Bannister v. State Farm Mutual Automobile Insurance ( 2012 )


Menu:
  •                                                                           FILED
    United States Court of Appeals
    PUBLISH                          Tenth Circuit
    UNITED STATES COURT OF APPEALS             September 5, 2012
    Elisabeth A. Shumaker
    TENTH CIRCUIT                         Clerk of Court
    JAMES BANNISTER,
    Plaintiff-Appellant/Cross-
    Appellee,
    v.
    STATE FARM MUTUAL                                Nos. 11-6174 and 11-6186
    AUTOMOBILE INSURANCE
    COMPANY,
    Defendant-Appellee/Cross-
    Appellant.
    Appeal from the United States District Court
    for the Western District of Oklahoma
    (D.C. No. 5:09-CV-01244-F)
    Jacob L. Rowe, Oklahoma City, Oklahoma (Jim Buxton of Buxton Law Group,
    Oklahoma City, Oklahoma, and Billy Coyle, Oklahoma City, Oklahoma, with him on the
    briefs), for Plaintiff-Appellant/Cross-Appellee.
    Reid E. Robison of McAfee & Taft, P.C., Oklahoma City, Oklahoma (Mark D. Spencer,
    Michael K. Avery, McAfee & Taft, P.C., Oklahoma City, Oklahoma, and Daniel C.
    Andrews, Oklahoma City, Oklahoma, Jones, Andrews & Oritz, P.C., Of Counsel, with
    him on the briefs) for Defendant-Appellee/Cross-Appellant.
    Before TYMKOVICH, EBEL, and HOLMES, Circuit Judges.
    EBEL, Circuit Judge.
    Plaintiff James Bannister was injured in a motorcycle accident on the freeway near
    Oklahoma City in 2009. According to Bannister, he was forced to lay down and slide his
    motorcycle at a high speed when a car in front of him braked suddenly, that car having
    been cut off by another car. Bannister slammed into the wall of the freeway and suffered
    substantial injuries. He did not collide with any other vehicle; neither of the
    aforementioned cars remained at the scene of the accident; and no witnesses besides
    Bannister ever gave an account of the crash.
    Bannister filed an insurance claim with his insurer, defendant State Farm
    Automobile Insurance Company (“State Farm”). State Farm denied Bannister’s claim,
    finding him to be majority at fault in the accident, which precluded recovery under his
    insurance policy. Bannister subsequently filed suit in Oklahoma state court, and State
    Farm removed the case to the Western District of Oklahoma. By the time the case went
    to trial, Bannister sought relief solely on a tort theory: that State Farm violated its duty of
    good faith and fair dealing in denying his claim.
    The jury found in favor of Bannister, but the district court granted State Farm’s
    renewed motion for judgment as a matter of law (“JMOL”), ruling essentially that the
    evidence showed that State Farm’s denial of Bannister’s claim was based on a reasonable
    dispute regarding whether Bannister was majority at fault, and that no evidence suggested
    that further investigation would have undermined the reasonableness of that dispute.
    Meanwhile, the district court conditionally denied State Farm’s alternative motion for a
    2
    new trial based on the jury’s irregular calculation of damages as well as on allegedly
    improper prejudicial statements by Bannister’s counsel at trial.
    Exercising jurisdiction under 
    28 U.S.C. § 1291
     over this diversity action, we
    AFFIRM the district court’s JMOL ruling in favor of State Farm. Accordingly, we
    DISMISS AS MOOT State Farm’s cross-appeal, regarding whether the district court
    abused its discretion in conditionally denying State Farm’s alternative motion for a new
    trial.
    I. BACKGROUND
    1. The motorcycle accident
    On Thursday afternoon, January 22, 2009, Bannister was driving his motorcycle in
    the left-hand lane on Interstate 40, en route to pick up his wife at work. According to his
    own testimony at trial, Bannister was driving at a speed of approximately 65-70 mph, or
    5-10 mph above the legal limit. Bannister had been drinking beforehand at a motorcycle
    bar in Oklahoma City, where he had two beers and a shot of whisky over the course of
    perhaps two and a half hours. The police report for the accident indicated that Bannister
    was driving under the influence of alcohol.1
    1
    The actual notation on the police report was “ALCOHOL-DUI/DWI.” We note
    that State Farm, which had obtained the police report prior to denying Bannister’s claim,
    used “DUI” in its claim logs. Also, Bannister testified that he had been criminally
    charged with “driving under the influence”—a charge that was later dismissed for reasons
    not disclosed in the record. Bannister’s hospital record, which does not appear to have
    been introduced at trial, reveals that when Bannister was admitted to the hospital, his
    blood-alcohol level was 0.09.
    3
    At trial, Bannister claimed that, just prior to his accident, he was following a beige
    car at a safe distance when a red car, apparently in the right lane, came alongside him and
    began to encroach into his lane.2 Bannister revved his engine so as to send an auditory
    warning to the driver of the red car (in effect, a honk). The red car stayed in its lane,
    passed the beige car, and then cut off the beige car, causing it to brake suddenly. The
    beige car slid into the emergency lane, and Bannister locked his brakes.
    Bannister’s recollection of the crash is foggy. He testified that, to avoid hitting the
    vehicle in front of him,3 he executed a ‘lay down’ of his motorcycle—essentially a
    deliberate (though necessarily somewhat uncontrolled) fall-and-slide. The maneuver sent
    Bannister into the freeway’s barrier wall at a high speed. Bannister hit his head against
    the wall and suffered head trauma, broken bones, bruising, lacerations, and other injuries
    that kept him in the hospital for eight days. Neither the red car nor the beige car
    remained at the scene of the accident; an unidentified other vehicle stopped to assist
    Bannister.
    2. The insurance claim
    2
    No other witnesses from the scene of the accident were ever interviewed. As
    noted below, at one point during a discussion with a State Farm representative, Bannister
    referenced a witness from a car behind him who he said stopped and helped him.
