Mitsubishi Motors Corporation and Mitsubishi Motors North America, Inc. appeal a district court judgment in a products liability action arising out of an accident involving a vehicle rollover. Finding no error, we affirm.

I.

In the early morning hours of October 11, 2002, Chefik Simo was a passenger in a 2000 Mitsubishi P45 Montero Sport that had been designed, manufactured, and sold by Mitsubishi. He suffered severe injuries when the vehicle rolled over on Interstate 85 near Spartanburg, South Carolina, after the driver suddenly steered left to avoid another vehicle and then attempted to correct his course by quickly turning back to the right. While the vehicle was on its side, it was struck by a Federal Express tractor trailer.

At the time of his accident, Simo was an 18-year-old freshman on the varsity soccer team at Furman University in Green-ville, South Carolina. Simo presented testimony that he was the top soccer recruit in the country the year he entered college and among the best players on the United States’ “Under-20” national team. By all accounts, he possessed outstanding speed, size, athleticism, technical ability, instincts, *297and work ethic. In addition to these qualities, his left-footedness and his experience at the left back position placed him in extraordinarily high demand at the professional level. Simo had intended to begin his professional career in Europe following the conclusion of the soccer season at Fur-man. Many European teams, including some at the top levels, had expressed interest in signing Simo when he became available.

Simo’s injuries from the accident were severe. They included a fractured shoulder blade and pelvis, dislocated shoulder, ruptured small intestine, broken wrist and finger, a knee dislocation in his left leg involving a “complete separation of the thigh bone from the shin bone” and tearing of three of the four major ligaments in the knee. Supp. J.A. 5. He suffered irreparable damage to his peroneal nerve, resulting in a “drop foot.” Id. at 6. As a result of these injuries, Simo underwent a number of surgeries and incurred more than $277,000.00 in medical bills. Although Simo undertook arduous rehabilitation efforts in an attempt to resume his soccer career, when he returned to the field he ended up overcompensating for his injuries to his left side, leading to painful stress fractures that forced him to terminate his comeback.

Simo instituted the present action in federal district court on September 21, 2004, alleging claims of strict tort liability and negligence against Mitsubishi Motors Corporation and Mitsubishi Motors North American, Inc. (collectively “Mitsubishi”).1 As is relevant here, Simo claimed that the Montero Sport was unreasonably dangerous because its center of gravity was too high, causing it to roll over in certain circumstances on flat, dry pavement (to roll over “untripped”).

In this regard, Simo presented the expert testimony of David Bilek, a person with experience in mechanical engineering, on the subjects of vehicle stability and design. Bilek had run stability tests and utilized data to evaluate vehicle dynamics for over 20 years in a litigation-consultant capacity. Bilek explained the physics involved in vehicle rollovers and discussed stability tests he performed on the Montero Sport, a testable prototype,2 and various sport utility vehicles (“SUVs”) comparable to the Montero Sport. Bilek opined that in the well-designed vehicles, lateral force by a sudden turn would cause the vehicle’s tires to slide on the pavement to the extent that they could not continue to grip the road. On the other hand, he explained that a vehicle like the Montero Sport that is unreasonably top-heavy can roll over untripped when the lateral forces on the vehicle reach a certain level. He also opined that, in light of information that had been disseminated from other manufacturers, a reasonable manufacturer would have performed testing on its vehicles to ensure that they would not rollover untripped. He opined that “handling” tests performed by Mitsubishi, in which the drivers did not expose the vehicles to forces strong enough to roll the vehicles over, were not sufficient. J.A. 1839.

Also providing testimony for Simo was engineer Michael Gilbert, who, like Bilek, testified that the Montero Sport rolled over untripped under certain circumstances, whereas better designed SUVs on the market did not. He further testified *298that designing a stable SUV is not a difficult task and had the Montero Sport been designed to have the stability of other SUVs, the accident at issue here never would have occurred.

Simo further offered extensive testimony regarding earnings that he lost as a result of the accident. In particular, Simo offered the expert testimony of Patrick McCabe, a former collegiate and professional soccer player, and then-current FIFA-licensed soccer agent.3 McCabe testified that he performs scouting duties for various North American and European soccer teams and that his job requires him to identify soccer talent and determine its market value. In light of Simo’s exceptional qualities — which McCabe described in detail — and McCabe’s knowledge of how Simo was regarded in the soccer world generally as well as by specific European teams, McCabe testified that Simo “was destined to become one of the top American players of his generation” before his accident. Id. at 1058. Based on his specialized knowledge of the market for professional soccer players in Europe, McCabe testified concerning the increasing income that Simo would have likely earned as his career progressed. Conservatively estimating that Simo would play for 15 years in Europe, McCabe estimated that Simo’s career earnings likely would have fallen within the range of $3 million to $10 million.

