Full opinion text
W. EUGENE DAVIS, Circuit Judge: Carlos Trevino. (“Trevino”) seeks a certificate of appealability (“COA”) to appeal the district court’s dismissal, on the pleadings and without an evidentiary hearing, of his habeas corpus petition under 28 U.S.C. § 2254, claiming that he was deprived his Sixth Amendment right to effective assistance of counsel when his trial counsel allegedly failed to adequately investigate and present mitigation evidence at the punishment phase of his capital murder trial. The district court held that Trevino’s claim is procedurally barred, even under Martinez v. Ryan, — U.S. -, 132 S.Ct. 1309, 182 L.Ed.2d 272 (2012), and Trevino v. Thaler, — U.S. -, 133 S.Ct. 1911, 185 L.Ed.2d 1044 (2013), because Trevino failed to sufficiently allege that his initial state habeas counsel rendered ineffective assistance. In the alternative, the district court held that his claim must be dismissed on the merits because the new mitigating evidence Trevino seeks to develop and admit, which the district court characterized as “double-edged,” could not outweigh the substantial aggravating evidence. Trevino seeks a COA, arguing that reasonable jurists could debate whether the district court properly dismissed his claims on the pleadings and whether it erred by failing to hold an evidentiary hearing. For the reasons set out below, we conclude that reasonable jurists could debate whether the district court correctly dismissed his habeas claim with respect to potential evidence of his fetal alcohol syndrome (“FAS”)' or, more broadly, fetal alcohol spectrum disorder (“FASD”). Indeed, reasonable jurists would agree that the district court erred in prematurely dismissing that claim. We also conclude that no reasonable jurist could debate whether the. district court erred in dismissing his habe-as claim with respect to his additional character witness testimony that is not relevant to an FASD diagnosis or whether the district court erred by failing to hold an evidentiary hearing before its dismissal on the pleadings. I. Factual BaCkground and Procedural History This is the second time this case has come before us and the first in which we reach beyond the procedural default issue to address the merits. Trevino contends that if his trial counsel had conducted a constitutionally sufficient investigation, he not only would have located more witnesses to testify about his character but also would have been able to discover and introduce evidence that Trevino suffers from FASD. The background is set out more fully in the prior opinions in this case, but we summarize them here for convenience. A. Trial Trevino was convicted for the June 9, 1996 gang rape and murder of 15-year-old Linda Salinas in San Antonio, Texas. One of the other participants, Juan Gonzales (Trevino’s cousin), testified on behalf of the prosecution that he was with the group that night, though he walked away briefly during the murder. He testified that although Trevino did not rape Salinas himself, he held her down while another participant raped her; that Trevino encouraged Gonzales to rape her (though Gonzales refused); that Trevino discussed the need to get rid of Salinas as a witness; and that Trevino later appeared with blood on his shirt. Gonzales also testified that Trevino bragged after the murder that he had “learned how to kill in prison” and “learned how to use a knife in prison.” Thus, although Gonzales did not witness the murder itself, he presented substantial testimony of Trevino’s involvement in the crime. Salinas’s body was discovered the day after her murder, and an autopsy revealed that she suffered two stab wounds to her neck, one of which was fatal, as well as other injuries consistent with sexual assault. The prosecution presented this autopsy evidence at Trevino’s trial. The prosecution also presented testimony from forensic and DNA experts establishing that fibers found on Salinas’s clothing were consistent with fibers from Trevino’s slacks and that Trevino’s DNA could not be excluded as a source of DNA found in Salinas’s panties. Based on this and other evidence, the jury returned a guilty verdict. During the punishment phase, the prosecution presented substantial evidence regarding Trevino’s culpability and future dangerousness, including his former arrests and admitted membership in a violent street gang. Trevino’s state trial counsel called only one witness, Trevino’s aunt, who testified generally about his rough childhood and his mother’s alcoholism. Her testimony, comprising approximately five pages, made up the entirety of his mitigation case. At the conclusion of the punishment phase, the jury found (1) that he constituted a future risk of dangerousness, (2) that he had actually caused the death of Salinas or, if he did not actually cause her death, that he intended to kill her or another, or anticipated a loss of life, and (3) that there were insufficient mitigating circumstances to warrant a sentence of life imprisonment. In accordance with the verdict, the state trial court imposed a sentence of death. B. Post-Conviction Proceedings The Supreme Court summarized the post-conviction proceedings, which are central to this COA application, as follows: Eight days later the judge appointed new counsel to handle Trevino’s direct appeal. Seven months after sentencing, when the trial transcript first became available, that counsel filed an appeal. The Texas Court of Criminal Appeals then considered and rejected Trevino’s appellate claims. Trevino’s appellate counsel did not claim, that Trevino’s trial counsel had been constitutionally ineffective during the penalty phase of the trial court proceedings. About six months after sentencing, the trial judge appointed Trevino a different new counsel to seek state collateral relief. As Texas’ procedural rules provide, that third counsel initiated collateral proceedings while Trevino’s appeal still was in progress. This new counsel first sought postconviction relief (through collateral review) in the trial court itself. After a hearing, the trial court denied relief; and the Texas Court of Criminal Appeals affirmed that denial. Trevino’s postconviction claims included a claim that his trial counsel was constitutionally ineffective during the penalty phase of Trevino’s trial, but it did not include a claim that trial counsel’s ineffectiveness consisted in part of a failure adequately to investigate and to present mitigating circumstances during the penalty phase of Trevino’s trial. [See] Wiggins v. Smith, 539 U.S. 510, 523, 123 S.Ct. 2527, 156 L.Ed.2d 471 (2003) (counsel’s failure to investigate and present mitigating circumstances deprived defendant of effective assistance of counsel). Trevino then filed a petition in federal court seeking a writ of habeas corpus. The Federal District Court appointed another new counsel to represent him. And that counsel claimed for the first time that Trevino had not received constitutionally effective counsel during the penalty phase of his trial in part because of trial counsel’s failure to adequately investigate and present mitigating circumstances during the penalty phase. App. 438, 456-478. Federal habeas counsel pointed out that Trevino’s trial counsel had presented only one witness at the sentencing phase, namely Trevino’s aunt. The aunt had testified that Trevino had had a difficult upbringing, that his mother had an alcohol problem, that his family was