(dissenting).
This is a tragic case. Jocelyn Dickhoff was born with a rare and aggressive form of childhood cancer that puts her young life at risk. I am acutely aware that the disputes between the parties in the courtroom are insignificant in comparison to the struggle that Jocelyn faces outside of it. But against this weighty background, we must decide the relatively modest issue of whether, on the present record, there is sufficient expert medical testimony to prove that Dr. Rachel Tollefsrud directly caused any of Jocelyn’s injuries. It is a cardinal principle of tort law and fundamental fairness that a “defendant should be responsible only for the injuries that are legally caused by the defendant’s negligence.” Rowe v. Munye, 702 N.W.2d 729, 742 (Minn.2005) (emphasis added). The majority disregards this cardinal principle and introduces speculation by concluding that a physician may be liable for harms not directly caused by the physician’s negligence, but caused by the patient’s underlying disease. In so doing, the majority sua sponte overrules long standing precedent and upsets an unbroken tradition of tort law in this state. Therefore, I must respectfully dissent.
I.
To properly analyze this case, it is important first to identify the precise injuries or harms for which the Dickhoffs seek compensation. The Dickhoffs concede that Dr. Tollefsrud should not have to compensate them for Jocelyn’s initial cancer treatments because she would have required those same treatments even if the cancer had been diagnosed immediately. But the Dickhoffs contend that Dr. Tollefsrud should have to compensate them for two other categories of damages. Specifically, they allege that Dr. Tollefsrud’s negligence caused a delay in the diagnosis and treatment of Jocelyn’s cancer, which decreased her odds of ultimately surviving her disease from 60% to 40%, and made her more likely to suffer a recurrence of her cancer. Therefore, the Dickhoffs seek compensation for (1) the cost of medical treatment related to the 2010 recurrence of Jocelyn’s cancer, as well as pain and suffering that Jocelyn experienced as a result of that treatment; and (2) Jocelyn’s decreased chance of surviving the cancer in the future and the probability that she will die from the disease.
As the majority aptly puts it, “the injury that lies at the heart of the Dickhoffs’ medical malpractice action is a claim that Dr. Tollefsrud’s alleged negligence in*339creased the risk that Jocelyn’s cancer would recur and decreased her chances of survival.” Although the Dickhoffs do not characterize it as such, I agree with the majority that this is a quintessential “loss of chance” claim. But in light of our precedent, I disagree with the majority’s conclusion that a loss of chance claim is cognizable in Minnesota. We have explicitly rejected loss of chance as a theory of recovery in two cases, and we are bound by that precedent. Fabio v. Bellomo, 504 N.W.2d 758 (Minn.1993); Leubner v. Sterner, 493 N.W.2d 119 (Minn.1992). The majority overrules these decisions without either party requesting that we do so, and without providing any compelling justification for the abandonment of our precedent. But even if we were unconstrained by precedent, I would decline to adopt the loss of chance doctrine because it unfairly holds physicians like Dr. Tollefsrud liable for harms that may never materialize and, if they do occur, are not proximately caused by the physician’s negligence.
A.
In considering whether the Dickhoffs can recover for loss of chance, we do not write on a clean slate. We have considered the viability of a loss of chance claim in two previous cases: Leubner v. Sterner, 493 N.W.2d 119 (Minn.1992) and Fabio v. Bellomo, 504 N.W.2d 758 (Minn.1993). In both cases we expressly declined to adopt the loss of chance doctrine in Minnesota under facts comparable to the present case.
In Leubner, the plaintiff was referred to Dr. Ronald Jensen after her family doctor confirmed two small lumps in her breast. 493 N.W.2d at 120. Dr. Jensen examined plaintiff but did not order a biopsy. Id. At a follow-up appointment 6 months later, Dr. Jensen noticed the nodules were enlarged and advised a biopsy, which revealed cancer. Id. A partial mastectomy was performed. Id. Unfortunately, plaintiff experienced several recurrences of cancer, ultimately resulting in a total mastectomy and the removal of two additional lesions on the wall of her chest. Id. Plaintiff sued Dr. Jensen and presented an expert witness who opined that, had plaintiff been diagnosed and treated for cancer after her first visit to Dr. Jensen, “she would possess an increased chance of survival from her disease of breast cancer, in contrast to [her] present chance of survival.” Id. In particular, plaintiff’s expert opined that, as a result of the delay, plaintiff’s 5-year survival rate had decreased from approximately 96% to 80%.1
Dr. Jensen moved for summary judgment and the district court granted the motion. Specifically, the court rejected plaintiffs claim under a loss of chance theory and concluded that plaintiff generally “failed to establish that it was more likely than not that the specified consequences resulted from the alleged negligence.” Leubner, 493 N.W.2d at 120. The court of appeals reversed, and Dr. Jensen sought review from this court, asserting that plaintiff failed to present a prima facie case of causation. Id. at 120-21.
