Full opinion text
ORDER JAMES G. CARR, Senior District Judge. This is a products liability action brought by plaintiffs Linda and Daniel Buck (Buck), against defendant Ford Motor Company (Ford). In 2006, Linda Buck was injured when a 1999 Ford Expedition, driven by J.D. White, crashed through a wall of the bakery in which Linda Buck was working — allegedly due to an electronic malfunction that suddenly seized control of the throttle. Jurisdiction is proper under 28 U.S.C. § 1332. Pending is Buck’s motion to exclude Ford’s expert Vincent Declercq. [Doc. 56]. Also pending is Ford’s motion to exclude Buck’s experts: Samuel Sero [Doc. 58]; Keith Armstrong, [Doc. 60]; and William Berg [Doc. 59]. For the reasons that follow, I grant in part and deny in part the parties’ Daubert motions. Background On April 27, 2006, White pulled his 1999 Ford Expedition into the parking lot of a Nickles Bakery in Toledo, Ohio. As or soon after Mr. White pulled into the lot in front of the store, the vehicle suddenly accelerated over the curb, traveled into the store through the front window, crashed through a brick wall and struck bakery employee Linda Buck, pinning her against a back wall. Mr. White, who was sixty seven years old, was cited and convicted for failure to control. He has since passed away. In 2008, Buck sued Ford in the Lucas County, Ohio, Common Pleas Court, asserting that the Expedition had suddenly accelerated because of a design defect that rendered it susceptible to unintended throttle opening due to the impact of electromagnetic interference (EMI) on the electronic throttle control system. The Expedition had approximately 98,-000 miles on it at the time of the incident. It was equipped with a Next Generation Speed Control system (NGSC). Ford removed the action to the district court and answered the complaint, denying that there was any defect in the subject vehicle and claiming that the accident was due to driver error. The plaintiffs designated two electronics experts, Keith Armstrong and Samuel Sero, and one human factors and accident reconstruction expert, Dr. William Berg, in support of their claim of defect. Ford has designated its former employee, Victor Declercq, as its expert to rebut that claim. Discussion Federal Rule of Evidence 702 requires me to perform a “gate-keeping role” when considering the admissibility of expert testimony. Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579, 597, 113 S.Ct. 2786, 125 L.Ed.2d 469 (1993). Rule 702 provides: If scientific, technical, or other specialized knowledge will assist the trier of fact to understand the evidence or to determine a fact in issue, a witness qualified as an expert by knowledge, skill, experience, training, or education, may testify thereto in the form of an opinion or otherwise, if (1) the testimony is based upon sufficient facts or data, (2) the testimony is the product of reliable principles and methods, and (3) the witness has applied the principles and methods reliably to the facts of the case. Rule 702 applies not only to scientific testimony, but also to other types of expert testimony based on technical or other specialized knowledge. See Kumho Tire Co., Ltd. v. Carmichael, 526 U.S. 137, 147, 149, 119 S.Ct. 1167, 143 L.Ed.2d 238 (1999). My gate-keeping function here is threefold. First, I must determine whether the witness is qualified as an expert. “When making a preliminary finding regarding an expert’s qualifications under Fed.R.Evid. 104(a), the court is to examine ‘not the qualifications of a witness in the abstract, but whether those qualifications provide a foundation for a witness to answer a specific question.’ ” Smelser v. Norfolk Southern Ry. Co., 105 F.3d 299, 303 (6th Cir.1997) (quoting Berry v. City of Detroit, 25 F.3d 1342, 1351 (6th Cir.1994)). Second, I must determine whether the testimony is reliable. See Daubert, supra, 509 U.S. at 590, 113 S.Ct. 2786. The Court in Daubert listed several factors for consideration in assessing the reliability of scientific testimony, including: • WTiether a “theory or technique ... can be (and has been) tested”; • Whether it “has been subjected to peer review and publication”; • Whether, in respect to a particular technique, there is a high “known or potential rate of error” and whether there are “standards controlling the technique’s operation”; and • WThether the theory or technique enjoys “general acceptance” within a “relevant scientific community.” Kumho Tire, supra, 526 U.S. at 149-50, 119 S.Ct. 1167 (quoting Daubert, supra, 509 U.S. at 592-94, 113 S.Ct. 2786). The test of reliability is, however, “flexible, and Daubert’s list of specific factors neither necessarily nor exclusively applies to all experts or in every case.” Id. at 140, 119 S.Ct. 1167. “[Wjhether Daubert ’s specific factors are, or are not, reasonable measures of reliability in a particular case is a matter that the law grants the trial judge broad latitude to determine.” Id. at 153, 119 S.Ct. 1167. The focus must be on the principles and methodologies on which the expert’s opinion is based, and not on the merits of the expert’s conclusions. Daubert, supra, 509 U.S. at 594-595 n. 12, 113 S.Ct. 2786; United States v. Bonds, 12 F.3d 540, 556 (6th Cir.1993) (district courts “are not to be concerned with the reliability of the conclusions generated by valid methods, principles and reasoning.”). Finally, I must determine whether the expert’s reasoning or methodology properly applies to the facts at issue: i.e., whether the opinion is relevant. See Daubert, supra, 509 U.S. at 591-93, 113 S.Ct. 2786. To be relevant, the testimony must “assist the trier of fact to understand the evidence or to determine a fact in issue.” Fed.R.Evid. 702. This relevance requirement ensures that there is a “fit” between the testimony and the issue to be resolved at trial. United States v. Bonds, 12 F.3d 540, 555 (6th Cir.1993). Rejection of expert testimony “is the exception rather than the rule.” In re Scrap Metal Antitrust Litigation, 527 F.3d 517, 531 (6th Cir.2008) (quoting Fed. R.Evid. 702 Advisory Committee’s Note, 2000 Amend.). My role as gatekeeper “is not intended to serve as a replacement for the adversary system: ‘[vigorous cross-examination, presentation of contrary evidence, and careful instruction on the burden of proof are the traditional and appropriate means of attacking shaky but admissible evidence.’ ” U.S. v. 14.38 Acres of Land, 80 F.3d 1074, 1078 (5th Cir.1996) (quoting Daubert, supra, 509 U.S. at 597, 113 S.Ct. 2786). In assessing expert testimony, I “should also be mindful of other applicable rules.” Daubert, supra, 509 U.S. at 595, 113 S.Ct. 2786. Federal Rule of Evidence 703 provides that “[i]f the underlying data are so lacking in probative force and reliability that no reasonable expert could base an opinion on them, an opinion which rests entirely upon them must be excluded.” In re Paoli RR. Yard PCB Litig., 35 F.3d 717, 748 (3d Cir.1994) (quoting In re “Agent Orange” Prod. Liab. Litig., 611 F.Supp. 1223, 1245 (E.D.N.Y.1985)). The proponent of the evidence has to establish that all of the pertinent admissibility requirements are met by a preponderance of the evidence. See Fed.R.Evid. 104(a); see also Bourjaily v. United States, 483 U.S. 171, 175-76, 107 S.Ct. 2775, 97 L.Ed.2d 144 (1987). Buck offers the expert testimony of Keith Armstrong, Samuel J. Sero, and William Berg. Ford urges the court to exclude all of these experts’ opinions, arguing that they are not qualified to offer their opinions, they rely on evidence that is insufficient as a matter of law to establish causation, and their opinions are unreliable. Ford offers the expert testimony of Vincent Declercq. Buck moves to exclude that testimony, arguing that Declercq is unqualified and that his testimony lacks a sufficient factual basis. 