Van Natta v. Great Lakes Reinsurance(UK) SE

UNITED STATES DISTRICT COURT DISTRICT OF CONNECTICUT LIETTE VAN NATTA, et al., Plaintiffs, No. 3:18-cv-438 (SRU) v. GREAT LAKES REINSURANCE (UK) SE, Defendant. MEMORANDUM OF DECISION On March 16, 2016, Steven Van Natta (“Van Natta”) arrived at his second home, which was located at 58 Lebanon Road in Bethany, Connecticut (the “Property”). Van Natta found the Property completely ruined: Water and mold were everywhere. (I will refer to that damage as the “Loss.”) This case concerns an insurance dispute arising from the Loss. Van Natta and his mother1 (together, the “Plaintiffs”) sue Great Lakes Reinsurance (UK) SE, now known as Great Lakes Insurance SE (“Great Lakes”), for breach of an insurance contract (the “Policy”). Great Lakes has refused to provide insurance coverage for the Loss. After the Loss, the Plaintiffs paid out-of-pocket to gut the Property’s interior, sold the Property at a significant discount in September 2017, and filed this suit seeking damages for what the cost of reconstruction would have been (they did not actually perform the reconstruction). In May 2020, I granted in substantial part Great Lakes’ motion for summary judgment. In December 2020, I held a two-day bench trial on the surviving portion of the Plaintiffs’ claim. At trial, the Plaintiffs argued that they are entitled to $32,273.86 under the Policy. Great Lakes argued that the Plaintiffs were entitled to nothing because Van Natta had not taken reasonable 1 At all relevant times, Liette Van Natta resided in Florida and was not involved in the events at issue in this case. See Joint Trial Memorandum, Doc. No. 83-3, at ¶ 9. care to maintain heat at the Property. In the alternative, Great Lakes argued that, even if coverage was due under the Policy, the amount of damages totaled between $1,700 and $3,400. Because the Plaintiffs have not established that they took reasonable care to maintain heat at the Property, the Loss was not covered under the Policy. Thus, Great Lakes is not liable for breach of contract. Judgment shall enter in favor of Great Lakes. The following constitutes my findings of fact and conclusions of law. See Fed. R. Civ. P. 52(a). I. The Policy At the time of the Loss, the Plaintiffs were insureds on the Policy, a homeowners’ insurance policy issued by Great Lakes. See Joint Trial Memorandum, Doc. No. 83-3 (“JTM”), at ¶ 16; Policy, Trial Ex. 1. The Policy covered the Loss because the Loss was a “direct physical loss” to “the dwelling on the residence premises.” Policy, Trial Ex. 1; Trial Tr. (Day 2) at 4;2 JTM, Doc. No. 83-3, at ¶¶ 16–18. The Policy contained two relevant exclusions and one relevant exception. The first relevant exclusion was the “Mold Exclusion,” which read, in part: Notwithstanding any other provision in this Policy, there is no coverage . . . for any loss or damage involving in any way the actual or potential presence of mold, mildew or fungi of any kind whatsoever, whether or not directly or indirectly caused by or resulting from an insured peril. Policy, Trial Ex. 1. Much of my summary judgment ruling was concerned with interpreting the Mold Exclusion. See Ruling, Doc. No. 69, at 23–30; Van Natta v. Great Lakes Reinsurance (UK) SE, 462 F. Supp. 3d 113 , 129–33 (D. Conn. 2020). However, as explained further below, this memorandum of decision does not address the Mold Exclusion. The second relevant exclusion was the “Freezing Exclusion,” which explained that Great Lakes did not insure 2 Although the bench trial transcript is unavailable on this case’s public docket, in this decision I rely on a draft of that transcript. The draft of each day’s transcript re-starts page numbers at one. In addition, the draft contains no line numbers. Thus, my citations to the trial transcript take the form: Trial Tr. (Day [1 or 2]) at [page]. for loss . . . [c]aused by . . . [f]reezing of a plumbing, heating, air conditioning or automatic fire protective sprinkler system or of a household appliance, or by discharge, leakage or overflow from within the system or appliance caused by freezing. Policy, Trial Ex. 1. However, the Freezing Exclusion did not exclude coverage in two situations: if [the insureds] have used reasonable care to: (a) Maintain heat in the building; or (b) Shut off the water supply and drain all systems and appliances of water. Id. Only the first exception (the “Heat Exception”) is relevant here.3 II. Procedural Background On February 8, 2018, the Plaintiffs filed a complaint in Connecticut Superior Court. See Compl., Doc. No. 1-1, at 5. On March 14, Great Lakes removed the case to this court. See Notice of Removal, Doc. No. 1, at 1. Discovery concluded in late spring 2019, and, in August 2019, Great Lakes made a motion for summary judgment. See Mot. for Summ. J., Doc. No. 44. On October 9, 2019, this case, which had initially been assigned to District Judge Warren W. Eginton, was transferred to me. A. Summary Judgment Ruling After holding a hearing in January 2020, see Min. Entry, Doc. No. 68, in May 2020, I issued a ruling granting in substantial part and denying in part Great Lakes’ motion for summary judgment. See Ruling, Doc. No. 69; Van Natta v. Great Lakes Reinsurance (UK) SE, 462 F. Supp. 3d 113 (D. Conn. 2020). First, I held that the Freezing Exclusion applied to exclude coverage because “[t]here is no genuine dispute that the cause of the Loss was a freeze-up.” Ruling, Doc. No. 69, at 18. However, I also held that a genuine dispute of material fact existed regarding whether the Heat 3 During his last visit to the Property before the Loss, Van Natta neither shut off the water supply nor drained the pipes. See Trial Tr. (Day 1) at 24 (“Q. I’m sorry. For this specific time before the incident happened, did you shut the water off? A. I did not. Q. Did you drain the pipes? A. I never drained the pipes, no.”). Exception applied, which, if it did apply, would reinstate coverage under the Policy. See id. at 16, 18–22. I noted that my conclusion was based in large part on the fact that “one c[ould] . . . (weakly) infer from the evidence”—“[d]espite the strong contradictory evidence”—that Van Natta “was at the Property on February 11, 2016.” Id. at 22. Thus, I denied Great Lakes’ motion for summary judgment based on the Freezing Exclusion. Second, I held that the Mold Exclusion applied to bar most, but not all, coverage under the Policy. Although Great Lakes asked me to treat the Mold Exclusion as an anti-concurrent causation (“ACC”) clause, I declined to do so and, instead, chose to interpret the Mold Exclusion “according to its unambiguous, plain terms.” Id. at 24. Boiled down, the Mold Exclusion read: “There is no coverage for any loss or damage involving in any way the actual presence of mold.” Id. at 26 (cleaned up). I then examined the photographic and other evidence—room-by-room and area-by-area—to determine what parts of the Loss involved the presence of mold. See id. at 27–30. The only parts of the Loss that possibly did not involve mold—and so the only areas of the Property for which the Plaintiffs might be entitled to coverage under the Policy—were (1) the garage, (2) the storage area above the garage, (3) the crawl area through the master bedroom, and (4) the “office.” See id. at 30. Thus, I granted Great Lakes’ motion for summary judgment with respect to all portions of the Loss other than those four areas. Finally, I denied Great Lakes’ motion insofar as it claimed the Plaintiffs’ damages were too speculative to establish the “damage” element of a breach of contract claim. Id. at 30–37. B. Motion for Reconsideration One week after I issued my summary judgment ruling, the Plaintiffs made a motion for reconsideration. See Mot. for Reconsideration, Doc. No. 71. The Plaintiffs agreed with my conclusion that the Mold Exclusion was not an ACC clause, but argued that my decision to give effect to the unambiguous terms of the Mold Exclusion was error because I “overlooked, and thus did not apply, Connecticut’s Efficient Proximate Cause Test.” Id. at 4. In September 2020, I issued an order denying the Plaintiff’s motion for reconsideration because “it assume[d] that the only way to contract around the efficient proximate cause doctrine is through an ACC clause,” which, in my view, is an incorrect interpretation of the law. See Order, Doc. No. 75, at 9–13. C. Motion in Limine Following my ruling on the Plaintiffs’ motion for reconsideration, I set this matter down for a bench trial in December 2020. See Conf. Mem. and Order, Doc. No. 78. On November 11, Great Lakes submitted a motion in limine. See Mot. in Limine, Doc. No. 80. On December 2, I issued an order denying Great Lakes’ motion. See Order, Doc. No. 82. In its motion in