Full opinion text
OPINION AND ORDER BUCHWALD, District Judge. Plaintiff Kathleen Madaline Jarvis (“plaintiff’ or “Jarvis”) filed this diversity action alleging that she was injured on July 14,1991 when the 1991 Ford Aerostar she was operating suddenly accelerated and traveled into a ditch. Plaintiff alleged that a design defect in the automobile’s cruise control system caused the sudden acceleration. After a two-week trial, the jury returned a verdict finding that plaintiffs Aerostar was not defectively designed but that Ford Motor Company (“defendant” or “Ford”) nevertheless had been negligent in the design of the vehicle. Additionally, the jury found Jarvis’ own negligence to have been 35% responsible for her accident. The jury awarded damages for past and future medical insurance premiums, loss of earnings, and pain and suffering. Presently pending is defendant’s motion of August 11, 1999 (“Deft. 50(b) Memo”), pursuant to Fed.R.Civ.P. 50(b), for an order granting judgment to defendant as a matter of law, or, in the alternative, granting a new trial. In support of its motion, defendant argues, first, that the jury’s verdict was inconsistent because its finding that Ford was not strictly liable negates its finding of negligence and, second, that the negligence finding is against the weight of the evidence. Defendant also moved on August 11, 1999 (“Deft. 4545 Memo”) for an order reducing the verdict by the amount of collateral source payments plaintiff has received and will receive in the future pursuant to CPLR § 4545. Finally, defendant has asked the Court to rule on its motion to dismiss plaintiffs punitive damages claim originally filed on June 6, 1997 (“Deft. Punitives Memo”). Plaintiff has offered a response to each of these motions. (Respectively, “PI. 50(b) Mem.”; “PI. 4545 Mem.”; and “PL Punitives Reply”). We will discuss' each of these motions in turn, with particular attention first to the issue of the jury verdict inconsistency and then to the sufficiency of the plaintiffs evidence. BACKGROUND Plaintiff filed this products liability action pleading design defect causes of action under both strict liability and negligence theories. Plaintiff testified during the trial that she started her six-day-old Aerostar on July 14, 1991 in the driveway of her home in Woodstock, New York, with her foot on the brake and the car in “park.” Tr. 73. According to the testimony, the vehicle proceeded to suddenly accelerate down the driveway. Tr. 74-75. Plaintiff testified that she attempted to stop the vehicle by stepping on the brake with both feet, but that the vehicle did not stop. Tr. 83-84. The vehicle traveled approximately 330 feet down the driveway, came to a stop in a drainage ditch, and overturned. Tr. 695. The essence of plaintiffs claim is that the stand-alone cruise control in the 1991 Ford Aerostar was defectively designed because, upon ignition, battery voltage will be supplied to the cruise control system (in particular, the servo component) and thus that the simultaneous occurrence of two transient electrical events could result in the car’s throttle going to a wide open condition, causing a sudden, unintended acceleration of the vehicle without driver input. It was further plaintiffs position that the mechanical features of the car were not adequate to overcome an electrically-caused wide open throttle condition, i.e., that the design lacked a failsafe mechanism. Ford’s position was, first, that the vehicle was properly designed to overcome even the statistically remote possibility that two simultaneous electrical events could occur, resulting in a wide open throttle. Second, Ford’s position was that there was no evidence that the two simultaneous electrical faults postulated by the plaintiffs electrical engineering expert ever occurred in the plaintiffs vehicle, or that the vehicle did not have adequate mechanical back-ups, and indeed that the physical evidence was to the contrary. Moreover, Ford presented another explanation for the events which plaintiff described and which considered the testimony of other witnesses to the events as well as the physical evidence. DISCUSSION I. Verdict Inconsistencies Defendants initially make this application to set aside the verdict as inherently inconsistent based on the jury’s answer to two special verdict questions. First, the jury answered “no” to the question, “Do you find by a preponderance of the evidence that the cruise control system of the 1991 Ford Aerostar was designed in a defective manner?” Court Ex. 12. Second, the jury answered “yes” to the question, “Do you find by a preponderance of the evidence that the defendant Ford Motor Company was negligent in the design of the cruise control system in the 1991 Ford Aerostar?” Id. Defendant reasons that the jury’s finding that the cruise control was not defective “precludes a finding, legally and logically, that Ford was negligent in designing the cruise control system.” Deft. 50(b) Memo at 3. Plaintiff argues that the Court must make every attempt to reconcile the jury’s answers on the verdict form to find a consistent way to interpret the verdict, even if such a reading is “strained.” PI. 50(b) Mem. at 7 (citing McGuire v. Russell Miller, Inc., 1 F.3d 1306, 1311 (2d Cir.1993)). We find that the two jury determinations cannot be reconciled. For nearly twenty years, lower courts in New York have found that the two theories of negligence and strict liability for design defect are “almost functionally equivalent.” DeRosa v. Remington Arms Co., 509 F.Supp. 762, 766-67 (E.D.N.Y.1981) (Weinstein, J.) (citing Lancaster Silo & Block v. Northern Propane Gas, 75 A.D.2d 55, 427 N.Y.S.2d 1009, 1013 (4th Dep’t 1980)). In Denny v. Ford Motor Co., 87 N.Y.2d 248, 257-58, 639 N.Y.S.2d 250, 255-56, 662 N.E.2d 730, 735-36 (1995), New York’s Court of Appeals formally endorsed this proposition for all design defect claims, citing several cases and law review articles to establish that negligence and strict liability design defect claims are “functionally synonymous” since the strict liability analysis is “negligence-inspired.” See also Gonzalez v. Morflo Industries, Inc., 931 F.Supp. 159, 164 n. 3 (E.D.N.Y.1996) (citing Denny for proposition that “negligence and strict liability design defect claims are virtually indistinguishable”); Enright v. Eli Lilly & Co., 77 N.Y.2d 377, 387, 568 N.Y.S.2d 550, 555-56, 570 N.E.2d 198, 203-04 (1991) (failure to warn claim, “though it may be couched in terms of strict liability, is indistinguishable from a negligence claim”). As a consequence of this functional equivalence, defendant’s position that a finding of no defect and a finding of negligence are inconsistent as a matter of law in a product liability case is well supported. The Restatement (Third) of Torts instructs that “two or more factually identical defective-design claims should not be submitted to the trier of fact in the same case under different doctrinal labels.” Restatement (Third) Torts § 2, cmt. n (warning that the submission of the same claim on “multiple theories of recovery ... would generate confusion and may well result in inconsistent results”). Accordingly, in New York, as in virtually all jurisdictions, in a products liability action premised on a design defect theory, a jury finding that the product is not defective precludes a parallel finding that the manufacturer is negligent in designing the product. Lundgren v. McColgin, 96 A.D.2d 706, 464 N.Y.S.2d 317 (4th Dep’t 1983). After a jury finding of no strict 'liability but negligence in a pure design defect case, the Lundgren court held, “[t]he jury could not have concluded that [defendant] negligently designed the override system and at the same time conclude that it was reasonably safe for its intended use.” 96 A.D.2d at 706, 464 N.Y.S.2d at 318. Similarly, in Tipton v. Michelin Tire Co., 101 F.3d 1145, 1150 (6th Cir.1996), the Sixth Circuit held that “proof of a defective product is essential to the products liability or the negligence claim.” According to the Sixth Circuit, the distinction between the two claims is of “no practical significance” and a jury finding of negligence without a finding of strict liability is “legally inconsistent.” Id. (citations omitted). Another example is the case of Witt v. Norfe, Inc., 725 F.2d 1277 (11th Cir.1984). In Witt, the jury found for the defendant on the strict liability question as to whether a shower door was defective but, at the same time, found the defendant negligent. The Eleventh Circuit regarded the two findings as “irreconcilably contradictory.” 