We noted that, in order to establish a prima facie case of medical malpractice, a plaintiff must prove “that it is more probable than not that his or her injury was a result of the defendant health care provider’s negligence.” Id. at 121. But we observed that “causation cannot be discussed intelligently without reference to the injury claimed to be caused.” Id. If the injury claimed is plaintiff’s death, we stated, the offer of proof failed because “there is no *340proof it is more probable than not that plaintiff will not survive her cancer.” Id. Indeed, we noted that the plaintiffs own expert believed that death is overwhelmingly improbable. Id. We also addressed whether plaintiffs claim could proceed based on the reduction of her odds of survival (ie., loss of chance):
Arguably, the injury claimed to be caused is a decreased percentage chance of surviving, whether or not the patient, in fact, has survived. Here the difficulty is perhaps not so much in proving causation as “more probable than not, ” but in what appears to be the amorphous and speculative nature of the asserted “injury,” especially as it applies to a particular patient.
Id. (emphasis added). In other words, we concluded that the mere reduction in the patient’s odds of survival due to the delay in diagnosis was too speculative of an injury to be independently compensable.2 Finally, we concluded that plaintiff could not recover for the recurrence of her cancer and resulting surgery and treatment, because her expert evidence did not show that it was more probable than not that her recurrence was caused by Dr. Jensen’s negligence. Id. at 122.
We were even more explicit in our rejection of the “loss of chance” doctrine in Fabio. In that case, Dr. James Bellomo twice examined a lump in plaintiffs left breast but told her not to worry about it because it was a fibrous mass. Fabio, 504 N.W.2d at 760. Plaintiff later saw another physician who ordered a mammogram. Id. The mammogram revealed two tumors, and a biopsy revealed that the tumors were cancerous and had metastasized to her lymph nodes. Id. Plaintiff sued Dr. Bellomo, alleging that he negligently failed to diagnose her breast cancer, which caused her to suffer three forms of damages, including damages arising from a “ ‘loss of chance’ of life expectancy and a greater risk of recurrence of cancer.” Id. at 761. Plaintiff offered expert testimony that, because of the delay in treating her cancer, her risk of recurrence had risen from as low as 15% to 30% or greater,3 and her chance for surviving had fallen but remained above 50%. The district court ruled that no cause of action existed for loss of chance, and the court of appeals affirmed. Fabio, 504 N.W.2d at 761.
We granted review and affirmed. Id. at 760-61. We directly addressed plaintiffs allegation that “loss of chance” was a com-pensable “form[ ] of damage” from the failure to diagnose the cancer. Id. at 761-62. But we again rejected loss of chance as a theory of recovery:
Fabio’s second theory of recovery is for “loss of chance.” She argues that her *341increased chance of recurrence of cancer and her decreased chance of living another 20 years are compensable injuries. We have never recognized loss of chance in the context of a medical malpractice action, and we decline to recognize it in this case.
Id. at 762. We acknowledged that a plaintiff may recover for future damages or consequences of a defendant’s negligence when the probability of the future damage was proven with reasonable medical certainty and the future damage was caused solely by the initial injury inflicted by the defendant. See id. at 763. But we noted that Fabio’s “initial ‘injury,’ her cancer, did not result from a misdiagnosis by Dr. Bel-lomo, and a misdiagnosis by Dr. Bellomo could not have been the sole cause of any future ill effects.” Id. Moreover, we noted that, “even if we were to adopt loss of chance as a theory of recovery in medical malpractice actions,” it would not apply to plaintiff because there was no evidence that it was more likely than not that her cancer would recur or that she would not survive. Id. Indeed, the evidence showed that she had at least a 50% chance of survival and about a 70% chance that her cancer would not recur. See id. Therefore, we held that the district court properly granted summary judgment on the loss of chance claim.4 Id. at 763.
In this case, the majority concedes that our opinion in Fabio “intimated” that we were conclusively rejecting loss of chance as a theory of tort recovery. Nevertheless, the majority concludes that it is not bound by our rejection of the doctrine for two related reasons. First, the majority claims our rejection of loss of chance in Fabio was motivated by the specific facts of that case. But the majority conspicuously fails to explain how the facts of the Diekhoffs’ case are meaningfully distinguishable from the facts of Fabio. Both cases involve allegations that a doctor negligently failed to diagnose an unidentified “lump” as cancer and that, as a result of the delay, the risk of both cancer recurrence and death increased. Even the percentage chance lost is nearly identical: as much as 15% chance lost in Fabio versus a 20% chance lost here. The majority fails to acknowledge any of these similarities, and does not cite a single distinguishing fact between this case and Fabio. Because Fabio is factually indistinguishable from this case, our rejection of the loss of chance theory “in that case” also controls here.
