OPINION

PERLUSS, P. J.—

Sukhsagar Pannu suffered a severe spinal injury, resulting in quadriplegia, when his Land Rover Discovery (Series 1) sport utility vehicle rolled over following a chain of collisions on the 118 Freeway near Simi Valley. Pannu sued Land Rover North America, Inc., Jaguar Land Rover North America, LLC, and Terry York Motor Cars, Ltd., doing business as Land Rover Encino (collectively Land Rover) alleging claims, among others, for strict liability based on defective design.

Following a bench trial, the court entered a judgment for $21,654,000 against Land Rover, finding stability and roof defects in the Discovery had caused Pannu's injury. On appeal Land Rover contends a new trial is warranted because the trial court erred as a matter of law in applying the "consumer expectation" test for product liability, misapplied the alternative "risk-benefit" test and abused its discretion in excluding certain evidence proffered by Land Rover. Land Rover also contends the court's ruling was not supported by substantial evidence because no skid marks were found at the accident site, evidence it asserts necessarily must have been present had the rollover been caused by the alleged stability defect. We affirm.

FACTUAL AND PROCEDURAL BACKGROUND

1. The Accident and Pannu's Injury

On December 14, 2003 Pannu was driving his 1998 Discovery westbound on the 118 Freeway, travelling about 65 miles per hour. Although a light mist had started to fall, the road was dry. Bret Lusis, a teenager driving an Acura Legend about 75 miles per hour, approached Pannu's vehicle from the rear on the driver's side and collided with the Discovery.1 The collision forced the Discovery across the freeway toward the far right lane, where it collided with a Chevrolet Blazer driven by David Beres. Beres was forced off the shoulder and up the embankment adjacent to the freeway. As he steered the Blazer up the embankment, Beres saw the Discovery rolling over several times along the right shoulder of the freeway. The Discovery came to a stop on its roof, which was crushed.

Pannu suffered a bilateral facet dislocation of the C-6 and C-7 vertebrae, resulting in quadriplegia, as well as a fracture of the fifth spinous process and a teardrop fracture of the C-7 vertebra. At the time of the accident Pannu was 47 years old and physically fit; he was an active runner and field hockey player. He had a college degree, was married with three children and owned two 7-Eleven stores and two Subway stores he and his wife personally managed. Pannu also managed two 7-Eleven stores owned by his parents. He worked eight to 10 hours a day, seven days a week.

As a result of his injuries Pannu is paralyzed below his chest, has limited mobility or dexterity of his arms and hands, cannot drive or groom himself and suffers from spasms, fevers, urinary tract infections, pressure sores, incontinence and constant pain. He requires the daily care of an attendant who dresses him, cares for his medical needs and drives him to his stores. Although he visits one store a day, he is no longer involved in the personal management of his family's stores. After the accident, Pannu's older son abandoned his studies at medical school to care for his father and the family stores. Pannu's daughters attend local colleges in order to be close to their father.2

According to a forensic economist who evaluated his lost earning capacity, Pannu's postdisability worklife expectancy is only one to six years due to his deteriorating medical condition and his inability to perform the necessary functions of most jobs. During those years he will be able to maintain about 19 percent of the original open market value of his preaccident earning capacity. In essence, he is incapable of contributing to the value of his businesses, and his earning capacity is limited to a return on his preaccident investment in the family stores.

2. The Alleged Stability Defect

a. Pannu's evidence

California Highway Patrol Officer Diane Nunes investigated the accident. By the time Officer Nunes arrived at the scene, the light mist had turned to rain. Using a roll meter to estimate distances, she measured several points of physical evidence, including a scuff mark on the center divider she attributed to the Acura hitting the divider after colliding with the Discovery; tire marks in the dirt of the embankment she attributed to the Blazer running up the embankment after colliding with the Discovery; a shallow, foot-long scrape on the shoulder adjacent to the far right lane; and a scrape and green paint transfer within the far right lane of the roadway, 20 feet in length and surrounded by broken glass, she attributed to the Discovery sliding on its roof near the end of its roll. She observed no tire marks and reported no other gouges or scrapes in the vicinity.

Based on Officer Nunes's measurements and observations, his inspection of the Discovery and his 2007 visit to the accident site, Ted Kobayashi, Pannu's accident reconstruction expert, opined the Discovery rolled because of friction between the tire and the roadway. Kobayashi asserted the impacts between the Discovery and the Acura and the Discovery and the Blazer were insufficient to cause the Discovery to roll and, in the absence of a tripping mechanism, he concluded the vehicle rolled as a result of a tire slip. Explaining why the roll occurred, he posited that Pannu began a series of five rapid steering maneuvers in an attempt to control his vehicle after it was struck by the Acura. The first maneuvers came in a failed effort to avoid colliding with the Blazer in the far right lane. The collision between the Discovery and the Blazer sent the Blazer to the right toward the embankment, while the Discovery rebounded to the left. Additional steering inputs by Pannu caused the Discovery to yaw. The resulting tire friction caused the left side wheels to lift from the ground, and the vehicle rolled three and a half times before coming to rest on its roof.3

In support of Kobayashi's reconstruction, Pannu presented another expert, Ed Heitzman, who had devised a protocol to test vehicles for stability. Using a comparable production Discovery, Heitzman equipped it with outriggers to prevent rollover and a steering mechanism to replicate steering inputs and drove it through a test course. Pursuant to the test protocol, the Discovery accelerated to a speed of 50 miles per hour, the speed Kobayashi estimated the Discovery was travelling after its collision with the Blazer, and the steering mechanism executed severe consecutive reverse steering inputs to simulate a driver attempting to avoid an object or collision. The test vehicle's wheels lifted when the steering mechanism executed consecutive left-right steering inputs of 165 degrees, comparable with the steering parameters suggested by Kobayashi.4 Heitzman recorded the same results in subsequent tests with the same steering inputs. Testing whether the production Discovery could be modified slightly to improve rollover resistance, Heitzman added spacers to extend the track width by one and a half inches and utilized low-profile tires to lower the center of gravity by 0.44 inches.5 The wheels of the modified Discovery did not lift up under equivalent steering inputs and resisted rollover when subjected to far more drastic steering inputs.