1. Keith Armstrong Armstrong intends to opine that: 1) the design of Ford’s NGSC system is defective in that EMI can open the throttle without a signal from the driver; 2) the system is not failsafe; 3) Ford ignored its own guidelines for electromagnetic compatibility (EMC) and other available alternative measures that would have improved the system’s EMC; and 4) Ford’s EMC testing protocols and testing results are insufficient bases to ensure the functional safety of the cruise control system. [Doc. 60-7, at 6]. Armstrong does not intend to testify as to the specific cause of the accident giving rise to this litigation. Ford moves to exclude Armstrong’s testimony on the grounds that: 1) his testimony is unreliable under Fed.R.Evid. 702(2); 2) his general causation testimony fails to assist the trier of fact; and 3) he had ex parte communications with a Ford employee and obtained Ford documents relating to the subject matter of this case from that employee during the pendency of this case, despite being told not to do so. [Doc. 60], On review of his testimony, I find that Armstrong’s general causation testimony — that EMI can cause a vehicle equipped with a Ford NGSC cruise control to suddenly accelerate and that the NGSC system is therefore defective — is unreliable. As each of Armstrong’s other proposed opinions necessarily incorporate this foundational opinion, his entire testimony shall be excluded A. Background Keith Armstrong is a chartered electrical engineer in the United Kingdom. Ford does not challenge Armstrong’s qualifications. B. Reliability under Rule 702(2) Applying the Daubert guidelines, Ford argues that Armstrong’s proposed testimony in this case is unreliable under Fed. R.Evid. 702(2) because: 1) his theories are untested; 2) his theories have not been peer reviewed through publication; 3) his methodology has not gained general acceptance; and 4) he cannot express his opinions within a reasonable degree of scientific certainty. Though no one element is dispositive, I find that Armstrong’s testimony is unreliable because his theory has not been tested and it has not been formally peer-reviewed. i. Failure Analysis Ford contends that Armstrong has not identified any actual methodology that he used to reach his conclusion that EMI could cause a vehicle with a NGSC system to suddenly accelerate. Buck states that Armstrong “employed failure analysis, an engineering design methodology that has been the standard in the field of EMC for many years.” [Doc. 80 at 4], This is insufficient, as there are multiple methods of failure analysis, and merely announcing that an expert applied failure analysis does not demonstrate that the methodology is reliable. For example, in Kumho Tire, supra, 526 U.S. at 155-56, 119 S.Ct. 1167, the Court found that an expert’s method of tire failure analysis that employed a visual/tactile inspection was unreliable vis-a-vis specific causation. Armstrong also testified that he relied on the scientific methodology of IS060000-1-2, a protocol written by a committee of acknowledged EMC experts and approved by the National Standards Committee on EMC. This protocol is entitled “Methodology for the achievement of functional safety of electrical and electronic systems including equipment with regard to electromagnetic phenomena” [Doc. 62-1 at 9]. Armstrong did not provide a complete copy of the document, nor does he explain the methodology espoused therein. Armstrong’s explanation of his methodology is: “[y]ou identify all possible hazards that could result from malfunctions, and then you go through your design and what likelihood will come of it, and you compare that likelihood with acceptable ... risk.” [Doc. 74 at 17-18]. It remains unstated, therefore, how Armstrong determined that EMI could activate a NGSC servo. Having read Armstrong’s report and testimony, Armstrong’s methodology seems in fact to be an application of his education and experience with EMI and electronics. When pressed on how he “can say beyond it’s possible but that [EMI-induced sudden acceleration] actually can happen here,” Armstrong responded, “[t]he engineering experience encompassed all over the world for 60 or more years.” [Doc. 74 at 93]. He seems to have inferred, from his understanding of general engineering principles, electromagnetic compatibility, and printed circuit board design, that a NGSC will behave in a predictable way. Armstrong states that EMI is inevitable in all electronics, and explained the effects EMI can have on a microprocessor. Accordingly, he testified that EMI can activate a speed control “[t]hrough interference getting into the microprocessor.” [Doc. 74 at 90]. The signal “could come in on the output lines, the output driver to the motor and coupled into the microprocessor and cause it to glitch, put the software into a loop. It’s hard to be more precise than that because there’s about a million ways which this could happen.” [Doc. 74 at 92]. He explained that he can reliably infer that the NGSC system will react to EMI in certain ways, because “the microprocessor is just a microprocessor. They use them all over the world. They all behave the same way.” [Doc. 74 at 93]. The design of the microprocessor has “built in protection” to minimize the effects of EMI. [Doc. 74 at 94]. In the case of the NGSC, those protections are the stepper motor and the aluminum enclosure. Armstrong testified to the weaknesses he perceives in those design elements that make them inefficient. He further opines that the NGSC system is “not as robust as it could be. It doesn’t follow any good EMC engineering design principles, including those of the Ford Motor Company. And it also doesn’t have a proper failsafe.” [Doc. 75 at 14-15], Based on his understanding of how EMI works in a general engineering sense, and having reviewed the design of the NGSC systems, Armstrong concluded that the design is susceptible to EMI-induced sudden acceleration. So far, so good. But as noted by Buck, “[t]he process is to first do a risk estimation, come up with a design specification, and then to verify it.” [Doc. 62 at 5]. As discussed below, I find Armstrong’s general causation opinion unreliable in part because he has failed to verify it, and he can not point to others who have. E.g. Smelser, supra, 105 F.3d at 304 (applying Daubert to exclude the testimony of a biomechanical engineer who failed to conduct pertinent testing). ii. Testing Ford contends that Armstrong’s general causation theory — that EMI can cause a vehicle to accelerate — is unreliable because it is untested. Ford argues that Armstrong must verify, through testing: 1) the creation of a transient EMI signal by a source within the vehicle; 2) the existence of pathway through which the signal can travel into the speed control electronics; 3) that the signal can activate the stepper motor; and 4) that, even if such a signal or combination thereof could engage the stepper motor, it could be both strong enough and last long enough to maintain a wide-open throttle through a sudden acceleration incident. [Doc. 60 at 13]. Armstrong acknowledges that he has never attempted to replicate or test a transient signal activating a servo in an automobile, and states that he is unaware of anyone who has ever been able to get EMI to actually activate a servo. Based on these statements, Ford argues that this untested hypothesis fails the Daubert test. Buck asserts that “testing is not an appropriate methodology to ensure safety.” [Doc. 62 at 6], This statement is not only illogical, it is unrelated to the issue at hand, ie. verification of a theory. Buck also argues that Armstrong’s opinion “is not the type of opinion that can or should be subjected to testing; design verification and failure modes and effects analyses are used instead, for the simple reason that there could be millions, if not trillions, of possible pathways to failure.” [Doc. 80 at 6]. Valid scientific methodology usually involves “generating hypotheses and testing them to see if they can be falsified.” Daubert, supra, 509 U.S. at 593, 113 S.Ct. 2786. Indeed, “Daubert and its progeny make clear that ‘proposed [expert] testimony must be supported by appropriate validation.’ ” Pride v. BIC Corp., 218 F.3d 566, 578 (6th Cir.2000) (citing Daubert, supra, 509 U.S. at 591, 113 S.Ct. 2786). Though the Daubert standard is flexible, that Armstrong failed to verify his theory through testing weighs against finding his opinion reliable. E.g. Pride, supra, 218 F.3d at 578 (in a product liability case, the “failure of [plaintiffs] experts to test their hypotheses in a timely and reliable manner” was one basis for holding their testimony inadmissible); see also, Smelser, supra, 105 F.3d at 304. Without testing, all Armstrong has done is identify a hypothesis. It may be a sound one, but the courts must necessarily lag behind science. Untested hypotheses, even if plausible, are insufficient to satisfy Rule 702. E.g., Rider v. Sandoz Pharm. Corp., 295 F.3d 1194, 1202 (11th Cir.2002) (“The courtroom is not the place for scientific guesswork, even of the inspired sort.”); see also Tamraz v. Lincoln Elec. Co., 620 F.3d 665, 670 (6th Cir.2010) (explaining that a “working hypothesis” is not “admissible scientific knowledge”). Armstrong provides no explanation for the analytical leap that the general engineering principles he describes apply to motor vehicles in general and the NGSC system in particular. “Nothing in either Daubert or the Federal Rules of Evidence requires a district court to admit opinion evidence which is connected to existing data only by the ipse dixit of the expert. A court may conclude that there is simply too great an analytical gap between the data and the opinion offered.” Gen. Elec. Co. v. Joiner, 522 U.S. 136, 146, 118 S.Ct. 512, 139 L.Ed.2d 508 (1997). I conclude, therefore, that Armstrong’s theory has not been sufficiently tested by himself or others to sustain a finding of sufficient reliability to allow the jury to hear his testimony. iii. Peer Review and Publication Ford argues that Armstrong’s theory is unreliable because Armstrong has not submitted that theory to a peer-reviewed publication, nor is he aware of any peer-reviewed article which has ever found that EMI has actually caused a speed control system in a vehicle to activate. Armstrong has written numerous articles and given many presentations on electromagnetic compatibility. In 2008, he presented a paper entitled “EMC for the Functional Safety of Automobiles; why EMC Testing is Insufficient and What is Necessary” at the IEEE EMC symposium in Detroit. [Doc. 74 at 52], Armstrong testified that it was peer-reviewed in the sense that “they reviewed it, and the could have told me to change it, or they could have not accepted it.” [Doc. 74 at 52]. Armstrong also testified that he was, in the near future, planning to give a paper called Sudden Acceleration in Automobiles and Control the Risks Due to EMI at the IEEE Product Safety Engineering Society annual symposium in Boston. Id. At the Daubert hearing, Ford asked Armstrong whether the 2008 paper had been “peer-reviewed by people who sat down, picked it apart, asked questions about it.” [Doc. 74 at 65]. Armstrong responded, “I understood that it had. It had to be accepted by the papers committee and reviewed ... these were safety engineers who are very well thought of safety engineers who reviewed my paper. To me that’s peer-reviewed.” Id. Ford dismisses Armstrong’s papers and publications as “symposiums where people make PowerPoint presentations,” and asserts that Daubert peer review means formal submission and publication through an established journal. Ford contends that “the law recognizes that simply presenting at a conference meets neither the letter nor the spirit of the Daubert peer review requirement because a presentation does not subject the theory to the active, critical rigors that a true peer review process involves.” [Doc. 84 at 21]. Ford’s is precisely the ossified analysis that Daubert rejected. The Supreme Court stated that peer review and publication are not identical and need not overlap: “Another pertinent consideration is whether the theory or technique has been subjected to peer review and publication. Publication (which is but one element of peer review) is not a sine qua non of admissibility; it does not necessarily correlate with reliability.” Daubert, 509 U.S. at 593-94, 113 S.Ct. 2786. Peer review through publication in an established journal is the most significant and meaningful form of peer review. Armstrong has not had his theory reviewed in that manner. This -is, under Daubert, “a relevant, though not dispositive, consideration in assessing” his testimony. Id. at 594, 113 S.Ct. 2786. I also find, as indeed I must based on Armstrong’s testimony, that his theory has been — at least to some modest degree— reviewed by his peers, which is some indication of reliability. It is but one indication that in his field, his theory is not immediately dismissed as junk science. iv. Probability Ford argues that because Armstrong cannot express his opinion as to the existence of a defect or the cause of the incident in terms of probability, his testimony should be excluded. Ford contends that to be admissible, an expert must present facts from which a jury could infer that one theory is more likely than not the actual cause in fact. Therefore, Ford argues that Armstrong’s opinion must be excluded as speculative. “Causation can be divided into general causation and specific causation, with proof of general causation being a prerequisite to proving specific causation.” In re Bausch & Lomb Contact Lens Solution Prods. Liab. Litig., 2010 WL 1727807 *1 (D.S.C. April 6, 2010). Ford conflates general. and specific causation. Armstrong is a general causation witness who intends to testify that EMI can cause sudden acceleration, not that it caused the sudden acceleration in this or any other case. The admission of expert testimony is fact specific, and the proper inquiry is the helpfulness of the opinion to the trier of fact. See Daubert, supra, 509 U.S. at 591-92, 113 S.Ct. 2786. Doubts regarding usefulness should generally be resolved in favor of admissibility. Aluminum Co. of America v. Sperry Prods., Inc., 285 F.2d 911, 916 (6th Cir.1960); Miles v. General Motors Corp., 262 F.3d 720, 724 (8th Cir. 2001). Were Armstrong’s testimony about EMI reliable, the general causation testimony Armstrong offers would likely be helpful to the jury in understanding what appears to me to be many highly technical and complicated engineering principles. C. Ex