725 F.2d at 1280. See also Romero v. International Harvester Co., 979 F.2d 1444 (10th Cir.1992) (jury verdict that defendant had “failed to exercise reasonable care” irreconcilable with jury verdict in favor of defendant on negligent design and strict liability claims). In this case, the jury was instructed with respect to the strict liability cause of action that: A product is defective if it is not reasonably safe — that is, if the product as designed is so likely to be harmful to persons or property that a reasonable person who had actual knowledge of its potential for producing injury would conclude that it should not have been marketed as designed. Tr. 1589-90. With respect to the negligence cause of action, the jury was instructed: If you find that the 1991 Aerostar which plaintiff claims caused her injury was defective when put on the market by the Ford Motor Company, that with such a defect the vehicle was reasonably certain to be dangerous when put to normal use, that Ford failed to use reasonable care in designing the vehicle or in inspecting and testing it for defects, or that even though Ford used reasonable care in designing, inspecting, and testing the 1991 Aerostar, that Ford learned of the defect before putting the product on the market but did nothing about it, you will find that Ford was negligent. Tr. 1592-93. As is evident in the charge, in order for the jury to find Ford negligent, it necessarily had to find that some defect existed in the cruise control. Thus, the jury’s finding that there was no defect in the cruise control is inconsistent with its finding that Ford was negligent in designing the cruise control. In apparent recognition of the force of Ford’s argument, plaintiff has endeavored to recast her position from the pretrial and trial phases of this case. She now attempts to argue that a negligence finding is compatible with the jury’s finding of no design defect because although the cruise control “in and of itself is not defective when operated in accordance with its intended design,” the negligence lay in “designing the components of the cruise control system ... in such a way that those components can suddenly activate the throttle in a stationary vehicle.” PL 50(b) Mem. at 12. Assuming that this distinction is more than semantics, plaintiff is attempting to distinguish a defect in the intended design from the plaintiffs use of the Aerostar. This interpretation, though, contradicts plaintiffs earlier position that she got in the car and turned it on according to its intended design. She testified, “I got up into the van, you know, I closed the door, put the seat belt on ... and I had my right foot on top of the brake, because that’s how dad taught me back in 1967.” Tr. 73. Plaintiff never introduced any evidence that she operated the vehicle in an unintended but reasonably foreseeable manner. This entire trial centered around whether the cruise control could be activated in a way that was unintended, regardless of driver input. Moreover, throughout the pretrial and trial phases, plaintiffs counsel was adamant that the jury could only reach one of two possible conclusions. As he asked the jurors, “this all comes down to was this driver error? Or was this a malfunction of the cruise control? In other words did [Jarvis] malfunction, or did the machine?”. Tr. 32. If, as plaintiffs counsel had maintained, these were the only options, then the jury’s finding that the cruise control was not designed defectively surely negates Ford’s negligence. As a result, the verdicts are irreconcilable. A. Waiver Plaintiff alternatively argues that, even if the jury’s verdict was inconsistent, Ford waived its challenge because it failed to object to either the charge or the special verdict questions. Citing Lavoie v. Pacific Press & Shear Co., 975 F.2d 48 (2d Cir.1992), plaintiff argues that defendant was required to object to the jury instructions prior to the jury retiring. PI. 50(b) Mem. at 13. Defendant, who made the objection immediately after the jury’s verdict and before the jury was discharged, cites Auwood v. Harry Brandt Booking Office, Inc., 850 F.2d 884 (2d Cir.1988), in which the Second Circuit refused to find a waiver where counsel made the objection after the jury .was discharged but before the trial court entered judgment. We need not attempt to reconcile the Lavoie and Auwood decisions because, even assuming plaintiffs interpretation of the law is correct, defense counsel’s failure to object cannot constitute a waiver because we clearly ruled in a pretrial conference that we intended to charge both negligence and strict liability. Pretrial Conference of July 12, 1999 at 27-28 (hereinafter “Pretrial Conf.”). At that conference, and at times previous to the conference, defense counsel made it clear that they believed the case should have gone to the jury on either a negligence or strict liability theory, but not both. Although it is true that defense counsel did not specifically object to our charge and special verdict questions on the ground that they could lead to inconsistent results, we cannot find a waiver because the defense’s fundamental position on this issue had previously been expressed to the Court and, after our ruling at the pretrial conference, it was reasonable for defendant to conclude that future efforts to object would be unavailing. See Anderson v. Branen, 17 F.3d 552, 556-57 (2d Cir.1994) (citing Ostrowski v. Atlantic Mut. Ins. Cos., 968 F.2d 171, 178 (2d Cir.1992)). B. Inconsistent Verdicts and Judgment as a Matter of Law Having decided that the verdict is fatally inconsistent, the next question is whether we should order a new trial or grant judgment as a matter of law to defendant. Many courts faced with inconsistent design defect jury verdicts have found their plaintiffs’ cases insufficient to sustain a jury verdict under any theory. For example, in the Tipton case described above, the court, finding that “the negligence theories that were not eliminated by the verdict of no defect ... were devoid of evidentiary support,” directed the district court to enter judgment in favor of the defendant. 