provide more than just conclusory state-
ments of admissibility or inadmissibility to show that it ade-
quately performed its gatekeeping function.” Gayton v.
McCoy, 

, 616 (7th Cir. 2010). “A more searching
Daubert analysis is required.” 

. In
No. 17-1810                                                     17

Metavante Corp. v. Emigrant Savings Bank, for example, we con-
cluded that the district court did not properly adhere to the
Daubert framework where the judge issued a one sentence ad-
missibility determination that did not even reference Daubert
by name. 

, 760 (7th Cir. 2010). We issued the same
holding in Naeem, where we noted “the district court’s one
sentence, stating that [the expert] ha[d] sufficient expertise,
[was] not enough to show that the district court applied the
Daubert 

. Likewise, in Fuesting,
“[a]lthough the district court offered a relatively lengthy dis-
cussion of [the expert’s] credentials, the court’s Daubert factor
analysis,” which merely made a passing reference to the ex-
pert’s deposition, “was not 

.
    In contrast, we have found that the district court applied
the proper legal standard when it “accurately outlin[ed] the
Daubert framework” at the outset of its analysis and “con-
ducted an in-depth review of the relevant studies the experts
relied upon.” See 

.
    Here, the district court applied the proper legal standard.
The court explicitly recognized that “[t]he admissibility of ex-
pert testimony is governed by Federal Rule of Evidence 702
and Daubert.” Gopalratnam, 

, at *4. It prefaced
its Daubert analysis with a two-and-a-half-page discussion of
the applicable test, which highlighted Daubert’s dual focus on
relevance and reliability, including the most commonly uti-
lized reliability factors. The court’s application of Daubert’s re-
liability standard to the facts of this case continued for another
six pages. All told, such an inquiry stands in stark contrast to
cases like Metavante, Naeem, and Fuesting. Thus, we will apply
an abuse of discretion standard to our review of the court’s
18                                                   No. 17-1810

ultimate determination to exclude Doughty and Hill’s testi-
mony.
     C. The district court did not abuse its discretion in ex-
        cluding the expert witness testimony of plaintiffs’ ex-
        perts
        1. The Testimony of Dr. Daniel Doughty
   Based upon his written report and deposition testimony,
we can fairly characterize Doughty’s expert opinions as fol-
lows:
        (1) An “internal fault” in one of the battery cells
        (Cell A) in Arun’s computer caused an internal
        short circuit that triggered thermal runaway
        such that the cell ejected its internal contents,
        which created a “potent fire hazard”; and
        (2) The internal fault was caused “by either a
        manufacturing defect in the cell” or “a failure of
        the computer’s control/safety circuitry.”
Said another way, Doughty rendered expert opinions regard-
ing both the cause of the fire (an internal fault in Cell A) and
the cause of the internal fault itself (a manufacturing defect in
the cell or failure of the computer’s control/safety circuitry).
    Pursuant to Daubert, it was “not the trial court’s role to de-
cide” whether these opinions were ultimately correct. See Ford
Motor 

. Rather, the court was limited to de-
termining “whether the methodology underlying that testi-
mony [was] sound.” See 

score, the district court
ruled that Doughty’s opinions were unreliable because their
“underlying bases” were improperly based upon “specula-
tion” and “unfounded inferences.” Gopalratnam, 2017 WL
No. 17-1810                                                                19

1067768, at *5, *10. We conclude that this ruling was not an
abuse of discretion, and address each of Doughty’s opinions
in turn.
            a. Doughty’s opinion that an “internal fault” in one of
               the battery cells caused the fire
   During his deposition, Doughty explained that his opin-
ion that an internal fault led to the failure of Cell A was based
upon three facts:
        (1) Only Cell A ejected its internal contents,
        whereas Cells B and C, as well as the cell phone
        battery, did not;
        (2) Cell A warped into an elliptical shape,
        whereas Cells B and C, as well as the cell phone
        battery, retained their dimensions; and
        (3) Cell A acted as a projectile.
    Central to the relevancy of Doughty’s first two factual un-
derpinnings were their differential nature. That is, what mat-
tered (at least to Doughty) was not just that Cell A ejected its
contents and warped into an elliptical shape, but also that the
other battery cells did not. 2 Admittedly, these findings were
not speculative; they are well supported by the physical evi-
dence recovered from plaintiffs’ home. Indeed, based upon


