¶ 1. This is a review of a published decision of the court of appeals affirming the judgment of the Washington County Circuit Court, Annette K. Ziegler, Judge.1 After a jury trial, the circuit court rejected plaintiffs' negligence and strict products liability personal injury claims. Plaintiffs moved for a new trial, challenging the accuracy of the jury instructions on the strict products liability claim. The circuit court concluded that the jury was properly instructed and denied the motion for a new trial. The court of appeals affirmed.
¶ 2. The jury instructions were based on Wisconsin Jury Instruction — Civil 3260 with a supplemental statement regarding bystander claims. The jury was informed that a bystander personal injury claim in strict products liability is only available if the product is unreasonably dangerous based on the expectations of an ordinary user or consumer (the "consumer contemplation test"). Plaintiffs claim that this jury instruction was an incorrect statement of the law. They contend *151that when a product is dangerous only to a bystander and not to a user or consumer, the consumer contemplation test is inappropriate. Rather, the jury should be instructed that a product is unreasonably dangerous based on the contemplation and expectations of an ordinary bystander. They call this a "bystander contemplation test," and assert that this is and should be the law in Wisconsin.
¶ 3. Thus, the main question before us is whether Wisconsin has adopted or should adopt a "bystander contemplation test." If the bystander contemplation test is the law, we must determine whether the circuit court's jury instructions were a misstatement of the law, and if so, whether the error was prejudicial.
¶ 4. We hold that the consumer contemplation test, and not a bystander contemplation test, governs all strict products liability claims in Wisconsin, including cases where a bystander is injured. While bystanders may recover when injured by an unreasonably dangerous product, the determination of whether the product is unreasonably dangerous is based on the expectations of the ordinary consumer.2 Therefore, the jury was properly instructed, and the decision of the court of appeals is affirmed.
I. BACKGROUND
¶ 5. The facts of this case are horrific. On the afternoon of May 2, 2004, the Horst family returned *152home from an overnight trip to Wisconsin Dells. Two-year-old Jonathan and his older brother went to play outside in the yard. Jonathan's mother, Kara, was planning to watch Jonathan as she hung laundry on an outdoor clothesline, but stopped to use the restroom first. Before Kara arrived outside, Jonathan's father Michael decided to mow the lawn using their John Deere LT160 riding lawn mower. As Michael began to cut the lawn, he decided to mow in reverse along the rear of the house, looking over his right shoulder. Jonathan, however, had moved behind the lawn mower to Michael's left, out of Michael's line of sight. As Michael proceeded backwards, he saw Jonathan's shoe come out the other side. Michael screamed, realizing that he had severed both of Jonathan's feet. Kara called 911, and Jonathan was flown to Children's Hospital. There he received multiple surgeries, and now wears prosthetics on both legs.
¶ 6. The John Deere LT160 mower Michael was using came equipped with a no-mow-in-reverse safety feature that stops both the engine and mower blades when an operator begins to travel in reverse while the mower blades are engaged. However, the lawn mower also had what amounts to an override feature, the Reverse Implement Option ("RIO"), which allows an operator to mow in reverse with the mower blades in operation.
¶ 7. To implement the RIO feature, an operator must depress the brake pedal and press the RIO switch. Once engaged, the RIO system allows an operator to mow in reverse without stalling either the engine or the mowing device. When reverse mowing is complete, the operator can continue to mow forward without shutting off the mowing device. When the operator begins mowing forward again, the lawn mower returns to its *153default position, which requires the operator to manually engage the RIO device again to mow in reverse.
¶ 8. Michael Horst engaged the RIO device twice before the accident in this case. He first engaged the RIO to mow toward the Horst home along his gravel driveway. He then moved forward along the back of the house. He engaged the RIO device again to mow in reverse along the back of the house. That is when Jonathan was injured.
¶ 9. The LT160 lawn mower operator's manual contained numerous warnings relating to mowing in reverse and mowing in the presence of children or bystanders. The warnings included the following:
• Before backing up, stop mower blades or attachments and look down and behind the machine carefully, especially for children.
• CAUTION: Avoid injury! Children or bystanders may be injured by runover [sic] and rotating blades. Before backing up, carefully check the area around the machine.
• NOTE: Backing up while the mower is engaged is strongly discouraged.
• The Reverse Implement Option should be used only when operating another implement (attachment) or when the operator deems it necessary to reposition the machine with the mower engaged.
The parties agree that Michael read but disregarded these warnings, choosing to mow in reverse in the presence of his young children.
¶ 10. Following the accident, the Horsts filed a lawsuit against Deere & Company ("Deere") in Washington County Circuit Court, bringing negligence and strict products liability claims. On the strict products *154liability claim, the Horsts argued that designing a mower to operate in reverse is unreasonably dangerous and that the mower should have had an alternative design. The Horsts asserted that the lawn mower should not have been equipped with the RIO, thus preventing an operator from ever mowing in reverse. The Horsts also sought punitive damages, alleging that the design demonstrated a deliberate disregard for safety.
¶ 11. Deere moved for summary judgment on the grounds that the "consumer contemplation test" barred Jonathan's strict products liability claim. The circuit court denied the motion for summary judgment. It concluded that while bystanders injured by unreasonably dangerous products may recover under Howes v. Hansen, 56 Wis. 2d 247, 201 N.W.2d 825 (1972) (hereafter "Howes I"), the question of whether the product was unreasonably dangerous and whether punitive damages should be awarded were issues of fact for the jury.
¶ 12. At trial, the Horsts requested that Wisconsin Jury Instruction — Civil 3260, which does not mention bystanders, be supplemented to reflect the availability of recovery for bystanders. They specifically proposed that the instruction include the phrase "or bystander" following most occurrences of "user" and "consumer" in the standard instruction. The circuit court denied the Horsts' proposed instructions, choosing to give the standard instructions supplemented with the following statement: "The law in Wisconsin imposes a duty on a manufacturer to a bystander, if the bystander is injured by a defective product, which is unreasonably dangerous to the ordinary user or consumer."
*155¶ 13. The Horsts also requested a special verdict question asking the jury: "Do you find from the evidence that the subject lawn tractor, when it left the hands of Defendant, Deere & Company, was in a defective condition so as to be unreasonably dangerous to a prospective user/consumer or bystander?" The circuit court denied this request and submitted the question to the jury without the "or bystander" language.3
*156¶ 14. The jury ultimately found both Michael and Kara Horst, but not Deere, negligent in the injury to their son, Jonathan. The jury also found that the lawn mower in question was not in a defective condition so as to be unreasonably dangerous to a prospective user or consumer. Accordingly, Deere was not strictly liable for Jonathan's injuries.
