DISSENT
DIETZEN, Justice(dissenting).
In reversing the postconviction court, the majority dismisses crucial portions of the record, misconstrues the relevant case law, and misapplies the standard of review. Carnell Harrison and William Brooks did not recant their trial testimony or state that they did not testify truthfully at trial. Instead, when asked whether they testified truthfully at trial, Carnell stated he could *779not remember and Brooks did not answer the question. Citing Pippitt v. State, 737 N.W.2d 221, 228-29 (Minn.2007), the post-conviction court concluded that an alleged inability to remember is legally insufficient to satisfy the first prong of the Larrison recantation test. The court alternatively concluded that Caldwell failed to satisfy the second prong of the Larrison test because, even in the absence of the allegedly false testimony, the jury would have reached the same verdict based on the testimony of several other important witnesses. Without affording any deference to the postconviction court, the majority concludes that Caldwell is entitled to an evidentiary hearing. In reaching that conclusion, the majority relies on minor inconsistencies in the -witnesses’ trial testimony and taped interviews, and dismisses the trial testimony of six important witnesses.
The majority’s analysis of the first Lar-rison prong is flawed for three reasons. First, the majority dismisses Carnell and Brooks’s responses to the crucial question of whether they testified truthfully at trial. Second, the majority misconstrues the relevant case law by suggesting that a simple contradiction in a witness’s trial testimony and subsequent statement is sufficient, by itself, to satisfy the first prong of the Larrison test. Third, the majority affords no deference to the postconviction court’s application of Pippitt to the facts alleged in the taped interviews.
The majority’s analysis of the second Larrison prong is also flawed for two reasons. First, the majority dismisses the trial testimony of six important witnesses. Second, the majority affords no deference to the postconviction court’s conclusion that, even in the absence of the allegedly false testimony, the jury would have reached the same verdict based on the testimony of other important witnesses. This lack of deference is especially troubling because the postconviction court had the opportunity to personally observe all the trial testimony, and therefore was uniquely qualified to assess the impact of the allegedly false testimony.
In my view the record plainly supports the decision of the postconviction court to deny the petition without an evidentiary hearing. To explain the basis for my dissent, I will first clarify the record before our court, and then discuss my disagreement with the majority opinion.
I.
Caldwell argues that he is entitled to a new trial because the State’s key witnesses, Carnell and Brooks, recanted their trial testimony or otherwise testified falsely at trial. To fairly address Caldwell’s argument, an accurate understanding of the witnesses’ trial testimony, their subsequent taped interviews, and the postcon-viction court’s analysis is required.
At trial, Carnell testified that Caldwell owned the gun that killed Brian Cole and that Caldwell “gave [the gun] to [the shooter].” In describing how Caldwell passed the gun to the shooter, Carnell said, “I think over the seat.” During cross-examination, Carnell admitted that he was drunk at the time of the incident, and admitted he told a police investigator “[he] was like really drunk, like kind of blurry.” Brooks testified at trial that Caldwell passed the gun to the shooter “between the [car] door and the chair.” But when asked if he saw the transfer “with [his] own eyes,” Brooks provided a nonresponsive answer. When asked, “Did you see where [Caldwell] pulled the gun from ... before he handed it back to [the shooter],” Brooks replied, “No.” During cross-examination, Brooks admitted he told the grand jury, “I guess [Caldwell] passed [the shooter] the gun.”