Pannu also called a stability and handling engineering expert, John Marcosky, who opined that, when a vehicle traveling on a smooth roadway rolls over as a result of steering input and not as a result of a tripping mechanism, the vehicle is defective. Under steering duress a vehicle should have sufficient rollover resistance to slide out rather than roll over.

b. Land Rover's reconstruction of the accident

Land Rover's reconstruction expert, Lee Carr, agreed with Pannu's experts the impacts between Pannu's vehicle and the Acura and Blazer were insufficient to cause the Discovery to roll over. According to Carr, however, the Discovery rolled not because of tire friction but because of a tripping mechanism. In other words, Carr concluded the roll was triggered when the right rear tire of the Discovery hit the asphalt curb of the shoulder after colliding with the Blazer.

Reconstructing the vehicle movements that led to the Discovery striking the curb, Carr postulated a series of movements and speeds just as had Kobayashi. Where Kobayashi accepted Officer Nunes's measurements, however, Carr determined she must have made several sizeable errors in locating the vehicle marks she described in her report. In particular, on his second visit to the site in October 2006, nearly three years after the accident, Carr located what he believed to be the scrape in the asphalt of the shoulder described by Nunes, approximately 90 feet west of the point measured by Nunes.6 He also discovered a semicircular gouge in the nearby curb of the shoulder at a distance he placed approximately 100 inches from the scrape, a measurement that correlated to the wheelbase of the Discovery. Based on these findings, Carr concluded the Discovery slid on the wet pavement after colliding with the Blazer and then rolled after striking the curb with its right rear wheel. Using data flowing from what he concluded was the trigger point of the roll, Carr postulated the Discovery struck the curb at such a speed and angle it executed a pirouette before coming down again on the curb (creating a corresponding dent on the vehicle's roof), as well as a portion of the adjacent embankment, and then rolled once more.7

Carr was especially critical of Kobayashi's reconstruction based on his assertion that tire friction rolls always leave skid marks. The lack of evidence of any skid marks indicated to Carr that the Discovery necessarily rolled because of some triggering mechanism, most likely the curb, and did not roll as a result of severe steering inputs by Pannu.

Following his testimony, Carr was impeached by a witness who located and measured the distance between the curb gouge and the scrape described by Carr. According to the impeachment witness, the distance between the two marks was 160 inches, not the 100 inches described by Carr. Carr then retook the stand and admitted he had erred in his testimony and concurred the correct measurement was 160 inches. He did not explain how this error would affect his reconstruction of the accident.

3. The Alleged Roof Defect

a. Pannu's evidence

As a result of the accident, the roof of the Discovery suffered 13 inches of plastic deformation at the A pillar on the driver's side. Elastic deformation, that is the extent of dynamic deformation during the rollover, ranged from 16 to 17 inches of intrusion into the occupant space. To measure the crush resistance of the Discovery's roof, Pannu's expert, Brian Herbst, performed a drop test on a comparable production Discovery. Using factors gleaned from Kobayashi's reconstruction of the crash, the production Discovery was positioned upside down 18 inches above a load plate at a pitch and angle consistent with Kobayashi's estimation of the position of Pannu's vehicle at the moment of roof failure and then dropped. The roof of the production Discovery suffered 14 inches of deformation, substantially comparable to the damage suffered by Pannu's vehicle.

Having established the weak points of the roof structure, Herbst reinforced the roof pillars and roof bows of a second production Discovery with tubular sections of steel and strengthened some of the steel plating on the roof, integrating the additions into the existing support structure of the roof. As Herbst explained, he added approximately 109 pounds of steel tubing, sheet metal and rigid polyurethane foam filling at a cost of $116. The reinforced Discovery was then dropped from the same position as the first Discovery. This time, the roof deformation was limited to three inches at the A pillar, instead of the 16 to 17 inches of deformation suffered by the unreinforced Discovery. Assuming economies of scale and manufacturing, Herbst estimated the true cost of modifying the roof design of the Discovery as approximately $76 and the additional weight to be in the range of 72 pounds.

Pannu then called a medical expert, Joseph L. Burton, to describe his injury and opine as to its cause. According to Dr. Burton, Pannu suffered a bilateral facet dislocation of the C-6 and C-7 vertebrae, as well as a fracture of the fifth spinous process and a teardrop fracture of the C-7 vertebra. Dr. Burton opined that these injuries demonstrated that Pannu's neck was hyperflexed into his chest by the deforming roof (and corresponding loss of occupant space), which caused the spinal injury that paralyzed him.

b. Land Rover's evidence

Land Rover's defense of the alleged roof defect focused primarily on causation.8 Land Rover's causation expert, Elizabeth Raphael, a medical doctor, testified Pannu's injury resulted not from hyperflexion of the neck but from axial loading on the spine resulting from Pannu's head, which was forced to the roof of the Discovery by centrifugal forces during the rollover, striking the ground (through the roof) in the milliseconds before roof deformation began. According to Dr. Raphael, Pannu's injury would have happened regardless of the strength of the roof. She cited studies in which testers had been able to replicate bilateral facet dislocation on cadaver necks only through axial loading caused by force impact and not through hyperflexion of the neck. She also cited as support for her opinion, over Pannu's objection, results obtained by the automobile industry in two crash test studies: an early 1980's series from General Motors using Chevrolet Malibu sedans (the Malibu test); and a 2000 to 2001 series from Ford using the Controlled Rollover Impact System (the CRIS test).9 These two studies concluded catastrophic injury in rollover crashes resulted not from deformation of vehicle roofs but from the initial impact of the head with the ground, not unlike the force experienced by a person in a falling elevator.

