PER CURIAM.
Pablo Ibar appeals an order of the circuit court denying his motion to vacate his convictions of first-degree murder and sentences of death filed under Florida Rule of Criminal Procedure 3.851 and petitions this Court for a writ of habeas .corpus. We 'have jurisdiction. See ■ .art. V, § 3(b)(1), (9), Fla. Const.
In this case, there was a lack of physical evidence connecting Ibar to the triple murders. Ibar’s DNA was not found on a blue t-shirt recovered from the crime- scene which was allegedly used to partially cover the face of the perpetrator whom the State claimed to have been Ibar.' Ibar never confessed to the crime as he steadfastly proclaimed his innocence, presented an alibi as to his whereabouts, and has always maintained his innocence.
The crux of the State’s case was a grainy video of the murders taken by a video surveillance camera installed in the home of Casmir Sucharski, who was one of the victims. Ibar was identified as one of the perpetrators from photographs distilled from this videotape. Consequently, identification was key.
Initially, Ibar was tried with codefen-dant Seth Penalver, but the first trial ended with a hung jury. Thereafter, Penal-ver, in a separate trial, was convicted of committing these murders and sentenced to death. This Court reversed Penalver’s murder convictions based on numerous errors, Penalver v. State, 926 So.2d 1118, 1137-38 (Fla.2006), and Penalver was acquitted on retrial. An essential part of Penalver’s defense was the assertion that he was not the subject in the videotape and in support of this he utilized an expert in forensic anthropology. In Ibar’s trial, Ibar’s private defense attorney, Kayo Morgan, failed to present a facial identification expert or forensic anthropologist despite Ibar’s request and his defense lawyer’s agreement to do so. At the postconviction evidentiary hearing, Ibar’s attorney, who detailed a litany of personal and professional issues that were occurring at the time of trial, testified that he did not understand “why [he] failed in this absolutely critical feature of the case”-in not having a facial identification expert testify, among other failings. <
[1015]*1015As this record bears out, there was simply no excuse for the numerous deficiencies and failures of Ibar’s defense attorney. None of the failures can be attributed to strategic moves nor could remotely constitute acceptable conduct for an attorney defending a first-degree murder charge with the death penalty being sought. Under any definition of “deficient performance,” Morgan could not be deemed to be functioning as defense counsel must perform to fulfill his or her crucial obligations to the defendant under the Sixth Amendment. While there were numerous deficiencies in performance, the most salient was the failure of trial counsel to present a facial identification expert to explain the physical differences between Ibar and the perpetrator alleged to have been him in the video, and to demonstrate that the quality of the images were so poor that they were inadequate to make a reliable identification. As we more fully explain, Ibar has established prejudice, given the relatively weak case against Ibar with no physical evidence linking him to the crime, the critical role of his identification derived from the video, and the errors we previously identified in Ibar’s direct appeal. Simply put, we cannot and do not have confidence in the outcome of this trial. Accordingly, we reverse the trial court’s denial of postconviction relief and remand for a new trial.
BACKGROUND
The facts of this case are set forth in Ibar’s direct appeal of his first-degree murder convictions and sentences of death:
On August 25, 1994, Pablo Ibar and Seth' Penalver were charged with three counts of first-degree murder, one count of burglary, one count of robbery, and one count of attempted robbery. ' Penal-ver and Ibar were initially tried together. . The first jury trial ended with a hung jury. Ibar and Penalver were eventually tried separately. Both Ibar and Penalver were ultimately convicted and sentenced to death.
On Sunday, June 26, 1994,' a Palm Beach County police 'officer discovered a Mercedes SL convertible on fire on a road twelve miles south of South Bay. The car was registered to Casmir Suc-harski, owner of a nightclub called Casey’s Nickelodeon. The officer who discovered the car notified the Miramar Police Department. A Miramar police officer went to Sucharski’s home to tell him that his car had beén found: The officer knocked on the door and received no answer. He stuck his card in the door and left.