¶ 15. After the jury verdict, the Horsts moved for a new trial on the grounds that the jury was improperly instructed. The circuit court concluded that the instruction was in accord with the facts and existing case law, and thus was not prejudicial. The circuit court therefore denied the motion for a new trial and dismissed the Horsts' claims with prejudice.
¶ 16. The Horsts filed a notice of appeal with the court of appeals, and also filed a petition to bypass with this court, which was denied. On its review, the court of appeals agreed with the circuit court, concluding that the consumer contemplation test is the proper test for unreasonably dangerous products, and that the jury instructions constituted an accurate statement of the *157law. Horst v. Deere & Co., 2008 WI App 65, ¶ 20, 312 Wis. 2d 421, 752 N.W.2d 406. The Horsts then petitioned this court for review.
II. STANDARD OF REVIEW
¶ 17. This case asks us to evaluate the sufficiency of the circuit court's jury instructions. Generally, "a trial court has broad discretion when instructing a jury." White v. Leeder, 149 Wis. 2d 948, 954, 440 N.W.2d 557 (1989). If the jury instructions fully and fairly explain the relevant law, there are no grounds for reversal. Green v. Smith & Nephew AHP, Inc., 2001 WI 109, ¶ 25, 245 Wis. 2d 772, 629 N.W.2d 727. The question of whether the jury instructions accurately state the law is a question of law, which we review de novo. Id.
¶ 18. If the jury instructions were an erroneous statement of the law, a new trial will be ordered only if the court's error "affected the substantial rights of the party." Wis. Stat. § 805.18(2) (2007-08). An error affects the substantial rights of the party if it undermines confidence in the outcome. State v. Dyess, 124 Wis. 2d 525, 544-45, 370 N.W.2d 222 (1985). An error undermines confidence in the outcome if there is a reasonable probability4 the outcome would have been different but for the error. Id. at 544.
*158III. DISCUSSION
¶ 19. This case is a dispute over both what the law is — that is, what Wisconsin courts have said — and over what the law should be. The parties here dispute the meaning and relevance of several cases, and underlying this, disagree about the proper scope of strict products liability law in Wisconsin.
¶ 20. To address these questions, we first, in subsection A (¶¶ 21-31), briefly introduce the development of and theory behind strict products liability and the consumer contemplation test. In subsection B (¶¶ 32-35), we explain the proposed bystander contemplation test as proffered by the Horsts. In subsection C (¶¶ 36-67), we review the significant cases relied on by the parties. Finally, in subsection D (¶¶ 68-81), we analyze and answer the ultimate question before us— whether the consumer contemplation test is the proper standard for determining whether a product is unreasonably dangerous when a bystander is injured.
A. Strict Products Liability and The Consumer Contemplation Test
¶ 21. Historically, with the exception of the sale of food, a supplier of a product was generally not liable for injuries caused by that product without a showing of negligence or privity of contract. See Restatement (Second) of Torts § 402A (hereafter "§ 402A") cmt. h (1965) .5 This began to change in the 1950s and 1960s as courts developed theories of liability, often based on warranty-like concepts, to hold manufacturers or sellers liable for *159injuries even without negligence or privity of contract. See id. (describing some of the early decisions as displaying "considerable ingenuity in evolving more or less fictitious theories of liability to fit the case"). There was, in short, a growing recognition that consumers needed protection from defective products that caused harm to consumers or their property. See id.
¶ 22. In the early 1960s, during this same time, the American Law Institute was drafting what became the Restatement (Second) of Torts. The Restatement (Second) attempted to capture this emerging line of cases by creating a new category of tort claims — strict products liability — which it set forth in the newly created § 402A. Although strict products liability was still in its intellectual infancy, § 402A was remarkably influential in speeding the adoption of this emerging area of law in courts around the country. See Douglas A. Kysar, The Expectations of Consumers, 103 Colum. L. Rev. 1700, 1711 (2003); George W Conk, Punctuated Equilibrium: Why Section 402A Flourished and the Third Restatement Languished, 26 Rev. Litig. 799, 808-09 (2007). We joined this trend and adopted § 402A in 1967 in Dippel v. Sciano, 37 Wis. 2d 443, 155 N.W.2d 55 (1967).6
*160¶ 23. This strict products liability structure, whereby a manufacturer bears the costs for injuries resulting from product use, even when the manufacturer was not negligent, arose for at least three important policy reasons.
¶ 24. First, strict products liability serves as a cost shifter. See § 402A cmt. c. It takes the usually overwhelming cost of injury off of the injured person and places it on the manufacturer. The manufacturer generally passes the costs for injuries and preventative safety measures on to all consumers through higher product prices. This liability system, then, spreads the cost of the injury risk to all consumers. Id. The justification relied upon by courts is that companies have the capacity to bear the costs and the ability to assume them more efficiently than individuals.
¶ 25. A second rationale underpinning strict products liability is fundamental fairness to the injured person. Id. If manufacturers can reasonably design a safer product or a product that better accords with the safety expectations of consumers but choose not to do so, they should be held liable for the resultant injuries.
*161¶ 26. Finally, a third reason for strict products liability is that it provides a strong incentive for deterrence. Restatement (Third) of Torts: Products Liability (hereafter "Restatement (Third)") § 2 cmt. a (1998). When a manufacturer can reasonably prevent an injury, strict products liability gives them a strong incentive to do so. This litigation threat promotes manufacturer investment in safer designs, quality control, and the furnishing of adequate warnings to the purchasers and users of products. Id.; § 402A cmt. j.
¶ 27. Strict products liability is not, however, absolute liability. Howes I, 56 Wis. 2d at 253. We do not want to hold businesses liable for every injury involving their products. Such an approach would eliminate some useful products from the market that cannot possibly be made completely safe (such as knives, guns, medicine, and even sugar in the case of a diabetic). § 402A cmts. i, k. Hence, all of the strict products liability analytical frameworks — including a risk-utility analysis7 and our own focus on consumer expectations — have at least a partial grounding in the necessity of guarding against absolute liability.