    However, Bannister did not provide any identifying information about that individual,
    and neither the police report nor any other account identified the individual.
    3
    Presumably Bannister was referring to the beige car. Bannister meanwhile said
    that the beige car had moved into the emergency lane, but perhaps Bannister had also
    steered into the emergency lane.
    4
    The day after the crash, January 23, 2009, Bannister’s wife reported the accident
    to State Farm. State Farm’s first relevant substantive log entry in Bannister’s claim file
    (entry No. 7) was recorded by State Farm claim representative Wendy Jeffus based on
    Jeffus’s conversation with Bannister’s wife. That log entry characterized the crash as
    having involved a vehicle in front of Bannister slamming on its brakes, and Bannister
    swerving to avoid that vehicle, and rolling4 his motorcycle. The log further noted that
    investigation was needed to determine whether Bannister’s uninsured motorist (“UM”)
    coverage—coverage that could apply to unknown “miss-and-run” vehicles—would
    apply, as well as whether Bannister was majority at fault, which would negate coverage.
    The log indicated that Jeffus explained as much to Bannister’s wife.
    Three days later, on January 26, 2009, another claim representative, Edwina
    Kelley-Gilliam of the UM division of State Farm, updated Bannister’s claim log after
    speaking with a team manager named Collins, who worked in the auto claims, or liability,
    4
    Bannister has emphasized that he did not “roll” his motorcycle but rather “laid
    [it] down”—lay-down being a more controlled, deliberate (and so presumably less
    negligent) maneuver. E.g., ROA v. III at 835; Aplt. Br. at 7. At trial, Jeffus
    acknowledged that rolled would tend to indicate a flipping of the motorcycle, rather than
    a sliding. ROA v. II at 531. Jeffus testified that Bannister’s wife did in fact use the term
    “rolled” in describing the accident, 
    id.,
     although Bannister’s wife testified that she did not
    recall using that term and that she probably said “laid the motorcycle over” instead, 
    id. at 481
    . In any event, the manner in which the motorcycle fell does not appear to have made
    a material difference to State Farm’s evaluation of Bannister’s comparative negligence in
    the accident or to its investigation of Bannister’s claim. Rather, State Farm’s claim
    denial focused on other facts, primarily the single-vehicle nature of the accident and that
    the vehicle in front of Bannister was able to avoid a collision with the vehicle in front.
    5
    division. 5 Kelley-Gilliam’s log entry (entry no. 14) reflected: “[T]he liability
    [investigation] does not appear complete. In addition, the insured appears to be majority
    at fault based on the facts of the loss. Moving file back to [the auto claims division]
    pursuant to our discussion. UM retains nothing . . . .” ROA v. V at 1500. It apparently
    took Kelley-Gilliam three minutes to enter that determination.
    The next significant event occurred on February 9, 2009, when another State Farm
    claim representative, Gloria Mercado of the motorcycle collision division, met with
    Bannister. Mercado’s purposes were to examine the crashed motorcycle and to complete
    a vehicle inspection report concerning the damage incurred by the motorcycle. In other
    words, Mercado’s primary mission was to evaluate damages to the motorcycle as relevant
    to Bannister’s collision coverage, rather than to investigate Bannister’s liability.
    However, in addition to evaluating the motorcycle, Mercado discussed the accident with
    Bannister over the course of their meeting. And at trial, Bannister testified that the
    account of the accident that he related to Mercado was a truthful one. Mercado did not
    electronically record the conversation, but she summarized it in a claim log (entry no.
    37). As reflected in that log, Bannister told Mercado that
    2 other cars flew by him on the right and switched over in front of the car
    directly ahead of him. Then the car in front of him had to hit brake [sic]
    5
    Different divisions within State Farm serve different functions. For instance, the
    UM division determines whether a claim will be a UM claim versus whether another
    motorist does have insurance that will cover an incident; the vehicle collision division
    seems chiefly to evaluate damages to vehicles; and the auto claims (liability) division
    investigates the level of fault of claimants such as Bannister.
    6
    and [Bannister] wasn’t able to get slowed fast enough so he laid [his
    motorcycle] over to keep from hitting car in front of him, he has over 30
    years[’] experience riding [a motorcycle] & this was his 1st wreck. He’s
    unsure why cars slammed on brakes as none of them stopped for him. A
    car behind him saw what happened and stopped his car sideways to keep
    any other traffic from running over [him]. . . .
    
    Id. at 1495
    . This log entry was largely consistent, then, with the earlier-made entry no.
    7.6 Meanwhile, this new entry added the information about Bannister’s riding experience
    as well as the reference to the other car, behind Bannister, whose driver allegedly
    witnessed the accident. However, Bannister never gave Mercado or anyone else any
    identifying information about that alleged individual.
    At trial, Mercado testified that she believed that based on these facts, Bannister
    would be at fault. She reasoned that due to the fact that the car in front of Bannister was
    able to stop without collision, Bannister—driving behind that car—likewise should have
    been able to stop safely, if Bannister had been following at a reasonable distance per his
    duty as a motorist.
    On February 17, 2009, State Farm obtained a copy of the police report for
    Bannister’s accident. The next substantive entries in the claim log were a pair of entries
    (entries nos. 73-74) by Patrick Dreier of State Farm’s auto claims division on February
    25. Dreier’s entries essentially summed up State Farm’s knowledge to date—the history
    6
    The only inconsistency is log entry no. 37’s reference to “2 cars” passing the
    vehicle in front of Bannister, whereas according to log entry no. 7 (as well as Bannister’s
    trial testimony), only one car (the red car) cut off the car in front of him (the beige car).
    That discrepancy is immaterial, however.