limine, Great Lakes asked me to exclude evidence regarding two topics. First, Great Lakes requested that I preclude “the submission of evidence regarding property damages to [the] ‘office’ . . . because the damage to this area involved mold and therefore is precluded from coverage.” Mem. in Supp. Mot. in Limine, Doc. No. 80-17, at 2. Great Lakes argued that “based on the clear photographic evidence, it cannot be disputed that the mold was present in the office area of the Premises following the pipe burst.” Id. at 8. I denied Great Lakes’ request because Great Lakes was “simply ask[ing] me to determine the sufficiency of the evidence and to resolve a factual dispute,” which “is not the proper function of a motion in limine.” Order, Doc. No. 82 (citing, inter alia, Williams v. Rushmore Loan Mgmt. Servs. LLC, 2017 WL 822793 , at *1 (D. Conn. Mar. 2, 2017)). Second, Great Lakes requested that I preclude both expert reports prepared by Jim Maguire—the Plaintiffs’ expert on reconstruction costs—and “any testimony with regard to the opinions expressed therein” because “the Maguire Reports do not meet the standard of admissibility set forth in Federal Rule of Evidence 702.” Mem. in Supp. Mot. in Limine, Doc. No. 80-17, at 2, 8. More specifically, Great Lakes argued that Maguire’s “[e]xpert [r]eports, testimony and the opinions expressed [therein] must be excluded because Maguire provides no facts or data to support his conclusions or indicate that his opinions are based on reliable data or methodology.” Id. at 13. I denied without prejudice Great Lakes’ request. I explained that “Daubert and its progeny do not apply straightforwardly in the context of bench trials” because “in a bench trial, a party’s attempt to exclude an expert’s testimony under Daubert is tantamount to asking the Court to gate-keep expert testimony from itself.” Order, Doc. No. 82 (quoting Kortright Capital Partners LP v. Investcorp Inv. Advisers Ltd., 392 F. Supp. 3d 382, 397 (S.D.N.Y. 2019); 720 Lex Acquisition LLC v. Guess? Retail, Inc., 2014 WL 4184691 , at *10 (S.D.N.Y. Aug. 22, 2014)) (cleaned up). Although I denied Great Lakes’ motion in limine with respect to Maguire, I noted that Great Lakes could re-raise the objection at trial. See id. Great Lakes did not re-raise its objection to Maguire’s two expert reports: In fact, those reports were admitted as full exhibits by stipulation. See Maguire’s First Expert Report, Trial Ex. 9; Maguire’s Second Expert Report, Trial Ex. 8. However, at the bench trial, Great Lakes offered a separate objection to Maguire’s qualifications to testify regarding “the damages that were sustained at the home from water independent from mold.” Trial Tr. (Day 1) at 70. No matter whether that objection was new or renewed,4 I overruled it. See id.5 4 In my view, Great Lakes’ attorney mistakenly characterized his objection to Maguire’s testimony as a renewed objection. See Trial Tr. (Day 1) at 70 (“I’m renewing my objection to this particular witness’s qualifications as an expert.”). Great Lakes’ motion in limine did not argue that Maguire was not qualified to testify. 5 I ruled that “by his experience alone,” Maguire had “the qualifications that are necessary to testify in this case.” Trial Tr. (Day 1) at 71. Expert witnesses may be qualified by their “extensive practical experience.” McCullock v. H.B. Fuller Co., 61 F.3d 1038, 1043 (2d Cir. 1995); see also Fed. R. Civ. P. 702 (explaining that expert witnesses may be qualified by their “knowledge, skill, experience, training, or education”). Maguire testified that, for 40 years, he ran his own property damage, mitigation, and reconstruction business in Connecticut. See Trial Tr. (Day 1) at 61. In his 40 years in the business, Maguire estimated that his company had remediated about 200 to 300 homes per year that suffered water and mold damage. See id. at 69. As I explained, Great Lakes’ objection went to the weight I should afford Maguire’s testimony, not its admissibility. See id. at 70. As it turns out, I need D. Bench Trial In December 2020, I held a two-day bench trial. See Min. Entries, Doc. Nos. 91 and 92. At that bench trial, four witnesses testified: (1) Van Natta, (2) Maguire, (3) Brenda Badders, and (4) Thomas Mierzwa. See Witness List, Doc. No. 94. The parties also admitted 32 exhibits by stipulation. See Exhibit List, Doc. No. 93. The parties agreed that there were (up to) two issues to address at the bench trial: (1) Whether the Freezing Exclusion applied to bar coverage for the Loss, and, if not, (2) the amount of damages due. See JTM, Doc. No. 83-3, at ¶ 54. III. Standard of Review6 “An insurance policy is to be interpreted by the same general rules that govern the construction of any written contract and enforced in accordance with the real intent of the parties as expressed in the language employed in the policy.” Hammer v. Lumberman’s Mut. Cas. Co., 214 Conn. 573, 583 (1990) (quoting Schultz v. Hartford Fire Ins. Co., 213 Conn. 696, 702 (1990)) (cleaned up); see also Buell Indus., Inc. v. Greater New York Mut. Ins. Co., 259 Conn. 527, 538 (2002). “The determinative question is the intent of the parties, that is, what coverage the plaintiff expected to receive and what the defendant was to provide, as disclosed by the provisions of the policy.” Hammer, 214 Conn. at 583 (cleaned up). In an insurance policy, “[t]he burden of proving that an exclusion applies is on the insurer, but the insured has the burden of proving that an exception to an exclusion reinstates coverage.” Capstone Bldg. Corp. v. Am. Motorists Ins. Co., 308 Conn. 760 , 787–88 n.24 (2013) (citing Buell Indus., 259 Conn. at 551 ). IV. Facts not weigh Maguire’s testimony at all. Because I hold that the Loss is excluded from coverage under the Policy, the Plaintiffs are owed no damages, and so it is unnecessary to determine the correct amount of any damages, which was the subject of Maguire’s testimony. 6 The parties agree that Connecticut law governs this contractual dispute. See JTM, Doc. No. 83-3, at ¶ 1. A. The Discovery of the Loss and Subsequent Events On March 16, 2016, Van Natta discovered the Loss. See JTM, Doc. No. 83-3, at ¶ 23. When he arrived that day, Van Natta noticed that the windows were “fogged over” and “steamed up” and that the Property’s side door “was swollen” shut. Trial Tr. (Day 1) at 13. Van Natta went to the local police barracks; police officers came to the Property and kicked down the door so that Van Natta could enter. See id. When Van Natta walked in, he testified that the Property “looked like a hothouse,” “a greenhouse,” and a “steam bath.” Id. Water and mold were everywhere. See id. at 13–14; Van Natta’s Examination Under Oath, Trial Ex. 27 (“Van Natta’s EUO”), at 76:13–14, 77:3–10; Photographs, Trial Ex. 25 (five photographs taken by Van Natta on March 16). Great Lakes found out about the Loss on March 16 (the same day Van Natta did), when the Plaintiffs’ insurance broker sent Great Lakes’ managing general agent (Connecticut Underwriters) a fax reporting the Loss. JTM, Doc. No. 83-3, at ¶ 24; Trial Tr. (Day 2) at 8–9. That fax noted that the “kind of loss” was “Burst Pipe/Water Damage” and indicated that Van Natta had been “on vacation and a pipe burst.” JTM, Doc. No. 83-3, at ¶ 25. Connecticut Underwriters immediately retained a field adjuster, Brian Rollinson, to investigate the claim, inspect the Property, and interview Van Natta. See id. at ¶ 26. On March 18, Rollinson met with Van Natta, inspected the Property, and took dozens of photographs. See id. at ¶ 27. Also on March 18, Rollinson issued a report regarding the Loss. (I will refer to that report as the “Rollinson Report.”) The Rollinson Report described extensive water and mold damage to the Property’s interior. See Rollinson Report, Trial Ex. 19, at Bates 1536 (noting that “[t]he interior needs to be gutted” and “[a]ny contents inside the home are a total loss”). Rollinson also mentioned that the Property “shows a lack of regular maintenance.” Id. at Bates 1535. Rollinson wrote that the Loss occurred “due to a freeze up.” Id. Rollinson also memorialized several things that Van Natta told him. Van Natta apparently explained to Rollinson that he “call[ed] for oil when needed,” that “there was oil in the tank prior to the loss,” and that “there has been continuous electric service” at the Property. Id. at Bates 1536. However, Van Natta also told Rollinson that he “believe[d] that the power may have shut off while he was away because this has happened 2-3 times previously.” Id. On April 3, Van Natta completed a claim form questionnaire. See Claim Form, Trial Ex. 21. That claim form’s header read: “Insured Claim Form (Freeze-Up).” See id. In the claim form, Van Natta explained that he was “unsure” when he was last at the Property before March 16, but he estimated “early February.” Id. Van Natta confirmed that, while he was away, no one had checked on the Property’s interior. See id. At one point, the claim form asked: “How was heat left on at the site?” Van Natta answered: “Warm Winter / varied / usually between 40 + 60.” Id. Sometime in early May, Connecticut Underwriters referred the file to E&S Claims Management, which is a claims management firm that does work for Great Lakes. See Trial Tr. (Day 2) at 2. Rollinson, on Great Lakes’ behalf, then retained an engineer “to go out and inspect and determine the cause and source of the loss.” Id. at 10; JTM, Doc. No. 83-3, at ¶ 28. That engineer was Thomas Mierzwa.7 7 Mierzwa obtained a B.S. in civil engineering from the University of Massachusetts Amherst in 1998 and has been a licensed professional engineer in Connecticut since 2008. Trial Tr. (Day 2) at 34–35; see also Mierzwa Resume, Ex. A to Joint Trial Memorandum, Doc. No. 83-1. Mierzwa owns his own engineering firm—T.A.M. Engineering and Associates, Inc.