(Dissenting).
Mrs. Stamp’s ability to recover most of the insurance benefits at issue hinges on whether her husband’s death was an accident within the meaning of the MetLife policies. It is clear to me that Mr. Stamp’s death while driving under the influence of alcohol was the result of an accident. As the Supreme Court recently noted, “DUI ... involves ‘accidental or negligent conduct.’ ” Begay v. United States, — U.S.-, 128 U.S. -, 1581, 1587, 170 L.Ed.2d 490 (2008) (quoting Leocal v. Ashcroft, 543 U.S. 1, 11, 125 S.Ct. 377, 160 L.Ed.2d 271 (2004)) (emphasis added). In Wickman v. Northwestern Nat’l Ins., 908 F.2d 1077, 1084 (1st Cir.1990), we announced anstandard for determining whether a death was “accidental” within the meaning of an ERISA-qualified policy. Noting that terms like “accident” “should be judged from the viewpoint of the insured,” inpromulgated a three-prong test because “actual expectation is often difficult, diffinot impossible to determine.” Id. (1) “[The] reasonable expectations of the insured when the policy was purchased is the proper starting point for a determination determinawhether an injury was accidental under its terms.” Id. at 1088. (2) “If the fact-finder determines that the insured did not expect an injury similar in type or kind to that suffered, the fact-finder must then examine whether the suppositions which underlay that expectation were reasonable.” reason(3) “Finally, if the fact-finder, in attempting to ascertain the insured’s actual actufinds the evidence insufficient insuffiaccurately determine the insured’s subjective expectation, the fact-finder should then engage in an objective analysis of the insured’s expectations.” Id.
The present case marks our first opportunity to apply the Wickman analysis to a factual scenario that involves a policy-holder who died with a heightened Blood Alcohol Content (“BAC”) level in a one-car collision. Deferentially following the footsteps of other circuits that have examined this question,14 the majority concludes that *95Mr. Stamp’s death was not accidental. I cannot disagree more vehemently with that outcome.
I do not feel compelled to follow these prior opinions because I believe that in reaching this outcome, those cases misapplied the third prong of Wickman, an error duplicated by the majority, which also focuses too heavily on a “reasonably foreseeable” analysis. See slip op. at 89. In Wickman, we said that “[i]n [our] analysis, [we are required to] ask whether a reasonable person, with background and characteristics similar to the insured, would have viewed the injury as highly likely to occur as a result of the insured’s intentional conduct.” 908 F.2d at 1088 (emphasis added).
I believe that those circuits that have examined insurance policy exclusions similar to the ones at issue here, using a “highly likely” or “substantially certain” standard, are the ones that have reached the correct answer to the question. See Padfield v. AIG Life Ins. Co., 290 F.3d 1121 (9th Cir.2002); Santaella v. Metro. Life Ins. Co., 123 F.3d 456 (7th Cir.1997) Todd v. AIG Life Ins. Co., 47 F.3d 1448 (5th Cir.1995). Those circuits applied Wickman in a proper fashion by engaging in the following analysis:
[F]or death under an accidental death policy to be deemed an accident, it must be determined (1) that the deceased had a subjective expectation of survival, and (2) that such expectation was objectively reasonable, which it is if death is not substantially certain to result from the insured’s conduct.
Santaella, 123 F.3d at 463 (quoting Todd, 47 F.3d at 1456) (emphasis added).
The administrator, which was affirmed by the majority, created out of whole cloth an exclusion ex post facto denying coverage where the decedent’s BAC is above legal limits when death' occurs. Rather than attempting to insert an exclusion by inference, what should be done in those cases in which the parties have not agreed to such a provision beforehand, is to have a clause specifically establishing such an exclusion. See Marcus Wilbers, Note, Alcohol-Related Car ‘Accidents’? The Eighth Circuit Moves Toward Policy Change in ERISA Litigation, 71 Mo. L.Rev. 471, 491 (2006) (“To eliminate any uncertainty, plan administrators should include an alcohol-related exclusion in all accidental death insurance plans.”).
