Justice PARRISH,
opinion of the Court:
INTRODUCTION
T1 Petitioner David Vincent Gregg appeals the dismissal of his petition for post-conviction relief, The district court dismissed Mr. Gregg's claims as procedurally barred under the Post-Conviction Remedies Act (PCRA). We hold that Mr. Gregg qualifies for an exception to the procedural bar because he received ineffective assistance of trial and appellate counsel. Therefore, we vacate his conviction and remand for new trial.
BACKGROUND
12 On July 18, 2003, David Vincent Gregg was convicted of rape, a first-degree felony. The trial court sentenced him to an indeterminate sentence of five years to life. Mr. Gregg has served nearly nine years in the Utah State Prison, where he is currently incarcerated. This matter is before us on appeal from the district court's summary judgment order dismissing Mr. Gregg's amended petition for post-conviction relief. Because this case comes before us after a jury verdict, "we recite the facts from the record in the light most favorable to the jury's verdict and present conflicting evidence only as necessary to understand issues [399]*399raised on appeal." State v. Bluff, 2002 UT 66, 12, 52 P.3d 1210 (internal quotation marks omitted).
T3 Mr. Gregg was tried by jury in July 2008. At trial, both parties agreed that the only contested issue was whether the alleged victim, Ms. S., consented to sexual intercourse with Mr. Gregg. At sentencing, the trial court also acknowledged that "[t]he only issue before the court and before the jury was the question of consent," and noted that "Itlhis was a highly difficult issue."
[ 4 Seven months earlier, in January 2003, Mr. Gregg and Ms. S8. met online and had begun corresponding through an online dating service, LDSSinglesOnline.com (LDSSO). A few weeks later on February 15, 20083, Mr. Gregg and Ms. S. began chatting on LDSSO shortly after midnight. After chatting online with Mr. Gregg, Ms. S. gave him her phone number and the two talked on the phone for the first time. Around 2:00 a.m., Ms. S. invited Mr. Gregg to her apartment. Just before Mr. Gregg arrived, Ms. S. called her friend Matt and asked him to call her in one hour. Ms. S. testified that she asked Matt to call her because she was concerned that Mr. Gregg, whom she had never met before, was coming over late at night. Mr. Gregg arrived at her home at 2:20 a.m.
5 After Mr. Gregg arrived, he and Ms. S. talked and then engaged in consensual kissing and sexual foreplay. They were interrupted when Matt called at 3:20 a.m. Ms. S. answered Matt's call and pretended the call was from a girlfriend. After ending the call, Ms. S. told Mr. Gregg that a girlfriend was coming over because she was having problems with her boyfriend. In reality, Matt had indicated that he would drive up to Ms. S.'s apartment in Bountiful onee his girlfriend Jess got off of work in Draper. At this point, Mr. Gregg's and Ms. S.'s description of events diverge.
16 According to Ms. S.'s testimony, once she ended the call with Matt, she and Mr. Gregg resumed kissing. She testified that she stopped and told Mr. Gregg that she was uncomfortable because they had just met. Mr. Gregg asked her if they could continue with sexual activity for another five minutes and promised they would not have intercourse. Ms. S. testified that she told Mr. Gregg "no" several times before he suddenly grabbed her ankles and began to rape her. She testified that shortly thereafter, Matt called again, interrupting intercourse.
17 Mr. Gregg's version of events is quite different.1 According to Mr. Gregg, after Ms. S. received the phone call at 8:20 a.m., they resumed kissing, which progressed to consensual foreplay. They paused for a brief conversation about how far they wanted sexual activity to progress. Ms. S. agreed to continue, and Mr. Gregg agreed that they would stop whenever she wanted. They resumed consensual sexual activity, which gradually escalated to the point of intercourse. According to Mr. Gregg, Ms. S. did not say "no" or stop foreplay after their conversation about boundaries. Rather, Ms. S. asked Mr. Gregg to turn off the lights and helped him pull off her pants before the two began consensual intercourse.
8 Both Mr. Gregg and Ms. S. agree about the events after this point. Matt's phone call interrupted intercourse. Ms. S. answered the phone and asked Matt and his girlfriend to wait outside a few minutes before coming to the front door. Police records indicate that this phone call occurred .at 4:07 a.m. With Matt and Jess waiting outside, Mr. Gregg and Ms. S. got dressed. Then Ms. S. answered the door, invited Matt and his girlfriend into her apartment, and introduced the couple to Mr. Gregg. Matt and Jess stayed in the apartment while Ms. S. walked Mr. Gregg to his car, hugged him, gave him a kiss, and said goodnight.
