Gini Nester suffered permanent injuries when an unmanned utility vehicle ran her over. Gini and her husband sued Textron, the vehicle’s manufacturer, alleging that the vehicle was defectively designed. A jury agreed and awarded the Nesters a sizeable verdict. Textron then appealed, complaining of the substance and form of the jury charge, the admissibility of two pieces of evidence, and the district court’s refusal to bifurcate the trial. We affirm.I.E-Z-GO, a division of Textron, Inc., makes various utility vehicles. This case involves one such vehicle, the ST 350 Workhorse, derived from E-Z-GO’s ubiquitous golf-cart designs but intended for use in other contexts (like farms and ranches). The Workhorse’s pedal layout is illustrated below.A key turns on the Workhorse’s electrical system, but it does not start the engine. Rather, there is a microswitch in the accelerator pedal, which means pushing the accelerator turns on the engine and propels the vehicle. The Workhorse also has a two-part brake pedal: The lower pedal is the vehicle’s service brake, and the upper pedal engages the parking brake. Textron recommends disengaging the parking brake by depressing it again. But, important here, there is a second way to disengage the parking brake: by pushing the accelerator—the two pedals are mechanically linked. Thus, if the parking brake is engaged, depressing the accelerator will (1) turn on the engine; (2) release the parking brake; and (3) accelerate the vehicle.Textron foresaw certain risks in this scheme, and it warned operators to always turn the key to the off position before exiting the vehicle. Yet, Textron was likewise aware that users disobeyed this directive from time to time. All this is to say, should one neglect to turn off the key, all it takes is something to fall with sufficient force on the accelerator to create a runaway vehicle. That is this case.Virginia (“Gini”) and Robert Nester owned and used a Workhorse on their Texas ranch. One afternoon, Gini set out on the Workhorse to move a group of cattle, all the while storing a 50-pound bag of cattle feed on the Workhorse’s floorboard. Once Gini reached a gate, she applied the parking brake and exited the vehicle. As far as the parties can surmise, the cattle then knocked over the feed bag onto the accelerator, causing the Workhorse to run over Gini and render her a quadriplegic.The Nesters filed a diversity suit against Textron, bringing claims for design defect, marketing defect, and gross negligence (for punitive damages) under Texas law. The design-defect claim alleged the pedal configuration— specifically, the link between the accelerator and parking brake—created an unreasonable risk of unintended acceleration. At trial, the Nesters proposed the following four safer alternative designs.First, and principally, the Nesters suggested that removing the link between the accelerator and the parking brake would cause the brake to remain engaged even if something hit the accelerator.Textron responded that the revving engine would overpower the parking brake, causing excessive wear on the brakes. Second, the Nesters suggested that a pedal guard (a shield of sorts) would stop objects from falling on the pedal. Textron responded that such a design would create a tripping hazard. Third, the Nesters suggested that a weight-sensitive switch in the driver’s seat would turn off the engine when the operator exited the vehicle. Textron responded that the switch would render the Workhorse unusable in the farm context because the bumpiness of the terrain would cause the vehicle to constantly stop and start. And fourth, the Nesters suggested that a second, hand-operated parking brake would prevent future accidents. Textron responded that such a design, used only on vehicles with hydraulic brakes, could not be adapted to the Workhorse’s mechanical-brake system.Following the Nesters’ case in chief, Textron moved for judgment as a matter of law. The district court denied the motion. The jury then found for the Nesters only on the design-defect claim. Furthermore, the jury found Gini negligent in her own right, apportioning 50% of fault to Gini and 50% to Textron. And finally, the jury awarded the Nesters over $15 million in pre-apportionment actual damages. After trial, Textron renewed its motion for judgment as a matter of law and coupled it with an alternative motion for new trial. The district court denied both.Textron appealed, abandoning its arguments for rendition and seeking only reversal and a new trial on four grounds: (1) the jury received an erroneous definition of “safer alternative design”; (2) a single-answer jury question erroneously commingled both supported and unsupported alternative-design theories; (3) the district court incorrectly admitted two key pieces of evidence; and (4) the district court erred in refusing to bifurcate the trial.II.Textron’s four