Second, the majority reasons it is not bound by Fabio’s rejection of loss of chance because it was “mere dictum.” The majority notes that we offered two alternative justifications for our holding in Fabio: (1) that we declined to recognize loss of chance, and (2) that even if we were to recognize loss of chance, the plaintiff had not presented sufficient evidence to make out such a claim. The majority therefore reasons that “[njeither justification was necessary to our decision in Fa-bio,” and thus the rejection of the loss of chance doctrine is “dictum.” I disagree with this analysis.
Under the majority’s reasoning, both of the alternative grounds in Fabio would be dictum, because neither justification was necessary to the decision in light of the existence of the other. But that result defies logic and would mean that no rule of law can be discerned from Fabio at all. In fact, we have explicitly rejected such reasoning. In State ex rel. Foster v. Naftalin, we stated:
*342Where ... two or more issues are before the court and are argued by counsel, and the court places its decision on both even though a decision on one issue might have been sufficient to dispose of the case, the decision is equally binding as to both issues.
246 Minn. 181, 208, 74 N.W.2d 249, 266 (1956);5 see also State v. Rainer, 258 Minn. 168, 178, 103 N.W.2d 389, 396 (1960) (stating that “even though a case might have been decided on another theory it does not render what was said in the court’s opinion ‘obiter dictum’ if what was said bears directly upon the theory upon which the decision proceeded and upon an issue of law treated as decisive”). “Dicta” is a term of art that generally refers to expressions in an opinion “which go beyond the facts before the court and therefore are the individual views of the author of the opinion.” Naftalin, 246 Minn. at 208, 74 N.W.2d at 266. The statements in Fabio clearly do not meet this definition of dicta, as they squarely addressed the facts and legal issues before the court and were not the mere views of the author of the opinion.
The majority’s disregard for Fabio is even more egregious in light of the fact that the majority fails to acknowledge that we also rejected loss of chance in Leubner, where we noted that a plaintiff cannot recover for the decreased percentage chance of surviving itself as an item of damage, because the reduced chance is too “amorphous and speculative” of an injury. Leubner, 493 N.W.2d at 121. We are bound by principles of stare decisis to follow Leubner and Fabio. In short, the majority is simply wrong to conclude that we have never definitively decided whether to recognize a loss of chance claim.
B.
Given that we rejected the loss of chance doctrine in Leubner and Fabio, the Dick-hoffs can only recover for such a claim if we overrule those precedents. But we have consistently stated that we are “extremely reluctant” to overrule our precedent under principles of stare decisis, Oanes v. Allstate Ins. Co., 617 N.W.2d 401, 406 (Minn.2000), and “ ‘require a compelling reason’ before overruling a prior decision,” Cargill, Inc. v. Ace Am. Ins. Co., 784 N.W.2d 341, 352 (Minn.2010) (quoting State v. Martin, 773 N.W.2d 89, 98 (Minn.2009)). This is because stare decisis promotes stability in the law, see Cargill, 784 N.W.2d at 352, as well as “ ‘the evenhanded, predictable, and consistent development of legal principles, ... reliance on judicial decisions, and ... the actual and perceived integrity of the judicial process,’ ” Seminole Tribe of Florida v. Florida, 517 U.S. 44, 63, 116 S.Ct. 1114, 134 L.Ed.2d 252 (1996) (quoting Payne v. Tennessee, 501 U.S. 808, 827, 111 S.Ct. 2597, 115 L.Ed.2d 720 (1991)). As we stated more than 50 years ago:
Government by law instead of by man, which is the main bulwark to our democratic form of government, demands a decent respect for the rule of stare deci-sis in order that citizens of this state will *343be assured that decisions of the court are good for more than “one trip and one day only.”
Naftalin, 246 Minn. at 205, 74 N.W.2d at 264. Although the majority overrules Fa-bio and Leubner on the issue of the loss of chance doctrine, it does not explicitly acknowledge that it is doing so. As such, the majority sidesteps the need to provide a “compelling reason” for departing from our precedent.