The policies at issue here do not include a definition for “accident.” In Wickman, we said that “[c]ase law is fairly consistent in defining an accident, using equally ambiguous terms such as undesigned, unintentional, and unexpected.” Wickman, 908 F.2d at 1087; Santaella, 123 F.3d at 464 (“Interpreting the policy terms in an ordinary and popular sense, in the way that a person of average intelligence and experience would, we treat the term ‘accidental’ as it is commonly defined, as ‘unexpected or unintentional.’ ” (footnote and citations omitted)).15 Applying Wickman, Mr. Stamp’s death can be termed nothing other than an unfortunate accident. The question is not, as the majority frames it, whether Mr. Stamp was severely intoxicated. See slip op. at 91. The question is whether he intended to kill himself by becoming intoxicated and driving while in this condition. There is absolutely no evidence that Mr. Stamp had any such intention.
*96I. Wickman Analysis
The first prong of the Wickman analysis requires us to look at the reasonable expectation of the insured when the policy was purchased. Wickman, 908 F.2d at 1088. As we said in Wickman, “[generally, insureds purchase accident insurance for the very purpose of obtaining protection from their own miscalculations and misjudgments.” Id. Accepting that Mr. Stamp had a BAC above the legal limit and that he was in a one-car collision with no other evidence of why he veered off the road and hit a tree, it is fair to say that, at most, Mr. Stamp miscalculated and misjudged his ability to drive that fateful evening. He purchased insurance in order to protect himself and his family against such unforeseen accidents.
There is no evidence that Mr. Stamp’s actions were the result of anything but misjudgments, not intentional actions. According to Joe Kingsley, he and Mr. Stamp discussed life insurance at some point pri- or to August 2. Mr. Stamp told Kingsley that he wanted to make sure that his wife and daughter would be “taken care of if anything happened” to him, adding that he had taken out significant life insurance for their benefit. On the night of his death, all evidence indicates that if Mr. Stamp would have continued along the route to his parents’ house, he would have reached his destination. There is no indication that he attempted or intended to kill himself. There is no evidence that Mr. Stamp was distraught or that he was experiencing any personal problems. A few hours before the accident, he had positive, upbeat conversations with his wife, co-workers, and friend. His decision to drive was unlike the game of Russian roulette the majority discusses because the point of that game is to see if you are lucky enough to live, see slip op. at 92, or that as in Wick-man, he was doing something which objectively looked like suicide.
The second prong of the Wickman analysis instructs us to examine the reasonableness of the insured’s expectations. See id. “The determination of what suppositions are unreasonable should be made from the perspective of the insured, allowing the insured a great deal of latitude and taking into account the insured’s personal characteristics and experiences.” Id. (internal citations omitted). Much more often than not, driving while under the influence has a non-fatal outcome. Cf. Todd, 47 F.3d at 1457 (“the risk of death from autoerotic practice is ‘not of such a nature that [the decedent] knew or should have known that it probably would result in death. Death was not a normal expected result of the behavior.’ ” (citation omitted)(emphasis added)). In 2002, the year Mr. Stamp died, only forty-one percent of people who died in automobile-related collisions did so in collisions that involved a driver with a BAC above the legal limit. See U.S. Dep’t of Transp., Traffic Safety Facts 2004 32 (2004). Almost sixty percent of the people who died that year did so from car accidents involving a driver with a BAC of zero. See id. Given this, it can be said that “a reasonable person would not view death as ‘highly likely’ to result from driving drunk.’ ” Michael E. Gardner, Note, Accidental Death Insurance Coverage of Drunk Drivers, 69 Mo. L.Rev. 235, 246 (2004). Instead, “[i]t is not highly likely for an impaired driver to die in an alcohol-related wreck, and those deaths are therefore accidents.” Id. at 472. Smoking cigarettes is responsible for eighty-seven percent of lung cancer deaths, see Am. Cancer Soc’y, Cancer Facts & Figures 2008 48 (2008), and nearly twenty percent of smokers die from heart disease, see Am. Heart Ass’n, Cigarette Smoking and Cardiovascular Diseases, (June 16, 2008). Yet it would be highly *97unusual where the administrator, or this Court, to conclude that anyone who smokes is engaging in a suicidal act and is thus excludable from coverage under this policy.