Parte Communications Ford argues that Armstrong’s testimony should be excluded as a sanction for his ex parte communications with a Ford employee during the pendency of this case. While a district court may issue sanctions to protect its integrity and prevent abuses of the judicial process, such action is unnecessary here. I disagree with Ford’s characterization of Armstrong’s contact with Mr. Gunsaya as a “brazen violation” of the ex parte communications rule. Armstrong’s deposition shows that he did not believe himself to be violating the rule when he repeated a request for documents from Mr. Gunsaya: Q. And notwithstanding you being told by those plaintiffs’ lawyers not to contact Mr. Gunsaya, you nonetheless reached out to him again in March to try to obtain additional training materials; is that correct? A. Well, maybe I was wrong, but it seemed that we had a business arrangement. * * * * * H? Q. Mr. Armstrong, were you requesting of Mr. Gunsaya materials regarding EMI or EMC in his capacity as an employee of Ford Motor Company? A. Yes, I was, but they were nothing to do with the case. Q. Does anything in this case, the Buck ease, have anything to do with EMI or EMC? A. Yes, it does. Q. And were you attempting to obtain documents related to EMI or EMC from Mr. Gunsaya, an employee of Ford Motor Company? A. Yes, I was, but they weren’t to do with the case. [Doc. 60 at 15]. Ford cites no case imposing sanction it seeks. Although Armstrong should have known better, his violation does not appear intentional. More importantly, it does not appear to have prejudiced Ford in any way. 2. Samuel J. Sero Ford moves to exclude plaintiffs expert Samuel J. Sero, an electrical engineer. Sero has a bachelor of science degree in electrical engineering. He graduated from Carnegie Mellon University in 1967 and worked twelve years with an investor-owned power company. Since 1975, he has performed private consulting on engineering and design of facilities. Since 1989, he has done forensic consulting for litigants, investigating electrocutions, fires, work place injuries, power tool accidents, consumer product accidents, slip and falls, and vehicle related problems including unintended vehicle accelerations as alleged in this case. He is a registered professional engineer in Pennsylvania. Sero intends to opine that: 1) as a matter of general causation, EMI can induce sudden acceleration; and 2) the sudden acceleration of the White vehicle was more likely than not caused by EMI. Ford argues that Sero’s general causation opinion is unreliable because: 1) it is untested; 2) it has not been peer-reviewed; 8) his methodology has not gained general acceptance; and 4) it is impermissibly based on possibility, not probability. Ford contends that Sero’s specific causation opinion should be excluded because he cannot reliably rule out driver error. Finally, Ford claims that Sero’s testimony is inadmissible because it is not based on sufficient facts or data. A. General Causation Sero intends to testify that EMI can cause a vehicle with an NGSC system to suddenly accelerate. Specifically, Sero opines that “electromagnetic interference generated in and between components in the engine compartment” creates a “signal which activates the cruise control output to the stepper motor [that] would cause the component to pull the cruise control cable to the wide open position.” [Doc. 65-1 at 3]. According to Sero, Ford’s NGSC electromagnetic clutch is energized from the moment the vehicle is started. [Doc. 76 at 10-11.] The cruise control’s microprocessor contains microelectronic components that operate on small voltage values. Id. at 14. The processor takes signals and makes a decision as to whether to activate the clutch and open the throttle. In his view, the system has inadequate filtration and isolation, making it susceptible to electromagnetic interference. “And all it that takes is one voltage input to cause the [stepper motor] to go into operation.” [Doc. 76 at 17]. Ford contends that Sero’s general causation opinion should be excluded because his opinion fails the threshold requirements for reliability under Rule 702(2). I agree. On review of his proposed testimony, I find that Sero’s opinion general causation opinion is unreliable because: 1) his methodology is not reliably applied; 2) his theory is untested; and 3) his theory has not been submitted for peer review or publication. i. Process of Elimination Methodology Standard Sero uses the same methodology for his opinions as to both general and specific causation: a process of elimination. Ford contends that Sero is applying “differential diagnosis,” in which he eliminates potential causes of sudden acceleration until the only remaining cause is EMI. Ford argues that Sero has not reliably applied his methodology, because he cannot rule in EMI as a potential cause in the first place. Buck argues that Sero’s methodology is not a differential diagnosis, but rather “engineering failure analysis” — an analytical tool that identifies the consequences of a failure in a particular component. [Doc. 65 at 7]. According to Buck, Sero’s methodology is a “FMEA” — a Failure Modes and Effects Analysis — in which one first hypothesizes a failure in that component and then ascertains the effect of that failure. Sero also uses the Ford Ishikawa diagram, in which instead of hypothesizing a component failure and extrapolating forward to determine its potential effects, one starts with a malfunction and works backwards to ascertain the specific types of failures that can cause such a malfunction. Buck distinguishes differential diagnosis, arguing that “that technique does not establish a direct link between exposure to a substance and an injury and is usually confined to the world of medicine ... It is a far cry from reconstruction and engineering failure analysis where causal links are accurately determined by applying the laws of physics.” [Doc. 81 at 3]. The difference, at least with respect to Sero’s methodology, appears to be tautological. A brief review of Sero’s testimony makes clear that Sero’s methodology is in essence that which Buck defines as differential diagnosis: eliminating each of the potential causes until one that cannot be ruled out is isolated. Q. And you have no evidence as you sit here today that a transient signal could actually actuate the stepper motor in a Next Generation system, do you? A. Depends on what you want to call evidence. Having eliminated every other possibility, it’s all that’s left. So yes, I do have evidence. [Doc. 76 at 101-102], When pressed again, Sero gave the same explanation: Q. You don’t have any evidence that any particular EMI signal has ever caused a sudden acceleration in his vehicle, have you? A. As I stated before, if the evidence is that nothing else has done it, then it has to be EMI. That’s the evidence that you have. [Doc. 76 at 105-106]. Though Buck argues that Sero does not apply a differential diagnosis, the validity of that technical distinction is ultimately irrelevant to this issue because Sero admittedly applies a process of elimination. Having determined that Sero’s methodology is essentially the same as differential diagnosis, this Circuit’s differential diagnosis standard guides determination of the reliability of Sero’s opinion. In Best, supra, 563 F.3d at 179, the court adopted the following standard for a reliable differential diagnosis: 1) the [witness] must objectively ascertain the