Derek BRASWELL, Plaintiff-Appellant, v. CINCINNATI INCORPORATED, a foreign corporation, Defendant-Appellee, and Hardy Machinery, a foreign corporation, Defendant, 731 F.3d 1081

TYMKOVICH, Circuit Judge. Derek Braswell suffered a horrific workplace accident. While he was operating a press brake, a heavy machine tool, manufactured by Cincinnati, Inc., his right arm was crushed, and eventually had to be amputated. Signs on the machine warned the operator not to reach into the die area, where a hydraulic-powered ram descends to bend sheet metal. Despite these warnings, Braswell reached into the die area to remove a jammed piece of metal. While doing so, he accidentally stepped on a pedal triggering the ram’s descent.

More unfortunate, the machine’s safety equipment designed to prevent this type of accident had been removed or disabled sometime prior to the accident. After his injury, Braswell filed a suit against Cincinnati on theories of strict products liability and negligence. The suit was asserted as a diversity claim, and that basis for subject-matter jurisdiction is not challenged. The district court granted

summary judgment for Cincinnati on the grounds that a subsequent owner had modified the press brake to create the danger and that the gated pedal on the original model made the press brake not unreasonably dangerous. We agree that the press brake was not unreasonably dangerous: with its warnings and safety devices, the machine did not pose a danger beyond that which the ordinary operator of the machine would appreciate. Exercising our jurisdiction under 28 U.S.C. § 1291, we AFFIRM the district court’s judgment. I. Background A.

The Machine Cincinnati manufactures a hydraulic press brake, a machine tool commonly used to shape sheet metal. The tool consists of a ram, which presses the metal, and a die, into which the metal is pressed. The operator feeds sheet metal between the die and the ram, and may hold the sheet until the ram has lifted. The machine is operated by a foot pedal, known as a footswitch, though other operator controls are usually included. The machine is designed to have a long lifetime, and often has multiple owners.

Given the diverse and complex nature of various manufacturing processes, many of the machine’s features may be added or removed according to an owner’s needs. For example, the machine could be set up to work on thick pieces of metal, requiring a wide opening to feed, or thin pieces, requiring an opening as small as a half inch. Derek Braswell’s employer, Ventaire, Inc., purchased a Cincinnati press brake in 2007 from Hardy Machinery. At the time of the purchase, the machine was almost twenty years old. Cincinnati had first sold the press brake in 1989 to a company named Steelgard.

Though the machine passed to many successive owners, Cincinnati continued to supply repair services. For example, one year before the accident a Cincinnati service technician visited Ven-taire to install new software in the press brake and reprogram the machine. As originally sold, the machine was equipped with a footswitch that had a front flap, or gate, to prevent its accidental depression. The operator had to lift the gate with his foot to access the enclosed pedal, and the gate would be resting on top of his foot while he used the machine.

Moreover, the machine came equipped with two footswitehes, each of which had to be depressed simultaneously by two different operators in order to trigger the ram. Sometime between the press brake’s original sale to Steelgard and its sale to Ventaire, the original footswitehes were removed and replaced with ones that did not have a gate. At the time of Braswell’s accident, one of the replacement foot-switches had been disabled, such that the machine could be operated with a single footswitch unprotected by any gate.

Another safety feature on the original press brake was dual palm stations, a feature designed to prevent the accidental activation of the machine by a sole operator. The palm stations required two operators to each place their respective hands on a different button simultaneously in order to activate the machine, thereby making it nearly impossible for the machine to be activated while one operator’s limb was inside the die area. This feature could be used in conjunction with the foot-switches.

The downside of this feature, however, was that an operator would not be able to hold the piece of sheet metal being fed into the machine. After Ven-taire acquired the press brake, it disabled the palm stations so that the machine could be operated by one footswitch alone. A final safety feature was a control panel, featuring an emergency stop button, that was mounted on the face of the machine. This emergency button on the panel could be reached by anyone standing close to the die area.

By the time Ven-taire purchased the press brake, however, the panel had been removed: it had been unbolted, rewired, and moved to a “remote pendant.” Aplt.App. 153. The machine also came equipped with warnings displayed on prominent places on the machine: THE FOOT SWITCH YOU ARE USING CANNOT PROTECT YOU FROM SERIOUS INJURY. YOUR HANDS OR FINGERS CAN BE CRUSHED OR CUT OFF IF THE MACHINE YOU ARE OPERATING DOES NOT HAVE A GUARD OR OTHER WAYS TO KEEP YOU AWAY FROM DANGEROUS MOVING MACHINE PARTS. TO REDUCE THE POSSIBILITY OF INJURY ... ... DO NOT PLACE YOUR HANDS IN THE DIE AREA. ...