In my view, this case is a poor vehicle for departing from our precedent because the Dickhoffs themselves never requested that we recognize loss of chance or overrule Fabio and Leubner. Indeed, in their brief to this court, the Dickhoffs repeatedly assert that they “have presented a traditional tort negligence claim.” They explicitly disavow a loss of chance theory and state that they “are not presenting what is referred to as a loss of chance case, either under a causation or damages analysis.” Because the Dickhoffs do not purport to assert a loss of chance claim, neither party has briefed the issue of whether Fabio and Leubner should be overruled or whether loss of chance should be adopted. The majority has undertaken to overrule these cases sua sponte despite the fact that the issue was not squarely presented for our consideration and we do not have the benefit of the adversarial process or arguments of the parties in making our decision.6 Sound judicial principles dictate that we should not render a decision on important legal issues that have not been properly developed. See Broehm v. Mayo Clinic Rochester, 690 N.W.2d 721, 728 (Minn.2005) (noting that we generally decline to consider issues that are not adequately briefed on appeal); Balder v. Haley, 399 N.W.2d 77, 80 (Minn.1987) (stating that issues not argued in briefs must be deemed waived on appeal); see also In re Welfare of K.T., 327 N.W.2d 13, 16-17 (Minn.1982) (stating that “a party may not raise for the first time on appeal a matter not presented to the court below”).
C.
But even if the Dickhoffs had asked that we revisit our precedent and reconsider our rejection of loss of chance, I can discern no compelling reason to overrule Fa-bio and Leubner. Specifically, the loss of chance doctrine that the majority adopts today undermines traditional tort principles of causation and violates fundamental fairness by holding physicians liable for harms that are not caused by their negligence.
1.
There are three essential elements of a medical malpractice claim. A plaintiff must prove: “(1) the standard of care recognized by the medical community as applicable to the particular defendant’s conduct, (2) that the defendant in fact departed from that standard, and (3) that the defendant’s departure from the standard was a direct cause of [the patient’s] injuries.” MacRae v. Grp. Health Plan, Inc., 753 N.W.2d 711, 717 (Minn.2008) (quoting Plutshack v. Univ. of Minn. Hosps., 316 N.W.2d 1, 5 (Minn.1982)). For the pur*344pose of this appeal, only the third element — that the physician’s negligence was a direct cause of the patient’s injuries — is in dispute. We have stated that, in order to prove causation, a plaintiff must introduce expert medical testimony that establishes that it was more probable than not that the injuries for which compensation is sought resulted from the physician’s negligence. Cornfeldt v. Tongen, 295 N.W.2d 638, 640 (Minn.1980). The injuries or harms can be things that have already materialized or come to pass at the time suit is brought; or, in certain circumstances, the injuries or harms can be things that the plaintiff proves are more likely than not to materialize in the future (i.e., future damages). See Pietrzak v. Eggen, 295 N.W.2d 504, 507 (Minn.1980); see also Carpenter v. Nelson, 257 Minn. 424, 428, 101 N.W.2d 918, 921 (1960); 4A Minn. Dist. Judges Ass’n, Minnesota Practice-Jury Instruction Guides, Civil, CIVJIG 91.25 (future damages for bodily harm), 91.30 (future medical expenses), 91.35 (future earnings) (5th ed.2006).
But whether the doctor caused the (present or future) injury in medical malpractice cases based on a physician’s failure to timely diagnose is difficult to determine because the patient is suffering from a preexisting condition — namely, the underlying disease — in this case, cancer. The plaintiff in a failure-to-timely-diagnose case does “no[t] claim the disease itself, the cancer, was caused by the physician,” Leubner, 493 N.W.2d at 122; see also Fabio, 504 N.W.2d at 763, and it is the underlying disease, not the physician, that directly inflicts the actual harm on the patient.
In light of this problem, we have stated that it is the “policy” of this state, and a fundamental principle of tort law, that a “defendant should be responsible only for the injuries that are legally caused by the defendant’s negligence.” Rowe, 702 N.W.2d at 739, 742. In cases where a plaintiff has a preexisting disease or condition, it is unfair to force the defendant to pay for the effects of that preexisting condition. See id. at 742. Therefore, we have held time and again that the “plaintiffs have the burden of showing that ‘it was more probable that [injury] resulted from some negligence for which defendant was responsible than from something for which he was not responsible.’ ” Plutshack v. Univ. of Minn. Hosps., 316 N.W.2d 1, 7 (Minn.1982) (alteration in original) (emphasis added) (quoting Silver v. Redleaf 292 Minn. 463, 465, 194 N.W.2d 271, 273 (1972)); accord Harvey v. Fridley Med. Ctr., P.A., 315 N.W.2d 225, 227 (Minn.1982).