    7
    of State Farm’s investigation, one might say. The entries indicated that the accident was
    a single-vehicle wreck, that Bannister had been driving under the influence of alcohol,
    and that no second vehicle was involved in any collision. Further, they concluded that
    Bannister was 100 percent at fault and that, accordingly, he was not entitled to UM
    coverage, to which an insured is not entitled if the insured is more than fifty percent at
    fault. The entries did not specify which prior logs/information their conclusions were
    based on, or to what extent. At trial, Bannister’s attorney emphasized that, judging from
    the claim log, Dreier appeared to take eleven minutes to reach these conclusions. Also,
    the February 25 date of these entries is the date that Bannister’s attorney stressed at trial
    as the cutoff date of when State Farm stopped “investigating” Bannister’s claim and
    “denied” coverage. See, e.g., ROA v. III-IV at 700, 752, 755, 806, 827, 943, 952.7
    On March 17, 2009, State Farm communicated to Bannister that his UM claim
    would be denied, though the exact context of this communication is unclear. Bannister
    testified he did “recall calling [State Farm after his doctor told him that he had not
    received payment for Bannister’s treatment] and speaking to somebody [who said] that
    State Farm wasn’t liable to pay.” ROA v. III at 878.
    On April 23, 2009, Bannister’s attorney wrote to State Farm formally to request a
    7
    Dreier, who entered the February 25 logs, did not testify at trial. Dreier had been
    transferred to a Missouri division of State Farm by then, and Bannister did not depose
    him. Thus, it was then-liability-agent Eddie Walker (who subsequently left State Farm
    but who nonetheless testified at trial) who testified in regard to the import of Dreier’s
    entry.
    8
    written explanation for the basis of State Farm’s denial of Bannister’s claim. On April
    30, presumably in reaction to that request, State Farm asked one of its liability agents,
    Eddie Walker, to review Bannister’s case. Walker, based on his review of the claim logs,
    recorded (in entries nos. 83-84) that the investigation of Bannister’s case was finished,
    and that
    all evidence would indicate [Bannister] himself was responsible for this
    collision and no one [sic] else. No support on [police report] or even in
    [Bannister’s] own statement that would indicate someone, other than
    himself, is liable for this accident. . . . UM [coverage] would not apply as
    [Bannister] was the proximate cause of this accident. . . . [Police report]
    also notes that there was drinking involved . . . . When we obtained facts
    from [Bannister], he stated he left roadway as vehicle in front of him
    slammed brakes and he swerved to avoid hitting the rear of this vehicle and
    rolled motorcycle. . . .
    ROA v. V at 1488.
    On June 9, 2009, State Farm UM claim representative Dani Conover sent a letter
    to Bannister’s attorney stating the following:
    At this time, we have no evidence Mr. Bannister is legally entitled to
    collect from an uninsured motorist. The police report we have indicates
    [Bannister’s] vehicle left the roadway for an unknown reason, and notes an
    improper start from alcohol-DUI/DWI. If you are aware of any information
    that does support [that Bannister] is legally entitled to collect [on his UM
    policy], please let me know, and we will be happy to review it.
    
    Id. at 1526
    .
    State Farm apparently did not receive a response to that letter, so on July 7,
    2009—after Bannister had filed his complaint in this lawsuit—Conover sent another
    letter to Bannister’s attorney to the same effect.
    9
    3. The lawsuit and the trial
    Bannister filed his complaint in Oklahoma state court on June 24, 2009. He
    alleged both that State Farm had breached its insurance contract with respect to
    Bannister’s UM policy, and that State Farm had breached its duty of good faith and fair
    dealing in denying Bannister’s UM claim. State Farm removed the case to the Western
    District of Oklahoma in November 2009. In September 2010, State Farm moved for
    summary judgment on both of Bannister’s claims, but the district court denied the
    motion. However, before trial, Bannister dropped his breach of contract claim.
    Bannister’s bad faith claim was thus the only claim presented at trial.8
    8
    Bannister acknowledges that he initially “proceed[ed] with both his breach of
    contract and his bad faith causes of action,” but “chose to proceed to trial only on the bad
    faith cause of action.” Aplt. Br. at 5. This tactical decision was reflected as early as the
    December 2010 Final Pretrial Report (Dist. Ct. docket, Doc. 49), which superceded all
    previous pleadings. It is later reflected in the “Statement of the Case” jury instruction.
    ROA v. II at 400.
    We reject Bannister’s argument on appeal that even if he is not entitled to recover
    on his bad-faith claim, he should still recover damages on a breach-of-contract theory in
    the amount allegedly owed under his insurance policy ($125,000). Bannister reasons that
    the jury, in finding State Farm liable on the tort of bad faith, necessarily found in his
    favor on the contract issue. See infra Section II-A-2 (noting that the first element of a
    bad-faith claim is that the insurer was required to pay under the insurance policy).
    Bannister’s assertion that he “did not need to try his breach of contract cause of action
    because the bad faith cause of action, which he did try, included the contract claim,”
    Aplt. Br. at 32-33, is half-correct and half-incorrect. Breach of contract was indeed one
    element the jury needed to find in determining that State Farm was liable for the tort of
    bad faith; and breach of contract can be a standalone theory of recovery. However, it
    does not follow that Bannister may therefore recover based on the jury’s finding of a
    single element of Bannister’s sole asserted claim—the tort of bad faith—when Bannister
    chose to abandon his earlier-asserted contract claim.
    Continued . . .
    10
    A three-day trial occurred at the end of March 2011. Bannister’s attorney called
    seven witnesses to testify: Bannister’s wife, four State Farm claims representatives from
    various departments who in some way dealt with Bannister’s claim, a corporate
    representative from State Farm, and Bannister himself. State Farm cross-examined each
    witness but did not call any witnesses of its own. At that point, State Farm moved for
    ______________________________________
    Cont.
