¶ 164. (dissenting). The majority opinion reverses the court of appeals decision affirming the circuit court's order denying Ralph Armstrong's motion to vacate his judgment of conviction, and it then concludes that Armstrong is entitled to a new trial. Majority op., ¶ 2. The majority opinion does so based on its conclusion that the results of DNA tests that have been recently completed prove the real controversy was not fully tried. Majority op., ¶ 2. However, the actual issue in this case is whether the DNA evidence, which is newly discovered, creates a reasonable probability of a different outcome at a new trial. See State v. McCallum, 208 Wis. 2d 463, 474, 561 N.W.2d 707 (1997). The majority opinion is able to side-step our well-established jurisprudence for newly discovered evidence and conclude that Armstrong is entitled to a new trial only by avoiding the crucial analysis of whether this DNA evidence creates a reasonable probability that a different result would be reached at a new trial. Because I conclude that this evidence does not create a reasonable probability that a different result would be reached at a new trial and because I conclude that the real controversy, whether Armstrong raped and murdered Charise Kamps, was fully tried in 1981, I respectfully dissent from the majority opinion. Accordingly, I would affirm the court of appeals.
I. DISCUSSION
¶ 165. Armstrong's claim of newly discovered evidence sufficient to vacate his conviction is based on recent DNA testing of two hairs found in Kamps' *706apartment and semen stains found on Kamps1 bathrobe showing neither the hair nor the semen is his. DNA testing was not available in 1981 when Armstrong was tried.
A. Newly Discovered Evidence
¶ 166. In order to set aside a judgment of conviction, newly discovered evidence must be sufficient to establish that a defendant's conviction was a manifest injustice. State v. Krieger, 163 Wis. 2d 241, 255, 471 N.W.2d 599 (Ct. App. 1991). The test for determining whether the proffered evidence is "newly discovered" and whether it meets the test of "manifest injustice" has been explained many times. The court of appeals clearly set out the criteria a defendant must meet in order to overturn a conviction based on an allegation of newly discovered evidence as follows:
(1) The evidence must have come to the moving party's knowledge after a trial; (2) the moving party must not have been negligent in seeking to discover it; (3) the evidence must be material to the issue; (4) the testimony must not be merely cumulative to the testimony which was introduced at trial; and (5) it must be reasonably probable that a different result would be reached on a new trial.
State v. Avery, 213 Wis. 2d 228, 234, 570 N.W.2d 573 (Ct. App. 1997) (quotation omitted). If a defendant proves the first four criteria set out above by clear and convincing evidence, then "the circuit court must determine whether a reasonable probability exists that a different result would be reached" at a new trial. McCallum, 208 Wis. 2d at 473. The first four criteria are questions of fact that are not contested here. See id. at 473.
*707¶ 167. The fifth criterion sets up a question of law, i.e., whether the facts of the case meet the legal standard of a reasonable probability of a different outcome at a new trial.1 Id. In assessing this legal standard, we must determine whether there is a reasonable probability that a jury, looking at all the relevant evidence in regard to whether the defendant did or did not commit the crime, would have reasonable doubt as to the defendant's guilt. See id. at 474. This examination requires an assessment of all the evidence to determine what effect, if any, the newly discovered evidence would be reasonably probable to have on a jury's verdict at a new trial. See id.
¶ 168. Part of the new evidence proffered by Armstrong is DNA testing that shows that two head hairs found on the belt of Kamps' bathrobe did not come from him. One of these hairs had previously been characterized as "consistent" with Armstrong's hair and the other had been characterized as "similar" to Armstrong's hair. At trial, Coila J. Wegner, the State's expert, testified about her examination of eight exhibits that contained hair samples taken from Kamps' apartment. She explained that the tests she ran could exclude donors of the hair, but not identify them. When she described the hair from eight exhibits, she explained that for 36 of the hairs she could not exclude Kamps as the donor, but she *708could exclude Armstrong as the donor. For six of the hairs,2 she could not exclude Armstrong as the donor, but she could exclude Kamps. Only two of the six hairs for which Armstrong could not be excluded as the donor were subjected to DNA testing. For thirteen hairs, both Kamps and Armstrong were excluded as donors, and nine of the hairs she examined were animal hairs. A review of Wegner's testimony about what she said she could determine relative to the donors of the hair samples is helpful to a consideration of how important this newly discovered evidence is in the context of all the evidence presented at trial. Wegner testified as follows:
Q: And what was the result of that comparison?