appellate issues are distinct, each subject to its own standard of review. We take them in turn.A. The Definition of “Safer Alternative Design”“[S]tate law governs the substance of the jury instructions,” but “the manner of giving instructions is controlled by federal law.” Turlington v. Phillips Petroleum Co., 795 F.2d 434, 439 (5th Cir. 1986) (internal quotations and emphasis omitted). A district court must “correctly and adequately instruct the jury as to the law to be followed in deciding the issues, however; the court is not compelled to give even every correct instruction offered by the parties.” Alexander v. Conveyors & Dumpers, Inc., 731 F.2d 1221, 1227 (5th Cir. 1984) (per curiam).Under Texas law, proof of a design defect requires proof of a “safer alternative design,” which the district court defined in the following manner:“Safer alternative design” means a product design other than the one actually used that in reasonable probability— would have prevented or significantly reduced the risk of the injury in question without substantially impairing the product’s utility; and was economically and technologically feasible at the time the product left the control of Textron by the application of existing or reasonably achievable scientific knowledge. This definition tracked both the relevant Texas statute[1] and pattern jury instruction (“PJC”). See Tex. Civ. PRAC. & REM. CODE § 82.005(b); Texas Pattern Jury Instruction, Design Defect § 71.4B (2016).Textron argues, however, that the court erred by refusing to give a more expansive definition. Specifically, Textron requested inclusion of the longstanding Texas rule that a safer alternative design is not just one that would have prevented the injury in question but must also be one that would not have “impose[d] an equal or greater risk of harm” under other circumstances. Uniroyal Goodrich Tire Co. v. Martinez, 977 S.W.2d 328, 337 (Tex. 1998); Because the district court’s definition failed to reference this “overall safety” component, Textron requests either reversal or certification to the Supreme Court of Texas to clarify Texas law on the matter.We review jury-instruction decisions for abuse of discretion. Janvey v. Dillon Gage, Inc., 856 F.3d 377, 388 (5th Cir. 2017). In general, reversal is appropriate only when the “charge as a whole leaves [the court] with substantial and ineradicable doubt whether the jury [was] properly guided in its deliberations” and the challenged instruction “affected the outcome of the case.” Id. (alterations in original and internal quotations omitted). And we have a particular three-part rubric for cases that involve requested-but-denied instructions (as this case does). We reverse only if the requested instruction “1) was a substantially correct statement of law, 2) was not substantially covered in the charge as a whole, and 3) concerned an important point in the trial such that the failure to instruct the jury on the issue seriously impaired the [party's] ability to present a given [claim].” Id. (alterations in original and internal quotations omitted).At step one, the Nesters do not dispute that Textron’s requested instruction is a substantively correct statement of Texas law. In other words, everyone agrees that a safer alternative design must not render the product less safe in other circumstances.The dispute hinges instead on the second step of our three-part test: Did the district court’s instruction “substantially cover” the overall-safety component? Id. Textron’s argument here is twofold. Textron points out first that the Supreme Court of Texas has, on rare occasion, found a PJC-derived instruction erroneous. See Ford Motor Co. v. Ledesma, 242 S.W.3d 32, 42 (Tex. 2007); And Textron then anchors its argument almost entirely on a concluding sentence from a recent design-defect case:To impose liability on [the defendant], the plaintiffs must have presented evidence of an alternative design that (1) would have been safer for [the injured person] and prevented or significantly reduced his risk of injury, (2) would not have been less safe in other circumstances and increased the risks to other users, (3) would not have substantially impaired the [product]‘s utility, and (4) was economically and technologically feasible at the time.Genie Indus., Inc. v. Matak, 462 S.W.3d 1, 9 (Tex. 2015). Because Genie lists the overall-safety component as numerically separate from those other components of the district court’s definition, Textron contends that overall safety was unaccounted for.But by looking with tunnel vision to the above excerpt from Genie, Textron misses the broader picture. The overall-safety component originated not from Genie but from a Restatement comment adopted by the Supreme Court of Texas in 1998. See Uniroyal, 977 S.W.2d at 337 (citing RESTATEMENT (Third) of Torts: Products Liability § 2 cmt. f (1998)). Ever since, overall