Six other important witnesses testified for the State at trial, namely Cey Barber, *780Troy Young, B.M., E.F., P.P. and C.P. Collectively, these witnesses established the following facts. Caldwell and the shooter were members of a street gang that “ha[d] a problem” with the “One-Nine” gang.1 When Caldwell and his group stood on the corner of Eighth Street and Penn Avenue talking with some girls, one of the girls said, “[H]ere come the One-Nines, you all better leave,”2 and they observed a large group of One-Nines approaching on foot.3 Because Caldwell and his group were outnumbered, they fled one block west along Eighth Street and climbed into Caldwell’s SUV.4 Meanwhile, the One-Nines had turned east on Eighth Street.5 Caldwell drove east along Eighth Street.6 Moments later the shooter shouted, “[T]here go the One-Nines.”7 Caldwell then drove slowly past the One-Nines as the shooter fired several shots.8 Immediately after the shooting, Caldwell sped away without saying anything or otherwise expressing surprise that the shooter had fired on the One-Nines.9 A few weeks after the shooting C.P. overheard Caldwell bragging about killing the kid on Eighth and Penn.10 More specifically, C.P. told police that Caldwell “was bragging about shooting a dude saying that he shouldn’t have shot him because he was just a nobody, just a plain basketball kid. He was really looking for other Mook, One-Nine Mook.”11 When defense counsel cross-examined C.P. regarding his use of the pronoun “he,” C.P. clarified that Caldwell had said, “[W]e shouldn’t have shot him.”12 P.P. also testified that Caldwell told him “I’m the reason why you got that shirt,” referring to a t-shirt honoring the victim, Brian Cole.13
In closing argument, the prosecutor argued the State had established aiding and abetting liability, emphasizing that Caldwell drove “right to the One-Nines,” “slowed down for that shooting,” and then “sped off to get away.” The prosecutor also relied on Carnell and Brooks’s testimony that Caldwell had passed the gun to the shooter. Defense counsel told the jurors they should believe the testimony of three of the State’s witnesses, namely Cey Barber, Troy Young and E.F., because they had nothing to gain.14 Defense counsel then vigorously attacked the credibility of Brooks and Carnell. With regard to Brooks, defense counsel told the jurors,
On September 14th of 2007, he tells law enforcement, I guess [Caldwell] passed [the shooter] the gun. Let’s see, this act happens on June 17th of 2006, so *781what’s this, about a year and a couple months after the event, he first says, I guess. Then under oath he’s questioned again, I guess [Caldwell] passed the gun to [the shooter]. [The shooter] leaned out the window and just opened fire. Then later he says, yes, I saw [Caldwell] pass the gun. I guess it’s one of those things that happened in society that if you start saying stuff over and over again, it comes true? That makes no sense. Under oath, I guess. To the police, I guess. Then later, oh, yeah, I saw it....15
As for Carnell, defense counsel emphasized that Carnell admitted that at the time of the shooting he ivas “[d]ozing in and out drunk” and that he told the police, “I am prepared to tell the truth if I’m not getting charged for nothing. Drop my charges.”
The jury found Caldwell guilty as charged. The district court convicted Caldwell of the most serious count, first-degree murder for the behefit of a gang, and sentenced him to life in prison without the possibility of release. On direct appeal, petitioner did not challenge the sufficiency of the State’s evidence on the issue of aiding and abetting. Instead, lie argued the shooter lacked the necessary premeditation and intent to conimit first-degree murder. State v. Caldwell, 808 N.W.2d 373, 383-84 (Minn.2011). We affirnied. Id. at 397.
On May 29, 2012, Caldwell filed his current petition for postconviction relief. In support of his petition, Caldwell submitted transcripts of taped interviews of Carnell and Brooks that were conducted by a defense investigator.16 During the interview with Carnell, the following colloquy occurred.
Investigator: Ok and the testimony, just to be clear, the testimony you gave in court was that inaccurate?
Carnell: What I say in the court? Investigator: What you testified in court, was that truthful testimony?
Carnell: I say what I say, are you talking about everything I said?
Investigator: Yeah when you testified against [Caldwell] was that truthful?
Carnell: Ah I can’t remember. I can’t like it’s been so long ago. I believe some of it was but like some of it wasn’t cause the part that wasn’t was the part that I didn’t remember like I was like drunk through the situation, you know what I mean. So I didn’t really remember I was just trying to put it together like how I thought I went it or whatever. But I really didn’t remember and I can barely, it’s been so long that I can barely put it together again now cause it’s like the first time it came back up, you know what I mean, six years. But I am putting it together the best of my ability now without being under no pressure, without being threatened with no time, you know what I mean, or threatened with a murder or nothing like that, I can put it together I think a little better than I put it together back then because I’m under no pressure.
Exhibit B (emphasis added). A similar colloquy occurred during the interview with Brooks.
Investigator: And part of the reason he got the sentence is because of your testimony, is that correct?
Brooks: Yeah.
Investigator: And was that testimony truthful?
*782Brooks: Uh page 3 was, hold on a sec.... You talking about I’m trying to find where the part is that says about them arguing and then Kurt was arguing about something. I wasn’t like by there to hear them arguing about anything. Cause I was like kinda drunk cause they had a little liquor.
Exhibit A. Neither Brooks nor the investigator ever return to the question of whether Brooks testified truthfully at Caldwell’s trial. They did, however, have the following discussion about the gun transfer. Investigator: Where did [the shooter] get the gun?