A: The head hair was consistent in microscopic characteristics with the standard head hair from Miss Kamps. It was not consistent in microscopic characteristics with the hair from Mr. Armstrong.
Q: Now let me understand something. With hair analysis when you say it was consistent, you can say that testimony or she cannot be eliminated as the source of the hair?
A: That is correct.
Q: And when you say it is not consistent, you are saying that that person or that standard is eliminated as a source of the hair?
*709A: That is correct.
Q: So this is a hair or examination which only excludes a person, it never includes a person?
A: It could include them, but not identify them.
Q: It can't identify them?
A: No, sir.
Q: Is there any — any method whatsoever similar to fingerprints for identifying a given hair with a given person where you can say with a certainty or to a reasonable degree of scientific certainty that this hair came from this person?
A: In ninety-nine point nine (99.9%) percent of the time you could not say that a specific hair came from a specific individual.
Wegner was very clear about the probative value of the hair analyses she completed.
¶ 169. In closing argument, the district attorney argued that the hairs for which Armstrong could not be excluded as a donor were his and tied him to Kamps1 murder. Armstrong's attorney argued that the hair analyses did not identify Armstrong as the donor of the hairs, only that he could not be excluded as the donor. He also argued that Armstrong admitted to being in Kamps' apartment earlier in the evening of her murder and because hairs move freely from place to place, the presence of those hairs did not show Armstrong committed the crimes. He said that the movement of hair was demonstrated by the presence of animal hairs in Kamps1 apartment when she never had a pet.
*710¶ 170. Armstrong also proffered DNA testing of semen samples from Kamps1 bathrobe that showed they came from Brian Dillman, Kamps' flaneé, as newly discovered evidence. At trial, Wegner testified that a semen stain that was found on Kamps' bathrobe was made by a Type-A secretor. She testified that both Armstrong and Dillman are Type-A secretors. Wegner also testified that 80% of the population are Type-A secretors. In closing argument, the district attorney said that the semen stains were made by a Type-A secretor and that Armstrong was a Type-A secretor. Armstrong's attorney carefully explained that Dillman and Kamps were lovers, and because Dillman was a Type-A secretor, the semen was his. He also repeated Wegner's testimony that 80% of the population are Type-A secretors. Therefore, the jury could not have given this evidence much weight in reaching its verdict that Armstrong raped and murdered Kamps.
¶ 171. There is another fact that bears on the testimony about semen. Kamps was raped anally and vaginally with a hard object. Her injuries were not caused by being raped with a body part. Therefore, it is understandable that semen from the perpetrator of this crime was not left at the crime scene.
¶ 172. It is important to keep in mind that the DNA evidence Armstrong proffers is not exonerating evidence as DNA evidence can sometimes be. Instead, this evidence affects only one part of one of the five categories of evidence the State presented to the jury. And, it does not affect Armstrong's defense: that he was at Kamps' apartment, but not at the time of her murder. Physical evidence that was presented to the jury included Armstrong's fingerprint on a bong in *711Kamps' apartment;3 six hairs; possible connection to the semen stains on Kamps1 bathrobe; human blood around all 10 of Armstrong's fingers and around his toes, except for his two little toes, and blood on his watch; blood smearing on Kamps' body and face, as though she had been "finger-painted" with her own blood;4 Armstrong's deposit of $315 later on the morning of the murder, when the $400 Armstrong had paid to Kamps was missing from her apartment after the murder; and the lack of a forced entry into Kamps' apartment the night of the murder. The DNA test results do not affect most of this physical evidence.