T9 At trial, Ms. S. testified that onee she got back to her apartment, she began to ery. She told Matt and Jess what had happened, and they urged her to go to the hospital or call the police. She resisted, saying that she was fine and that she was responsible for not stopping things with Mr. Gregg sooner. Matt called a rape crisis line, suggesting that Ms. S. had been assaulted. The crisis line told Matt that if Ms. S. went to the hospital, [400]*400they would file a report, but she would not have to press charges. Ms. S. was still resistant until she called another friend of hers, Sean, whom she had dated and met on LDSSO. Sean was able to convince her to go to the hospital to get the morning-after pill to prevent a potential pregnancy.
110 While at the hospital, the examining nurse brought Ms. S. to a room to wait for a doctor. The nurse then returned with a rape victim advocate and a detective. Ms. S. was hesitant to talk to them because she did not feel that it was necessary. In her own words, Ms. S. testified that she "didn't know-[there were] so many feelings going through my mind and I wanted to make sure what I was feeling was valid ... But I'm not sure, you know, if what I'm feeling is just my imagination." At this time, Ms. S. had never claimed she was raped in her conversations with Matt, Jess, Sean, or the examining nurse.
T11 The rape victim advocate and the detective encouraged Ms. S. to tell them about her encounter with Mr. Gregg, suggesting that she could use a fake name so that no one would be implicated. Ms. S. finally acquiesced and described the events of the previous evening. After Ms. S. told them about her encounter with Mr. Gregg, she testified that the rape victim advocate and the detective "told [her] that they ha[ld] never heard of someone that was more manipulative, and that what happened was a date rape. And [Mr. Gregg was] a person who plays mind games." At that point, Ms. S. decided to press charges against Mr. Gregg.
1 12 Six months later, after a two-day trial, the jury convicted Mr. Gregg of one count of rape, a first-degree felony. After trial and before sentencing, Mr. Gregg fired trial counsel based on counsel's "serious errors" at trial Mr. Gregg filed a pro se motion to arrest the judgment, which the trial court denied. At that same hearing, the trial court sentenced Mr. Gregg to an indeterminate term of five years to life.
183 Mr. Gregg retained appellate counsel, who appealed the district court's denial of the motion to arrest the judgment and filed a rule 23B remand motion with the Utah Court of Appeals.2 The court of appeals denied the motion and affirmed the conviction. We denied Mr. Gregg's petition for certiorari.
[ 14 Mr. Gregg then obtained new counsel and filed a petition for post-conviction relief. The district court dismissed this petition, but the Davis County Attorney's office stipulated that Mr. Gregg could amend his original petition. Mr. Gregg did so. In his amended petition, Mr. Gregg made eleven arguments alleging ineffective assistance of trial and appellate counsel.3 The Davis County Attorney's office filed a motion for summary judgment and the district court granted the motion, dismissing all of Mr. Gregg's claims as procedurally barred under the PCRA. Mr. Gregg then filed the instant appeal. We have jurisdiction under section 78A-3-102(3)(i) of the Utah Code.
STANDARD OF REVIEW
¶ 15 "We review an appeal from an order dismissing or denying a petition for [401]*401post-conviction relief for correctness without deference to the lower court's conclusions of law." Taylor v. State, 2012 UT 5, ¶ 8, 270 P.3d 471 (internal quotation marks omitted).
ANALYSIS
I. MR. GREGGS AMENDED PETITION FOR POST-CONVICTION RELIEF IS A FIRST PETITION RATHER THAN A SUCCESSIVE PETITION
¶ 16 The State argues for the first time on appeal that Mr. Gregg's amended petition for post-conviction relief should be treated as a successive petition and urges us to dismiss his petition on that basis alone. The State argues that it was procedurally improper for the Davis County Attorney to stipulate to Mr. Gregg's amended petition because the State Attorney General's office is the only party who can permit an extension under rule 65C(i) of the Utah Rules of Civil Procedure. Thus, the State argues that Mr. Gregp's petition should be dismissed as an improper successive petition. We disagree.