Brooks: I really don’t know if the gun was just sitting on the side or if [Caldwell] passed it to him. I really don’t know.
[[Image here]]
Investigator: Ok did you see [Caldwell] pass [the shooter] the gun?
Brooks: I couldn’t really see anything. Cause if you going towards the door and the chair way you can’t even see him pass nothing.
Exhibit A (emphasis added). Brooks told the investigator: “I don’t know if [Caldwell] even passed the gun to Kirk.... [T]hat’s why ... I said in my testimony, ‘I guess he passed him the gun on the side.’ ” Exhibit A (emphasis added). The postconviction court accurately stated that a request for a new trial based on recantation, or false testimony of a witness, is evaluated under the Larrison test, which requires (1) the court be reasonably well satisfied that the trial testimony was false, (2) without the false testimony, the jury might have reached a different conclusion, and (3) the petitioner was taken by surprise at trial or did not know of the falsity until after trial. State v. Caldwell, 322 N.W.2d 574, 584-85 (Minn.1982). Indeed, to recant means to withdraw or renounce prior statements or testimony formally or publicly. Black’s Law Dictionary 1295 (8th ed.2004). Citing Ferguson v. State, 645 N.W.2d 437, 445 (Minn.2002), the post-conviction court expressly acknowledged that the showing required to obtain a post-conviction evidentiary hearing on a false-evidence claim was lower than the showing required to obtain a new trial. More specifically, the postconvietion court explained that Caldwell was entitled to an evidentia-ry hearing if he alleged facts that, if proved by a preponderance of the evidence, would entitle him to relief under the Larrison test. See Minn.Stat. § 590.04, subd. 1 (2012); Ferguson, 645 N.W.2d at 445.
The postconviction court relied upon Pippitt v. State, 737 N.W.2d 221, 228-29 (Minn.2007) to conclude that an “inability to recall” is legally insufficient to satisfy the first prong of the Larrison test that the court must be reasonably well-satisfied that the testimony in question was false. The court determined that “[Carnell] never recants his trial testimony,” emphasizing that “when specifically asked if his trial testimony was truthful, [Carnell] states that he cannot remember and he is just trying to remember what happened to the best of his ability.” Similarly, the court determined “Brooks never recants his trial testimony or admits that he lied in the recorded statement; he simply indicates that parts of his testimony may have been inaccurate because he no longer remembers the events of that day.”17
*783With regard to the second Larrison prong, the postconviction court concluded that even in the absence of the allegedly false testimony, the jury would have reached the same verdict because the testimony of several other important witnesses provided independent proof of Caldwell’s intent to further the commission of the crime. The court emphasized the following testimony. B.M. and P.P. testified Caldwell was driving the SUV. E.F. and P.P. testified the group that was shot at consisted of members of the One-Nine gang and that Caldwell had issues with that gang. Cey Barber testified the shooter “identified the group as members of the [One-Nine] gang and after that comment, [Caldwell] drove the car toward, the group.” Finally, C.P. and P.P. testified they heard Caldwell make comments following the murder indicating that he was involved.
Having clarified the record before the court, I now discuss my disagreement with the majority opinion.
II.
The majority dismisses the responses Carnell and Brooks gave in their taped interviews when they were directly asked whether their trial testimony was truthful. The majority contends, without citation, that a recanting witness need not publicly renounce his or her prior testimony with a statement like, “and by the way, I lied at trial.” Supra at 774. But this is not a case in which the witnesses simply failed to state that they lied at trial. Instead, it is a case in which the witnesses evaded direct questions on the issue of whether they testified truthfully at trial. In my view, the postconvietion court properly considered the responses Carnell and Brooks gave in their taped interviews when they were directly asked whether their trial testimony was truthful.
III.
Having dismissed the witnesses’ response to the critical question of whether their trial testimony was truthful, the majority plucks a handful of minor inconsistent statements from the interview transcripts that it contends support a claim of false trial testimony. In support of its contention, the majority cites Dobbins v. State, 788 N.W.2d 719, 735 (Minn.2010), and Ferguson v. State (.Ferguson II), 779 N.W.2d 555, 559-60 (Minn.2010), for the proposition that a contradiction in a witness’s trial testimony and subsequent statement is sufficient, by itself, to warrant an evidentiary hearing on the first prong of the Larrison test. Supra at 774 n. 3, 776.