—that “provid[es] forensic engineering services as well as design engineering services for . . . the insurance industry, as well as contractors, builders, owners, architects throughout the entire Northeast.” Trial Tr. (Day 2) at 34–35. At trial, I qualified Mierzwa as an expert in forensic engineering “for the purposes of determining the cause of the loss and the extent of the damage, the presence of mold and whether that can be separated – designated separately as water damage.” Id. at 37–38. On May 16, Mierzwa inspected the Property with a view towards determining the cause of the Loss. That day, Mierzwa spoke with Van Natta, inspected the Property, and took numerous photographs. See JTM, Doc. No. 83-3, at ¶ 29; see also Photographs, Trial Exs. 3–5, 17. In the following weeks, Mierzwa also reviewed relevant electricity bills, oil delivery tickets, thermostat installation guides, and weather data. On July 18, Mierzwa issued a report on the cause of the Loss. See Mierzwa’s First Expert Report, Trial Ex. 12. (I will refer to that report as the “Mierzwa Report.”) In the Mierzwa Report, Mierzwa concluded that (1) the source of the water damage was a pipe break near the kitchen sink; (2) the pipe break followed a freeze-up sometime in January or February 2016; (3) the Property was so cold because the thermostats’ batteries had died; and (4) the lack of oil deliveries since March 2015—and the fact that the oil tank was still over half full—indicated that very little, if any, heat was consumed at the Property during the winter of 2015–16, and so the batteries might have died before the winter began. See id. at 25. Following the Mierzwa Report, the Plaintiffs’ claim was referred to Great Lakes for further review. See Trial Tr. (Day 2) at 18. Great Lakes retained coverage counsel to evaluate the claim. See id. at 19. As part of that evaluation, coverage counsel examined Van Natta under oath on May 30, 2017. See JTM, Doc. No. 83-3, at ¶¶ 32–34; Van Natta’s EUO, Trial Ex. 27. On July 17, 2017,8 Great Lakes sent the Plaintiffs a letter disclaiming coverage based on the Freezing Exclusion and the Mold Exclusion. See JTM, Doc. No. 83-3, at ¶ 37. In February 2018, the Plaintiffs filed suit for breach of contract. See id. at ¶ 39. 8 Although in their joint trial memorandum the parties say that this letter was sent on July 17, 2018 (rather than 2017), it is clear that the letter was sent in 2017. First, this case was filed in February 2018, which was necessarily after coverage was denied. Second, in connection with their motion for summary judgment, Great Lakes submitted a copy of the letter, which was dated in the year 2017. See Letter, Ex. N to Badders Aff., Doc. No. 44-39. As part of this litigation, the Plaintiffs retained Jim Maguire as an expert on property damage restoration, and Maguire produced an expert report on July 11, 2018. See Maguire’s First Expert Report, Trial Ex. 9.9 In his first expert report, Maguire estimated reconstruction costs to be between $244,800 and $326,400. See id. After I issued my summary judgment ruling, the parties’ experts produced second expert reports. In Mierzwa’s second expert report—issued on June 12, 2020—Mierzwa “assess[ed] if direct physical water damage (without mold) could have been sustained to” (1) the garage, (2) the storage area above the garage, and/or (3) the crawl-space area through the master bedroom. Mierzwa’s Second Expert Report, Trial Ex. 2, at 1. After examining numerous photographs and relying on his own recollection from his site visit in May 2016, Mierzwa concluded that the following areas sustained only water (and not mold) damage: (1) a portion of the garage floor, (2) a wooden joist near the garage ceiling, and (3) a “small section of the landing/doorway within the upper storage room.” Id. at 12. Mierzwa estimated the cost of “perform[ing] work” to remedy that damage to be between $1,700 to $3,400. Id.10 In Maguire’s second expert report—issued on October 21, 2020—Maguire submitted a “revised estimate” of reconstruction costs “limited to those portions of the home that were damaged by water, but not by mold.” Maguire’s Second Expert Report, Trial Ex. 8. Ultimately, 9 Notably, the Plaintiffs’ relationship with Maguire did not begin with this litigation. Some months after the Loss was discovered, Van Natta asked Maguire to examine the Property. See Trial Tr. (Day 1) at 82–84. In the summer of 2016, the Plaintiffs hired Maguire to undertake mold remediation, completely gut the Property’s interior, demolish its interior structures, and remove and discard its contents. See Maguire Depo. Tr., Trial Ex. 31, at 60:20– 65:19. In late September or early October 2016, Maguire did that work, and the Plaintiffs paid him $34,055. See id. at 48:14–49:5; Maguire’s First Expert Report, Trial Ex. 9. 10 At trial, Mierzwa was qualified as an expert “for the purposes of determining the cause of the loss and the extent of the damage, the presence of mold and whether that can be separated – designated separately as water damage.” Trial Tr. (Day 2) at 37. Mierzwa admitted that he is neither a licensed contractor, electrician, nor estimator of any kind, and that he based his assessments of labor, construction, and replacement part costs on his “personal and professional” experiences doing work on his own properties and supervising work on other properties. See id. at 125–33. I need not decide what weight to afford Mierzwa’s testimony regarding costs because I do not reach the question; the Loss was not covered under the Policy because the Freezing Exclusion applies. Maguire estimated that the cost of remediation would be $32,273.86. See id. Maguire addressed the three areas that Mierzwa addressed in his second expert report and concluded that reconstruction costs would be $4,538.84—only $1,138.84 higher than the high end of Mierzwa’s estimate. Id.; see also Trial Tr. (Day 2) at 111–13 (Mierzwa indicating that he and Maguire were generally in agreement about those costs). However, Maguire also claimed that oak flooring throughout the Property and 13 interior wooden doors were damaged only by water and would need to be replaced. See Trial Tr. (Day 1) at 75–98. Those additional charges—plus 10 percent overhead, 10 percent profit, and 6.35 percent sales tax—accounted for the balance of Maguire’s revised estimate. See Maguire’s Second Expert Report, Trial Ex. 8. Mierzwa disputed the validity of those additional costs.11 B. Competing Interpretations: Events Before the Loss Since late 2014, Van Natta has lived primarily in West Harrison, New York. See JTM, Doc. No. 83-3, at ¶ 10; Trial Tr. (Day 1) at 3. At all times relevant to this case, Van Natta was responsible for maintaining the Property. JTM, Doc. No. 83-3, at ¶¶ 10–11. The major question in this case is whether Van Natta responsibly maintained heat at the Property in the months leading up to the Loss. Unsurprisingly, Van Natta argues that he did. According to Van Natta, he was present at the Property (1) sometime right before Christmas in December 2015,12 (2) on or around January 11 Maguire claimed that the damage to the oak floors and interior wooden doors involved only water—and not mold—because the wood was warped, and water (not mold) causes warping. See Trial Tr. (Day 1) at 75–98. Maguire further explained that to the extent mold was on the floors and doors, it could simply be wiped off. See id. Mierzwa admitted that water causes warping and that, because the wood floors had urethane covering, mold on the floors or other smooth and finished wood surfaces could most likely be wiped off. See id. (Day 2) at 114, 144, 150. However, Mierzwa raised several challenges to Maguire’s cost assessment, including to Maguire’s square footage calculation. See id. at 114–18, 140–41. In addition, Mierzwa identified several pictures of interior doors that depicted extensive mold damage that did not look as though it could simply be wiped off. See id. at 118–22; see also Servpro Report, Trial Ex. 6. 