¶ 173. The majority opinion implies that the testimony of Dr. Edward Blake that he could not detect blood when he examined a piece of cloth and accompanying slides prepared from the scrapings from Armstrong's thumbs and great toes undermines Wegner's testimony that she detected human blood around all Armstrong's fingers and around most of his toes. Majority op., ¶ 95. However, the circuit court found that Blake's testimony in this regard was not credible. This is a finding that we are not free to disregard. Micro-Managers, Inc. v. Gregory, 147 Wis. 2d 500, 512, 434 N.W.2d 97 (Ct. App. 1988) (the determination of a witness's credibility is for the circuit court). Therefore, the blood evidence potentially linking Arm*712strong to the murder has not been refuted by the defense, and remains part of the State's case.
¶ 174. In refusing to apply the newly discovered evidence test, the majority opinion improperly ignores the mountain of other evidence incriminating Armstrong that is not affected in any way by the DNA test results at issue here: (1) the time evidence presented by the State, showing that Armstrong could not have been at Kamps' apartment between 9:10 and 9:30 p.m. as he testified at trial that he was; (2) Armstrong's fingerprint in Kamps' apartment; (3) the eyewitness testimony that placed Armstrong's car at Kamps' apartment at the time of the murder; (4) the eyewitness testimony placing Armstrong at Kamps' apartment at the time of the murder; (5) the missing $400 from Kamps' apartment and Armstrong's deposit of $315 the next day;5 (6) the lack of a forced entry into Kamps' apartment, suggesting she voluntarily let in her murderer; (7) the romantic interest Armstrong had in Kamps and her rebuff of that interest; (8) the human blood around all 10 of Armstrong's fingers and on his toes and the blood on his watch; (9) the parking ticket showing Armstrong's car was not parked near the back door of May's apartment where he said he entered; (10) the testimony that someone was heard entering May's *713apartment building between 3:30 and 5:00 the morning Kamps was killed; (11) the repetitive lies Armstrong told to law enforcement and to the jury; (12) the lack of trial testimony by Armstrong's brother, who could have corroborated Armstrong's testimony that his brother was the source of the $315 Armstrong deposited and who could have verified part of Armstrong's alibi; and (13) the lack of an affidavit from Armstrong's brother for these motions. The majority opinion errs in its utter disregard of this mountain of evidence.
¶ 175. In regard to the issue of time, the majority opinion repeats Armstrong's mantra that he left his apartment in Fitchburg at 9:10 p.m., drove to May's apartment in Madison and dropped her off, drove to Kamps' apartment, spent about 15 minutes at Kamps' apartment drinking a beer, drinking some orange juice and playing music and then drove to Brent Goodman's house to buy cocaine, arriving there by 9:30 p.m.6 *714Majority op., ¶¶ 22-24. Armstrong had to place himself at Kamps' between 9:10 and his arrival at Goodman's at 9:30, if he was to cover for any evidence he may have left at Kamps' apartment at the time of the murder. However, it is not possible to do all Armstrong says he did in the 20 minutes between when he left his apartment and arrived at Goodman's. I agree with the circuit court that if the jury had believed Armstrong's time evidence, he would not have been convicted.
¶ 176. The majority opinion also ignores how the immediate and unwavering description of a car that matched Armstrong's car as the vehicle seen at Kamps' apartment at the time of the murder, strengthens the eyewitness's identification of Armstrong as the man who drove that car and went in and out of Kamps' apartment three times after midnight on the night Kamps was murdered. The majority opinion diminishes Riccie Orebia's identification of Armstrong as the man Orebia saw because Orebia was a reluctant witness and had been hypnotized. However, Orebia gave an accurate description of Armstrong's car long before being hypnotized. That description never changed.