DO NOT POSITION ANY PART OF YOUR BODY WHERE IT MAY BE STRUCK OR CRUSHED BY PART MOVEMENT. ... ALWAYS READ AND UNDERSTAND THE OPERATION, MAINTENANCE AND SAFETY MANUAL BEFORE INSTALLING DIES, OPERATING OR SERVICING THE PRESS BRAKE. Id. at 27. As part of its operation, the press brake could be equipped with blocks to prevent the ram from descending.

Though there is a factual dispute about whether Ventaire properly trained Braswell on using the blocks and whether it had made blocks available for employees, Braswell does not dispute that, in the press brake’s manual, Cincinnati recommended using blocks whenever anyone — in spite of the warnings against doing so — needed to reach into the die area.

Ventaire had programmed the machine to operate in a three-step cycle: The first compression of the footswitch caused the ram to lower so that the metal was pinned; the second one caused the ram to actually press and bend the metal into the die; and the third one caused the ram to lift up. B. The Accident On the day of the accident, Braswell was operating the press brake alongside a coworker. Braswell and the coworker were feeding in sheet metal, while only Braswell was operating the ram using a footswitch. Both operators were standing, and the functioning footswitch was mere feet from the press brake.

A piece of sheet metal became stuck in the machine, and Braswell reached into the machine to remove the metal. At that point, Braswell accidentally pressed the footswitch, causing the ram to pin his arm inside the machine. Braswell screamed for help, and his supervisor came to his assistance. Instead of pressing the ram lift button, however, the supervisor triggered the next movement in its cycle, further crushing Braswell’s arm. When the ram was finally lifted, Braswell was rushed to the hospital, where his right hand had to be amputated. C. The Lawsuit Braswell filed suit in district court against Cincinnati.

Braswell made three claims: (1) strict liability defective design, (2) negligent manufacture and design, and (3) failure to warn. After discovery, Cincinnati filed a motion for

summary judgment. In opposing the motion, Braswell alleged that the press brake had an unreasonably dangerous design because, among other reasons, it was not equipped with an anti-trip footswitch, which requires a complete depression of the pedal each time the operator wants to reinitiate the machine’s movement — a mechanism, Braswell contended, that would have prevented his injury. The district court noted that Braswell, to defeat

summary judgment, had to show there were disputed issues of material fact on each element of his products liability claim. Of the three elements in a products liability claim — that the press brake was (1) the cause of the injury, (2) defective at the time it left Cincinnati’s control, and (3) unreasonably dangerous — the court concluded Braswell could not show a genuine issue on elements two or three. The press brake was not defective at the time it left Cincinnati’s control because it contained a gated footswitch, which, the district court reasoned, would have prevented the accident.

And the inclusion of the gated foot-switch made the brake not unreasonably dangerous. The district court also stated that Oklahoma’s product liability test encompassed both negligent and strict liability design theories; as a result, the court concluded, the negligent design claim failed as well. Finally, the district court concluded the inclusion of warnings on the machine satisfied Cincinnati’s duty to warn. Accordingly, the district court entered

summary judgment in Cincinnati’s favor. II. Analysis Braswell challenges the district court’s rulings on (1) the products liability claims, and (2) the negligence claim. “We review a grant of

summary judgment de novo, applying the same standard as the district court.” Oldenkamp v. United Am. Ins. Co., 619 F.3d 1243, 1246 (10th Cir.2010). Oklahoma substantive law applies to this diversity action. See Ahrens v. Ford Motor Co., 340 F.3d 1142, 1145 (10th Cir.2003). Applying Oklahoma law, we conclude there is no genuine issue of material fact on the products liability claims, and that Braswell waived his negligence arguments below by failing to sufficiently plead them or raise them in response to Cincinnati’s motion for

summary judgment. A. Products Liability Claims — Defective Design & Failure To Warn Oklahoma, like most states, has adopted a strict liability regime for products that are defectively manufactured or designed. The Oklahoma Supreme Court has identified three elements to a products liability claim: the defect must have (1) caused the injury in question, (2) existed at the time it left the manufacturer’s control, and (3) made the product unreasonably dangerous. Kirkland v. Gen. Motors Corp., 521 P.2d 1353, 1363 (Okla.1974) (adopting § 402A of the Restatement (Second) of Torts).