on welfare, and that he had dropped out of high school. She had added that Trevino had a child, that he was good with children, and that he was not violent. Id., at 285-291. Federal habeas counsel then told the federal court that Trevino’s trial counsel should have found and presented at the penalty phase other mitigating matters that his own investigation had brought to light. These included, among other things, that Trevino’s mother abused alcohol while she was pregnant with Trevino, that Trevino weighed only four pounds at birth, that throughout his life Trevino suffered the deleterious effects of Fetal Alcohol Syndrome, that as a child Trevino had suffered numerous head injuries without receiving adequate medical attention, that Trevino’s mother had abused him physically and emotionally, that from an early age Trevino was exposed to, and abused, alcohol and drugs, that Trevino had attended school irregularly and performed poorly, and that Trevino’s cognitive abilities were impaired. Id., at 66-67. The federal court stayed proceedings to permit Trevino to raise this claim in state court. The state court held that because Trevino had not raised this claim during his initial postconviction proceedings, he had procedurally defaulted the claim, id., at 27-28; and the Federal District Court then denied Trevino’s • ineffective-assistance-of-trial-counsel claim, id., at 78-79. The District Court concluded in relevant part that, despite the fact that “even the most minimal investigation ... would have revealed a wealth of additional mitigating evidence,” an independent and adequate state ground (namely Trevino’s failure to raise the issue during his state postcon-viction proceeding) barred the federal habeas court from considering the ineffective-assistance-of-trial-counsel claim. Id., at 131-132. See Coleman v. Thompson, 501 U.S. 722, 729-730, 111 S.Ct. 2546, 115 L.Ed.2d 640 (1991). Trevino appealed. The Fifth Circuit, without considering the merits of Trevino’s ineffeetive-assistance-of-trial-coun-sel claim, agreed with the District Court that an independent, adequate state ground, namely Trevino’s procedural default, barred its consideration. 449 Fed. Appx. at 426. In 2011, when the panel decided Trevino’s appeal, there was no applicable exception to the procedural default rule under any state habeas scheme. In 2012, however, the Supreme Court decided Martinez, which held that a federal habeas petitioner was not barred from asserting an ineffective-assistance-of-trial-counsel claim if (1) the state habeas scheme (such as Arizona’s in Martinez) required a defendant convicted at trial to raise that claim during his first state habeas proceeding, and (2) defendant’s counsel during his initial state habeas proceeding was ineffective. Trevino filed a. petition for a writ of certiorari, seeking a determination that the Martinez rule should also apply to the Texas habeas scheme. The Supreme Court explained that although the Texas scheme did not require a defendant to raise an ineffective-assistance-of-trial-counsel claim in his first state habeas proceeding, the result should be the same: [W]e believe that the Texas procedural system — as a matter of its structure, design, and operation — does not offer most defendants a meaningful opportunity to present a claim of ineffective assistance of trial counsel on direct appeal. What the Arizona law prohibited by explicit terms, Texas law precludes as a matter of course. And, that being so, we can find no significant difference between this case and Martinez. The very factors that led this Court to create a narrow exception to Coleman in Martinez similarly argue for the application of that exception here. Thus, the Court applied the rule of Martinez to Texas’ scheme for postconviction relief, i.e.: “[A] procedural default will not bar a federal habeas court from hearing a substantial claim of ineffective assistance at trial if, in the initial-review collateral proceeding, there was no counsel or counsel in that proceeding was ineffective.” Accordingly, it remanded to the Fifth Circuit: Given this holding, Texas submits that its courts should be permitted, in the first instance, to decide the merits of Trevino’s ineffective-assistance-of-trial-counsel claim. We leave that matter to be determined on remand. Likewise, we do not decide here whether Trevino’s claim of ineffective assistance of trial counsel is substantial or whether Trevino’s initial state habeas attorney was ineffective. For these reasons we vacate the Fifth Circuit’s judgment and remand the ease for further proceedings consistent with this opinion. We remanded to the district court as follows: In light of the Supreme Court’s decision in Trevino v. Thaler, — U.S. -, 133 S.Ct. 1911, 185 L.Ed.2d 1044 (2013), we remand to the district court for full reconsideration of the Petitioner’s ineffective assistance of counsel claim in accordance with both Trevino and Martinez v. Ryan, — U.S. -, 132 S.Ct. 1309, 182 L.Ed.2d 272 (2012). If the Petitioner requests it, the district court may in its discretion stay the federal proceeding and permit the Petitioner to present his claim in state court. C. District Court Proceedings on Remand On remand, the district court, in April to July 2014, “granted Petitioner’s multiple requests for additional time to investigate and develop Petitioner’s remaining claims for relief and authorized Petitioner to expend resources in excess of the statutory cap set forth in 18 U.S.C. Section 3599(g) (2) for investigative and expert assistance.” On November 13, 2014, the district court held a status conference concerning pending motions, and it entered an order granting in part Trevino’s motions for additional time and for expert funding that same day, which reads, in part: After hearing arguments from both parties, for the reasons discussed at length during the hearing, the parties are directed to file amended pleadings designed to clarify the issues remaining in this cause and Petitioner should be permitted to proceed with some, but not all, of the expert examination of Petitioner requested in the motion for expert assistance. Once the parties have clarified their positions and the issues are more focused, the Court will hold another hearing to ascertain how best to proceed with the remainder of this cause. On February 2, 2015, Trevino filed his second amended federal habeas petition, and the state filed its response on May 26, 2015. On June 11, 2015, without holding a hearing or otherwise alerting the parties to its impending decision, the district court sua sponte issued its 36-page memorandum opinion and order, based on' the pleadings, denying all relief under the second amended habeas petition and denying a COA. The court noted that it had rejected all five claims presented in Trevino’s first amended habeas petition on the merits and had alternatively held that two of them were procedurally defaulted, including the ineffective-assistance-of-trial-counsel claim now presented in his second amended petition. In its new order, it reasoned that Trevino failed to show cause for excusing his procedural default even under Martinez/Trevino, but even if he could overcome the procedural default, his claim would still be subject to dismissal on the merits because none of the “new” mitigating evidence referred to in the second amended petition changed the district court’s analysis set out in its earlier opinion, as discussed below. 