Barry William MUTH, Sr., Individually and as Next Friend of Barry William Muth, Jr., Ashley Diana Muth, and Evan Paul Muth, minors; Dee Anne Muth, Individually and as Next Friend of Barry William Muth, Jr., Ashley Diana Muth, and Evan Paul Muth, minors, Plaintiffs-Appellees, v. FORD MOTOR COMPANY; et al., Defendants, Ford Motor Company, Defendant-Appellant, 461 F.3d 557

PATRICK E. HIGGINBOTHAM, Circuit Judge: A jury returned a nearly $9 million judgment against Ford Motor Company for injuries sustained by Barry William Muth, Sr. while traveling in a 1996, four-door Ford Crown Victoria. Ford appeals, challenging the sufficiency of the evidence, evi-dentiary rulings, and the conduct of the trial judge. We affirm. I After finishing a pick-up basketball game, plaintiff Barry W. Muth, Sr. and Julius Wineglass, both Majors in the United States Army, got in a 1996, four-door Ford Crown Victoria and headed back to Escon village, site of a U.S. Army base, in Riyadh, Saudi Arabia.

Wineglass was driving with Muth in the front passenger seat, both men wearing seatbelts. Traveling along a four-lane highway, they approached a right-hand curve going approximately ten miles per hour over the speed limit. Loose in the turn, Wineglass lost control of the car and ran it into a three-foot high “Jersey barrier” separating the two sides of the highway.

Although the precise movement of the car. was disputed, generally the left front wheel climbed the side of the barrier, causing the car to slide along the barrier for a short distance and, ultimately, to flip, landing on its roof and coming to rest about 209 feet from where it initially hit the barrier. Muth sustained a subluxation injury of the C5-C6 vertebrae in his spinal cord, leaving him a quadriplegic with only limited use of his arms and hands. Wineglass received minor injuries and is not party to this litigation.

Muth and his family sued Ford in federal district court, bringing negligence and strict product liability claims. Muth alleged two design defects: first, that the 1996 Ford Crown Victoria contained “inadequate rollover/roof crush protection”; and second, that the 1996 Ford Crown Victoria contained an “inadequate occupant restraint system.” During the seven day trial, Muth focused on the roof strength defect, contending that a stronger and economically practical roof would have prevented the injury.

Keith Friedman, Muth’s expert witness, testified that the roof was defective because it collapsed twelve to fifteen inches on the passenger side. Friedman testified that increasing the thickness of the steel in several parts of the roof structure could have reduced the “roof collapse” to three inches for $9 per car or two inches for $31 per car. Ford did not dispute that a stronger roof would be feasible. Rather, Ford contended that a stronger roof would do little, if anything, to prevent injuries in rollover accidents.

According to Ford, during a rollover accident, the body drops toward the ground — in other words, toward the roof. Because a normal seatbelt system allows the body to drop five inches, which is more than the normal three-to-four inches of clearance between head and roof, the only way to prevent injuries in rollover accidents is to use a five-point, NASCAR-style seatbelt with crotch strap, an impossibility in commercial vehicles. In short, Ford contended that a stronger roof would not help prevent head-and-neck injuries in rollover accidents.

Attempting to prove this counterintui-tive point, Ford relied on data from two crash tests: an early 1980s series from General Motors using Chevy Malibu sedans (“the Malibu test”); and a 2000-2001 series from Ford using the Controlled Rollover Impact System (“the CRIS test”). Both tests used slow-motion video and high-speed cameras to record the precise movements of cars and dummies during rollovers. Although the district court allowed Ford’s expert witnesses to discuss the data and conclusions drawn from the tests, the court excluded the demonstrative evidence — video and photograph — illustrating those results.

At the close of all the evidence, Muth withdrew his negligence claim and only submitted his design defect claim to the jury. The jury answered “yes” to the question of whether there was “a design defect in the Crown Victoria at the time it left the possession of Ford Motor Company that was a producing cause of the injury” and awarded Muth and his family nearly $9 million in damages. Ford timely appealed. We have jurisdiction under 28 U.S.C. § 1291. II Ford raises four issues.