The next morning, Monday, June 27, 1994, Marie Rogers’ mother reported her missing to the Broward County Sheriff’s Department. • Rogers had gone to Casey’s. Nickelodeon on-.Saturday, June 25, 1994, with her friend, Sharon Anderson, and did not1 return home. Deputy Christopher Schaub went to Casey’s Nickelodeon and learned that Suc-harski left the club early Sunday morning with Rogers and Anderson.' Schaub then went to Sucharski’s residence. Anderson’s car was in the driveway but no one answered the door. Schaub found a Miramar Police Department business card in the door and a blue T-shirt on the porch. He peered inside and saw three bodies. .
The police identified; the individuals found in the 'residence as . Sucharski, Rogers, and Anderson; All three died of gunshot wounds. ■ Because Sucharski had recently installed-.a video surveillance camera in his home, there was a videotape of the actual murders. The tape revealed that on Sunday, June 26, 1994, at 7:18 a.m., two. men entered through the back sliding door of Suchar-ski’s home. The intruder alleged to be Ibar initially had something covering his face, but he eventually removed it. The other intruder, alleged to be Seth Penal-[1016]*1016ver, wore a cap and sunglasses, which were never removed, and carried a firearm. The videotape showed that one of the intruders had a Tec-9 semiautomatic handgun with him when he entered the home. The other intruder displayed a handgun only after he went into another room and left the camera’s view. At one point, the intruder alleged to be Penal-ver hit Sucharski with a Tec-9 in the face, knocked him to the floor, and beat him on the neck, face, and body. This attack ■ on Sucharski lasted for nearly twenty-two minutes. The man later identified as Ibar shot Sucharski, Rogers, and Anderson in the back of the head. The intruder alleged to be Penal-ver then shot Anderson and Sucharski in the back.
During this time, the intruders searched Sucharski’s home. They rummaged through the home and entered the bedrooms and the garage. Suchar-ski was searched and his boots removed. Sucharski struggled and was repeatedly hit by both intruders. The intruders were seen putting things in their pockets. The State presented evidence that Sucharski kept ten to twenty thousand dollars in cash, carried a gun, and owned a Cartier watch. The watch was not found and Sucharski’s gun holster was empty.
Police took frames from the videotape and produced a flyer that was sent to law enforcement agencies. Three weeks after the murders, the Miramar police received a call from the Metro-Dade Police Department informing them that they had a man in custody on a separate and unrelated charge who resembled the photo on the flyer. The man in custody at the Metro-Dade Police Department was Pablo Ibar. Ibar was interviewed by Miramar investigators. He told police he lived with his mother, and that on the night of the murders he had been out with his girlfriend, whom he called both Latasha and Natasha.
Ibar actually lived with several friends in a rented home on Lee Street in Hollywood, Florida. One of his roommates was Jean Klimeczko. Klimeczko initially identified Ibar and Penalver as the men on the videotape. Klimeczko told police that early on the morning of the murders, Ibar and Penalver rushed into the Lee Street home, grabbed a Tec-9 that was kept at the house; and left. At the second trial, however, Klimeczko had no memory of his earlier statements. Other witnesses who had given earlier statements to police that the men in the photo looked like Ibar and Penalver also denied making identifications.
The jury found Ibar guilty on each charge.
Ibar v. State, 938 So.2d 451, 457-58 (Fla.2006) (footnotes omitted), cert. denied, 549 U.S. 1208, 127 S.Ct. 1326, 167 L.Ed.2d 79 (2007). After a penalty phase with a jury recommendation of death, the trial court sentenced Ibar to death concluding that the aggravators outweighed the mitigators. Id.