¶ 28. Section 402A describes what has been called the "consumer contemplation test" for determining whether a product is unreasonably dangerous. Comment g explains that a manufacturer is strictly liable "only where the product is, at the time it leaves the seller's hands, in a condition not contemplated by the ultimate consumer, which will be unreasonably dangerous to *162him." § 402A cmt. g. Similarly, comment i states that the product "must be dangerous to an extent beyond which would be contemplated by the ordinary consumer who purchases it, with the ordinary knowledge common to the community as to its characteristics." Id. cmt. i. This standard requires manufacturers to anticipate what consumers will expect, and to take safety precautions in accordance with those expectations. Those safety precautions might include giving adequate instructions and warnings (see id. cmts. h and j) and/or implementing reasonably available safety features.
¶ 29. One of the implications of the consumer contemplation test is that consumers can and do contemplate open and obvious dangers, and are not protected when injured by such dangers. See id. cmt. j (warnings are not required "when the danger, or potentiality of danger, is generally known and recognized"); Tanner v. Shoupe, 228 Wis. 2d 357, 367, 596 N.W.2d 805 (Ct. App. 1999) ("In order for a defective design to render a product unreasonably dangerous, the defect must be hidden from the ordinary consumer, that is, not an open and obvious defect."). Consumers are also not protected if they proceed in the face of a known, though not open and obvious, danger. See § 402A cmt. n ("If the user or consumer discovers the defect and is aware of the danger, and nevertheless proceeds unreasonably to make use of the product and is injured by it, he is barred from recovery."). Thus a manufacturer may be able to avoid liability under § 402A by making consumers aware of dangers through warnings or simply by virtue of selling a product with open and obvious dangers.
¶ 30. Section 402A is explicitly neutral as to whether persons other than users or consumers should be able to recover under strict products liability. Id. Caveat (1). Comment o explains:
*163Thus far the courts .. . have not gone beyond allowing recovery to users and consumers .... Casual bystanders, and others who may come in contact with the product, as in the case of employees of the retailer, or a passer-by injured by an exploding bottle, or a pedestrian hit by an automobile, have been denied recovery. There may be no essential reason why such plaintiffs should not be brought within the scope of the protection afforded, other than that they do not have the same reasons for expecting such protection as the consumer who buys a marketed product.
In relatively short order, however, courts around the country did extend protection to injured bystanders. See Peter Zablotsky, Eliminating Proximate Cause As An Element of the Prima Facie Case For Strict Products Liability, 45 Cath. U. L. Rev. 31, n.79 (1995). This court extended such protection in Howes I, which is discussed below.
¶ 31. No one in the case at bar disputes that bystanders may recover if a product is unreasonably dangerous. The issue in this case is the proper legal standard for determining whether a product is unreasonably dangerous when a bystander is injured.
B. The Proposed Bystander Expectations Test
¶ 32. The Horsts argue that the jury instructions in this case, which asked whether the lawn mower was unreasonably dangerous based on the expectations of the ordinary user or consumer, were incorrect as a matter of law. They maintain that the law in Wisconsin is, or should be, what they call a "bystander contemplation test."
¶ 33. The bystander contemplation test asks exactly the same question as the consumer contemplation test, but replaces the expectations of the user or con*164sumer with the expectations of an ordinary bystander. Accordingly, the Horsts assert that "when a bystander is injured by a product, the question is whether the product was as reasonably safe as an ordinary bystander would contemplate or expect, not whether the user or consumer understood and appreciated the risk." Like the consumer contemplation test, the bystander contemplation test is an objective test and not dependent on an injured party's knowledge.
¶ 34. The bystander contemplation test, according to the Horsts, applies when a bystander is injured and "where a manufacturer designs and sells a product that poses a unique risk of bodily harm to bystanders alone." They submit that the consumer contemplation test is still proper when a bystander is injured and when the danger is present for both the user or consumer and the bystander.
¶ 35. The Horsts further contend that the bystander contemplation test is not only the law, but that it is necessary to provide meaningful protection to bystanders. They argue that bystanders need greater protections than users and consumers because they have less information about the product, and less access to warnings and instructions. They also point to the language in Howes I (which extended protection to bystanders injured by unreasonably dangerous products), stating that "[t]here is no essential difference between the injured user or consumer and the injured bystander." 56 Wis. 2d at 255.
C. Prior Wisconsin Case Law
¶ 36. The Horsts argue that the bystander contemplation test is the law in Wisconsin even though it has not been formally announced in any of the cases as such. They point in particular to Howes I, Howes v. *165Deere, 71 Wis. 2d 268, 238 N.W.2d 76 (1976) (hereafter "Howes II"), and Komanekin v. Inland Truck Parts, 819 F. Supp. 802 (E.D. Wis. 1993), asserting these cases show that when a bystander is injured, and when the threat of danger is to the bystander alone, the determination of the unreasonableness of the danger is based on an ordinary bystander's expectations, not a user or consumer's expectations.8
¶ 37. Deere, on the other hand, points to the standard jury instructions as well as our prior decisions in Vincer v. Esther Williams All-Aluminum Swimming Pool Co., 69 Wis. 2d 326, 230 N.W.2d 794 (1975) and Green to argue that, while bystanders may recover if injured by an unreasonably dangerous product, the consumer contemplation test applies in all strict products liability cases.9
*166¶ 38. Therefore, we will briefly discuss the principal cases cited by the parties and the arguments offered for each.
1. Howes I10
¶ 39. Two-year-old Richard Howes II11 ("Richard") lived with his parents in Lake Geneva, Wisconsin in a two-family dwelling owned by neighbor Naomi Schatzman. Howes I, 56 Wis. 2d at 250. Ms. Schatzman delegated lawn care for the dwelling to her adult son, who in turn hired a twelve-year-old boy to mow the lawn. Id. One day, while the twelve-year-old boy was mowing with a Deere riding lawn mower, Richard came into contact with the mower blades and suffered serious injuries, including the loss of his right foot. Id. The Howes sued Deere (among others) for the child's injuries, one claim of which sounded in strict products liability. Id. Because the child was an injured bystander (i.e., not a user or consumer), the main issue in the case was whether an injured bystander can maintain a strict products liability claim against a manufacturer, or *167whether strict products liability claims may only be advanced by injured users and consumers.
¶ 40. After acknowledging that our prior decision in Dippel only allowed users and consumers to recover, the court chose to extend coverage to bystanders, holding that a manufacturer is strictly liable "when he places a defective article on the market 'that causes injury to a human being.'" Id. at 260 (quoting Greenman v. Yuba Power Prods., Inc., 377 P.2d 897, 900 (Cal. 1963)).
