¶ 98. {dissenting). The majority reverses the court of appeals not because of any legal or factual error. Rather, the majority in essence disregards the standard of review and reverses the court of appeals simply because it does not agree with the outcome.
¶ 99. The standard of review in this case requires us to determine whether the court of appeals erroneously exercised its discretion because it committed a legal or factual error. In order to appear as if it reviews this matter for an erroneous exercise of discretion, the majority takes the court of appeals to task for failing to analyze whether this case is "truly exceptional." Majority op., ¶ 55.
¶ 100. The problem with the majority's analysis is that what constitutes an "exceptional" case is in the eye of the beholder. What the majority is really doing is substituting its own evaluation of whether this is such an exceptional case in place of the court of appeals' reasoned determination that this case warrants a new trial in the interest of justice.
¶ 101. By conducting what is really an independent review, the majority subverts the court of appeals' discretion to order a new trial in the interest of justice. Vollmer v. Luety, 156 Wis. 2d 1, 15, 456 N.W.2d 797 (1990). Because a straightforward review for an erroneous exercise of discretion would reveal that the court *454of appeals has committed no error of fact or law that requires reversal, I respectfully dissent.
I
¶ 102. The majority correctly sets forth the two-part test for determining whether a new trial should be granted in the interest of justice when the controversy has not been fully tried. Majority op., ¶ 38 n.18. This court has concluded that the real controversy was not fully tried in two situations. State v. Hicks, 202 Wis. 2d 150, 160, 549 N.W.2d 435 (1996). First, when the jury was erroneously denied the opportunity to hear important evidence bearing on an important issue in the case. Id. Second, when the jury had before it evidence not properly admitted that "so clouded" a crucial issue that it may be fairly said that the real controversy was not fully tried. Id.
¶ 103. The majority also correctly states the standard of review for an order of the court of appeals granting a new trial in the interest of justice: we review to determine whether the court of appeals erroneously exercised its discretion. See Majority op., ¶ 38. Nevertheless, in the very next paragraph, the majority concludes that "[t]he real controversy here was fully tried," in essence deciding the issue on its merits de novo. Id., ¶ 39.
¶ 104. Although the majority pays lip service to the correct standard of review, its approach to this case is evident from its declaration that Avery asks this court to "grant him a new trial in the interest of justice." Id., ¶ 54. Here, however, there is a fly in the majority's ointment — a new trial has already been granted. The court of appeals analyzed this very issue and concluded *455that the real controversy was not fully tried. State v. Avery, 2011 WI App 148, 337 Wis. 2d 560, ¶ 37, 807 N.W.2d 638.
¶ 105. The majority mines through the facts of Hicks and Armstrong1 in an attempt to distinguish the facts in those cases from the facts here. Majority op., ¶¶ 43-53. Ultimately it concludes that unlike Hicks and Armstrong, this is not a "truly exceptional" case. Majority op., ¶ 55.
¶ 106. The majority then proceeds to independently evaluate the merits of the photogrammetry evidence and its effects on the trial. Id., ¶ 56. Although it acknowledges that the jury saw the video in this case but did not hear the photogrammetry evidence, it dismissively concludes that the photogrammetry evidence does not "discredit[] a pivotal piece of evidence" at trial. Id. Because the State "did not meaningfully rely on the video at trial," the majority dismisses the photogrammetry evidence's effect on the proceeding entirely. Id. Finally, although it cloaks its conclusion with the term "erroneous exercise of discretion," it determines that this is not an exceptional case and that Avery is not entitled to a new trial in the interest of justice. Id., ¶ 59.
II
¶ 107. The problem with the majority's analysis is that one person's "exceptional" is another person's "routine." The "exceptional" is to be found in the eyes of the beholder. The exact outline of what constitutes an exceptional case is merely a matter of opinion based on one's view of the facts. That is precisely the reason this *456court has traditionally accorded the court of appeals "substantial discretion" and has "liberally construed" its power of discretionary reversal under Wis. Stat. § 752.35.2 Vollmer, 156 Wis. 2d at 15.
¶ 108. In Vollmer, this court recognized that the discretionary power of reversal is "compatible with doing justice in an individual case, which is primarily the duty of the court of appeals." Id. That power is consistent with the court of appeals' error-correcting function, which this court has recognized means that the court of appeals is the court of last resort in many cases. State v. Schumacher, 144 Wis. 2d 388, 408, 424 N.W.2d 672 (1988). The court of appeals is thus accorded "substantial discretion" when it orders a new trial in the interest of justice under Wis. Stat. § 752.35. Vollmer, 156 Wis. 2d at 15.