1. Martinez/Trevino Issue With respect to the Martinez/Trevino issue, the district court concluded that Trevino still had failed to overcome the procedural default bar. Specifically, it held that Trevino failed to sufficiently allege that his state habeas counsel was ineffective, on the ground that the evidence at issue was not available to the first state habeas counsel at the time. The court explained that none of the “new” mitigating evidence (including testimony from Trevino’s mother and evidence about his background and history) had been gathered by his state trial counsel. The district court reasoned that Trevino’s state habeas counsel cannot reasonably be faulted, much less declared “ineffective,” for failing to develop and present an ineffective assistance claim during Petitioner’s initial state habeas corpus proceeding premised upon “new” mitigating evidence absent some showing this “new” mitigating evidence was reasonably available to said counsel at the time of Petitioner’s initial state habeas corpus proceeding. On this basis, the district court held that Trevino had failed to show cause under Marbinez/Trevino for his procedural default, and that his claim was still subject to dismissal on this ground alone. The district court also held that even if Trevino had overcome the procedural default bar, his claim should be dismissed on the merits. 2. Merits of the Claim On its merits determination, the district court relied extensively on its alternative holding in its 2009 opinion that the “new” mitigating evidence (concerning Trevino’s character, childhood abuse and neglect, alcohol and narcotics abuse, school performance, and possible FASD) could not outweigh the substantial aggravating evidence presented at trial. The district court listed some of the aggravating evidence, including Trevino’s “callous comments” following the murder of Salinas, his participation in the violent assault, and his gang membership, but the district court placed special emphasis on “the complete and total absence of any indication the Petitioner has ever expressed sincere contrition or genuine remorse over Salinas’ murder.” In the district court’s estimation, Trevino’s apparent lack of remorse seemed to be the primary piece of aggravating evidence: The latter point cannot be over-emphasized. Salmas’ murder was particularly brutal and senseless. Yet Petitioner has consistently refused to acknowledge his role in her murder, even to his own trial counsel, claiming instead to have been “too stoned” to remember exactly what happened that evening. Petitioner’s own affidavit, executed June 11, 2004, contains not even a scintilla of sincere contrition; instead Petitioner expresses hostility and blames his trial counsel for allegedly misrepresenting the terms of a proffered plea bargain for a life sentence without accepting any responsibility for his own rejection of the offer after it was accurately described to Petitioner. Absent some indication the Petitioner has willingly accepted responsibility for his role in Salinas’ brutal rape and murder, the evidence showing Petitioner’s long history of alcohol and drug abuse, long history of criminal misconduct, and membership in violent street and prison gangs precludes this Court from finding this aspect of Petitioner’s ineffective assistance claims herein satisfies the prejudice prong of Strickland. There is simply no reasonable probability that, but for the failure of Petitioner’s trial counsel to present Petitioner’s capital sentencing jury with the additional, double-edged, mitigating evidence now before this Court, the outcome of the punishment phase of Petitioner’s capital trial would have been different. The district court concluded that the second amended petition did not change the balance because the “new” mitigating evidence was fundamentally the same “double-edged” evidence it had addressed in its earlier opinion. The district court summarized the “new” evidence as follows: Petitioner asserts that trial counsel could have called various witnesses who would have offered supportive testimony (e.g., Janet Cruz, the mother of his two children; Mario Cantu, friend; Ruben Gonzalez, employer; Jennifer DeLeon, his sister). One of the experts recently retained opines that Petitioner “presents with characteristics of Fetal Alcohol Affect”, and a “low average range of intellectual functioning.” She further opines that his “history of Fetal Alcohol Affect, along with his history of physical and emotional abuse” contributed to his “inability to make appropriate decisions.” She opines that this may also have contributed to Petitioner rejecting the plea offer made to him that would have spared him from the death sentence. Another expert opines that based on his preliminary assessment, Petitioner suffers from “8 domains” of poor “cognitive functioning,” (i.e., academics, verbal and visuospatial memory, visuospatial construction, processing speed, executive functioning, communication skills, daily living skills and socialization skills). This expert states that although his assessment is a “critical component in the FASD diagnostic process,” the diagnosis of FASD must be made by a medical doctor. According to this expert, yet unexamined is whether Petitioner’s “FASD has resulted in an organic brain disorder.” In summary, Petitioner argues that, had the “jury been able to consider [Petitioner’s] mixed up and unexplainable turbulent and chaotic life history on the mitigating side of the scale, there is unquestionably a reasonable probability that at least one juror would have struck a different balance.” The expert “recently retained” was Dr. Rebecca A. Dyer, Ph.D., of Forensic Associates of San Antonio, whose 18-page report dated May 6, 2004, was attached to Trevino’s earlier habeas petition that was the subject of the district court’s 2009 opinion, and was again attached to his second amended habeas petition. The district court correctly set out the standards for uncalled witnesses and uninvestigated facts as follows: “To prevail on an ineffective assistance claim based upon uncalled witnesses, an applicant must name the witness, demonstrate that the witness would have testified, set out the content of the witness’s proposed testimony, and show that the testimony would have been favorable.” Gregory v. Thaler, 601 F.3d 347, 352 (5th Cir.), cert. denied, 562 U.S. 911, 131 S.Ct. 265, 178 L.Ed.2d 175 (2010). “An applicant ‘who alleges a failure to investigate on the part of his counsel must allege with specificity what the investigation would have revealed and how it would have altered the outcome of the trial.’ ” Id. Applying these standards, the district' court addressed two distinct categories of proposed evidence: (1) character witness testimony and (2) evidence pertaining to Trevino’s possible FASD. The district court explained that although the character witness testimony did contain some mitigating evidence, it contained a great deal of aggravating evidence as well, which would serve to bolster the prosecution’s case. Among • the aggravating evidence was testimony that Trevino was “always high” from sniffing spray paint, that he was abusive to the mother of one of his children and had two sides to his personality, that he was “always jealous,” “angry,” “violent,” and “impulsive” even when he was not drunk, and that he always had a gun. Thus, the district court concluded that the “new” character witness testimony was “double-edged” and, if introduced, could not have affected the outcome. Next, the district court found that the FASD evidence was also “double-edged,” though the court’s language suggested that the FASD evidence may be more mitigating than aggravating: Finally, this Court has previously noted the double-edged nature of a diagnosis of Fetal Alcohol Syndrome or Fetal Alcohol Effects. This Court has also noted that a diagnosis of Fetal Alcohol Syndrome or Fetal Alcohol Effects or Fetal Alcohol Spectrum Disorder was not within the mainstream of psychological diagnosis and treatment at the time of Petitioner’s 1997 capital murder trial. In sum, the “new” evidence presented by Petitioner, while admittedly containing some mitigating aspects (particularly those concerning Petitioner’s mother’s alcoholism and the likelihood Petitioner suffers from Fetal Alcohol Spectrum Disorder), also contains a plethora of information which would have assisted the prosecution in obtaining an affirmative answer to the Texas capital sentencing scheme’s future dangerousness special issue. After characterizing all of the proposed “new” evidence as “double-edged,” the district court turned to the question of whether Trevino’s trial counsel rendered ineffective assistance, setting out the same standards we will apply here: To establish ineffective assistance of counsel, a petitioner must show that counsel’s representation fell below an objective standard of reasonableness, and to establish prejudice he must show that there is a reasonable probability that, but for counsel’s unprofessional errors, the result of the proceeding would have been different. Williams v. Taylor, 529 U.S. 362, 390-91, 120 S.Ct. 1495, 146 L.Ed.2d 389 (2000) (citing Strickland v. Washington, 466 U.S. 668, 104 S.Ct. 2052, 80 L.Ed.2d 674 (1984)). A few highlights from Strickland should be noted. “Counsel also has a duty to bring to bear such skill and knowledge as will render the trial a reliable adversarial testing process.” Strickland, 466 U.S. at 688, 104 S.Ct. 2052. “Prevailing norms of practice as reflected in American Bar Association standards” are mere guides. Id. The district court concluded that Trevino had failed to satisfy either the performance or prejudice prong of Strickland. On the performance prong, the district court essentially concluded that the evidence simply was not available to his trial counsel at the time of his trial and therefore counsel’s failure to investigate it could not constitute deficient performance. The district court stated that “trial counsel was not wholly inattentive to developing mitigating evidence,” in that he interviewed Trevino’s stepfather, and Trevino “failed to assist his trial counsel in identifying any family members or others who may have provided mitigating testimony.” With respect to the evidence of Trevino’s mother’s alcohol abuse during pregnancy, the court noted that Trevino’s mother did not provide a sworn statement until 2004 and did not state that she would have been available to testify at the 1997 trial. “Accordingly, it is difficult to understand how trial counsel could reasonably be blamed for not locating Ms. Trevino prior to Petitioner’s 1997 capital murder trial and presenting potentially mitigating evidence from this witness.” The court also found that the FASD claim should be denied primarily on the performance prong because Trevino failed to show that the evidence was available at the time of trial. In sum, the district court found that there was no evidence that Trevino’s state trial counsel knew or should have known about additional character witnesses or about the factual basis for a possible FASD claim. On the second prong of Strickland, the district court concluded that, even if Trevino could show that his trial counsel’s performance fell below an objectively reasonable standard, that failure did not result in prejudice. Again, the district court set out the correct legal standards, which are also applicable to this COA application, but the district court focused primarily on the character witness testimony: In evaluating prejudice in the context of the punishment phase of a capital trial, a federal habeas court must re-weigh all the evidence in aggravation against the totality of available mitigating evidence (had the Petitioner’s trial counsel chosen a different course). Wong v. Belmontes, 558 U.S. 15, 20, 130 S.Ct. 383, 175 L.Ed.2d 328 (2009); Wiggins v. Smith, 539 U.S. 510, 534, 123 S.Ct. 2527, 156 L.Ed.2d 471 (2003). Strickland does not require the State to “rule out” or negate a sentence of life in prison to prevail; rather, it places the burden on the defendant to show a “reasonable probability” that the result of the punishment phase of a capital murder trial would have been different. Wong v. Belmontes, 558 U.S. at 27, 130 S.Ct. 383. The prejudice inquiry under Strickland requires evaluating whether there is a “reasonable probability that, but for counsel’s unprofessional errors, the result of the proceeding would have been different.” Strickland, 466 U.S. at 694, 104 S.Ct. 2052. “The likelihood of a different result must be substantial, not just conceivable.” Brown v. Thaler, 684 F.3d 482, 491 (5th Cir. 2012) (citing Harrington v. Richter, 562 U.S. 86, 131 S.Ct. 770, 178 L.Ed.2d 624 (2011)), cert. denied, — U.S. -, 133 S. Ct. 1244, 185 L.Ed. 2d 190 (2013). Federal habeas corpus petitioners asserting claims of ineffective assistance based on counsel’s failure to call a witness satisfy the prejudice prong of the Strickland analysis only by naming the witness, demonstrating the witness was available to testify and would have done so, setting out the content of the witness’ proposed testimony, and showing the testimony would have been favorable to a particular defense. Woodfox v. Cain, 609 F.3d 774, 808 (5th Cir. 2010); Day v. Quarterman, 566 F.3d 527, 538 (5th Cir. 2009). The district court found it sufficient that Trevino’s trial counsel presented the testimony of his aunt, “albeit in a cursory fashion,” to explain “the facts that Petitioner’s mother was an alcoholic and Petitioner’s family lived on welfare in public housing.” The court concluded that the proposed “new” character witness testimony could not have changed the result because, “in addition to noting that Petitioner was raised in a very troubled household and neighborhood, and that he was kind and caring at times, these individuals also have described Petitioner as a man quickly prone to angry and violent outbursts.” In the district court’s view, the character witness testimony is only weakly mitigating and contains highly aggravating evidence, so there is no reason to believe it would serve any meaningful mitigation purpose. In- summing up its conclusion that the “new” evidence could not have changed the outcome (i.e., the failure to introduce it was not prejudicial under Strickland), the district court focused on the heinous nature of the crime, the fact that there was a great deal of aggravating evidence, and the fact that the proposed character witness testimony contained additional aggravating evidence in addition to fairly inconsequential mitigating evidence. In short, it concluded that the “new” mitigating evidence simply could not outweigh the aggravating evidence because it was “double-edged,” placing greater emphasis on the character witness testimony. Because the district court concluded that Trevino failed to satisfy either prong of Strickland, it denied relief on the merits. 