    In arguing for a contrary conclusion, Bannister invokes the concept of lesser-
    included offenses from the criminal law context. However, we are unaware of any
    precedent extending that criminal doctrine to this civil context, such that a forsaken
    contract claim would be transformed into an independent sub-claim of a separate tort
    claim, upon which recovery could be independently awarded. We do not interpret the
    Court of Civil Appeals of Oklahoma’s decision in Cales v. Le Mars Mut. Ins. Co., 
    69 P.3d 1206
     (Okla. Civ. App. 2002), to compel a contrary conclusion. Cales held that a
    new trial was warranted in light of the trial court’s improper decision to bifurcate the
    plaintiff’s breach of contract and bad faith claims into separate trials. 
    Id. at 1208-09
    . In
    doing so, the appellate court reflected:
    We further note the trial court incorrectly describes Cales’ suit as “two
    causes of action.” Cales has but one cause of action: for damages arising
    out of Insurer’s failure to pay Cales’ claim. In support of that cause of
    action Cales has two interrelated theories of recovery. The first, sounding
    in contract, is for damages arising out of Insurer’s failure to pay the claim
    in breach of the insurance contract. The second theory of recovery sounds
    in tort, based on Cales’ allegation that Insurer acted in bad faith by ignoring
    relevant information in its investigation of the claim, leading to its decision
    not to pay. These theories are connected and, as set out below, should not
    be bifurcated.
    
    Id. at 1208
    . However, notwithstanding Cales’s “not[ing]” that the plaintiff’s breach of
    contract claim and bad faith claims comprised “one cause of action,” the actual holding
    of Cales was that it was improper to bifurcate the consideration of the “two interrelated
    theories of recovery” when both theories had been asserted. 
    Id.
     Cales did not hold that a
    plaintiff could recover under the ‘lesser-included’ theory of breach of contract when he
    had earlier chosen to abandon that theory.
    11
    JMOL under Fed. R. Civ. P. 50(a), arguing that the evidence presented at trial did not
    provide a sufficient factual basis upon which the jury could find in favor of Bannister.
    The court took the motion under advisement and allowed the case to proceed. Jury
    instructions and closing arguments followed.
    The jury returned an award of $125,000 in compensatory damages for Bannister.
    Additionally, the jury also found that State Farm had acted recklessly. That finding of
    recklessness triggered consideration of punitive damages, in line with Oklahoma’s
    system of bifurcated consideration of compensatory and punitive damages.9 The jury
    9
    Oklahoma law sets forth a bifurcated procedure that the jury must follow in
    considering actual and punitive damages, when the jury finds that an insurer recklessly
    disregarded its duty of good faith in denying a claim. Under such circumstances,
    the jury, in a separate proceeding after the jury has made such a finding
    [regarding recklessness] and awarded actual damages, may award punitive
    damages in an amount not to exceed the greater of [$100,000 or the amount
    of actual damages awarded]. Any award of punitive damages under this
    subsection awarded in any manner other than as required in this subsection
    shall be void and reversible error.
    
    Okla. Stat. Ann. tit. 23, § 9.1
    (B)(2); see Lierly v. Tidewater Petroleum Corp., 
    139 P.3d 897
    , 906 (Okla. 2006).
    In this case, the jury was informed, prior to its initial calculation of compensatory
    damages:
    following the reading of your verdict in open court, there will be a brief
    additional hearing in which the parties will have the opportunity to present
    evidence and arguments with respect to punitive damages. At the
    conclusion of that additional hearing, you would return to the jury room to
    determine whether to award punitive damages and, if so, in what amount.
    ROA v. II at 413-14 (Jury Instruction no. 13).
    12
    therefore then heard brief testimony relevant to punitive damages. The jury was
    subsequently instructed on punitive damages, with the court explaining the bases on
    which they could award punitive damages and advising the jury that “[i]n no event should
    the punitive damages exceed the amount of actual damages ($125,000) you have
    previously awarded.” ROA v. II at 421 (Supplemental Jury Instruction no. 1). Counsel
    for the parties then made closing arguments with respect to the issue of punitive damages.
    Afterwards, the jury withdrew to deliberate on punitive damages, and then sent the
    following note to the court:
    We the jury were not clear on the dollar amount awarded in actual damages
    on this claim. We thought we could only award the cap amount we kept
    hearing about, which was $125,000. Nor did we know that it would cap the
    amount of our punitive damages.
    
    Id. at 428
    . It is unclear what prompted the jury to realize that it could have awarded more
    than $125,000 in compensatory damages; the jury had not expressed confusion with that
    instruction during the first stage of deliberations.
    The court discussed the note with the parties and heard their input about how to
    proceed. Bannister advocated allowing the jury to recalculate compensatory damages,
    then letting the jury proceed with calculating punitive damages, reflecting that it would
    be wasteful to order a new trial. In contrast, State Farm’s urged the court to declare a
    mistrial, arguing that allowing the jury to recalculate compensatory damages with the
    knowledge that its compensatory award would cap its potential punitive award would
    circumvent the statutory purpose behind bifurcation of the respective considerations. The
    13
    court denied State Farm’s motion and allowed the jury to re-deliberate on compensatory
    damages and then calculate punitive damages afterwards. The jury ultimately returned a
    verdict awarding $350,000 in compensatory damages and $350,000 in punitive damages.
    On April 27, 2011, State Farm renewed its motion for JMOL pursuant to Fed. R.
    Civ. P. 50(b), and moved in the alternative for a new trial pursuant to Fed. R. Civ. P. 59.
    State Farm argued that procedural irregularity in the jury’s consideration of damages, as
    well as allegedly improper references made by Bannister’s counsel through trial,
    warranted a new trial. After the motions were briefed and a hearing was held, the court
    granted State Farm’s renewed motion for JMOL, concluding that Bannister’s bad faith
    claim could not properly be submitted to the jury as a matter of law. The court
    determined that the evidence showed that State Farm denied Bannister’s claim based on
    facts that reasonably supported a legitimate dispute as to whether Bannister was majority
    at fault in his accident; and that no evidence suggested that further investigation would
    have undermined the State Farm’s legitimate basis for disputing the claim.