Finally, economist Ken McCoin provided expert testimony concerning the concept of “present value,” and he calculated the present value of the earnings that McCabe had projected Simo would have enjoyed had the accident not occurred.

At the close of the evidence, Mitsubishi moved for judgment as a matter of law on the negligence and strict liability claims. The district court granted the motion as to Simo’s claim for negligent failure to warn, and as to his request for punitive damages, but otherwise denied the motion. Subsequently, the jury returned a verdict in Simo’s favor for $7 million in compensatory damages.

Following the verdict, Mitsubishi moved for judgment as a matter of law on the basis that Simo had failed to present sufficient evidence that an alternative feasible design existed for the Montero Sport on the date of its manufacture and sale that would have prevented the accident here. Mitsubishi also moved for a new trial on the grounds that the district court improperly admitted testimony from Simo’s experts and that Simo’s counsel made improper argument during trial and. closing arguments, resulting in prejudice to Mitsubishi. Mitsubishi finally moved for a new trial nisi remittitur on the grounds that the verdict was excessive, that Simo failed to offer sufficient evidence regarding an alternative feasible design, and that Mitsubishi was entitled to a set-off in the amounts Simo had received in settlement with other parties for claims involving the accident. The district court granted Mitsubishi’s motion for a set-off but otherwise denied its post-trial motions, and entered final judgment against Mitsubishi for $6,050,000.00.

II.

Mitsubishi first argues that the district court erred in denying its motion for judgment as a matter of law on the ground that Simo failed to establish the existence of an alternative feasible design that would have prevented or reduced Simo’s injuries. We disagree.

Under South Carolina law, which the parties agree applies in this diversity suit, “in order to find liability under any prod*299ucts liability theory, a plaintiff must show: (1) he was injured by the product; (2) the injury occurred because the product was in a defective condition, unreasonably dangerous to the user; and (3) that the product at the time of the accident was in essentially the same condition as when it left the hands of the defendant.” Bragg v. Hi-Ranger, Inc., 319 S.C. 531, 462 S.E.2d 321, 326 (1995). Proving the existence of an alternative feasible design is a “crucial aspect” of this required showing. Little v. Brown & Williamson Tobacco Corp., 243 F.Supp.2d 480, 495 (D.S.C. 2001). We review de novo a district court’s denial of a Rule 50(b) judgment as a matter of law, viewing the evidence in the light most favorable to the prevailing party and drawing all reasonable inferences in his favor. See Konkel v. Bob Evans Farms Inc., 165 F.3d 275, 279 (4th Cir. 1999).

Here, Simo presented expert testimony that designing an SUV that will not rollover untripped is not difficult so long as the issue is addressed early in the design process. See Restatement (Third) of Torts: Prod. Liability § 2 cmt. f (1998) (explaining that qualified expert testimony may establish that an alternative feasible design existed “if it reasonably supports the conclusion that [such a] design could have been practically adopted at the time of sale”). Indeed, Simo presented evidence that, at the time Mitsubishi sold the Montero Sport at issue here, several other SUVs already on the market had centers of gravity sufficiently low that the vehicles would not roll over untripped. Thus, the district court correctly denied Mitsubishi’s motion. See id. (“[Ojther products already available on the market ... may serve as reasonable alternatives to the product in question.”).

III.

Mitsubishi also contends that the district court erred in admitting expert testimony from David Bilek. We disagree. The admissibility of expert testimony is governed by Federal Rule of Evidence 702, which provides:

If scientific, technical, or other specialized knowledge will assist the trier of fact to understand the evidence or to determine a fact in issue, a witness qualified as an expert by knowledge, skill, experience, training, or education, may testify thereto in the form of an opinion or otherwise, if (1) the testimony is based upon sufficient facts or data, (2) the testimony is the product of reliable principles and methods, and (3) the witness has applied the principles and methods reliably to the facts of the case.