¶ 41. The court extended protection because the rationale supporting protection for users and consumers was equally applicable to bystanders. See Howes I, 56 Wis. 2d at 255. The court noted the reasons articulated in Dippel for protecting users and consumers, including the fairness of compensating for injury, the cost and risk-distribution effects of strict products liability, and the deterrent effect on manufacturers. Id. These policy goals support the extension of recovery to bystanders, the court reasoned, because there is "no essential difference between the injured user or consumer and the injured bystander."12 Id. However, plaintiff bystanders still needed to prove the product was unreasonably dangerous in order to recover. Id. at 258.
¶ 42. The Horsts claim that Howes I is dispositive, and suggest that we cannot legitimately reconcile a rejection of their proposed bystander contemplation test with this case.
¶ 43. Deere, on the other hand, concedes that Howes I allows an injured bystander to pursue a strict *168products liability claim, but disputes that it creates or adopts a bystander contemplation test. Deere points out that the case contains no language indicating that the question of whether a product is unreasonably dangerous in bystander cases is evaluated from the perspective of the ordinary bystander. Howes I merely holds, according to Deere, that a manufacturer may be strictly hable in tort when he places a defective article on the market that causes an injury to any human being, including bystanders. The case did not, Deere contends, modify the test for whether the product was defective/unreasonably dangerous.
¶ 44. We acknowledge the broad language Howes I occasionally uses, especially the statement that there is "no essential difference between the injured user or consumer and the injured bystander." Id. at 260. But this statement cannot be pulled out of its issue-sensitive context and used to support an entirely different legal proposition. With respect to who may recover, there truly are no good reasons to limit recovery to injured users and consumers. But with respect to determining whether a product is unreasonably dangerous, there are significant differences (outlined more fully below) between a standard based on the expectations of an ordinary consumer and a standard based on the expectations of an ordinary bystander.
¶ 45. In short, Howes I did not purport to address a proposition greater than the legal question before that court. Our holding today, rejecting a bystander contemplation test, leaves intact and indeed reaffirms the basic holding of Howes I: Bystanders injured by an unreasonably dangerous product may assert a strict products liability claim against the manufacturer or seller.
*1692. Howes II13
¶ 46. Following the decision in Howes I, all other defendants except Deere settled out of court. Howes II, 71 Wis. 2d at 270. The claims against Deere went to trial. At the end of trial, the circuit court directed the plaintiffs to elect between one of the two theories of liability for submission to the jury — either negligence or strict products liability. Id. at 271. While indicating a preference for strict products liability, the plaintiffs submitted a proposed special verdict covering both theories. Id. The circuit court rejected the dual theory verdict form, and instead submitted a final special verdict question regarding liability, which read as follows: "Was the John Deere mower in question, when it left the possession of the manufacturer, Deere & Company, defective in design so as to be unreasonably dangerous to a bystander?" Id. The jury found that Deere was not liable, and the Howes appealed. Id. at 272. After review, we reversed and remanded for a new trial, holding that plaintiffs were incorrectly forced to choose between alternative theories. Id. at 272-75.
¶ 47. The Horsts find this case especially pertinent because of the language used in the special verdict question, asking whether the product was "defective in design so as to be unreasonably dangerous to a bystander?" This special verdict question was nearly identical to the Horsts' proposed special verdict question that was denied by the trial court.
¶ 48. Deere counters that the court in Howes II did not consider or analyze the propriety of the special verdict question. Rather, the case was reversed because of the trial court's erroneous decision requiring plain*170tiffs to choose between strict products liability and negligence theories of liability.
¶ 49. In Howes II, the special verdict question was admittedly almost identical to the Horsts' proposed special verdict question. But the special verdict question was not affirmed or intentionally addressed by the court. We also do not know how the trial court instructed the jury to determine whether the product was unreasonably dangerous.14 On this point, which is the central question we address today, Howes II is no more helpful or determinative than Howes I.
3. Vincer15
¶ 50. In July of 1970, two-year-old Curt Vincer was visiting his grandparents' house when he fell into the unsupervised swimming pool in their backyard. Vincer, 69 Wis. 2d at 327. As a result, Curt was severely brain damaged and became totally disabled. Id. The above-ground swimming pool had a retractable ladder that was allegedly left in the down position. Id. This allowed Curt to climb the ladder and fall into the pool. Id.
¶ 51. Curt (through his guardian ad litem) and his parents sued the swimming pool company and the company that sold and installed the pool. Id. They claimed that the swimming pool's design was defective *171because there was no self-latching gate preventing entry into the pool. Id. at 331. The legal issue before the court was whether the complaint stated a cause of action against the two defendants. Id. at 329. The court's opinion was primarily focused on the determination of when a product is unreasonably dangerous. This court affirmed that the consumer contemplation test was the proper framework for analyzing design defect cases:
Thus, the test in Wisconsin of whether a product contains an unreasonably dangerous defect depends upon the reasonable expectations of the ordinary consumer concerning the characteristics of this type of product. If the average consumer would reasonably anticipate the dangerous condition of the product and fully appreciate the attendant risk of injury, it would not be unreasonably dangerous and defective. This is an objective test and is not dependent upon the knowledge of the particular injured consumer. Id. at 332.
¶ 52. This court ultimately found that the dangers inherent in a swimming pool were obvious, and that the average consumer would be completely aware of the risk of harm to unsupervised small children when a ladder for the pool is left in a down position. Id. at 333. The court concluded that the pool was as safe as it could reasonably be, that it was not defective as a matter of law, and that the complaint therefore did not state a cause of action. Id. at 331.
¶ 53. The Horsts contend that Vincer does not stand for the proposition that the consumer contemplation test is applicable in all § 402A cases. They argue that Vincer is not a bystander injury case at all, much less one where the product poses a risk of harm to bystanders alone. Rather, the danger in Vincer was the same for everyone, and therefore the consumer contem*172plation test was appropriate. In any event, the Horsts contend use of the bystander contemplation test would have yielded the same result.
¶ 54. Deere argues that this case is far weightier than the Horsts would like to admit. They submit that this is a bystander injury case in which this court explicitly adopted and applied the consumer contemplation test, decisively confirming it as the proper standard for Wisconsin products liability actions. Deere maintains that the court correctly focused on the risk of harm to a child bystander from the perspective of the "average consumer." Deere also notes that the Standard Jury Instructions post-Vmcer (1975) and post -Howes II (1976) describe the consumer contemplation test, and do not describe a bystander contemplation test or contain any unique language for when the injured person is a bystander.