Of course, a plaintiff may recover “where a pre-existing disease is aggravated by the negligence of another person.” Nelson v. Twin City Motor Bus Co., 239 Minn. 276, 280, 58 N.W.2d 561, 563 (1953). But in such cases, “we have required the defendant to pay only for the damages he or she caused over and above the consequences that would have occurred from the preexisting injury if the [defendant’s negligence] had not occurred.” Rowe, 702 N.W.2d at 736; see also Schore v. Mueller, 290 Minn. 186, 189, 186 N.W.2d 699, 701 (1971); Nelson, 239 Minn. at 280, 58 N.W.2d at 563 (stating that “the victim’s recovery in damages is limited to the additional injury caused by this aggravation over and above the consequences which the pre-existing disease, running its normal course, would itself have caused if there had been no aggravation by negligent injury”). The burden of proof, however, remains on the plaintiff to segregate her damages and identify which specific compensable harms were the result of defendant’s aggravation of her disease, as opposed to those that were the result of the disease itself. See Rowe, 702 N.W.2d *345at 736; Leubner, 493 N.W.2d at 122. A jury must not be permitted to speculate as to the actual cause of the harm: the disease or the doctor. See Cornfeldt, 295 N.W.2d at 640-41.
Here, the injury or harm for which the Diekhoffs seek compensation is, in their own words, “the shortening of [Jocelyn’s] life” or her “imminent death.” But this harm has not yet come to pass. I will assume for the sake of argument that it could, in theory, be a compensable form of future damage under traditional tort principles.7 The Diekhoffs have offered expert medical evidence that this harm will more likely than not materialize in the future (ie., that death is likely). Dr. Edwin Forman opined that Jocelyn’s likelihood of survival is 40%, versus a 60% chance that she will die. But where the Diekhoffs’ proof fails is in showing that, if Jocelyn dies, her death or shortened life will have been caused by something for which Dr. Tollefsrud is responsible (the allegedly negligent delay in diagnosis) rather than caused by something for which Dr. Tollefs-rud is not responsible (the natural progression of the disease itself). See Plutshack, 316 N.W.2d at 7. Put differently, the Diekhoffs improperly attempt to use expert testimony that it is more likely than not that Jocelyn will have a shortened life to satisfy the requirement that they prove that the shortened life (if it occurs) will have been directly caused by the failure to diagnose.
In Dr. Forman’s opinion, Dr. Tollefs-rud’s actions reduced Jocelyn’s likelihood of survival from 60% to 40%. In other words, Jocelyn always had at least a 40% chance of death even if she had been timely diagnosed, but she gained an additional 20% chance of death due to the delay in diagnosis. Jocelyn now has an overall 60% chance of dying from the cancer, of which two-thirds (40/60) is attributable to the cancer itself even without the delay in diagnosis, and one-third (20/60) is attributable to the allegedly negligent delay in diagnosis. This means that, even under the Diekhoffs’ own evidence, Jocelyn’s death, if it occurs, is very probably (66.7% or 40/60) attributable to the underlying cancer that she was born with, and unlikely (33.3% or 20/60) attributable to anything that Dr. Tollefsrud did. Put simply, the Diekhoffs are seeking to hold Dr. Tollefs-rud legally responsible for something that she likely did not cause.
The Diekhoffs’ proof of causation similarly fails with respect to the April 2010 recurrence of Jocelyn’s cancer and resulting treatments. On this point, Dr. For-man’s third affidavit admitted that it was “impossible to put precise statistics” on the likelihood of recurrence with or without timely care, but nonetheless asserted that Dr. Tollefsrud’s failure to timely diagnose the disease was “to blame for the recurrence” because it “changed the likelihood of recurrence and need for additional care from unlikely to probable.” But Dr. For-man’s opinion is purely conclusory. It *346does not set out the factual or scientific basis for the opinion or explain how he was able to conclude that the recurrence is more likely than not due to Dr. Tollefsrud rather than the natural progression of the cancer even with timely diagnosis. As such, the opinion does not begin to meet the minimum requirements of “foundational reliability” required by our rules. Minn. R. Evid. 702. We require more from experts in medical malpractices cases. See Lindberg v. Health Partners, Inc., 599 N.W.2d 572, 577-78 (Minn.1999) (noting that an expert’s affidavit must contain more than “broad and conclusory statements as to causation” and should “outline a chain of causation” connecting the negligence with the injury); Stroud v. Hennepin Cnty. Med. Ctr., 556 N.W.2d 552, 556 (Minn.1996). Therefore, the Dickhoffs failed to prove that Dr. Tollefsrud caused Jocelyn’s April 2010 recurrence of cancer.