¶ 177. The jury's decision reflects its consideration of all five categories7 of evidence the State presented. It cannot be ignored that if the jury had believed *715Armstrong's trial testimony About his being at Kamps' earlier in the evening, while also believing the two strands of hair were his, the jury would have acquitted him. This is so because the jury was offered an explanation of how Armstrong's hairs could have attached to the bathrobe belt. As the crime scene photo shows, the bathrobe belt was placed over Kamps after she was murdered and smeared with her own blood. Therefore, that belt must have been elsewhere in the apartment, where it could easily have picked up the hairs that were found on it, prior to its being placed on Kamps' body. If the jury had believed Armstrong, they would have believed the explanation Armstrong's attorney provided. However, the jury saw Armstrong testify. The jury did not believe him. By refusing to apply the newly discovered evidence test, the majority opinion ignores that crucial credibility determination.
¶ 178. The State presented an extraordinary amount of evidence, from a fingerprint to eyewitness identification of both Armstrong and his car, to prove that Armstrong was Kamps' murderer. The question the evidence presented at trial was, "Given the evidence before you, did Armstrong murder Kamps?" Taking away a piece of evidence from all that was presented in this case does not change the ultimate question. Armstrong testified that he was in Kamps' apartment. His defense was that he was not there when she was killed. He said he was with his brother for part of the time and with May for part of it. His brother did not testify and May could not say when he returned to her apartment. Evidence that would show he could not have been at Kamps' apartment when the eyewitness said he was would be significant in regard to the results at a new *716trial. The DNA evidence presented here does not affect the time testimony and the eyewitness testimony of both Armstrong's car and of him, which were critical to his conviction.
¶ 179. The newly discovered evidence offered here is much different in its impact from the evidence that was presented in State v. Hicks, 202 Wis. 2d 150, 549 N.W.2d 435 (1996).8 In Hicks, the question was whether Hicks had ever been in the victim's apartment. There were only two pieces of evidence tying him to that apartment at trial, one of which was later disproved. The impact of the evidence in Hicks is a far cry from the impact of the DNA evidence Armstrong proffers because of the overwhelming amount of evidence that was presented to the jury in Armstrong's case. Accordingly, I conclude that the newly discovered DNA evidence does not make it reasonably probable that a different result would be reached.
B. Real Controversy Not Fully Tried
¶ 180. Instead of applying the newly discovered evidence test as I have above, a test that Armstrong fails to pass, the majority reverses Armstrong's judgment of conviction on the theory that the real controversy was not fully tried. Majority op., ¶ 156. In doing so, the majority opinion misapplies our precedent and equates the idea of the "matter not being fully tried" with new scientific identification procedures in a way that threatens to reopen convictions statewide every *717time a scientific improvement occurs, regardless of the lack of a probable effect on the issues underlying the jury's verdict. Because the facts of this case do not meet the criteria necessary to reversing a conviction under our long-standing jurisprudence regarding the reed controversy being fully tried, I would not reverse the court of appeals on this basis.
¶ 181. The ability of this court to set aside a conviction through the use of our discretionary-reversal powers has often been discussed.9 In State v. Schumacher, 144 Wis. 2d 388, 424 N.W.2d 672 (1988), we identified two avenues for its use: when the real controversy has not been fully tried and when there has been a miscarriage of justice. Id. at 400. The proper analysis of a motion to set aside a conviction based on our discretionary-reversal powers was carefully laid out in Schumacher and many cases since then.
[Ujnder the "real controversy not fully tried" category, two different situations were included: (1) Either the jury was not given an opportunity to hear important testimony that bore on an important issue in the case, or (2) the jury had before it testimony or evidence which had been improperly admitted, and this material obscured a crucial issue and prevented the real controversy from being fully tried.
Under the second prong of the discretionary-reversal statute, the "miscarriage of justice" prong, the case law made clear that, in order to grant a discretionary reversal under this prong, the court would have to conclude that there would be a substantial probability that a different result would be likely on retrial.
*718Id. at 400-01 (citing State v. Wyss, 124 Wis. 2d 681, 741, 370 N.W.2d 745 (1985)). As we explained in Schumacher and have repeated many times since, "this broad discretionary-review power ... is ... to be used sparingly, and only in exceptional circumstances." Schumacher, 144 Wis. 2d at 407 (citing State v. Cuyler, 110 Wis. 2d 133, 141, 327 N.W.2d 662 (1983)).