nature of the patient’s injury or disease; 2) he or she must “rule in” one or more causes of the injury using a valid methodology; and 3) engage in “standard diagnostic techniques” to rule out alternative causes to reach a conclusion as to which cause is the most likely. In using this methodology to come to a conclusion regarding general causation, “[i]t is important to realize that a fundamental assumption underlying [differential diagnosis] is that the final, suspected ‘cause’ ... must actually be capable of causing the injury.” Clausen v. M/V New Carissa, 339 F.3d 1049, 1057-58 (9th Cir.2003) (internal citations omitted). As noted by the Second Circuit, differential diagnosis might not support a general causation opinion because, “like any process of elimination, it assumes that the final, suspected cause remaining after this process of elimination must actually be capable of causing the injury.” Ruggiero v. Warner-Lambert Co., 424 F.3d 249, 254 (2d Cir.2005) (emphasis supplied). Thus, where an expert employs differential diagnosis to “rule out other potential causes for the injury at issue, he must also rule in the suspected cause, and do so using scientifically valid methodology.” Ruggiero, supra, 424 F.3d at 254 (internal quotations omitted); Tamraz, supra, 620 F.3d at 674 (expert’s differential diagnosis testimony excluded where he could not reliably conclude that manganese could cause Parkinson’s disease). Buck argues that the “ruling in/ruling out” analysis described in medical diagnosis/etiology cases is not germane because here Sero testified that there is only one explanation for a stationary car that suddenly accelerates at a high rate of speed on its own — an EMI-induced fault in the electronics. This is obviously circular, as Sero must have some methodology to conclude that EMI-induced fault is a possibility in the first place. Buck contends, in any case, that “in this case Mr. Sero reliably ‘ruled in’ EMI as a possible cause of sudden acceleration, which is the same thing as establishing general causation.” [Doc. 81 at 4], Ford disagrees, and so do I. ii. Application of the Methodology— Ruling in EMI Sero concludes that EMI can cause sudden acceleration on the basis of two Ford documents, and application of general principles of engineering. Sero has not reliably ruled in EMI. In brief, Sero’s opinion is that EMI can cause sudden acceleration because: 1) numerous components under the hood generate an infinite variety of transient electronic signals; 2) these signals can potentially travel along the numerous interconnections between and among the various wiring harnesses, ground connections and power connections in the car; and 3) because of these interconnections, transient electronic impulses can sometimes invade pathways not intended by the design, and some of those pathways, in turn, could lead to the output transistors on the cruise control’s printed circuit board, signaling the throttle to open to a near wide-open condition. Sero thus testifies persuasively about the scientific and engineering principles suggesting the existence of EMI and its ability to trigger sudden acceleration. But although appeals to general scientific principles are appropriate as part and parcel of a reliable methodology, a wholesale, unexplained reliance on those principles as the sum total of an expert’s methodology does not pass Daubert scrutiny. E.g., Meadows v. Anchor Longwall & Rebuild, Inc., 306 Fed.Appx. 781, 789 (3d Cir.2009) (affirming the exclusion of expert evidence solely on “generally accepted principles of basic physics (recognized since the time of Sir Isaac Newton).”). “[T]he courtroom is not the place for scientific guesswork, even of the inspired sort.” Rosen v. Ciba-Geigy Corp., 78 F.3d 316, 319 (7th Cir.1996). Rather, “[t]he important thing is not that experts reach the right conclusion, but that they reach it via a sound methodology.” Tamraz, supra, 620 F.3d at 675 (citing Daubert, supra, 509 U.S. at 595, 113 S.Ct. 2786). Sero has not reliably ruled in EMI as a potential cause of sudden acceleration, because he has not “supplemented his conclusions based on general engineering principles with reliable methodology.” Meadows, supra, 306 Fed.Appx. at 789. As discussed below, Sero’s opinion lacks the indicia of reliability as set forth in Daubert. Sero’s theory has not been: 1) verified through testing; 2) published or peer reviewed; 3) generally accepted. Finally, Sero’s theory is not based on sufficient facts or data. “Nothing in either Daubert or the Federal Rules of Evidence requires a district court to admit opinion evidence that is connected to existing data only by the ipse dixit of the expert.” Joiner, supra, 522 U.S. at 146, 118 S.Ct. 512. Therefore I find that “there is simply too great an analytical gap between the data and the opinion proffered”, id., to permit Sero’s opinions to go to the jury. (a). Testing Ford contends that Sero’s general causation opinion fails the Daubert standard because his theory is untested. Either Sero’s general causation opinion should be excluded for failure to test a clearly testable theory, argues Ford, or because an untestable theory is unreliable. Sero acknowledges that he has never found a signal capable of activating the servo in any of the speed control systems he has tested. Nor is Sero aware of anyone who has ever found such a signal. But Sero asserts that he has “tested for the impact of EMI upon the Next Generation cruise control system and I have simulated the effects. However, it is not feasible economically or practically to conduct the type of testing that would be all inclusive.” [Doc. 65-2 at 3-4], Sero testified that he could “trigger throttle opening due to an injection of EMI.” [Doc. 76 at 41]. “I can take a device and put it on a bench, and knowing what the operating parameters are of the signals that it takes to operate the device, I can do that. And I can inject them at the signal point and have the device take off and do what it’s suppose to do.” Id. This is entirely insufficient to verify Sero’s theory. Sero does not, for example, describe: what he did; how he did it; what, if any controls he used; what voltages he used; what “simulated the effects” means; or whether the effects can be reliably analogized to a cruise control in a car. Indeed, at the Daubert hearing he also testified that he had never “simulated a fault” on an NGSC system, and that he has never attempted to try to get any electrical transients to activate a NGSC system at all. [Doc. 76 at 101], We are left to wonder how Sero knows what he says he knows. For example, Sero states in his report that “[w]hen an EMI-induced failure sends an unintended signal to the throttle, there are usually no detectable marks.” [Doc. 65-1 at 3]. But he testified that he has also never been able to get a transient signal to activate any other kind of speed control. [Doc. 76 at 102], Without any person ever having found a signal that could activate a servo, how can Sero state what usually occurs? “The criterion of the scientific status of a theory is its falsifiability, or refutability, or testability.” Daubert, supra, 509 U.S. at 593, 113 S.Ct. 2786. Without providing any guidelines by which Sero’s simulation might be replicated, its results verified and critiqued, his testing cannot meet this element of the Daubert standard. E.g. Smelser, supra, 105 F.3d at 304-305. But Ford’s argument that an untestable or untested theory is per se