The majority misconstrues Dobbins and Ferguson II. Specifically, the majority ignores the broader totality of the circumstances approach adopted in Dobbins and Ferguson II, and substitutes a wooden rule that all minor inconsistent statements merit an evidentiary hearing regardless of context. In Dobbins, we acknowledged that the conflict between the witness’s trial testimony and subsequent statement made “one or the other ... false.” 788 N.W.2d at 735. But we did not end our analysis there. Instead, we considered “the context of the two inconsistent statements,” which included the reduced sentence the witness received in exchange for his trial testimony and the fact that the witness made the subsequent statement to a friend. Id. We observed that the reduced sentence provided the witness an incentive to blame the murder on Dobbins and there was “likely a high probability that [the witness] would tell a friend the truth.” Id. We held that “under these circumstances,” a court could be reasonably well satisfied that it was the trial testimony, not the subsequent statement, that was false. Id. Similarly, we considered the context of the inconsistent statements in Ferguson II. *784See 779 N.W.2d at 560. Those circumstances included a claim that the police secured the witness’s trial testimony by threatening to take away his children and imprison the mother of his children. Id. at 558. Like the circumstances in Dobbins, these alleged threats could cause a court to be reasonably well satisfied that it was the trial testimony, not the subsequent statement, that was false.
Here, Brooks’s subsequent statement was not made to a friend, and nobody offered him a deal. Rather, Brooks was motivated by his belief that Caldwell was sentenced to “life for something he didn’t really do.... He was like involved a little but at the same time you can’t just give a man life. He didn’t kill no one. He got more time than the person who killed him.” Essentially, Brooks questions the policy judgment of the Legislature regarding aiding and abetting liability. The post-conviction court correctly concluded “feeling sorry that your friend is in prison is not sufficient assurance that you are now telling the truth, especially since Brooks still admits that [Caldwell] was involved in the murder.” Thus, the circumstances surrounding the inconsistent statements in Dobbins and Ferguson are not present in Caldwell’s case.18 By remanding this case for an evidentiary hearing, the majority effectively holds that any minor contradiction in a witness’s trial testimony and subsequent statement is sufficient, by itself, to warrant an evidentiary hearing. Such a holding, however, overrules our conclusion in Opsahl v. State that “a simple statement contradicting earlier testimony is not sufficient” to satisfy the first prong of the Larrison test. 710 N.W.2d 776, 782 (Minn.2006); see also Ferguson II, 779 N.W.2d at 559 (reaffirming that “a simple statement contradicting earlier testimony is insufficient to establish the petitioner’s right to a new trial”).
IV.
Finally, the majority misapplies the standard of review when it affords no deference to the postconviction court’s determination that the facts alleged in the taped interviews, if believed, would not have changed the result obtained at trial. A postconviction court’s decision not to hold an evidentiary hearing is reviewed for abuse of discretion, not de novo. Riley v. State, 819 N.W.2d 162, 167 (Minn.2012). In Reed v. State, we identified three ways in which a postconviction court might abuse its discretion. 793 N.W.2d 725, 729 (Minn.2010). Under the abuse of discretion standard, a matter will not be reversed unless the postconvietion court (1) based its ruling on an erroneous view of the law, (2) made clearly erroneous findings, or (8) exercised its discretion in an arbitrary or capricious manner. Id. at 729; see Montanaro v. State, 802 N.W.2d 726, 731 n. 6 (Minn.2011). Put differently, to determine whether the postconviction court abused its discretion, an appellate court must examine whether the postcon-viction court erred in identifying the controlling law for the case, determining the applicable facts, and applying the proper law to the relevant facts. I will discuss each of the Reed categories in turn.
*785A.
The question of whether the postconviction court correctly identified the controlling law for a case is a question of law that we review de novo. Riley, 819 N.W.2d at 167. The posteonviction court correctly identified the Larrison test and Our decisions in Ferguson, 645 N.W,2d at 445, and Pippitt, 737 N.W.2d at 228-29, as relevant to deciding whether to grant an evidentia-ry hearing. The majority concedes, and I agree, that the postconviction court correctly identified the Larrison test and Ferguson as controlling law in this case. In my view, the postconvictiori court also correctly relied upon Pippitt to conclude that the “inability to recall” is legally insufficient to satisfy the first prong of the Lar-rison test.