¶ 17 Rule 65C(i) governs service of petitions for post-conviction relief and requires that the court direct the clerk of the court to "serve a copy of the petition, attachments and memorandum by mail upon the respondent." Utah R. Civ. P. 65C(i). The rule notes that where "the petition is a challenge to a felony conviction or sentence, the respondent is the state of Utah represented by the Attorney General." Id. Thus, rule 65C(i) directs the court clerk to serve PCRA petitions on the proper respondent. In this case, the clerk mailed Mr. Gregg's petition for post-conviction relief to the Davis County Attorney rather than the Utah State Attorney General. And rather than forwarding the petition to the State Attorney General, the Davis County Attorney filed a motion for summary judgment and then stipulated that Mr. Gregg could amend his original petition. We will not hold Mr. Gregg accountable for a procedural error on behalf of the clerk of the court or for the county attorney's failure to forward the petition. Because of these errors, it was reasonable for Mr. Gregg to believe that the Davis County Attorney had authority to stipulate to the filing of his amended petition. Therefore, we will treat Mr. Gregg's petition as a first amended petition and address the merits of his claims.
II. MR. GREGG RECEIVED INEFFECTIVE ASSISTANCE OF COUNSEL
{[ 18 The PCRA "establishes a substantive legal remedy for any person who challenges a conviction or sentence for a eriminal offense and who has exhausted all other legal remedies, including a direct appeal." Utah Cope § 78-35a-102(1) (2007).4 The Act includes a legal remedy for a petitioner who received ineffective assistance of counsel. Id. §§ 78-35a-104(1)(d), -106(2). To succeed on an ineffective assistance of counsel claim in a post-conviction petition for relief, the petitioner must prove that he received ineffective assistance from both his trial counsel and his appellate counsel. See id. § 78-352a-106(2); Lafferty v. State, 2007 UT 73, ¶ 44, 175 P.3d 530.
¶ 19 The Sixth Amendment to the U.S. Constitution guarantees eriminal defendants a right to effective assistance of counsel. U.S. Const. amend. VI; Strickland v. Washington, 466 U.S. 668, 684-87, 104 S.Ct. 2052, 80 L.Ed.2d 674 (1984).5 We evaluate whether a defendant has received ineffective assistance of counsel under the two-part test established in Strickland v. Washington, 466 U.S. 668, 104 S.Ct. 2052, 80 L.Ed.2d 674 (1984). "First, the defendant must show that counsel's performance was deficient.... See-[402]*402ond, the defendant must show that the deficient performance prejudiced the defense." Id. at 687, 104 S.Ct. 2052. We hold that Mr. Gregg received ineffective assistance of both trial counsel and appellate counsel.
A. Mr. Gregg Received Ineffective Assistance of Trial Counsel
¶ 20 To meet the first part of Strickland, the defendant must demonstrate that "counsel's performance was so deficient as to fall below an objective standard of reasonableness." Myers v. State, 2004 UT 31, ¶ 20, 94 P.3d 211 (citing Strickland, 466 U.S. at 687-88, 104 S.Ct. 2052) (internal quotation marks omitted). Generally, there is "a strong presumption that counsel's conduct falls within the wide range of reasonable professional assistance." Strickland, 466 U.S. at 689, 104 S.Ct. 2052.
¶ 21 Under the second part of Strickland, "the defendant must show that [counsel's] deficient performance prejudiced the defense." Id. at 687, 104 S.Ct. 2052. Prejudice is shown where there is "a reasonable probability that, but for counsel's unprofessional errors, the result of the proceeding would have been different. A reasonable probability is a probability sufficient to undermine confidence in the outcome." Id. at 694, 104 S.Ct. 2052. We have noted that "iIn making this determination, an appellate court should consider the totality of the evidence, taking into account such factors as whether the errors affect the entire eviden-tiary picture or have an isolated effect and how strongly the verdict is supported by the record." State v. Templin, 805 P.2d 182, 187 (Utah 1990) (citing Strickland, 466 U.S. at 696, 104 S.Ct. 2052).
{22 Mr. Gregg has met his burden under Strickland. We hold that it was ineffective assistance of trial counsel to fail to investigate and present evidence of (1) Ms. S.'s LDSSO e-mail correspondence after the alleged rape and (2) the 47-minute period of consensual sexual activity immediately before the alleged attack.6
1. Ms. S.'s LDSSO E-mails
¶ 23 Mr. Gregg argues that it was ineffective assistance of trial counsel to fail to investigate facts that would have undermined the credibility of the victim, who was the only witness that provided direct evidence of his guilt. Specifically, Mr. Gregg argues that his trial counsel failed to investigate Ms. S.'s LDSSO e-mails, which would have revealed that Ms. S. contacted other men and sent four cheerful, light-hearted e-mails through LDSSO only two days after the alleged rape. Mr. Gregg claims that this evidence would have affected the outcome at trial because it would have severely undermined Ms. S.'s credibility. In particular, the e-mails would have undermined Ms. S.'s testimony that she only logged on to LDSSO after the alleged rape to aid the police investigation, and the e-mails would have undermined her credibility regarding the severe anxiety and panic attacks she claimed to suffer as a result of the incident. We agree that it was ineffective assistance of counsel to fail to investigate and present this evidence to the jury.