101 F.3d at 1151. Similarly, in North American Catamaran Racing Assoc., Inc. v. McCollister, 480 So.2d 669, 671 (Fla.Dist.Ct.App.1985), review denied, 492 So.2d 1333 (Fla.1986), a case in Florida in which a jury found a catamaran not defective but nonetheless found for the plaintiff, the court reversed the judgment and remanded for entry of judgment for defendant. See also Galdieri-Ambrosini v. National Realty & Dev. Corp., 136 F.3d 276, 286 (2d Cir.1998) (lack of specificity of jury verdict was harmless in light of insufficiency of evidence); Romero, 979 F.2d at 1446 (entering judgment for defendant despite split jury verdict for plaintiff on “reasonable care” claim and for defendant on negligent design and strict liability claims). Other courts have remanded cases with inconsistent verdicts back to the trial court for a new trial. See Witt, 725 F.2d at 1280 (“Because these findings are irreconcilably contradictory, we must remand for a new trial on the issue of liability”); Lambert v. General Motors, 67 Cal.App.4th at 1186, 79 Cal.Rptr.2d 657 (“The proper disposition, in our view, is to remand for a new trial”); Lundgren, 96 A.D.2d at 707, 464 N.Y.S.2d at 318 (“Judgment unanimously reversed on the law and facts ... and a new trial granted”). Because the case law is unclear as to whether the appropriate result is to grant a new trial or grant defendant judgment as a matter of law based on the inconsistent verdict, we instead analyze the evidence presented to determine whether, assuming that the jury had found for plaintiff on the strict- liability theory, we would have granted defendant judgment as a matter of law based on the sufficiency of the evidence adduced at trial. II. Sufficiency of the Evidence Our discussion of the sufficiency of Jarvis’ evidence is structured as follows: first, we will discuss the legal standards for judgment as a matter of law; second, we will discuss the controlling case law for design defect claims in New York; third, we will discuss the need for expert evidence in scientifically complex design defect cases such as this one; and fourth, we will discuss whether plaintiff has met her burden of proving that the claimed defect actually existed and caused plaintiff’s injury. A. The Legal Standard For Judgment as a Matter of Law “Judgment as a matter of law may not properly be granted under Rule 50 unless the evidence, viewed in the light most favorable to the opposing party, is insufficient to permit a reasonable juror to find in her favor.” Williams v. County of Westchester, 171 F.3d 98, 101 (2d Cir.1999) (quoting Galdieri-Ambrosini, 136 F.3d at 289). In considering a Rule 50 motion, the trial court “must give deference to all credibility determinations and reasonable inferences of the jury, and it may not itself weigh the credibility of witnesses or consider the weight of the evidence.” Galdieri-Ambrosini, 136 F.3d at 289; Kirsch v. Fleet Street Ltd., 148 F.3d 149, 161 (2d Cir.1998). According to another formulation of the Second Circuit test, “judgment as a matter of law should not be granted unless (1) there is such a complete absence of evidence supporting the verdict that the jury’s findings could only have been the result of sheer surmise and conjecture, or (2)there is such an overwhelming amount of evidence in favor of the movant that reasonable and fair minded persons could not arrive at a verdict against it.” Id. (internal quotation marks and citations omitted). See also, Doctor’s Associates, Inc. v. Weible, 92 F.3d 108, 111-12 (2d Cir.1996) (judgment as a matter of law is inappropriate unless there is “either an utter lack of evidence supporting the verdict, so that the jury’s findings could only have resulted from pure guesswork, or the evidence [is] so overwhelming that reasonable and fair-minded persons could only have reached an opposite result”) (internal quotation marks and citations omitted). “A mere scintilla of evidence is insufficient to present a question for the jury.” Fane v. Zimmer, 927 F.2d 124, 128 (2d Cir.1991) (quoting, Epoch Producing Corp. v. Killiam Shows, Inc., 522 F.2d 737, 743 (2d Cir.1975)). B. The Law of Design Defect To prevail, plaintiff must prove that the 1991 Ford Aerostar was defectively designed. If plaintiff cannot establish that the vehicle was defectively designed for purposes of strict liability analysis, the case law on inconsistent verdicts (see supra, at section I) precludes a negligence finding based upon the same design defect. Therefore, our discussion focuses on the sufficiency of the strict liability claim. Only if the evidence supporting strict liability is sufficient would a new trial be ordered. The governing law for design defect product liability in New York was summarized by the state’s Court of Appeals in Voss v. Black & Decker Manufacturing Co., 59 N.Y.2d 102, 107, 463 N.Y.S.2d 398, 402, 450 N.E.2d 204, 208 (1983): In order to establish a prima facie case in strict products liability for design defects, the plaintiff must show that the manufacturer breached its duty to market safe products when it marketed a product designed so that it was not reasonably safe and that the defective design was a substantial factor in causing plaintiffs injury. See also McCarthy v. Olin Corp., 119 F.3d 148, 155 (2d Cir.1997) (same). Generally, a plaintiff must provide enough evidence for a jury to conclude that a series of propositions are more likely true than not: (1) the product is “defective” because it is not reasonably safe as marketed; (2) the product was used for a normal purpose; (3) the defect was a substantial factor in causing the plaintiffs injuries; (4) the plaintiff by the exercise of reasonable care would not have both discovered the defect and apprehended its danger; (5) the plaintiff would not have otherwise avoided the injury by the exercise of ordinary care. Urena v. Biro Manufacturing Co., 114 F.3d 359, 363 (2d Cir.1997) (citing Fane, 927 F.2d at 128; Wolfgruber v. Upjohn Co., 72 A.D.2d 59, 62, 423 N.Y.S.2d 95, 97 (4th Dep’t 1979), aff'd, 52 N.Y.2d 768, 436 N.Y.S.2d 614, 417 N.E.2d 1002 (1980)). See also Jones v. Lederle Laboratories, 785 F.Supp. 1123, 1127 (E.D.N.Y.1992) (“Once [the] conclusion [that one of these propositions could not have reasonably been established based on the evidence] is reached, there is nothing left to the case”). New York’s Court of Appeals has explained, “[s]ince no product may be completely accident proof, the ultimate question in determining whether an article is defectively designed involves a balancing of the likelihood of harm against the burden of taking precaution against that harm.” Robinson v. Reed-Prentice Div. of Package Mach. Co., 49 N.Y.2d 471, 479, 426 N.Y.S.2d 717, 720, 403 N.E.2d 440, 443 (1980) (citing Micallef v. Miehle Co., Div. of Miehle-Goss Dexter, Inc., 39 N.Y.2d 376, 386, 384 N.Y.S.2d 115, 121, 348 N.E.2d 571, 577 (1976); 2 Harper and James, Torts, § 28.4). See also Urena, 114 F.3d at 363 (same). Courts have alternatively referred to the relevant balancing as that of “risk/utility analysis.” McCarthy, 119 F.3d at 155; Liriano v. Hobart Corp., 92 N.Y.2d 232, 239, 677 N.Y.S.2d 764, 767-68, 700 N.E.2d 303, 306-07 (1998). Briefly, under the latter moniker, the test requires that a manufacturer be held strictly liable for harm caused by its product only if the risk of harm outweighs its utility. Id.; Robinson, 49 N.Y.2d at 479, 426 N.Y.S.2d at 720, 403 N.E.2d at 443 (citing Restatement (Second) of Torts § 402A). Factors to be considered in determining whether a product is reasonably safe include: (1) the likelihood that the product will cause injury; (2) the product’s utility to the public as a whole and to the individual user; (3) the technological