unreliable is an overstatement. The test of reliability is “flexible, and Daubert’s list of specific factors neither necessarily nor exclusively applies to all experts or in every case.” Kumho Tire, supra, 526 U.S. at 140, 119 S.Ct. 1167. “[W]hether Daubert’s specific factors are, or are not, reasonable measures of reliability in a particular case is a matter that the law grants the trial judge broad latitude to determine.” Id. at 153, 119 S.Ct. 1167. Though whether a techñique can be and has been tested is ordinarily a “key question” to be answered under Daubert, supra at 593-594, 113 S.Ct. 2786, a hypothesis may satisfy Daubert even if it is untested so long as the expert provides sufficient alternative indicia of reliability. Ford relies on the Sixth Circuit’s opinion in Pride, supra, 218 F.3d 566 for the contention that an untested hypothesis fails the Daubert test. But Ford’s reliance on Pride is not persuasive. There, while the expert’s failure to test was significant, more important was the fact that the expert’s theory was actually contradicted by the physical evidence in the case. Id at 578. The fact that Sero has not verified his theory through any testing on the subject vehicle, an exemplar vehicle, or any other vehicle with an NGSC system weighs against finding his opinion reliable, but it is not a per se bar. (b). Peer Review As the Court indicated in Daubert, peer review and publication are good indicators of reliability. Submission of a theory to the “scrutiny of the scientific community” is, generally, a part of “good science,” and therefore whether or not a proposed theory has been published for peer review “will be a relevant, though not dispositive, consideration in assessing the scientific validity of a particular technique or methodology on which an opinion is premised.” Daubert, supra, 509 U.S. at 593, 113 S.Ct. 2786. Neither Sero’s EMI theory nor any of his work on sudden acceleration has ever been published or peer reviewed. [Doc. 76 at 78-79]. Nor is Sero aware of any peer-reviewed article in a recognized journal finding that a transient signal can cause a cruise control to activate and result in sudden acceleration. Id. at 105. Buck does not dispute this, but instead argues that “the presence of a peer-reviewed article is only one facet of determining reliability of a methodology.” [Doc. 81 at 7], While Buck is correct that the lack of peer review is not dispositive of reliability, the fact that Sero, who has worked in this field for decades, has never had this theory reviewed weighs heavily against admitting his testimony. (c). Factual Basis Under Federal Rule of Evidence 702, an expert witness’s testimony is admissible if it is based on “sufficient facts or data.” Ford argues that Sero’s opinion that EMI can ever cause sudden acceleration lacks the necessary factual predicate and should therefore be excluded. Specifically, Ford contends that Sero is unaware of key mechanical components of the NGSC system that make it impervious to EMI. According to Ford, Sero: 1) does not understand the NGSC three-phase system; 2) incorrectly believes that the NGSC system only has one filter to protect against transient signals; 3) erroneously claims that the NGSC system is not enclosed in aluminum; and 4) misunderstands the function of the brake on/off switch in vehicles equipped with the NGSC system. (i). Three-Phase System According to Sero, Ford’s cruise control is negligently designed because power is supplied to the cruise control immediately upon ignition. As such, Sero contends, it takes only one fault to open the throttle. [Doc. 76 at 16-17] According to Sero, as long as one transient signal enters the integrated chip, the pulse generator will “set in motion” the stepper motor. Id. at 16. Ford argues that Sero’s assumption that a single transient signal is capable of activating the NGSC system and can cause the vehicle to reach wide-open throttle is faulty. Ford insists that Sero “ignores the NGSC system’s unique EMI safeguard— namely, that the system requires three separate signals to interact with the system in a specific order and for a specific duration to cause the system to work at all.” [Doc. 83 at 8] Ford argues that Sero’s misunderstanding of the NGSC system demonstrates complete ignorance of the basic circuitry of the system. According to Ford’s expert, Sero’s single transient signal theory is scientifically impossible because the signals that actually cause the NGSC system to work come from three different locations. Sero responds that, although it is true that the system opens the throttle through the receipt of three signals in a sequence, all three phases are triggered by one input. In an affidavit submitted after the Daubert hearings, Sero contends that— based on the NGSC schematic — a single signal to the main integrated circuit chip, Ul, sets the three-step function into action. It is this signal which Sero contends could trigger the cruise control. According to Buck, “the problem resides in the fact that the input signal is unintended and triggers the sequence of outputs when it should not.” [Doc. 88 at 4]. Sero says that it is possible, Declercq says that it is not. Sero’s qualifications make him qualified to testify as to his reading of the NGSC schematic. It is not my role as a gatekeeper to determine whether Sero’s conclusion is correct. E.g., Jahn v. Equine Servs., PSC, 233 F.3d 382, 391 (6th Cir.2000) (“But comparing two pieces of evidence and determining which is more credible should be left for the finder of fact and should not be considered when ruling on Rule 702 admissibility.”). Were Sero’s testimony otherwise reliable, this dispute would be for the jury to resolve. (ii). Filter Ford argues that “Sero’s assumption that the NGSC system only has one capacitor to filter potential transient signals is incorrect. Rather, the NGSC system has a series of barriers that prevent EMI at varying frequencies from interfering with the system.” [Doc. 83 at 15]. A closer read of the deposition testimony to which Ford refers makes clear that Sero’s point was not that there was only one capacitor, but rather that capacitors are insufficient to filter every potential signal. Q. And one of design features of the Next Generation system that [Declercq] claims eliminates EMI effects is the following: All input and output circuits are filtered. Do you agree with that? A. I’ll agree that there’s probably some form of capacitor at the termination point of all of the wire connectors coming in. The problem with it, especially in a microcircuit like this, is that the capacitors are extremely small, have very low power ratings, and because of that, they are also only good for certain ranges of frequencies. So that anything above or below that range goes right past it. It doesn’t — you know, you can’t put in a single capacitor conductor or even capacitor conductor combination that’s going to filter out everything. [Doc. 81-1 at 56]. As Ford’s argument is based on a misunderstanding of Sero’s point, I need not address it. (iii). Aluminum Enclosure The parties and their experts spent considerable time both at the Daubert hearing and in their briefs debating whether the cruise control’s electromagnetic clutch is fully enclosed in aluminum-shielded package that eliminates any external effects of EMI. According to Sero, the cruise control is not entirely protected by metal. According to Ford, though a portion of the enclosure is made of plastic, the