3. Denial of Evidentiary Hearing The district court did not hold an evi-dentiary hearing because it found Trevino “has failed to allege specific facts which, if proven, would entitle Petitioner to federal habeas corpus relief in this cause.” Because it based its decision on the pleadings, an evidentiary hearing could not affect the outcome. 4. Denial Of COA Based on all the above, the district court denied all relief under the second amended petition. It also denied a COA. Although it noted that “[i]n death penalty cases, any doubt as to whether a CoA should issue must be resolved in the petitioner’s favor,” it concluded there was no such doubt here, at least with respect to the Martinez/Trevino procedural default issue (i.e., whether Trevino sufficiently alleged that his state habeas counsel rendered ineffective assistance for failing to raise the ineffective-assistanee-of-trial-counsel claim) and the prejudice prong of Strickland. II. APPLICABLE Law Before turning to the particular claims asserted by Trevino, we first address the applicable law; the framework for Strickland claims generally and the Wiggins inadequate investigation claim at issue here; and the law concerning “double-edged” evidence. A. Jurisdiction and Standard of Review We have jurisdiction in this application for a COA from the district court’s denial of habeas relief under 28 U.S.C. § 2254 pursuant to 28 U.S.C. §§ 1291 and 2253(c)(1)(B). As we set out in our 2011 opinion, the standard of review under AEDPA is usually highly deferential. AEDPA deference does not apply here, however, because the district court was not reviewing a state court decision on the merits of Trevino’s claim but rather addressing the merits for the first time. Thus, AEDPA’s deferential standard of review does not apply, and we review the merits de novo. Under § 2254(b)(2), “An application for a writ of habeas corpus may be denied on the merits, notwithstanding the failure of the applicant to exhaust the remedies available in the courts of the State.” Because Trevino’s claim may be dismissed on either procedural default grounds or on the merits, he must demonstrate that reasonable jurists would debate the correctness of the district court’s dismissal on both grounds. B. Strickland and Wiggins Strickland analysis is, of course, central to this COA application, especially as applied in Wiggins v. Smith, 539 U.S. 510, 523, 123 S.Ct. 2527, 156 L.Ed.2d 471 (2003). Because Wiggins demonstrates the basis for Trevino’s entire claim, we examine the case in some detail. The petitioner, Wiggins, was convicted in August 1989, after a four-day jury trial, for a murder he committed in 1988. Prior to sentencing, his trial counsel moved for bifurcation of sentencing into two phases: in the first, counsel proposed to prove that Wiggins did not act as the principal in the murder, and in the second, they intended to present mitigating evidence. Counsel argued that bifurcation would prevent mitigation evidence from undercutting their argument that Wiggins was not primarily responsible for the murder. The trial judge denied the motion, and sentencing commenced in a single phase. On October 12, the court denied the bifurcation motion, and sentencing proceedings commenced immediately thereafter. In her opening statement, Nether-cott told the jurors they would hear evidence suggesting that someone other than Wiggins actually killed Lacs. Counsel then explained that the judge would instruct them to weigh Wiggins’ clean record as a factor against a death sentence. She concluded: “ ‘You’re going to hear that Kevin Wiggins has had a difficult life. It has not been easy for him. But he’s worked. He’s tried to be a productive citizen, and he’s reached the age of 27 with no convictions for prior crimes of violence and no convictions, period. ... I think that’s an important thing for you to consider.’ ” During the proceedings themselves, however, counsel introduced no evidence of Wiggins’ life history. Before closing arguments, Schlaich made a proffer to the court, outside the presence of the jury, to preserve bifurcation as an issue for appeal. He detailed the mitigation case counsel would have presented had the court granted their bifurcation motion. He explained that they would have introduced psychological reports and expert testimony demonstrating Wiggins’ limited intellectual capacities and childlike emotional state on the one hand, and the absence of aggressive patterns in his behavior, his capacity for empathy, and his desire to function in the world on the other. At no point did Schlaich proffer any evidence of petitioner’s life history or family background. The jury returned a sentence of death. In 1998, Wiggins sought state habeas relief, “arguing that his attorneys had rendered constitutionally defective assistance by failing to investigate and present mitigating evidence of his dysfunctional background.” In support, he submitted testimony by a licensed social worker who had prepared an extensive social history report detailing severe physical and sexual abuse by his own father and mother as well as various foster parents. During these proceedings, one of Wiggins’ trial attorneys testified that he did not recall retaining a forensic social worker to prepare a social history, even though the State'of Maryland made funds available for that purpose, and he testified that the trial team had, “well in advance of trial, decided to focus their efforts on ‘retrying] the factual case’ and disputing Wiggins’ direct responsibility for the murder.” The state habeas courts denied relief on the ground that the decision not to investigate was “a matter of trial tactics” and therefore did not constitute deficient performance under Strickland. The state appellate court focused on the fact that trial counsel knew at least the general contours of Wiggins’ childhood, and that at least one mitigating factor, Wiggins’ lack of prior convictions, was presented to the jury. Wiggins filed a habeas petition in federal court, arguing that the state habeas courts’ rejection of his ineffective-assistance-of-trial-counsel claim was based on an unreasonable application of clearly established federal law. The federal district court agreed, concluding that trial counsel’s decision not to investigate Wiggins’ social history further could only be reasonable if it was “based upon information the attorney has made after conducting a reasonable investigation.” Reviewing de novo, the Fourth Circuit reversed. The Supreme Court granted certiorari and reversed the Fourth Circuit. After setting out the Strickland standards and emphasizing the “heavy measure of deference” accorded to the judgments of trial counsel, the Supreme Court explained the limits of that deference: Our opinion in Williams v. Taylor is illustrative of the proper application of these standards. In finding Williams’ ineffectiveness claim meritorious, we applied Strickland and concluded that counsel’s failure to uncover and present voluminous mitigating evidence at sentencing could not be justified as a tactical decision to focus on Williams’ voluntary confessions, because counsel had not “fulfilled] their obligation to conduct a thorough investigation of the defendant’s background.” 