and economic feasibility of a safer design; (4) the plaintiffs awareness of the danger and the ability to have avoided injury with careful use of the product; and (5) the manufacturer’s ability to spread the cost of any safety-related design changes. Gonzalez, 931 F.Supp. at 164 (citing Denny, 87 N.Y.2d at 257, 639 N.Y.S.2d at 255, 662 N.E.2d at 735; Voss, 59 N.Y.2d at 109, 463 N.Y.S.2d at 402, 450 N.E.2d at 208; Fallon v. Clifford B. Hannay & Son, Inc., 153 A.D.2d 95, 550 N.Y.S.2d 135 (3d Dep’t 1989)). Of these factors, the one that is clearly the most crucial in this case is the first. New York’s Court of Appeals has said that any design defect plaintiff is “under an obligation” to present evidence that the product presents a “substantial likelihood of harm.” Voss, 59 N.Y.2d at 108, 463 N.Y.S.2d at 402, 450 N.E.2d at 208. If a plaintiff cannot show that a product poses a substantial likelihood of harm, then she cannot make out her prima facie case. Gonzalez, 931 F.Supp. at 165 (citing id.; Fallon, 153 A.D.2d at 99, 550 N.Y.S.2d at 137). Thus, plaintiffs first burden is to demonstrate that there truly was a defect in the design. Second, a design defect plaintiff is also under an affirmative burden to prove that the defect was a substantial factor in causing the plaintiffs injuries. Urena, 114 F.3d at 363. Where this burden is not met, the plaintiffs claim equally fails. Jones, 785 F.Supp. at 1127. Thus, even if a reasonable jury could find the product to be defectively designed, the plaintiff must establish by sufficient evidence in the record that the asserted defect was the proximate cause of the plaintiffs injuries. Fane, 927 F.2d at 130-131; Doll v. Digital Equipment Corp., No. 93 Civ. 0359, 1996 WL 107111, at *1 (W.D.N.Y. Mar. 6, 1996) (citing Cover v. Cohen, 61 N.Y.2d 261, 473 N.Y.S.2d 378, 461 N.E.2d 864 (1984); Voss, 59 N.Y.2d 102, 463 N.Y.S.2d 398, 450 N.E.2d 204); Crump v. Times Square Stores, 157 A.D.2d 768, 769, 560 N.Y.S.2d 373, 374-75 (2d Dep’t 1990) (entering directed verdict where “plaintiffs failed to present sufficient evidence to indicate that the accident was caused by a design defect”). See also McCarthy, 119 F.3d at 155 (design defect cannot be shown simply on the basis of a product’s inherent risks). “If there is no causal relationship between a defendant’s product and a plaintiff’s injury, the defendant cannot be held liable on any product liability theory advanced by the plaintiffs complaint.” Lawrence v. Sofamor, S.N.C., No. 95 Civ. 1507, 1999 WL 592689, at *5-6 (N.D.N.Y. Aug.2, 1999) (causation is “indispensable element for each element” of plaintiffs design defect claims). C. The Necessity of Expert Evidence In a technical, scientific case such as this, expert evidence to support plaintiffs claim is essential. See, e.g., Barnes v. Anderson, 190 F.3d 47, 54-57 (2d Cir.1999) (to survive judgment as a matter of law, plaintiffs must produce expert evidence to prove proximate causation of medical injury); In Re Joint Eastern & Southern District Asbestos Litigation, 52 F.3d 1124, 1133 (2d Cir.1995) (noting need for expert evidence which can overcome the “unique difficulties” in complex tort cases); In re “Agent Orange” Product Liability Litigation, 611 F.Supp. 1223, 1262-63 (E.D.N.Y.1985) (expert testimony as to probabilities “is particularly important when establishing causation in a product liability action”); Gold v. Daikon Shield Claimants Trust, No. B-82-383, 1998 WL 351456, at *3 (D.Conn. June 15, 1998) (without expert testimony “as to causation to link the defect to the injury, a reasonable jury could not find that the plaintiff has proved that the defect caused her specific injuries”). The Second Circuit’s decision in Fane, 927 F.2d at 130-132, which upheld a trial court’s directed verdict in a design defect case where the plaintiff failed to provide sufficient evidence of causation, is illustrative. Reviewing the trial record for claims of both negligence and strict liability, the court reasoned that the issue of causation was so scientifically complicated that it was “beyond the sphere of the ordinary juryman and required expert testimony.” Id. Distinguishing the facts of Voss, the court pointed out that unlike a relatively straightforward case, such as an injury caused by a circular power saw, an expert is required where “no one knows for sure what happened.” Id. at 131. In Fane, there were alternate theories as to precisely what started the chain of events that led to the plaintiffs injury. Id. Since the plaintiffs experts could not tie their theory to what actually happened to the plaintiff by any evidence, the Second Circuit found the plaintiffs testimony to be insufficient to establish causation. Id., at 131-32 (“Nonexpert witnesses can only theorize as to how the leg was fractured and the device broken”). D. The Plaintiff’s Evidence It is with these legal principles, concerning judgment as a matter of law, design defect, and the need for expert testimony, that we will review the evidence presented by the plaintiff. Specifically, we will review the evidence from the dual perspectives of: (a) whether plaintiff has established that the cruise control system of the 1991 Ford Aerostar was defectively designed and (b) whether, assuming that defect has been established as a matter of theory, plaintiff has sufficiently proven that the defect actually caused the plaintiffs injury. While these two issues are analytically distinct, it is difficult to parse the testimony here into the two subject areas without being either confusing or repetitive. Accordingly, our discussion is initially melded, but will be followed by a discussion of the basis for our conclusions on each issue. 1. Plaintiff’s Description of the Accident According to the plaintiff, around 4:45 or 4:50 p.m. on July 14,1991, she got into her Aerostar in order to move it from the middle of the driveway, where her father had left it. Tr. 69-70. She testified that, at the time, she was “ticked [ ] off’ due to the sight of her father operating a chainsaw nearby, close to her eight-year-old son. Tr. 71-72. She said that she got into the Aerostar, closed the door, put on her seatbelt, put her right foot “lightly on the brake” and turned the ignition key. Tr. 73. She testified that she never moved the gear out of “park.” Tr. 258-59. According to her testimony, “all of a sudden, the car went vroom and took off,” kicking up gravel from the driveway behind it. Tr. 73-74. Jarvis testified that, looking down to verify that her foot was on the brake, “pounding” the brake with “two feet pumping” and “holding onto the steering wheel with all of [her] might,” she could not stop the vehicle. Tr. 74, 83-84, 255-59. At the same time, she testified that she maneuvered the vehicle to avoid three pedestrians ahead of her on the curved gravel driveway and a tree to one side of it. Tr. 74-75, 260. Along the way, she said she glanced down to the dash board, where she saw the speedometer “going up” and a brightly colored emergency light that read “brake.” Tr. 80-81, 279-80. Jarvis testified that she remembered the car driving off the road and brushing up against saplings before she blacked out. Tr. 75. Plaintiff called an accident reconstruc-tionist, George David Pope (“Pope”), to explain the events in a more technical fashion. According to Pope, the car traveled approximately 330 feet before it came to a rest in a drainage ditch and tipped over. Tr. 695. Assuming a wide open throttle condition, he hypothesized that, “she was able to slow the vehicle to, I believe, 15 to 20 miles an hour as she went into the ditch and overturned.” Tr. 702. Initially, Pope had examined the physical evidence for the possibility of brake failure (“Her initial comments were, my brakes weren’t working and my brakes failed... that was the point of my initial investigation to evaluate was there a brake failure or not?”) Tr. 700. Pope concluded, though, that there was nothing wrong with the braking system of the Jarvis vehicle. Tr. at 710. 