In the

    United States Court of Appeals
                 For the Seventh Circuit
                     ____________________
No. 17-3212
VARLEN CORPORATION,
                                                  Plaintiff-Appellant,
                                 v.

LIBERTY MUTUAL INSURANCE COMPANY,
                                                 Defendant-Appellee.
                     ____________________

         Appeal from the United States District Court for the
           Northern District of Illinois, Eastern Division.
           No. 1:13-cv-05463 — Joan B. Gottschall, Judge.
                     ____________________

     ARGUED DECEMBER 3, 2018 — DECIDED MAY 16, 2019
                ____________________

   Before SYKES, BARRETT, and ST. EVE, Circuit Judges.
   BARRETT, Circuit Judge. Varlen Corporation owned and op-
erated two industrial sites that were found to have significant
amounts of groundwater contamination related to the sites’
operations. When its insurer, Liberty Mutual Insurance Com-
pany, refused to indemnify it, Varlen sued. Varlen’s case
turned on testimony from an expert witness, who was ex-
cluded by the district court because he didn’t use reliable
2                                                   No. 17-3212

methods. We agree with the district court’s exclusion and af-
firm its grant of summary judgment to Liberty Mutual.
                               I.
    Varlen, an Illinois corporation, owned and operated two
industrial facilities related to railroad operations during the
time period relevant to this appeal. At the first, which the par-
ties call the LASI site, Varlen performed operations such as
plating parts for locomotive engines in chrome. At the second,
the Silvis site, Varlen’s operations included refueling diesel
engines. Varlen discovered contamination at both sites.
    The LASI site was equipped with a sump that held
wastewater from the chrome plater. When the water in the
sump reached a certain level, a pump would engage, pump-
ing the water to a holding tank. Varlen found a chemical
called hexavalent chromium contaminating the area around
the sump.
    At the Silvis site, Varlen discovered two types of ground-
water contamination. It found a chlorinated solvent by a tank
into which metal parts were dipped to degrease them. It also
found diesel fuel near a large tank where locomotives would
refuel.
   The contamination at these two sites cost Varlen millions
of dollars in damages and remediation expenses. Varlen
sought indemnification from its insurer, Liberty Mutual. But
Varlen’s policy with Liberty Mutual had an exclusion for any
property damage arising out of chemical leaks or discharges,
and Liberty Mutual denied coverage on this ground.
   Varlen sued Liberty Mutual. To overcome the pollution
exclusion, it pointed to a policy provision stating that, despite
the exclusion, Liberty Mutual would cover chemical leaks or
No. 17-3212                                                      3

discharges that were “sudden and accidental.” Lacking direct
evidence of how the damage occurred, it proffered the expert
testimony of geologist Daniel Rogers to prove that the con-
tamination of the LASI and Silvis sites occurred suddenly and
accidentally.
   Rogers testified that the contaminants at the LASI site
were released because the concrete sump leaked. He opined
that the releases were “sudden and accidental” because they
were not intended and occurred in sudden spurts each time
that the sump failed. When asked about his basis for these
opinions, he explained that he had experience working with
sumps and had personal knowledge of these sumps in partic-
ular.
    Rogers also testified that the releases at the Silvis site were
likely “sudden and accidental.” There are two relevant zones
at the Silvis site: the area around the diesel refueling station
and the area where the chlorinated solvents were stored. Rog-
ers asserted that the contamination around the diesel refuel-
ing area was too large to have occurred by minor leakage. In-
stead, he testified that the contamination was “consistent with
overfills of diesel locomotives” and suggested that “tens of
gallons to hundreds of gallons [] would have been released
before it was noticed.” He also said that the value of the fuel
made it unlikely that such a fuel spill would have occurred
intentionally. Turning to the contamination at the chlorinated
solvent storing area, Rogers surmised that it was “indicative
of a drum overturning and suddenly leaking out rather than
from operations.” He based this opinion in part on the fact
that the contamination was found around where the solvent
was stored, not where it was used.
4                                                            No. 17-3212

   Both parties moved for summary judgment. Liberty Mu-
tual also moved to strike Rogers’s testimony. The district
court granted the motion to strike, holding that Rogers’s opin-
ions were unreliable and speculative under Federal Rule of
Evidence 702. It then granted Liberty Mutual’s motion for
summary judgment. Varlen appealed.
                                    II.
   Liberty Mutual is entitled to summary judgment against
Varlen if no reasonable jury could find that the releases were
“sudden and accidental” at either the LASI or Silvis sites. See
FED. R. CIV. P. 56(a); Anderson v. Liberty Lobby, Inc., 

, 255–56 (1986). And Liberty Mutual correctly asserts that
Varlen has pointed to no admissible evidence that would per-
mit a reasonable jury to make that finding.
    Rogers’s expert testimony is the only evidence that Varlen
offered as to whether the contamination occurred in a sudden
and accidental fashion.1 But before Rogers’s expert testimony
can be admitted, it must be deemed reliable under Rule 702 of
the Federal Rules of Evidence, which tracks the Supreme
Court’s opinion in Daubert v. Merrell Dow Pharmaceuticals, Inc.,

(1993). The court must decide that the witness is
“qualified as an expert by knowledge, skill, experience, train-
ing, or education”; the testimony will “help the trier of fact to
understand the evidence or to determine a fact in issue”; “the