¶ 55. We need not decide whether Vincer is a bystander case. It may be that the injured child in Vincer was a "user" of the pool by virtue of being in the pool when the injury occurred. On the other hand, § 402A explains that a user "includes those who are passively enjoying the benefit of the product" (§ 402A cmt. 1), and it is self-evident that the drowning two-year-old was neither using the pool for its intended purpose nor enjoying the benefit of the pool. However the Vincer case is construed, it is clear that it strongly supports the notion that even when the injured person is not an ordinary user or consumer (which the child surely is not), the consumer contemplation test, which looks at the expectations of the ordinary adult user or consumer, is the framework under which a product's defectiveness is to be governed. We do not vary the test depending on the status of the injured person.
*1734. Komanekin16
¶ 56. In this case from the Eastern District of Wisconsin, five-year-old Jamie Komanekin was severely injured by a propane delivery truck that had come to his home. Komanekin, 819 F. Supp. at 804-05. While playing unsupervised, Jamie somehow snagged his clothing near the connection to the pump, getting his arms tangled in the rotating drive-shaft of the truck's pumping system. Id. at 805. The accident was not caused by any malfunction of the pumping system itself, which worked as intended. Id.
¶ 57. The Komanekins sued a number of companies involved in the manufacture and assembly of the pumping system device that was attached to the propane delivery truck. Id. at 804. They alleged that the defendants should be found strictly liable for selling defectively designed products, and liable for negligence in the design of those products. Id. at 808. The defendants then moved for summary judgment. Id. at 811.
¶ 58. In its opinion, the Eastern District of Wisconsin described the injured child as a bystander, and rightly stated that he may assert a strict products liability claim just like a user or consumer under Howes I. Id. at 809. After acknowledging Wisconsin's reliance on the consumer contemplation test, the court stated:
In a bystander case, presumably, a product is unreasonably dangerous if it presents dangers not apparent to the ordinary bystander. Thus, a product not unreasonably dangerous to the ordinary user or consumer might well be unreasonably dangerous to the ordinary bystander.
*174Id. at 809. The court ultimately denied defendants' motion for summary judgment because it believed there was a genuine issue of material fact as to whether an ordinary bystander would appreciate the danger posed by the truck's pumping system. Id. at 809-11.
¶ 59. Though not a decision of this court, the Horsts highlight this case because the Komanekin court used a bystander contemplation test, concluding that it was the logical outgrowth of Howes I.
¶ 60. Deere counters that the court's analysis was dicta, and in any event, the court's use of "presumably" shows a hesitance and lack of clarity regarding the proper test.
¶ 61. The court in Komanekin clearly did apply a bystander contemplation test, but in our reading, did so without any analysis or consideration of the implications of such an approach. The court erred in assuming that because bystanders can recover, the perspective of the ordinary bystander should control. Howes I does not compel this conclusion, and we reject it.
5. Green17
¶ 62. Plaintiff Linda M. Green filed suit when she suffered injuries from allergic reactions to the proteins in latex medical gloves manufactured by Smith and Nephew AHp Inc. ("S & N"). Green alleged that S & N should be held strictly liable for her injuries. Green, 245 Wis. 2d 772, ¶ 1. After trial, a jury returned a verdict in favor of Green, finding that S & N's gloves were defective and unreasonably dangerous. Id., ¶ 19. The court of appeals affirmed. Id., ¶ 2.
¶ 63. Our review addressed two evidentiary questions and, more importantly for our purposes here, *175whether the jury was properly instructed that the consumer contemplation test is (and should be) the law for strict products liability cases in Wisconsin. We answered in the affirmative. We rejected a move to the standard outlined in § 2(b) of the Restatement (Third), and reiterated that "Wisconsin strict products liability law applies the consumer-contemplation test and only the consumer-contemplation test in all strict products liability cases." Id., ¶¶ 34-35.
¶ 64. The Horsts acknowledge that Green reaffirmed Wisconsin's adherence to the consumer contemplation test, but point out that the issue in Green had nothing to do with bystanders. The Horsts insist that the bystander contemplation test is simply a variation of the consumer contemplation test affirmed in Green.
¶ 65. Deere counters that Green's broad holding is applicable here. The court unambiguously stated that the consumer contemplation test "and only the consumer-contemplation test" applies "in all strict products liability cases."
¶ 66. In our view, the analysis in Green is instructive, though not determinative. The Horsts are correct that the facts in Green did not involve bystanders, and thus, we cannot say that Green answers the question before us today. That said, we were well aware when we decided Green in 2001 that bystanders can recover in strict products liability.18 Yet, the opinion made no concession and left no room for exceptions to the rule that the expectations of the consumer or user guide the determination of whether the product was defective/ unreasonably dangerous.
*1766. Case Law Conclusion
¶ 67. This review leads us to the conclusion that while the language in these prior cases is suggestive, no Wisconsin Supreme Court case directly answers the question before us.19 Our holding today clarifies the law related to the consumer contemplation test; it impinges on no precedent and does not require us to overturn or modify the holding of any prior cases.
D. The Consumer Contemplation Test Governs All Strict Products Liability Cases.
¶ 68. We reject the proposed bystander contemplation test and reiterate that the consumer contemplation test is the proper standard for all strict products liability cases.
¶ 69. At its root, the bystander contemplation test is inherently unworkable. While an ordinary consumer or user of a product can be said to have some objective expectations regarding a product, the same cannot be said of bystanders. The consumer contemplation test was developed in recognition of the fact that it is reasonable for users and consumers of products to hold certain expectations regarding the products they use and the products they buy. See Rebecca Korzec, Dashing Consumer Hopes: Strict Products Liability and the Demise of the Consumer Expectations Test, 20 B.C. Int'l *177& Comp. L. Rev. 227, 232 (1997) ("[T]he consumer expectations test is the natural, logical outgrowth of strict products liability as the extension of implied warranty law."). Thus, the concept of an "ordinary consumer" has some reasonably objective content. But bystanders may have no familiarity with a product. This is especially so in complex design defect cases. What does a bystander expect of the technical design and reasonably available safety features of a product he or she does not buy, does not use, and may not even be aware of?