On direct appeal,1 we outlined the evidence against Ibar:
[1017]*1017In addition to the statements of [Roxa-na] Peguera, [Marlene] Yindel, [Maria] Casas, and [Jean] Klimeczko identifying Ibar, which Ibar concedes was proper as impeachment evidence but not substantive evidence, there were other witnesses and items of evidence from which the jury could conclude that Ibar was one of the perpetrators of this triple homicide. First, there was a videotape of the murders. The perpetrator identified as Ibar removed his disguise and his face was visible on the videotape. This videotape was played for the jury. Gary Foy, one of Sucharski’s neighbors, testified .that he saw two men leaving in Sucharski’s Mercedes-Benz. He stated that he did not get a good look at the driver of the car, but he got a good look at the passenger. Foy identified Ibar as the passenger in the Mercedes. Klimec-zko testified that at some point both Penalver and Ibar came to the residence on Lee Street in a big, black, shiny new car. Although [Ian] Milman denied that he had ever positively identified Ibar as the person in the still photograph made from the videotape, he did say that the person in the photograph resembled Ibar. Moreover, the trial judge admitted as substantive evidence Milman’s grand jury testimony in which he positively identified Ibar. [Melissa] Munroe’s statement placing Ibar and Penalver together during the weekend of the murder was also admitted as substantive evidence. On the issue of identification, the jury also heard evidénce from Kimberly San and David Phillips that' placed Ibar and Penalver in the Mercedes. Both Peguera and her mother testified that the person ⅛ the photograph resembled Ibar. We conclude that any error in admitting some of these identification statements as substantive evidence rather than as impeachment evidence was harmless error. !
Id. at 463. Significantly for our analysis in this case, we found error in allowing certain identification testimony to be admitted in evidence but concluded that it was “harmless.” Having concluded that there was no reversible error demonstrated on appeal, we' affirmed Ibar’s convictions of first-degree murder and his sentences of death. Id. at 457.
Ibar thereafter filed his initial motion for postcdnviction relief, raising claims of ineffective assistance of trial counsel, due process violations including Brady v. Maryland, 373 U.S. 83, 83 S.Ct. 1194, 10 L.Ed.2d 215 (1963), and a claim under Ring v. Arizona, 536 U.S. 584, 122 S.Ct. 2428, 153 L.Ed.2d 556 (2002).2 Following an evidentiary hearing, the trial court denied Ibar postconviction relief. In addition to appealing the denial of relief, Ibar filed a petition for a writ of habeas corpus and for extraordinary relief.
ANALYSIS
Ibar claims that his guilt-phase counsel provided ineffective assistance as a result of numerous deficiencies in performance and asserts that the deficiencies resulted in prejudice. Ineffective assistance of counsel claims aré evaluated in accordance with the Supreme Court’s decision in Strickland v. Washington, 466 U.S. 668, 104 S.Ct. 2052, 80 L.Ed.2d 674 (1984). [1018]*1018“[T]he test when- assessing the actions of trial counsel is not how, in hindsight, present counsel would have proceeded.” Bradley v. State, 33 So.3d 664, 671 (Fla.2010). To succeed on. a claim that trial counsel was ineffective, a defendant must establish two criteria:
First, counsel’s performance must be shown to be deficient. Strickland[, 466 U.S. at 687, 104 S.Ct. 2052], Deficient performance in this context means that counsel’s performance fell below the standard .guaranteed by the. Sixth Amendment. Id. When examining counsel’s performance, an objective standard of reasonableness applies, id. at 688, 104 S.Ct. 2052, and great deference is given to counsel’s performance. Id. at 689, 104 S.Ct. 2052. The defendant bears the burden to “overcome the presumption that, .under the circumstances, the challenged action ‘might be considered sound trial strategy.’ ” Id. (quoting Michel v. Louisiana, 350 U.S. 91, 101, 76 S.Ct. 158, 100 L.Ed. 83 (1955)). This Court has made clear that “[strategic decisions do not constitute ineffective assistance of. counsel.” See Occhicone v. State, 768 So.2d 1037, 1048 (Fla.2000). There is a strong presumption that trial counsel’s performance was not ineffective. See Strickland, 466 U.S. at 669, 104 S.Ct. 2052.
Second, the deficient performance must have prejudiced the defendant, ultimately depriving the defendant of a fair trial with a reliable result. [Id.] at 689, 104 S.Ct. 2052. A defendant must do more than speculate that. an error affected the outcome. Id. át 693, 104 S.Ct. 2052. Prejudice is met only if 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. Both deficient performance and prejudice must be shown. Id.