First, Ford, contending that Muth failed to meet his burden on either theory of design defect, argues that the district court erred when it denied Ford’s motion for judgment as a matter of law. Second, Ford contends that even if there was sufficient evidence, the district court erred when it did not ask the jury to unanimously agree on one particular design defect. Third, Ford objects to the exclusion of demonstrative evidence from the Malibu and CRIS tests. Finally, Ford contends that remarks of the trial judge in front of the jury were improper, warranting reversal. We address each in turn.

A Ford moved for judgment as a matter of law at the close of all the evidence, so our review is de novo. We apply the same legal standard as the district court — that is, judgment as a matter of law will only be granted if “the facts and inferences point ‘so strongly and overwhelmingly in the movant’s favor that reasonable jurors could not reach a contrary conclusion.’ ” Ford objects to the sufficiency of the evidence on both theories of design defect.

As to the inadequate roof strength, Ford contends that Muth failed in his burden to establish that the vehicle was in substantially the same condition at the time of the accident as it was at the time of manufacture, pointing to evidence suggesting that the windshield had been replaced prior to the accident. As to the inadequate restraint system, Ford contends that Muth failed to establish any safer alternative designs, a requirement imposed by law.

We disagree with the former, but agree with the latter. 1 Although cast as an insufficient evidence charge, Ford’s objection to the jury’s finding of defective roof strength is really that Muth did not put on sufficient evidence of an essential element in his prima fade case — namely, that he failed to establish that the 1996 Crown Victoria was in substantially the same condition at the time of the accident as it was at the time of manufacture.

Muth questions Ford’s reading of Texas law, citing to cases placing the burden on the defendant to show that the injury was caused by a substantial alteration in the product. Ford cites two cases for the proposition that Texas law requires the plaintiff to prove that the product was in substantially the same condition at the time of accident as at the time of manufacture: Uniroyal Goodrich Tire Co. v. Martinez, from the Supreme Court of Texas, which only states that Texas follows

Section 402A of the Restatement (Second) of Torts; and Syrie v. Knoll International, from our Court, which reads

Section 402A to' put the burden on the plaintiff to show that “the product reached the consumer without substantial change in its condition from the time of original sale.” But the Restatement says nothing about the burden of proof, and subsequent cases, some cited by Muth, suggest that the burden of showing a substantial alteration in the product lies with the defendant. Both arguments miss the mark. Both arguments focus on the windshield, but we need not resolve who has the burden of proof to show that the vehicle involved in the accident had the same (or a similar) windshield as the vehicle off the assembly line if the windshield played no

part in Muth’s theory of design defect. We do not read Ford’s argument to suggest that Muth has the burden to show that the vehicle had not changed in any respect from the time it left Ford’s manufacturing plant. All products, especially complex products like cars, change between the time of purchase and the time of accident, but not every change would obviate a manufacturer’s liability. Muth could have replaced the tires on the car, which may mean it was not in substantially the same condition as at the time of manufacture, but the replacement tires are only relevant if they were a cause of the accident.

Put another way, regardless of who carries the burden of proof on a substantial alternation, the supposed alteration must be relevant to the theory of defect. All of this is subsumed by the basic elements of a design defect products liability claim: design defect and producing cause. Inherent in the notion of a “design defect” is a “defect” in the “design” of the product — that is, a defect existing when the product was manufactured. The second element, causation, ties the past to the present, linking the specific defect, existing at the time of manufacture, to the particular injury.

To this end, Syñe is at least partially right, as Muth must show that the alleged defect, existing at the time of manufacture, was in substantially the same condition at the time of the accident. And this is done by showing that the “design defect” was the “producing cause” of the accident. So framed, our question is whether the windshield was a component of Muth’s alleged design defect. If it was, as Ford contends, then, yes, Muth had the burden to show that, the windshield was in substantially the same condition at the time of accident as at the time of manufacture.

Only then would there be a “design defect” that was a “producing cause” of Muth’s injuries. If the windshield was not part of the theory, then it is only relevant if Ford can show that the replacement windshield actually caused Muth’s injuries, rather than the allegedly defectively designed roof structure. Ford argues that the windshield had been replaced, that the original windshield contributed to the roof strength and that there was no evidence it was replaced with a Ford-manufactured product, that the replacement conformed to Ford’s design specifications, or that the replacement was installed correctly.