¶ 70. In addition, it is difficult, if not impossible, to discern who an "ordinary" bystander is and what they know. To illustrate, if a bystander is injured by a combine on a farm, is the "ordinary" bystander a neighboring farmer or a life-long urbanite who cannot tell you what a combine does?20 In short, the notion of an "ordinary bystander" is a concept without content.
¶ 71. One of the basic requirements of a coherent legal test is that it offer a framework for analyzing claims that provides some measure of predictability. See Antonin Scalia, The Rule of Law as a Law of Rules, 56 U. Chi. L. Rev. 1175, 1179 (1989) ("Rudimentary justice requires that those subject to the law must have the means of knowing what it prescribes."). Predictability is important in the law because it allows citizens and businesses to shape their behavior accordingly.
¶ 72. This is particularly important in strict products liability, where one of the main purposes is to incentivize manufacturers to research and implement *178safer designs. If a manufacturer cannot predict with some degree of accuracy when its product is and is not unreasonably dangerous, it will not be able to efficiently adopt appropriate safety precautions. This uncertainty may lead some manufacturers to needlessly remove useful products from the market. The bystander contemplation test would move our state further away from the goal of an efficient deterrent effect on manufacturers by creating unpredictability. Manufacturers and all members of the business community need the ability to anticipate how their choices will be adjudicated in a court of law. Because it does not provide the requisite predictability, the bystander contemplation test fails this basic measure of a legal test.
¶ 73. Although noted earlier, it is important to emphasize that strict products liability is not absolute liability. And a bystander contemplation test comes dangerously close to absolute liability by adopting an amorphous, ambiguous, and standard-less test that effectively gives a jury the power to find a manufacturer liable under almost any conceivable fact situation. This would impose huge and unjustified burdens on businesses (and through businesses, consumers as well). That is not what strict products liability is about, and that is not where we will take it.
¶ 74. Another problem with the bystander contemplation test is that it changes the focus from the product to the injured party, creating discordant results. A bystander contemplation test would in effect create different levels of duty for strict products liability purposes, blurring the line between negligence and strict products liability. Manufacturers would owe a certain level of duty to the user or consumer, and a different, likely higher level of duty to a bystander. This is true because a jury could plausibly find a bystander to *179have higher expectations with respect to the safety and design features of a product than the user or consumer of that same product. This opens the door to a jury finding the same product unreasonably dangerous under some circumstances, but not others, depending on who was injured. Under the Horsts' scheme, a manufacturer could be strictly liable if its product causes injury to a bystander, but not be strictly liable if a consumer suffers the same injury from the same product. Strict products liability, particularly design defect cases, should ensure that products are not unreasonably dangerous, not create different levels of duty and incongruous liability, depending on who is injured.
¶ 75. The Horsts' suggested application of the bystander contemplation test is equally puzzling. They argue it applies only when the threat of injury is to a bystander alone. It is difficult to conceive of a situation where a product is dangerous to a bystander, yet poses no danger to a user or consumer. The Horsts say that is the case here, but surely there is some danger to the user of a riding lawn mower who is driving in reverse. Undoubtedly, the risk of danger is much lower than the danger of running over an unsuspecting child, but it does exist. This means that, under the Horsts' proposal, the bystander contemplation test would not be appropriate here, and may never be.
¶ 76. Perhaps the Horsts do not mean to suggest a danger must be to bystanders "alone" for the bystander contemplation test to apply. It may be that merely a greater risk of danger to the bystander should trigger application of the bystander contemplation test. But this approach would likely require a legal and factual determination by the judge before instructing the jury as to whether the danger posed by a product is greater *180to bystanders than to users or consumers.21 It is not clear how a judge would make such a determination or what procedures he or she would use. It seems that if a bystander contemplation test were adopted, it ought to govern whenever a bystander is injured, avoiding any sort of early judicial fact-finding as to the potential danger posed by a product.22 This, however, is not what the Horsts propose.
¶ 77. The Horsts also argue in their brief that if we reject their proposed test, we "completely eliminate[] any duty owed by manufacturers to bystanders in cases where the risk of bodily injury is unique to bystanders." It is true that under our holding today, where a product is not unreasonably dangerous based on the expectations of the ordinary user or consumer, the bystander does not receive additional protections. But if a product is unreasonably dangerous in light of the expectations of the ordinary user or consumer and a bystander is injured, a strict products liability claim remains available. In addition, the Horsts ignore the availability of recovery under negligence. They are understandably dissatisfied with that avenue of recovery because the jury considered it in this case, and found that Jonathan's parents, and not John Deere, were negligent in the tragic injury to Jonathan.
*181¶ 78. Additionally, a user or consumer's expectations regarding a product will often include safety expectations relating to bystanders. That is, users and consumers do not just have expectations regarding their own safety; they expect that a product will be reasonably safe for bystanders as well. Juries can certainly take this into account in their deliberations and evaluation of whether a product is unreasonably dangerous.23
*182¶ 79. The Horsts also argue that our holding today impermissibly delegates the duty to make a product safe to the user or consumer. But this is begging the question; it assumes a product is unreasonably dangerous. If a product is not unreasonably dangerous (based on consumer expectations), the user or consumer has no duty to make the product safe for bystanders because, by definition, it is sufficiently safe for strict products liability purposes. Users and consumers simply have a duty to use the potentially, but not unreasonably dangerous product with the appropriate standard of care toward their fellow citizens, as they do in all of life.
¶ 80. Finally, even if we accept the Horsts' proposed test, we have difficulty seeing how an ordinary bystander's contemplation (to the extent it exists) would be significantly different than a consumer's contemplation in this case.24 Everyone is aware that a lawn *183mower is very dangerous. While a bystander may or may not be aware or expect a lawn mower to go in reverse, a bystander would certainly know that it is dangerous to allow an unsupervised two-year-old child to play in the yard while it is being mowed.25 The risks *184here are known to all people, including bystanders. The jury confirmed this in finding that Jonathan's parents were the negligent actors. To argue that a bystander is less aware of the very real dangers present here strains credulity.
¶ 81. This truly was a tragic injury, but we cannot simply invent legal theories to make the manufacturer pay for the injuries. This was a horrible accident caused by the negligent use of the lawn mower and negligent supervision of the boy by his parents. At the end of the day, as has been noted elsewhere in a case involving almost precisely the same legal claims, a lawn mower is a lawn mower26 — it is dangerous, and accidents happen. A bystander contemplation test is a creative, but ultimately unsupported, unwise, and unfair attempt to create liability where none exists.