    2  For example, Doughty’s report states “[o]nly Cell A had ejected its
contents, whereas all the other cells (Cell B, Cell C or the Nokia cell phone
battery) did not eject their contents,” and “[t]he [CT scan of the cells taken
after the fire] shows the distortion of the Cell A can, which now has an
elliptical shape instead of being circular. The external cell dimensions of
Cells B & C appear to have no distortion, and dimensions appear nearly
the same as as-manufactured cells.”
20                                                  No. 17-1810

the present record, it does not appear that either fact is dis-
puted by the parties.
    Nevertheless, Doughty’s reliability fails when it comes to
the method by which he derived conclusions from these un-
derlying events. In his expert report, Doughty determined
that the cells’ differential appearance suggested that one
cell—Cell A—“had a different failure mechanism.” This infer-
ence, however, relied almost entirely upon Doughty’s prem-
ise that the exposure of cells to external fire (as opposed to an
internal fault) “cause[s] predictable results” amongst the cells.
Doughty reasoned that, during the fire in plaintiffs’ home, all
of the cells would have been exposed to the same external
heat conditions. Thus, according to Doughty, “one would ex-
pect to see relatively uniform temperature response of all
three cells.” By extension, because the condition of one cell
was “clearly different” than the other cells, Doughty con-
cluded that the “temperature history of [that cell] was
unique.” Having previously excluded other potential causes
(such as electrical and mechanical abuse), “the only trigger …
that remain[ed was] an internal flaw in the cell that [led] to
rapid thermal runaway.” (emphasis added).
   However, the record indicates that Doughty’s central un-
derlying premise—that exposure of cells to external fire
causes predictable, uniform results—was not only unsup-
ported, but in fact contrary to generally accepted battery sci-
ence. Defendants’ own battery expert testified to the follow-
ing:
       Q. Why was the thermal runaway of Cell A
       more energetic given that it should have been at
       the same or similar charge as Cell B and C?
No. 17-1810                                                                    21

         …
         A. … [W]hen you expose cells to a fire there’s a
         certain amount of randomness or stochasticity
         to the failure event … . [T]here is a certain
         amount of randomness to how batteries fail
         when they’re exposed to fire, and especially an
         uncontrolled event [such as the fire at issue
         here] … . So I think that, you know, you’ve got
         the stochasticity of how lithium-ion batteries
         fail, you’ve got the, the unknown variables as
         far as how the fire encroached on the laptop,
         which cells got heated up first. So I mean, when
         you ask me how to explain it, I explain it by say-
         ing that there’s a lot of randomness and differ-
         ences other than state of charge that are going to
         affect whether a cell expels its contents or does
         not.
   In his written report, Doughty supported his premise only
by citing to his own 2004 article, which he co-authored with
E. Peter Roth, where he tested the thermal response of lith-
ium-ion cells to external heat. Defendants’ expert, however,
challenged the compatibility of Roth’s findings with
Doughty’s premise:
         [I]n fact, in the Roth report that we were just
         talking about … they took two identical 18650
         cells, 3 they exposed them to exactly the same
         heat source and exactly the same state of charge,
         and one expelled its contents and one did not.
         They did the same thing at 80 percent state of

   3   The laptop battery cells at issue in this case were also 18650 cells.
22                                                      No. 17-1810