529 U.S. at 396, 120 S.Ct. 1495 (citing 1 ABA Standards for Criminal Justice 4-4.1, commentary, p. 4-55 (2d ed.1980)). While Williams had not yet been decided at the time the Maryland Court of Appeals rendered the decision at issue in this case, cf. post, at 2546 (SCALIA, J., dissenting), Williams’ case was before us on habeas review. Contrary to the dissent’s contention, ibid., we therefore made no new law in resolving Williams’ ineffectiveness claim. See Williams, 529 U.S. at 390, 120 S.Ct. 1495 (noting that the merits of Williams’ claim “are squarely governed by our holding in Strickland”); see also id., at 395, 120 S.Ct. 1495 (noting that the trial court correctly applied both components of the Strickland standard to petitioner’s claim and proceeding to discuss counsel’s failure to investigate as a violation of Strickland’s performance prong). In highlighting counsel’s duty to investigate, and in referring to the ABA Standards for Criminal Justice as guides, we applied the same “clearly established” precedent of Strickland we apply today. Cf. 466 U.S. at 690-691, 104 S.Ct. 2052 (establishing that “thorough investigation[s]” are “virtually unchallengeable” and underscoring that “counsel has a duty to make reasonable investigations”); see also id., at 688-689, 104 S.Ct. 2052 (“Prevailing norms of practice as reflected in American Bar Association standards and the like ... are guides to determining what is reasonable”). In light of these standards, our principal concern in deciding whether Schlaich and Nethercott exercised “reasonable professional judgmen[t],” id., at 691, 104 S.Ct. 2052, is not whether counsel should have presented a mitigation case. Rather, we focus on whether the investigation supporting counsel’s decision not to introduce mitigating evidence of Wiggins’ background was itself reasonable. Ibid. Cf. Williams v. Taylor, supra, at 415, 120 S.Ct. 1495 (O’CONNOR, J., concurring) (noting counsel’s duty to conduct the “requisite, diligent” investigation into his client’s background). In assessing counsel’s investigation, we must conduct an objective review of their performance, measured for “reasonableness under prevailing professional norms,” Strickland, 466 U.S. at 688, 104 S.Ct. 2052, which includes a context-dependent consideration of the challenged conduct as seen “from counsel’s perspective at the time,” id., at 689, 104 S.Ct. 2052 (“[E]very effort [must] be made to eliminate the distorting effects of hindsight”). The Court noted that trial counsel drew its mitigation case from three sources: an IQ test conducted by a psychologist, which revealed Wiggins had an IQ of 79; a pre-sentence investigation report (“PSI”), which included a brief summary of his miserable personal history; and records kept by the Baltimore City Department of Social Services (“DSS”), which showed his various placements in the foster care system. They did not, however, develop any further social history, despite the availability of funds for that purpose. The Court held that this constituted constitutionally deficient investigation in light of not only Maryland’s standards but ABA Guidelines in place prior to his sentencing, including the admonition that “investigations into mitigating evidence ‘should comprise efforts to discover all reasonably available mitigating evidence and evidence to rebut any aggravating evidence that may be introduced by the prosecutor.’ ” Despite these well-defined norms, however, counsel abandoned their investigation of petitioner’s background after having acquired only rudimentary knowledge of his history from a narrow set of sources. Cf. id., 11.8.6, p. 133 (noting that among the topics counsel should consider presenting are medical history, educational history, employment and training history, family and social history, prior adult and juvenile correctional experience, and religious and cultural influences (emphasis added)); 1 ABA Standards for Criminal Justice 4-4.1,. commentary, p. 4-55 (2d ed. 1982) (“The lawyer also has a substantial and important role to perform in raising mitigating factors both to the prosecutor initially and to the court at sentencing .... Investigation is essential to fulfillment of these functions”). Moreover, the Court found that “the investigation was also unreasonable in light of what counsel actually discovered in the DSS records,” including the fact that Wiggins’ mother was an alcoholic, that he had spent time in different foster homes, that he displayed emotional difficulties, that he was frequently absent from school, and that he was left without food for days at a time. As the Federal District Court emphasized, any reasonably competent attorney would have realized that pursuing these leads was necessary to making an informed choice among possible defenses, particularly given the apparent absence of any aggravating factors in petitioner’s background. 164 F.Supp.2d at 559. Indeed, counsel uncovered no evidence in their investigation to suggest that a mitigation case, in its own right, would have been counterproductive, or that further investigation would have been fruitless; this case is therefore distinguishable from our precedents in which we have found limited investigations into mitigating evidence to be reasonable. See, e.g., Strickland, supra, at 699, 104 S.Ct. 2052 (concluding that counsel could “reasonably surmise ... that character and psychological evidence would be of little help”); Burger v. Kemp, 483 U.S. 776, 794, 107 S.Ct. 3114, 97 L.Ed.2d 638 (1987) (concluding counsel’s limited investigation was reasonable because he interviewed all witnesses brought to his attention, discovering little that was helpful and much that was harmful); Darden v. Wainwright, 477 U.S. 168, 186, 106 S.Ct. 2464, 91 L.Ed.2d 144 (1986) (concluding that counsel engaged in extensive preparation and that the decision to present a mitigation case would have resulted in the jury hearing evidence that petitioner had been convicted of violent crimes and spent much of his life in jail). Had counsel investigated further, they might well have discovered the sexual abuse later revealed during state postconviction proceedings. In sum, the Court concluded that Wiggins’ trial counsel’s investigation was constitutionally inadequate under the performance prong of Strickland: In finding that Schlaich and Nether-cott’s investigation did not meet Strickland’s performance standards, we emphasize that Strickland does not require counsel to investigate every conceivable line of mitigating evidence no matter how unlikely the effort would be to assist the defendant at sentencing. Nor does Strickland require defense counsel to present mitigating evidence at sentencing in every case. Both conclusions would interfere with the “constitutionally protected independence of counsel” at the heart of Strickland, 466 U.S. at 689, 104 S.Ct. 2052. We base our conclusion on the much more limited principle that “strategic choices made after less than complete investigation are reasonable” only to the extent that “reasonable professional judgments support the limitations on investigation.” Id., at 690-691, 104 S.Ct. 2052. A decision not to investigate thus “must be directly assessed for reasonableness in all the circumstances.” Id., at 691, 104 S.Ct. 2052. The Court then turned to the prejudice prong of Strickland: In order for counsel’s inadequate performance to constitute a Sixth Amendment violation, petitioner must show that counsel’s failures prejudiced his defense. Strickland, 466 U.S. at 692, 104 S.Ct. 2052. In Strickland, we made clear that, to establish prejudice, a “defendant must show