¶ 24 Trial counsel's failure to conduct a reasonable investigation into the underlying facts of a case constitutes deficient performance under the first Strickland prong. In State v. Templin, we vacated a conviction of rape and remanded for new trial where the defendant's trial counsel failed to investigate facts that would have contradicted the vie-tim's testimony. Id. at 188-89. We noted that when "counsel does not adequately investigate the underlying facts of a case, ... counsel's performance cannot fall within the 'wide range of reasonable professional assistance' " because "a decision not to investigate cannot be considered a tactical decision." Id. at 188 (quoting Strickland, 466 U.S. at 689, 104 S.Ct. 2052).
¶ 25 Similarly, trial counsel's failure to investigate Ms. S.'s LDSSO e-mails constituted [403]*403deficient performance. Trial counsel was aware that Ms. S. logged on to the dating website at least seventeen times after the alleged rape, but counsel failed to investigate the contents of her online interaction. The specific content of Ms. S.'s LDSSO e-mails was critical to Mr. Gregg's defense because Mr. Gregg and Ms. S. met on LDSSO, the two chatted on that same dating service immediately before the alleged attack, and the e-mails revealed crucial details about Ms. S.'s state of mind after the alleged rape. Yet trial counsel did not attempt to investigate or subpoena Ms. S.'s LDSSO e-mail records in the six months leading to trial, even though he easily could have done so. In fact, Mr. Gregg subpoenaed the records himself after he fired trial counsel because of his counsel's "serious errors" and "failure to present key facts" at trial. Had trial counsel investigated Ms. S.'s correspondence on LDSSO, he would have discovered that Ms. S. actually sent light-hearted e-mails to four different men through LDSSO only two days after she claimed she was raped by Mr. Gregg. Trial counsel's decision not to investigate this evidence fell outside the range of reasonable professional assistance under the first prong of Strickland.
126 Under the second Strickland prong, Mr. Gregg must show prejudice. Strickland, 466 U.S. at 687, 104 S.Ct. 2052. We have held that when trial counsel fails to reasonably investigate and present evidence that was crucial to the defense, it amounts to prejudice when this evidence would have "affected] the 'entire evidentiary picture."" Templin, 805 P.2d at 188 (quoting Strickland, 466 U.S. at 696, 104 S.Ct. 2052). To determine the overall effect of such evidence, we have instructed appellate courts to "consider the totality of the evidence, taking into account such factors as whether the errors affect the entire evidentiary picture or have an isolated effect and how strongly the verdict is supported by the record." Id. at 187. For instance, in Templin we noted that the prospective witness's testimony was important because "it reflect[ed] upon the credibility of [the alleged victim}." Id. at 188. We held that the alleged victim's credibility was particularly important "because [her] testimony [was] the only direct evidence of Temp-lin's guilt," and the failure to investigate a prospective witness amounted to prejudice because that witness's testimony would have affected "the entire evidentiary picture." Id. (internal quotation marks omitted).
¶ 27 The second Strickland prong has been met here. The LDSSO e-mails would have severely undermined Ms. S.'s credibility, which would have affected the entire eviden-tiary picture because her testimony was the only direct evidence of Mr. Gregg's guilt. At trial, Ms. S. testified that she only logged on to LDSSO after her encounter with Mr. Gregg to aid the police in their investigation. But in reality, only two days after the alleged rape, Ms. S. sent four e-mails to various men through LDSSO, giving them her private email address and encouraging them to contact her. One e-mail read:
Hey Coz!
I just wanted to let you know that I am not going to renew my membership ..[.] it expires today. So, I think that you still have my number and email address and I would love to keep in touch. :) Hope all is going well! :) Talk to you soon!
