Nelson v. Davis

Before JONES, SMITH, and DENNIS, Circuit Judges.^[1] JAMES L. DENNIS, Circuit Judge: Steven Nelson seeks a Certificate of Appealability (COA) to challenge his 2012 Texas capital conviction, alleging multiple claims of ineffective assistance of trial counsel as well as unconstitutional juror strikes under Batson v. Kentucky, 476 U.S. 79 (1986). Nelson also appeals the district court’s denial of his motions for investigative funding under 18 U.S.C. § 3599(f) and for stay and abatement of his federal proceedings pending exhaustion of claims in state court. As discussed below, a COA is hereby GRANTED in part and DENIED in part. We AFFIRM in part the district court’s denial of Nelson’s other motions and defer adjudication in part until our full consideration of the merits of Nelson’s appeal. I. Background In 2012, Steven Nelson was convicted of the capital murder of Clinton Dobson on March 3, 2011, in Arlington, Texas. Nelson v. State, No. AP-76,924, 2015 WL 1757144 at *1 (Tex. Crim. App. Apr. 15, 2015). Dobson, a pastor, had been violently assaulted and then suffocated with a plastic bag, and his secretary, Judy Elliot, was badly beaten and almost did not survive. Id. at *1- *2. A laptop, cellphone, car, and credit cards were stolen from the victims. Id. Nelson was arrested and indicted after information from his acquaintances, forensic evidence from the scene, and surveillance video of him with the victims’ possessions linked him to the crime. Nelson confessed that he had agreed to participate in the robbery, but denied assaulting Elliot or murdering Dobson. Id. at *3. A jury convicted Nelson after receiving a law-of-the-parties instruction to return a guilty verdict if it found either that Nelson had murdered Dobson or that Nelson had joined a conspiracy to commit the robbery and should have anticipated the murder of another in furtherance of that robbery. At the punishment phase, the State provided substantial evidence of Nelson’s past violence and criminal history, which the Court of Criminal Appeals of Texas (“TCCA”) summarized in detail in its opinion on direct appeal. See Nelson, 2015 WL 1757144, at *4-7. Relevant here, punishment phase evidence included evidence that, while awaiting trial, Nelson “killed Jonathon Holden, a mentally challenged inmate.” Id. at *6. “According to a fellow inmate who witnessed the incident, Holden had angered inmates when he mentioned ‘the N word under his voice.’” Id. After Nelson “talked Holden into faking a suicide attempt to cause Holden to be moved to a different part of the jail. . . . Holden came to the cell bars, and [Nelson] looped a blanket around Holden’s neck.” Id. Nelson strangled Holden, and after his death, “did a ‘celebration dance’ in the style of Chuck Berry,” using “a broom stick, which he had previously used to poke another mentally challenged inmate in the eye, as a guitar.” Id. The defense at the punishment phase presented mitigation testimony from Nelson’s family, a social worker who counseled him when he was a child, and Dr. Antoinette McGarrahan, a forensic psychologist hired as an expert witness to evaluate Nelson. Id. at *7. The state court summarized Dr. McGarrahan’s mitigation testimony as follows: [Dr. McGarrahan] testified that, although appellant had no current learning disability or cognitive impairment, he had a past history of learning disabilities. Dr. McGarrahan explained that, when, as a three-year-old, appellant set fire to his mother’s bed with intent to cause harm, it was essentially a cry for attention and security. She believed that there was “something significantly wrong with [appellant's] brain being wired in a different way, being predisposed to this severe aggressive [sic] and violence from a very early age.” She testified that, by the time appellant was six years old, he had had at least three EEGs, meaning that people were already “looking to the brain for an explanation” of his behavior. The test results did not indicate a seizure disorder, but Dr. McGarrahan said that they did not rule out appellant having one. Risk factors present in appellant’s life included having ADHD, a mother who worked two jobs, an absent father, verbal abuse, and witnessing domestic violence. Id. After answering Texas’s three special questions required at the capital punishment phase, the jury sentenced Nelson to death. See TEX. CODE CRIM. Pro. Art. 37.071. In his direct appeal, Nelson argued, as relevant here, that the State unconstitutionally used its peremptory strikes to eliminate as jurors racial minorities. Nelson, 2015 WL 1757144 at *10. The TCCA denied relief. Id. at *15. Nelson then filed a state habeas application alleging, among other claims, ineffective assistance of trial counsel for failure to adequately investigate and present mitigating evidence from “other family members, friends, and former teachers” at the punishment phase of trial. The state court denied Nelson’s claims, adopting the State’s