All of that may be true, but it is beside the point. The windshield’s contribution, if any, to the roof strength was not part of Muth’s theory of design defect, as explained by Keith Friedman, Muth’s expert witness. His analysis did not turn on the role of the windshield. He stated plainly “that the roof structure was defectively designed” because “it had a very weak roof rail and A pillar, B pillar system, header system,” and when testifying as to safer alternative designs, he never mentioned any changes to the windshield.

Ford’s treatment of the case confirms that the windshield was not a relevant aspect of Muth’s design defect case. Ford did not cross-examine Friedman on any contribution of the windshield to the overall strength of the roof, and Kenneth Orlowski, Ford’s expert witness, testified that during testing, “the roof peak strength relies more on the ... the metal structure and the B pillar behind the A pillar.” Ford cannot reinvent Muth’s theory of design defect on appeal and then contend that Muth’s evidence was insufficient.

If Ford was proceeding on a substantial alteration theory — that is, the replaced windshield was the real cause of Muth’s injuries — -then it likely failed in its proof. Ford’s evidence that the windshield was replaced is scant at best. Orlowski testified that it was his “understanding” that the windshield had been replaced, no more. It is doubtful that this is sufficient to warrant an instruction on substantial alteration, but the answer to that inquiry is of no moment here.

Procedurally and even more probative, Ford did not raise substantial alteration as an affirmative defense, nor did it request an instruction on the issue during the charge conference. The bottom line is that factual issues relating to the windshield and its contribution, if any, to the roof strength and Muth’s injuries, were not part of this case, and Muth presented sufficient evidence that the design defect, properly construed, was a producing cause of his injuries.

The district court did not err when it denied Ford’s motion for judgment as a matter of law on the roof strength defect. 2 As to the inadequate restraint system, we agree with Ford that Muth did not present sufficient evidence of a safer alternative design, a necessary element of Texas law on which Muth had the burden. Muth does not contest this point, stating in his brief that the inadequate roof strength design defect was the “entire focus” of his case. The district court judge erred when it denied Ford’s motion for judgment as a matter of law on the restraint system defect.

B The question becomes, however, whether that error makes any difference in this case. Ford contends that it does, citing the general rule, first recognized by the Supreme Court in Maryland v.

Baldwin, that “when a case is submitted to the jury on a general verdict, the failure of evidence or a legal mistake under one theory of the case generally requires reversal for a new trial because the reviewing court cannot determine whether the jury based its verdict on a sound or unsound theory.” That occurred here, at least according to Ford, because Muth proceeded on two design defect theories, both supposedly submitted to the jury, one illegitimate. We agree that if both theories are put to the jury, a new trial is generally necessary when the evidence is insufficient on one.

But this Court, as well as many others, have engrafted a sort-of harmless error gloss onto the basic principle. Braun v. Flynt is our case. There, we upheld a general verdict for invasion of privacy, despite an instruction authorizing recovery on either a “false light” theory, that was supported by the evidence, or on a “appropriation” theory, that was not.

We concluded that the “entire focus” of the plaintiffs case was her claim that the publication of an indecent photograph of her in Chic Magazine, a Larry Flynt production, “created a false impression of her and damaged her reputation.” On review, we were both “totally satisfied” and “reasonably certain” that the verdict in the plaintiffs favor was not based on the erroneously submitted appropriation theory. Braun applies here. True, as Ford points out, Muth’s fourth amended complaint alleged two design defects: inadequate roof strength and an inadequate restraint system.

That being said, the inadequate restraint system played little role during the trial. Muth made no mention of that theory during voir dire or during opening arguments. During closing argument as Muth’s counsel was summarizing the case, he referenced only the inadequate roof strength defect. He stated, “Ladies and gentlemen of the jury, the answer to Question 1, Was this vehicle defectively designed in not having a stronger roof? Yes.” Muth’s counsel never mentioned the allegedly defective restraint system. Furthermore, evidence of a defect in the Crown Victoria’s restraint system was minimal.

Keith Friedman, Muth’s expert witness, was asked whether other changes could be made to the Crown Victoria to improve the occupant protection system. Friedman then discussed changes to the seatbelt system, recognizing how the seat-belt can work in tandem with the roof to prevent injuries in rollover accidents. Yet, Friedman testified that he did not take into account changes in the seatbelt system when evaluating the design modifications to the roof structure, and Ford did not question his comments about the restraint system.