[Ms. S.]7
This evidence would have undermined Ms. S.'s credibility because it directly conflicted with her testimony that she only logged on to LDSSO after the rape to help the police investigation. And the e-mails demonstrating her cheerful demeanor also would have conflicted with her story that she was so distraught after her encounter with Mr. Gregg that she suffered an anxiety attack and swore off the LDSSO dating service altogether.8
[404]*404128 Had Mr. Gregg's trial counsel presented these e-mails, the jury would have been able to fully assess Ms. S.'s eredibility. And Ms. S.'s e-mails would have undermined the entirety of her testimony because they would have demonstrated that she sent four light-hearted e-mails to various men through LDSSO only two days after she claimed to have been raped by a man whom she had met on that same website. Her cheerful demeanor also would have undermined her credibility regarding the severe anxiety she claimed to have suffered after her attack. And the e-mails would have directly refuted her claim that she swore off the dating service and only accessed her LDSSO account to further the police investigation.
129 The State contends that Ms. S.'s LDSSO activity was sufficiently addressed at trial because Mr. Gregg's trial counsel presented Ms. S.'s unauthenticated LDSSO lo-gon history. But Mr. Gregg's trial counsel never investigated or presented Ms. S.'s emails at trial. And the specific content of Ms. S.'s LDSSO e-mails was essential to determining her credibility, given that they provided insight into Ms. S.'s state of mind after the alleged rape. Furthermore, the emails would have directly rebutted Ms. S.'s claim that she accessed her LDSSO account only to further the police investigation.
11 80 As we noted in Templin, it is unreasonable for trial counsel to fail to investigate critical facts relevant to the victim's eredibility, especially in a case such as this one where the only direct evidence of the defendant's guilt was from the testimony of the alleged victim. Id. Ms. S's LDSSO e-mails would have affected the entire evidentiary picture at trial because Ms. S.'s testimony was the only direct evidence of her lack of consent, which was key to Mr. Gregg's guilt or innocence. There was no independent physical evidence that supported or contradicted Ms. S.'s testimony, and therefore, the conviction is not strongly supported by the record. And "although it is undisputed that a person can be convicted of rape solely on the testimony of the victim," we have nevertheless held that where the conviction is not strongly supported by the record and trial counsel fails to investigate and present evidence impacting the victim's credibility, Strickland is met. Id. at 188-89. This evidence affected the overall evidentiary picture and is sufficient to undermine our confidence in the trial's outcome. Therefore, we hold that Mr. Gregg was prejudiced by his trial counsel's deficient performance under Strickland.
2. The 47-Minute Window
¶ 31 Mr. Gregg argues that his trial counsel also failed to present evidence as to the timing of events that would have undermined Ms. S.'s credibility, called into question her account of the alleged attack, and would have corroborated Mr. Gregg's story that the two engaged in consensual intercourse after nearly an hour of sexual foreplay. We agree that trial counsel's failure to investigate and present this evidence to the jury constituted ineffective assistance of counsel.
¶ 32 To meet the first part of Strickland, Mr. Gregg must demonstrate that "counsel's performance was so deficient as to fall below an objective standard of reasonableness." Myers, 2004 UT 31, 120, 94 P.3d 211 (citing Strickland, 466 U.S. at 687-88, 104 S.Ct. 2052) (internal quotation marks omitted). In Templin, we held that it was deficient performance of trial counsel to fail to investigate facts that would have undermined the vie-tim's credibility and her story regarding the alleged rape, especially when that evidence tended to corroborate the defendant's story. 805 P.2d at 187-88. We determined that this was deficient performance because the attorney failed to investigate a prospective witness even though the defendant had provided the name of the witness to the attorney prior to trial. Id.
133 Similarly, in this case, Mr. Gregg's trial counsel was aware of the 47-minute time period prior to trial. It was clear from [405]*405the police report that there was a 47-minute window of time between the two phone calls during which the rape allegedly occurred. This directly conflicted with Ms. S.'s story that the rape occurred in a matter of minutes. Yet Mr. Gregg's trial counsel failed to investigate that time period even though it would have undermined Ms. S.'s overall ered-ibility and her testimony regarding the rape. And trial counsel's failure to investigate this evidence was particularly troubling given that the 47-minute window would have corroborated Mr. Gregg's account that he and Ms. S. engaged in consensual intercourse after nearly an hour of sexual foreplay. This deficiency is especially egregious given that Mr. Gregg's guilt or innocence rested on the issue of consent.