proposed findings of facts and conclusions of law without alteration. The TCCA affirmed without further reasoning. Ex Parte Steven Lewayne Nelson, No. WR-82,814-01, 2015 WL 6689512, at *1 (Tex. Crim. App. Oct 12, 2015). With the assistance of different counsel, Nelson then filed the instant federal habeas action in the district court, asserting five grounds with multiple subparts. The district court denied relief on all claims on the merits and some on the alternative grounds that they were procedurally barred, and then denied a COA. Nelson now seeks a COA on his claims that 1) his trial counsel was ineffective in failing to adequately investigate and present three different categories of mitigating evidence, 2) the State used race to select the jury in violation of Batson, and 3) his trial counsel was ineffective for failing to adequately litigate his Batson claim during voir dire. Additionally, Nelson directly appeals the district court’s denial of his three motions seeking funding for investigative services claim under 18 U.S.C. § 3599(f). Nelson also appeals the district court’s denial of his motion for stay and abatement to permit him to exhaust in state court his claims of ineffective assistance of counsel and an additional claim that the State knowingly presented false testimony at the punishment phase. II. Standard of Review To appeal the district court’s denial of his habeas claims, Nelson must first seek a COA from this court pursuant to 28 U.S.C. § 2253(c)(1). See Miller- El v. Cockrell, 537 U.S. 322, 335-36 (2003). To obtain a COA, Nelson must demonstrate “a substantial showing of the denial of a constitutional right.” 28 U.S.C. § 2253(c)(2). For a claim that the district court decided on the merits, he must show that “jurists of reason could disagree with the district court’s resolution of his constitutional claims or that jurists could conclude the issues presented are adequate to deserve encouragement to proceed further.” Miller- El, 537 U.S. at 327 (citing Slack v. McDaniel, 529 U.S. 473, 484 (2000)); see 28 U.S.C. § 2253(c)(2). For claims denied on procedural grounds, Nelson must show that “jurists of reason would find it debatable whether the petition states a valid claim of the denial of a constitutional right and that jurists of reason would find it debatable whether the district court was correct in its procedural ruling.” Segundo v. Davis, 831 F.3d 345, 350 (5th Cir. 2016) (quoting Slack, 529 U.S. at 484). The COA standard is less burdensome in capital cases, as “in a death penalty case any doubts as to whether a COA should issue must be resolved in the petitioner’s favor.” Clark v. Thaler, 673 F.3d 410, 425 (5th Cir. 2012). When a state court has reviewed a petitioner’s claim on the merits, our review is constrained by the deferential standards of review found in the Antiterrorism and Effective Death Penalty Act (“AEDPA”). See 28 U.S.C. § 2254. Under these circumstances, we may not issue a COA unless reasonable jurists could debate that the state court’s decision was either “contrary to, or involved an unreasonable application of, clearly established Federal law,” § 2254(d)(1), or “was based on an unreasonable determination of the facts in light of the evidence presented in the State court proceeding,” § 2254(d)(2). “For claims that are not adjudicated on the merits in the state court, however, we do not apply the deferential scheme laid out in § 2254(d) and instead apply a de novo standard of review.” Ward v. Stephens, 777 F.3d 250, 256 (5th Cir. 2015), abrogated on other grounds by Ayestas v. Davis, 138 S. Ct. 1080 (2018) (citations and internal quotation marks omitted). A petitioner does not require a COA to appeal the district court’s denial of funding under § 3559(f) or denial of petitioner’s motion to stay proceedings. See Ayestas v. Stephens, 817 F.3d 888, 895 (5th Cir. 2016), vacated on other grounds by Ayestas v. Davis, 138 S. Ct. 1080 (2018) (COA not required to appeal denial of funding under § 3599(f)); Williams v. Thaler, 602 F.3d 291, 309 (5th Cir. 2010) (COA not required to appeal denial of a motion for stay and abatement). We review the district court’s denial of these motion for abuse of discretion. Ayestas v. Davis (Ayestas II), 933 F.3d 384, 388 (5th Cir. 2019) (§ 3599(f)); Williams, 602 F.3d at 309 (stay and abatement). III. Ineffective Assistance of Trial Counsel at the Punishment Phase Nelson seeks a COA on claims that his trial counsel was ineffective at the punishment phase of his trial in failing to investigate and develop three different kinds of potential mitigating evidence. To prevail on a claim of ineffective assistance of counsel, Nelson must demonstrate both deficient performance and prejudice. Strickland v. Washington, 466 U.S. 668, 687 (1984). Deficient performance is only that which “fell below an objective standard of reasonableness.” Id. at 688. Courts “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 have been considered