    Meanwhile, with respect to State Farm’s alternative motion for a new trial,
    pursuant to Fed. R. Civ. P. 59(c) the court conditionally denied that motion in the event
    that the court’s JMOL decision were reversed. The court concluded that the complained-
    of matters—including the peculiar sequence of events in the jury’s calculation of
    damages, and allegedly improper statements by Bannister’s counsel at trial—did not
    warrant a new trial. The court further noted that many of counsel’s statements had not
    been objected to during trial.
    14
    Finally, the court denied Bannister’s motion for attorney fees. The Court observed
    that Bannister was not a prevailing party and determined that, even if he were, State
    Farm’s defense in the lawsuit was neither asserted in bad faith, ungrounded in fact, nor
    unwarranted by existing law, as required by statute for recovery of attorney’s fees in this
    case.
    Bannister appealed the district court’s grant of JMOL to State Farm, as well as the
    court’s denial of Bannister’s motion for attorney’s fees. State Farm cross-appealed the
    court’s conditional denial of its alternative motion for a new trial.
    II. DISCUSSION
    A. Legal Standards
    1. Appellate review of JMOL generally
    We review de novo the district court’s grant of State Farm’s renewed motion for
    JMOL, applying the same standard as the district court. See Bristol v. Bd. Of Cnty.
    Comm’rs of Cnty. of Clear Creek, 
    312 F.3d 1213
    , 1216 (10th Cir. 2002) (en banc).
    Accordingly, we will affirm the district court if we determine that “a reasonable jury
    would not have [had] a legally sufficient evidentiary basis to find for” Bannister on his
    bad faith claim. Fed. R. Civ. P. 50(a)(1); see also Bristol, 
    312 F.3d at 1216
    . In making
    that determination, we “construe the evidence and inferences most favorably to the
    nonmoving party, and refrain from weighing the evidence, passing on the credibility of
    witnesses, or substituting our judgment for that of the jury.” Magnum Foods, Inc. v.
    Cont’l Cas. Co., 
    36 F.3d 1491
    , 1503 (10th Cir. 1994) (citation omitted). Further,
    15
    “[a]lthough federal law dictates [the procedural question of] whether a judgment as a
    matter of law is appropriate, . . . in a diversity case we examine the evidence in terms of
    the underlying burden of proof as dictated by state law.” 
    Id.
     (citations omitted).
    2. The tort of bad faith
    The law of bad faith was properly encapsulated by Jury Instruction no. 10 in this
    case. The instruction on the elements of Bannister’s bad faith claim (i.e., breach of the
    duty good faith and fair dealing) was that “[Bannister] must prove each of the following
    elements by the greater weight of the evidence”:
    FIRST: That State Farm was required under the insurance policies to pay
    Mr. Bannister’s uninsured motorist claim10;
    SECOND: That State Farm’s refusal to pay the claim was unreasonable
    under the circumstances because
    1) State Farm did not perform a proper investigation,
    2) State Farm did not evaluate the results of the investigation properly,
    or
    3) State Farm had no reasonable basis for the refusal.
    THIRD: That State Farm did not deal fairly and in good faith with Mr.
    Bannister; and
    FOURTH: That the violation by State Farm of its duty of good faith and
    fair dealing was the direct cause of the damages sustained by Mr. Bannister
    and sought to be recovered in this action.
    ROA v. II at 408 (emphases added, footnote added). The instruction went on to state:
    10
    State Farm argues that it is deserves JMOL on the independent basis of this first
    element, asserting that “undisputed evidence demonstrates that Bannister was majority at
    fault.” Aple. R. Br. at 1. Because of our other rulings in this appeal, we need not address
    that argument.
    16
    In determining whether the insurer had a good faith belief in some
    justifiable reason for denying payment at the time it made its decision on
    the insurance claim, you [the jury] may only consider evidence which the
    insurer had at the time it decided to deny the claim. In this action there is a
    factual dispute about when that decision was made.
    An insurer’s refusal to pay a claim is not bad faith when there is a
    legitimate dispute concerning coverage; however, merely because there is a
    reasonable basis that an insurance company could invoke to deny a claim
    does not necessarily immunize the insurer from a bad faith claim if, in fact,
    it did not actually rely on that asserted reasonable basis and instead took
    action in bad faith. The insurer is not required to show that its good faith
    belief was correct.
    Id. at 409 (emphasis added).
    That instruction properly stated the elements of the tort of bad faith. See Badillo
    v. Mid Century Ins. Co., 
    121 P.3d 1080
    , 1093 (Okla. 2005) (citing Oklahoma Uniform
    Jury Instructions Civ (2d) 22.3). The district court properly recognized that it was not
    bad faith per se for State Farm to resort to the judicial forum to settle legitimate disputes
    over insurance claims. See Garnett v. Gov’t Employees Ins. Co., 
    186 P.3d 935
    , 944
    (Okla. 2008). The court correctly acknowledged that the decisive questions are whether
    State Farm’s denial of coverage was based on a good-faith reason at the time it decided to
    deny coverage, and also whether State Farm conducted an investigation reasonably
    appropriate under the circumstances to determine the validity of Bannister’s claim. See
    Buzzard v. Farmers Ins. Co., Inc., 
    824 P.2d 1105
    , 1109 (Okla. 1991).
    3. Appellate review of JMOL in bad faith cases specifically
    “[A]s a matter of law . . . no reasonable inference of bad faith arises”—and hence
    JMOL is warranted for the insurer—“when an insurer denies a claim solely because of
    17
    the existence of a legitimate dispute.” Oulds v. Principal Mut. Life Ins. Co., 
    6 F.3d 1431
    ,
    1442 (10th Cir. 1993). However, “a legitimate dispute as to coverage will not act as an
    impenetrable shield against a valid claim of bad faith.” Timberlake Constr. Co. v. U.S.