that 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. In assessing prejudice, we reweigh the evidence in aggravation against the totality of available mitigating evidence. The Court concluded that the failure to investigate and discover the “powerful” mitigation evidence was indeed prejudicial, in that it showed a history of severe abuse, starting with his “alcoholic, absentee mother” and continuing through an unbroken series of extreme hardships. In the Court’s words, Wiggins “thus has the kind of troubled history we have declared relevant to assessing a defendant’s moral culpability.” Given both the nature and the extent of the abuse petitioner suffered, we find there to be a reasonable probability that a competent attorney, aware of this history, would have introduced it at sentencing in an admissible form. While it may well have been strategically defensible upon a reasonably thorough investigation to focus on Wiggins’ direct responsibility for the murder, the two sentencing strategies are not necessarily mutually exclusive. Moreover, given the strength of the available evidence, a reasonable attorney might well have chosen to prioritize the mitigation case over the direct responsibility challenge, particularly given that Wiggins’ history contained little of the double edge we have found to justify limited investigations in other cases. Burger v. Kemp, 488 U.S. 776, 107 S.Ct. 3114, 97 L.Ed.2d 638 (1987); Darden v. Wainwright, 477 U.S. 168, 106 S.Ct. 2464, 91 L.Ed.2d 144 (1986). Accordingly, the Supreme Court held that the mitigating evidence, considered as a whole, could have resulted in a different sentence, and it reversed and remanded. C. “Double-Edged” Evidence The district court focused on the fact that Wiggins does not necessarily apply when the proposed “new” evidence is “double-edged,” as Wiggins itself explained. Therefore, the two cases the Supreme Court cited in Wiggins for such evidence, Burger and Darden, are worth examining briefly. In Burger, the petitioner’s trial counsel was aware of some, but not all, of his troubled family history, including his “unhappy and unstable childhood,” one of his stepfathers getting him involved in marijuana and alcohol, his running away from home and being placed in a juvenile detention home, and similar facts. During his investigation, trial counsel had talked to the petitioner’s mother, an old friend of the petitioner’s, a psychologist counsel had employed to examine him prior to trial, and others, before deciding not to present evidence of his childhood. Counsel also decided not to have the petitioner testify on the ground that he showed no remorse and might actually brag about the crime on the witness stand. The petitioner argued that his attorney should have conducted more of an investigation, but the Court concluded that the proposed “new” testimony could not have helped. The proposed testimony contained only meager mitigation evidence and a substantial amount of aggravating evidence, including the fact that he had spent time in juvenile detention, which had not been disclosed at trial, and that he had violent tendencies and seemed to have a split personality that resulted in unpredictable angry outbursts. As the Court noted, “Even apart from their references to damaging facts, the papers are by no means uniformly helpful to petitioner because they suggest violent tendencies that are at odds with the defense’s strategy of portraying petitioner’s actions on the night of the murder as the result of Stevens’ strong influence upon his will.” In short, the petitioner’s trial counsel in Burger had conducted a fairly extensive investigation into mitigation evidence and had made considered judgments in choosing not to present some seemingly mitigating evidence. The evidence trial counsel failed to discover through his investigation contained a great deal of aggravating evidence and therefore its absence could not have prejudiced him. In Darden, the petitioner argued that he had received ineffective assistance of trial counsel on the ground that his attorney spent insufficient time preparing the mitigation case and had opted to “rely on a simple plea for mercy from petitioner himself.” The Court found that his trial counsel had spent hundreds of hours preparing his case, including mitigation. The problem was that there simply was no mitigating evidence that would not have permitted the state to bring in even stronger aggravating evidence to rebut it. Any argument that he was nonviolent would have allowed the state to bring in evidence of his prior convictions, which had not previously been admitted in evidence, and any argument that he was incapable of committing the crimes would have allowed the state to introduce a psychiatric report indicating he very well could have based on his “sociopathic type of personality,” among other damaging rebuttal evidence. Accordingly, the Court concluded that the trial counsel’s decision to rely on a simple plea of mercy, following the investigation and consideration of potentially mitigating evidence, constituted a defensible trial strategy under Strickland, III. Analysis A. MartineziTrevino Issue Trevino argues (1) that he is entitled to an evidentiary hearing on the Martinez/Trevino issue, and (2) that he properly established cause for his procedural default under the Supreme Court’s Martinez/Trevino rule. Based on his pleadings alone, we conclude he has at least alleged sufficient cause, so we need not address his evidentiary hearing argument. As noted above, the Supreme Court in Trevino stated, “[A] procedural default will not bar a federal habeas court from hearing a substantial claim of ineffective assistance at trial if, in the initial-review collateral proceeding, there was no counsel or counsel in that proceeding was ineffective.” The district court, citing standards for appellate counsel, found that Trevino failed to show that his state habeas counsel was ineffective because he failed to show that the proposed “new” evidence was even available at the time. Martinez suggests that a similar standard should apply to both state trial counsel and state habeas counsel. There, the Supreme Court explained that the purpose of the exception is to recognize that the initial state habeas proceeding is virtually the same as a direct appeal for some purposes: Where, as here, the initial-review collateral proceeding is the first designated proceeding for a prisoner to raise a claim of ineffective assistance at trial, the collateral proceeding is in many ways the equivalent of a prisoner’s direct appeal as to the ineffective-assistance claim. This is because the state habeas court “looks to the merits of the clai[m]” of ineffective assistance, no other court has addressed the claim, and “defendants pursuing first-tier review ... are generally ill equipped to represent themselves” because they do not have a brief from counsel or an opinion of the court addressing their claim of error. Here, the crux of Trevino’s claim is not that his trial counsel made an informed decision not to present certain evidence following a constitutionally sufficient investigation, but that his trial- counsel failed to conduct such an investigation in the first place. Trevino argues that the state trial counsel’s failure to investigate would have been obvious to his state habeas counsel as well. Thus, he argues that his state habeas counsel’s failure to investigate the possibility