The defect can stem from either a dangerous design or an inadequate warning about the product’s dangers. Cincinnati argues Braswell’s design claim fails because the press brake was not in its original condition when Braswell was injured. As originally built, the press brake was wired with two dual-palm stations and two gated footswitches, either of which, had they been used, would have prevented Braswell’s injuries.

Braswell, by contrast, contends that subsequent alterations, such as removing safety devices, do not insulate a manufacturer from liability if the alteration was reasonably foreseeable — as the replacement of the footswitch and the disabling of the palm stations here supposedly were. Consequently, Braswell argues, the product was still defective. For defective design claims, Oklahoma law exempts a manufacturer from liability if “modifications or alterations [to the product] are responsible for the defect and are the intervening and superseding cause of the injuries.” Dutsch v.

Sea Ray Boats, Inc., 845 P.2d 187, 191 (Okla.1992) (emphasis added). That is, a subsequent alteration can sever proximate cause. But, on the other hand, modifications that are reasonably foreseeable are not supervening causes. “A supervening cause is a new, independent and efficient cause of the injury which was neither anticipated nor reasonably foreseeable.” Akin v. Mo. Pac. R.R. Co., 977 P.2d 1040, 1054-55 (Okla.1998) (emphasis added); see also Saupitty v. Yazoo Mfg.

Co., 726 F.2d 657, 659 (10th Cir.1984) (noting that a manufacturer is liable “if the subsequent modification was foreseeable” but not “when an unforeseeable subsequent modification alone causes the plaintiffs injury”). The question of subsequent alteration of a product often collapses elements one and two of a products liability claim— causation and existence of defect at the time of manufacture. This is because if a subsequent alteration created the defect then, by extension, the defect did not exist at the time it left the manufacturer’s control.

But if the alteration was reasonably foreseeable then the product was still defective when originally sold — in the sense that it could be altered into a dangerous state. At the same time, the question of proximate causation also turns on the reasonable foreseeability of the alteration. Given that most complex products can be made dangerous by modification, courts have sought to define foreseeability to avoid imposing a form of absolute or near-absolute liability on the original manufacturer, especially when safety devices are modified or removed.

There is no Oklahoma case law on what constitutes a reasonably foreseeable modification — let alone case law on how to apply such a rule to the removal of safety devices. The question is especially important in this context because a manufacturer may not expect safety features to be disabled or removed even if there is an incentive for a user to do so. In a sense, any safety feature can be removed or disabled, even the ones proposed by Braswell in this case.

Thus, foreseeability must limit the scope of a manufacturer’s liability- Without Oklahoma law to guide us, we could turn to other jurisdictions that have considered a reasonable foreseeability rule. Some of those eases suggest we look to a range of factors when evaluating the removal of a safety device: For example, (1) the ease with which a safety device can be removed; (2) the benefits of removing a safety device to the machine’s operation; (3) the obviousness of the danger once the safety feature was removed; and (4) the existence of warnings against making the modifications. See, e.g., Hood v. Ryobi Am.

Corp., 181 F.3d 608, 612 (4th Cir.1999) (applying Maryland law and holding that removing blade guards when warnings cautioned against doing so was not reasonably foreseeable); Perez v. Sunbelt Rentals, Inc., 360 Ill.Dec. 329, 968 N.E.2d 1082, 1085 (Ill.App.Ct.2012) (looking to fact that “guard gate was easily removed and hindered the use of the lift”); Johnson v. Black & Decker U.S., Inc., 701 So.2d 1360, 1365 (La.Ct.App.1997) (use of blade while guard was removed was not reasonably foreseeable when instruction manual warned against it); Davis v.

Berwind Corp., 547 Pa. 260, 690 A.2d 186, 190-91 (1997) (removal of safety device not reasonably foreseeable when instruction manual warned against it). This approach has much to offer in channeling legitimate design defect claims. But rather than speculate on what precise formulation (if any) the Oklahoma Supreme Court would adopt, we have an alternate basis, rooted in Oklahoma precedent, that resolves this case. As noted above, a plaintiff must satisfy three elements to bring a products liability claim. We can dispose of this case on the third element: unreasonable dangerousness.

Under Oklahoma law, a product is unreasonably dangerous if it poses a danger “beyond that which would be contemplated by the ordinary consumer who purchases it, with the ordinary knowledge common to the community as to its characteristics.” Woods v. Fruehauf Trailer Corp., 765 P.2d 770, 774 (Okla.1988). And, considering the ordinary operator of Cincinnati’s press brake, we conclude the machine was not unreasonably dangerous. As explained by Oklahoma courts, merely because a product could be made “safer” does not mean it posed an unreasonable danger to the ordinary consumer who used it.