4. The Trial Court's Statement of Decision and Judgment

The trial court issued an 11-page proposed statement of decision finding in Pannu's favor. After reviewing the evidence related to the cause of the rollover, the court concluded it was more likely than not the rollover occurred as described by Pannu's expert, Kobayashi. The court based its conclusion on several factors: Kobayashi's acceptance of Officer Nunes's measurements, which had been rejected in whole by Land Rover's expert, Carr; the impeachment of Carr as to the distance between the scrape and gouge he had claimed matched the Discovery's wheelbase; and the court's conclusion Carr's curb-trigger theory was "too speculative." In particular, the court rejected Carr's premise that tire skid marks would necessarily still be visible at the site when he inspected it more than two years after the accident.

Applying the consumer expectation test to these facts, the court concluded Land Rover was liable for both stability and roof defects because the Discovery "did not perform as safely as an ordinary consumer would have expected at the time of the accident." As the court explained, "Consumers reasonably expect that manufacturers will anticipate negligent accidents occurring that can put the automobile into uncontrollable movements. If those dynamics result in causing harm, then there is a design defect and there is strict liability under the [c]onsumer [e]xpectation test." The court rejected Land Rover's attempt to attribute full responsibility for the accident and Pannu's injuries to Lusis, the driver of the Acura, and apportioned 5 percent fault to Lusis and 95 percent to Land Rover.

Applying the alternate risk-benefit test, the court ruled Pannu had carried his burden of proving the roof and stability design of the vehicle was a substantial factor in causing his injury, and Land Rover had failed to establish the benefits of the design outweighed its inherent risks: Options to strengthen the roof were available at reasonable cost, and, although there is no way to make vehicles immune to rollovers or roof crush, "the challenge is to make vehicles have safer rollover resistance and safer roofs."

The court identified the dispute between Drs. Burton and Raphael as to the timing of Pannu's injury as "the most interesting part of defendant's case" and acknowledged this issue has engendered "an ongoing debate among experts." Nonetheless, the court concluded Pannu's injury was most likely caused by the roof crushing inward during the rollover, producing flexion and fracture. The court noted that Land Rover's position necessarily assumed the weakness of the roof (and defective design) and, by focusing on the timing of the injury, challenged only its causation.

The court also found Land Rover strictly liable for a failure to warn of the dangerous propensities of the Discovery. According to the vehicle's window sticker, Pannu was informed the Discovery had a "steel inner body cage" and a "steel roof panel," among other things. Based on earlier testing of the Discovery, Land Rover had sufficient knowledge about the possibility of rollovers and roof crush that it should have warned consumers, including Pannu, of these risks. Instead, Land Rover highlighted these aspects of the vehicle as safety features.

The court assessed damages in the amount of $21,654,000, including $11,654,000 in economic damages and $10 million in noneconomic damages. Although Land Rover submitted objections to the proposed statement of decision, the court issued an order on May 18, 2009 adopting the proposed statement of decision as the final decision and entered judgment in Pannu's favor. Land Rover moved for a new trial, which the court denied on July 10, 2009.

CONTENTIONS

Land Rover contends the trial court erred as a matter of law in applying the consumer expectation test and misapplied the alternate risk-benefit test by failing to undertake the considered analysis required to impose liability. Land Rover also argues Pannu failed to carry his burden of proof in establishing his injury was caused by stability and crashworthiness defects and that the trial court's findings were not supported by substantial evidence. Land Rover further contends it was prejudiced by the improper exclusion of certain expert testimony, including information concerning the Malibu and CRIS rollover tests, and that a new trial should therefore be granted. Finally, Land Rover claims the trial court's award of economic damages was not supported by the record.

DISCUSSION

1. The Trial Court Properly Imposed Strict Liability on Land Rover for Stability and Roof Defects

a. The governing standards

(1) A manufacturer may be held strictly liable for placing a defective product on the market if the plaintiff's injury results from a reasonably foreseeable use of the product. (Soule v. General Motors Corp. (1994) 8 Cal.4th 548, 560 [34 Cal.Rptr.2d 607, 882 P.2d 298] (Soule).) Products liability may be premised upon a theory of design defect, manufacturing defect or failure to warn. (Anderson v. Owens-Corning Fiberglas Corp. (1991) 53 Cal.3d 987, 995 [281 Cal.Rptr. 528, 810 P.2d 549].) A design defect exists when the product is built in accordance with its intended specifications, but the design itself is inherently defective. (Barker v. Lull Engineering Co. (1978) 20 Cal.3d 413, 429 [143 Cal.Rptr. 225, 573 P.2d 443] (Barker).)

(2) As we explained in McCabe v. American Honda Motor Co., Inc. (2002) 100 Cal.App.4th 1111, 1120-1121 [123 Cal.Rptr.2d 303] (McCabe), "[T]he Supreme Court [has] recognized two tests for proving design defect. The `consumer expectation test' permits a plaintiff to prove design defect by demonstrating that `the product failed to perform as safely as an ordinary consumer would expect when used in an intended or reasonably foreseeable manner.' [Citation.] This test, rooted in theories of warranty, recognizes that implicit in a product's presence on the market is a representation that it is fit to do safely the job for which it was intended. [Citations.] If the facts permit an inference that the product at issue is one about which consumers may form minimum safety assumptions in the context of a particular accident, then it is enough for a plaintiff, proceeding under the consumer expectation test, to show the circumstances of the accident and `the objective features of the product which are relevant to an evaluation of its safety' [citation], leaving it to the fact finder to `employ "[its] own sense of whether the product meets ordinary expectations as to its safety under the circumstances presented by the evidence."' [Citations.] Expert testimony as to what consumers ordinarily `expect' is generally improper. [Citation.] [Fn. omitted.]