The majority essentially concludes that expert medical testimony that Dr. Tollefs-rud directly caused the shortened lifespan and recurrence of cancer is no longer necessary. Instead, proof that Jocelyn’s chances of survival may have dropped 20% plus rank speculation regarding the cause of the recurrence of cancer is deemed sufficient, thus undermining this state’s policy that a “defendant should be responsible only for the injuries that are legally caused by the defendant’s negligence.” See Rowe, 702 N.W.2d at 742. In practice, the majority’s opinion allows plaintiffs to recover merely upon proof that they have suffered any non-de minimis reduction in their chance of recovery or survival, even if it is exceedingly unlikely that the physician can be shown to have caused the ultimate harm. For example, the majority’s opinion would seemingly allow a plaintiff to recover whose chance of survival has been reduced from 20% to 10%, even though, if the plaintiff dies, there will be only an 11% chance (10/90) that the harm is attributable to the doctor. Holding physicians liable when they are highly unlikely to have caused the ultimate harm is contrary to cardinal tort principles and is fundamentally unfair. Although I do not deny the harshness of denying compensation to a young child who alleges that her doctor’s negligence diminished her chance of survival and recovery, the importance of causation as a limiting principle for tort liability cannot be ignored. As one jurist aptly stated:
The combination of the loss of a loved one to cancer and a doctor’s negligence in diagnosis seems to compel a finding of liability. Nonetheless, justice must be dealt with an even hand. To hold a defendant liable without proof that his actions caused plaintiff harm would open up untold abuses of the litigation system.
Herskovits v. Grp. Health Coop, of Puget Sound, 99 Wash.2d 609, 664 P.2d 474, 491 (1983) (Brachtenbach, J., dissenting).
2.
As a justification for departing from these traditional and fundamental rules, the majority attempts to recast the “injury” at issue. According to the majority, the injury is not the actual shortening of Jocelyn’s life or the actual recurrence of her cancer; rather, the “reduction of a patient’s chance of recovery or survival [is] a distinct injury.” In other words, with respect to the decrease in Jocelyn’s odds of survival, the majority holds that the 20% reduction in her chance of survival is itself the harm, regardless of whatever actually comes to pass or whether Jocelyn ultimately survives. And because, according to the majority, Dr. Tollefsrud’s negligence caused the 20% reduction, the Dick-hoffs can recover for that lost 20% even if the ultimate harm (death or shortened life) never materializes or the physician did not *347cause the ultimate harm. In sum, the majority seeks to mask its alteration of traditional causation requirements by casting its opinion instead as a redefinition of the concept of recoverable damages.
I cannot accept this premise. While I agree with the majority in the abstract that a patient “values” her chances for survival, that fact alone does not render every reduction of that chance a compen-sable “injury.” Imagine, for example, a patient who undergoes surgery for which the doctor negligently fails to disclose a 20% chance of death on the operating table. But the patient survives the surgery and does not die on the operating table. Although the patient surely would not have been indifferent to the increased 20% chance of death in choosing to undergo the surgery, having escaped harm the patient has not suffered a compensable injury. See K.A.C. v. Benson, 527 N.W.2d 553, 561 (Minn.1995) (plaintiff must show that risk “actually materialized in harm”). That is because the “chance” only has meaning with reference to the actual outcome. The true harm or injury to the plaintiff is the death. See Kramer v. Lewisville Mem,'Il Hosp., 858 S.W.2d 397, 403 n. 5 (Tex.1993); see also Fennell v. S. Md. Hosp. Ctr., Inc., 320 Md. 776, 580 A.2d 206, 213 (1990) (“[I]t would seem that the true injury is the death.”). And if the doctor cannot be said to have caused the ultimate injury (if it ever occurs), of what relevance is it that the doctor can be said to have caused the loss of chance?
In other words, the majority has concocted a legal fiction in order to obscure the very real causation problem in this case. The loss of chance is not an injury that the law should recognize as compensa-ble. We recognized this is Leubner, when we rejected the idea that the “decreased percentage chance of surviving” may itself be “the injury claimed to be caused.” See 493 N.W.2d at 121. We stated that “[hjere the difficulty is perhaps not so much in proving causation as ‘more probable than not,’ but in what appears to be the amorphous and speculative nature of the asserted ‘injury.’ ” Id. (emphasis added).