enclosure creates a metal seal around the entire unit. Again, Ford is imprecise. Declercq, Ford’s expert, did not testify that the aluminum box entirely seals the unit when closed. Instead, he said: A. [W]hen you close the thing, there’s a spacing of perhaps a few thousandths of an inch, 5/1000 of an inch, something like that ... So your wavelength is about an inch and a half long. And so the wavelength is not going to penetrate or get into that small crack. Similarly, all the way around the pulley outlet, that is not dramatically sealed either, and it also is an opening; but it, again, takes a very specific frequency directed in a very specific direction because when you get above 400 megahertz, we are talking about line of sight transmission .... The signals that could possibly get in here would have to be well into the gigahertz region, and there are very few of those kind of emitters. Q. You mean they just don’t exist? A. They do exist. Q. They don’t exist inside the vehicle? A. In general. Then if they do exist— they can be generated inside the vehicle, the wiring harness and the componentry inside the vehicle, they can be generated. But if they are generated externally, then the shield and various sheet metal of the vehicle is a super good shield to prevent it from entering the vehicle. [Doc. 81-3 at 47-49] (emphasis supplied). In short, while Declercq is clear that he does not believe that any EMI would penetrate the aluminum box, he does not testify that the NGSC system is fully enclosed in aluminum. Though the crack is small, Declereq’s testimony is that there is a crack. Ford’s argument on this ground is misplaced. (iv). BOO Switch Ford also disputes Sero’s description of the brake on/off, or BOO, switch. Ford states that “[u]nlike past systems, where the speed control system theoretically could reactivate after the driver hits the brake and then releases it, the NGSC system functions differently, by disengaging the speed control system as soon as the driver applies the brake, and does not re-engage when the brake is released, regardless of the cause of the speed control’s initial activation.” [Doc. 75 at 74-75]. Sero cites two sources for his theory that when EMI causes a vehicle to suddenly accelerate, applying the brake would not deactivate the speed control system and would allow the system to re-engage when the brake is released. First, Sero argues that his opinion is confirmed by Ford engineer Casey Mulder. In a 1998 email, Mulder explains that in wide open throttle, the engine no longer produces vacuum. Thus, an operator of a vehicle with power assist brakes would only have one press of the brakes with the assist, and then subsequent depressions of the brakes would have substantially less braking power. This loss of braking assist would make it much harder “to push hard enough to open the brake pressure switch.” [Doc. 81-5]. Buck characterizes this email as a description of “a sudden acceleration event in an Explorer equipped with the NGSC system where he found both that the BOO did not work and that the brakes were impeded — placing the driver in ‘big trouble.’ ” [Doc. 88 at 4], Buck neglects to note that Mulder specifically limited discussion to situations in which the BOO switch independently malfunctioned: Why would the BOO switch not work? If you ground the indicator light line .pin4, like when we download calibrations to the module) [sic] BOO will not respond. So if an ‘event’ occurred where speed control went to WOT and for some reason pin 4 were grounded, incidents just as described by customers could occur. [Doc. 88-2 at 2]. This email does not support Sero’s contention that applying the brake would not deactivate the speed control system. Instead, Mulder says that it would be significantly more difficult to trigger the BOO switch on a second application of the brakes due to a loss of power assist. This email also does not support Sero’s contention that the system might re-engage with the brake is released — it is simply not addressed. Sero states that he did testing on the White vehicle: he floored the accelerator and braked at the same time, and found that he was unable to get the vehicle to stop even though he was applying approximately 150-200 lbs. of force. [Doc 81 at 10-11]. Sero does not explain whether and why this “test” — holding down the accelerator — is analogous to an errant transient signal. Additionally, this test does not seem analogous to the issue in question— depletion of the power assist — in the Mulder email. Accordingly, I find that Sero’s opinion regarding the BOO switch is not reliable. B. Specific Causation Sero intends to opine “that the most likely cause of the crash of the subject 1999 Ford Expedition on April 27, 2006 was a failure of the vehicle’s cruise control system.” [Doc. 58-7 at 68]. Ford argues that Sero’s specific causation opinion should be excluded because his methodology is not reliably applied. I agree, i. Methodology Sero relies on the same “engineering failure analysis” to determine that EMI caused the incident in this case. [Doc. 81 at 14]. Buck asserts that this methodology is generally accepted and reliable. As discussed above, Sero’s opinion relies on his ability to rule in EMI as a potential cause. His specific causation opinion also requires that he reliably ruling out mechanical problems and driver error. Tamraz, supra, 620 F.3d 665. Buck contends that “differential diagnosis ... is not applicable to engineering failure analysis. As to the latter, there is no authority requiring an expert to definitively rule out all possible causes of the failure of a machine.” [Doc. 81 at 3-4]. The eases Buck cites for this proposition are unpersuasive and distinguishable. The major case on which Buck relies, John, supra, 233 F.3d 382, is, in fact, a medical diagnosis case, not an engineering failure analysis. In that case, the experts’ opinions were “based on undisputed objective medical facts,” id. at 392. The court noted that “[Booking at the records o£ test results and physical symptoms to infer the presence of an infection is not a methodologically unsound ‘assumption’ or ‘guess’— it is a diagnosis.” Id. at 391. The court emphasized that “[c]ertainty [was] not to be found in this case [was] due in considerable part to the lack of medical records kept by the defendants.” Id. at 390. The decision in Christie v. Mazda Motor of Amer. Inc., 2006 WL 2128897, at *4 (E.D.Tenn. July 27, 2006), is likewise distinguishable. The expert in that case used reliable methodology to narrow the possible causes of the plaintiffs injuries to two potential design defects. He reliably eliminated the plaintiffs negligence as a possible cause. Id. The court held that the expert’s inability to further narrow his opinion to determine which of the two defects was the actual cause did not make his testimony inadmissable. Id. at *3. This issue is instead directly analogous to that in Viking Yacht Co. v. Composites One LLC, 615 F.Supp.2d 327 (D.N.J.2009). In that case, the expert, an engineer, concluded that a product was defective “by eliminating other possible causes by a process of elimination.” Id. at 335. The court, analogizing this methodology to differential diagnosis in medical cases, excluded the expert’s causation testimony because he was not qualified to reliably rule out environmental causes. Id. Thus, the law does not support Buck’s proposed distinction between differential diagnosis and engineering failure