OPINION

SUTTON, Circuit Judge.

Robert Fabian, a representative of a yet-to-be-certified class, seeks recovery from a helmet manufacturer for misrepresenting the safety of its helmets. The district court granted the defendant's motion to dismiss the suit, holding that Fabian's complaint fails to state a cognizable claim. Because Fabian's factual allegations, when construed in his favor, state a plausible claim for relief, we reverse.

I.

Tucked within the Department of Transportation is the National Highway Traffic Safety Administration (NHTSA), which regulates the performance of motorcycle helmets (among other vehicle products) under the National Traffic and Motor Vehicle Safety Act of 1966. 49 U.S.C. § 30101 et seq. In accordance with the Safety Act, NHTSA promulgated Standard No. 218, which spells out the testing procedures that helmets sold in the United States must satisfy. 49 C.F.R. § 571.218. One of these procedures is an "impact attenuation test," which involves the dropping of a helmet from a minimum height of six feet onto an anvil to measure the effect of the impact on the helmet. Id. at S7.1. Another test applies force to a helmet's chin strap to determine whether the helmet will remain in place during a crash. See id. at S7.3.

Standard 218 relies on self-certification, which means that companies test and certify their own helmets rather than having NHTSA do it for them. When helmets pass the test, the companies place a "DOT" label on them. 49 C.F.R. § 571.218 at S5.6. 1(e). NHTSA enforces these requirements by randomly purchasing helmets, employing independent companies to run compliance tests on them and publishing the results.

Headquartered in Memphis, Tennessee, Fulmer Helmets designs, manufactures and distributes the AF-50 Trooper motorcycle helmet. The AF-50 helmet comes in at least two sizes: small and large.

In 2000, NHTSA selected the large AF-50 helmet for testing. The helmet passed each component of the test. In 2002, NHTSA selected the small AF-50 for testing and it failed two components of the test. It failed the impact attenuation test (because the helmet's absorption of impact, measured in time, exceeded the regulatory requirement of 2.0 milliseconds by 0.2 milliseconds), and it failed the labeling requirement (because the "DOT" symbol appeared an eighth of an inch too high from the bottom of the helmet). Fulmer Helmets did not issue a recall, take any action to inform purchasers or retailers that it had failed the test, or fix or remove the DOT symbol.

On July 22, 2004, Fabian bought two large Fulmer AF-50 helmets. In 2007, Fabian sold one of the helmets to a friend, who later died of severe brain trauma in a motorcycle crash while wearing the helmet.

In April 2009, Fabian filed a complaint against Fulmer Helmets in Tennessee state court, which was later removed to federal court on diversity grounds. The federal complaint alleged (1) fraudulent misrepresentation, (2) negligent misrepresentation, (3) breach of the implied warranty of merchantability and (4) unjust enrichment. Fabian claimed he had "relied on Fulmer's material misrepresentations that such helmets were `DOT approved,'" causing him to purchase an "unsafe," "inferior-quality" helmet that created a "heightened risk of serious physical injury or death." R.5 ¶¶ 3, 4, 23. Fabian sought class certification for all persons who had purchased the AF-50 since the failed 2002 test, while excluding "anyone seeking to recover for physical injuries suffered due to the failure of the subject helmets." R.5 ¶ 24. Fabian asked for damages in the form of a refund and disgorgement of profits.

Fulmer Helmets filed a Rule 12(b)(6) motion to dismiss the case for these reasons, among others: (1) Fabian failed to state a claim as a matter of law; (2) the Safety Act preempted the lawsuit; and (3) Tennessee's statute of limitations barred the claims.

The district court granted Fulmer Helmets' motion to dismiss. It held that Fabian failed to state a claim because Fabian had purchased two large helmets, and only small helmets failed the 2002 test. The court reasoned that Fabian had not purchased "helmets similar in all respects to the helmets which allegedly failed the August 2002 testing forming the basis for [Fabian's] claims." R.24 at 27. At the same time, however, the court rejected one of Fulmer Helmets' alternative arguments, holding that the Safety Act did not expressly or impliedly preempt Fabian's claims. It reasoned that the Act's savings clause carves state common law claims from the preemptive scope of the statute and that the state law claims would not "present an obstacle to the federal objective of `self-certification.'" Id. at 24. Lastly, the court held that Tennessee's statute of limitations barred Fabian's claims for breach of implied warranties.

Fabian appeals the failure-to-state-a-claim ruling but not the breach-of-implied-warranties ruling. Fulmer Helmets challenges the preemption ruling.

II.

To survive a motion to dismiss, Fabian must plead "enough factual matter" that, when taken as true, "state[s] a claim to relief that is plausible on its face." Bell Atl. Corp. v. Twombly, 550 U.S. 544, 556, 570, 127 S.Ct. 1955, 167 L.Ed.2d 929 (2007). Plausibility requires showing more than the "sheer possibility" of relief but less than a "probab[le]" entitlement to relief. Ashcroft v. Iqbal, ___ U.S. ___, 129 S.Ct. 1937, 1949, 173 L.Ed.2d 868 (2009).