12 In his examination under oath, Van Natta actually claimed that he had been at the Property twice in December 2015. See Van Natta’s EUO, Trial Ex. 27, at 46:9–22. 29, 2016, and (3) on or around February 11, 2016. See Trial Tr. (Day 1) at 20–21, 50–51. On all those occasions, Van Natta claims that the Property looked fine. See id. at 18–19, 32. Because Van Natta was at the Property as late as February 11, 2016 and everything seemed fine, Van Natta believes that whatever caused a pipe to burst happened between February 11 and March 16. There is no corroboration in the record regarding Van Natta’s December 2015 visit. And there is very little corroboration for Van Natta’s January and February 2016 visits. The only piece of evidence that directly corroborates Van Natta’s presence at the Property on or around January 29, 2016 and on or around February 11, 2016 comes in the form of a one-page affidavit from Jamie Verab. See Verab Aff., Trial Ex. 20. Verab plowed Van Natta’s driveway “during the winter months from 2006 to 2016.” Id. at ¶ 4. Verab swears that Van Natta paid him “in person and in cash in Bethany, CT on February 11, 2016.” Id. at ¶ 7. A rudimentary ledger— which does not bear any indication of the date on which it was created—also suggests that Van Natta paid Verab in cash on January 29. See also Ledger, Ex. A to Verab Aff., Trial Ex. 20. As Van Natta testified at trial, he and Verab are friends. This is how Van Natta described the process of paying Verab for his services: I would meet him in different places, over – we would meet sometimes, we would grab dinner sometimes. I would – I would meet him in his – at his – at his house. His brother, his brother had a – had a greenhouse basically around the corner. Various places. He’d meet me at the house. I’d pay in cash at the house, sometimes in the driveway. One time I paid him when he was on a job. . . . I went around the corner and paid him there. Well, you know, different – different places. But, you know, I had dinner with him occasionally, or breakfast with him the following morning, and I would go up there and pay him in person. Trial Tr. (Day 1) at 20. In contrast to Van Natta’s version of events, Great Lakes suggests that Van Natta was not present at the Property at all between November 2015 and March 16, 2016. In Great Lakes’ estimation, sometime during that absence—most likely around January 4–5 (explained further below)—a hot water pipe under the kitchen sink froze and ruptured. The pipe froze because the Property’s two battery-powered thermostats died and so could no longer trigger the furnace to heat the Property. After the frozen pipe thawed, it began leaking water. That leaking went unabated for over two months, while Van Natta was absent from the Property. In support of its theory, Great Lakes offers numerous pieces of probative circumstantial evidence, including (1) cell phone records, (2) electricity bills, (3) thermostat-related information, (4) historical temperature data, and (5) oil delivery tickets. In Great Lakes’ view, that evidence all adds up to the one cohesive story described above. Although, as discussed below, the theory is not airtight, I generally agree with Great Lakes. 1. Cell Phone Records I received into evidence at trial Van Natta’s Verizon cell phone records from November 7, 2015 to April 7, 2016. See Verizon Records, Trial Ex. 24. The number associated with the Verizon records corresponds to Van Natta’s only mobile phone; the Property had no landline at the relevant times. See Van Natta’s Depo. Tr., Trial Ex. 28, at 48:19–50:18. The Verizon records list every voice call that Van Natta made and received during the covered period. The records list the “origination” and “destination” location of each call, as well as each call’s duration. For calls that Van Natta placed, the “origination” location corresponds (roughly) to where Van Natta was when he placed the call. For calls that Van Natta received, the “origination” location corresponds (roughly) to where Van Natta was when he received the call.13 Thus, whether Van Natta placed or received a call, the “origination” location in the Verizon records corresponds (roughly) to Van Natta’s location. 13 Although the parties did not address exactly how to read the Verizon records with respect to incoming calls, the parties’ arguments assumed that the “origination” location for incoming calls corresponded (roughly) to On March 16, 2016, the Verizon records reflect that several calls originated in Bethany, Connecticut or a nearby location (such as Beacon Falls, Southbury, and Prospect). See Verizon Records, Trial Ex. 24, at VANNATTA000689. Of course, Van Natta confirmed that he was in Bethany on March 16—that is the date on which he discovered the Loss. The same goes for March 18. On that day, too, Van Natta was at the Property. See Trial Tr. (Day 1) at 29–30 (“Q. So did you return to your property in Bethany on March 18? A. Yes.”). And, indeed, the Verizon records reflect several calls that day originating in Bethany, Beacon Falls, and Southbury. See Verizon Records, Trial Ex. 24, at VANNATTA000690. In contrast, the Verizon records do not reflect a single call originating from (or near) Bethany, Connecticut between December 1, 2015 and March 15, 2016. See Verizon Records, Trial Ex. 24. More pointedly, the Verizon records seem to contradict Van Natta’s assertions that he was present at the Property sometime right before Christmas in 2015, around January 29, 2016, and around February 11, 2016. For instance, from December 20 to 24, Van Natta made or received 51 calls, none of which originated near Bethany and many of which originated in southern States. See id. at VANNATTA000717–18.14 From January 28 to 30, Van Natta made or received 22 calls, all of which originated in either White Plains or Port Chester, New York. See id. at VANVATTA000708–09. And from February 10 to 12, Van Natta made or received 15 Van Natta’s location upon receiving the call. For instance, as discussed below, Van Natta himself used the “origination” location of one incoming call to suggest that he was in or around that “origination” location at the time of the call. See Verizon Records, Trial Ex. 24, at VANNATTA000717 (incoming call on December 21, 2015 at 8:39 p.m. “originating” in Riverside, Connecticut); Trial Tr. (Day 2) at 55 (Van Natta testifying that that call corroborated his recollection of being in Riverside on that date and time); see also infra n.14. 14 On December 21, one call originated in Riverside, Connecticut. See Verizon Records, Trial Ex. 24, at VANNATTA000717. That is the only call that originated in Connecticut between December 20 and 24. Although Van Natta points to that 8:39 p.m. call in an attempt to corroborate his assertion that he travelled to the Property that night, see Trial Tr. (Day 1) at 55, I do not view the call in the same way. Riverside is far from Bethany, in the southwest corner of Connecticut, and it is close to West Harrison, where Van Natta had his primary residence. calls, all of which originated in White Plains, Rye, Port Chester, or Harrison, New York. See id. at VANNATTA000698–99. At trial, Van Natta conceded that the Verizon records do not place him in Bethany at any time between December 1, 2015 and March 15, 2016. Still, Van Natta attempted to harmonize the records with his theory of the case. In Van Natta’s estimation, it was not surprising that the Verizon records reflected no calls originating in Bethany during that period because, in general, he “didn’t necessarily make phone calls from Connecticut once I was in Connecticut.” Trial Tr. (Day 1) at 51. Instead, Van Natta claimed that he often had long phone calls with his girlfriend on the way to Bethany, which the Verizon records reflect. See id. For instance, on February 11, Van Natta points to two calls to his girlfriend’s number at 7:42 pm (for 24 minutes) and at 8:08 pm (for 52 minutes)—both originating in White Plains, New York. See Verizon Records, Trial Ex. 24, at VANNATTA000699. The next call that Van Natta made was on the morning of February 12 at 10:47 am, originating in Port Chester, New York. See id. Van Natta claimed that those calls are consistent with his having spent the night of February 11 in Bethany, Connecticut. Similarly, Van Natta identified a 99-minute call to his girlfriend on the night of January 29 originating in White Plains as indicative of a call that he made on his way to Bethany. See Trial Tr. (Day 1) at 53–54. Van Natta also points to calls from late on November 30 and early on December 1 originating in Danbury, Southbury, Redding, and Stamford, Connecticut, to suggest that he travelled to the Property that night. See Trial Tr. (Day 1) at 55–56; Verizon Records, Trial Ex. 24, at VANNATTA000727–28. As discussed below, I find the absence of calls originating near Bethany, Connecticut at the relevant times to be highly probative evidence, and I do not find Van Natta’s contrary explanations persuasive. 