Fed.R.Evid. 702. We review a decision to admit or exclude expert testimony for abuse of discretion. See Gen. Elec. Co. v. Joiner, 522 U.S. 136, 141, 118 S.Ct. 512, 139 L.Ed.2d 508 (1997); Cooper v. Smith & Nephew, Inc., 259 F.3d 194, 200 (4th Cir. 2001).

Bilek was experienced in applying mechanical engineering principles. After obtaining his bachelor of science degree in mechanical engineering technology, Bilek received training in vehicle stability issues for several years from Dr. Michael Kaplan, who possesses a Ph.D. in mechanical engineering as well. In doing “litigation consulting-type work,” Bilek gained extensive experience and knowledge over twenty years concerning how to perform stability testing on vehicles. J.A. 1732. He has performed “hundreds and hundreds of tests associated with vehicle stability [and] vehicle rollover resistance.” Id. at 1734. As a result, he has specialized knowledge concerning the tests that manufacturers employ and experience in evaluating the effectiveness of different design modifications in protecting against rollovers. Bilek also has “reviewed thousands of pages of internal documents, test data, test reports, test video tapes, deposition testimony of *300in-house engineers and other materials related to the design and development of various vehicles,” most of which have been written by and for engineers. Id. at 937. He has also reviewed many documents authored by the National Highway Traffic Safety Administration, as well as publications from the Society of Automotive Engineers, and many others relating to vehicle design and rollovers.

All of this training qualified Bilek to testify, as he did, regarding the physics involved in rollovers, his testing of the various vehicles to determine whether they roll over untripped, and the state of knowledge of the risk of SUV rollovers. See Clay v. Ford Motor Co., 215 F.3d 663, 668-69 (6th Cir. 2000) (holding that district court did not abuse its discretion in admitting testimony of engineer, who had never worked in the automotive industry and never tested a two-wheel drive Bronco II before the suit in question, that the instability of that vehicle rendered it defective). Mitsubishi appears to argue, however, that Bilek was not qualified to offer expert testimony concerning whether his notion of designing the Montero Sport to be lower and wider could be feasibly implemented. Even assuming that Bilek was not qualified to offer testimony concerning the feasibility of his own design — the prototype— he did not purport to do so. As we have explained, there was no need for Bilek to theorize about whether Mitsubishi could design an SUV with utility equal to the Montero Sport that would not roll over untripped because Simo presented testimony that several such vehicles were already on the market.

rv.

Mitsubishi next maintains that the district court erred in admitting the expert testimony of Patrick McCabe. Mitsubishi argues that the testimony violated the standard of admissibility established in Federal Rule of Evidence 702 and in Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579, 113 S.Ct. 2786, 125 L.Ed.2d 469 (1993). Mitsubishi also argues that the district court erred in admitting Ken McCoin’s testimony regarding Simo’s future earnings because his testimony relied on McCabe’s testimony. We conclude that the district court was within its discretion in admitting McCabe’s and McCoin’s testimony.

A district court determining the admissibility of expert testimony has a gate-keeping responsibility to “ensur[e] that an expert’s testimony both rests on a reliable foundation and is relevant to the task at hand.” Kumho Tire Co. v. Carmichael, 526 U.S. 137, 141, 119 S.Ct. 1167, 143 L.Ed.2d 238 (1999) (internal quotation marks omitted). McCabe testified that an important part of a job as a sports agent was evaluating the worth of soccer players on the market. In estimating the income that Simo could have been expected to earn absent the injury, McCabe drew on his own evaluation of Simo’s abilities as well as those of others involved with the sport who believed that Simo “was destined to become one of the top American players of his generation” before his accident. J.A. 1058. Based on this evaluation, as well as his awareness of the fact that Simo had wanted to pursue his soccer career in Europe, McCabe utilized his specialized knowledge of the earnings opportunities Simo would likely have had in his career. In so doing, he noted as a point of comparison the salaries of eight then-current or former left-footed players from the American Senior National Team. Cf. Drews Co. v. Ledwith-Wolfe Assocs., Inc., 296 S.C. 207, 371 S.E.2d 532, 536 (1988) (noting with approval the “yardstick” method of proving future lost profits by comparing business to one of similar size, nature, and location).