(3) "The second test for design defect is known as the `risk-benefit test.' Under this test, products that meet ordinary consumer expectations nevertheless may be defective if the design embodies an `"excessive preventable danger."' [Citations.] To prove a defect under this test, a plaintiff need only demonstrate that the design proximately caused the injuries. Once proximate cause is demonstrated, the burden shifts to the defendant to establish that the benefits of the challenged design, when balanced against such factors as the feasibility and cost of alternative designs, outweigh its inherent risk of harm. [Citations.] [¶] (4) The two tests provide alternative means for a plaintiff to prove design defect and do not serve as defenses to one another. A product may be defective under the consumer expectation test even if the benefits of the design outweigh the risks. [Citation.] On the other hand, a product may be defective if it satisfies consumer expectations but contains an excessively preventable danger in that the risks of the design outweigh its benefits. [Citation.] [¶] Whether a plaintiff may proceed under the consumer expectation test or whether design defect must be assessed solely under the risk-benefit test is dependent upon the particular facts in each case."

b. Liability under the consumer expectation test

Land Rover contends the trial court erred in applying the consumer expectation test to the alleged stability and roof defects, arguing the question of defect under the facts of this case is far too complicated to decide based on the perceptions of the ordinary driver. (See Soule, supra, 8 Cal.4th at pp. 562, 567 [because "`"[i]n many situations . . . the consumer would not know what to expect, because he would have no idea how safe the product could be made,"'" the consumer expectation test is "reserved for cases in which the everyday experience of the product's users permits a conclusion that the product's design violated minimum safety assumptions, and is thus defective regardless of expert opinion about the merits of the design"].) In a footnote that has spawned substantial debate rather than quelled it, the Soule court remarked, "the ordinary consumers of modern automobiles may and do expect that such vehicles will be designed so as not to explode while idling at stoplights, experience sudden steering or brake failure as they leave the dealership, or roll over and catch fire in two-mile-per-hour collisions." (Id. at p. 566, fn. 3.)10

(5) While the mishaps described in this footnote certainly fall within the ambit of the consumer expectation test, we doubt it states the outer boundaries of activities that may be found subject to the test. As we observed in McCabe, supra, 100 Cal.App.4th 1111, 1124, "The critical question, in assessing the applicability of the consumer expectation test, is not whether the product, when considered in isolation, is beyond the ordinary knowledge of the consumer, but whether the product, in the context of the facts and circumstances of its failure, is one about which the ordinary consumers can form minimum safety expectations." We concluded the consumer expectation test could properly be applied to the failure of an airbag to deploy in a head-on collision. (Id. at p. 1122; see also Campbell v. General Motors Corp. (1982) 32 Cal.3d 112, 126 [184 Cal.Rptr. 891, 649 P.2d 224] [passenger on bus injured during sharp turn could use consumer expectation test to prove absence of "grab bar" was design defect; public transport is matter of common experience and required no expert testimony]; Saller v. Crown Cork & Seal Co., Inc. (2010) 187 Cal.App.4th 1220, 1236 [115 Cal.Rptr.3d 151] [reversing trial court's refusal to instruct on consumer expectation test in claim involving asbestos exposure]; Sparks v. Owens-Illinois, Inc. (1995) 32 Cal.App.4th 461, 474-475 [38 Cal.Rptr.2d 739] [insulation containing asbestos and causing fatal disease was a product within ordinary experience and understanding of consumer]; Arnold v. Dow Chemical Co. (2001) 91 Cal.App.4th 698, 726 [110 Cal.Rptr.2d 722] [home pesticides causing disability were products within common knowledge of consumer, and, therefore, defect could be assessed under consumer expectation theory].) In the context of rollovers, one court cautioned nearly four decades ago, "Emergency situations requiring severe turning movements arise every day. Cars should not be built just to coincide with normal driving conditions. While the car, driver and road are all interrelated, situations of peril do arise daily requiring heroic turning maneuvers. (6) While strict liability should not be imposed upon the manufacturer (or distributor) when injury results from the use of its product that is not reasonably foreseeable, and while a collision may not be the normal or intended use of a vehicle, vehicle manufacturers must take accidents into consideration as reasonably foreseeable occurrences involving their products." (Culpepper v. Volkswagen of America, Inc. (1973) 33 Cal.App.3d 510, 518 [109 Cal.Rptr. 110] (Culpepper).) The Culpepper court concluded a Volkswagen driven at freeway speeds over a smooth road surface and subjected to a steering input to avoid a perilous situation constituted a factual situation the manufacturer should have foreseen and accounted for in its design of the car. (Ibid.)