¶ 109. Therefore, the proper role of this court on review is solely to determine whether the court of appeals committed an error of fact or law that requires reversal. See Stivarius v. DiVall, 121 Wis. 2d 145, 153 n.5, 358 N.W.2d 530 (1984) (discretionary reversals by the court of appeals are reviewed for an error of law). Because no one argues that an error of fact was committed by the court of appeals, the review in this *457case should be confined to whether an error of law was committed when the court of appeals granted a new trial because the real controversy was not fully tried. The majority appears to abandon the proper standard of review simply because it disagrees with the result in this case.
¶ 110. Ultimately, the majority today has sent a message to the court of appeals that regardless of the statutory authority given to the court of appeals to grant a new trial in the interest of justice pursuant to Wis. Stat. § 752.35, it should not be entrusted to exercise that discretion. Instead, this court will not be bound by the standard of review and will second-guess the court of appeals' discretionary determination even though there is no mistake of law or fact in its decision.3
*458Ill
¶ 111. A straightforward application of the correct standard of review reveals no error of fact or law that requires reversal. As I note above, the court of appeals' grant of a new trial in the interest of justice is properly reviewed solely for an erroneous exercise of discretion. No one argues that the court of appeals committed an error of fact, so the sole question on review should be whether the court of appeals' exercise of discretion was based on an error of law.
¶ 112. This court has concluded that the real controversy was not fully tried in two situations. Hicks, 202 Wis. 2d at 160. First, when the jury was erroneously denied the opportunity to hear important evidence bearing on an important issue in the case. Id. Second, when the jury had before it evidence not properly admitted that "so clouded" a crucial issue that it may be fairly said that the real controversy was not fully tried. Id.
¶ 113. Here, the first situation is arguably satisfied. The photogrammetry evidence was not heard by the jury but bears on a critical issue in the case. Identification is nearly always an important issue in a criminal case, but in this case it was Avery's principal defense. He testified that he was not a participant to the robberies and that his confession was coerced. Avery produced several alibi witnesses who corroborated his version of the events.
¶ 114. In contrast, the State's case was heavily dependent on two eyewitnesses to the robberies. This court has been critical of the reliability of eyewitness identification testimony, observing that studies confirm that eyewitness testimony is often "hopelessly unreli*459able."4 State v. Dubose, 2005 WI 126, ¶ 30, 285 Wis. 2d 143, 699 N.W.2d 582. In Dubose, we cited to a United States Department of Justice study reporting that 85% of wrongful convictions in the study were based primarily on misidentification. Id. This court stated:
The research strongly supports the conclusion that eyewitness misidentification is now the single greatest source of wrongful convictions in the United States, and responsible for more wrongful convictions than all other causes combined. In a study conducted by the United States Department of Justice of 28 wrongful convictions, it determined that 24 (85 percent) of the erroneous convictions were based primarily on the misidentification of the defendant by a witness.
Id. (Citations omitted.)
¶ 115. Studies also reveal an alarming number of false confessions. Since 1992 the Innocence Project has used DNA evidence to help exonerate 301 people who *460were wrongly convicted of a crime.5 About 25 percent were found to have given a false confession or plead guilty to the crime for which they were subsequently exonerated.6 Given the "hopelessly unreliable" nature of eyewitness identification testimony and the alarming incidence of false confessions, photogrammetry evidence that tends to show Avery was not a robber at Malone's Fine Foods certainly could reasonably be considered evidence not heard by the jury that bears on an important issue in the case.7 Hicks, 202 Wis. 2d at 160.
¶ 116. It is not an error of law to conclude if the jury believed the photogrammetry evidence, Avery could not have been present for the Malone's Fine Foods robbery. His alibis for both robberies are likely to be viewed much more favorably by a jury if he could show, as the photogrammetry evidence arguably does, that he was not involved in at least one of the two robberies.8
*461¶ 117. Because it committed no legal or factual error, I conclude that the court of appeals' decision was not an erroneous exercise of discretion. Rather, the error here lies in the majority's substitution of its own evaluation of this case in place of the court of appeals' reasoned determination that this case warrants a new trial in the interest of justice. Accordingly, I respectfully dissent.