    Fidelity & Guar. Co., 
    71 F.3d 335
    , 343 (10th Cir. 1995). Thus, in “cases in which the
    question of bad faith [is] required to be submitted to the jury, the evidence of the
    insurer’s defense to the underlying claim [is] so weak that a reasonable inference could
    be drawn that the insurer denied the claim in bad faith.” Oulds, 
    6 F.3d at 1442
    ; see also
    Timberlake, 
    71 F.3d at 343
     (“In sum, ‘in order to establish such a [bad faith] claim, the
    insured must present evidence from which a reasonable jury could conclude that the
    insurer did not have a reasonable good faith belief’ [for denying the claim].” (quoting
    Oulds, 
    6 F.3d at 1436
    ) (alteration marks omitted)). In other words, if the evidence at trial
    demonstrates that “there was a legitimate dispute as to coverage under the policy, and
    that [the insurer’s] position was reasonable in light of the facts known or knowable to it at
    the time it denied [the] claim,” then “as a matter of law [the insurer] did not breach the
    duty of good faith merely by refusing to pay [the] claim.” Timberlake, 
    71 F.3d at 344
    .
    Rather, the claimant must “produce additional evidence of bad faith in order to send the
    issue to the jury.” 
    Id.
    To that end, a jury may decide the issue of bad faith, even when the evidence
    reveals a legitimate possible basis for a dispute, if the claimant submitted evidence that
    the insurer did not actually rely on that legitimate basis but rather denied the claim for an
    illegitimate reason, such as a “systematic, bad faith scheme of canceling policies without
    18
    . . . good cause,” Vining v. Enter. Fin. Grp., Inc., 
    148 F.3d 1206
    , 1214; see also Capstick
    v. Allstate Ins. Co., 
    998 F.2d 810
    , 814-15 (10th Cir. 1993) (affirming denial of JMOL
    where “from the very beginning without any investigation, [the insurer] treated the claim
    as a ‘suspicious loss’” and “denied coverage without making any other bona fide
    investigation”).
    Another instance in which the jury may decide the issue is if there is evidence that
    the insurer “failed to adequately investigate [the] claim.” Timberlake, 
    71 F.3d at 345
    .
    Crucially, however, “when a bad faith claim is premised on inadequate investigation, the
    [claimant] must make a showing that material facts were overlooked or that a more
    thorough investigation would have produced relevant information” that would have
    delegitimized the insurer’s dispute of the claim. 
    Id.
     That is, evidence of inadequate
    investigation must “suggest a sham defense or an intentional disregard of
    uncontrovertible facts” in order to be put to a jury. 
    Id.
     To illustrate, where an insurer had
    interviewed a claimant, but had failed to question key individuals and therefore “had not
    completed an investigation [but rather] had only gotten one side of the story,” JMOL was
    still warranted when such questioning “would not have changed the underlying facts
    already known to [the insurer], facts from which [the insurer] was entitled to form a
    reasonable belief” regarding its justification for denying the claim. 
    Id.
    B. Analysis
    We hold that a reasonable jury could not find, based on the evidence produced in
    this case, that State Farm did not actually rely on a legitimate reason in disputing
    19
    Bannister’s insurance claim. Furthermore, we discern no evidence showing that State
    Farm failed adequately to evaluate or to investigate Bannister’s insurance claim such that
    additional investigation would have materially altered the legitimate factual basis on
    which State Farm disputed Bannister’s claim.
    1. Whether State Farm actually based its denial of Bannister’s claim on a
    legitimate dispute as to whether Bannister’s policy covered the accident
    First, to evaluate both the reasonableness of State Farm’s denial of Bannister’s
    claim in light of State Farm’s knowledge at the time, it is necessary to identify the date of
    that denial.11 See Timberlake, 
    71 F.3d at 344
     (evaluating reasonableness “in light of the
    facts known or knowable to [the insurer] at the time it denied [the insured’s] claim”
    (citing Buzzard v. McDanel, 
    736 P.2d 157
    , 159 (Okla. 1987))); see also ROA v. II at 409
    (Jury Instruction no. 10). On this record and our required standard of review, we
    conclude that State Farm denied Bannister’s claim on February 25, 2009, although if this
    issue were reviewed de novo we might well conclude that the proper date to use for the
    denial was March 17, 2009, or even June 9, 2009. Throughout the trial, Bannister
    11
    On appeal, Bannister emphasizes that he was never sent a proper claim-rejection
    letter by State Farm that explained the reasons for his denial, in violation of the
    Oklahoma Unfair Claims Settlement Practices Act, 
    Okla. Stat. Ann. tit. 36, § 1250.7
    (A).
    However, an insurer’s violation of the regulation under that Act does not give rise to an
    inference of bad faith in the context of this tort action. See Beers v. Hillory, 
    241 P.3d 285
    , 294 (Okla. Civ. App. 2010) (“[A] violation of the Act does not necessarily establish
    bad faith. An insurer may carelessly fail to perform some duty required by the statute
    with such frequency to warrant administrative sanction, but that does not establish more
    than negligent conduct in any individual case.”); Badillo, 121 P.3d at 1094 (holding that,
    in bad faith cases, “the minimum level of culpability necessary for liability against an
    insurer to attach is more than simple negligence”).
    20
    continually invoked the February 25 date as the end of State Farm’s affirmative
    investigatory efforts and the day it definitively concluded that Bannister was at fault.