The trial court concluded the rollover here was subject to the consumer expectation test on the ground the vehicle was used as intended, and it was reasonably foreseeable that freeway accidents occur and unpredictable forces can cause a vehicle to "act erratically." Although the trial court's observation about the foreseeability of accidents is undoubtedly true, in our view the applicability of the consumer expectation test to the alleged stability defect under the circumstances of this case is an exceedingly close question.11

Moreover, while it might seem easier to conceive of the alleged roof defect falling within the experience of ordinary consumers, the circumstances of this case fall well beyond the examples set forth in Soule. Would a reasonable consumer expect the roof of the Discovery to intrude so dramatically on the occupant survivor space in the event of a rollover following multiple collisions at freeway speed? There are few reported California cases on point; and, even then, much of the direction on this issue is dicta. The Third District once rejected General Motors's contention "the subject of roof collapse is esoteric and only amenable to proof by an analysis of the physical forces acting upon the roof, e.g., by means of expert testimony involving a calculation of the forces acting upon the driver's-side roof and the impact of such forces upon the vehicle in its defective condition." (Doupnik v. General Motors Corp. (1990) 225 Cal.App.3d 849, 869 [275 Cal.Rptr. 715].) As the court responded, "We see no reason why the mode of proof must be so constrained. There is a saying that, according to the science of aeronautical engineering, a bumble bee cannot fly. But if that were an issuable fact, the defects in the theory could be shown by the observations of a beekeeper or an entomologist." (Ibid.; cf. Bell v. Bayerische Motoren Werke Aktiengesellschaft (2010) 181 Cal.App.4th 1108, 1129 [105 Cal.Rptr.3d 485] [stating, in dictum, the potential risk of injury resulting from rollover of sports car was within realm of consumer expectation test]; Gordon v. Nissan Motor Co., Ltd. (2009) 170 Cal.App.4th 1103, 1117 [88 Cal.Rptr.3d 778] [finding prejudicial error in trial court's refusal to give proffered jury instruction on consumer expectation test for alleged roof defect].)

We need not resolve these difficult questions, however. The trial court's alternative finding of strict liability under the risk-benefit test is amply supported by the record and fully justifies the judgment in favor of Pannu.

c. Liability under the risk-benefit test

(7) The risk-benefit test for defective product design requires the fact finder to "`consider, among other relevant factors, the gravity of the danger posed by the challenged design, the likelihood that such danger would occur, the mechanical feasibility of a safer alternative design, the financial cost of an improved design, and the adverse consequences to the product and to the consumer that would result from an alternative design.'" (Merrill v. Navegar, Inc. (2001) 26 Cal.4th 465, 479 [110 Cal.Rptr.2d 370, 28 P.3d 116].) "In such cases, the jury must consider the manufacturer's evidence of competing design considerations [citation], and the issue of design defect cannot fairly be resolved by standardless reference to the `expectations' of an `ordinary consumer.'" (Soule, supra, 8 Cal.4th at p. 567.) Once the plaintiff has made a prima facie showing that his or her injury was caused by the product's defective design, the burden shifts to the defendant to establish that, in light of the relevant factors, the product is not defective. (Barker, supra, 20 Cal.3d at p. 431.)

(8) Land Rover first asserts the trial court committed legal error by failing to consider the necessary factors in applying the risk-benefit test. Although its analysis is not detailed, the trial court's statement of decision adequately addresses the factual basis for the court's findings under this theory of strict liability.12 With respect to stability design, Pannu established that the production Discovery would tip under evasive steering maneuvers and that slight modifications to the track width and center of gravity of the vehicle dramatically improved its rollover resistance. Similarly, modest enhancement of the roof support of the production Discovery yielded substantial gains in roof strength. Pannu proved these improvements could be achieved at a modest cost. Land Rover did not rebut any of these showings. Moreover, Land Rover's senior engineer who testified about the design goals of the Discovery acknowledged these modifications were available and could have been made at the time Pannu's vehicle was manufactured. While he also spoke of the specialized needs of sport utility vehicles for high road clearance and traction, he did not state those goals would preclude implementation of these safety-enhancing modifications.13 This evidence was more than sufficient to establish the Discovery's design presented an "`excessive preventable danger'" and that "the benefits of the . . . design" did not "outweigh the risk of danger inherent in such design." (Barker, supra, 20 Cal.3d at pp. 430, 432; see McCabe, supra, 100 Cal.App.4th at p. 1121.)

Land Rover's principal challenge to the court's finding of strict liability on the risk-benefit theory of design defect is based not on the evidence of a stability defect itself but instead on the predicate issue of causation. Land Rover asserts the finding of a stability defect is fully refuted by the lack of evidence of skid marks, which, in the opinion of Carr, Land Rover's reconstruction expert, demonstrates the rollover must have resulted from a tripping mechanism. Absent evidence of skid marks, Land Rover argues, Pannu failed to carry his threshold burden of causation. The trial court dismissed this assertion as "speculative," noting that Land Rover failed to prove "weather and continual driving over the area . . . could not have obliterated any skid tire marks in the two-year-four-month period" between the accident and Carr's inspection of the accident site. Moreover, Pannu's experts testified newer tires leave lighter skid marks that are more difficult to see and any marks left would have been difficult to detect in the rain following the accident, particularly on the black asphalt of the shoulder. The trial court concluded, notwithstanding the lack of evidence of skid marks, Kobayashi's reconstruction was more likely than Carr's.

Land Rover argues on appeal that the trial court's finding on this point improperly eliminated Pannu's burden of proving causation because, without skid marks, Kobayashi's reconstruction of the accident is pure speculation. In other words, according to Land Rover, the lack of evidence of skid marks precluded the court from finding in Pannu's favor.

Land Rover's argument fundamentally misperceives the nature of Pannu's burden.14 Pannu did not need to prove skid marks were present at the accident site; instead, he bore the burden of establishing in the mind of the trial court "a requisite degree of belief"—meaning more probably than not—his injury was caused by a design defect. (See Barker, supra, 20 Cal.3d at p. 431.) He was entitled to rely on a full array of evidence, including the observations of percipient witnesses and the opinion testimony of expert witnesses. The testimony by Beres and Lusis that Pannu's vehicle rolled on the roadway and not on the curb appears at least as probative of causation in this case as the unresolved question about the presence of skid marks at the time of the accident.15 In any event, as we must, we resolve all conflicts in favor of the prevailing party. (Hasson v. Ford Motor Co. (1982) 32 Cal.3d 388, 398 [185 Cal.Rptr. 654, 650 P.2d 1171]; accord, Campbell v. General Motors Corp., supra, 32 Cal.3d at pp. 117-118.) In short, the trial court's finding Pannu carried his burden of proving his injury was caused by a stability defect is unquestionably supported by substantial evidence.