Id. at 671-72.
Ineffective assistance claims are examined under a mixed standard of review because the performance and prejudice elements of Strickland present mixed questions of law and 'fact. Bradley, 33 So.3d at 672. Postconviction courts hold a superior vantage point with respect to questions of fact, -evidentiary weight, and observations of the demeanor and credibility of witnesses. See Cox v. State, 966 So.2d 337, 357-58 (Fla.2007). As a result, this Court ‘ defers to thé postconviction court’s factual findings so long as those findings are supported by competent, substantial evidence.' See Bradley, 33 So.3d at 672. However, this Court reviews the postconviction court’s legal conclusions de novo. Id. Deficiency and prejudice are mixed questions of law and fact, so this Court reviews those findings de novo. Frances v. State, 143 So.3d 340, 349 (Fla.2014).
In his claim for ineffective assistance of counsel, Ibar identifies numerous deficiencies, particularly in that trial counsel failed to procure a facial identification expert.3 [1019]*1019The unrefuted testimony as to the reasons for these multiple, failures came from Ibar’s own defense lawyer, Kayo, Morgan. Although we do not simply accept, trial counsel’s concession that he was deficient in performance, the details of his testimony overwhelmingly bear this out.
As to the issue of facial, identification experts, Morgan recalled that Dr. Iscan, Penalver’s expert witness, testified at Ibar and Penalver’s joint trial that the facial features of the individual in the video did not comport with the.same indicators of Penalver, and consequently, Dr. Iscan was unable to make a definitive identification that the perpetrator in the video was Pen-alver. Morgan testified that prior to the joint trial, Dr. Iscan told him that it was even less likely that Ibar was the perpetrator. During the joint trial, counsel for Penalver requested Morgan to stay away from Dr. Iscan, and Morgan honored that request.
After the joint trial resulted in a hung jury, Morgan became involved with a drug addict who became pregnant with his child. Although Morgan'was sick with reoccurring bouts of pneumonia and other complications, he dedicated himself to' the woman. Morgan’s personal life impacted his practice. .In January 1999, during the time of jury selection in Ibar’s case, Morgan was charged with .aggravated battery on the woman. Morgan was in emotional and physical pain, suffering from extreme duress, and “was not there” mentally. As Morgan explained during the postconviction evidentiary hearing, his whole concern was the woman, her plight, and the health of their child.
In defending Ibar,. Morgan knew it was of upmost importance for the defense to draw issues with the witness identification infirmities. From the beginning, Morgan knew it was incumbent upon him to have a forensic anthropologist or a facial identification expert, which was “vital” to attack the unreliability of the identification. Recognizing that Dr. Iscan’s testimony “had to. be” in Ibar’s case, Morgan attempted to have. Dr. Iscan meet with Ibar, but Dr. .Iscan was out of the country. Morgan recalled consulting with two other forensic anthropologists, but they were unable to work for indigent fees.. Ibar told Morgan that he wanted a “forensic thing,” but, despite Morgan’s understanding of the critical nature of such evidence, Morgan talked Ibar out of it.
In January 1999, per Morgan’s request, Barbara Brush entered her appearance as Ibar’s second-chair and penalty phase counsel. Morgan intended for Brush tb take some of the load off of him. As a result, Brush participated in the trial every day unless she was covering for Morgan elsewhere. Delegating the garnering of expert witnesses to Brüsh, Morgan instructed Brush that the defense needed forensic anthropologist Dr. Anthony Fal-setti. On January 31,2000, less than three months before Ibar’s separate trial was to begin, Morgan faxed Brush a note which said as follows:
We spoke about getting monies approved for experts. Here are three we can start with — We will also need, the anthropol[o]gist, [Dr.] Falsetti, who can establish discrepancies in the culprit and Ibar.... As usual I have put my back to the wall. We need to move ÁSAP.