    That theory that the denial occurred on February 25 is supported by evidence (namely,
    the claim log as well as testimony), so drawing all reasonable inferences in favor of
    Bannister, see Magnum Foods, 
    36 F.3d at 1503
    , it is appropriate to use February 25 for
    our bad faith analysis.12
    On February 25, the facts known to State Farm included that Bannister had been
    involved with a single-vehicle accident in which Bannister was unable safely to stop
    when the car in front of him braked suddenly; that the police report showed that
    Bannister had been driving under the influence of alcohol; and that there were no
    identified witnesses. These facts make State Farm’s dispute of Bannister’s eligibility for
    recovery reasonable. For one, Bannister had a duty to leave a safe, appropriate distance
    between himself and the vehicle ahead. See 
    Okla. Stat. Ann. tit. 47, § 11-310
    (a) (“The
    driver of a motor vehicle shall not follow another vehicle more closely than is reasonable
    12
    A State Farm employee in the UM division, Dani Conover, testified that
    Bannister’s claim was still “open” as late as May 1, 2009, insofar as she was still
    reviewing Bannister’s claim log at that point. However, Conover more or less conceded
    that she was aware of no further investigation done since February 25. Moreover, State
    Farm’s corporate representative at trial, Jason Taylor, agreed that a determination of
    100% liability on the part of Bannister had been made by February 25, although Taylor
    added that State Farm was open to considering any additional information and that it had
    asked Bannister to provide any he had. But in any case, drawing all reasonable
    inferences from the evidence in favor of Bannister, February 25 seems to be the
    appropriate date upon which to review the basis for State Farm’s decision to deny
    coverage.
    21
    and prudent . . . .”).13 Bannister’s inability timely to stop when the vehicle in front of him
    applied its brakes tends to show that Bannister failed to observe his duty to leave a safe
    stopping distance, and that he was therefore at fault in his accident. The reasonableness
    of deeming Bannister to be at fault based on that fact is bolstered by the fact that the
    driver in front of Bannister was able to stop safely even though that driver, not Bannister,
    was the one who was allegedly cut off by another vehicle.14 Additionally, the police
    report indicated that Bannister had been driving under the influence of alcohol. While
    that fact did not dispositively preclude coverage under Bannister’s policy, it certainly
    added to the reasonableness of finding Bannister negligent. Meanwhile, there were no
    other identified witnesses who might have contradicted any of those facts.
    Next, the evidence showed that State Farm, from its initial reaction to Bannister’s
    claim to its ultimate denial of the claim, actually relied on the legitimate reasons listed
    above. It is clear that State Farm relied at least on the facts about Bannister not having
    sufficient space between him and the vehicle in front of him timely to brake and avoid
    13
    At trial, Bannister agreed that he had a duty to leave sufficient space between
    himself and the vehicle ahead to be able timely to react, and also said that he knew such
    space was greater for motorcycles, which cannot decelerate as swiftly as cars.
    14
    State Farm liability representative Walker testified that a motorist might not be
    negligent if he hits something that jumps directly in front of him to cause an accident:
    e.g., in the present scenario, the beige car might not have been negligent if it had hit the
    red car that cut it off. Walker emphasized that such a situation is crucially different, for
    the purposes of negligence evaluations, from Bannister’s case, where the vehicle
    immediately in front of him simply stopped suddenly, and Bannister crashed as a result of
    that vehicle braking.
    22
    laying down his motorcycle. That itself is sufficient to give rise to a reasonable dispute
    about whether Bannister was majority at fault. State Farm may also have relied on the
    police report’s notation that Bannister was driving under the influence; and that would
    only have bolstered the already reasonably supported conclusion that Bannister was
    majority at fault.
    The first material entry in Bannister’s claim log (entry no. 7, on January 23, 2009)
    recorded the facts of the accident as reported by Bannister and relayed to State Farm
    through his wife the day after the crash. At trial, Wendy Jeffus, who spoke with
    Bannister’s wife and entered that log, said that the facts she recorded—i.e., that a vehicle
    in front of Bannister slammed on its brakes, causing Bannister to crash—were
    insufficient at that point to dictate who was at fault, since it was just one person’s version
    of what had happened. However, Jeffus testified that if those facts were confirmed
    through subsequent investigation, then there would be reason to believe that Bannister
    was at least majority at fault for the accident, which would have disqualified him from
    UM coverage. The next material log entry (entry no. 14, on January 26, 2009), said that
    there was insufficient information on file to make a determination at that point and that
    further investigation was appropriate; but that it appeared that Bannister was majority at
    fault based on reasonable inferences from the facts of a single-vehicle crash where the car
    in front of the claimant stopped suddenly, and the claimant had insufficient room to stop
    himself.
    After those entries, State Farm continued to rely on the crash scenario, but it also
    23
    became aware that the police report, obtained on by State Farm on February 17, 2009,
    indicated that Bannister had been driving under the influence of alcohol. State Farm
    representative Walker testified that Bannister’s alcohol consumption was “a piece of the
    investigation,” though not something upon which State Farm did, or could have, based its
    determination of liability exclusively. ROA v. III at 710. It is uncertain the extent to
    which State Farm took the police report’s driving-under-the-influence notation into
    account in assessing Bannister’s fault. To that end, the February 25 entry only noted that
    fact in the “comment” section of the log rather than the “analysis” section, though it is
    unclear how much of a difference, if any, such placement makes. 
    Id. at 767
    . Walker—
    who reflected on but did not himself make the February 25 log entry determining
    Bannister’s fault—testified that he would have denied Bannister’s claim based solely on
    the crash scenario, independent of the driving-under-the-influence factor.
    In sum, from the initial consideration of Bannister’s claim through the February 25
    denial of it, State Farm actually at least relied on the facts of the crash scenario—a single-
    vehicle accident where the car in front of the claimant braked suddenly, and the claimant
    had insufficient space timely to stop—and possibly also on the police report’s notation
    that Bannister had been driving under the influence. The former reason alone is
    sufficient to create a legitimate dispute as to whether Bannister was at least majority at
    fault, and the latter reason would only add to the reasonableness of deeming Bannister at
    fault. Meanwhile, Bannister put forth no evidence that State Farm arbitrarily prejudged
    his claim or was otherwise engaged in systematic bad-faith denials of claims. See
    24
    Vining, 148 F.3d at 1214; Capstick, 
    998 F.2d at 815
    . Thus, we conclude that Bannister
    failed to produce evidence that State Farm did not actually dispute his insurance claim in
    good faith, as necessary to submit his legal claim to the jury under this theory of bad
    faith. See Timberlake, 
    71 F.3d at 344
    .
