Summary
holding that district courts "act as a gatekeeper . . . ensuring only reliable evidence makes it to the jury"
Summary of this case from L.V. Nagle & Assocs. v. Tubular Steel, Inc.Opinion
22-3222
01-31-2023
JACQUELYNN WILLIAMS, et al., Plaintiffs-Appellants, v. EDWARD SYPHAN; PITT OHIO EXPRESS, LLC, Defendants-Appellees.
NOT RECOMMENDED FOR PUBLICATION
ON APPEAL FROM THE UNITED STATES DISTRICT COURT FOR THE NORTHERN DISTRICT OF OHIO
Before: SUTTON, Chief Judge; STRANCH and DAVIS, Circuit Judges.
OPINION
SUTTON, Chief Judge.
A semitruck struck and killed James Williams on the Ohio Turnpike. Williams' wife Jacquelynn and their children (collectively, "Plaintiffs") sued the semitruck driver, Edward Syphan, and his employer, Pitt Ohio Express, on negligence and other grounds under Ohio law. The success of the claims hinged on whether Williams was in the road or on the shoulder when the accident happened. Plaintiffs' expert placed Williams on the shoulder. The district court excluded that opinion as unreliable and granted summary judgment to Syphan and Pitt Ohio. Plaintiffs appeal. We affirm.
I.
A tragic series of events occurred in the early morning of July 17, 2018, ending in Williams' death. Mr. and Mrs. Williams had been having marital problems. Mrs. Williams wanted a divorce. Williams wanted to reconcile. Everything came to a head when, in the "wee hours of the morning on the 17th," Williams went to Mrs. Williams' home to convince her to stay together. R.46-2 at 13. Unable to persuade her, Williams became violent. Mrs. Williams called the police and left the home with her children to stay in a hotel.
Williams drove east on the Ohio Turnpike on I-80. Then, at about 3:18 a.m., he drove off the road, onto the shoulder, into a guardrail, through a ditch, through a fence, into a field, back through the fence, back across three lanes of traffic, and into the road's concrete center divider. Williams exited his GMC, leaving it running with his keys, wallet, ID, money, and cell phone inside. He did not call 911, as his cell phone records confirmed. He walked about four tenths of a mile, still heading east along the turnpike.
Meanwhile, Syphan had been traveling east on the turnpike in his Freightliner semitruck. Syphan drove in the right lane at 67 miles per hour, the speed set by his truck's computer. At about 3:30 a.m. the truck struck Williams. Syphan did not see anything before the collision and thought he had hit a deer. The truck's computer, upon examination after the accident, did not show that the truck had switched lanes before or immediately after the accident. Syphan drove about an eighth of a mile further before pulling over and calling his terminal to report the accident and request a tow truck. He waited there for five hours before assistance came. At that point, Williams' body was discovered in a ditch alongside the turnpike.
Plaintiffs sued Syphan and Pitt Ohio in Ohio state court, raising negligence, survivorship, wrongful death, and gross negligence claims. The defendants removed the case to federal court. One turning point for liability was Williams' location at impact. For Plaintiffs' tort claims to succeed, Williams must have been on the shoulder and not in the roadway. And without any physical evidence or witness testimony placing Williams on the shoulder, Plaintiffs had to rely on expert testimony. They retained Henry Lipian, an accident reconstructionist. He issued a report placing Williams on the shoulder.
Syphan and Pitt Ohio moved for summary judgment. At the motion hearing, the district court explored the bases for Lipian's conclusions. Syphan and Pitt Ohio filed a motion to preclude Lipian's testimony. The district court held a Daubert hearing and granted the motion. The district court then granted the motion for summary judgment because at that point, as Plaintiffs conceded, no other evidence showed that Syphan drove negligently.
II.