of a Wiggins claim constitutes ineffective assistance, satisfying Martinez/Trevino. Trevino’s second amended petition’s section titled “Petitioner’s State Habeas Counsel was Ineffective” is mostly devoted to the many failings of his state trial counsel, but it also squarely addresses his state habeas counsel’s alleged ineffective assistance: Failing to raise such a claim after investigation, and making a thoroughly informed decision that there was no merit in raising that issue for review was an option to Attorney Rodriguez. Never investigating the possibility or merits of such a claim was not. AS [sic] thoroughly demonstrated in the foregoing sections of this petition, there was an immense amount of material not included in the record indicating that trial counsel had indeed been ineffective at the punishment phase of trial. State habeas counsel had a duty and obligation to undertake an investigation to at least determine whether such a claim was a viable one. Had such an investigation been undertaken, the magnitude of the error would have become evident. At that time, there was simply no scenario in which state habeas counsel’s actions and performance could be considered effective representation of any client — especially one sentenced to death who was relying on state habeas counsel for his one and only possible opportunity in existence at that time. State habeas counsel was undoubtedly ineffective in his failure to raise a claim that Petitioner’s trial counsel was ineffective at the punishment phase of Petitioner’s trial. Trevino essentially argues that the facially deficient investigation by the state trial counsel should have put his state ha-beas counsel on notice to investigate a claim for failure to investigate. The district court’s approach, on the other hand, suggests that Trevino’s state habeas counsel could not have rendered ineffective assistance for failing to assert a claim based on his trial counsel’s failure to investigate because there was no record evidence of what mitigating evidence his trial counsel failed to discover. We conclude Trevino has the better argument here. If state habeas counsel is not subject to the same Strickland requirement to perform some minimum investigation prior to bringing the initial state habe-as petition, the Martinez/Trevino rule would have limited utility (if any) in addressing Wiggins claims. There is a serious danger, under the district court’s reasoning, that a state trial counsel’s failure to investigate (and put into the record) mitigation evidence could insulate state ha-beas counsel from an ineffective assistance claim simply because the evidence was missing. That would only compound the problem with state trial counsel’s failure to conduct a reasonable investigation in the first place, and Wiggins claims for deficient investigation might be effectively un-reviewable under Martinez/Trevino. In this ease, Trevino’s state trial counsel presented only one mitigation witness and no other evidence during the punishment phase. The deficiency in that investigation would have been evident to any reasonably competent habeas attorney. Thus, we conclude that reasonable jurists not only could debate the correctness of the district court’s conclusion on the Martinez/Trevino issue, but would agree that the district court reached the wrong conclusion. Trevino at least sufficiently pleaded that his state habeas counsel was ineffective so as to excuse his procedural default in failing to raise the ineffective-assistance-of-trial-counsel failure-to-investigate claim earlier. B. Wiggins Claim — Strickland Performance Prong Turning to the merits of Trevino’s ineffective-assistance-of-trial-counsel claim under Wiggins, we must determine whether Trevino satisfied both prongs of Strickland. First, we must determine whether Trevino’s trial counsel’s performance was deficient. To demonstrate deficient performance, “the defendant must show that counsel’s representation fell below an objective standard of reasonableness” as measured by “prevailing professional norms.” Our scrutiny of counsel’s performance is highly deferential. We “must indulge a strong presumption that counsel’s conduct falls within the wide range of reasonable professional assistance; that is, the defendant must overcome the presumption that, under the circumstances, the challenged action ‘might be considered sound trial strategy.’ ” To overcome this presumption, “[a] convicted defendant making a claim of ineffective assistance must identify the acts or omissions of counsel that are alleged not to have been the result of reasonable professional judgment.” Of central importance here, “choices made after less than complete investigation are reasonable [only] to the extent that reasonable professional judgments support the limitations on investigation.” Factors affecting whether it is reasonable not to investigate include whether counsel has “reason to believe that pursuing certain investigations would be fruitless or even harmful,” resource constraints, and whether the information that might be discovered would be of only collateral significance. As set out above, the district court held that Trevino had failed to allege facts showing that the performance of his trial counsel was deficient and instead concluded that Trevino’s “trial counsel was not wholly inattentive to developing mitigating evidence,” in that he interviewed Trevino’s stepfather, and Trevino “failed to assist his trial counsel in identifying any family members or others who may have provided mitigating testimony.” Moreover, the court emphasized that because Trevino’s mother drank heavily during the time of trial in 1997, Trevino’s trial counsel could not be blamed for failing to locate her or discover evidence pertaining to FASD. Reasonable jurists could debate the correctness of the district court’s determination that Trevino failed to plead that his trial counsel conducted a constitutionally deficient investigation into mitigation evidence. The record shows that Trevino’s trial counsel only put forward one mitigation witness, Trevino’s aunt, and that he interviewed her briefly only on the day of her testimony. As Trevino argued in his COA application: The relevant legal question is not whether counsel were “wholly inattentive” to developing mitigation evidence. Nor is it whether counsel’s client meaningfully assisted in the mitigation investigation. Nor is it whether one particular witness was easily locatable. Nor is it whether counsel successfully managed to investigate so little so as to remain completely ignorant about significant aspects of their client’s background. It is significant to note here that the one witness the trial counsel did present, Appellant’s aunt Juanita DeLeon, testified that Appellant’s mother could not be present to testify because she “had alcohol problems” and lived “in Elgin [Texas].” Clearly, Ms. DeLeon had current knowledge of where Appellant’s mother was living, and of her current state of health. Had counsel simply asked that question of Ms. DeLeon during the trial preparation phase, instead of when she was on the stand, and followed up with a diligent investigation, significant mitigation evidence could have, and would have, been uncovered. It is also significant that, in the state habeas hearing, trial counsel testified that he knew Appellant’s mother had been in court — or at least in the courthouse — -at some time before the a