analysis. Buck’s assertion is further undermined both by the fact that Buck is unable to articulate any difference between Sero’s ruling-out process of elimination analysis and differential diagnosis in practice, because Buck defines Sero’s methodology as a differential diagnosis in her opposition to Ford’s motion in limine, and because Buck acknowledges that “plaintiffs have the burden of ruling out driver error.” [Doc. 81 at 4]. Thus, the alleged distinction between Sero’s ruling-out analysis and a differential diagnosis appears to be a distinction without a difference for the specific causation question in this case. Regardless of what it is called, Ford characterizes Sero’s methodology as a flawed process of elimination that purportedly rules out all potential causes of sudden acceleration. Ford argues that Sero’s opinion does not have a reliable basis and therefore his opinion that EMI most likely caused this incident should be excluded. I agree. In short, Sero determined that there are three potential causes of sudden acceleration, eliminated two of those causes, and therefore concluded that the one cause remaining is the likely cause. There is nothing inherently unreliable about a process of elimination methodology, call it what you will. E.g., Carmichael v. Samyang Tires, Inc., 923 F.Supp. 1514, 1520 (S.D.Ala.1996) (“the Court perceives no inherent flaw in a process-of-elimination form of proof per se, so long as the underlying methodology is scientifically valid”), rev’d, 131 F.3d 1433 (11th Cir.1997), rev’d sub nom Kumho Tire Co., Ltd. v. Carmichael, 526 U.S. 137, 119 S.Ct. 1167, 143 L.Ed.2d 238 (1999). However, as discussed in the next section, Sero applied this methodology in an unreliable manner on the basis of insufficient facts or data. ii. Reliable Bases for Ruling out Driver Error Sero intends to testify that he reliably eliminated two of his purported three potential causes for the vehicle’s sudden acceleration: namely, vehicle mechanical problems and driver error. Thus, according to Buck, Sero has reliably concluded that an electrical malfunction is the most likely cause. [Doc. 58-7 at 11, 73]. Having found no mechanical fault, Sero concluded that the cause of the sudden rapid acceleration could only be one of two explanations: “either the driver mistakenly put ‘pedal to the metal’ or the cruise control system electronically failed.” [Doc. 65 at 13]. Sero then proceeded to rule out driver error as a potential cause. To make this determination, Sero relied on: 1) witness testimony; 2) brake pedal wear; and 3) driver habit. Sero’s specific causation opinion must be excluded because he has not reliably ruled out driver error. (a). Witness testimony In the Daubert hearing, Sero testified that he relied on witness testimony of individuals who were present at the Nickles Bakery during the incident who saw White’s feet on the ground while the vehicle continued to accelerate. Sero repeatedly testified that this deposition testimony was a major predicate for his opinion: THE COURT: you rule out operator error because the brake pedal was not worn in a manner indicative of a, quote, two-footed driver? THE WITNESS: Well, partly. And also the fact when the vehicle came to a stop and the man was not on any of the pedals, the vehicle was still trying to accelerate. [Doc. 76 at 73-74], THE COURT: At some point in expressing your opinion you expressed the view that — well, what view do you express about the brake pedal and the unlikelihood or likelihood that Mr. White was a two-footed driver? THE WITNESS: I think that the fact that when the car came to rest that— and I don’t remember the name of the person that came over to the car while he was in it and actually turned the car off, noticed that he was actually not on any pedals, and yet the vehicle was still going high speed, the wheels were turning. So that whatever Mr. White may have thought he was doing or anything else, the simple fact is that the vehicle, when it came to rest, was still under idle and throttle condition without anybody’s feet on any pedals, which only leaves the car. [Doc. 76 at 137-138], Ford asserts that no such testimony exists, and neither Buck nor Sero has identified where in the record any witness made this statement. I therefore find that Sero cannot rely on this witness testimony to rule out driver error. (b). Brake Pedal Wear Sero intends to testify that his investigation of White’s car revealed wear on the brake pedal consistent with a right-footed driver, and that a right-footed driver would have hit the brake, not the accelerator in the circumstances of this incident. Sero opines that the vehicle’s pedal wear is not consistent with a two-footed driver. Nothing in Sero’s curriculum vitae indicates that he is qualified to make such a determination. Nor has Sero conducted any testing to determine how many times a driver must drive with two feet before observable brake pedal wear would appear. [Doc. 76 at 139]. Even if Sero were qualified to testify as to the pedal wear on the White vehicle, he could not reliably base any opinion about White’s driving habits on that pedal wear. Sero testified that he did not know: 1) how many miles were on the White vehicle; 2) whether White was the original owner of the vehicle; 3) whether White was the primary driver; 4) whether the wear on the pedals of this vehicle came from White or some other driver; 5) how many times White had driven the vehicle. [Doc. 76 at 81 — 82;139]. (c). Human Habit Relying on his conclusion that White was a right-footed driver, Sero testified that as White was slowing down, he would have had his foot hovering over the break and would not be likely to hit the accelerator. This is true, Sero contends, because “[fit’s a matter of habit of what you do in a vehicle.” [Doc. 65 at 13]. At deposition, when asked how he had eliminated driver error as a potential cause of the crash, Sero explained: So he took his foot off the brake and then the vehicle takes off ... And it had to be going more than idle speed in order to just get over the curb ... So the vehicle had to have gone to an acceleration rate. His foot as it came off the brake would have still been in the brake position where he could have hit the brake and would have hit the brake rather than going over to the accelerator; would never have thought — been an occurrence that he would have thought he was hitting the brake and was hitting the accelerator. [Doc. 58-7 at 12]. Sero is not a human factors expert and is not qualified to give this opinion. Buck intends to call a separate expert to testify with respect to human factors. Buck argues that “the analysis is the province of Dr. William Berg” and “[t]here is no requirement that one expert opine as to all facets of a specific causation determination.” [Doc. 81 at 14]. But Sero did not rely on Dr. Berg’s analysis, and therefore Dr. Berg’s conclusions cannot supply a post-hoc reliable basis for Sero’s specific causation determination. 3. Dr. William Berg Dr. Berg intends to testify to the following: 1) that a human factors analysis and reconstruction of subject accident demonstrates that it is not probable that the incident was caused by a pedal error on the part of Mr. White; 2) that an analysis of Ford’s Updegrove investigation demonstrates that the substantial majority of the 2,877 events catalogued in that study were not caused by driver error; 3) that the design of the NGSC is defective in that it does not provid