Id. at 775. “[E]ven where a product’s design defect makes the product unreasonably dangerous, Oklahoma law does not impose liability if the product contains a warning that adequately addresses the known risks of use.” McPhail v. Deere & Co., 529 F.3d 947, 958 (10th Cir.2008); see also Treadway v.

Uniroyal Tire Co., 766 P.2d 938, 947 n. 14 (Okla.1988) (“Where warning is given, the seller may reasonably assume that it will be read and heeded; and a product bearing such a warning, which is safe for use if it is followed, is not in defective condition, nor is it unreasonably dangerous.” (emphasis omitted) (quoting Restatement (Second) of Torts § 402A emt. j. (1965))). In its seminal products liability case, Kirkland, the Oklahoma Supreme Court explicitly adopted the test from the Restatement (Second) of Torts. 521 P.2d at 1362-63.

The Second Restatement’s test for whether a product is unreasonably dangerous became known as the consumer expectations test. See Restatement (Second) of Torts § 402A, cmt. i. The test requires that the product “be dangerous to an extent beyond that which would be contemplated by the ordinary consumer who purchases it, with the ordinary knowledge common to the community as to its characteristics.” Id. Unlike a negligence standard — which looks to the reasonable precautions taken by the manufacturer— the Restatement’s strict liability test considers only the product’s features and the user’s knowledge.

And, by pegging the standard to the expectations of the ordinary consumer, the test was understood by many courts to excuse manufacturers from liability for design defects where the design’s dangers are known or obvious — either because the ordinary user has expert knowledge, and thus appreciates the product’s risks, or because the manufacturer unambiguously warned of the danger. See David G. Owen, M. Stuart Madden & Mary J.

Davis, Madden & Owen on Products Liability § 5:6 (2013) (“[B]ecause a person’s safety expectations will almost never be frustrated by an obvious product danger, the expectancy test plainly precludes liability in most such cases.”). Thus, under this approach a manufacturer would not have to correct a highly dangerous feature, however effortless or economical the fix, as long as the peril was obvious to the normal user. As a result of this limitation on liability, the consumer expectations test quickly drew criticism.

Many commentators believed a manufacturer should not be excused from the duty to eliminate a defect merely because it was obvious or because the manufacturer warned against it; ordinary human carelessness would still result in injuries despite numerous warnings or the presence of conspicuous danger. See, e.g., W. Page Keeton, Products Liability— Design Hazards and the Meaning of Defect, 10 Cumb.

L.Rev. 293, 302 (1979) (faulting consumer expectations test because “a victim could never recover for harm suffered as a result of a design hazard that was open and obvious or one with respect to which the purchaser was adequately informed”); Dan B. Dobbs, Paul T. Hayden & Ellen M. Bublick, Dobbs’ Law of Torts § 455 (2013) (same). Others have found the test unworkable for complex or constantly evolving designs.

For example, “consumers comprehend that automobiles are not completely crashproof, but they have no meaningful expectations as to the extent to which a vehicle may or may not be compromised in the event of a collision at highway speeds.” Madden & Owen on Products Liability § 8:3. Commentators proposed an alternate rule, known as the risk-utility test, under which manufacturers had a duty to eliminate dangers where an alternative design could do so without imposing unreasonable costs or impairing the functionality of the product. See e.g., John W.

Wade, On the Nature of Strict Torts Liability for Products, 44 Miss. L.J. 825 (1973) (proposing the weighing of the costs and benefits of a design feature); David Owen, Defectiveness Restated: Exploding the “Strict” Products Liability Myth, 1996 U. Ill. L.Rev. 743, 754 (“[T]he goal of both design engineers and the law should be to promote in products an ideal balance of product usefulness, cost, and safety.”). But consumer expectations could help determine the costs and benefits of an alternative design even if such evidence was no longer necessary.

In response to this criticism, the Third Restatement of Torts, released in 1998, explicitly jettisoned the consumer expectations test and adopted the risk-utility test. See Restatement (Third) of Torts: Products Liability § 2 emt. g(1998) (noting that “consumer expectations do not play a determinative role in determining defectiveness” as “[they], standing alone, do not take into account whether the proposed alternative design could be implemented at reasonable cost, or whether an alternative design would provide greater overall safety”).

The touchstone of a products liability claim would be a reasonable alternative design not consumer expectations. “A broad range of factors may be considered in determining whether an alternative design is reasonable and whether its omission renders a product not reasonably safe.” Id. cmt. f. A number of jurisdictions have either adopted the risk-utility test or use it in conjunction with the consumer expectations test. See, e.g., Branham v. Ford Motor Co., 390 S.C. 203, 701 S.E.2d 5, 16-17 (2010) (remanding case for retrial using “risk-utility test and not the consumer expectations test”); Potter v.