The majority contends that treating the lost chance of survival as a compensable injury in itself promotes the fundamental aims of tort law, “including compensation for victims of medical negligence and the deterrence of unsafe conduct by health care providers.” I disagree. As to compensation, the loss of chance doctrine as articulated by the majority has several problems. First, the majority allows plaintiffs to recover damages for the reduction in their life expectancy but gives no guidance to litigants, juries, and our state’s district courts on how such a novel claim should be valued. Under the majority’s decision, juries award damages by assigning a dollar value to plaintiffs reduced life expectancy. But I see no framework that would allow a jury to assess damage fairly and logically, without inviting speculation, guess, passion, or prejudice. 4A Minn. Dist. Judges Ass’n, Minnesota Practice-Jury Instruction Guides, Civil, CIVJIG 90.15 (5th ed.2006) (stating that the jury “must not decide damages based on speculation or guess”).8
*348Second, the majority’s adoption of loss of chance will actually overcompensate plaintiffs in certain circumstances. The problem stems from the fact that the majority allows a plaintiff to recover damages from her doctor for loss of chance before knowing whether that chance will ever materialize in harm. Take, for example, a young plaintiff who originally had a life expectancy of 80 years. A delayed diagnosis, however, created a situation in which there is a 20% chance the plaintiff will die by age 10, but she retains an 80% chance of beating the cancer and living to the originally expected age of 80. In other words, the plaintiffs new life expectancy considering the risk of early death is 66 years instead of 80 years, a reduction of 14 years. Under the majority’s theory, that plaintiff may sue her doctor while still living and recover the full amount of damages for these 14 years of life that have been theoretically lost. If a jury determined these 14 years of potential lost life to be worth $1,000,000, she could recover that $1,000,000 from her doctor even though it is highly probable (80%) that she will live as long as she would have without the negligence. And if the plaintiff does survive, then she has received $1,000,000 as “compensation” even though no ultimate harm ever befalls her. Allowing a plaintiff to recover damages when it is 80% likely that that the harm will never materialize defies our requirement that the plaintiff prove “future damages to a reasonable certainty.” Pietrzak, 295 N.W.2d at 507. Moreover, such an award is not “compensation” in any meaningful sense of the word, but a windfall at the expense of physicians and the healthcare system and an invitation to abuse.
As to deterrence, I agree that the prevention of medical negligence is a laudable goal. But it is not the only goal. Fairness and the tailoring of liability to those that have actually directly caused harm is also important. As the supreme courts of other states have aptly observed, “ ‘[i]f deterrence were the sole value to be served by tort law, we could dispense with the notion of causation altogether and award damages on the basis of negligence alone.’ ” Weymers v. Khera, 454 Mich. 639, 563 N.W.2d 647, 654 (1997) (quoting Kramer, 858 S.W.2d at 406).
But even if deterrence were the principal objective, I disagree with the majority’s contention that traditional tort liability fails to deter medical negligence. The majority seems to assume that, at the time negligent conduct is committed, physicians know whether their conduct will result in harm to the patient, and thus some physicians will not be deterred because there is no potential for liability. This is, of course, incorrect. Traditional tort principles deter all medical negligence because the physician is aware that any negligence could result in harm to the patient and open up the possibility for liability as a result.9 Further, there is little deterrent *349value in extracting a penalty from a physician if it cannot be shown that she in fact directly caused any harm. See Mohr v. Grantham, 172 Wash.2d 844, 262 P.3d 490, 499 (2011) (Madsen, C.J., dissenting) (stating that “[deterrence of negligence that does not cause actual harm is a meaningless proposition”). The prospect of liability for harms that the physician did not cause is not likely to deter negligence, but rather is more likely to “encourage the practice of costly defensive medicine in an attempt to avoid practically certain liability in the event of an unfavorable outcome.” Falcon v. Mem’l Hosp., 436 Mich. 443, 462 N.W.2d 44, 66 (1990) (Riley, C.J., dissenting). As such, the aim of deterrence does not justify the result today.
II.
In sum, the majority’s decision greatly expands the liability of medical professionals in this state and unfairly holds physicians liable for harms that may never materialize and, if they do occur, are not caused by the physician’s negligence. In so doing, the majority sua sponte overrules two well-established cases and undermines unbroken and fundamental principles of tort law. Such a drastic expansion of liability, especially in the healthcare field, implicates serious policy considerations, which are better addressed by the Legislature. Therefore, I would reverse the court of appeals and reinstate the order of the district court.
. These figures do not appear in the court’s opinion, but are set forth in the opinion of the court of appeals. Leubner v. Sterner, 483 N.W.2d 518, 520 (Minn.App.1992).