Land Rover also contests causation with respect to Pannu's injury and the alleged roof defect, asking us in effect to reweigh the evidence that injury occurred because of axial loading rather than hyperflexion. Again, the trial court reviewed the conflict between Pannu's and Land Rover's medical experts and concluded it was more likely than not that Pannu's injury resulted from hyperflexion of the neck caused by deformation of the roof. That finding, too, was supported by substantial evidence; and we may not disturb it.

In sum, substantial evidence supports the trial court's findings of strict liability on Pannu's claims of stability and roof defects.

2. Substantial Evidence Supports the Failure to Warn Finding

(9) "Generally speaking, manufacturers have a duty to warn consumers about the hazards inherent in their products. [Citation.] The requirement's purpose is to inform consumers about a product's hazards and faults of which they are unaware, so that they can refrain from using the product altogether or evade the danger by careful use. [Citation.] Typically, under California law, we hold manufacturers strictly liable for injuries caused by their failure to warn of dangers that were known to the scientific community at the time they manufactured and distributed their product." (Johnson v. American Standard, Inc. (2008) 43 Cal.4th 56, 64 [74 Cal.Rptr.3d 108, 179 P.3d 905]; accord, Brown v. Superior Court (1988) 44 Cal.3d 1049, 1065-1066 [245 Cal.Rptr. 412, 751 P.2d 470]; Anderson v. Owens-Corning Fiberglas Corp., supra, 53 Cal.3d at p. 1000.) To establish strict liability for failure to warn, the plaintiff must prove the defendant "did not adequately warn of a particular risk that was known or knowable in light of the generally recognized and prevailing best scientific and medical knowledge available at the time of the manufacture and distribution. . . . [T]he reasonableness of the defendant's failure to warn is immaterial." (Anderson, at pp. 1002-1003, fn. omitted; accord, Saller v. Crown Cork & Seal Co., Inc., supra, 187 Cal.App.4th at p. 1239.)

Contrary to Land Rover's argument no evidence at trial supported the trial court's finding that Land Rover had failed to warn of the stability and roof defects in the Discovery, Pannu proved not only that the defects existed, but also that Land Rover knew of the dangers of rollover by the time his vehicle was manufactured in 1998 and corrected them in the redesign of the Discovery Series II, which debuted in the United States in 1999. Rather than warn of the dangers of rollover or roof crush, however, the vehicle's window sticker informed Pannu the Discovery had a "steel inner body cage" and a "steel roof panel," among other things, leading him to believe the vehicle was not defectively designed. Pannu testified he read the window sticker, as well as other marketing statements, and understood the Discovery was a solid car that could withstand damage. Nothing more was required to establish Land Rover's liability under this additional theory.

3. The Trial Court Did Not Abuse Its Discretion in Excluding Land Rover's Evidence

Trial court rulings on the admissibility of evidence, whether in limine or during trial, are generally reviewed for abuse of discretion. (People v. Williams (1997) 16 Cal.4th 153, 196-197 [66 Cal.Rptr.2d 123, 940 P.2d 710] ["In determining the admissibility of evidence, the trial court has broad discretion.. . . On appeal, a trial court's decision to admit or not admit evidence, whether made in limine or following a hearing pursuant to Evidence Code section 402, is reviewed only for abuse of discretion."]; accord, People v. Alvarez (1996) 14 Cal.4th 155, 203 [58 Cal.Rptr.2d 385, 926 P.2d 365] ["appellate court reviews any ruling by a trial court as to the admissibility of evidence for abuse of discretion"]; Zhou v. Unisource Worldwide (2007) 157 Cal.App.4th 1471, 1476 [69 Cal.Rptr.3d 273].) "The trial court's error in excluding evidence is grounds for reversing a judgment only if the party appealing demonstrates a `miscarriage of justice'—that is, that a different result would have been probable if the error had not occurred." (Zhou, at p. 1480; see Evid. Code, § 354; Code Civ. Proc., § 475.)

Land Rover challenges three evidentiary rulings it claims were prejudicial: (1) the exclusion of the Malibu/CRIS test results demonstrating severe injuries suffered in rollover accidents are caused by axial loading at the point of impact rather than hyperflexion resulting from roof deformation; (2) the exclusion of federal safety standards, especially Federal Motor Vehicle Safety Standard No. 216 (see 49 C.F.R. § 571.216 (2009)) (FMVSS 216); and (3) the exclusion of the results of drop tests performed by Land Rover expert Jeff Croteau relating to roof strength.

a. The Malibu/CRIS tests

As discussed, the Malibu and CRIS tests were conducted by General Motors and Ford Motor Company, respectively, to demonstrate that catastrophic injury in rollover crashes results not from deformation of vehicle roofs but from the initial impact of the head with the ground. Land Rover sought to introduce this evidence through its causation expert, Dr. Raphael, who intended to use the studies to support her assertion Pannu's injury resulted from axial loading at the moment of impact rather than hyperflexion during roof deformation.16 Pannu successfully moved in limine to exclude evidence of these tests on the ground the vehicles and the dynamics of the test were not substantially similar to the Discovery or the circumstances of the accident at issue in this case.

Land Rover contends this ruling was error because the testing was offered to demonstrate general rollover dynamics rather than to recreate what happened to Pannu's vehicle. To support its argument, Land Rover relies on the decision in Culpepper, supra, 33 Cal.App.3d 510, in which the appellate court affirmed the trial court's ruling allowing evidence of testing performed on a Ford Pinto for the purpose of disputing causation in an accident involving a Volkswagen Beetle. (Id. at p. 522.)