The issue in Pippitt was whether a witness’s “postconviction testimony constitute[d] a recantation.” 737 N.W.2d at 228 (emphasis added). We concluded that it did not. Id. The witness “did not state at the postconviction hearing that his trial testimony was false.” Id. Instead, the witness testified that he could not remember whether the door in question had a dead bolt lock. Id. We held that such testimony did not constitute a recantation for purposes of the Larrison test. Id. More specifically, we said, “[w]hen we compare the level of detail [the witness] provided about the front door during his trial testimony with his inability to remember, at the post-conviction hearing, whether the door even had a dead bolt, we cannot be reasonably ‘well-satisfied’ that his trial testimony was false, as required by the Larrison test.” Id. at 228-29.
We acknowledged the witness later reported his memory of the lock had been refreshed by a photograph. Id. at 228. Nevertheless, we held that the conflict between his refreshed memory and his trial testimony did not constitute a recantation. We concluded that the conflicting testimony in Pippitt did not “indicate that [the witness’s] trial testimony was false for purposes of the Larrison test.” Id. In doing so we relied upon the Opsahl rule that “a simple statement contradicting earlier testimony” does not satisfy the first prong of Larrison. 710 N.W.2d at 782.
In sum, we held in Pippitt that the witness’s postconviction testimony was legally insufficient to satisfy the Larrison test because the witness never renounced his trial testimony. We reasoned that a recantation requires more than a mere inability to remember or a simple statement contradicting earlier trial testimony.19 Consequently, Pippitt supports the *786postconviction court’s legal conclusion that an “inability to recall” is insufficient to satisfy the first prong of the Larrison test. Because the postconviction court’s decision was not based on an erroneous view of the controlling law, the first Reed category does not warrant a reversal in this case.
B.
The second Reed category requires us to consider whether the postconviction court made any clearly erroneous findings of fact. Generally, to determine whether an evidentiary hearing is necessary under section 590.04, a court must consider whether the facts alleged in the petition, if proved by a preponderance of the evidence, would entitle the petitioner to relief. See Minn.Stat. § 590.04, subd. 1 (2012); Ferguson, 645 N.W.2d at 445. The evidence presented to the postconviction court consisted of taped interviews of Carnell and Brooks. Consistent with section 590.04, the postconviction court did not make any findings of fact to which we must defer. Instead, the court assumed that Brooks and Carnell were telling the truth when they told the defense investigator that they could not clearly remember the circumstances surrounding the shooting. When asked if he testified truthfully at trial, Carnell answered: “Ah I can’t remember. I can’t like it’s been so long ago.” Brooks answered the question this way: “I wasn’t like by there to hear them arguing about anything. Cause I was like kinda drunk cause they had a little liquor.” Because the court did not make any factual findings, much less any clearly erroneous findings, the second Reed category does not warrant a reversal.20
C.
Third, I consider whether the postcon-viction court’s decision reflects an arbitrary or capricious application of the law to the facts presented in the petition to the postconviction court. The decision to grant an evidentiary hearing rests upon the determination of whether the petitioner has presented sufficient evidence that, if believed, would probably change the result obtained at trial. We review that determination for an abuse of discretion.21
*787The postconviction court applied the legal principle expressed in Pippitt — that an “inability to remember” is legally insufficient to satisfy the first prong of the Lar-rison test — to the facts of Caldwell’s case. Notably, neither Carnell nor Brooks renounced their trial testimony, or stated that they testified falsely at trial. In fact, both witnesses are very sketchy over whether they intended to recant or change any portion of their testimony. Based on the record in this case, I conclude that the postconviction court’s decision to deny Caldwell’s request for an evidentiary hearing does not reflect an arbitrary or capricious application of Pippitt to the facts alleged in the petition, and therefore the third Reed category does not warrant a reversal in this case.22
V.
Moreover, even if the postconviction court’s reliance on Pippitt was misplaced, a reversal is not warranted because the minor inconsistencies between the witnesses’ trial testimony and taped interviews do not satisfy the second prong of the Larrison test.