2. Plaintiff’s Sudden Acceleration Theory Plaintiff called Samuel J. Sero (“Sero”) in its effort to explain to the jury the theoretical basis for how a vehicle could suddenly accelerate without driver input. While Sero posed two theories, both involving simultaneous electrical faults, as to how an unintended wide open throttle condition could occur, Sero testified that his second scenario, involving the simultaneous grounding of the automobile’s vent and vacuum wires, did not happen here. Tr. 955. Thus, we will focus only on the first theory. Under Sero’s first (hereafter described without that adjective) theory, there must be both: (a) an open ground wire from the speed amplifier and (b) a grounding of either the vent or the vacuum (“vac”) wires. Tr. 944. An open ground wire can result from looseness (usually due to corrosion or breakage), either at the point where the ground wire attaches to the amplifier or where the ground wire attaches to the metal of the car. A grounding of the vent or the vac wire can, according to Sero, occur if there is: (a) moisture or contaminate on the circuit board or at the connector to the speed amplifier (Tr. 873-74); (b) overheating of the circuit board causing it to “expand” or “bow up” (Tr. 875); or (c) a nick in the wire insulation that allows either wire to make contact with metal. For the postulated wide open throttle condition to result in an accident (or, put another way, to constitute a defect), there must be no built-in mechanism to close the throttle, even assuming that the simultaneous, random electrical events could occur in “real life.” According to Sero, “[tjhere is no mechanism existent in this device to kill the power, take it to a failsafe condition, not one.” Tr. 877. Sero rejects the dump valve — which vents the servo, releases vacuum pressure, and causes the throttle to close — as a failsafe mechanism because: (a) he asserts that unless you keep the brakes depressed, the electrical 'fault will not clear; (b) the dump valve itself may be misadjusted; and (c) the dump valve will not perform its function unless the driver knows to brake. In addition, Sero maintains that the braking system of the ear will not function well if there is a wide open throttle condition because the vacuum assist to the brakes will be lost. In sum, for Sero’s theory to actually cause the accident that occurred here, four things or events must occur, two of them simultaneously and the others within seconds thereafter: (1) there must be an open ground wire from the speed amplifier; (2) there must be a grounding of the vent or the vac wire which involves both the exposure of the wire and its contact with metal; (3) there must be a failure of the dump valve; and (4) there must be a loss of braking power. 3. An Evaluation of Plaintiff’s Evidence We will examine first the evidence relating to the simultaneous, random electrical events that Sero posits. At the outset, we note that it is basic (and convenient) to Sero’s theory that there will be no physical evidence that the electrical events occurred. Sero testified, “There will be no arc, no physical evidence left. There will be nothing left to prove it ever happened. It will be a random event” (Tr. 860) that “heals itself.” Tr. 975. Sero does not dispute, though, that shorts to ground can occur from breaks in the insulation, disconnections, etc., and that there would physical evidence of such an event. Focusing first on the ground wire from the speed amplifier, Sero agreed that there was no evidence that the ground wire was not making full and firm electrical contact during the relevant time period, or that there was any physical damage to the ground wire or the insulation surrounding it. Tr. 926. Victor DeClercq (“DeClercq”), a retired Ford engineer called by both sides to explain the components of the cruise control system, also testified that there was no physical evidence that the ground wires had shorted or become loose. Tr. 1366-67. Under Sero’s theory, the loose connection could be found where the ground wire attaches to the amplifier or where the ground wire attaches to the car’s metal body. However, in contrast to DeClercq, who tested the wires electrically and found them sound, Sero did not perform electrical tests on the wire (Tr. 916) or x-ray the amplifier to see if the insulation was intact. Tr. 926. Further, Sero postulated that the connection between the ground wire and the car could be flawed because of corrosion. However, Sero never inspected the vehicle and could only surmise that the Aerostar, owned by plaintiff for one week and with just over four hundred miles on it could have been corroded because Ford forgot to galvanize the car. Tr. 937. In short, there was no tangible or even inferential evidence to support the notion that the ground wire was loose or disconnected. The second, and simultaneous, electrical event that must occur is that there must be a grounding of either the vent or the vac wires. The vent and vac wires run between the servo and the cruise control module. They control the opening and closing of the throttle, which affects the speed of the car. For a grounding of either of these wires to occur, there must be: (a) moisture or debris on the printed circuit of the cruise control module, causing a fault; (b) moisture or debris in the connectors of the wires to the speed amplifier, causing a fault; or (c) a break in the insulation of the wire and the exposure of the wire to metal. Again, there was no evidence in the trial record from which a reasonable jury could conclude that any of these situations occurred and led to a grounding of the vent or the vac wires. As for the moisture possibility, Sero did not explore what the weather actually was for the week that plaintiff owned the car, but did acknowledge that plaintiff had stated that the day of the accident was dry and sunny. Tr. 939. Nor did Sero test the speed amplifier to ascertain whether a direct infusion of water on the speed amplifier would create the condition postulated. In contrast, Ford sprayed the inside of the engine to test Sero’s hypothesis, with negative results. Tr. 1169-71, 1437. Moreover, the notion of debris in the circuit board is unsupported by any evidence at trial and is especially speculative in a car as new as the Jarvis vehicle. As for the possibility of a break in the insulation of the wires, Sero also testified that there was no evidence of physical problems with the vent or the vac wires (Tr. 943), and he acknowledged that if there was no problem with the insulation of the wires, they cannot go to ground. Tr. 915. Similarly, DeClercq testified that he examined the vent and vac wires and that there was no evidence that they were abraded or had gone to ground. Tr. 1367. DeClercq also tested the wires by applying one thousand volts to each and established that the insulation was good. Tr. 1388. At trial, Sero proposed yet a fourth scenario for how