A pair of four-factor tests govern the admissibility of expert opinions. Evidence Rule 702 requires courts to decide if the expert's testimony "will help the trier of fact," is "based on sufficient facts or data," is produced from "reliable principles and methods," and "reliably applie[s] the principles and methods to the facts of the case." Fed.R.Evid. 702. Daubert v. Merrell Dow Pharmaceuticals, Inc., in turn, sets out four factors for assessing a methodology's reliability: testability; peer review; error rate; and general acceptance in the relevant scientific community. 509 U.S. 579, 593-94 (1993). The Daubert inquiry is "flexible," id. at 594, meaning that each factor may not bear on every case, Kumho Tire Co. v. Carmichael, 526 U.S. 137, 151 (1999). Because non-exhaustive, multi-factor tests "run the risk of obscuring the core inquiry," courts must not lose their grip on the two "key handholds" of Rule 702-a reliable methodology, reliably applied. United States v. Gissantaner, 990 F.3d 457, 463 (6th Cir. 2021).
The district court acts as a gatekeeper under this framework, ensuring only reliable expert evidence makes it to the jury. Daubert, 509 U.S. at 597. We review a district court's admissibility decision for abuse of discretion. Kumho Tire, 526 U.S. at 152. Recognizing the district court's "considerable leeway" in acting as the gatekeeper, id., we reverse "only where we are left with a definite and firm conviction that it committed a clear error of judgment," Conwood Co. v. U.S. Tobacco Co., 290 F.3d 768, 781 (6th Cir. 2002).
The district court did not abuse its discretion by excluding Lipian's testimony. Lipian calculated that Williams stood on the shoulder, within one-to-two feet of the fog line, and that the right front side of the truck also crossed onto the shoulder, between one-to-three feet past the fog line. And he gave himself a one-to-three-foot margin of error. The key problem for the district court, and the key problem for us, is that no reliable methodology connects these ranges to the known facts. His report outlines various intermediary findings and how he determined them with a good deal of specificity. But Lipian does not explain how those inputs add up to his output: location ranges for Williams and the truck and a margin-of-error range. As a result, Lipian's opinion is not "the product of reliable principles and methods." Fed.R.Evid. 702(c).
Take a look at Lipian's specific findings, and the missing link emerges. Lipian classified the crash as a forward projection trajectory type. In other words, after the truck hit Williams, he flew straight forward. Lipian calculated Williams' center of mass, evaluated the truck's profile, and inspected the damage to the truck. Based on these pieces of evidence, Lipian concluded that the truck hit Williams above his center of mass. And if a vehicle with a "high or flat front" hits a pedestrian above his center of mass, a forward projection trajectory crash "common[ly]" results. R.87-1 at 4. For throw distance, Lipian used "the Barzely forward projection formula," and its key input (the speed of the truck at impact), to calculate a 230-foot distance. Id. at 7. And as to trajectory, Lipian found it curved to the right after Williams hit the ground and slid off the shoulder into a ditch because of irregularities in the road's slope and surface.
These inputs are helpful starting points, to be sure. But they do not explain why or how Lipian could reach the conclusion that Williams was one-to-two feet outside the fog line or why the margin of error was one-to-three feet. The problem is that these preliminary methodologies- the center-of-mass analysis, the Barzely formula, the unspecified slope where he landed-do not explain his conclusion about where the crash occurred. At most, the inputs indicate that Williams flew forward and then to the right. But they cannot quantify the accident's location as Lipian purports to do.
This problem is not fixed by Lipian's statement in his report that the crash location ranges rest on "the totality of available data." Id. at 12. He never explains what this phrase means or why it fills the gap left by his intermediary findings about crash type, throw distance, and trajectory. With no articulated underlying methodology to pull everything together on this point, we are left with crash location ranges based on equal parts extrapolation and speculation. See Newell Rubbermaid, Inc. v. Raymond Corp., 676 F.3d 521, 527 (6th Cir. 2012) (noting that "[r]ed flags" in expert reports include "improper extrapolation"); Pluck v. BP Oil Pipeline Co., 640 F.3d 671, 680 (6th Cir. 2011) (affirming exclusion when expert could not explain methodology underlying calculations).