safety has been a recognizable feature of Texas law. And all the while, the PJC utilized here has remained the same, citing to new caselaw but retaining its core definition. Thus, Textron’s argument asks us, in effect, to hold that the PJC has been wrong—so wrong that it is not substantially correct—for the last two decades.Textron’s request faces a couple obstacles at the outset: (1) Texas courts routinely utilize the PJC’s definition of safer alternative design[2] and (2) we have not once held that reliance on a Texas PJC was an abuse of discretion. See Bagby Elevator Co. v. Schindler Elevator Corp., 609 F.3d 768, 772-73 (5th Cir. 2010) (“[W]e find it persuasive that Texas courts have repeatedly approved the Texas Pattern Jury Charges as a correct statement of the law.”).Nor does Genie itself upend the propriety of the PJC’s core definition. At no point did the Genie Court purport to alter the existing parameters of safer alternative designs. Instead, Genie‘s substantive, introductory legal discussion reaffirms the long-understood statutory definition: “A safer alternative design is one that would have prevented or significantly reduced the risk of the injury, would not substantially impair the product’s utility, and was economically and technologically feasible at the time.” 462 S.W.3d at 7.[3]Of course, the Court followed up its definition with additional clarification about what really constitutes a safer alternative design, including the overall-safety component from Uniroyal and the rule that the “design need not be actually built and tested.” Id. But therein lies the rub: Federal judges often face the workaday dilemma of how much state law to consolidate expressly into the jury charge. Must this one include overall safety? How about the caveat about conceptual designs versus those actually built? The list of conceivable additions goes on. But, as our prior cases indicate, a commonly administered PJC is often an entirely sensible place to draw the line. Bagby Elevator, 609 F.3d at 772-73.More fundamentally, the extant definition of a safer alternative design— specifically, its requirement that such designs cannot substantially impair the product’s utility—gave Textron the practical means to raise its concerns about the Nesters’ proposed designs. Take, for instance, the Nesters’ primary fix, the link-removal design. Textron’s rebuttal to that theory was, in the words of one of its witnesses, that removing the link between the parking brake and accelerator was a “bad idea . . . [b]ecause people are going to drive with the brakes engaged on the car and they’re going to destroy their brakes.” The semantic consequences therefrom—a vehicle with useless brakes versus a vehicle with useless, unsafe brakes—are one in the same. And Textron’s closing argument confirms just that. There, Textron’s counsel explained that removing the link “will create more injuries than having it on there” and, on the flip side of the same coin, that the link helps “increase[] braking effectiveness.” The district court’s instruction therefore gave counsel “the opportunity to emphasize the matters in his favor contained in [the] proposed instructions during jury argument”—a final consideration that weighs in favor of a substantial-coverage finding. Alexander, 731 F.2d at 1227.At the end of the day, Textron asks us to hold that the district court erred by refusing to deviate from a standard Texas instruction. That definitionpermitted Textron to make its arguments about various tradeoffs to the jury (it did so) and gave those jurors a means to find in Textron’s favor (they balked). Those circumstances do not instill in us substantial and ineradicable doubt about whether the jury was properly guided, and we do not find an abuse of discretion. Janvey, 856 F.3d at 388.B. The Commingling IssueTextron does not dispute that the Nesters needed to prove only one safer alternative design to recover. Nor does Textron contest that two of the Nesters’ four designs (the link-removal and pedal-guard designs) had sufficient evidentiary support. The dilemma, rather, is about how the jury question was formulated. Because the district court commingled the Nesters’ four alternative designs into one broad-form question, Textron suggests that we cannot discern which of the designs served as the basis for the jury’s “yes” verdict. Such a grouping was harmful, Textron concludes, because the other two designs (the weight-sensitive switch and hand-brake designs) lacked evidentiary support. The Nesters respond (in part) that, in this circuit, such a presumed-harm reversal is proper only when the charge commingles legally invalid theories, not those without evidentiary support.Assuming for argument’s sake that the Nesters offered insufficient proof on at least one of their alternative designs, Textron can prevail only if its understanding of (what we will label) the “commingling rule” is correct—i.e., that the rule encompasses theories that lack evidentiary support. In this circuit, the rule is not so broad. A bit of history explains why.Over a century ago, the Supreme Court first applied the commingling rule to legally invalid theories, holding that a multi-theory verdict “prevents us from perceiving upon which plea [the jury] found” and “therefore, [if] upon any one issue error was committed, either in the admission of evidence or in the charge of the court, the verdict cannot be upheld.” Maryland v. Baldwin, 112 U.S. 490, 493 (1884). A few years later, the Supreme Court applied the rule to theories without evidentiary support, explaining that when there are “counts of a declaration which have not been supported by any evidence, and where it is impossible from the record to say upon which of the counts of the declaration the verdict was based,” the judgment cannot be sustained. Wilmington Star Mining Co. v. Fulton, 205 U.S. 60, 79 (1907). In the years that followed, this circuit applied the rule in a similar fashion. See, e.g., Lyle v. Bentley, 406 F.2d 325, 327-28 (5th Cir. 1969) (If “the court’s instructions permit a verdict to be based on an issue not supported by sufficient evidence, the jury verdict must be set aside.”).But something happened in 1991 that caused this circuit to reconsider the rule’s breadth. The Supreme Court handed down Griffin v. United States, 502 U.S. 46, 59-60 (1991), in which the Court clarified in the criminal context that the commingling rule would apply only to legally invalid theories. This was so, Justice Scalia said, because when “jurors have been left the option of relying upon a legally inadequate theory, there is no reason to think that their own intelligence and expertise will save them from that error” but that “[q]uite the opposite is true . . . when they have been left the option of relying upon a factually inadequate theory, since jurors are well equipped to analyze the evidence.” Id. at 59.In short order, a panel of this court applied Griffin to a civil case. See Walther v. Lone Star Gas Co., 952 F.2d 119, 126 (5th Cir. 1992). Walther acknowledged that Griffin arose in the criminal context but applied it nonetheless: “[W]e will not reverse a verdict simply because the jury might have decided on a ground that was supported by insufficient evidence.” Id. Since then, this circuit has followed in Walther‘s footsteps and applied the Griffin rule. See, e.g., Advocare Int’l L.P. v. Horizon Labs., Inc., 524 F.3d 679, 696 n.67 (5th Cir. 2008); Wellogix, Inc. v. Accenture, L.L.P., 716 F.3d 867, 878 n.4 (5th Cir. 2013).Textron argues, however, that Walther violates our rule of orderliness because it is incongruous with cases that came before it, see, e.g., Lyle, 406 F.2d at 327, and those that came after it.[4] True enough, Walther would ordinarily be without authority to contradict past cases (like Lyle), for “a panel may not overturn a controlling precedent absent an intervening change in law, such as by . . . the Supreme Court.” Vaughan v. Anderson Reg’l Med. Ctr., 849 F.3d 588, 591 (5th Cir. 2017) (internal quotations omitted). But that is precisely what Walther relied upon: the Supreme Court’s opinion in Griffin. See Walther, 952 F.2d at 126. Such reliance on intervening high-court precedent is reason enough to create a rift in our orderliness framework. Moreover, Textron’s suggestion that Griffin did not actually change the commingling rule for civil cases is nothing more than an indictment of Walther—and this panel may not entertain a challenge to a prior panel opinion. McClain v. Lufkin Indus., Inc., 649 F.3d 374, 385 (5th Cir. 2011).Indeed, we need not evaluate Walther s authority on a blank slate; a panel of this court has already answered the question:This Court recently held, reversing a number of our earlier cases, that a jury verdict may be sustained even though not all the theories on which it was submitted had sufficient evidentiary support. Walther purported to be applying the Supreme Court’s decision in Griffin v. United States, which the Fifth Circuit believed overruled cases such as Neubauer v. City of McAllen. Although its reasoning may be questioned, Walther appears authoritative in our Court at this time.Prestenbach v. Rains, 4 F.3d 358, 361 n.2 (5th Cir. 1993) (citations omitted). Prestenbach tells us all we need to know.In light of Walther, the district court’s grouping does not require reversal. No one suggests that any one of the Nesters’ alternative-design theories was legally invalid. And Textron concedes by its silence that sufficient evidence supports at least two of the Nesters’ designs. We therefore trust the jury to have sorted the factually supported from the unsupported.C. Admissibility IssuesTextron challenges next the admissibility of two pieces of evidence. “A trial judge’s ruling on the admissibility of evidence