(10) Land Rover's reliance on Culpepper is misplaced. An appellate court's ruling that a trial court did not abuse its discretion in admitting a certain type of evidence is not authority for the proposition that it is an abuse of discretion to exclude similar evidence in another case. Indeed, the admission of testing results relating to a Pinto rollover was not challenged on appeal in Culpepper; the appellate court merely cited the evidence as an example of the leeway given by the trial court to allow Volkswagen to prove its case, even though the trial court had excluded other vehicle testing based on the lack of similarity. (Culpepper, supra, 33 Cal.App.3d at p. 522.) Notably, Culpepper applied the same standard for admission of experimental evidence used by the trial court in this case: "Admissibility of experimental evidence depends upon proof of the following foundational items: (1) The experiment must be relevant (Evid. Code, §§ 210, 351); (2) the experiment must have been conducted under substantially similar conditions as those of the actual occurrence (Andrews v. Barker Brothers Corp.[ (1968)] 267 Cal.App.2d 530, 537 [73 Cal.Rptr. 284]); and (3) the evidence of the experiment will not consume undue time, confuse the issues or mislead the jury (Schauf v. Southern Cal. Edison Co.[ (1966)] 243 Cal.App.2d 450, 455 [52 Cal.Rptr. 518])." (Culpepper, at p. 521, italics added; accord, People v. Boyd (1990) 222 Cal.App.3d 541, 565 [271 Cal.Rptr. 738].) The trial court properly applied this standard; Culpepper thus adds nothing to Land Rover's position.

Alternatively, Land Rover relies on several federal decisions that have affirmed trial court rulings admitting the results of Malibu/CRIS tests "even in cases involving different vehicles, so long as the evidence is not used for accident reconstruction and the finder of fact is appropriately instructed." (See, e.g., Champeau v. Fruehauf Corp. (8th Cir. 1987) 814 F.2d 1271, 1278 [neither test]; Muth v. Ford Motor Co. (5th Cir. 2006) 461 F.3d 557, 566-567 [Malibu and CRIS tests]; Crossley by Crossley v. General Motors Corp. (7th Cir. 1994) 33 F.3d 818, 821-823 [Malibu test]; Harvey v. General Motors Corp. (10th Cir. 1989) 873 F.2d 1343, 1355-1356 [Malibu test].) These decisions do not establish the trial court applied the wrong standard in ruling on the motions in limine; to the contrary, they illustrate the federal courts' persistent concern experimental evidence not conducted under substantially similar conditions be properly limited if it is to be admitted in a jury trial. (See, e.g., Champeau, at p. 1278 [declining to find error in admission of experimental evidence conducted under insufficiently similar conditions when trial court required parties to submit lists of differences and similarities; evidence did not attempt to reconstruct accident and was proffered to demonstrate general principles of physics]; Muth, at pp. 566-567 [although demonstrative evidence offered as illustration of general scientific principles and not as reenactment need not pass substantial similarity test, it must not be misleading; no abuse of discretion in trial court ruling allowing expert to discuss Malibu and CRIS tests but excluding demonstrative evidence of those tests]; Crossley, at p. 822 [no abuse of discretion in admitting portion of Malibu test with proper limiting instruction explicitly informing jury evidence was not intended as reenactment of accident and preventing jury from reviewing tape during deliberations]; Harvey, at p. 1355 [no abuse of discretion in admission of experimental evidence with proper limiting instruction when evidence did not purport to reconstruct accident].)

The unifying theme in these cases is the broad discretion afforded trial courts in deciding whether particular experimental evidence, including the Malibu and CRIS tests, may appropriately be shown to a jury if there is a risk of misleading the jury. That discretion, of course, permits the trial court to exclude, as well as to admit, experimental evidence depending on the circumstances presented. For example, in Shipp v. General Motors Corp. (5th Cir. 1985) 750 F.2d 418, the Fifth Circuit found no abuse of discretion in a trial court's ruling excluding the Malibu test: "After balancing the film's acknowledged relevance against the danger its admission would mislead, confuse, or prejudice the jury, [citation], and determining that a limiting instruction would not defuse such improper influence, the district court excluded GM's film on the ground the jury would likely consider it as more than a simple demonstration of general principles." (Id. at p. 427.)

In a comparable context, the First Circuit similarly affirmed the trial court's exclusion of experimental evidence offered to illustrate "scientific principles." (Fusco v. General Motors Corp. (1st Cir. 1993) 11 F.3d 259, 264.) As the court observed, "General Motors says that the tapes here were designed not to recreate the accident but rather to illustrate general `scientific principles,' presumably the behavior of a car with a disconnected ball stud. Admittedly, `when this [scientific principles] label is attached,' courts often do ask not about similarity of conditions but only whether the test was properly conducted. [Citation.] We think it would be a great stretch to call the tapes an abstract demonstration of scientific principles, but the critical point is not one of labels. The issue for us is whether the demonstration is sufficiently close in appearance to the original accident to create the risk of misunderstanding by the jury, for it is that risk that gives rise to the special requirement to show similar conditions." (Ibid.)

This case required a similar inquiry. At the time the trial court granted the motion in limine excluding evidence concerning the Malibu and CRIS tests, the parties had not yet waived their right to a jury trial. Among other considerations, the court was concerned the jury might be confused by the proffered evidence. In the course of evaluating this issue, the parties submitted extensive briefing that explained the nature and circumstances of both the Malibu and CRIS tests.17 The court's exercise of its discretion to exclude the evidence, after carefully evaluating its probative value and potential for confusion, was not an abuse of that discretion.