34 The record reveals that Mr. Gregg's trial counsel never attempted to introduce the police report into evidence, and the State never moved to exclude it. The record also indicates that trial counsel never attempted to further investigate the timing of the phone calls. Trial counsel never subpoenaed Matt's phone records. And at trial, Mr. Gregg's counsel did not even attempt to elicit testimony from Matt that would have clarified when he placed the calls. This is particularly troubling given that the little testimony that Matt did provide regarding the timing of his phone calls appeared implausible.9 As we noted in Templin, it can never be a tactical decision to fail to investigate and introduce evidence that would undermine the eredibility of the only witness who presented direct evidence of the defendant's guilt. Id. at 188. In a case where consent is the only contested issue at trial, this detail was essential in determining whether Mr. Gregg's or Ms. S.'s account more accurately portrayed what occurred, especially because this detail would have further undermined Ms. S.'s overall credibility.
1 35 The State argues that the essence of this evidence was presented to the jury and Mr. Gregg's trial counsel merely failed to emphasize the 47-minute time frame. And the State contends that trial counsel cannot be found ineffective for points of emphasis. But the police report was neither presented at trial nor submitted to the jury. And Mr. Gregg's trial counsel failed to produce any evidence of the 47-minute period. His trial counsel made only one vague reference to a 45-minute time period during closing argument, and that reference was completely unsupported by any evidence presented at trial. Thus, Mr. Gregg's trial counsel performed deficiently by failing to investigate and present this evidence to the jury, despite its obvious presence in the police report.
¶ 36 Mr. Gregg has also demonstrated prejudice under the second Strickland prong. In Templin, the defendant was con-viected of rape after a trial where the only disputed issue was consent. Id. at 183-84. On appeal, we held that it was ineffective assistance of trial counsel to fail to contact and interview a prospective witness who "would have testified that she saw [the] defendant and the [alleged] victim kissing passionately for over fifteen minutes ... within an hour of the rape reported by the victim." Id. at 188-89. We noted that this evidence was critical because the witness's testimony, "reflect[ed] upon the credibility of [the alleged victim]." Id. at 188. Additionally, we noted that it was important because "[the witness's] testimony, although not completely consistent with [the defendant's] testimony, contradicted] several aspects of [the alleged victim's] testimony." Id. at 188. Ultimately, we determined that the defendant had established prejudice because the undiscovered evidence would have "affect[ed] the entire evi-dentiary picture" given that the evidence would have undermined "the credibility of [the alleged victim,] the only witness who gave direct evidence of [the] defendant's guilt." Id. (internal quotation marks omitted).
[406]*406¶ 37 Here too, Mr. Gregg's counsel failed to investigate and present crucial evidence that would have affected the entire evidentia-ry picture. Evidence of the 47-minute window would have undermined Ms. S.'s overall credibility, which is particularly important because her testimony provided the only direct evidence of Mr. Gregg's guilt. And those facts also would have contradicted Ms. S.'s testimony that the alleged rape occurred in a matter of minutes between the two phone calls, and it would have corroborated Mr. Gregg's claim that consensual intercourse culminated after almost an hour of sexual foreplay.
¶ 38 Ms. S.'s testimony portrayed a sudden attack that occurred in a matter of minutes. She testified that the rape began shortly after Matt's 3:20 a.m. phone call. Specifically, Ms. S. testified that after she got off the phone with Matt she and Mr. Gregg briefly resumed kissing. She stopped Mr. Gregg, and said "no" several times. According to her testimony, Mr. Gregg then immediately grabbed her ankles and commenced noneon-sensual intercourse. She testified that intercourse was quickly interrupted by Matt's see-ond phone call.10 The prosecution's closing argument further suggested that this scene played out in a matter of minutes by reiterating that shortly after the 8:20 a.m. phone call, Ms. S. told Mr. Gregg "no" several times, whereupon Mr. Gregg "commenced immediately to place her down on the couch and start that process.... Here was a man who promised he wouldn't have sex and all of a sudden, it's going on." Thus, Ms. S.'s testimony and the prosecution's closing argument gave the jury the distinct impression that this was a sudden attack that occurred in a matter of minutes.
¶ 39 But Ms. S.'s testimony is directly con-tradiected by the police report. It was not merely a matter of minutes between Matt's phone calls. In fact, the police report indicates that there was actually a 47-minute window of time between the two calls. According to the police report, Matt first called Ms. S. at 3:20 am. and then again at 4:07 a.m. when he arrived at her apartment. Ms. S. testified that the alleged rape began shortly after the first phone call and was interrupted by the second phone call. But her testimony also suggested that the time between the two phone calls was merely a matter of minutes. Thus, evidence of the 47-minute window between the two phone calls would have undermined Ms. S.'s credibility as well as the plausibility of her account of the rape. This is particularly troubling given that her testimony was the only direct evi-denee of Mr. Gregg's guilt.