either the vent or the vac wires could have gone to ground: the circuit board could have warped from the heat. While Sero posed this scenario, he agreed that the circuit board showed no evidence whatsoever of warping, suggesting instead that after warping, the circuit board would “contract” or “go back” into precisely the same shape. Tr. 876. Moreover, Sero had no knowledge of what the temperature was on the date of the accident, no idea if the car’s windows were opened or closed, or whether the air conditioner had been on. In fact, he conceded that he had no evidence “to support the fact that it was warm enough inside the car to cause a warp in the [circuit] board.” Tr. 959. Plaintiffs counsel’s promise to connect up this meltdown theory to the facts of the case (Tr. 876) remained unfulfilled. Plaintiff never proffered evidence that any such moisture, contaminate, nicks, or “bowing” was present in plaintiffs Aeros-tar on the day of her accident. Tr. at 911-27. There was no direct or circumstantial evidence upon which a jury could reasonably conclude that either the vent or the vac wires in the Jarvis vehicle went to ground on July 14,1991. As noted earlier, it is an essential part of Sero’s theory that the vehicle did not include mechanical devices that could have overcome the electrical events that allegedly caused the throttle to open significantly and result in an acceleration that was not intended. Ford maintains that the existence of the Aerostar’s dump valve, a spring loaded plunger which is opened when the brake pedal is depressed approximately one half of an inch, is a sufficient mechanical redundancy. If the dump valve is open (i.e., the brake pedal is depressed one half of an inch), it will mechanically release the vacuum and close the throttle, terminating any wide open throttle condition and returning the throttle to neutral. This undisputed mechanical sequence is of importance in several contexts. First, it undermines plaintiffs own version of the accident, which was that she started the car with her foot on the brakes, and that with her foot on the brake, the car accelerated. Indeed, plaintiffs expert witness, Pope, testified ■ that with a functioning dump valve and plaintiffs foot on the brake, as she testified it was when she started the car, the vehicle could not have gone to wide open throttle or even moved from a standing start. Tr. 754-55. The only evidence in the record was that the dump valve was in working order. Tr. 1396-97. Thus, again, assuming the two simultaneous electrical events, for the vehicle to have gone to a wide open throttle, plaintiff would have needed to release the brake, in turn releasing the dump valve, for the car to move. According to Sero, with the release of the dump valve, the effects of the electrical fault will be reinstated, and with a closed dump valve, the vacuum will stay in the system permitting an open throttle. Even assuming that “reinstated” electrical faults and the closed dump valve combine to create a wide open throttle condition, plaintiff still needs to explain why re-application of the brakes would not open the dump valve, return the throttle to idle, and with additional braking, bring the car to a stop. Sero’s explanation is that there will be no vacuum remaining in the vacuum booster to assist the driver and thus the driver will not be able to use the other mechanical device, namely the brakes, to bring the vehicle to a halt. Sero further postulated that the vacuum assist failed because the plaintiff, unaccustomed to anti-lock brakes, pumped the brakes instead of applying steady pressure in accordance with the driver’s manual. Tr. 882. Sero offered his braking theory despite the fact that he had no experience designing brakes, never tested the brake system in the Jarvis Ae-rostar, did not know how many pumps of the brake can be made before the vacuum reservoir is exhausted (Tr. 963-64), and never tested the dump valve. Tr. 972. In fact, plaintiffs other expert witness, Pope, who did have extensive experience in designing brakes, gave testimony totally contrary to Sero’s theory: Q [to Pope]: Well, if the dump valve was fully operable, and if the dump valve is engaged at approximately half an inch of pedal travel and before the brakes are engaged, and you are at normal pumping phase, the dump valve is always going to be open, so you are never going to be out of vacuum assist, isn’t that so? A: As you phrase it, that’s correct. Q: Did you consider that when you were doing your analysis? A: I considered that the dump valve was operating properly, the braking action would have brought it to a stop well before the ditch. Q: Alternatively, if she believed she was applying the brake, and in fact was applying the accelerator, by putting the accelerator all the way to the floor in a pumping fashion with the parking brake engaged, the vehicle would continue on down its path, wouldn’t it? A: Yes. Tr. 745-46. Thus, Pope rejects Sero’s theory that you would lose vacuum assist under one version of events testified to by the plaintiff. Pope’s testimony also totally undermined Sero’s hypothesis that this accident occurred because plaintiff pumped the brakes rather than applying steady pressure: Q [to Pope]: if plaintiff indicated then [sic] as the vehicle moved unexpectedly, she was pumping the brake under ordinary pumping, the dump valve would have remained open, and the vehicle would have stopped. Do you agree with that? A: Yes. Q: So she would have to have been mistaken on that application of brakes, right? A: Or the dump valve malfunctioning. Q: Absent a malfunctioning dump valve? A: True. Tr. 755. Again, it will be recalled that the only testimony in the record vias that the dump valve was functioning properly. In fact, plaintiff never made a claim that the dump valve was not functioning properly— a claim far different from one for a design defect. Thus, based on the evidence offered by plaintiffs own expert witness, there is no support for Sero’s assumption that the mechanical redundancies of the dump valve and brakes were insufficient to overcome a wide open throttle condition. 4. Judgment as a Matter of Law A review of the evidence demonstrates that it is completely insufficient to support liability under either a strict liability or a negligence theory and that reasonable and fair minded persons could only have reached a verdict for Ford. Plaintiff did not establish that the postulated simultaneous electrical faults were substantially likely to occur or, even assuming they did, that the mechanical features of the car would not have overcome the resulting wide open throttle condition. Further, even if a jury could have reasonably found for the plaintiff on the existence of a design defect, plaintiff has still failed to establish that the alleged defect in fact caused Jarvis’ accident. Finally, defendant has presented an explanation of the events of July 14, 1991 that is so overwhelmingly in accord with the evidence adduced at trial that no reasonable jury could have found otherwise. a. Substantial Likelihood of Harm Focusing first on the prima facie element of substantial likelihood of harm, it is important to note that plaintiff has offered no evidence to suggest how frequently the alleged design defect is likely to occur. Sero could not even state the mathematical possibility of the two electrical faults occurring at the same time. See Jarvis, 1999 WL 461813, at *6 n. 16. Plaintiff has absolutely no evidence