Defense counsel obtained an order that provided payment for expert fees to Dr. Falsetti and listed Dr. Falsetti .as a-potential witness. While Morgan denied having [1020]*1020any contact with Dr. Falsetti - himself, Brush’s billing statement reveals that about a month after the fax, Brush had a twelve-minute telephone conference with Dr. Falsetti. No other contact was apparently made.'
Despite this limited contact with Dr. Falsetti, Morgan understood that presenting Dr. Falsetti’s perspective to the jury was “vital” — the “heart of the case” — that he did not understand how he did not “perfect” the participation of Dr. Falsetti at Ibar’s trial, that he “failed in this, absolutely critical feature of the case,” and that not having-a-forensic expert was a “significant omission” and not a strategic decision. Moreover, Morgan did not inform the judge about the extent of his sickness or about all of the problems in his life in seeking a continuance. Morgan concluded that he was “defective” in his representation of Ibar.
In April 2000, -Ibar’s separate trial commenced. Morgan suffered from recurring bouts of pneumonia, breathlessness, depression,. recurring sinusitis, bronchitis-like asthma, fatigue, insomnia, nausea, and fluctuating weight. Morgan’s medical problems worsened as Ibar’s trial progressed. In addition, Morgan was involved in a custody situation regarding his daughter, and Morgan wanted to focus on saving his daughter’s mother from her plight. Morgan claimed he could not withdraw from the case because of a promise he made to Ibar’s mother.
Morgan acknowledged that during his opening statement he told the jury they would see significant distinctions between the perpetrator and Ibar, including differences with their hairlines and eyes and that Ibar had a cut on his eye which was not seen in the video. Morgan explained that his mind was not fully working throughout Ibar’s trial, that he conducted poor cross-examinations, that he looked for shortcuts, and that it was difficult to concentrate. After the jury rendered its verdict, Morgan was hospitalized. He did not attend Ibar’s sentencing.
Dr. Falsetti'testified at the evidentiary hearing that in 2000, facial recognition was accepted in Florida courts for the comparisons of photographs' of an unknown to a known on a case-by-case basis by anthropologists and anthropometrists. Dr. Fal-setti was certain that he did not consult in Ibar’s case in '2000 and that' he never received' any material from either Morgan or Brush relating to the case. Dr; Falsetti stated that had he been contacted then, there would have been no reason why he would not have been willing to provide his services on identification issues.
Facial identification expert Raymond Evans testified at the evidentiary hearing that his work in facial identification is based upon scientific principles and is accepted as a valid and reliable scientific discipline within the scientific community — and was so recognized in 2000. Evans explained that because poor images have some resemblances to a .referenced image, lay persons — who are generally unable to factor in disc.olorati.on, distortion, or other factors — may be lulled into believing that the images have.to be depicting the same person.
Evans found the crime scene- images distilled from the surveillance ■ videotape had very poor quality and lighting and very low resolution. Evans maintained that the images were not adequate to make a reliable identification. In comparing the facial proportion's of Ibar with the perpetrator alleged to have been him, Evans found discrepancies with their respective jaw lines, right eyebrows, the width of the mouth's, and dorsal ridge shape. Although he was not able to completely exclude Ibar because of general similarities, Evans opined. that it is not possible to conclude that the perpetrator and Ibar are [1021]*1021the same person because of the noted-differences.
Deficiency
Ibar claims that his trial counsel were deficient for failing to procure a: facial identification expert or forensic anthropologist to establish the difficulty in being able to positively identify Ibar as one of the perpetrators in the crime surveillance video and photo distillations and. tó show the physical discrepancies between Ibar and the perpetrator. After carefully reviewing the record, we conclude that trial counsel Morgan was deficient for failing to present a facial identification' expert to challenge the State’s charge that Ibar was the perpetrator seen in the videotape committing the murders at the Sucharski residence.4 This videotape, and the images distilled therefrom, were instrumental to the State’s case. Even as early as the investigation, the police questioned numerous people who knew Ibar and showed them photographs created from the video to determine whether they identified the person in the photo as somebody who resembled Ibar. The jury watched the video — which was of very poor quality — and heard about countless improper lay witness identifications of Ibar based on these images. While defense counsel objected.to this type of identification evidence .being used as substantive evidence, a point to which we generally agreed on direct appeal, defense counsel did not present other substantive evidence to challenge this important piece of evidence, even though he knew it was successfully used in Ibar’s codefendant’s trial. See Ibar, 938 So.2d at 463 (holding that the trial court erred in allowing several prior identifications to be considered as substantive evidence).