    1 In its order granting summary judgment against Varlen, the district
court noted that Varlen “relie[d] exclusively on Rogers’ expert opinions”
to “ward off Liberty Mutual’s request for summary judgment under the
[pollution] exclusion.” On appeal, Varlen makes a passing attempt at re-
butting the district court’s statement, offering a few conclusory claims and
theories—none of which are enough to lead us to disagree with the district
court’s assessment. Varlen’s case rises or falls on Rogers’s testimony.
No. 17-3212                                                     5

testimony is based on sufficient facts or data” and “reliable
principles and methods”; and the expert has “reliably applied
the principles and methods to the facts of the case.” FED. R.
EVID. 702. An expert’s proponent has the burden of establish-
ing the admissibility of the opinions by a preponderance of
the evidence. See Lewis v. CITGO Petroleum Corp., 

,
705 (7th Cir. 2009).
    The district court determined that Rogers’s testimony did
not meet these requirements because it was not based on reli-
able methods or principles. It did not abuse its discretion in
reaching that conclusion. See Brown v. Burlington N. Santa Fe
Ry. Co., 

, 772 (7th Cir. 2014) (“If the court properly
[followed the Daubert framework], we then review its ulti-
mate decision to exclude expert testimony for an abuse of dis-
cretion.”).
     In Rogers’s report and testimony, he asserted that the dis-
charges at the LASI site must have been “sudden and acci-
dental.” He suspected that the contamination occurred in con-
nection with a failure in the sump pump in the 1970s. He
claimed to base this on his experience with sumps, his site vis-
its, and his knowledge of the sites’ operations. This type of
evidence is not necessarily unreliable. See Metavante Corp. v.
Emigrant Sav. Bank, 

, 761 (7th Cir. 2010) (“An ex-
pert’s testimony is not unreliable simply because it is founded
on his experience rather than on data.”). But Rogers still
needed to show how his experience or expertise led to his con-
clusions.
   Rogers attempted to base his conclusions on inferences
from qualities of the “plume” of contamination. That is, he
looked at the size and scope of the contamination and worked
backward to surmise how it must have occurred. For instance,
6                                                   No. 17-3212

Rogers testified that at the LASI site, the contaminant mass
must have been very large to create a plume with the size and
concentration of the one at that site. He concluded that the
contamination must have therefore been sudden and acci-
dental.
   But Rogers failed to explain why this data mattered or
why his inferences were justified. When pressed specifically
on the connection between the contaminant mass and the cir-
cumstances of the release, he simply stated that the data was
“an indication” but acknowledged that it was “not conclu-
sive.” He opined that the sump itself had a “sudden and acci-
dental nature.” But it’s not clear what he meant, considering
that the sump is just a basin in the ground. Rogers did say in
passing that the contamination concentrations were “not uni-
form,” but he didn’t explain why that was significant.
    Rogers’s testimony about the Silvis site was equally lack-
ing. He said that the volume of the contamination at the fuel-
ing location was inconsistent with “minor leakage” and
opined instead that it may have resulted from overfills of die-
sel locomotives. As to the chlorinated solvent storing area,
Rogers suggested that perhaps a drum got punctured and
caused the contamination. But Rogers offered no methodol-
ogy to explain how he drew those conclusions.
    In short, Rogers failed to demonstrate that his conclusions
were anything more than guesses. To satisfy Daubert, Rogers
needed to provide an explanation of how the evidence led to
his conclusions. He had to articulate a justification for his in-
ference that the chemical spills were sudden and accidental
beyond a simple say-so. If Rogers made an argument based
on a reliable methodology, then Varlen has not pointed it out,
either to us or to the district court. And courts do not have to
No. 17-3212                                                                   7

scour the record or make a party’s argument for it. See D.Z. v.
Buell, 

, 756 (7th Cir. 2015). The district court did
not abuse its discretion in excluding Rogers’s testimony. And
without Rogers’s testimony and report, there is no issue of
material fact as to whether the contamination occurred sud-
denly and accidentally.2 See Celotex Corp. v. Catrett, 

, 322 (1986) (“Rule 56(c) mandates the entry of summary
judgment … against a party who fails to make a showing suf-
ficient to establish the existence of an element essential to that
party’s case, and on which that party will bear the burden of




    2 This is true regardless of whether Illinois or New York law applies.
In its opening brief, Varlen took no position on the choice-of-law issue,
instead arguing that it ought to prevail under either state’s law. In Liberty
Mutual’s response brief, it asserted that New York law applies. Varlen
then made the argument in its reply brief that Illinois law applies. Even if
Varlen has not forfeited this argument, the question is purely academic.
Under either law, Varlen bears the burden of proving that the exception
to the pollution exclusion applies. See Santa's Best Craft, LLC v. St. Paul Fire
& Marine Ins. Co., 

, 347 (7th Cir. 2010) (noting that under Illi-
nois law, “[i]nsurers have the burden of proving that an exclusion ap-
plies,” while insureds “have the burden to prove that an exception to an
exclusion restores coverage”); Mahl Bros. Oil Co. v. St. Paul Fire & Marine
Ins. Co., 

, 494 (W.D.N.Y. 2004) (explaining that under
New York law, “[o]nce an insurer satisfies its burden of proof that the
claims are within the pollution exclusion, the burden shifts to the insured
to demonstrate, either through a reasonable interpretation of the underly-
ing complaint or extrinsic evidence, that the discharge was in fact ‘sudden
and accidental’”). Varlen can’t carry its burden under either law without
Rogers’s testimony.
8                                                         No. 17-3212

proof at trial.”). We thus AFFIRM the district court’s grant of
summary judgment to Liberty Mutual.3




    3Varlen also suggests in passing that it might have a claim under a
theory of “wrongful entry … or other invasion of the right of private oc-
cupancy.” This theory is inadequately briefed and we do not consider it.