of her own or support from external scientific literature to allow a rational jury to even begin to engage in a balancing of this risk with the cruise control design’s utility. Liriano v. Hobart Corp., 132 F.3d 124, 131 n. 12 (2d Cir.1998); McCarthy, 119 F.3d at 155. See also United States v. Carroll Towing Co., 159 F.2d 169, 173 (2d Cir.1947) (Judge Learned Hand’s famous expression of the need for mathematical balancing in negligence-cases). In Gonzalez v. Morflo Industries, 931 F.Supp. at 165-66, the court noted that even though the plaintiffs experts offered a theory as to how the accident happened, they provided no evidence to suggest how often such an accident occurs and thus “failed to demonstrate even a meaningful, let alone ‘substantial’ likelihood” that the product would cause the type of harm experienced. See also Aghabi v. Sebro, 256 A.D.2d 287, 288, 681 N.Y.S.2d 333, 334 (2d Dep’t 1998) (expert’s “bare conclusory assertions ... which consisted primarily of speculative allegations with no independent factual basis, were insufficient to raise a triable issue of fact and defeat [defendant’s] motion for summary judgment”). Even if a jury could find that the 1991 Aerostar’s cruise control bore a substantial likelihood of unintended activation, plaintiff still has not provided proof that the fault could not be overcome by the minivan’s mechanical redundancies, namely its dump valve and brake system. First, Pope, plaintiffs knowledgeable brake witness, testified that if plaintiff started the car with her foot on the brake, the Aeros-tar could not have gone to a wide open throttle or even moved because the dump valve would have prevented such a scenario. Second, Pope testified that if plaintiff was pumping the brakes as she testified, there would have been no loss of vacuum assist and the brakes would have stopped the car well before she reached the ditch. Thus, both the dump valve and the brakes as designed would have prevented this accident even assuming the unproven simultaneous undetectable electrical faults. Sero’s inexpert hypothesizing to the contrary is simply not evidence. On this record, judgment must be entered for the defendant. It is not tenable that a jury could rationally find a substantial likelihood of harm from the Aerostar’s cruise control design.since plaintiff has not offered any evidence -that the attenuated chain of events necessary to result in an accident is likely to occur, let alone with significant regularity. We note that, in her 50(b) motion papers, plaintiff never challenged the jury’s finding that the product was not defective as being unsupported by the evidence. In fact, the. plaintiff admitted, “there is nothing illogical in the jury finding that the system was not designed in a defective manner.” PI. 50(b) Mem. at 11. This also fatally undermines plaintiffs claim. See, e.g., Tipton, 101 F.3d at 1151. Other courts have entered judgment under similar circumstances. The Fifth, Sixth, and Ninth Circuits have all' either upheld' or entered directed verdicts for defendants in cases of sudden acceleration injuries allegedly caused by faulty cruise control systems. In Dupper v. General Motors Corp., 887 F.2d 1089, 1989 WL 123650, at *3, the Ninth Circuit reviewed petitioner’s claim that an additional “on/ off’ switch for the cruise control would have made the mechanism safer and concluded that “[a]s a matter of law, the absence of such a switch does not make the system unreasonably dangerous.” In another unpublished opinion, the Sixth Circuit focused on a different aspect of the design defect claim, pointing out that the sequence of events that allegedly made the cruise control system defective was so unlikely as to allow for the district court’s finding as a matter of law that the design was not defective. Cole v. General Motors Corp., 852 F.2d 568, 1988 WL 76522, at **3-5 (6th Cir.1988) (unpublished opinion). See also Lawrence v. General Motors Corp., 73 F.3d 587, 590 (5th Cir.1996) (reversing jury verdict where plaintiff did not elaborate as to likelihood that alternate design could have avoided accident); Ashley v. General Motors Corp., 666 So.2d 1320 (La.Ct.App.1996) (reversing trial court’s design defect verdict for plaintiff where plaintiffs failed to offer proof that the accelerator system deviated in any fashion from the identical system installed on .similar cars and examination of car revealed no evidence of defect). But cf. Cole v. Ford Motor Co., 136 Or.App. 45, 900 P.2d 1059, 1063 (1995) (upholding a jury verdict for plaintiff in a sudden acceleration case after finding sufficient evidence in that record to sustain the outcome). b. Causation We now turn to the analysis of the evidence from a causation or “substantial factor” perspective. For this discussion, we will assume that which we have found was not proven, namely that two, undetectable simultaneous electrical faults could have occurred in the Jarvis vehicle and overcome its dump valve and braking system on July 14, 1991. Even making this dubious assumption, plaintiffs evidence does not establish that it is more likely than not that the design of the Aerostar’s cruise control system was a substantial factor in plaintiffs accident. While Sero’s testimony is replete with hypotheses and proffers of explanations of how plaintiffs accident could have happened, the record is empty of evidence that would support the proposition that the theoretical defect in fact caused the accident described by the plaintiff. It is undisputed that there is no physical evidence that any of the wires became loose or shorted to ground. Nor did Sero or any other witness called by plaintiff utilize available tests to ascertain the soundness of the wire’s connections or insulation in the Jarvis Aerostar. Thus, Ford’s testing evidence demonstrating that all wires and the circuit board were sound and showed no evidence of insulation breakage or grounding was uncontradicted. Likewise, Sero’s hypotheses of moisture, contaminate and “bowing” were not only not proven, but were rendered merely speculative at best by the absence of evidence as to the actual weather conditions on the day of the incident, the newness of the vehicle, and Ford’s tests on the Aerostar. In these circumstances, a verdict for the defendant is required. Matthews v. Hyster Co., 854 F.2d 1166, 1169 (9th Cir.1988) (upholding a directed verdict where the plaintiffs expert “theorized that contaminants may have gotten into the brake system and caused the brakes to malfunction” but did not offer any proof to support this theory). See also Browne v. McDonnell Douglas Corp., 698 F.2d 370, 371 (9th Cir.1982) (“where proximate causation is at issue [in a directed verdict case], the nonmoving party will not have met its burden if the evidence merely suggests the possibility that proximate cause exists”); Mazzola v. Chrysler France, S.A., 470 F.Supp. 24, 26 (E.D.N.Y.1978) (jury notwithstanding the verdict in automobile design defect case where expert had testified as to theory of how accident could have happened but “had never examined the [plaintiffs] car” to verify theory). The plaintiffs theory as to the dump valve and braking system is similarly troubled. For a jury to find that the Aerostar was defectively designed, plaintiff would need proof that the dump valve and braking system actually were unable to overcome the hypothesized electrical events. However, plaintiffs proof was to the contrary. Pope’s testimony was that a working dump valve and brake system would have stopped the car and prevented the incident. And the only evidence in the record is that both the brakes and the dump valve were in good working order. In fact, Pope testified that if plaintiff had been pressing hard on the brakes, as she claimed, the car would have stopped in 30 to 35 feet, approximately one-tenth of the 330 foot distance she actually traveled. Sero’s theory that the brakes were inadequate because Jarvis pumped them instead of applying steady pressure is belied by his own lack of expertise in braking systems and human factors, the fact that he never tested the brakes or dump valve, and the directly contradictory testimony of Pope. In cases similar to the one at bar, judgment has been entered for the defendant. For example, in Doll v. Digital Equipment, the court entered judgment for the defendant based on the failure of the plaintiffs experts to “link the plaintiffs injuries to the design of defendant’s [product]” or “to any faulty feature thereof.” 