¶ 118. I am authorized to state that CHIEF JUSTICE SHIRLEY S. ABRAHAMSON joins this dissent.State v. Armstrong, 2005 WI 119, 283 Wis. 2d 639, 700 N.W.2d 98.
Wisconsin Stat. § 752.35 states the following:
Discretionary Reversal. In an appeal to the court of appeals, if it appears from the record that the real controversy has not been fully tried, or that it is probable that justice has for any reason miscarried, the court may reverse the judgment or order appealed from, regardless of whether the proper motion or objection appears in the record and may direct the entry of the proper judgment or remit the case to the trial court for entry of the proper judgment or for a new trial, and direct the making of such amendments in the pleadings and the adoption of such procedure in that court, not inconsistent with statutes or rules, as are necessary to accomplish the ends of justice.
The court of appeals has granted only a handful of new trials in the interest of justice in recent years. See, e.g., State v. Tanon, No. 2009AP491, unpublished slip op. (Ct. App. Mar. 18, 2010); State v. Vandenberg, No. 2009AP1242, unpublished slip op. (Ct. App. Aug. 24,2010); State v. Louis, No. 2009AP2502-CR, unpublished slip op. (Ct. App. Mar. 15, 2011); State v. Davis, 2011 WI App 147, 337 Wis. 2d 688, 808 N.W.2d 130; State v. Jeffrey A.W., 2010 WI App 29, 323 Wis. 2d 541, 780 N.W.2d 231; State v. Berard, No. 2010AP2439-CR, unpublished slip op. (Ct. App. Jan. 25, 2012); State v. Howland, No. 2011AP532-CR, unpublished slip op. (Ct. App. Dec. 14, 2011). There is no evidence that the court of appeals utilizes its power of discretionary reversal except in limited and extreme circumstances.
Not one example cited above mentions the word "exceptional," much less analyzes what exceptional circumstances led the court of appeals to grant a new trial. It appears that under the majority's analysis, nearly every instance where the court of appeals has ordered a new trial in the interest of justice in recent years was an erroneous exercise of discretion. See also State v. Plude, 2008 WI 58, ¶¶ 52-102, 310 Wis. 2d 28, 750 N.W.2d 42 (Ziegler, J., concurring) (concluding that granting a new trial in the interest of justice was required without analyzing whether the case was exceptional).
In light of the voluminous body of scientific knowledge that questions the reliability of eyewitness identifications, the Oregon Supreme Court recently established new procedures for the admissibility of eyewitness identifications. These procedures include shifting the burden of proof to the State to show that eyewitness identifications are sufficiently reliable to be admissible as evidence. State v. Lawson, 291 P.3d 673, 2012 WL 5955056 (Or. 2012). The Oregon Supreme Court decision followed on the heels of a decision by the New Jersey Supreme Court which also recognized the need to establish new procedures for the admissibility of eyewitness identifications. State v. Henderson, 27 A.3d 872 (N.J. 2011). See also Honorable Stuart Rabner, Evaluating Eyewitness Identification Evidence in the 21st Century, 87 N.Y.U. L. Rev. 1249 (2012); Office of the Wisconsin Attorney General, Model Policy and Procedure for Eyewitness Identification, http://www.doj.state.wi.us/dles/tns/ eyewitnesspublic.pdf (last visited Dec. 19, 2012).
Innocence Project, http://www.innocenceproject.org/index.php (last visited Nov. 20, 2012).
Innocence Project, False Confessions, http://www.innocenceproject.org/understand/False-Confessions.php (last visited Nov. 20, 2012).
Several academic studies find that false confessions are surprisingly easy to induce from innocent confessors. See, e.g., Jennifer T. Perillo & Saul M. Kassin, Inside Interrogation: The Lie, The Bluff, and False Confessions, 35 Law & Hum. Behav. 327 (2011); False Confessions: Silence is Golden, The Economist, Aug. 12, 2011.
The court had the benefit of a non-party brief filed by distinguished academics and forensic scientists. The brief criticizes the methodology used by the prosecutor's expert for the "radically non-blind" nature of his evaluation. It concludes that "[a]ny rational factfinder should give it little weight, and it certainly should not be treated as dispositive in this case."
The majority opinion repeatedly refers to and applies the "reasonable probability test." Its holding in footnote 16 leaves for another day the standard for the reasonable probability test. *461I am concerned about the effect of this holding on existing case law that has adopted a standard.