2. Electricity Bills I also received into evidence at trial the Eversource electricity bills associated with the Property between February 24, 2014 and June 8, 2017. See Eversource Bills, Trial Exs. 22 and 32. Great Lakes emphasizes an eye-popping rise in electricity charges from December 22, 2015 to March 22, 2016. The charge for December 22, 2015 to January 22, 2016 was $374.48. See Eversource Bills, Trial Ex. 22. The charge for January 22 to February 23, 2016 was $718.94. See id. And the charge from February 23 to March 22, 2016 was $506.48. See Eversource Bills, Trial Ex. 32. Before those astronomical bills, no single electricity charge (at least since February 2014) had exceeded $126.55. See id. For the same period from just a year before—December 22, 2014 to March 23, 2015—the aggregate charge was $323.85, which is about five times smaller than the aggregate charge just one year later ($1,599.90). Great Lakes posits that the extraordinary jump in electricity usage owed to the enormous amount of electricity being used by the Property’s (1) water pump, which was pumping an abnormal amount due to the burst pipe that was leaking, and (2) water heater, which was working continuously because it was a hot water pipe that had burst. See Trial Tr. (Day 2) at 78, 183 (Mierzwa testifying that that was “the most likely cause” of what happened). Van Natta was aware of the high electricity charges at the time they accrued (in 2015 and 2016). Van Natta’s testimony about his response to the bills was perplexing. On the one hand, Van Natta confirmed that the bills stood out to him “as unusually large” and that he “questioned Eversource about” them. Trial Tr. (Day 1) at 45–46; see also Van Natta’s EUO, Trial Ex. 27, at 39:20–44:14. On the other hand, Van Natta claimed that the bills did not raise a red flag because they fit into a years-long pattern of erratic billing by Eversource about which he had repeatedly questioned Eversource. See Trial Tr. (Day 1) at 46 (Van Natta explaining that he had “questioned Eversource . . . for years” regarding “discrepancies in bills”). Van Natta testified: “I had that problem with them for a long time, and when I saw this, yeah, it stood out compared to some of the others,” but “it’s not even a greater factor than other months on this summary.” Id. 3. Thermostat-Related Information Closely related to the electricity bills is information regarding the Property’s two thermostats. See Trial Tr. (Day 2) at 49–50 (Mierzwa describing the location of the two thermostats, one on each level of the Property). Van Natta was in the habit, during the winter, of leaving the thermostats set between the “high 40s” and “low 50s” when he was absent from the Property. See id. (Day 1) at 10. Given that explanation, one would expect that, if the thermostats were working, no freeze-up would have occurred. In Great Lakes’ view, though, at some point before the freeze-up occurred, the batteries in the thermostats died, and so the thermostats were unable to signal the Property’s furnace to turn on. See id. (Day 2) at 49 (Mierzwa noting that the furnace is “an oil-fired furnace, which has signals sent to it by thermostats that are located on the base floor, as well as the main floor”). Importantly, it is beyond dispute that the thermostats were, in fact, battery powered. Although the thermostats had the capability of being hard-wired to the furnace, that capability was not enabled at the Property. See Honeywell Installation Guide, Att. to Mierzwa’s First Expert Report, Trial Ex. 12, at Mier-000046 (explaining that “[t]o wire the thermostat for AC power, connect the common side of the cooling transformer to the ‘C’ terminal as shown at left.”); id. (explaining that “[t]he thermostat can be powered by batteries alone or, if used with AC power, can provide backup power”); see also Mierzwa’s First Expert Report, Trial Ex. 12, at Photos 16 and 26 (showing that the “C” terminal was not hardwired on either thermostat); Trial Tr. (Day 2) at 50. When inspecting the Property on May 16, 2016, Mierzwa concluded that the batteries powering the thermostats had died, in part, because the thermostats’ digital faces were entirely blank. See Trial Tr. (Day 2) at 51, 53; Mierzwa’s First Expert Report, Trial Ex. 12, at 11, 15. Van Natta claims that he always believed the thermostats were hardwired into the Property’s electrical system and that the batteries served merely as backups. See Trial Tr. (Day 1) at 59; Van Natta’s EUO, Trial Ex. 27, at 31:10–15. Nevertheless, Van Natta was apparently in the habit, as a general matter, of “check[ing] batteries in different things in the house – thermostat, the fire/the smoke detectors, the security system, that sort of thing.” Trial Tr. (Day 1) at 6. Van Natta normally replaced those batteries approximately every two years. See Van Natta’s EUO, Trial Ex. 27, at 29:3–17. But Van Natta did not know the last time he replaced the batteries in the thermostats before March 16, 2016. See Trial Tr. (Day 1) at 41. Relatedly, Van Natta claims that, in general, he took good care of the Property’s heating system.15 Importantly, though, Van Natta acknowledged that at least twice before March 16, 2016—and as recently as “one or two years before” the Loss—the furnace had stopped working because its pilot light had gone out. See id. at 8–10. 4. Historical Weather Data As part of the Mierzwa Report, Mierzwa examined historical weather records for Bethany, Connecticut. See Mierzwa’s First Expert Report, Trial Ex. 12, at Mier-000026–37. After analyzing those data, Mierzwa concluded that three periods of extended below-freezing 15 Van Natta had the heating system (furnace and two thermostats) installed in 2005, and the furnace was serviced yearly “to make sure the filters were okay on it and that it was operating normally.” Trial Tr. (Day 1) at 6– 7; Van Natta’s EUO, Trial Ex. 27, at 17:18–19:24. Indeed, as part of an inspection that Great Lakes conducted before renewing homeowners’ coverage on the Property, Great Lakes concluded that the furnace “was in ‘good’ condition” and noted that it was “serviced annually.” JTM, Doc. No. 83-3, at ¶ 21. weather in January and February 2016 were the most likely times when a freeze-up might have occurred: (1) January 4–5, (2) January 18–23, and (3) February 11–15. See Mierzwa’s First Expert Report, Trial Ex. 12, at 24; Trial Tr. (Day 2) at 86. Because the electrical bill from December 22, 2015 to January 22, 2016 was high ($374.48), Mierzwa concluded that the most likely time for the freeze-up was January 4–5, and that the pipe started leaking water after thawing sometime around January 7. See id. 5. Oil Delivery Tickets On March 16, the Property’s oil tank was over half full with oil. See Mierzwa’s First Expert Report, Trial Ex. 12, at Photo 7. (The furnace was oil-fired.) The parties dispute the significance of that fact. Great Lakes points to oil delivery receipts received into evidence at trial. See Oil Delivery Receipts, Trial Ex. 23; see also Mierzwa’s First Expert Report, Trial Ex. 12, at 22; Trial Tr. (Day 2) at 84 (Mierzwa testimony). Those receipts evidence the following deliveries: Date Amount (in gallons) January 19, 2013 100 February 1 100 September 20 250 December 3 251.5 January 7, 2014 258.2 February 8 261 December 16 200 January 20, 2015 250 February 20 200 March 20 200 No receipts evidence an oil delivery later than March 20, 2015. Van Natta confirmed that he usually received oil on a monthly basis during the winter months. See Van Natta’s EUO, Trial Ex. 27, at 47:5–23. In Great Lakes’ view, that fact fits into their theory that Van Natta was not at the Property during the winter of 2015–16. At trial, Van Natta did not offer an explanation for the lack of oil delivery receipts after March 20, 2015. Instead, Van Natta appeared to argue that there may have been further oil deliveries that were simply not memorialized in receipts. See Trial Tr. (Day 2) at 174–75. In Van Natta’s deposition testimony, which was admitted by stipulation as a full exhibit at trial, Van Natta testified that he believed he received further oil deliveries, that he paid the