In any objective analysis, we must “ask whether a reasonable person, with background and characteristics similar to the insured, would have viewed the injury as highly likely to occur as a result of the insured’s intentional conduct.” Wickman, 908 F.2d at 1088 (internal citations omitted). In 2005, nearly 1.4 million drivers were arrested for driving under the influence of alcohol or narcotics. See United States Dep’t of Justice, Crime Prevention in the United States 2005, Table 29. This number is much higher, both empirically and as a percentage, than the number of people who died as a result of driving under the influence. Thus, “[wjhat common knowledge ... actually tell[s] a person driving while intoxicated is that he or she is far more likely to be arrested for driving while intoxicated than to die or be injured in an alcohol-related automobile crash, and far more likely to arrive home than to be either arrested, injured, or killed.” West v. Aetna Life Ins. Co., 171 F.Supp.2d. 856, 904 (N.D.Iowa 2001).
The likelihood of death is even less probable when we take into account that the number of people arrested for drunk driving is lower than one percent of the 159 million self-reported cases of alcohol-impaired driving among adults in the United States each year. See Kyran P. Quinlan et al., “Alcohol-Impaired Driving Among U.S. Adults, 1993-2002,” 28 Am. J. of Preventive Med., 346, 348-49 (2005). These statistics do not demonstrate that it is highly likely that someone who is driving under the influence will die.
Thus, Mr. Stamp “had a subjective expectation of survival and ... such an expectation was objectively reasonable because death was not certain or even highly likely to result from [his] conduct.” Santaella, 123 F.3d at 464 (citing Todd, 47 F.3d at 1456). “Drunk driving is, to be sure, a foolish and reckless act. However, when an insured dies as a consequence of his driving while intoxicated, his death should still be considered accidental because a reasonable person would not view death as a natural and probable consequence of drunk driving.” Gardner, supra at 253. It cannot be said that Mr. Stamp expected or intended to kill himself. See Santaella, 123 F.3d at 463 (“[T]he record simply would not support a jury determination that a reasonable person in [the decedent’s] situation would have considered death either ‘highly likely to occur’ ... or ‘substantially certain to result from the insured’s conduct.’ ” (citations omitted)). I believe that appellees abused their discretion when they denied Mrs. Stamp’s claims because of a determination that Mr. Stamp’s death was not an accident. Mrs. Stamp’s claims, however, were denied on other grounds as well. We examine those grounds below.16
II. Intentionally Self-Inflicted Injury Exclusion
The inquiry for examining policy exclusions for intentional self-destruction and intentionally self-inflicted injury mirrors the analysis above. Given the facts, and knowing that Mr. Stamp was on his way to his parents’ house to meet his family, it cannot be said that his death was the result of a self-inflicted injury. In Wick-man, the decedent drove onto a bridge, got out of his car, and went to a part of the bridge that he only would have gone to if *98he intended to kill himself. See Wickman, 908 F.2d at 1079-80. We found that his death was not the result of an accident. See id. at 1089. “[A] reasonable person in Wickman’s shoes would have expected to die or be seriously injured as result of [his actions].” Id. at 1088. That cannot be said here. There is no evidence that he intentionally struck the tree which caused his death. “[T]he record [does] not support a determination that the insured ‘did anything other than make a fatal mistake.’ ” Padfield, 290 F.3d at 1121 (quoting Santaella, 123 F.3d at 465).
As with the term “accident,” there is no definition in the policy of “intentionally self-inflicted injury.” In ERISA cases, we use federal common law rules to “ ‘interpret the terms of the policy in an ordinary and popular sense, as would a person of average intelligence and experience,’ and ‘construe all plan ambiguities in favor of the insured.’ Plan language may only be deemed ambiguous where ‘it is subject to more than one reasonable interpretation.’ ” Santaella, 123 F.3d at 461 (quoting Life Ins. Co. of N. Am. v. Von Valtier, 116 F.3d 279, 283 (7th Cir.1997)). A person of average intelligence and experience would not think that someone who dies in a one-car collision, even with a BAC above the legal limit, intentionally caused his own death.17 No doubt, Mr. Stamp’s decision to drive while intoxicated was risky, “but voluntary risky acts resulting in injury are not necessarily acts that result in ‘intentionally self-inflicted injury.’ ” Padfield, 290 F.3d at 1129. Accordingly, I believe that Mrs. Stamp’s claim for Basic AD & D insurance benefits should not have been denied by Appellees.