is generally reviewed for an abuse of discretion.” H & W Indus., Inc. v. Occidental Chem. Corp., 911 F.2d 1118, 1121 (5th Cir. 1990).1. The Cowboys Stadium VideoTwelve days after Gini’s injury, another unintended-acceleration event unfolded during a high-school football game at Dallas Cowboys Stadium. There, an unoccupied E-Z-GO cart (with the same pedal design as the Workhorse) collided with a group of people after two end-zone pylons depressed the cart’s accelerator pedal. Textron produced a Youtube.com video of the incident during discovery. At trial, the Nesters sought to authenticate that video through Jim Fisher, a Textron corporate representative who watched video of the incident and later investigated and spoke with certain non- eyewitness officials from Cowboys Stadium. The district court admitted the video, and Textron now raises authentication and relevance challenges.We begin with authentication. “This Court does not require conclusive proof of authenticity before allowing the admission of disputed evidence.” United States v. Jimenez Lopez, 873 F.2d 769, 772 (5th Cir. 1989). Rather, Rule 901(a)[5] “merely requires some evidence which is sufficient to support a finding that the evidence in question is what its proponent claims it to be.” Id.The Nesters presented no eyewitness authentication testimony. Yet, “[a]lthough authentication by eye witnesses is certainly preferable, we do not conclude under the circumstances of this case that the [video] lacked proper authentication.” United States v. Clayton, 643 F.2d 1071, 1074 (5th Cir. Unit B 1981). First is the matter of Fisher’s voir dire testimony; he testified about the content of the video and about what he learned from his investigation—the two sets of facts matched. Textron takes issue with the district court’s characterization that Fisher “relied” on the video in coming to his conclusions. But the court’s phrasing is unimportant. That Fisher reached a conclusion about what happened and it corresponds with what unfolds in the video is some circumstantial proof that the video is what it purports to be. FED. R. EVID. 901(a). Second, Textron’s failure to offer any rebuttal to authenticity bolsters admission. See McConathy v. Dr. Pepper/Seven Up Corp., 131 F.3d 558, 562 (5th Cir. 1998) (per curiam) (identifying opponent’s failure to “claim that the document is not authentic” as one of several bases for authentication); Buziashvili v. Inman, 106 F.3d 709, 717 (6th Cir. 1997) (same). And finally, Textron’s own production of the video during discovery further corroborates its authenticity. See McConathy, 131 F.3d at 562 (affirming authentication based in part on the fact that the document’s opponent produced it during discovery).[6]Although the district court did not rely on the final of these three reasons, they collectively support the district court’s decision. Josyln Mfg. v. Koopers Co., 40 F.3d 750, 753 (5th Cir. 1994) (“[W]e may affirm for reasons other than those relied upon by the district court.”).As for relevance, Textron argues first that the Cowboys incident is dissimilar to the one at hand. And second, Textron argues the video serves no relevant purpose.When offered for a purpose “other than to show notice,” evidence of other incidents must be “closely similar to the facts and circumstances at issue.” Johnson v. Ford Motor Co., 988 F.2d 573, 579 (5th Cir. 1993) (per curiam) (internal quotations omitted). In hewing to that directive here, the district court limited admissible events to those where “(1) a kick-off operational [brake] system was involved; (2) an unsecured object fell or was placed on the accelerator pedal; (3) the driver had moved away from and was not in the vehicle or cart; and (4) there was powered forward movement of the vehicle or cart.” Such a limitation was both reasonable and narrowly tailored, and there is no serious argument that the Cowboys Stadium incident did not qualify thereunder. The most Textron can say is that the video itself “does not show clearly how the pylons became wedged against the accelerator.” Even if that were true, Fisher learned through his investigation (and then conveyed to the jury) that someone threw the pylons into the cart. How the incident came about—and its corresponding similarity to this case—was no great secret.As for the video’s purpose, Textron points out that the later-occurring incident could not be used to show notice of a defect. We agree. But the video was nevertheless probative of the existence of a design defect—that is, whether the Workhorse’s pedal design was unreasonably dangerous given the magnitude of the risk. See Nissan Motor Co. v. Armstrong, 145 S.W.3d 131, 138-39 (Tex. 2004) (similar accidents may be “relevant to show whether . . . a product is unreasonably dangerous”). Yes, everyone at trial agreed that unintended acceleration could happen. But Textron in no way conceded that the risk was an unreasonable one, arguing instead that the design was a “good, safe system” only dangerous to those who took “unreasonable and negligent actions,” thus making the possibility of “unintended acceleration very, very low.” A video depicting just how two misplaced pylons could lead to a runaway vehicle allowed the jury to evaluate that risk for themselves. Accordingly, we find no abuse of discretion.