Under the second prong, a petitioner must allege facts that, if proved, would demonstrate that the jury might have reached a different conclusion absent the allegedly false testimony. Martin v. State, 825 N.W.2d 734, 740 (Minn.2013). By “might” we mean “something more than an outside chance although much less than ... would probably.” Caldwell, 322 N.W.2d at 585 n. 8 (internal quotation *788marks omitted). In applying “the second prong, we consider only the impact that [the witness’s] allegedly false testimony had on the jury, rather than guess what impact the substance of [the] recantation would have on the jury.” Dobbins, 788 N.W.2d at 733 n. 4. The existence of ample evidence independent of the false testimony will support a conclusion that the defendant “has failed to establish that, without [the false testimony], the jury might have reached a different conclusion.” Hooper v. State, 680 N.W.2d 89, 96 (Minn.2004). In Hooper, the false testimony “went directly to [the defendant’s] guilt.” Id. Nevertheless, we concluded the defendant failed to establish that, without the false testimony, the jury might have reached a different conclusion because two other witnesses testified that the defendant had confessed to them and the State’s other evidence placed the defendant at the scene of the murder. Id.
Having carefully reviewed the record in this case, I conclude the district court did not abuse its discretion when it concluded that Caldwell failed to establish that in the absence of the allegedly false testimony, the jury might have reached a different verdict. The impact of the allegedly false testimony of Brooks and Carnell was blunted by defense counsel’s cross-examination of those witnesses. During the questioning, Carnell conceded that he was drunk at the time of the shooting and that he later told a police investigator that “[he] was like really drunk, like kind of blurry.” Similarly, Brooks conceded that he told the grand jury, “I guess [Caldwell] passed Kirk the gun.”
Additionally, the jurors heard the testimony of Cey Barber, Troy Young and E.F. Defense counsel repeatedly told the jurors that they should believe Barber, Young and E.F. The cumulative testimony of Barber, Young and E.F. established that Caldwell and the shooter were members of a street gang that “ha[d] a problem” with the “One-Nine” gang. On the date in question, Caldwell and his group were at the corner of Eighth Street and Penn Avenue and observed a large group of One-Nines approaching them on foot. Caldwell’s group saw they were outnumbered, fled west along Eighth Street and climbed into Caldwell’s SUV, and Caldwell then drove east along Eighth Street. Moments later the shooter in Caldwell’s SUV shouted, “[T]here go the One-Nines.” In response, Caldwell drove toward the One-Nines “going slow,” and as he was “going past” the shooter fired several shots. Caldwell then sped away without saying anything or otherwise expressing surprise that the shooter had fired on the One-Nines. The jurors also heard the testimony of B.M., who said that Caldwell was driving the vehicle from which the shots came. And the jury heard from P.P. and C.P., who told the jurors that they heard Caldwell make separate comments following the murder indicating that he was responsible for the victim’s death.
As in Hooper, 680 N.W.2d at 96, the existence of ample evidence independent of the false testimony supports the district court’s conclusion that Caldwell failed to establish that without the allegedly false testimony of Brooks and Carnell, the jury might have reached a different conclusion. Caldwell was convicted of aiding and abetting, which requires that Caldwell knew his alleged accomplice was going to commit a crime and that Caldwell intended his presence or actions to further the commission of that crime. State v. Bahtuoh, 840 N.W.2d 804, 810 (Minn.2013). The evidence of Caldwell’s guilt of aiding and abetting was abundant even without the gun evidence at issue in Caldwell’s petition. Specifically, Caldwell had a motive for the murder. He drove the shooter to the scene after the shooter identified the location of the intended victims. And *789Caldwell immediately sped from the scene of the crime, not pahsing to render aid or even investigate whether any assistance was needed. Consistent with Hooper, I would affirm the district court’s analysis of the second Larrison prpng.
The majority reaches a different result by ignoring defense counsel's substantial attack on the testimony that Caldwell passed the gun to the shpoter. The majority also errs when it summarily dismisses Caldwell’s inculpatory statements to P.P. and C.P., where Caldwell admitted his participation in the crime, the undisputed testimony of Barber, Young and E.F. regarding Caldwell’s on-going dispute with the One-Nines, which established Caldwell’s motive, and Caldwell’s driving conduct, which positioned the shooter.23 The majority’s analysis is faulty and inconsistent with our precedent. Moreover, by affording no deference to the posteonviction court’s assessment of the impact of the allegedly false testimony, the majority effectively holds that a posteonviction court’s opportunity to observe the witnesses’ testimony and the jurors’ reactions to that testimony has no value when assessing the impact of false testimony on the jurors. In my mind, a posteonviction court is uniquely qualified to assess the impact of the false testimony. Because the record in this case supports the posteonviction court’s assessment, I dissent.
VI.