Ford also points to statements by Muth’s counsel during an in camera conference on the jury instructions, in which Muth’s counsel argued that the restraint system was still part of the case. We find these statements, made outside the presence of the jury, insufficient to bring the restraint system back into the case. Importantly, here, the jury instructions did not identify the two different design defect theories. From the jury’s perspective, they had no reason to think the restraint system was at issue, aside from the minimal comments of Friedman.

The allegedly defective restraint system was not mentioned during voir dire, opening argument, or closing argument. This case illustrates how an evidentiary deficiency can work in tandem with the rule concerning general verdicts in multi-claim lawsuits. Without a doubt, a party can present evidence sufficient to invoke Baldwin’s rule, which requires reversal, but insufficient to sustain even a favorable jury verdict.

Muth presented sufficient evidence of a design defect in the roof strength, and we are “totally satisfied” or “reasonably certain” that the jury decided in Muth’s favor on that defect and that defect alone. C Ford next objects to the district court’s exclusion of demonstrative evidence — video and photograph — from the Malibu test, a rollover crash test conducted by General Motors in the early 1980s using a Chevrolet Malibu, and the CRIS test, a rollover crash test conducted by Ford in 2000-2001 using a 1998-2000 model Crown Victoria.

The Malibu test was one of the first attempts to determine the relationship between roof deformation and injury. Improving on the Malibu test, the CRIS test was conducted after the National Highway Traffic Safety Administration sought comments on Federal Motor Vehicle Safety Standard 216, which set requirements on the amount of weight the roof structure in passenger cars must withstand. The CRIS test controlled the position, momentum, and point of impact of the vehicle’s first contact with the ground.

According to Ford, both tests illustrate how a stronger roof would do little, if anything, to prevent injuries in rollover accidents. Ford offered the visual evidence from the tests to assist the jury in understanding their expert’s testimony regarding the general dynamics of rollover accidents. Muth objected, pointing to several differences between the conditions involved in the tests and the conditions, at least as Muth saw them, involved in the accident.

The court excluded the demonstrative evidence, noting that the tests were not conducted “under substantially the same conditions as those that [were] involved in this particular litigation.” We review the exclusion of demonstrative evidence for an abuse of discretion. No one seriously contests that the video and photographs help the jury understand the general dynamics involved in rollover accidents. The evidence illustrates Ford’s claim that during rollover accidents, head-and-neck injuries can occur prior to any roof deformation.

Importantly here, however, Ford’s expert witness testified at length to this conclusion. In other words, the jury heard the evidence; the only question is whether the district court abused its discretion when it forced Ford’s expert witness to testify without his visual aids. When the demonstrative evidence is offered only as an illustration of general scientific principles, not as a reenactment of disputed events, it need not pass the substantial similarity test.

Such demonstrative aids, however, must not be misleading in and of themselves, and one such way that a demonstration might mislead is when, as here, the demonstration resembles the disputed accident. Indeed, it is this resemblance which gives rise to the requirement of substantial similarity. As the First Circuit has explained, “Scientific principles, when demonstrated in a fairly abstract way, are quite unlikely to be confused with the events on trial.

The more troublesome cases, however, are ones like this one where some principles of some kind may be demonstrated but in a fashion that looks very much like a recreation of the event that gave rise to the trial.” The district court rejected Ford’s demonstration as not quite similar enough, yet that same demonstration too closely resembles the disputed accident to effectively present abstract principles without misleading the jury.

One of the central disputes in this case concerned the precise movement of the Crown Victoria as it went from an upright position on top of the Jersey barrier to upside-down on the pavement below. To Muth, as the car rolled to the driver’s side coming off the barrier, the left wheel struck the pavement briefly, causing the back end of the car to bounce back up and the car to come down on the left front wheel, the car then rolling onto the roof from the driver’s side.

Ford’s accident reconstructionist, in contrast, contended that the car was airborne for twelve feet while coming off the barrier, rolling 90 or more degrees around its lengthwise axis at a rate of 202 degrees per second. The car then hit the pavement with its left front tire, wheel assembly, and fender, the forward movement causing it to pivot around and twist around that point, hitting the pavement first on the front passenger’s side of the roof.

Ford characterized the CRIS test as essentially depicting Ford’s theory of the accident, all the while maintaining that it was offered, not as a reenactment, but only to show general scientific principles. The CRIS test shows a car dropped directly onto the roof over the front passenger seat, consistent with Ford’s theory of the accident; Muth contended, however, that the car landed first on its front left side before falling onto the front passenger side.