       charge, one expelled its contents, one didn’t … .
       These are laboratory tests, so where things are
       done as best as possible to keep everything the
       same. And you still see a difference on how
       things fail.
  This observation is confirmed by the Roth article itself,
which states:
       Two cells at 100% [state-of-charge] and two cells
       at 60% [state-of-charge] were run in heating
       tests to 200°C. In these tests, two cells vented, sav-
       ing the respective cans, and two cells did not, causing
       the can lid to fail and the roll to eject. There was a
       result each way, at each [state-of-charge]. It appears
       that in instances where there was can, or seal, or
       vent damage which could cause the cell to leak
       prematurely, then the cell did leak prematurely,
       and the cell was eventually observed to vent
       through the machined vent, and the can was
       saved from overpressure. In the other two, nor-
       mal cases where there was no premature leak-
       age, the machined vent never opened, and the
       case and roll were destroyed by overpressure.
E. Peter Roth et al., Sandia Nat’l Labs., Advanced Technology
Development Program for Lithium-Ion Batteries: Thermal Abuse
Performance of 18650 Li-Ion Cells 35 (2004), https://pdfs.seman-
ticscholar.org/7c66/2ecf8d3c4830c84283f225504e5b2f454ba8.p
df (emphasis added). Doughty’s reliance on the Roth study,
therefore, is misplaced.
  During his deposition, Doughty expanded his list of aca-
demic support beyond the Roth article to include a 2011 study
No. 17-1810                                                   23

conducted by Celina Mikolajczak and others at Exponent Fail-
ure Analysis Associates, that also tested the response by lith-
ium-ion cells to external fire:
       Q. All right. Is there any other literature that you
       can identify by actual name that someone could
       go look at and review that also supports that
       other than the paper you co-wrote [with Roth]
       in 2004?
       A. Celina’s report has a number of tests, fire
       tests of batteries intended to evaluate shipping
       conditions, and she has—they have a number of
       tests at Exponent.
       Q. And you’re saying that that report, if I go
       read it, and you’re referring to, “Lithium-Ion
       Batteries Hazard and Use Assessment,” will in-
       clude in it the proposition that lithium-ion bat-
       tery cells that are exposed to the same condi-
       tions in a fire should behave the same way?
       A. That is—that—the consistency of her results
       points to that conclusion. I don’t know that she
       specifically mentions that conclusion, but the
       consistency of the results will point to that con-
       clusion.
As with the Roth study, however, Mikolajczak’s study indi-
cates the opposite. Mikolajczak’s article includes statements
such as:
       [1] In a few tests, some cells ruptured their cases
       and expelled their contents. … [and]
24                                                   No. 17-1810

       [2] Tests were also conducted on laptop battery
       packs that contained 18650 cells. In the first test,
       a single pack containing eight cells was tested.
       In this test, the packaged plastic began to be
       consumed by the propanol flame. Eventually,
       the cells began to vent with flames. Ultimately,
       some of the cells ruptured, ejecting and dispers-
       ing their contents.”
Celina Mikolajczak et al., The Fire Prot. Research Found., Lith-
ium-Ion Batteries Hazard and Use Assessment 94–96 (2011),
http://www.prba.org/wp-content/uploads/Exponent_Re-
port_for_NFPA_-_20111.pdf (emphasis added).
    Finally, during his deposition, Doughty also referenced an
Underwriters Laboratories (“UL”) safety standard for lith-
ium-ion cells used in laptop computers. Part of this standard
includes a “projectile” test where a cell is positioned on a
metal screen over an open flame and heated “until it explodes
or the cell or battery has ignited and burned out.” Underwrit-
ers Labs. Inc., Standard for Safety: Lithium Batteries 20–21
(5th             ed.            2012),          http://www.by-
choice.com/UL_1642_Ed_5_2012.pdf. To pass the test, no part
of an exploding cell’s contents may “penetrate the wire screen
such that some or all of the cell or battery protrudes through
the screen.” 

Doughty testified that the “good uni-
formity” of the results of such tests supported his theory that
cells react uniformly when exposed to external flame.
     Doughty, however, failed to provide any details regarding
the UL test results. Regardless, the very premise of the projec-
tile test undermines Doughty’s assertion, since the test explic-
itly contemplates that, in reaction to external heat, a subject
battery may either “explode” or “ignite and burn out.” This
No. 17-1810                                                   25