In granting the motion to dismiss, the district court used the following chain of reasoning: (1) NHTSA performed a safety test on a large AF-50 helmet in 2000, and the helmet passed all components of the test; (2) NHTSA performed a safety test on a small AF-50 helmet in 2002, and the helmet failed at least one component of the test; and (3) because Fabian premises his claim on the purchase of large AF-50 helmets, his claim is implausible on its face given that Fulmer Helmets passed a 2000 NHTSA test on a large AF-50 helmet.

The problem with this chain of reasoning is that it turns on potential inferences, not necessary ones. There are at least two legitimate ways to think about the significance of the NHTSA tests, and they point in opposite directions when it comes to the merits of this lawsuit. One is that the difference between the 2000 and 2002 test results turns on differences between the performance of the small and large AF-50 helmets. If so, that would support the district court's ruling that the disparity between the size of the helmet bought and the size of the helmet tested is fatal to Fabian's claims. The other reasonable inference, however, is that helmets of the same model, even if differently sized, perform the same. Two differently sized helmets, for example, may be no more distinct as a matter of performance than two differently sized pairs of shoes or two differently sized pairs of pants. If so, the failed 2002 test potentially exposed a defect in all AF-50 helmets, no matter their size.

In the absence of further development of the facts, we have no basis for crediting one set of reasonable inferences over the other. Because either assessment is plausible, the Rules of Civil Procedure entitle Fabian to pursue his claim (at least with respect to this theory) to the next stage—to summary judgment or, if appropriate, a trial after the parties have engaged in any relevant discovery to support one or the other interpretation. So long as we can "draw the reasonable inference that the defendant is liable for the misconduct alleged," Iqbal, 129 S.Ct. at 1949, a plaintiff's claims must survive a motion to dismiss. That inference is reasonable here because "common sense," id. at 1950, tells us that a mass-manufactured consumer product, whether it is shoes, pants or helmets, may utilize the same design (and carry the same flaw) regardless of its size.

Fulmer Helmets stresses that Fabian's large helmet has "passed all tests" and that the 2002 test is irrelevant. Fulmer Helmets Br. 16. But that does not necessarily end the inquiry. The company may have changed its design or manufacturing process for all AF-50s between 2000 and 2002, giving rise to a defect in all of its helmets and negating the relevance of the successful 2000 test result. Or the same test conducted on two randomly selected helmets (otherwise exactly the same) might yield different outcomes due to nothing more than natural statistical variances. The successful 2000 test thus may reflect an aberration unrelated to helmet size, while the failed 2002 test may point to a real flaw in all AF-50s. Because Fabian has "nudged his claims ... across the line from conceivable to plausible," Iqbal, 129 S.Ct. at 1950-51, he deserves a shot at additional factual development, which is what discovery is designed to give him.

III.

Fulmer Helmets seeks to affirm the judgment on alternative grounds, namely that the Safety Act preempts Fabian's lawsuit. We disagree.

Although Fulmer Helmets' position is not a picture of clarity, the company seems to raise the following multi-step argument. Step one: the Safety Act contains several enforcement mechanisms, including recalls initiated by NHTSA or actions initiated by the Attorney General to enjoin violations of the Safety Act and to collect civil penalties for violations of the Act. Step two: the Safety Act contains no private enforcement provision. Step three: Fabian cannot do indirectly what he cannot do directly by couching his private efforts to enforce the provisions of the Safety Act as state law misrepresentation claims and the like. Step four: Fabian's claims therefore must be dismissed either because they are unauthorized by the Act or are preempted by it.

We can make short work of one aspect of this argument. That the Safety Act does not expressly create a private enforcement action does not by itself defeat Fabian's claims. He has filed each claim under Tennessee law, which authorizes private enforcement actions for fraudulent and negligent misrepresentation. Fabian has no need for, and thus need not invoke, a private right of action under the Safety Act.

The other aspect of Fulmer Helmets' argument—its preemption claim—deserves a longer response. The Safety Act contains a preemption provision, 49 U.S.C. § 30103(b)(1) ("When a [federal standard] is in effect ... a State may prescribe or continue in effect a standard ... only if the standard is identical to the [federal] standard."), and a savings clause, id. § 30103(e) ("Compliance with a [federal standard] does not exempt a person from liability at common law."). The savings clause "excludes common-law tort actions" from the express preemption clause, allowing "state tort law to operate—for example, where federal law creates only a ... minimum safety standard." Geier v. Am. Honda Motor Co., 529 U.S. 861, 868, 120 S.Ct. 1913, 146 L.Ed.2d 914 (2000). At the same time, however, the savings clause does not "foreclose or limit the operation of" implied conflict preemption principles, id. at 869, 120 S.Ct. 1913, namely the possibility that a state common law action nonetheless would be preempted because it conflicts with, or "stand[s] as an obstacle" to, id. at 886, 120 S.Ct. 1913, the Safety Act or Standard 218.