Mitsubishi challenges the district court’s conclusion that McCabe’s methodology was *301sufficiently reliable. It argues that McCabe “made no reference to objective sources or outside information to explain how he determined [Simo’s] supposed career path playing professional soccer” and that Simo’s' prospects for future success were “simply unknown and unknowable.” Appellants’ Cross Br. at 43, 46. It further maintains that the district court did not specifically consider the non-exclusive list of factors set out in Daubert for judging the reliability of methodologies underlying expert opinions. We find no fault with the district court’s admission of McCabe’s testimony. The factors identified in Daubert “do not constitute a definitive checklist or test.” Kumho Tire, 526 U.S. at 150, 119 S.Ct. 1167 (internal quotation marks omitted). Rather, the inquiry into the reliability of an expert’s methodology must be flexible and. case-specific. See Maryland Cas. Co. v. Therru-O-Disc, Inc., 137 F.3d 780, 784-85 (4th Cir. 1998). Here, the district court reasonably accepted that a soccer player’s value can be reliably estimated by the personal observations and experience of a person whose job requires him to evaluate players’ abilities and determine their value.4 Cf. Kumho Tire, 526 U.S. at 150, 119 S.Ct. 1167 (recognizing that “the relevant reliability concerns may focus upon personal knowledge or experience” because “there are many different kinds of experts”).

While neither McCabe nor anyone else could predict with certainty what the future would have held for Simo, South Carolina damages law did not require such certainty. See South Carolina Fin. Corp. of Anderson v. W. Side Fin. Co., 236 S.C. 109, 113 S.E.2d 329, 336 (1960) (“The law does not require absolute certainty of data upon which lost profits are to be estimated, but all that is required is such reasonable certainty that damages may not be based wholly upon speculation and conjecture, and it is sufficient if there is a certain standard or fixed method by which profits sought to be recovered may be estimated and determined with a fair degree of accuracy.” (internal quotation marks omitted)). McCabe explained that his projections encompassed “a range of averages,” rather than a precise prediction of Simo’s future. J.A.2025. And, it is noteworthy that even Mitsubishi’s expert testified that he was sufficiently informed to offer a “probable career path” for Simo. Id. at 1092; cf. Correa v. Cruisers, 298 F.3d 13, 26 (1st Cir. 2002) (“Acceptance of the methodology by the other party’s expert may give additional credence to the reliability of the proffered testimony.”).5

V.

On cross-appeal, Simo argues that the district court erred in granting Mitsubishi’s motion for judgment as a matter of *302law on the issue of punitive damages. However, Simo requests that his case be remanded for a trial on punitive damages “if and only if the issue of punitive damages can be tried without disturbing [his] award of actual damages.” Appellee’s Br. at 77. Because we determine that the issue of punitive damages could not be tried without disturbing the compensatory damage award, we need not decide whether the district court erred in granting judgment as a matter of law to Mitsubishi on the punitive damages issue.

Although the propriety of granting a new trial on fewer than all issues of fact is well established, we should not order a new trial on the sole issue of punitive damages if “the evidence relating to wilful misconduct is so inextricably tied up with that relating to primary negligence that a fair trial upon either issue requires a trial of both issues together.” Atl. Coast Line R.R. Co. v. Bennett, 251 F.2d 934, 939 (4th Cir. 1958). In Bennett, the plaintiffs recovered compensatory and punitive damages in an action involving an accident which occurred when a freight train engineer failed to slow the train to an appropriate speed. See id. at 936. We held that the punitive damage award could not stand because it was based on an improper jury instruction. See id. at 938. Nevertheless, we refused to grant a new trial on punitive damages only, reasoning that the amount, if any, of punitive damages to be awarded could be intelligently determined only in connection with the jury’s consideration of the entire case. See id. at 938-39.

As in Bennett, the evidence that Simo relies on here to support his punitive damages claim is largely the same evidence on which he relies to establish the defectiveness of Mitsubishi’s vehicle. His argument for punitive damages boils down in large part to the proposition that not only was the vehicle defective, but it was so obviously defective that Mitsubishi’s failure to remedy the defect amounted to at least recklessness. We therefore conclude that remand for a trial on punitive damages only would not be appropriate. See id.; but cf. Atlas Food Sys. & Servs., Inc. v. Crane Nat’l Vendors, Inc., 99 F.3d 587, 599-600 (4th Cir. 1996) (holding that the district court’s grant of new trial on sole issue of punitive damages did not amount to an abuse of discretion).

VI.

In sum, we affirm the judgment of the district court.