Although experts may "extrapolate from existing data," the "analytical gap between the data and the opinion proffered" here is "simply too great." Gen. Elec. Co. v. Joiner, 522 U.S. 136, 146 (1997). Lipian confirmed as much at the Daubert hearing. When asked to explain how he determined these ranges, Lipian responded that they were his "opinion" based on his "training and experience." R.99 at 78. When asked if the error rate could instead be two-to-four feet, Lipian replied: "I don't think it's 4. I think it's, as I said, within 1 to 3. That's my opinion." Id. And when pressed for an underlying citation, methodology, or body of scientific research for these ranges, Lipian could say only, "I recognize it has to be given within a range, and that's my opinion." Id.
Expert opinions, it's true, can be based on the expert's "practical experiences." First Tenn. Bank Nat'l Ass'n v. Barreto, 268 F.3d 319, 335 (6th Cir. 2001). But when, as here, "summary judgment turns largely on factual issues found in an expert report or opinion," an expert report "must include how and why the expert reached a particular result, not merely the expert's conclusory opinions." Thacker v. Ethicon, Inc., 47 F.4th 451, 459 (6th Cir. 2022) (quotations omitted). Lipian never explained how or why his experiences caused him to reach his ultimate crash location opinion. That is to say, Lipian has not demonstrated that this opinion stemmed from a reliable methodology that he reliably applied. Fed.R.Evid. 702. On the key point that matters, it is nothing more than Lipian's "ipse dixit." Joiner, 522 U.S. at 146. And the district court was not obligated to accept it. See id.
Even if Lipian's process for calculating the crash location ranges could be classified as a methodology, it could not be classified as reliable under the Daubert factors. Think of testability. Nothing in the record indicates that either the "totality of available data" or Lipian's "training and experience" can be or has been empirically tested or falsified. R.87-1 at 12; R.99 at 78; see Daubert, 509 U.S. at 593. As for peer review, Lipian testified that his report satisfied this factor because a coworker reviewed it. But peer review in the Daubert sense means submission of a methodology "to the scrutiny of the scientific community," usually through "[p]ublication in a peer-reviewed journal." Gissantaner, 990 F.3d at 464. Turn to error rate. Lipian calculated it to be one-to-three feet. As the district court noted, that's the whole "ball game." R.81 at 22. The error rate could swallow the finding that Williams stood one-to-two feet past the fog line and move the accident onto the road. And general acceptance fares no better. Lipian's statements that he applied the "usual protocol" for crash reconstruction and followed authoritative texts are not enough to show the "[w]idespread" use of his "totality of available data" or "training and experience" methodologies in the accident reconstruction community. R.87-1 at 3, 12; R.99 at 78; Daubert, 509 U.S. at 594. The district court did not abuse its discretion in excluding Lipian's testimony.
Plaintiffs raise numerous counterarguments.
They argue, to start, that the district court exceeded its gatekeeping function during the summary judgment and Daubert hearings. No doubt, the judge engaged heavily with the issues in the case. As "a former tort lawyer defending railroads and trucking companies," the judge acknowledged that he was "well acquainted with what accident reconstructionists do." R.99 at 2. "I love this stuff," he proceeded to say, "and you can't make me unlearn a bunch of stuff that I've learned." R.81 at 30. He then outlined his view of the court's gatekeeping role, noting "I have a duty under Daubert and its progeny, as a gatekeeper, to make sure that you're not putting up evidence that's not scientifically reliable," id. at 24-25, or evidence that he concluded, "as a matter of just physics, isn't right," id. at 31. At the summary judgment hearing, the judge expressed concern that Lipian "really didn't analyze the potential for Mr. Williams to have left the front of that truck at a bit of an angle." Id. at 36. He also voiced his frustration that the defendants had not yet challenged Lipian's report, saying "you're making me do it, and that's not really my job." Id. at 34. Later on, in the midst of the Daubert hearing, the judge thoroughly questioned Lipian about his report and methodology.