All of this means two things: (1) the savings clause spares Fabian's misrepresentation claims from express preemption; and (2) to the extent preemption exists, it is because these state law actions "stand as an obstacle" to Standard 218. We see no such obstacle, however.

In passing the Safety Act, Congress sought to "reduce traffic accidents and deaths and injuries resulting from traffic accidents" by authorizing the Secretary of Transportation to "prescribe motor vehicle safety standards." 49 U.S.C. §§ 30101, 30111. The Act defines a "safety standard" as a "minimum standard for ... motor vehicle equipment performance." Id. § 30102. In 1972, NHTSA decided to address the "dramatic increase in both motorcycle registrations and motorcyclist fatalities" in the previous five years and its finding that "two-thirds to three-fourths of motorcycle fatalities result from head injuries." 37 Fed.Reg. 10,079, 10,079 (May 19, 1972). At the time, helmet companies relied on industry-established quality standards, but the agency found that "some helmets, contrary to their own certification labels, do not meet the requirements of industry specifications, or otherwise exhibit unacceptable characteristics." Id.

NHTSA, as a result, promulgated Standard 218 with the hope of "reduc[ing] deaths and injuries to motorcyclists and other motor vehicle users resulting from head impacts" by establishing "minimum performance requirements for helmets." 38 Fed.Reg. 22,390, 22,391 (Aug. 20, 1973). Manufacturers must label their helmets with a "DOT" sticker, which "constitut[es] the manufacturer's certification that the helmet conforms to the applicable Federal motor vehicle safety standards." 49 C.F.R. § 571.218 at S5.6.1(e). The label represents to the public that the manufacturer has satisfied Standard 218's testing requirements. NHTSA enforces this self-certification regime through "testing, inspection, investigation, or research," 49 U.S.C. § 30118, and has the authority to demand that the manufacturer remedy any defect, such as by issuing a recall, see id. §§ 30118(b)(2)(B), 30120.

The upshot is that Standard 218 creates minimum performance standards that helmet manufacturers must meet in order to place "DOT" labels on their helmets, all subject to NHTSA oversight. Fabian's claims do not conflict with this regulatory regime. The premise of Fabian's common law claims is not the creation of a new standard, whether one below, at or above Standard 218. It is that Fulmer misrepresented its helmets as "DOT approved" through its marketing materials, website and catalogues, as well as by the placement of the "DOT" symbol on the helmets, even after knowing that it failed the 2002 safety test. Liability, if it exists at all, would turn on what Fulmer Helmets said about its products, not on whether its products meet a standard that conflicts with Standard 218.

These claims thus do not "actually conflict" with the requirements of, or the purposes of, the Safety Act or Standard 218. Geier, 529 U.S. at 869, 120 S.Ct. 1913. They do not change Standard 218's technical requirements. They do not disturb Standard 218's labeling requirements. And they do not add a new requirement that interferes with what Standard 218 already requires. All that the claims do is potentially impose liability based on representations about whether the Department of Transportation has approved the helmets, even after a failed government-sponsored test. Cf. Cipollone v. Liggett Grp., Inc., 505 U.S. 504, 528-29, 112 S.Ct. 2608, 120 L.Ed.2d 407 (1992) (plurality) (holding that a fraudulent misrepresentation claim was not expressly preempted by the Federal Cigarette Labeling and Advertising Act because it turned on a "general obligation"—the "duty not to deceive"—and because "State-law prohibitions on false statements of material fact do not create diverse, nonuniform, and confusing standards"); see also Altria Grp., Inc. v. Good, 555 U.S. 70, 129 S.Ct. 538, 545-49, 172 L.Ed.2d 398 (2008).

This approach is consistent with Geier. There, the Supreme Court held that Federal Motor Vehicle Safety Standard 208, which gave car manufacturers a "range of choices" in installing passive restraint systems, impliedly preempted a tort suit that created a duty for defendant-manufacturers to install air-bags (as opposed to other restraint systems). 529 U.S. at 874-75, 120 S.Ct. 1913. The Court was concerned that such a duty "presented an obstacle to the variety and mix of devices that the federal regulation sought." Id. at 881, 120 S.Ct. 1913. It thus reasoned that Standard 208 contemplated a policy objective that "safety would best be promoted if manufacturers installed alternative protection systems in their fleets rather than one particular system in every car," id., and an air-bag requirement conflicted with this objective. No analogous purpose exists with respect to Standard 218, which seeks only to establish "minimum performance requirements for helmets." 38 Fed.Reg. at 22,391.

IV.

For these reasons, we reverse and remand the case to the district court to consider Fulmer's other defenses in the first instance.