In sum, the majority dismisses the witnesses’ responses to the crucial question of whether they testified truthfully at trial. It also misconstrues the relevant case law by suggesting that a simple contradiction in a witness’s trial testimony and subsequent statement is sufficient, by itself, to satisfy the first prong of the Larrison test. Finally, the majority applies the wrong standard of review when it affords no deference to the posteonviction court’s determination that the facts alleged in the taped interviews, if believed, would not have changed the result obtained at trial. This lack of deference is especially troubling when the posteonviction court had the unique opportunity to observe the witnesses’ trial testimony, and when the minor inconsistencies cited by the majority do not satisfy the second prong of the Larrison test, which requires a defendant to demonstrate that the jury might have reached a different verdict. For these reasons, I dissent from the majority opinion. Moreover, because the posteonviction court’s denial of Caldwell’s request for an evidentiary hearing was not based on an erroneous view of the law, a clearly erroneous finding, or an arbitrary or capricious exercise of discretion, I would affirm.
. This fact was established by Cey Barber, Trial Tr. at 689, and C.P., Trial Tr. at 714-15.
. This fact was established by Troy Young, Trial Tr. at 563.
. This fact was established by Troy Young, Trial Tr. at 562, and Cey Barber, Trial Tr. at 692.
. This fact was established by Cey Barber, Trial Tr. at 694.
. This fact was established by E.F., Trial Tr. at 412.
. This fact was established by Troy Young, Trial Tr. at 571, Cey Barber, Trial Tr. at 696, and B.M., Trial Tr. at 521.
. This fact was established by Cey Barber, Trial Tr. at 697.
. This fact was established by P.P., Trial Tr. at 542-43.
. This fact was established by Troy Young, Trial Tr. at 591.
. This fact was established by Detective Nancy Dunlap, Trial Tr. at 444, and C.P., Trial Tr. at 725.
. Trial Tr. at 725-26.
. Trial Tr. at 726.
. Trial Tr. at 548.
. Trial Tr. at 793-94, 800-02, 809-11.
. Trial Tr. at 797.
. Caldwell also submitted an affidavit from the defense investigator in which he swore the transcripts accurately reflected the interviews.
. The postconviction court concluded that an affidavit from a third witness, S.T., did not warrant an evidentiary hearing because S.T.'s trial testimony was not material or pivotal. See Hooper v. State, 680 N.W.2d 89, 96 (Minn. 2004) (affirming the postconviction court's decision not to hold an evidentiary hearing because the recanting witness's testimony was not pivotal). That conclusion is sound.
. Moreover, the majority relies on several statements that are not actually contradictory. For example, Brooks testified at trial that Caldwell passed the murder weapon to the shooter, while at the time of his taped interview he said, "I really don’t know if the gun was just sitting on the side or if [Caldwell] passed it to [the shooter]. I really don't know.” Importantly, Brooks never said that Caldwell did not pass the gun to the shooter. Admittedly, Brooks stated that he has ”change[d] [his] mind” about some of his testimony at trial. Supra at 775 n. 4. But that statement does not suggest that Brooks provided testimony that he knew was false at the time of trial. Rather, it suggests that sometime after the trial ended he has begun to question some of his testimony.
. The majority's efforts to distinguish Pippitt on procedural grounds dre unavailing. It is true that in Pippitt our discussion of whether an inability to remember was legally insufficient to satisfy the first prong of the Larrison test was based on facts found after a postcon-viction evidentiary hearing, rather than facts alleged in a postconviction petition. 737 N.W.2d at 228-29. Here, that is a distinction without a difference. The postconviction court assumed that Brooks and Camell were telling the truth when they told the defense investigator that they could not remember. Had the postconviction court held an eviden-tiary hearing and found that Brooks and Car-nell were telling the truth when they told the defense investigator that they could not remember, the postconviction court's legal analysis would have remained unchanged. Put differently, the legal principle expressed in our discussion in Pippitt is the same whether the inability to remember is assumed to be true before an evidentiary hearing or found to be true after an evidentiary hearing. Moreover, the majority’s contention that the post-conviction court’s "credibility determination” was central to our decision in Pippitt is incorrect. Nowhere in our discussion of Pippitt’s Larrison claim did we use the words "credible” or “credibility.” Moreover, we did not mention, much less apply, the clearly erroneous standard, which controls our review of credibility determinations. Instead, we focused on the legal soundness of Pippitt’s argu*786ment that the witness's postconviction testimony "constitute[d] a recantation.”