¶ 182. The majority opinion seems to rely on a belief that the hair testimony "obscured a crucial issue," thereby preventing the real controversy from being fully tried. Majority op., ¶ 115. Its discussion focuses mainly on Hicks. Majority op., ¶¶ 117-35. The majority opinion bases its decision on what it characterizes as the "striking similarities" between Hicks and the present case. Majority op., ¶ 116.
¶ 183. I do not agree that Hicks and the present case are similar. Instead, as I explain below, the two cases are dissimilar in all respects that are material to whether the real controversy was fully tried. In Hicks, the issue the majority opinion turned upon was whether Hicks' claim that he had never been in the victim's apartment was fully tried due to Hicks' attorney choosing not to pursue DNA testing of hairs recovered there. Hicks, 202 Wis. 2d at 163-64.
¶ 184. Hicks' presence in the victim's apartment, or the lack of his presence, was pivotal to the case because Hicks is an African-American and the victim said that no other African-American male had been in her apartment except the perpetrator of the crime. Id. at 155. Five African-American hairs were found in the victim's apartment. It was the State's theory at trial that all five hairs came from the same person: the perpetrator. Id. at 165. It was Hicks' defense "that he had never been in [the victim's] apartment and could not have been the source of hairs that were found *719there." Id. at 163. Although DNA testing was available at the time of Hicks' trial, his trial counsel chose not to have the hair evidence DNA tested. Id. at 155.
¶ 185. After Hicks' conviction, DNA analysis was performed on the hair specimens. The results obtained from some of the specimens were inconclusive, but on two specimens, Hicks was ruled out as the source of the DNA. Id. at 156. Therefore, the following syllogism was set up: if all the hair came from the same person, the hairs were from an African-American, and the only African-American who had been in the victim's apartment was the perpetrator of the crime, then Hicks could not have been the perpetrator. Accordingly, we concluded that the issue of whether Hicks had been in the victim's apartment was not fully tried and Hicks was entitled to a new trial. Id. at 171-72.
¶ 186. In the present case, as in Hicks, hair recovered from the crime scene was inculpatory of the defendant at the time of trial, and some of it was later proved not to be the defendant's. However, there the similarity to Hicks ends. Armstrong said he had been in Kamps' apartment. Therefore, finding hair consistent with Armstrong's did not undermine his defense. The time evidence and the eyewitness identifications of his car and of him at Kamps' apartment during the time when she was murdered undermined his defense extensively.
¶ 187. There is another point that bears mentioning with regard to the DNA evidence at issue here. What was presented at trial was not "false evidence," a characterization of Armstrong's that the court of appeals picked up. State v. Armstrong, Nos. 2001AP2789 and 2002AP2979, unpublished slip op., ¶ 37 (Wis. Ct. App. May 27, 2004). Wegner's testimony about her analyses of the hair samples explained that Armstrong *720could not be excluded as the donor of six hairs by the tests that she ran. She also explained that those tests could not identify a hair's donor. There was nothing "false" about this testimony. It accurately described the capability of the tests she conducted relative to the hairs analyzed. That there are more accurate DNA tests now available does not change the capability of Wegner's tests or cause her testimony about them to become "false."
¶ 188. In the present case, the majority opinion states, "First the jury did not hear important DNA evidence that bore on an important issue of the case." Majority op., ¶ 115. Of course the jury did not hear the DNA evidence. It did not exist at the time of the trial. Likewise, the expert testimony regarding the hairs found in Kamps' apartment was properly admitted at the time of the trial. It strains the meaning of "fully tried" to suggest that Armstrong's case was not fully tried because the scientific bases for physical evidence set forth in the trial were only state-of-the-art at the time of the trial, but not state-of-the-art at present. Using the majority's standard, the real controversy can never be fully tried because scientific advances in evidence gathering and analysis will continue to improve. The majority opinion's explanation of how the DNA evidence fits into the theory that the real controversy was not fully tried shows the fallacy in using that test and is an additional reason why the proper avenue to handle cases where new evidence is obtained is, as I have explained above, the newly discovered evidence test. The newly discovered evidence test is best suited to analyzing the new evidence in the context of its impact on all the other evidence presented at trial. Therefore, for the foregoing reasons, I reject the use of *721discretionary reversal under the rubric of the real controversy not fully tried.