The vehicle in the CRIS test was spun at a rotational speed of 220 degrees per second, consistent with testimony from Ford’s expert that the vehicle rolled at a rate of 202 degrees per second as it came off the Jersey barrier; again, Muth disputed this point, contending that the car teetered off the concrete barrier, making only one-quarter of a roll.

As we have explained, the similarities between Ford’s theory of the accident and the conditions of the CRIS test heighten the visual evidence’s prejudicial effect, and this is sufficient to justify the district court’s exercise of discretion in limiting Ford’s expert to oral testimony only. D Ford’s final point concerns allegedly improper statements of the trial judge in front of the jury.

We review the entire record, not just individual comments, and even when certain conduct is inappropriate, we will not reverse unless the conduct so permeates the proceedings that it impairs substantial rights and casts doubt on the jury’s verdict. Although Ford cites many examples of supposedly improper conduct by the trial judge, only one deserves discussion. That incident concerns the same visual evidence from the CRIS test.

All in front of the jury, after Orlowski testified at length regarding the conclusions drawn from the two crash tests, Ford’s counsel began questioning Orlowski about the visual evidence supporting his conclusions. After Orlowski confirmed that the scientific conclusions drawn from the crash tests were derived from slow-motion video and high-speed photography, Muth’s counsel objected. The trial judge then stated: Counsel, I’m disturbed by your reference to these photographs. I had excluded these photographs from the consideration of the jury.

I find it inconceivable that you’d make reference to them after that ruling. Members of the jury, I viewed the photographs that he is referring to. And it was obvious to the Court that there was not sufficient similarity between the conditions that were — of the — that were used in making these photographs to the accident which is the basis of this — I mean, the accident that’s involved in this lawsuit. Then, the trial judge concluded the morning session of testimony and excused the jury for lunch. After lunch, Ford’s counsel moved for a mistrial, which the court denied.

We cannot say that the trial judge abused his discretion with this comment. The court reviewed the visual evidence and allowed Ford to make a proffer of it. The court’s ruling was clear, Ford’s counsel asked for certain clarifications, and the trial court made plain that he wanted no mention of the visual evidence. Ford’s counsel started down that line, and the trial judge stopped him, nothing more. We would not say that counsel’s questions crossed the line, but neither can we say that the court’s response did. There was no abuse of discretion. Ill Accordingly, the judgment of the district court is AFFIRMED. .

Coffel v. Stryker Corp., 284 F.3d 625, 630 (5th Cir.2002). . Flowers v. Southern Regional Physician Services, Inc., 247 F.3d 229, 235 (5th Cir.2001) (quoting Omnitech Int’l, Inc. v. Clorox Co., 11 F.3d 1316, 1322 (5th Cir.1994)). . 977 S.W.2d 328, 334-35 (Tex. 1998); see also McKisson v. Sales Affiliates, Inc., 416 S.W.2d 787, 788-89 (Tex. 1967). . 748 F.2d 304, 306 (5th Cir.1984). .

Section 402A provides, in part, “One who sells any product in a defective condition unreasonably dangerous to the user or consumer or to his property is subject to liability for physical harm thereby caused to the ultimate user or consumer, or to his property, if (

a) the seller is engaged in the business of selling such a product, and (

b) it is expected to and does reach the user or consumer without substantial change in the condition in which it is sold.” Restatement (Second) of Torts § 402A (1965). . See Olympic Arms, Inc. v. Green, 176 S.W.3d 567, 587 (Tex.App.—Hous. [1st Dist] 2004, no pet.) ("Substantial alteration of a product is a type of product misuse and an affirmative defense for which the defendant bears the burden of proof.” (citing Placencio v. Allied Indus. Intern., Inc., 724 S.W.2d 20, 22 (Tex.1987) ("In General Motors v.

Hopkins, 548 S.W.2d 344 (Tex.1977), this court treated alteration as a type of misuse and thus as an affirmative defense on which the defendant had the burden of proof.”))); see also Texas Jurisprudence, Products Liability § 109 (3d ed. 2006) ("There is no burden on the plaintiff to show that the product involved in an accident-producing event was not subjected to misuse or alteration after leaving the hands of the product supplier.”). . See Texas Pattern Jury Charges § 71.4B, p. 169 (2003 ed.). To find a "design defect,” the jury must conclude that the product was "unreasonably dangerous as designed.” Turner v.