Regarding the district court's questioning and the depth of its participation in the two motion hearings, it's possible that the judge, like all judges from time to time, might have said some of these things differently upon further reflection. But Daubert delegates a special gatekeeping function to district courts, which requires considerable engagement with an expert's testimony-to ensure it "both rests on a reliable foundation and is relevant to the task at hand." Daubert, 509 U.S. at 597. That explains why we review their decisions, including the process for making those decisions, for abuse of discretion. Kumho Tire, 526 U.S. at 152-53. All perspectives considered, the court's engagement with the issue did not overstep its discretion.
Plaintiffs next object that the district court improperly required Lipian to analyze Williams' "lateral launch angle." Appellants' Br. 12. Recall that Lipian classified the crash as a forward projection trajectory, in which Williams flew straight forward on impact. Yet the collision extensively damaged the truck's front right side. The district court queried if that damage could have caused Williams to fly off the truck at an angle-the so-called "lateral launch angle." Id. And the court asked Lipian if he could quantify that angle. Lipian replied that although he could not quantify a lateral launch angle, he had factored the truck's damage into his opinion, including through inspecting the damage in person and creating a 3D scan of the truck.
Plaintiffs identify several instances in which the district court pressed Lipian on the lateral launch angle during the Daubert hearing and, in their view, overstepped its gatekeeping role when it suggested that Williams flew off the truck at an angle. But the district court's discussion of the lateral launch angle, while interlaced with statements about the judge's understanding of its potential physical implications, was designed to explore a relevant question in a Daubert hearing: how preliminary findings about the distance Williams was thrown, the speed of the truck, and the place of landing, showed where, along the road surface, Williams was located at impact. That was the question then; it remains the question now. Unable to obtain a satisfactory answer from Lipian, the court concluded that his testimony contained "too much speculation and too little applied knowledge." R.93 at 8. The court did not abuse its discretion in conducting this line of inquiry.
Plaintiffs argue the district court improperly "questioned" Lipian's methodology after finding him so "well-qualified." Reply Br. 6. But an expert's qualifications are one thing, a methodology's reliability another. They are separate features of the Rule 702 inquiry. See, e.g., Newell Rubbermaid, 676 F.3d at 528. Good qualifications in other words cannot save a flawed methodology. See Kumho Tire, 526 U.S. at 153 (finding reasonable the exclusion of unreliable testimony "despite" expert's sound qualifications).
Plaintiffs point out that "rejection of expert testimony is the exception, rather than the rule." In re Scrap Metal Antitrust Litig., 527 F.3d 517, 530 (6th Cir. 2008) (quotations omitted). That is presumably because most experts satisfy each feature of the Rule 702 inquiry. But the whole point of Daubert-the whole point of delegating to district courts the gatekeeping role of ensuring that only scientific evidence that uses a reliable methodology may be submitted to lay jurors-is to ensure that is always so, not that it is sometimes so. Confirming the necessity of Daubert and the bedrock fairness role it plays, this court has affirmed many district court decisions that exclude expert testimony on this ground. See, e.g., United States v. Sammons, 55 F.4th 1062, 1072-74 (6th Cir. 2022); Madej v. Maiden, 951 F.3d 364, 375-77 (6th Cir. 2020); Wilden v. Laury Trans., LLC, 901 F.3d 644, 656 (6th Cir. 2018); United States ex rel. Tenn. Valley Auth. v. 1.72 Acres of Land in Tenn., 821 F.3d 742, 750-51 (6th Cir. 2016); Newell Rubbermaid, 676 F.3d at 529; United States v. Semrau, 693 F.3d 510, 523 (6th Cir. 2012); Pluck, 640 F.3d at 680; Ky. Speedway, LLC v. Nat'l Ass'n of Stock Car Auto Racing, Inc., 588 F.3d 908, 916-19 (6th Cir. 2009); Johnson v. Manitowoc Boom Trucks, Inc., 484 F.3d 426, 430-36 (6th Cir. 2007); Nelson v. Tenn. Gas Pipeline Co., 243 F.3d 244, 251-54 (6th Cir. 2001); Pride v. BIC Corp., 218 F.3d 566, 578 (6th Cir. 2000); see also Hamilton Cnty. Emergency Commc'ns Dist. v. Level 3 Commc'ns LLC, 845 Fed.Appx. 376, 384-88 (6th Cir. 2021); Mohney v. USA Hockey, Inc., 138 Fed.Appx. 804, 808-09 (6th Cir. 2005); Christ v. Sears, Roebuck &Co., 149 F.3d 1182, at *4 (6th Cir. 1998) (unpublished table decision) (per curiam).