IV CONCLUSION
¶ 82. We hold that the consumer contemplation test, and not a bystander contemplation test, governs all strict products liability claims in Wisconsin, including cases where a bystander is injured. While bystanders may recover when injured by an unreasonably dangerous product, the determination of whether the product is unreasonably dangerous is based on the expectations of the ordinary consumer. Therefore, the jury was properly instructed and the decision of the court of appeals is affirmed.
*185By the Court. — The decision of the court of appeals is affirmed.
¶ 83. ANNETTE K. ZIEGLER, J. did not participate.Horst v. Deere & Co., 2008 WI App 65, 312 Wis. 2d 421, 752 N.W.2d 406.
The dissent confuses and muddles the issue and our holding in this case. It begins: "I agree with the majority that bystanders can recover in strict liability for a product that is unreasonably dangerous to bystanders...." Our holding is more precise. The issue is not who is the product unreasonably dangerous to, but whether the product is unreasonably dangerous. This distinction is subtle, but important.
The trial court's instruction to the jury was substantially identical to Wis JI — Civil 3260. The instruction, as given, is here reproduced in its entirety:
A manufacturer of a product who places on the market a defective product which is unreasonably dangerous to the ordinary user or consumer and which is expected and does reach the consumer without substantial change in the condition in which it is sold, is regarded by law as responsible for harm caused by the product even though he or she has exercised all possible care in preparation and sale of the product, provided the product was being used for the purpose for which it was designed and intended to be used.
There is no claim in this case that the subject lawn tractor failed to perform its intended purpose of mowing the lawn. You may find the subject lawn tractor was dangerous beyond the reasonable contemplation by an ordinary user or consumer even if it served its intended purpose.
A product is said to be defective when it is in a condition not contemplated by the ordinary user or consumer, which is unusually dangerous to the ordinary user or consumer and the defect arose out of design, manufacture or inspection while the article was in the control of the manufacturer.
A defective product is unreasonably dangerous to the ordinary user or consumer when it is dangerous to an extent beyond that which would be contemplated by the ordinary consumer possessing the knowledge of the products' characteristics which were common to the community. A product is not defective if it is safe for normal use.
A manufacturer is not under a duty to manufacture a product which is absolutely free from all possible harm to every individual. It is the duty of the manufacturer not to place upon the market a defective product, which is unreasonably dangerous to the ordinary consumer.
*156The law in Wisconsin imposes a duty on a manufacturer to a bystander if the bystander is injured by a defective product which is unreasonably dangerous to the ordinary user or consumer.
Question 1 in the verdict form asks, when the subject lawn mower left the possession of — Question Number 1 asks, do you find from the evidence that the subject lawn mower, when it left the hands of Defendant Deere and Company, was in a defective condition so as to be unreasonably dangerous to a prospective user/consumer?
Before you can answer Question 1 yes, you must be satisfied that; one, the product was in a defective condition. Two, the defective condition made the product unreasonably dangerous to people. Three, the defective condition of the product existed when the product was under the control of the manufacturer. And four, the product reached the consumer without substantial change in the condition in which it was sold.
Some of our prior decisions have used the term "possibility" instead of "probability." We believe "probability" is a better statement of the test. See Green v. Smith & Nephew AHP, Inc., 2001 WI 109, ¶¶ 114-19, 245 Wis. 2d 772, 629 N.W.2d 727 (Crooks, J., concurring).
Section 402A provides as follows:
(1) One who sells any product in a defective condition unreasonably dangerous to the user or consumer or to his property is subject to liability for physical harm thereby caused to the ultimate user or consumer, or to his property, if
*159(a) the seller is engaged in the business of selling such a product, and
(b) it is expected to and does reach the user or consumer without substantial change in the condition in which it is sold.
(2) The rule stated in Subsection (1) applies although
(a) the seller has exercised all possible care in the preparation and sale of his product, and
Ob) the user or consumer has not bought the product from or entered into any contractual relation with the seller.
Dippel v. Sciano announced a five-part test to prove strict liability under § 402A:
*160(1) that the product was in defective condition when it left the possession or control of the seller,
(2) that it was unreasonably dangerous to the user or consumer,
(3) that the defect was a cause (a substantial factor) of the plaintiffs injuries or damages,
(4) that the seller engaged in the business of selling such product or, put negatively, that this is not an isolated or infrequent transaction not related to the principal business of the seller, and
(5) that the product was one which the seller expected to and did reach the user or consumer without substantial change in the condition it was when he sold it. 37 Wis. 2d 443, 460, 155 N.W.2d 55 (1967).
A risk-utility analysis requires a balancing of the risks and benefits of a product design "in light of the knowledge of risks and risk-avoidance techniques reasonably attainable at the time of distribution." Restatement (Third) of Torts: Products Liability § 2 cmt. a (1998).
The Horsts also suggest that our decision in Sumnicht v. Toyota Motor Sales, 121 Wis. 2d 338, 360 N.W.2d 2 (1984) is relevant. Sumnicht concerned a passenger in a car seat who was injured in an accident, and claimed the car seat was defectively designed and unreasonably dangerous. The Horsts point to the language used in the special verdict question in Sumnicht asking whether the defect in the seat system design was "unreasonably dangerous to a prospective passenger in the rear seat of the automobile." Id. at 370. However, the court actually affirmed that Wisconsin follows the consumer contemplation test, quoting comment i from § 402A. The court correctly noted that an unreasonably dangerous product "must contain a dangerous defect whose presence an ordinary consumer would not reasonably expect." Id. at 369 (quoting Arbet v. Gussarson, 66 Wis. 2d 551, 557, 225 N.W.2d 431 (1975)). The Horsts also ignore that § 402A explicitly includes automobile passengers within the definition of a "user":" 'User' includes those who are passively enjoying the benefit of the product, as in the case of passengers in automobiles." § 402A cmt. 1.
Deere also cites several other Wisconsin and non-Wisconsin cases to bolster its case that we will not discuss here *166in detail. Of note, in Adamany v. Cub Cadet Corp., 04-C-224—C, 2005 U.S. Dist. LEXIS 9612 (W.D. Wis. May 16, 2005), the Western District of Wisconsin applied the consumer contemplation test, and not a bystander contemplation test, to a strict liability claim arising from an injury to a five-year-old child bystander. Deere also cites Anderson v. Alfa-Laval Agri, Inc., 209 Wis. 2d 337, 564 N.W.2d 788 (Ct. App. 1997), which affirmed the use of Wis JI — Civil 3260 for the strict liability claim arising from an injury to a two-year-old child bystander.