presents disparate, not uniform outcomes. In fact, such results
are remarkably similar to the facts presented here, where Cell
A expelled its internal contents while Cells B and C merely
ignited. Thus, Doughty’s reliance upon the UL projectile test
is inapposite.
    Contrary to Doughty’s assertion, therefore, the Roth,
Mikolajczak, and UL tests stand for the countervailing prop-
osition that individual cells can react differently to external
fire; some may expel their contents, while others may not. The
central premise underlying Doughty’s conclusion is thus
faulty. Without this premise, Doughty cannot reliably draw
the inference that the differential appearance between the
cells in Arun’s laptop was caused by a “different failure mech-
anism.” In turn, Doughty’s ultimate conclusion—that an in-
ternal flaw created the fire hazard that led to Arun’s death—
collapses.
    Of course, Doughty was not bound to merely rely upon
the Roth, Mikolajczak, and UL standards as sources; he could
have cited to any reliable basis for his theory. He chose, how-
ever, to cite only contrary sources, and he did not conduct his
own independent testing despite acknowledging the feasibil-
ity of doing so. In short, Doughty failed to “bridge the analyt-
ical gap” between the accepted differential appearance of the
laptop battery cells and his contested conclusion that such dif-
ferential appearance was caused by an internal fault in Cell A.
See Fuesting v. Zimmer, Inc., 362 F. App’x 560, 563 (7th Cir.
2010). That is, on the present record, there is no “rational con-
nection” between Doughty’s data and his opinion. See Man-

; see also United States v. Moore, 

, 684 (7th Cir. 2008) (asking the question: “Are [the ex-
pert’s] inferential methods reliable?”); Korte v. Exxonmobil
26                                                    No. 17-1810

Coal USA, Inc., 164 F. App’x 553, 557 (7th Cir. 2006) (affirming
the district court’s exclusion of expert causation testimony
where the expert “formed his opinion without sufficient sci-
entific evidence confirming the validity of [his] premise”);
Baker v. Dalkon Shield Claimants Tr., 

, 253 (1st Cir.
1998) (indicating that “scientific premises” may be “so faulty
that [they cannot] even be tendered to the jury for its consid-
eration”).
    Equally problematic was the third factual basis for
Doughty’s “internal fault” opinion: that Cell A acted as a pro-
jectile. According to Doughty, the fact that Cell A was pro-
pelled away from the laptop served as further evidence of a
“substantial pressure rise” and “rapid” release of stored en-
ergy that, in his view, was only consistent with an internal
short circuit.
    Unlike the differential appearance of cells after the fire, the
issue of whether Cell A acted as a projectile was hotly con-
tested by defendants. Doughty testified during his deposition
that his finding was primarily based upon the fact that Cell A
was found apart from the other laptop remnants:
       It’s my understanding that Cell A was found
       distant from the computer in the room, and
       therefore, if it had explosively disassembled, it’s
       not uncommon to have the remnants of the cells
       be projectiles across the room … .
However, although it is true that Cell A was ultimately found
in one of the debris piles outside plaintiffs’ home, no one, in-
cluding Doughty, could establish precisely how it got there,
nor where it was located prior to cleanup. As a result, at least
two possible alternatives cut against Doughty’s theory. It is
No. 17-1810                                                     27

entirely possible that Cell A was actually located near the
other electronic remnants, but was simply missed by investi-
gators and subsequently discarded with the remaining fire
debris. It is equally possible that the cell originally came to
rest away from the bed, but was moved there by fire suppres-
sion efforts (such as a running fire hose) rather than an explo-
sive internal fault. Either alternative (or others) would under-
mine Doughty’s finding that Cell A acted as a projectile, which
would in turn further weaken his ultimate conclusion that an
internal fault led to its failure. Of course, the actual explana-
tion of how Cell A came to rest in the debris pile is irrelevant
to our Daubert analysis. What matters is that Doughty failed
to account for other possible explanations in arriving at his
conclusion. See Fed. R. Evid. 702 advisory committee’s note to
2000 amendment (finding “[w]hether the expert has ade-
quately accounted for obvious alternative explanations” to be
relevant “in determining whether expert testimony is suffi-
ciently reliable to be considered by the trier of fact”).
           b. Doughty’s opinion that the internal fault was caused
              “by either a manufacturing defect in the cell” or “a
              failure of the computer’s control/safety circuitry”
    Even less reliable was Doughty’s second opinion that the
internal fault in Cell A was specifically caused by a manufac-
turing defect in the cell or a failure in the computer’s electrical
circuitry. During his deposition, Doughty acknowledged that,
even though electronic manufacturing processes undergo
continuous development, he has never been to a laptop man-
ufacturing plant, and has not visited a lithium-ion battery
manufacturing facility since 2003 or 2004. He further admitted
that, aside from what he read in the depositions of defend-
28                                                 No. 17-1810