Chicago Pneumatic Tool Co., 241 Conn. 199, 694 A.2d 1319, 1333 (1997) (“We find persuasive the reasoning of those jurisdictions that have modified their formulation of the consumer expectation test by incorporating risk-utility factors into the ordinary consumer expectation analysis.”). Nevertheless, there is no sign that Oklahoma has backed away from the consumer expectations test since the release of the Third Restatement in 1998. See, e.g., K.M. ex rel. Arnold v.

Steger Lumber Co. of Durant, 296 P.3d 517, 519 n. 5 (Okla.Civ.App.2012) (noting that test was whether product was “dangerous to an extent beyond that which would be contemplated by the ordinary consumer”); see also McPhail, 529 F.3d at 958 (noting that consumer expectations test is the controlling rule in Oklahoma); Clark v. Mazda Motor Corp., 68 P.3d 207, 209 n. 4 (Okla.2003) (same). While we can see many of the merits of the risk-utility test, we are bound to interpret Oklahoma law in accordance with the opinions of its highest court.

And without any indication that the Oklahoma Supreme Court is inclined to adopt the risk-utility test, we will continue to apply the consumer expectations test to design defect cases such as this one. Previously, in applying Oklahoma’s consumer expectations test, we have evaluated the dangerousness of a product from the perspective of the typical user. In Wheeler v.

HO Sports Inc., 232 F.3d 754 (10th Cir.2000), for example, we held that a life vest was not defectively designed when it included only 7.1 pounds of floatation, rather than the 10 pounds necessary to keep afloat an average person, because the vest “was designed for use by experienced, skilled wakeboarders,” who are “willing to forego some degree of floatation for the sake of enhanced mobility.” Id. at 758. In Gaines-Tabb v.

ICI Explosives, USA Inc., 160 F.3d 613 (10th Cir.1998), we held that a fertilizer formulation was not unreasonably dangerous — notwithstanding the fact that an alternate formula, equally effective as a fertilizer, would have reduced the risk of explosion — because the ordinary user was a farmer who would not be expected to use it as an explosive. Here, we apply an objective test. The ordinary consumer (or user) of a press brake is a trained operator.

That individual will be trained as to its proper operation and how to use any available safety features, such as palm stations and foot-switches, and to safely remove pieces of metal that become jammed in the machine. Specifically, the ordinary user would be aware of the extreme danger and risk of reaching into the machine while having one’s foot remain anywhere near the foot-switch, at least without disengaging or blocking the ram. There has been no evidence presented that the press brake, if properly operated, poses a danger beyond that which an ordinary user would expect.

The fact that Braswell may have been improperly trained in the use of the machine — as he insists he was — may be the fault of Ventaire, but it does not change the scope of Cincinnati’s liability. See Woods, 765 P.2d at 774 (dangerousness evaluated according to “ordinary knowledge common to the community” that uses the product). The ordinary operator of a heavy machine tool would be properly trained. The ordinary operator of a press brake would also heed the warnings on the machine and in the instruction manual. See Hutchins v.

Silicone Specialties, Inc., 881 P.2d 64, 67 (Okla.1993) (manufacturer not required to foresee that professional users of product would ignore warnings and then use product in exact manner warned against). The warnings on the press brake were sufficient to make the machine not unreasonably dangerous, as it covered all the salient dangers accompanying its operation. See McPhail, 529 F.3d at 958. A sign warned that the footswitch could not protect the operator from injury, and that the operator’s fingers or hands could be crushed.

Another sign warned the operator not to place his hands in the machine and to read the instruction manual before operating the machine. A Cincinnati representative testified that the press brake’s manual warns that if, for some reason, an operator needs to place his hands in the die area, then he should use a block to prevent the ram from descending or turn off the operator control. (It appears this manual was chained to the machine for any operator to look through.) Had these instructions been followed, Braswell would not have been injured.

While the warnings, admittedly, did not cover the exact scenario presented here— the danger of accidental activation via the footswitch while removing a jammed piece of sheet metal — our prior cases do not require such granular specificity for warnings. This is especially true when the ordinary user would be aware of the specific danger. See Duane v. Okla. Gas & Elec.