¶ 40 This 47-minute time period also would have corroborated Mr. Gregg's account of events. Mr. Gregg maintains that consensual intercourse culminated after nearly an hour of sexual foreplay. According to Mr. Gregg, after the 8:20 a.m. phone call, Mr. Gregg and Ms. S. resumed consensual sexual activity. They paused for a brief conversation about how far they were willing to proceed, and he agreed that they would stop whenever she wanted. According to Mr. Gregg, Ms. S. agreed to resume kissing and engage in consensual sexual activity. Things gradually escalated to intercourse, at which point Ms. S. asked him to turn off the lights and helped him take off her pants. According to Mr. Gregg, Ms. S. did not stop consensual foreplay, and she never said "no" after their initial conversation following the 3:20 a.m. phone call.
¶ 41 The police report indicates that Matt called Ms. S. at 3:20 a.m. and again at 4:07 am. This 47-minute time frame would have corroborated Mr. Gregg's story that Ms. S. consented to intercourse after a gradual progression of consensual foreplay over the course of about an hour. This evidence regarding the timing of the phone calls is critical to determining whose account more accurately portrays the events of that night, especially in a case where consent was the only contested issue at trial. This evidence also would have undermined Ms. S.'s testi[407]*407mony, which suggested that nonconsensual intercourse occurred suddenly after Matt's 3:20 a.m. phone call. The 47-minute time frame, like the testimony in Templin, while not perfectly corroborating either Ms. S.'s or Mr. Gregg's story, is crucial to this credibility determination because it suggests that Mr. Gregg's account more accurately portrays what actually occurred that night. See id.
142 The failure to investigate this evidence establishes prejudice under Strickland's second prong because it "affect[ed] the entire evidentiary picture." Id. (internal quotation marks omitted). It would have undermined Ms. S.'s overall credibility, as well as the plausibility of her account of the alleged rape. This evidence also would have tended to corroborate Mr. Gregg's account of the events. As we noted in Templin, "[aln appellate court cannot discern the exact ef-feet such [evidence] would have had on the jury's judgment concerning the credibility of [the defendant] and [the alleged victim]." Id. However, where the evidence is of "sufficient import that we feel there is a reasonable probability that [had this evidence been investigated and presented at trial], the outcome of the trial would have been different," we may nevertheless conclude that Strickland has been met. Id. at 188-89. Evidence of the 47-minute time period between the calls, coupled with evidence of Ms. S.'s LDSSO e-mails, convince us that if this evidence had been presented at trial, there is a reasonable probability that Mr. Gregg would have been acquitted. Accordingly, we hold that Mr. Gregg received ineffective assistance of trial counsel.
B. Mr. Gregg Received Ineffective Assistance of Appellate Counsel
¶ 43 Mr. Gregg also argues that he received ineffective assistance from his appellate counsel when appellate counsel failed to argue that trial counsel performed defi-ciently with respect to the LDSSO e-mails and the 47-minute period. The State argues that Mr. Gregg's claims "could have been raised" on direct appeal, and thus are procedurally barred under the PCRA, Utah Code section 78-35a-106(1)(c). In its motion for summary judgment in the district court, the State argued that Mr. Gregg's claims regarding the 47-minute time frame and Ms. S.'s LDSSO e-mails were procedurally barred "because they could have been raised on direct appeal." And in its brief to this court, the State Attorney General's office again argues that Mr. Gregg "'could have' raised [these] ineffective assistance claim[s] on direct appeal." Thus, the entire thrust of the State's argument is that these claims are procedurally barred under the PCRA, Utah Code section 78-35a-106(1)(c).11 But this argument ignores the PCRA's exception, which provides that "a person may be eligible for [408]*408relief on a basis that the ground could have been but was not raised at trial or on appeal, if the failure to raise that ground was due to ineffective assistance of counsel." Utah CopE § 78B-9-106(2) (2007). And this provision controls because Mr. Gregg received ineffective assistance of both appellate counsel and trial counsel.
¶ 44 As with ineffective assistance of trial counsel claims, a petitioner claiming ineffective assistance of appellate counsel must prove that his appellate counsel provided ineffective assistance under Strickland's two-part test. Lafferty, 2007 UT 73, 189, 175 P.3d 530. Specifically, "to prevail on a claim of ineffective assistance of appellate counsel, a petitioner must prove that appellate counsel's representation fell below an objective standard of reasonable conduct and that the deficient performance prejudiced [him]." Id. (alteration in original) (internal quotation marks omitted). Where a petitioner is claiming ineffective assistance of counsel "for omitting a claim, he must show that the issue [was] obvious from the trial record and ... probably would have resulted in reversal on appeal." Id. (alterations in original) (internal quotation marks omitted).