AFFIRMED.

concurring in part and dissenting in part:

I concur in Parts I-IV of the majority opinion, affirming the jury’s verdict on liability and compensatory damages. However, I believe the district court erred in excluding Simo’s claim for punitive damages from the jury’s consideration. In addition, I believe it is entirely appropriate to remand this case for a trial solely on punitive damages. Accordingly, I dissent from Parts V and VI of the majority opinion.

Under South Carolina law, “punitive damages may be awarded to punish only those tortfeasors who have acted in a reckless, willful or wanton manner.” Jimenez v. DaimlerChrysler Corp., 269 F.3d 439, 449-50 (4th Cir. 2001) (internal punctuation omitted). In meeting this standard, a plaintiff need not show that a tortfeasor had actual knowledge of the danger his tortious behavior posed to another. Rather, a plaintiff can recover punitive damages “even when the wrongdoer does not *303actually realize that he is invading the rights of another, provided the act is committed in such a manner that a person of ordinary prudence would say that it was a reckless disregard of another’s rights.” Atlas Food Systems & Services, Inc. v. Crane Nat’l Vendors, Inc., 99 F.3d 587, 599 (4th Cir. 1996).

Simo produced evidence which indicates that Mitsubishi (1) knew or should have known of the risk that its Montero Sport would roll over because it could not sustain the tire forces which come into play following an emergency swerve and (2) recklessly disregarded those risks.6 More specifically, as to the risk of an accident and Mitsubishi’s knowledge thereof, Simo’s evidence shows: that drivers of Mitsubishi’s vehicles would, from time to time, have to undertake accident avoidance maneuvers; that such maneuvers could expose the vehicles to “limit tire forces”; that vehicles whose static stability factor is as low as 1.06 are likely to roll over, untripped in such circumstances; that the consequence of a rollover would be severe, inasmuch as rollovers are one of the deadliest accident scenarios; that Mitsubishi conceded it was on notice that its vehicles had been involved in roll overs; and that any reasonable manufacturer would have known of the risk based on the testing, research, and literature available in 2000. As to Mitsubishi’s disregard of these risks, Simo introduced evidence demonstrating: that Mitsubishi did not perform “limits” testing to determine whether the Montero Sport would remain stable in limit tire force conditions; that it knew how to perform such testing (and had done so for other vehicles); that it knew or should have known that only limits testing would allow it to evaluate whether the Montero Sport posed unnecessary risks of untripped rollovers; that it did not alter the design of the Montero Sport; and that it sold the Montero Sport despite these failures.

Mitsubishi, of course, vigorously contested Simo’s allegations and his evidence both in its case-in-chief and on cross-examination. Nonetheless, it was for the jury to determine whether to accept Simo’s evidence and credit his witnesses or to side with Mitsubishi instead. Because Simo’s evidence, if credited, supports a claim for punitive damages, the district court erred in excluding it from the jury’s consideration. See Atlas Food Systems, 99 F.3d at 597 (holding that a district court should not grant judgment as a matter of law “unless there was no evidence presented in the case that would authorize the jury’s verdict”) (emphasis added).

The majority does not consider whether Simo presented a viable claim for punitive damages because it concludes that a trial limited to that issue would be “inappropriate.” In reaching this conclusion, the majority relies on Atl. Coast Line R. Co. v. Bennett, 251 F.2d 934, 938 (4th Cir. 1958), in which we applied the established principle that a partial new trial should not be ordered unless the issue to be retried is separate and distinct from the remainder of the case such that “the trial of it alone may be had without injustice” to bar a new trial limited to punitive damages. The Bennett court based its holding on its observation that “the evidence relating to wilful misconduct is inextricably tied up with that relating to primary negligence.” Id. at 939. I do not believe, however, that Bennett established a firm rule that a new trial on punitive damages is inappropriate as a matter of law where a plaintiff relies on overlapping evidence to support his claims for compensatory and punitive damages. In fact, we implicitly rejected this *304interpretation of Bennett in Atlas Food Systems. There, we concluded that a district court did not abuse its discretion when it ordered a new trial on punitive damages because “[a]ll of the evidence relating to [the defendants’] willful or wanton conduct was before the second jury, enabling it to render a proper verdict on both the liability for and amount of punitive damages.” Atlas Food Systems, 99 F.3d at 599-600. Similarly, in this case, if a new trial on punitive damages were granted, a jury would be able to consider in full the evidence relating to Mitsubishi’s alleged willfulness and recklessness. Therefore, I would reverse the district court’s grant of judgment as a matter of law on Simo’s punitive damages claim and would remand for a new trial on that claim.