Co., 833 P.2d 284, 286 (Okla.1992) (“[Tjhere is no duty on a manufacturer or seller to warn of a product-connected danger which is obvious or generally known, and there is no duty to warn a knowledgeable user of the product of the dangers associated therewith.”); Daniel v. Ben E. Keith Co., 97 F.3d 1329, 1334 (10th Cir.1996) (no need to provide “warning labels to identify so specifically the consequences of misusing a product” when danger of “mixing all-purpose bleach with cooking oil and boiling water” was apparent “to an ordinary user from the nature of the product”).

It is undisputed that Braswell knew of the exact danger he faced. He admitted in his deposition that when a piece is jammed “[y]ou take your foot off of the pedal” before reaching into the machine. Aplt. App. 54. Unfortunately, Braswell neglected to do just that. Nevertheless, we have expressed concern where a professional user would have to ignore a warning designed to make the product’s operation safe while a lingering unknoum danger remained. For example, in one case we considered the adequacy of a warning on a tractor that stated, “Danger. Start only from seat in park or neutral.

Starting in gear kills.” McPhail, 529 F.3d at 960. That warning was insufficient because the manufacturer knew drivers would sometimes have to “bypass start” the tractor, which requires the driver to stand outside the tractor rather than remain seated. Id. at 959. The warning suggested that as long as the tractor was in park or in neutral, then the danger was eliminated. Id. at 960. Yet there was an additional unknown danger that the tractor could be in gear even though it appeared to be in neutral. Id.

Because the warning did not cover that common scenario, and the driver would occasionally be outside the seat upon starting (notwithstanding the warning against doing so), the warning provided was not sufficient. There are no similar concerns here. The press brake’s general warning to not place one’s hand in the die area and the manual’s instruction to use blocks sufficiently covered the scenario faced by Braswell. Unlike in McPhail, there is no evidence the press brake’s operation routinely required users to not utilize the safety device (i.e., the blocks).

And, unlike in McPhail — where the driver was unaware the tractor could be in gear when it appeared to be in neutral — Braswell was not ignorant as to the particular danger he faced. While Braswell insists he had to ignore the warnings to remove a piece of jammed metal, the record demonstrates this conduct was a result of either poor training or carelessness rather than necessity. Indeed, a Ventaire employee testified to training Braswell to use blocks to hold up the ram whenever he reached into the die area.

Another Ventaire employee testified that in his two and a half years operating the press brake he never once had to place his hand directly in the die area. If a piece of metal was inserted too far into the machine, he would pull it out from the front, rather than reach into the machine to adjust it from the back. While Braswell presents conflicting testimony that Ventaire never trained him on how to use the blocks, nor ensured they were available to the press brake’s operators, he does not dispute that Cincinnati’s manual advised using them.

In short, the ordinary operator of the press brake would have known of the machine’s dangers and how to avoid them. That Braswell may not have taken the correct precautions to avoid those dangers, or that Ventaire may have been negligent, is no fault of Cincinnati’s. The warnings, if followed, made operating the machine reasonably safe. Finally, Cincinnati’s inclusion of the gated footswiteh in its original model provides us further assurance that the press brake was not unreasonably dangerous.

Accidental depression of the footswiteh is a real danger, which is why Cincinnati included a gated footswiteh in the first place (among myriad other safety devices). With that feature, the ram would not descend unless an operator purposefully lifted the gate with his foot. Braswell contends the gated footswiteh would not have prevented his injury because he never testified to removing his foot back from the pedal — such that a gate, had it existed, would have fallen and prevented his foot from accidentally returning to the pedal area.

Only an anti-trip footswiteh, Bras-well contends, would have prevented his injury. The district court concluded otherwise, explaining that Braswell’s testimony “merely presents a metaphysical doubt as to the material facts rather than a genuine issue.” ApltApp. 408 (internal quotation marks omitted). But we need not answer the question of causation because Cincinnati had no duty to guard against the danger that a press brake operator would leave his foot on the pedal while removing a piece of jammed metal.

An ordinary user of a press brake would know not to have his foot on the footswiteh when he did not want the ram to descend, just as, say, an ordinary user of a gun would know not to have the safety off when he did not want the gun to fire. See Smith v. Ron’s Guns, Inc., No. MMXCV116004200, 2013 WL 453060, at *5 (Conn.Super.Ct. Jan. 8, 2013) (“[A]n average handgun owner would clearly be aware that if they carried a loaded gun, the gun could fire if its safety mechanism was not employed.”). And, once he removed his foot, the ordinary user would be protected from accidental depression by the gate.

That was the only protection Cincinnati needed to provide in this regard. While Braswell insists that Cincinnati is nevertheless liable because the gated footswiteh could be replaced by one without a gate, we are convinced that an otherwise safe product is not made unreasonably dangerous if the manufacturer fails to prevent the replacement of a part with a substandard aftermarket part. See, e.g., Baughman v. Gen.