II. CONCLUSION
¶ 189. Accordingly, because I conclude that the DNA evidence does not create a reasonable probability that a different result would be reached at a new trial, and because I conclude that the real controversy, whether Armstrong raped and murdered Charise Ka-mps, was fully tried in 1981,1 respectfully dissent from the majority opinion.
¶ 190. I am authorized to state that Justices JON R WILCOX and DAVID T. PROSSER join this dissent.
Hairs for which Armstrong could not be excluded as the donor were found in Exhibit 27 (hair taken from the belt of Kamps' bathrobe), Exhibit 29 (hair taken from the bathroom sink) and Exhibit 43 (hair taken from blood and fecal-like material near Kamps' body).
This is the only fingerprint of Armstrong's found in the apartment, which is curious given that Armstrong testified that he drank a half glass of orange juice, a bottle of beer and played music on Kamps' turntable when he was there.
Ms. Wegner testified that she did not check the bathroom for blood because she was not asked to do so. Therefore, the jury heard no testimony about whether the murderer cleaned up in Kamps' bathroom before leaving the scene.
Armstrong did not mention his trip to the bank, but instead said that after leaving May's apartment, he drove around in Brittingham and James Madison parks looking for a source of cocaine before going to Kamps’ apartment. He told police that he got some cocaine from a "well-dressed" African-American male who was about "five-seven, five nine" with a medium "afro" and "a mustache that turns into sideburns." Armstrong changed this detailed story after he learned that law enforcement knew that he had deposited $315 the morning Kamps’ body was discovered.
In his brief to this court, Armstrong includes "Mapquest" printouts showing driving times, which he implies, show that he could have done all that he alleges before going to Brent Goodman's at 9:30 p.m. The State properly objected to this presentation as evidence never presented at trial. However, even using the times from Mapquest, it is not possible to do what Armstrong said he did because from his apartment to Kamps' address, Mapquest lists 12 minutes; from Kamps' to May's one minute and from Kamps' to Goodman's 10 minutes, a total of 23 minutes. This is not the route that Armstrong testified he drove, in regard to his stop at Kamps' before he went to Goodman's, because he said he first dropped May off at her apartment and changed from his car to Kamps' car. However, even adding the numbers from Mapquest shows a lapse of 23 minutes after he left his apartment in Fitchburg, when he had only 20 minutes available before Goodman testified he was at his house. The 23 minutes from Mapquest also includes no time for getting people in and out of the car, changing cars and going *714into Kamps' apartment for 15 minutes as Armstrong said he did, coming out, starting up the car and driving to Goodman's. The 23 minutes also allocates nothing for all the traffic lights along the routes shown on the Mapquest printouts. However you slice it, Armstrong could not have done all that he said he did in 20 minutes.
In order to summarize the evidence the State presented, the district attorney suggested it represented five categories: time evidence, testimonial evidence of Riccie Orebia and Laura Chafee, physical evidence, testimonial evidence of Armstrong's *715interest in Kamps which she rebuffed and Armstrong's lies to police and on the witness stand.
State v. Hicks, 202 Wis. 2d 150, 549 N.W.2d 435 (1996), was not argued as a newly discovered evidence case. Nor could it have been, because DNA tests were available and known to Hicks' attorney, who chose not to do them for what he believed were tactical reasons.
Our discretionary power to reverse judgments arises from both statute and common law. Vollmer v. Luety, 156 Wis. 2d 1, 13, 456 N.W.2d 797 (1990).