Testimony from an expert on the subject matter of facial identification was certainly admissible. Dr. Iscan testified on behalf of Penalver in the joint trial which lead to a hung jury, and in Penalver’s separate trial after the defendants were severed from each other, after which Penalver was acquitted. Dr: Iscan, referring to the same video at issue in this case, testified that
because of the poor quality of the video and the lighting conditions, he could not reach a positive conclusion about whether the individual in the video was Penal-ver. Dr. Iscan noted that there were discrepancies in the lower, half of the face which led him to lean to a conclusion that the individual on the tape was not Penalver.
Penalver, 926 So.2d at 1125-26 (footnote omitted). Morgan, Ibar’s trial counsel, who was suffering from significant personal issues at the time of trial, was certainly aware of the importance of Dr. Iscan’s expert testimony. In comparing Ibar to Penalver, Dr.. Iscan told Morgan that it was even less likely that Ibar was one of the perpetrators.
[1022]*1022Ibar presented an alibi defense at trial, which was that he; was not at the Suchar-ski residence'at the.time of the.murders; instead, the defense presented that, at that time, he was with Tonya Quinones in her home. The calling of. a facial identification expert, therefore, would not have been inconsistent with Ibar’s alibi. Gaxy Foy’s identification of Ibar as the passenger in Sueharski’s vehicle leaving Sucharski’s home on thé morning of the murders was strong evidence presented by the State to suggest that Ibar was one of the perpetrators inside of the home at the time of the murders.5 Challenging the video by way of a facial identification expert to point out problems with the video and the purported identifications of Ibar was necessary in the face of Foy’s testimony.
In light of the foregoing, we conclude that’Morgan’s'performance fell'below the standard guaranteed by the 'Sixth Amendment. We emphasize that in conducting our review of Morgan’s performance we do not rely on his admissión that he was defective but rather on his complete failure to pursue the important defense that Ibar was not the perpetrator of the crime through discrediting the videotape and the State’s evidence as to that identification. See Harris v. Dugger, 874 F.2d 756, 761 n. 4 (11th Cir.1989) (“[Admissions of deficient performance by attorneys are not decisive.”). Ibar has therefore “overcome the presumption that, under the circumstances, the challenged action ‘might be considered sound trial strategy.’ ” Strickland, 466 U.S. at 689, 104 S.Ct. 2052 (quoting Michel v. Louisiana, 350 U.S. 91, 101, 76 S.Ct. 158, 100 L.Ed. 83 (1955)).
Prejudice
Having found that .Morgan was deficient, we must decide whether Ibar has shown that there is a reasonable probability that “but for counsel’s unprofessional errors, the result of the proceeding would have been different.” See id.. at 694, 104 S.Ct. 2052. Prejudice is not determined by a “more likely than not” standard but rather is expressed in terms of undermining confidence in the outcome. See Porter v. McCollum, 558 U.S. 30, 44, 130 S.Ct. 447, 175 L.Ed.2d 398 (2009); see also Wheeler v. State, 124 So.3d 865, 873 (Fla.2013); Jennings v. State, 123 So.3d 1101, 1113 (Fla.2013); Alcorn v. State, 121 So.3d 419, 425 (Fla.2013). We conclude that Ibar has met this burden.
■Raymond' Evans, Ibar’s postconviction expert, opined that the crime scene images were not adequate to -make a reliable identification due to the lack of quality of the images created from the videotape.- Finding discrepancies in the facial proportions of Ibar with that of the perpetrator alleged to have been- him, Evans opined that-it is not possible to conclude that the perpetrator and Ibar are the same person. ■ This type of testimony could have been extremely powerful to undercut the State’s reliance on the videotape.