oil delivery company in cash, and that the oil delivery company’s records were simply mistaken. See Van Natta’s Depo. Tr., Trial Ex. 28, at 20:15–21:1. However, Van Natta admitted that he could not locate any other oil delivery receipts. See id. at 17:25–18:11. V. Discussion This case turns mostly on one issue: Whether Van Natta was at the Property in the winter of 2015–16. Although it is not beyond all reasonable dispute, I find that the substantial weight of the evidence establishes that Van Natta was not present at the Property during the winter of 2015–16. That conclusion—that Van Natta left the Property unattended for nearly four months in the dead of winter—is critical. Van Natta left the Property unattended, even though he knew the Property’s electrical bills were astronomical and that the furnace’s pilot light had gone out in the past. Those facts establish that Van Natta failed to take reasonable care to maintain heat at the Property, and so the Loss is not covered under the Policy. Because I hold that the Loss was uncovered (and so Van Natta is owed no damages), I need not decide the amount of damages that Van Natta is owed due to damage allegedly caused only by water. When analyzing insurance claims under Connecticut law, courts should “begin [their] analysis with the initial grant of coverage . . . and then consider the effect of the exceptions and exclusions to the policy’s coverage.” Capstone Bldg. Corp., 308 Conn. at 774. Here, there is no question that the Loss falls under the Policy’s initial grant of coverage. The parties agree on that. See JTM, Doc. No. 83-3, at ¶¶ 17–18. The burden of showing that the Freezing Exclusion applies rests on the insurer, Great Lakes. See Capstone Bldg. Corp., 308 Conn. at 788 n.24. Great Lakes has easily satisfied its burden because the great weight of the evidence suggests that the cause of the loss was a freezeup. The parties agree that the Loss occurred because a water pipe burst. The parties also agree that the Loss occurred sometime during the winter of 2015–16. On March 18, 2016, after inspecting the property, Rollinson concluded that the Loss was “due to a freeze up.” Rollinson Report, Trial Ex. 19, at Bates 1535. In the Mierzwa Report, Mierzwa concluded that the Loss occurred due to a freeze-up of a pipe underneath the kitchen sink. To support his conclusion, Mierzwa pointed to (1) photographs of oxidation and rusting on pipes beneath the kitchen sink,16 (2) photographs of water stains and debris streaks on the floor joist below the kitchen sink as seen from the garage,17 (3) photographs of the thermostats, which showed blank digital faces and thus suggested that the batteries in the thermostats had died,18 (4) the remaining oil in the oil tank, which indicated that the oil-fired furnace had not been working, (5) the shockingly high Eversource electricity bills, which indicated that the water pump and water heater were working nearly continuously during those months, and (6) the lack of oil delivery receipts, which suggested that Van Natta had not used heat at the Property for some time. In short, both 16 Mierzwa’s First Expert Report, Trial Ex. 12, at Photos 29 and 30. 17 Mierzwa’s First Expert Report, Trial Ex. 12, at Photo 13; Trial Tr. (Day 2) at 80–81. 18 Mierzwa’s First Expert Report, Trial Ex. 12, at Photos 15 and 24. common sense and substantial evidence suggest that the cause of the Loss was a classic freezeup. In the face of that evidence, Van Natta offers only speculation and weak counterarguments that sound in possibility. For instance, Van Natta argues that the causal chain of the Loss may have been reversed: The batteries “may have suffered damage from the event rather than the batteries failing.” Trial Tr. (Day 2) at 152. In support, Van Natta points to a photograph that shows slight rusting on the negative node of one of the batteries in the upstairs thermostat. See id. at 154; Mierzwa’s First Expert Report, Trial Ex. 12, at Photo 25. Van Natta emphasizes that excessive moisture can cause corrosion, which, in turn, can cause batteries to die. Van Natta also highlights several things that Mierzwa did not do, which, in Van Natta’s view, degrade Mierzwa’s conclusion regarding the cause of the Loss. For example, Mierzwa did not find the actual hole that was the source of the leak or test the piping to determine where it was, test the batteries in the thermostats to see if they were truly dead, examine the furnace to see whether it was in good working order, inspect the well and pump, know the expansion tank’s pressure trip level, or know exactly how much power the pump and water heater pulled. See Trial Tr. (Day 2) at 62, 152–53, 155–58, 160–62, 170–73. Van Natta’s critiques are factually accurate, but they demonstrate only that the Mierzwa Report—and Mierzwa’s testimony—was not airtight. To be sure, the Mierzwa Report and Mierzwa’s testimony would have been even more persuasive if Mierzwa had, for instance, tested the thermostats’ batteries to show that they were dead, or estimated exactly how much power the pump and water heater might have been expected to draw under his theory.19 But, even without those extra forms of corroboration, I still find the Mierzwa Report—and Mierzwa’s testimony at trial—extremely credible. The evidence that Mierzwa assembled—the high electricity bills, the thermostat information, the historical weather data, and the lack of oil delivery receipts— combine to form one coherent whole, like pieces of a puzzle. The electricity bills, in particular, strongly corroborate Mierzwa’s theory that the leak began on January 7, assuming that the water pump and water heater were working at a constant per day rate after the leak began.20 The January 22 to February 22, 2016 electricity bill ($718.94) yields an average per day cost of $22.47, which is $718.94 divided by 32. Because the pipe was leaking on all those days, $22.47 represents the average per day electricity cost for the duration of the leak. Applying that per day cost to 16 days of the December 22, 2015 to January 22, 2016 bill (from January 7 to January 22, inclusive) yields a cost of $359.52. The cost of the entire month’s bill was $374.48. Similarly, applying that per day cost to 22 days of the February 23 to March 22 bill (February 23 to March 15, inclusive) yields a cost of $494.34. The cost of that entire month’s bill was $506.48. Breaking the electricity bills down into estimated per day costs only strengthens Mierzwa’s theory. In sum, Mierzwa’s expert report and testimony, along with all the other evidence and common sense, strongly suggest—and I find—that the cause of the Loss was a freeze-up. 19 Some of Van Natta’s challenges to the Mierzwa Report, though, are trivial. For example, I do not find it significant that Mierzwa could not locate the actual source of the leak. The parties agree that a huge water leak caused the Loss. Mierzwa’s inability to find the exact source of the leak is not particularly important. 20 I acknowledge that is a crude approximation. For instance, the hole might have started small and gotten bigger. Or, perhaps, the leaking pipe re-froze at some point after January 7, such as the two later periods (Jan. 18– 23 and Feb. 11–15) that Mierzwa identified as periods with extended sub-freezing temperatures. Still, though, it is a reasonable approximation because Mierzwa’s theory is that the cycle involving the water pump and water heater was repeating continuously over a period of months, until the water was shut off. See Trial Tr. (Day 2) at 161–62. The burden of establishing that the Heat Exception to the Freezing Exclusion applies is on the Plaintiffs. See Capstone Bldg. Corp., 308 Conn. at 788 n.24. The Heat Exception applies if Van Natta took reasonable care to maintain heat at the Property. Whether Van Natta acted reasonably in maintaining the heat is an objective inquiry. See McCartney v. Pawtucket Mut. Ins. Co., 1994 WL 723056 , at *3 (Conn. Super. Ct. Dec. 23, 1994). The question, which is highly fact-specific, is whether Van Natta acted as a reasonably prudent person would under the circumstances. See Doe v. Saint Francis Hosp. and Med. Ctr., 309 Conn. 146 , 167 (2013). For several reasons, Van Natta cannot establish that he used reasonable care to maintain heat at the Property. Most importantly, I find that the substantial weight of the evidence establishes that Van Natta was absent from the Property for four months during the winter of 2015–16. Because that conclusion contradicts Van Natta’s testimony, I will explain it in some detail. The only evidence suggesting that Van Natta was at the Property any time in December 201521 is his own testimony. For obvious reasons, courts often discredit a party’s