    2. Whether State Farm failed adequately to investigate Bannister’s claim
    Even though State Farm had a reasonable, actually-relied-upon basis for denying
    Bannister’s claim, the bad faith issue could still be sent to the jury to the extent that
    Bannister’s theory is “premised on inadequate investigation.” Timberlake, 
    71 F.3d at 345
    . However, to resist JMOL based on a theory of inadequate investigation, Bannister
    “must [have] ma[d]e a showing that material facts were overlooked or that a more
    thorough investigation would have produced relevant information.” 
    Id.
    We can reject this theory even assuming arguendo that State Farm’s investigation
    of Bannister’s claim was inadequate.15 Bannister’s argument fails on this ground because
    all that any further, reasonable investigation would have revealed, according to the record
    15
    On the record before us, State Farm’s investigation arguably was adequate.
    State Farm obtained Bannister’s side of the story through his wife’s initial account as
    well as through Mercado’s later conversation with him, although that was under the
    auspices of a collision evaluation, not a liability interview; it obtained the police report;
    and there were no other identifiable witnesses to consult. Also, it is not apparent that
    State Farm was required, either by company policy or by law, electronically to record an
    interview with Bannister, as Bannister argues should have been done.
    In any event, that does not change the fact that, to survive JMOL based on a theory
    of inadequate investigation, Bannister needed to put forth evidence showing that further
    investigation would have delegitimized State Farm’s basis for disputing the claim. See
    Timberlake, 
    71 F.3d at 345
    . As discussed below, he did not do that.
    25
    before us, is what Bannister himself testified to at trial; and that testimony “would not
    have changed the underlying facts already known to [State Farm], facts from which [State
    Farm] was entitled to form a reasonable belief” regarding its justification for denying
    Bannister’s claim. Timberlake, 
    71 F.3d at 345
    . That is, taking into account everything
    Bannister said at trial, State Farm still would have had before it the material facts of a
    single-vehicle accident with no identified witnesses (other than the claimant), where the
    claimant had not left sufficient space to brake and avoid crashing, and where the claimant
    was driving after the consumption of alcohol. Bannister’s testimony did not contradict
    any material facts upon which State Farm based its legitimate dispute regarding
    Bannister’s negligence.
    On the contrary, upon further investigation Bannister would only have appeared
    more negligent, since the fact that he was speeding—a fact to which Bannister testified,
    but which was never noted in the claim log—would have come out. Further, if State
    Farm had obtained Bannister’s hospital record from the aftermath of the accident (or had
    asked Bannister about the record and had received truthful answers), State Farm would
    have discovered that the hospital record indicated that Bannister’s blood-alcohol level
    was 0.09.
    On appeal, Bannister asserts that State Farm’s investigation was inadequate, but he
    does not explain how further investigation would have helped his case for recovery under
    his insurance policy. For example, Bannister argues that “had State Farm taken a
    recorded statement, it could have and should have asked Bannister whether he was (1)
    26
    drunk, (2) speeding and/or (3) following too closely.” Aplt. Br. at 23. However, he does
    not explain how the answers to those questions would have altered the factual basis on
    which State Farm reasonably disputed coverage. Critically, Bannister’s truthful answers
    to those questions would have confirmed that his blood-alcohol level was above the legal
    limit16; that he was speeding by 5-10 mph; and that he was following too closely, as he
    did not leave sufficient room in front of him to stop, whereas the car in front of him—the
    car that was allegedly cut off, and therefore had greater reason to crash—managed safely
    to brake without collision.
    Bannister argues that “[p]erhaps most significantly, a statement from Bannister
    would have provided State Farm with the opportunity to assess for itself Bannister’s
    credibility,” id. at 24, but Bannister’s credibility in itself was not a “fact” upon which
    claim coverage was disputed, see Timberlake, 
    71 F.3d at 345
    . Bannister asserts that “a
    situation like this . . . turns on how an accident happened and the only evidence of that is
    the insured’s testimony.” Aplt. Br. at 24. But again, assuming State Farm had deemed
    Bannister credible, and Bannister had recounted in an interview the same things he
    recounted at trial, State Farm still would have had the same material facts before it. And
    as discussed above, those facts support a good-faith dispute of Bannister’s claim-recovery
    eligibility, given the apparent degree of Bannister’s fault.
    Bannister also argues in his reply brief that State Farm should have made efforts to
    16
    In Oklahoma, it is unlawful to operate a motor vehicle with a blood-alcohol
    concentration of 0.08 or more. 
    Okla. Stat. Ann. tit. 47, § 11-902
    (A)(1).
    27
    contact the alleged driver from a truck behind Bannister who stopped to help him after
    the accident, as that person could have been a witness. However, that alleged individual
    was never noted in a police report, and Bannister never gave, nor at trial professed to
    possess, any contact information for that individual. Accordingly, State Farm—whose
    duty it was simply to undertake an investigation that was reasonable under the
    circumstances, see Buzzard, 824 P.2d at 1109—cannot be faulted for not seeking out
    some unknown and reasonably unknowable person.
    In conclusion, Bannister failed to “make a showing that material facts were
    overlooked or that a more thorough investigation would have produced relevant
    information” that would have delegitimized the insurer’s dispute of the claim.
    Timberlake, 
    71 F.3d at 345
    . As such, his inadequate-investigation theory of bad faith is
    without merit, and JMOL in favor of State Farm was appropriate.
    III. CONCLUSION
    For the foregoing reasons, we AFFIRM the district court’s grant of JMOL to State
    Farm.17 We therefore DISMISS AS MOOT State Farm’s cross appeal.
    17
    In light of our holding, Bannister is not a prevailing party, and therefore is not
    entitled to attorney’s fees under 
    Okla. Stat. Ann. tit. 23, § 103
    , as he contends.
    28