. The majority contends that we did not address loss of chance in Leubner because the issue was waived by plaintiff on appeal. Although we noted in Leubner that plaintiff did not file a petition for review to preserve their lost chance of survival claim, we nonetheless proceeded to address whether plaintiff could recover on that theory. In so doing, we rejected the argument that “the injury claimed to be caused is a decreased percentage chance of surviving, whether or not the patient, in fact, has survived." 493 N.W.2d at 121 (emphasis added). This is the precise injury that the majority recognizes as compensable in this case. The majority dismisses our determination in Leubner on the ground that it was mere "dicta.” But the fact remains that, in Leubner, we denied any recovery to a plaintiff on virtually identical facts to this case. No matter how the plaintiff chose to characterize her injury or legal theory, we held that it could not proceed because she could not show that the injury was the result of the defendant’s negligence. The majority’s decision ignores this rule of law.
. These figures do not appear in the court’s opinion, but are set forth in the opinion of the court of appeals. Fabio v. Bellomo, 489 N.W.2d 241, 243 (Minn.App.1992).
. We reaffirmed Fabio’s holding in MacRae v. Group Health Plan, Inc., 753 N.W.2d 711, 722 (Minn.2008) ("We rejected 'loss of chance’ due to reduced life expectancy and increased risk of recurrence as a theory of compensable damages....”).
. The majority's reference to a footnote in Curtis v. Altria Group, Inc., 813 N.W.2d 891 (Minn.2012) does not support its position. At best, that footnote states that, when a court is faced with alternative arguments, a decision resting on the first argument renders a decision on the second argument dictum. See Curtis, 813 N.W.2d at 901 n. 6. But in Fabio, the primary and first-articulated basis for our decision was our categorical rejection of the loss of chance doctrine. See 504 N.W.2d at 762. The second basis was our conclusion that "even if” we adopted the doctrine, plaintiff had not proven such a claim. See id. at 763. Therefore, even under the Curtis footnote, Fabio’s rejection of the loss of chance doctrine is the holding and the plaintiff’s failure to prove such a claim was dictum.
. The majority contends that the adoption of loss of chance was "squarely presented for our consideration” despite the fact that, in more than 100 pages of briefing, the parties did not address the primary rationales upon which the majority bases its decision. Specifically, neither the Dickhoffs nor Dr. Tollefsrud meaningfully addressed whether the court is bound by Fabio and Leubner; whether traditional tort rules adequately deter medical negligence: whether loss of chance should be treated as a distinct injury; and whether medical science has progressed to the point where physicians can accurately estimate chances of survival. Therefore, the majority's claim that the adoption of loss of chance has been squarely presented for our consideration lacks credibility.
. Appellants contend that a party cannot maintain a cause of action seeking compensation for her future death — no matter how probable — during her lifetime. In particular, appellants argue that such a claim- is essentially one for wrongful death, which did not exist at the common law and is entirely a creature of statute. See Beck v. Groe, 245 Minn. 28, 45, 70 N.W.2d 886, 897 (1955). While a party may recover for damages incurred before death (such as medical expenses and pain and suffering), the claim for death itself does not exist during a person’s lifetime because it "is created by the decedent’s death." Regie de l'assurance Auto. du Quebec v. Jensen, 399 N.W.2d 85, 89 (Minn. 1987). As noted above, I need not decide this issue now and I will assume for the sake of argument that damages for Jocelyn’s eventual death (or decreased length of life) could be recoverable as an item of future damages under traditional tort principles.
. The majority conspicuously ignores the other item of damage sought by the Dickhoffs— the medical expenses and pain and suffering related to the 2010 recurrence of Jocelyn’s cancer — and provides no guidance to the district court on how those damages should be analyzed in this case. As noted above, the Dickhoffs cannot prove that Dr. Tollefsrud caused the recurrence. Indeed, the Dickhoffs’ expert testimony stated only that Dr. Tollefs-rud changed the likelihood of recurrence "from unlikely to probable,” which could mean that the likelihood of recurrence merely changed from 49% to 51%. As such, recovery of 100% of the costs of that recurrence is inappropriate. Arguably, the plaintiffs could *348pursue "proportional recovery” by multiplying the total cost of recurrence by the percentage chance of recurrence caused by the doctor’s negligence. But plaintiffs' expert, Dr. Forman, could not explain what that percentage was and admitted that "[£ ]i is impossible to put precise statistics " on the likelihood of recurrence with or without timely care. This underscores the speculative nature of the Dickhoffs’ recovery here and demonstrates why we should reverse at least with respect to this item of damage.
. For a discussion of the relative deterrent value of traditional tort rules and the loss of chance doctrine, see generally David A. Fischer, Tort Recovery for Loss of a Chance, 36 Wake Forest L.Rev. 605, 631-32 (2001) (noting that most physicians are adequately deterred by the "all or nothing rule” because they "diagnose and treat a variety of patients with a variety of ailments, giving rise to widely varying chances of a cure”).