1996 WL 107111, at *2. Specifically, the court criticized the plaintiff for offering only experts that were not able or not qualified to show that the asserted defect in fact caused the injury. Id. “This Court draws a line between reasonable inferences on the one hand and scintillas of evidence and conjecture and surmise on the other.” Id. Such is the case here. The Sixth and Ninth Circuits reached similar conclusions on the issue of causation in cruise control cases. In Dupper, the Ninth Circuit found it dispositive that plaintiff had offered no evidence that the cruise control system was defective at the time it was purchased, that it was defective when the accident occurred, or that its wear characteristics made it likely to cause the accident described. 887 F.2d 1089, 1989 WL 123650, at *5. Second, the Ninth Circuit pointed out that the plaintiffs own testimony had not supported her expert’s theory. Id. The Sixth Circuit case, Cole v. General Motors Corp., is even closer to being on point. There, the court held that “the sequence of events alleged by the plaintiffs was so unlikely that no reasonable juror could have possibly found a causal connection between the alleged defect, i.e., the absence of an on/off switch on the cruise control mechanism, and the injuries suffered by [the plaintiff] as a result of the accident.” 852 F.2d 568, 1988 WL 76522, at *5. c. Overwhelming Evidence Alternatively, apart from the complete absence of evidence to support the jury verdict, there is such an overwhelming amount of evidence in favor of the defendant that' reasonable and fair minded persons could not arrive at a verdict against it. Galdieri-Ambrosini, 136 F.3d at 289; Kirsch, 148 F.3d at 161. Ford, in addition to demonstrating essentially from plaintiffs own witnesses that the July 14, 1991 incident was not the result of plaintiffs sudden acceleration theory, also proffered an alternative scenario that was consistent with the evidence offered by third parties and received support from plaintiffs own witness Pope and the NHTSA governmental study. Ford’s scenario was essentially as follows: Jarvis was unfamiliar with her brand new car, a minivan which was different from her previous low-riding car. Tr. 239-40. Under these circumstances, she started the car unaware that her father had set the parking brake (Tr. 280, 308), put her foot on the accelerator thinking it to be on the brake, and was startled when the engine started to race against the force of the parking brake. Continuing to believe that her foot was on the brake and not on the accelerator, plaintiff was unable to stop the car. This theory is supported by the following evidence: (1) plaintiffs father’s testimony that he set the parking brake (Tr. 308); (2) plaintiffs own testimony that she saw the word “brake” lighted on the dashboard, which according to Pope means that the parking brake is set (Tr. 755-56); (3) plaintiffs testimony that the speedometer was “going up” (Tr. 81) and read 30 to 35 m.p.h. (Tr. 278) or 40 to 45 m.p.h. (Tr. 279), which DeClercq testified was impossible if the ground wire was loose or disconnected since the ground wire for the amplifier and the speedometer are the same and the speedometer would have read zero (Tr. 1367-68); (4) the testimony of eye witness Joanne Valentine-Simonian who witnessed the event while walking on the driveway and testified that she saw no rear brake lights illuminated on plaintiffs vehicle when it passed her (Tr. 425, 752); (5) the absence of skid marks in the gravel driveway (which would have supported plaintiffs claim that her foot was forcefully on the brake), as testified to by the investigating police officer who walked by the length of the driveway right after the accident looking for evidence (Tr. 91-92 749-50.); (6) Pope’s testimony that Ford’s scenario of plaintiff pumping the accelerator with the parking brake on would have resulted in the engine racing sound that plaintiff and Valentine-Simonian reported (Tr. 730-31); (7) Pope’s testimony that the cessation of the revving sound that Valentine-Simonian described could have indicated that plaintiff had taken her foot off the accelerator (as if she were pumping the brakes), which would also explain the slowing of the vehicle (Tr. 756); (8)The testimony of Pope as well as other Ford witnesses that if plaintiff was pumping the brakes instead of the accelerator, the car would have stopped. Tr. 739-46,1150,1371-74. The reasonableness of Ford’s position was supported, both generally and in certain specifics, by a study conducted by the National Highway Traffic Safety Administration (“NHTSA”), a division of the United States Department of Transportation, which was introduced into evidence as defendant’s Exhibit F-4, and relied on by both sides during the trial. In 1989, NHTSA convened an independent panel “to identify and evaluate factors which could potentially cause or contribute to the occurrence of ‘Sudden Acceleration Incidents’ (SAI).” In addition to convening the panel of independent experts, NHTSA also collected literature, case documentation, interviewed drivers, studied the fuel-systems, braking systems, and driving controls of the vehicles, received complete schematics for cruise control and other vehicle systems, and performed tests and experiments at NHTSA’s Vehicle Research and Test Center. The panel selected ten vehicles about which there were numerous complaints (not including the Aerostar), noting that the specific vehicles chosen were representative of a much larger group. The study reached a number of conclusions of particular relevance to the issues here. In connection with the panel’s consideration of cruise control systems as a potential cause of SAI, the panel considered a Sero-like theory. With respect to the likelihood of simultaneous, intermittent electrical failures occurring, the report concluded: Extensive laboratory testing of the operation of cruise controls under stress from temperature extremes, power supply variations, EMI/FRI and high voltage discharges has demonstrated no failure modes of any relevance to SAI. Analysis of their circuitry shows that for nearly all controls designed in the past few years, two or more independent, intermittent failures would have to occur simultaneously to cause throttle opening in a way that would be difficult to detect after the incident. The occurrence of such simultaneous, undetectable failures is virtually impossible. Ex. F-4 at