Motors Corp., 780 F.2d 1131, 1133 (4th Cir.1986) (applying South Carolina law to hold that car manufacturer had no duty to “test[] and warn[ ] against any of a myriad of replacement parts supplied by any number of manufacturers”); Hoyt v. Wood/Chuck Chipper Corp., 651 So.2d 1344, 1352 (La.Ct.App.1995) (“A manufacturer’s duty to anticipate that users might replace certain worn out parts on its product does not include the risks occasioned by the use of improper or substandard replacement parts.”); Braaten v.

Saberhagen Holdings, 165 Wash.2d 373, 198 P.3d 493, 495-500 (2008) (no duty to warn about danger of asbestos-containing replacement parts manufactured by third-party). The ordinary user of a press brake would be aware of the risk of accidental depression, and the inclusion of a gated footswitch protects those users from all but the most careless of accidents. Accordingly, the press brake was not unreasonably dangerous. We agree with the district court’s grant of Cincinnati’s motion for

summary judgment on the design defect claim. And, by extension, we agree that Braswell’s inadequate warning claim fails too, as the warnings provided here apprised an ordinary user of the relevant dangers. B. Negligence Claim Braswell also contends the district court erred in granting

summary judgment on his negligence claim. He contends

(1) Cincinnati violated its duty of care to design a reasonably safe press brake, and (2) the service technician’s visit imposed a duty on Cincinnati to ensure the machine was being operated safely. We decline to reach these arguments because both have been waived. “An issue is waived if it was not raised below in the district court.” Wilburn v. Mid-South Health Dev., Inc., 343 F.3d 1274, 1280 (10th Cir.2003).

This rule, however, “is not inflexible and the matter of what questions may be taken up and resolved for the first time on appeal is one left primarily to the discretion of the courts of appeals, to be exercised on the facts of individual cases.” Forest Guardians v. U.S. Forest Serv., 495 F.3d 1162, 1170 n. 7 (10th Cir.2007) (quoting Anixter v. Home-Stake Prod. Co., 77 F.3d 1215, 1229 (10th Cir.1996)). In Braswell’s original complaint, he included an allegation that Cincinnati “negligently designed and manufactured” the press brake. ApltApp. 14. When Cincinnati moved for

summary judgment, however, there was no mention of the negligence claim, either in Cincinnati’s opening or reply briefs or in Braswell’s response. Nevertheless, the district court, in a footnote, addressed Braswell’s negligence theory, explaining that “[t]he manufacturer’s liability test of Kirkland encompasses defects alleged to have been caused by negligence.” Aplt.App. 406 n. 2 (citing Kirkland, 521 P.2d at 1365 (“[W]e believe that a defect ... would also include a defect caused by some form of negligence. The negligence action for products liability may thus be rendered unnecessary.”)). The district court entered

summary judgment for Cincinnati. Braswell filed a motion to reconsider, which the district court denied in a minute order. In the motion, Braswell took issue with the court’s resolution of his negligence claim. While acknowledging the district court’s conclusion that negligent manufacturing was captured under the standard product liability test, Braswell argued that he had made out an independent negligence claim based on the Cincinnati service technician’s visit in 2007 to Yentaire to replace a computer chip in the press brake. Yet this was an entirely new claim. See Muskrat v. Deer Creek Pub. Sch., 715 F.3d 775, 791 (10th Cir.2013) (“New liability theories after

summary judgment are discouraged.”). Braswell’s complaint merely alleges that the press brake had been “negligently designed and manufactured,” and mentions nothing about a duty of care arising from a technician’s visit. ApltApp. 14. Braswell did mention the technician’s visit in his response to the motion for

summary judgment, but never explained how it related to any theory of liability (negligence or otherwise). Unguided by the briefs, the district court analyzed the visit in relation to the products liability claim that was being disputed, and concluded the technician’s “work on the machine’s software did not impose a duty upon Cincinnati to ensure that material modifications to safety features had not been made.” ApltApp. 409.

Braswell’s attempt to repackage this fact in a new argument — that the technician’s visit imposed on Cincinnati an ordinary duty of care — in a motion to reconsider was not sufficient to preserve it for appeal. See Anderson v. Unisys Corp., 52 F.3d 764, 765 (8th Cir.1995) (declining to rule on argument raised for the first time in motion for reconsideration when district court did not address the argument). Braswell has also not sufficiently pursued his negligent design and manufacture claim. Cincinnati in its motion for

summary judgment did not mention the negligence claim, and neither did Braswell in his response. While the nonmoving party generally “need only respond to the arguments asserted,” there must be some evidence in the record that it pursued the claim prior to raising it in a motion for reconsideration. Muskrat, 715 F.3d at 790. The complaint included the phrase “negligently designed and manufactured,” Aplt.App. 14, but Braswell gave no indication that he viewed this claim as separate from the standard products liability claim, cf.