III. Crime Exclusion
Both the Voluntary and Occupational AD & D policies have exclusions for decedents who die while “committing or attempting to commit a felony or other serious crime.” Appellees denied Mrs. Stamp’s claim because Mr. Stamp had a BAC level above the legal limit while he was operating his vehicle.
I believe that the exclusion is inapplicable and find that the Appellees abused their discretion when they found that Mr. Stamp committed a serious crime. Mr. Stamp was operating his vehicle in Rhode Island when he died, and under Rhode Island law it is not a felony to operate a vehicle while under the influence of alcohol; it is a misdemeanor. See Rhode Island Motor Vehicle Offenses, § 31-27-2(a).18 By definition, a misdemeanor is not a serious crime. See United Union of Roofers, Waterproofers, & Allied Workers, Union No. 33 v. Meese, 823 F.2d 652, 656 (1st Cir.1987) (“the ‘misdemeanor’ label typically applies to crimes that are not very serious”) (citation omitted). Accordingly, I would find that the crime exclusion does not apply, and that Mrs. Stamp is entitled to Voluntary AD & D benefits. This does not answer whether Mrs. Stamp is entitled to Occupational AD & D because that policy requires death to occur “while at work.”
IV. “While at Work” Exclusion
In order for Mrs. Stamp to qualify for Occupational AD & D benefits she had to *99demonstrate that Mr. Stamp was ‘at work” when he died. The policy clearly states that “while at work” “does not include any personal time taken before, during or following a Mobil-approved business trip.” Mrs. Stamp argues that Mr. Stamp was “at work” because the meeting he attended was mandatory and he died on his way back from the meeting. Exxon-Mobil noted that Mr. Stamp was no longer “traveling from” the work-related business meeting when he died. ExxonMobil also noted Mr. Stamp stopped “for several hours for a separate, purely personal purpose.”
According to Mrs. Stamp, Mr. Stamp’s co-workers, and Kingsley, Mr. Stamp was not inebriated when he left his work function. Mr. Stamp was legally inebriated when he died; his BAC level was three-times the legal limit. Given the road conditions, the weather, and Mr. Stamp’s conversation with Kingsley at about 9:30 p.m., Mr. Stamp should have reached his parents’ house by 10:30 p.m. Yet, by midnight, he still had not reached his destination. The conclusion that Mr. Stamp became inebriated by reason of some non-work related activity after he left the work function is supported by circumstantial evidence. Thus, I would find that ExxonMo-bil did not abuse its discretion when it determined that Mr. Stamp’s collision occurred during “personal time taken ... following” his trip and that Mrs. Stamp is not entitled to recover under the Occupational AD & D policy.
V. Conclusion
For the reasons explained above, I dissent.
. See Lennon v. Metro. Life Ins. Co., 504 F.3d 617 (6th Cir.2007); Eckelberry v. Reliastar Life Ins. Co., 469 F.3d 340 (4th Cir.2006); King v. Hartford Life & Accident Ins. Co., 414 *95F.3d 994 (8th Cir.2005) (en banc); Buce v. Allianz Life Ins. Co., 247 F.3d 1133 (11th Cir.2001); Cozzie v. Metro. Life Ins. Co., 140 F.3d 1104 (7th Cir.1998).
. Both the police and the autopsy reports indicated that the manner of death was "accidental.” This is not a mere coincidence.
. The majority did not examine the other policies because they also require the death to be a result of an accident. See slip op. at 86-87 n. 3.
. As previously noted, both the police and the autopsy reports classified the manner of death as ''accidental."
. The Rhode Island statute reads in relevant part:
(a) Whoever drives or otherwise operates any vehicle in the state while under the influence of any intoxicating liquor, drugs, toluene, or any controlled substance as defined in chapter 28 of title 21, or any combination of these, shall be guilty of a misdemeanor except as provided in subdivision (d)(3) and shall be punished as provided in subsection (d) of this section.