¶ 45 Mr. Gregg clearly articulated his concern that trial counsel failed to fully develop the underlying facts of his case. After trial, Mr. Gregg requested that the court allow his trial counsel to withdraw. Mr. Gregg's letter to the court reads:
Last week I contacted [my trial counsel's] office [and] completely of my own voluntary choice, after much thought, I asked [my attorney] to withdraw as my counsel. My reasons for doing so were two-fold, but the primary and most important reason is because I do not want him to represent me because I have no confidence in him. I feel this way because [my attorney] made serious errors at trial, with the biggest being not presenting facts that are critical [and] essential to my case, and which may have made a difference at trial I discussed these facts with him multiple times over the time between my arrest in February and my trial in July. I asked him to present many of these points at trial, but he did not do so. The omissions/errors were blatant to the degree that I am not willing to place my future in his hands (unless forced by the Court to do so), and ... I am seriously considering filling] a complaint with the Utah State Bar.
These concerns, which were plainly apparent in the record, should have made appellate counsel acutely aware of potential ineffective assistance of counsel claims.12 And while "Lalppellate counsel is not obligated to raise every nonfrivolous issue on appeal ... [and may] [instead] winnow out weaker claims in order to focus effectively on those more likely to prevail," id. ¶ 49 (second and third alterations in original) (internal quotation marks omitted), this does. not exeuse appellate counsel from ignoring obvious errors that would have influenced the trial's outcome, see id. 1 39.
¶ 46 In this case, the record clearly indicated that trial counsel failed to present key facts at trial that would have likely had an effect on the trial's outcome. For instance, the record plainly shows that Mr. Gregg subpoenaed Ms. S.'s e-mail records from LDSSO pro se, that trial counsel had failed to request these records, and that the e-mail messages severely undermined Ms. S.'s ered-ibility, thereby undermining confidence in the trial's outcome. The record also reveals that Mr. Gregg's trial counsel failed to introduce evidence that the alleged assault occurred during a 47-minute period, despite its obvious presence in the police report and defense counsel's unsupported reference to the time frame in his closing argument. And the record shows that counsel's failure to introduce this evidence prejudiced the outcome at trial because the evidence would have corroborated Mr. Gregg's story and would have conflicted with Ms. S.'s depiction of events.
¶ 47 Mr. Gregg's appellate counsel failed to allege these ineffective assistance of trial counsel claims even though they were obvi[409]*409ous from the record. Instead, on direct appeal, appellate counsel merely reargued the points of Mr. Gregg's pro se motion to arrest the judgment. And in Mr. Gregg's motion for remand, appellate counsel made only the general allegation that trial counsel failed to "research, investigate, and use evidence ... to impeach the testimony of the victim," without reference to Ms. S.'s LDSSO e-mails or the 47-minute window of time.13
€48 Appellate counsel's failure to allege ineffective assistance of trial counsel regarding the LDSSO e-mails and the 47-minute time period prejudiced Mr. Gregg's direct appeal under the second prong of Strickland. The court of appeals affirmed Mr. Gregg's conviction based on Ms. S.'s testimony. Specifically, the court of appeals noted "the question of guilt or innocence often depends on the weighing of the credibility of the victim against that of the accused" and ultimately affirmed Mr. Gregg's conviction because there was sufficient evidence to support the verdict, "namely, [Ms. S.'s] testimony." State v. Gregg, 2005 UT App 258U, para. 4, 2005 WL 1314900 (internal quotation marks omitted). Had appellate counsel presented these claims of ineffective assistance of trial counsel, this would have seriously undermined Ms. S.'s credibility and likely changed the outcome on appeal. By failing to present and clearly articulate these obvious claims on direct appeal, Mr. Gregg's appellate counsel provided deficient performance that prejudiced Mr. Gregg's appeal. Therefore, we hold that Mr. Gregg received ineffective assistance of appellate counsel under Strickland.14
CONCLUSION
4[ 49 We conclude that Mr. Gregg received ineffective assistance from both his trial and appellate counsel in violation of the Sixth Amendment. Therefore, we vacate his conviction and remand for new trial.
Justice PARRISH authored the opinion of the Court, in which Chief Justice DURRANT, Associate Chief Justice NEHRING, and Justice DURHAM joined.
Justice LEE filed a dissenting opinion.