Even were we to conclude the court had erred in excluding the test evidence, there is no basis to find Land Rover was unfairly prejudiced by the ruling. There was extensive testimony at trial as to the parties' contrasting views of causation, that is, Dr. Burton's opinion Pannu's injury resulted from hyperflexion of the neck caused by roof deformation, on the one hand, and Dr. Raphael's opinion the injury was caused by axial loading, on the other hand. Indeed, as Pannu points out, the court allowed Dr. Raphael to identify the disputed tests as evidence upon which she relied in reaching her opinion. Although no demonstrative evidence of those tests was admitted, the court made clear in its statement of decision it had closely evaluated the respective theories of causation and had concluded Pannu's evidence was more compelling. There was no prejudicial abuse of discretion.

b. FMVSS 216

Land Rover next challenges the trial court's ruling excluding evidence relating to the Discovery's compliance with FMVSS 216, adopted by the National Highway Traffic Safety Administration to establish minimum safety standards for vehicle roof strength. FMVSS 216, however, applies only to vehicles weighing less than 6,000 pounds; Pannu's vehicle weighed slightly more than 6,000 pounds. Nonetheless, according to Land Rover, even though not directly applicable to the Discovery, FMVSS 216 is widely recognized in the automotive industry as the relevant standard for determining acceptable roof crush for passenger vehicles.

(11) Although the trial court granted Pannu's motion to exclude evidence of compliance with FMVSS 216, it did allow Land Rover to testify about its own testing results using the platen test protocol required by FMVSS 216. This was not an abuse of discretion. California courts have long held compliance with federal motor vehicle safety standards does not preclude liability for defective design. (See Ramirez v. Plough, Inc. (1993) 6 Cal.4th 539, 548 [25 Cal.Rptr.2d 97, 863 P.2d 167], citing Buccery v. General Motors Corp. (1976) 60 Cal.App.3d 533, 540-541 [132 Cal.Rptr. 605].) Land Rover's compliance with the standard, even if arguably relevant, was outweighed by the potential jury confusion posed by evidence of compliance with a standard that did not even apply to the Discovery. (See Evid. Code, § 352.)

c. Croteau's Drop Tests

Finally, Land Rover appeals the trial court's ruling excluding the results of drop testing performed by its roof strength expert, Jeff Croteau, because, unlike Pannu's vehicle, the production Discovery vehicles used by Croteau did not have roof racks. In support of his motion, Pannu submitted evidence, including Croteau's own admission, that the absence of a roof rack significantly reduced roof deformation because the roof rack acted as a force concentrator on the precise point of roof weakness.

While the differences attributable to the presence or absence of a roof rack would normally seem to go more to the weight of the evidence rather than its admissibility (see, e.g., People v. Bolin (1998) 18 Cal.4th 297, 322 [75 Cal.Rptr.2d 412, 956 P.2d 374]), in this case the trial court's ruling fell well within the range of its discretion. In the context of a jury trial, the drop test evidence may have been unduly confusing and insufficiently probative, justifying its exclusion under Evidence Code section 352, which was one of the bases for the court's ruling. As the record reveals again, moreover, the court as fact finder was well aware of Croteau's results and found them insufficiently probative to justify revisiting their exclusion.

4. The Award of Economic Damages Was Supported by Substantial Evidence

Appellate review of a fact finder's award of damages is limited. (See Johnson v. Stanhiser (1999) 72 Cal.App.4th 357, 361 [85 Cal.Rptr.2d 82].) In the absence of error in the admission of testimony supporting a claim of economic damages, including expert opinion about lost profits, we will affirm the judgment if substantial evidence supports the damage award. (Heiner v. Kmart Corp. (2000) 84 Cal.App.4th 335, 348 [100 Cal.Rptr.2d 854] [affirming personal injury award of lost profits to self-employed dentist].)18 "Damages, even economic damages, are difficult to measure in personal injury cases. There may be disputed facts regarding the amount of medical expenses or lost wages, or disputed inferences about the probable course of events such as the length of incapacitation or whether a continuing disability will worsen, plateau, or improve." (Abbott v. Taz Express (1998) 67 Cal.App.4th 853, 856-857 [79 Cal.Rptr.2d 360].) "Technical arguments about the meaning and effect of expert testimony on the issue of damages are best directed to the [fact finder]." (Heiner, at p. 347.)

In an attempt to avoid the consequences of our limited and deferential review of the trial court's findings and award of economic damages, Land Rover contends the court erred as a matter of law in accepting evidence of the anticipated lost profits of Pannu's businesses, rather than the value of his services to those businesses. In advancing this argument, Land Rover characterizes Pannu as essentially nothing more than the manager of the stores, akin to an employee who was readily replaced by family members and paid workers.

Land Rover's argument is predicated on a distortion of the record. Pannu introduced compelling evidence of the devastating impact of his injury on his ability to work and the length of time he was likely to be able to work. The award of economic damages was based not on his stores' lost profits but on the unrebutted testimony of a vocational expert assessing Pannu's lost earning capacity as a "franchise owner/vice president." Contrary to Land Rover's grudging argument, there was ample evidence of Pannu's entrepreneurial skills, his work ethic and his consistent success in growing his businesses. In addition to simply managing the various stores (with assistance from his wife), he participated in franchise and trade associations in an effort to expand his opportunities. He had accumulated four stores (two Subways and two 7-Elevens) at the time of his injury and had recently been approved to purchase a third 7-Eleven. He had the experience and motivation to assess employee performance, to prevent malfeasance and to maximize growth. His injuries deprived him of far more than his ability to earn a manager's salary; there was thus ample evidence to support the trial court's award of economic damages.

DISPOSITION

The judgment is affirmed. Pannu is to recover his costs on appeal.

Woods, J., and Zelon, J., concurred.