ants’ employees, he had no independent knowledge of de-
fendants’ individual manufacturing processes, including
their quality control measures. Indeed, he could not even
identify the specific plant where the battery cells in this case
were originally manufactured. In short, despite concluding
that a manufacturing defect led to the alleged internal fault in
this case, Doughty could not provide details as to what the
specific defect was; why it transpired; when it occurred in the
manufacturing process; or even where such manufacturing
took place. Rather, Doughty simply opined that, because sev-
eral manufacturing processes “can cause” an internal short
circuit, such must have occurred here. This is simply too spec-
ulative to pass muster under Daubert and Rule 702. See 

the expert has unjustifiably extrapolated
from an accepted premise to an unfounded conclusion” to be
relevant “in determining whether expert testimony is suffi-
ciently reliable to be considered by the trier of fact”).
    Plaintiffs’ fifty-three page brief attempts to explain away
Doughty’s various shortcomings. Here, however, it is im-
portant to remember the two key elements of our standard of
review. First, having found that the district court properly ap-
plied the Daubert framework, we review its ultimate exclusion
of Doughty’s testimony only for an abuse of discretion. 

. That is, the question is not whether we would
have admitted Doughty’s testimony in the first instance; the
relevant inquiry is whether any “reasonable person would
agree with the decision made by the trial court” 

(quoting 

). Second, it is the
party that proffers expert witness testimony that bears the
burden of demonstrating by a preponderance of the evidence
that such testimony is sufficiently reliable. 

. Given the myriad issues with Doughty’s testimony, the
No. 17-1810                                                   29

district court was not manifestly erroneous in determining
that plaintiffs failed to satisfy this burden.
       2. The Testimony of Michael Hill, Sr.
   Based upon his written report and deposition testimony,
we can fairly characterize Hill’s expert opinions as follows:
       (1) The fire originated on the top of the bed in
       the basement bedroom of plaintiffs’ home;
       (2) The “most probable” cause of the fire was the
       laptop battery; and
       (3) The fire was accidental.
    The district court’s ruling can be affirmed solely by exam-
ining Hill’s second opinion regarding the cause of the fire. In
arriving at this conclusion, Hill proceeded by the process of
elimination. After excluding the possibility that the fire was
caused by the home’s electrical circuitry, the ceiling light fix-
ture located above the bed, a printer found along the bed-
room’s western wall, or Arun’s cell phone, he came to rest
upon Arun’s laptop battery, which he concluded was a viable
fire source. Hill stated that “[l]iterature available at the time
of [his] report indicate[d that] failures of lithium ion battery
cells eject molten metals and flaming electrolyte, thereby pos-
ing fire hazards.” He further noted that one of the laptop’s
battery cells was “damaged consistent with an internal failure
which would be a competent ignition source for this fire.” Fi-
nally, Hill suggested that “[t]here [were] indications thermal
runaway occurred within one of the lithium ion battery cells
of [the] laptop, providing sufficient heat to ignite bedding ma-
terials on the bed.”
30                                                 No. 17-1810

    During his deposition, however, Hill acknowledged that
he was not an electronics or battery expert. Instead, those por-
tions of his report that discussed the laptop batteries as a
cause of the fire relied entirely upon the work of Doughty:
       Q. Are you—do you have any expertise or
       knowledge whereby you could render an opin-
       ion regarding the appearance of a lithium-ion
       battery cell and whether its appearance is con-
       sistent or inconsistent with an internal failure
       which would be a competent ignition source?
       A. No.
       Q. Okay. From where did you get that statement
       from?
       A. I believe that was … in discussion with Mr.
       Doughty.
       Q. Okay. So to the extent that information is in
       your report, that would be information you got
       from Mr. Doughty?
       A. I believe so, yes, sir.
       …
       Q. There it says, the next sentence says, “There
       are indications thermal runaway occurred
       within one of the lithium-ion battery cells of a
       laptop, providing sufficient heat to ignite bed-
       ding materials on the bed.” Do you see that?
       A. Yes, sir.
       Q. Where is that information from?
       A. Again, Mr. Doughty.
No. 17-1810                                                    31