Kirkland, 521 P.2d at 1365 (“[W]e believe that a defect ... would also include a defect caused by some form of negligence.”). In the absence of any discussion to the contrary in the briefs, the district court reasonably assumed (given the state of Oklahoma law) that the negligent design claim was subsumed under the products liability claim. Indeed, Braswell conceded as much in his motion to reconsider.

See ApltApp. 427 (acknowledging district court’s dismissal of negligent manufacture claim but arguing negligence claim was actually “with regard to the maintenance and service aspect of [technician’s] contact with the press brake”). The mere fact that Braswell included a throwaway reference to negligence in his complaint does not — without any other evidence in the record showing he pursued the claim below — entitle him to appellate review of the claim. Accordingly, we decline to review any of Braswell’s negligence arguments. III.

Conclusion Because the press brake was not unreasonably dangerous, Braswell’s products liability claims fail. And Braswell has waived any arguments concerning his negligence claim. Accordingly, we AFFIRM the district court’s judgment. . The only Oklahoma Supreme Court case we have been able to locate that relies on the alternative test, the risk-utility test, concerns the affirmative defense of an “unavoidably unsafe” product. See Tansy v. Dacomed Corp., 890 P.2d 881, 885-87 (Okla.1994); see also Restatement (Second) of Torts § 402A cmt. k.

But this test is not used by the plaintiff to prove that an alternative, safer design was feasible, but by the defendant to negate liability in light of the product’s benefits and the infeasibility of making it safer. See Tansy, 890 P.2d at 886. That is, the risk-utility test in Tansy is used as a defense to an otherwise valid design defect claim, and not a test for liability in the first instance. . Replacing parts necessary to a machine’s operation is a separate question from removing its safety features altogether, the foreseeability of which we need not address. .

Because Oklahoma has just one test for whether a product is unreasonably dangerous — the consumer expectations test — most cases treat design defects and inadequate warnings as merely different methods of proof for the same products liability claim. As a result, warnings sufficient to counteract an otherwise dangerous design usually satisfy the manufacturer’s duty to warn, as they do here. See Hutchins, 881 P.2d at 67 (warnings on water-proofing primer made product not unreasonably dangerous even though "other primers containing non-flammable ingredients were available on the market”); Smith v. U.S.

Gypsum Co., 612 P.2d 251, 255 (Okla.1980) (approving a jury instruction where design defect was defined as a dangerous feature "not readily apparent to the eye” which could be cured with an adequate warning); see also McPhail, 529 F.3d at 958 ("[E]ven where a product’s design defect makes the product unreasonably dangerous, Oklahoma law does not impose liability if the product contains a warning that adequately addresses the known risks of use.”). There are a few authorities that suggest otherwise, but they appear to rely on the risk-utility test, which, as already noted, is not the governing rule in Oklahoma.

Compare Clark, 68 P.3d at 209 n. 4 (acknowledging consumer expectations test), with Steele ex rel. Steele v. Daisy Mfg. Co., 743 P.2d 1107 (Okla.Civ.App.1987) (though warnings for air rifle were sufficient, design defect question remained because it may "have been economically feasible to have incorporated an automatic trigger safety device on the rifle”), and Vicki L.

Mac-Dougall, Oklahoma Product Liability Law § 8:1 ("If a feasible alternative to the design existed that would have avoided the risk of harm, the manufacturer would be liable for the injures caused by the design defect even if adequate warnings accompanied the sale of the product.”). . Notwithstanding Kirkland, Oklahoma has not eliminated a freestanding negligence claim for defective products. See Honeywell v.

GADA Builders, Inc., 271 P.3d 88, 96 (Okla.Civ.App.2011) ("Even with the advent of strict products liability, the negligence cause of action remains available to a plaintiff injured by a defective product.”); Cochran v. Buddy Spencer Mobile Homes, Inc., 618 P.2d 947, 949 (Okla.Civ.App.1980) ("Kirkland does not stand for the proposition that the only recourse a plaintiff has when a defective product causes injury to him or his property is an action in products liability.”).

As a practical matter, however, it is highly unlikely a plaintiff will prevail on a negligence claim when he cannot do so on a products liability claim.