General Motors Corp., 584 S.W.2d 844, 847-48 & n. 1 (Tex.1979). . See Texas Pattern Jury Charges § 70.5, p. 161 (2003 ed.) ("A product is not in a defective condition, thus not unreasonably dangerous when sold, if the unreasonably dangerous condition is solely caused by a substantial change or alteration of the product after it is sold, and but for which unreasonably dangerous condition the event would not have occurred.”); Woods v. Crane Carrier Co., Inc., 693 S.W.2d 377, 379-80 (Tex.1985) (finding the evidence sufficient to warrant a substantial alternation instruction); Ramirez v.

Volkswagen of Am., 788 S.W.2d 700, 701-02 (Tex. Civ.App.—Corpus Christi, 1990) (same); see also Fed. Pac. Electric Co. v. Woodend, 735 S.W.2d 887, 892-93 (Tex.Civ.App.—Ft.Worth, 1987) (holding that failure to request altered condition instruction waived any error). . See Tex. Civ. Prac. & Rem.Code § 82.005. . See Smith v. Louisville Ladder Co., 237 F.3d 515, 520 (5th Cir.2001) (insufficient evidence to establish safer alternative design under Texas law). . 112 U.S. 490, 493, 5 S.Ct. 278, 28 L.Ed. 822 (1884); see also Wilmington Star Mining Co. v.

Fulton, 205 U.S. 60, 78, 27 S.Ct. 412, 51 L.Ed. 708 (1907); United New York and New Jersey Sandy Hook Pilots Ass’n v. Halecki, 358 U.S. 613, 619, 79 S.Ct. 517, 3 L.Ed.2d 541 (1959). . Pan Eastern Exploration Co. v. Hufo Oils, 855 F.2d 1106, 1123 (5th Cir.1988); Nowell ex rel. Nowell v. Universal Electric Co., 792 F.2d 1310, 1312 (5th Cir.1986). . Question 1 of the verdict form stated, “Was there a design defect in the Crown Victoria at the time it left the possession of Ford Motor Company that was a producing cause of the injury in question?” . See Olney Savings & Loan Ass’n v.

Trinity Banc Savings Ass’n, 885 F.2d 266, 271-73 (5th Cir.1989) (upholding general verdict on fraud when jury charge listed eleven acts of fraud, each supported by sufficient evidence). . 731 F.2d 1205 (5th Cir.1984); see also Collum v. Butler, 421 F.2d 1257, 1260 (7th Cir.1970); Morrissey v. Nat’l Maritime Union of Am., 544 F.2d 19, 26-27 (2d Cir.1976); Mueller v. Hubbard Milling Co., 573 F.2d 1029, 1038-39 (8th Cir.1978); Asbill v. Housing Authority of the Choctaw Nation, 726 F.2d 1499, 1504 (10th Cir.1984). .

For the same reasons, we reject Ford's claim that they are entitled to a unanimous decision by the jury on the particular design defect. They got that here, as only one defect was presented. . Big John, B.V. v. Indian Head Grain Co., 718 F.2d 143, 146 (5th Cir.1983); Jon-T Chemicals, Inc. v. Freeport Chemical Co., 704 F.2d 1412, 1417 (5th Cir.1983). . Shipp v. General Motors Cotp., 750 F.2d 418, 428 (5th Cir.1985) ("GM was not deprived of an opportunity to present evidence; rather it was not allowed to present it in the way it preferred. So viewed and so weighed on the Rule 403 scale, there was no error.”). .

Four Corners Helicopters, Inc. v. Turbomeca, S.A., 979 F.2d 1434, 1442 (10th Cir.1992); 1 K. Broun, McCormick on Evidence § 202 (2006). . Fusco v. General Motors Corp., 11 F.3d 259, 264 n. 5 (1st Cir.1993). . Reese v. Mercury Marine Div. of Brunswick Corp., 793 F.2d 1416, 1423 (5th Cir.1986). . Bufford v. Rowan Cos., 994 F.2d 155, 157 n. 1 (5th Cir.1993). . Transcript, vol. 23, at 113-15.