    Of course, as a general matter, there is nothing objectiona-
ble about an expert relying upon the work a colleague. See
Dura Auto. Sys. of Ind., Inc. v. CTS Corp., 

, 613 (7th
Cir. 2002) (“[I]t is common in technical fields for an expert to
base an opinion in part on what a different expert believes on
the basis of expert knowledge not possessed by the first ex-
pert … .”); see also Carnegie Mellon Univ. v. Marvell Tech. Grp.,
Ltd., 

, 1303 (Fed. Cir. 2015) (“For areas outside
her expertise … the district court properly concluded that [the
expert] could, indeed must, rely upon … other experts having
such industry-specific experience.”). Such a scenario is explic-
itly contemplated by the Rules of Evidence. See Fed. R. Evid.
703 (“An expert may base an opinion on facts or data in the
case that the expert has been made aware of or personally ob-
served.” (emphasis added)).
    However, “[a]n expert is not entitled to testify to opinions
that rely on the opinion of another expert, simply because the
other is an expert.” Mooring Capital Fund, LLC v. Knight, 388 F.
App’x 814, 820 (10th Cir. 2010). Under Rule 703, the facts or
data relied upon must themselves be the kind that “experts in
the particular field would reasonably rely on … in forming an
opinion on the subject.” Fed. R. Evid. 703. As 

,
Doughty’s own expert opinions were not sufficiently reliable
to pass Rule 702 strictures. Thus, plaintiffs cannot reasonably
admit through Hill what they could not offer through
Doughty. As the Fifth Circuit stated in Sims v. Kia Motors of
America, Inc.:
       The district court’s order ... indicates that its ex-
       clusion [of Expert #2’s testimony] was, at least
       in part, due to [Expert #2’s] dependence on [Ex-
       pert #1’s] theory … . Since the court had already
32                                                   No. 17-1810

       deemed [Expert #1’s] testimony inadmissible, it
       concluded that [Expert #2’s] testimony … was
       also inadmissible. … [T]he district court
       properly excluded [Expert #2’s] theory … be-
       cause it relied on [Expert #1’s] inadmissible …
       theory.

, 404-05 (5th Cir. 2016); see also Zimmer, 362 F.
App’x at 564 (“[B]ecause [Expert #2’s] testimony on causation
primarily relies on an excluded expert opinion … , the district
court did not err in excluding it.”); Tajonera v. Black Elk Energy
Offshore Operations, LLC, No. 13-0366, 

, at *11
(E.D. La. June 7, 2016) (“[A]lthough Rule 703 undoubtedly al-
lows experts to testify on the basis of facts or data that ‘need
not be admissible’ in and of themselves, Rule 703 does not
necessarily allow a witness to rely on the methodology of an-
other expert, if that expert’s methodology would be deemed
unreliable under Daubert.” (quoting Fed. R. Evid. 703)). As
our previous analysis states, the district court did not abuse
its discretion by excluding Doughty’s testimony. Thus, we
find it also did not abuse its discretion when it excluded Hill’s
testimony regarding the cause of the fire, as his conclusions
were based on Doughty’s unreliable methodology.
    Once Hill’s causation theory is precluded, his remaining
opinions—regarding the fire’s place of origin and the extent
to which it can be classified as “accidental”—are rendered
moot. We concur with the district court’s finding that, because
the “inner workings of a laptop and its components, including
the battery pack and its cells, are highly technical,” expert tes-
timony is necessary for plaintiffs to prevail on their claims.
Gopalratnam, 

, at *3. Thus, without Doughty
and Hill’s causation theories, plaintiffs cannot prove that one
No. 17-1810                                                  33

of defendants’ products caused the fire. Therefore, summary
judgment in favor of defendants was appropriate. See Zimmer,
362 F. App’x at 564 (“Given that all causation testimony has
been excluded, [the plaintiff’s] strict liability and negligence
claims necessarily fail, and summary judgment in favor of
[the defendant] is appropriate.”).
                         III. Conclusion
    For the foregoing reasons, we AFFIRM the judgment of the
district court.