uncorroborated, self-serving testimony, particularly when other evidence in the record contradicts that testimony. See, e.g., Ketab Corp. v. Mesriani Law Grp., 2016 WL 2902333 , at *6 (C.D. Cal. May 18, 2016) (“Plaintiff’s uncorroborated and self-interested testimony is entitled to little weight.”); Malick v. J.P. Morgan Chase Bank, N.A., 2016 WL 2858772 , at *2 n.2 (D. Conn. May 16, 2016); Pure Power Boot Camp, Inc. v. Warrior Fitness Boot Camp, LLC, 813 F. Supp. 2d 489, 553 (S.D.N.Y. 2011). In the face of the strong contradictory evidence in this case, 21 The Verizon phone records indicate that Van Natta may have driven to the Property late in the evening on November 30 and early in the morning of December 1, 2015. See Verizon Records, Trial Ex. 24, at VANNATTA000727–28 (several calls in relatively quick succession originating in Danbury, Southbury, Redding, and Stamford, Connecticut). Notably, this is not when Van Natta said he was at the Property in December 2015. Rather, at trial, Van Natta testified that he was there “right before Christmas.” Trial Tr. (Day 1) at 21. In any event, I take as true that Van Natta was at the Property on December 1, 2015. Thus, I find his absence to have been between December 1, 2015 and March 16, 2016. as described above, I also afford little weight to Van Natta’s testimony regarding his visit to the Property just before Christmas in 2015. There is only slightly more evidence—one piece of evidence, in fact—to support Van Natta’s assertions that he was at the Property around January 2922 and February 11, 2016. That evidence is the Verab Affidavit. For several reasons, I afford the Verab Affidavit limited weight. First, the Verab Affidavit was executed in May 2019, years after the incidents at issue. See Verab Aff., Trial Ex. 20; cf. Singh v. Barr, 826 F. App’x 7 , 10 (2d Cir. 2020) (“The [Immigration Judge] accorded little weight to Rajinder Singh’s affidavit because, inter alia, it was written years after the incident it purported to describe, specifically for submission in Singh’s asylum proceedings.”). Relatedly, even though Verab attaches a ledger to his affidavit that purports to offer contemporaneous proof of Van Natta’s presence in Bethany, the ledger contains no indication—such as a date—to suggest when it was made. Second, the Verab Affidavit is extremely threadbare and lacking in detail. Third, although I do not imply nefarious motives, I note the potential bias arising from the fact that Van Natta and Verab are friends. See Trial Tr. (Day 1) at 19–20 (Van Natta explaining that they “would grab dinner sometimes,” or “breakfast . . . the following morning”). Fourth, Van Natta himself does not remember meeting Verab on either January 29 or February 11, 2016. See id. at 57–58. In sum, the only evidence supporting Van Natta’s claim that he was at the Property three times during the winter of 2015–16 is his own testimony and the Verab Affidavit, both of which I find relatively unpersuasive. 22 The Verab Affidavit does not actually say that Van Natta was in Bethany on January 29, but the ledger attached to the Verab Affidavit indicates that Verab was “pd 1/29/16 cash.” See Ledger, Ex. A to Verab Aff., Trial Ex. 20, at 4. Thus, I take the Verab Affidavit to mean that Van Natta was present in Bethany on January 29 to pay Verab in person. Construing the Verab Affidavit in that way (which is favorable to Van Natta) causes no prejudice to Great Lakes because, ultimately, I find that the Verab Affidavit is outweighed by other evidence. In contrast, I assign great weight to the contrary evidence, which suggests that Van Natta was absent from the Property between December 1, 2015 and March 16, 2016. The Verizon records are the most probative contrary evidence. The Verizon records’ “origination” data can provide objective and accurate corroboration regarding Van Natta’s whereabouts at certain points in time. For instance, on the two undisputed dates on which Van Natta was in Bethany— March 16 and 18, 2016—the Verizon records reflect calls originating in Bethany. See Verizon Records, Trial Ex. 24, at VANNATTA000689–90. And Van Natta himself relied on the Verizon records to establish that he was, in fact, in Arizona on a business trip from February 15 to 17, 2016. See Verizon Records, Trial Ex. 24, at VANNATTA000699; see also Trial Tr. (Day 1) at 18, 20, 30–32, 52; id. (Day 2) at 186–87 (Van Natta’s lawyer, in closing argument, claiming that Van Natta’s “presence at the business trip is substantiated by the phone records that were produced into evidence, clearly showing that he was there when he said he was there”). It is thus significant that, between December 1, 2015 and March 15, 2016, Van Natta made and received hundreds of calls, but none originated in (or near) Bethany, Connecticut.23 Van Natta’s strained attempts to square the Verizon records with his theory of the case are not persuasive. In general, Van Natta argued that it was unsurprising that the Verizon records reflected no calls originating in Bethany because he normally did not make or receive any calls when he was at the Property. That is not a strong argument. Although the Verizon records are consistent with Van Natta’s proffered explanation, they are also consistent with the (far simpler) conclusion that Van Natta was absent from the Property. The second conclusion is much more likely, too, given the sheer number and frequency of calls that Van Natta made and 23 As described above, see supra n.14, only one call during that nearly four-month period originated in Connecticut at all, and that call originated in Riverside, which is a town in the southwest corner of Connecticut, some 50 miles from Bethany. See Verizon Records, Trial Ex. 24, at VANNATTA000717 (Dec. 21, 2015). received between December 1, 2015 and March 16, 2016. Van Natta made or received hundreds of calls overall and at least one call on every day except two.24 Van Natta’s attempt at harmonization would require me to cast aside common sense and a much more obvious inference from extremely probative evidence. Van Natta’s attempts to construe the Verizon records in his support with respect to specific dates is even more strained and unconvincing. Van Natta testified: I mean, what I – what I usually did when I would drive up there is I would call my girlfriend for – for usually the length of the time that I was driving up, and I didn’t necessarily make phone calls from Connecticut once I was in Connecticut, but I usually spoke to her on the drive up. Trial Tr. (Day 1) at 51. Given that explanation, Van Natta then pointed to Verizon records from the evenings of January 29 and February 11 and argued that the call records supported his travelling up to the Property on those nights. Take February 11 first. Looking at the Verizon records for February 11, Van Natta identified a call at 8:08 p.m. to his girlfriend’s number that originated in White Plains, New York and lasted 52 minutes. See id.; see also Verizon Records, Trial Ex. 24, at VANNATTA000699. The next call was on February 12, at 10:47 a.m., originating in Port Chester, New York, a town very near West Harrison, New York, where Van Natta has his primary residence. Van Natta apparently suggests that he drove to Bethany at 8 p.m. on the evening of February 11, arrived at around 9:15 p.m.,25 and then left the following morning by around 9:30 a.m., at the latest. That is unlikely. It strains credulity to believe that Van Natta went to Bethany simply to sleep there. Furthermore, Van Natta claims that he paid Verab in person on February 11. Thus, according to 24 The two days on which Van Natta did not make or receive at least one call were December 26, 2015 and February 27, 2016. See Verizon Records, Trial Ex. 24, at VANNATTA000700, VANNATTA000718. 