Any issues with Lipian's report, Plaintiffs argue, should be addressed through "[v]igorous cross-examination" rather than exclusion. Daubert, 509 U.S. at 596. But the antiseptic of cross-examination applies when evidence is "shaky" but nevertheless "admissible." Id. Lipian's failure to do more than speculate about the crash location-his failure to use a reliable formula or a test, such as the Barzely formula, to explain where Williams was at the point of impact-cannot fairly be described in this way.
Plaintiffs cite four cases in which we have reversed a district court's exclusion of expert testimony. But two of them, Andler v. Clear Channel Broadcasting, Inc., 670 F.3d 717, 728-29 (6th Cir. 2012), and In re Southeastern Milk Antitrust Litigation, 739 F.3d 262, 279-82 (6th Cir. 2014), do not help because they turned on flawed expert data, not methodology.
Dilts v. United Group Services, LLC, 500 Fed.Appx. 440 (6th Cir. 2012), also is a step removed, and unpublished to boot. Two men working on a roof fell to their deaths after a large roof panel crashed. Id. at 441-42. An expert determined how the panel fell by inspecting physical evidence, examining photographs, and performing calculations and algebraic equations with a computer program. Id. at 445-46. The expert concluded that a crane lifted the panel, which struck a beam, rotated counterclockwise, and hit a handrail as it fell. Id. We held that the district court erred in excluding the report, reasoning that the expert had "performed the necessary calculations and sufficiently relied on the laws of physics and mathematics generally employed in accident reconstruction." Id. at 446. But the expert in Dilts explained how the physical evidence, scanned model, and calculations supported his conclusion about how the panel fell. Lipian in contrast cannot explain how his intermediary findings on crash type, throw distance, and trajectory show where the crash occurred. His general invocations of the totality of the evidence and personal experience are precisely the kinds of vague notions of scientific expertise that Daubert-through its admonition against admitting expert testimony based on mere "subjective belief or unsupported speculation"-is designed to prevent. Daubert, 509 U.S. at 590.
Turn to the fourth case, another unpublished case, Palatka v. Savage Arms, Inc., 535 Fed.Appx. 448 (6th Cir. 2013). It involved an alleged design and manufacturing defect, and the district court excluded expert testimony for failing to test a proposed alternative design. Id. at 455-56. We reversed because lack of testing went to the testimony's weight, not its admissibility. Id. But the district court did not exclude Lipian's testimony on this ground.
Plaintiffs identify similarities between Lipian's opinion and the opinions of the defendants' experts, suggesting that this proves the district court erred in excluding Lipian's evidence. But the district court's summary judgment decision did not turn on its acceptance of the defendants' experts and rejection of the plaintiffs' expert. It turned on the reality that, without evidence from Plaintiffs that Williams was on the shoulder of the road at impact, Plaintiffs could not make out a threshold case of negligence. Plaintiffs indeed admitted as much at the summary judgment hearing.
Plaintiffs object that the district court raised its concerns about the reliability of Lipian's report at the summary judgment hearing unprompted. But the court did not rule on the admissibility of the report until after defendants moved to exclude it, Plaintiffs filed a response, and the court held a Daubert hearing. This sequence of events gave Plaintiffs ample opportunity to argue the issue, avoiding any unfair prejudice.
We affirm.