Howes v. Hansen, 56 Wis. 2d 247, 201 N.W.2d 825 (1972).
The injured child in this case is Richard Howes II. However, because this case came before us twice, the naming convention we will use to describe the two cases is Howes I for the 1972 decision and Howes II for the 1976 decision.
The opinion noted that some courts suggested bystanders should have more protection than consumers because of their disadvantaged position. Id. at 260. This dicta, however, was neither adopted by the court nor made part of the holding.
Howes v. Deere, 71 Wis. 2d 268, 238 N.W.2d 76 (1976).
We note here and further address in footnote 23 below that, because bystanders may recover, it does not matter who a product is dangerous to (i.e. — a bystander versus a user or consumer). What matters is whether the product is unreasonably dangerous as an objective matter. Today we address the proper test for making that determination when bystanders are injured.
Vincer v. Esther Williams All-Aluminum Swimming Pool Co., 69 Wis. 2d 326, 230 N.W.2d 794 (1975).
Komanekin v. Inland Truck Parts, 819 F. Supp. 802 (E.D. Wis. 1993).
Green, 245 Wis. 2d 772.
It is indicative of the court's intent to adhere to the consumer contemplation test that the majority cited our opinion in Howes I on two occasions. Green, 245 Wis. 2d 772, ¶¶ 56, 69.
While Komanekin did answer the question, it was a federal case, not one of our precedents. No prior Wisconsin state cases have addressed this question directly. The Horsts admit as much in their brief, stating that "[n]o Wisconsin court has considered the § 402A liability standard to be applied where a product design creates a risk of harm to bystanders alone."
A combine is a "power-operated harvesting machine that cuts, threshes, and cleans grain." The American Heritage Dictionary of the English Language 377 (3d ed. 1992).
In fact, this legal/factual determination by the judge would probably be required even under the Horsts' suggested application of the test. Under their approach, a determination must be made before instructing the jury that the danger is to bystanders alone, and not to users or consumers.
To be clear, we reject the bystander contemplation test. Our point is to show that the Horsts' more limited suggested application of this test has many practical problems.
The special verdict question in this case asked: "Do you find from the evidence that the subject lawn mower, when it left the hands of Defendant Deere & Company, was in a defective condition so as to be unreasonably dangerous to a prospective user/consumer?" The trial court rejected a plea by the Horsts to add "or bystander" to the end of the special verdict question. The jury was further instructed that to answer "yes" to the special verdict question, they were required to find, among other things, that "the defective condition made the product unreasonably dangerous to people."
We think a better special verdict question need not query to whom the product is unreasonably dangerous. The question is not whether a product is unreasonably dangerous to a user or consumer versus unreasonably dangerous to a bystander. The question is simply whether a product is unreasonably dangerous. And the determination of whether a product is unreasonably dangerous is an objective inquiry based on the expectations of an ordinary user or consumer. Any injured person, whether a bystander or user or consumer, may recover if injured by an unreasonably dangerous product.
The dissent asserts that the jury instructions in Wis JI— Civil 3260, even with the trial court's insertion making clear the availability of recovery for bystanders, were wrong and misleading. Dissent, ¶¶ 120-128. Though the standard jury instructions are not as clear as we would prefer, when viewed as a whole, the jury instructions given here sufficiently conveyed that a bystander may recover if the product was unreasonably dangerous, and that such a determination is based on the expectations of the ordinary user or consumer.
*182The dissent fails to acknowledge that the jury instructions themselves define when a product is "unreasonably dangerous to the ordinary user or consumer," and that is "when it is dangerous to an extent beyond that which would be contemplated by the ordinary consumer possessing the knowledge of the products' characteristics which were common to the community." In other words, the instructions define the phrase that so troubles the dissent and do so by reference to the consumer contemplation test. Moreover, the inconsistency between the instruction's dual statements that the product must be unreasonably dangerous to "a prospective user/consumer" and later to "people" is harmless.
In other courts around the country, plaintiffs proceeding in strict liability upon the injury to a child have argued that it is the expectations of an ordinary child that should govern. Courts have resoundingly rejected this argument, in large part because children do not have expectations regarding the dangers of certain products. Courts have thus held that it is the *183ordinary consumer or user's expectations that govern. See Kelley v. Rival Mfg. Co., 704 F. Supp. 1039, 1043 (W.D. Okla. 1989) (in a case involving a small child who was injured hy pulling a crock-pot onto himself, the question of whether the crock-pot was unreasonably dangerous is determined by the perspective of the parent consumer who purchased the product, not the perspective of the minor child) (applying Oklahoma law); Curtis v. Universal Match Corp., Inc., 778 F. Supp. 1421, 1425 (E.D. Tenn. 1991) (in a case involving a two-year-old child whose diaper was set on fire by his three year old brother, the question of whether the lighter was unreasonably dangerous is determined by the contemplation of the ordinary adult consumer, not the viewpoint of the minor child) (applying Tennessee law); Welch v. Scripto-Tokai Corp., 651 N.E.2d 810, 814-15 (Ind. Ct. App. 1995) (in a case involving a child who suffered injuries after lighting his pajamas on fire when playing with a lighter, the question of whether the lighter was unreasonably dangerous is determined by the perspective of the ordinary adult consumer, not the perspective of the minor child) Bellotte v. Zayre Corp., 352 A.2d 723, 725-26 (N.H. 1976) (in a certified question from the First Circuit Court of Appeals involving a five-year-old whose pajamas were set on fire when playing with matches, the New Hampshire Supreme Court stated that the question of whether the pajamas were unreasonably dangerous should be based on the expectations of consumer parents) (note, New Hampshire has moved away from the consumer contemplation test and now engages in a risk-utility balancing test).
The Horsts state the question as follows: "[I]f the ordinary bystander would have expected Deere to design its mower so that it would be incapable of mowing in reverse, instead of simply instructing operators not to do so, Deere breached its § 402A duty to bystanders." But the issue is not whether a bystander (or consumer) would expect the lawn mower to go in *184reverse, but whether a bystander (or consumer) would appreciate the danger posed by the RIO-equipped lawn mower to a two-year-old child.
Transcript of Record at 4, Brown v. Sears, Roebuck & Co., 328 F.3d 1274 (10th Cir. 2003) No. 01-4226 ("A lawn mower is a lawn mower.").