25 Van Natta testified that the driving distance between West Harrison and Bethany was “about 70 minutes.” Trial Tr. (Day 1) at 57. Van Natta’s version of events, he would have paid Verab sometime after 9:15 p.m. on February 11. I am unaware of any reason why a snow plowing bill would be so urgent that it needed to be paid at 9:15 p.m. rather than sometime the next day. Further, such an unusual, late-night meeting would be the type of event that participants are likely to remember, but Van Natta did not remember where or when he allegedly met and paid Verab on February 11. See Trial Tr. (Day 1) at 57–58. Van Natta proffered a similar explanation regarding January 29. See id. at 54–55. Van Natta identified a 99-minute call to his girlfriend at 8:30 p.m. on the evening of January 29 that originated in White Plains, New York. See id. The next call was the following morning at 9:01 a.m., also originating in White Plains, New York. See id.; see also Verizon Record, Trial Ex. 24, at VANNATTA000708. Van Natta’s explanation regarding January 29 suffers from precisely the same infirmities as his explanation for February 11. If Van Natta paid Verab on January 29, then it would have been at 10 p.m. Additionally, the fact that Van Natta “usually” called his girlfriend late at night on his way up to Bethany does not suggest that the specific late-night calls that Van Natta identified were made on the way to Bethany. In fact, Van Natta’s Verizon records are replete with long, late-night calls to his girlfriend. See Trial Tr. (Day 1) at 53 (identifying such calls on the nights of January 21 and 23, when not even Van Natta suggests he was on his way to the Property); Verizon Records, Trial Ex. 24, at VANNATTA000689 (March 10 call at 8:24 p.m. for 105 minutes). Further to the point, on the evenings of January 28 and February 10—the evenings before his supposed trips to Bethany—Van Natta also was on long calls with his girlfriend that originated in White Plains, New York. See Verizon Records, Trial Ex. 24, at VANNATTA000708, VANNATTA000699. Of course, it is highly unlikely that Van Natta shuttled back and forth between Bethany and West Harrison twice on the same day. A much more likely explanation is the one suggested by Great Lakes: Van Natta simply was not there. In my view, Van Natta’s explanations at trial were not credible—instead, the Verizon records speak for themselves.26 Given that, in my view, the evidence establishes that Van Natta was absent from the Property between at least December 1, 2015 and March 16, 2016, the question becomes whether Van Natta still took reasonable steps to maintain heat in the Property during that time. In my view, Van Natta did not. Most importantly, Van Natta did not arrange for someone to check on the Property, even though he easily could have done so, for instance, by giving a key to Verab. In similar circumstances, courts have held that plaintiffs acted unreasonably when they left properties uninhabited for the winter months in cold areas without arranging for someone to check on the properties. See, e.g., Jugan v. Econ. Premier Assurance Co., 728 F. App’x 86 , 88– 91 (3d Cir. 2018) (affirming district court’s granting summary judgment to insurance company based on only six-week absence before loss); Stephenson v. Allstate Indem. Co., 73 N.Y.S.3d 804 , 806 (3d Dep’t 2018) (affirming lower court’s granting summary judgment to insurance company based on three-month vacancy before loss); Pazianas v. Allstate Ins. Co., 2016 WL 3878185 , at *5 (E.D. Pa. July 18, 2016) (granting summary judgment to insurance company based on four-month absence before loss); Evangelista v. Hingham Mut. Fire Ins. Co., 2005 WL 705840 , at *1–3 (Mass. Super. Ct. Feb. 14, 2005) (granting judgment in favor of insurance company following bench trial based on four-month absence before loss); McCartney, 1994 WL 26 I do not imply that Van Natta was necessarily trying to mislead me. It may be that, due to the passage of time, Van Natta’s memory has simply faded. I also do not categorically rule out the possibility that Van Natta’s testimony was truthful. I simply conclude that the better evidence presented at trial, as described above, suggests that Van Natta was absent from the Property during the winter of 2015–16. 723056, at *1, *3 (granting judgment in favor of insurance company following bench trial based on three-month absence before loss). Van Natta’s failure to monitor the Property during those winter months was particularly unreasonable because he admitted he had contemporaneous knowledge of the astronomical electricity bills spanning December 22, 2015 to March 22, 2016. See Van Natta’s EUO, Trial Ex. 27, at 40:18–22. A reasonably prudent person who received a $718.94 electricity bill for a second residence that was often vacant would investigate that bill with some urgency. That would be especially so if—as Van Natta testified was the case here—there were no apparent reason for the high electricity bill. See Trial Tr. (Day 1) at 47 (“It’s confusing to me, over that period of time, that it would be that high.”). Yet, on his own account, Van Natta did not investigate with any urgency. I found Van Natta’s trial testimony on this point particularly unpersuasive. Although Van Natta seemed to concede that the electricity charges were large,27 he claimed they were simply part of a long pattern of “discrepancies” that he had noticed in the Eversource billing. See id. at 46–47 (“[Y]ou know, their equipment was – never reflected, in my opinion, an accurate – an accurate reading.”); see id. (Day 2) at 193 (Van Natta’s counsel arguing that “he was having all kinds of problems with the – with the meter on the gauge of the house”). As described above, the Eversource records simply belie that testimony: The charges from December 22, 2015 to March 22, 2016 were five times higher than those from the same period the previous year. And, although Van Natta contacted Eversource regarding the high bills, Van Natta did not pursue a resolution to the issue. When Eversource told him that there was no problem with their 27 At times, Van Natta seemed even to dispute that fact and to downplay the magnitude of the charges. See Trial Tr. (Day 1) at 46 (“[I]t is an inordinate amount of money, but it’s – it’s not – it’s not even a greater factor than other months on this summary.”). equipment,28 Van Natta apparently accepted that explanation. Notably, he did not himself check up on the Property, even though he was just 70 minutes away by car. See Verizon Records, Trial Ex. 24, at VANNATTA000698–709 (scores of calls originating in White Plains, New York between January 8 and March 7, 2016). Finally, two facts that Van Natta cites to suggest that he used reasonable care to maintain heat actually cut against him. First, Van Natta emphasizes that there was oil in the oil tank at the time of the Loss. See Trial Tr. (Day 2) at 184. In Van Natta’s view, that fact suggests that he took good care of the Property’s heating system. But that is not a good fact for Van Natta. Combined with the lack of oil delivery receipts, it lends support to Mierzwa’s theory that not much oil was consumed at the Property during the winter of 2015–16. Similarly, Van Natta focuses on the fact that he was in the practice of replacing the thermostats’ batteries every couple of years, even though he thought they were backups. That fact, too, does not help Van Natta— especially because he could not remember the last time he replaced the thermostats’ batteries before the Loss—because it implies that Van Natta knew that the thermostats’ batteries might die. But still, Van Natta did not visit (or ask someone to check on) the Property between December 2015 and March 2016.29 VI. Conclusion 28 See Trial Tr. (Day 1) at 46 (“Q. And as to these two billing periods, did you get a response from Eversource as to why these -- . . . . A. They – they always told me that the equipment was fine, even though they never came out to check the equipment.”); see also Van Natta’s EUO, Trial Ex. 27, at 44:5–11. 29 Similarly, Van Natta admitted that he knew the furnace’s pilot light had gone out in the recent past. See Trial Tr. (Day 1) at 8–10; see also Rollinson Report, Trial Ex. 19, at Bates 1536 (“The insured believes that the power may have shut off while he [was] away because this has happened 2-3 times previously since he has owned the home.”). It is plausible that the furnace’s pilot light’s going out—rather than the thermostats’ batteries’ dying— caused the freeze-up. In that case, too, Great Lakes would still be entitled to judgment for the reasons I have identified. It was unreasonable for Van Natta, knowing that the furnace’s pilot light had extinguished in the recent past, not to check on the Property for nearly four months throughout the winter of 2015–16. Because Great Lakes satisfied its burden of proving that the Policy’s Freezing Exclusion applied to bar coverage for the Loss, the Plaintiffs bore the burden of proving that the Heating Exception to the Freezing Exclusion applied to reinstate coverage. The Plaintiffs failed to carry that burden because they have not shown that they (1) used reasonable care to maintain heat at the Property or (2) shut off the Property’s water supply and drained the pipes. As a result, the Policy’s Freezing Exclusion applies, and the Plaintiffs are not entitled to coverage because Great Lakes did not breach the contract. The Clerk is instructed to enter judgment for the defendant—Great Lakes Reinsurance (SE) UK—and to close the case. So ordered. Dated at Bridgeport, Connecticut, this 16th day of June 2021. /s/ STEFAN R. UNDERHILL Stefan R. Underhill United States District Judge