[7]2. The “Best Protection” LetterSix years before Gini’s injury, children at a United Kingdom resort entered a Textron vehicle (with the same pedal design as the Workhorse) and depressed the accelerator, disengaging the parking brake and causing the vehicle to roll downhill. An investigator wrote to Textron with concerns about the vehicle’s design. And, in response, Textron’s vice president of engineering authorized a reply that suggested a solution: removing the link between the accelerator and parking brake. In fact, the letter characterized the link-removal solution—the same fix proposed by the Nesters—as both “feasible” and the “best protection” for users. The district court admitted the letter.Here, Textron raises only a relevance challenge, and its substantial-similarity and relevant-purpose arguments tend to converge. In short, Textron argues the UK incident was not similar because it involved a gravity-induced occurrence as opposed to a forward-powered one. This distinction, according to Textron, means that the letter’s recommendations were likewise limited to preventing roll-away accidents and bore no relevance to this case.Two problems. First, the district court already agreed with Textron’s argument insofar as it excluded evidence of the underlying UK incident and redacted the letter accordingly. And second, the fact that the letter addressed a slightly different risk caused by the same alleged defect does nothing to diminish its most probative function: The letter contradicted Textron’s argument that the link-removal design would cause brake wear and tear. Put simply, the jury was free to wonder why Textron’s vice president of engineering would recommend a design as “feasible” and the “best protection” if, in fact, that design would produce such catastrophic results. The district court did not abuse its discretion in admitting the redacted letter.D. BifurcationBefore trial, Textron asked the district court to bifurcate punitive-damages issues from the rest of the case so that net-worth evidence would not come before the jury. The district court denied bifurcation, and Textron’s final appellate point challenges that decision.“For convenience, to avoid prejudice, or to expedite and economize, the court may order a separate trial of one or more separate issues . . . .” FED. R. CIV. P. 42(b). “A motion to bifurcate is a matter within the sole discretion of the trial court, and we will not reverse the court’s decision absent an abuse of that discretion.” First Tex. Sav. Ass’n v. Reliance Ins. Co., 950 F.2d 1171, 1174 n.2 (5th Cir. 1992).Textron’s argument on appeal is straightforward: Because Texas requires its own trial courts to bifurcate liability from the amount of punitive damages, and because federal courts should look to state law in determining whether to bifurcate, the district court erred by not doing so here.Textron’s argument is halfway correct. Texas state courts are statutorily required to bifurcate liability from punitive damages. TEX. CIV. PRAC. & REM. CODE § 41.009(a). But, in our federal system, bifurcation is a case-specific procedural matter within the sole discretion of the district court. Reliance, 950 F.2d at 1174 n.2. As this court and another have held before, a district court is simply not bound by state law when deciding whether to bifurcate. See Rosales v. Honda Motor Co., 726 F.2d 259, 260 (5th Cir. 1984); Getty Petroleum Corp. v. Island Transp. Corp., 862 F.2d 10, 15 (2d Cir. 1988) (upholding a district court’s refusal to bifurcate liability from punitive damages despite New York law requiring such bifurcation).[8]As for case specifics, Textron points to several instances in the record where the Nesters referred to Textron’s size or net worth. But when a punitive-damages case is not bifurcated, those references are precisely what one would expect—Texas law lists the defendant’s net worth as evidence the jury “shall consider” in determining the amount of punitive damages. TEX. CIV. Prac. & REM. CODE § 41.011(a)(6). To hold (as Textron seems to suggest) that the mere presence of such evidence necessitates bifurcation would be to upend our federal discretionary framework and replace it with a judicial analogue to Texas’s statutory mandate. That we will not do. Rather, on the circumstances of this case, we cannot see the district court’s balancing of the Rule 42(b) considerations as an abuse of discretion. We therefore reject Textron’s fourth and final appellate issue.AFFIRMED.