Moreover, there was a lack of any DNA or physical evidence linking Ibar’s involvement to the murders. The videotape reveals that the perpetrator, alleged to have been Ibar, at one point removed a t-shirt that had been partially covering his ‘face. See Ibar, 938 So.2d at 458 (“The intruder alleged to be Ibar initially had something covering his face, but he eventually removed it.”); State v. Ibar, No. 94CF13062 (Fla. 17th Jud.Cir.Ct. Aug. 28, 2000) (Sentencing Order at 3) (noting that Ibar “was observed on the videotape, removing a shirt that had been partially covering his [1023]*1023face”). This t-shirt was later discovered on Sucharski’s porch and was tested for DNA. Notably, although the shirt contained blood, hair, and cellular material on the collar, none of the evidence obtained from the shirt was consistent with Ibar’s DNA. In fact, Ibar was excluded as the source of the DNA derived from small amounts of blood obtained from the t-shirt cutting, the hairs taken from the t-shirt were not consistent with Ibar’s hair, standards, and there was no match between a partial profile and Ibar as to the cellular material of the t-shirt’s collar and the left and right underarm.
While identification of Ibar as the perpetrator was the crux of the State’s case, on direct appeal we held that the trial court “erred in allowing several of the identification statements to be considered as substantive evidence.” Ibar, 938 So.2d at 463. Specifically, at trial the State called police investigators to testify that Roxana ’Pegu-era, Marlene Vindel, Maria Casas, and Jean Klimeczko had previously confirmed the identity of the person in a photograph — created from the video surveillance tape — as Ibar. Id. at 459-60. We concluded that these prior identifications should not have been admitted as substantive evidence, but that the errors were harmless. Id, at 460-64. In our harmless error anafy^ we first relied on the fact that the videotape of the murders was played for the jury, and wé referred- to properly admitted identifications of.Ibar based on the images. Id. at 463. Our harmless error analysis would - have undoubtedly been.different in this case had the surveillance videotape and images been challenged by a facial identification, expert at trial.
There are also similarities 'between' Ibar and Penalver’s case, where we reversed Penalver’s convictions of three counts of first-degree murder and death sentences, and remanded for a new trial because he was denied a fair trial by the prejudicial admission of irrelevant and inadmissible evidence repeatedly elicited by the State over objections. Penalver, 926 So.2d at 1138. We reasoned that there was little evidence offered against Penalver:
While the State presented circumstantial evidence regarding Penalver’s involvement in the crime, only two pieces of direct evidence tying Penalver to the murders were presented: .a photographic still taken from, a grainy videotape depicting a person alleged to be Penal-ver who was attired in a cap and sunglasses that concealed his face; and a statement allegedly made by Penalver to another inmate .that h,e had a chance of being acquitted because he did not remove his mask. In light of the scant evidence connecting Penalver to ■ this murder and the cqnsequent importance of identifying .the individual depicted on the videotape in sunglasses and hat, we conclude that the improperly admitted evidence and the State’s suggestion that the defense tampered with or suborned perjury' by an identification witness meet the cumulative error requirements outlined above and require reversal.
Id. We also expressed that there was no physical evidence tying Penalver to the murders. Id. at 1125. Penalver later was acquitted in his retrial...
■ Similarly, “[i]n light of the scant evidence connecting” Ibar “to this -murder and the consequent importance of identifying the individual depicted on the videotape,” alleged to have been Ibar, see id. at 1138, we conclude that trial counsel’s deficiency, in failing to procure a facial identification expert, undermines our confidence in Ibar’s trial.6 Accordingly, we reverse [1024]*1024the trial court’s denial of postconviction relief.7
CONCLUSION
For the reasons expressed above, we reverse the trial court’s denial of postcon-viction relief. Ibar is entitled to a new trial.
It is so ordered.
LABARGA, C.J., and PARIENTE, POLSTON, and PERRY, JJ., concur.
QUINCE, J., dissents with an opinion, in which CANADY, J.,' concurs.
LEWIS, J., dissents.