. The majority incorrectly alleges that the postconviction court made factual findings. More specifically, the majority alleges that the postconviction court "conflated the requirements for a postconviction evidentiary hearing with the requirements for a new trial” when the postconviction court stated that it was not "reasonably well satisfied” that Carnell and Brooks's testimony was false. The majority contends that the postconviction court "failed to apply [the recited] standard correctly when it assessed the credibility of Caldwell’s witnesses without the benefit of an evidentiary hearing.” Supra at 773. Not only does the record fail to support the majority’s contention, it directly contradicts that contention. The postconviction court expressly acknowledged that (1) the showing required to obtain a postconviction evidentiary hearing on a false-evidence claim is lower than the showing required to obtain a new trial, and that (2) the decision to grant or deny a postconviction evidentiary hearing must be made without resolving factual disputes regarding credibility. For example, when discussing Brooks’s statement that he felt sorry for Caldwell, the postconviction court said “without making any credibility determinations, feeling sorry that your friend is in prison is not sufficient assurance that you are now telling the truth, especially since Brooks still admits that [Caldwell] was involved in the murder.” (Emphasis added.)
. The majority muddles the abuse of discretion standard of review of postconviction petitions. Specifically, the majority contends that the de novo standard of review applies to whether a petitioner is entitled to an evi-dentiary hearing on a claim of recantation, or false testimony of a witness. Indeed, the majority contends that the court must independently review the record to determine whether the alleged recantation contradicts the witness's trial testimony. This is not an abuse of discretion standard of review; instead it is de novo review. Put differently, if *787the only relevant question in deciding whether to hold an evidentiary hearing is reviewed de novo, what discretion does a postconviction court exercise when it decides not to hold an evidentiary hearing? The majority’s answer is none; instead, the majority proposes de novo review with no deference to the postconviction court. But we have repeatedly stated that we review a postconviction court's decision not to hold an evidentiary hearing for abuse of discretion. See, e.g., State v. Hooper, 838 N.W.2d 775, 786 (Minn. 2013); State v. Nicks, 831 N.W.2d 493, 503 (Minn.2013); Fort v. State, 829 N.W.2d 78, 82 (Minn.2013). The majority effectively overrules Hooper, Nicks, and Fort when it states that the legal sufficiency of Caldwell’s allegations is a question of law that does not require the exercise of any discretion.
. Affording no deference to the postconviction court’s determination, the majority finds that "Brooks substantively changed his testimony between the trial and his recorded statement.” Supra at 774 n. 3. In my view, the differences between Brooks’s trial testimony and the taped interview are minor and not sufficient to warrant an evidentiary hearing. As part of his trial testimony, Brooks conceded that during the grand jury proceedings he testified, "I guess [Caldwell] passed Kirk the gun.” In his recorded statement, Brooks said, "I don’t know if [Caldwell] even passed the gun to Kirk.... [T]hat’s why ... I said in my testimony, T guess he passed him the gun on the side.' ” Similarly, Brooks testified at trial that Caldwell passed the gun to the shooter "between the [car] door and the chair.” He did not testify that his view was not obstructed by the driver’s chair as the gun passed between the left side of the driver’s chair and the car door. In fact, when asked if he saw the transfer with his own eyes, Brooks provided a nonresponsive answer. In his taped interview, Brooks told the investigator that he couldn’t really see anything ”[c]ause if you going towards the door and the chair way you can’t even see him pass him nothing.” But that statement simply acknowledges the obvious fact that a driver’s chair blocks the rear middle passenger’s view as an object passes between left side of the driver’s chair and the car door. The fact that Brooks did not "see” the gun pass between the hands of Caldwell and the shooter does not make Brooks’s testimony that Caldwell passed the gun to the shooter "between the [car] door and the chair” false. This is especially true when one can reasonably infer that the driver passed his gun to the person directly behind the driver if the driver reaches between the driver’s chair and the car door and minutes later the person sitting behind the driver starts firing the driver's gun out the window. Thus, the difference between Brooks's trial testimony and his recorded statement is not as substantial as the majority suggests.
. In effect the majority holds that the verdict might have been different had the jurors known that Brooks was guessing when he said Caldwell passed the gun to the shooter and that Carnell was too drunk to remember the events surrounding the shooting. But as the record demonstrates, the jurors were made aware of both these facts and still found Caldwell guilty as charged.