sound trial strategy.” Id. (internal quotation marks omitted). However, this “does not eliminate counsel’s duty to ‘make reasonable investigations or to make a reasonable decision that makes particular investigations unnecessary.” Escamilla v. Stephens, 749 F.3d 380, 388 (5th Cir. 2014) (citing Strickland, 466 U.S. at 690-91). “[S]trategic choices made after less than complete investigation are reasonable precisely to the extent that reasonable professional judgments support the limitations on investigation.” Wiggins v. Smith, 539 U.S. 510, 521 (2003) (citing Strickland, 466 U.S. at 690-91). To satisfy the prejudice prong of an ineffective assistance of counsel claim, a petitioner “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.” Strickland, 466 U.S. at 694. Applying this two-prong inquiry, “the Supreme Court has found that trial counsel’s failure to adequately investigate available mitigating evidence . . . amounts to ineffective assistance of counsel.” Escamilla, 749 F.3d at 388 (citing Williams v. Taylor, 529 U.S. 362, 395 (2000)). Moreover, “we have explained that, ‘in investigating potential mitigating evidence, counsel must either (1) undertake a reasonable investigation or (2) make an informed strategic decision that investigation is unnecessary.’” Id. at 390 (quoting Charles v. Stephens, 736 F.3d 380, 389 (5th Cir.2013)). “‘[T]rial counsel must not ignore pertinent venues of investigation, or even a single, particularly promising investigation lead.’” Id. Where “the scope and adequacy of counsel’s mitigation investigation was debatably unreasonable,” we have granted a COA. Id. at 391 (citing Smith v. Dretke, 422 F.3d 269, 280 (5th Cir.2005)). A. Failure to Investigate and Present Mental Health History Nelson first alleges that his trial counsel was ineffective during the punishment phase for failing to adequately investigate and present his history of childhood trauma and its impact on his mental health. We refer to this claim as Nelson’s “IATC-Mental Health” claim. Nelson principally objects to counsels’ decision to select Dr. Antoinette McGarrahan, a neuropsychologist, to evaluate Nelson and testify as an expert witness at the punishment phase. He contends that, despite information in counsel’s possession indicating that Nelson was affected by severe trauma, counsel did not properly investigate these leads by retaining a trauma specialist or specifically instructing Dr. McGarrahan to consider whether he suffered from post-traumatic stress disorder (“PTSD”). Nelson emphasizes that counsel called Dr. McGarrahan to testify at the punishment phase even though she informed counsel before trial that “[i]f asked on cross, [she] [would] most likely agree that he has several traits associated with psychopathy,” and that, on cross-examination, she in fact conceded that Nelson “has many, many psychopathic characteristics.” Nelson argues that, had counsel properly investigated his abusive past and his resulting mental health problems, they would have secured an expert who would attribute his destructive behavior to severe PTSD, a potentially treatable condition for which he bears no fault, instead of psychopathy. In his petition, Nelson references his psychological evaluations from his pre-trial facility, which indicated that Nelson’s PTSD symptoms were nearly twice as severe as the average among its inmates—a group already comprised of people who have on average experienced more trauma than the general population. Nelson argues that, despite counsel’s awareness of these records and other “red flags” indicating severe trauma, they failed to properly investigate these leads. In support of his argument in the district court that such investigation would have revealed material mitigating evidence that trial counsel missed, federal habeas counsel hired Dr. Bekh Bradley, a clinical psychologist, to evaluate Nelson and conduct an initial inquiry into his background. Dr. Bradley’s report concluded that Nelson “suffered extreme childhood trauma and adversity, which has likely resulted in unrecognized and untreated trauma-related symptoms,” and that “a failure to take into account the influence of early trauma/adversity and PTSD is likely to have led to an inappropriate assessment of [Nelson] as having antisocial personality disorder.” Consistent with Dr. Bradley’s recommendations, Nelson also sought additional funding under 18 U.S.C. § 3599(f) for additional experts to further evaluate the impact on Nelson of 1) childhood and adolescent trauma and 2) “life-long incarceration.” 1. Procedural Hurdles The district court found, and the State argues before this court, that Nelson raised his IATC-Mental Health in his state habeas petition and that it was adjudicated on the merits. Nelson argues that his IATC-Mental Health claim is unexhausted, and that he can demonstrate cause for the resulting procedural default under Martinez v. Ryan, 566 U.S. 1, 9 (2012) and Trevino v. Thaler, 569 U.S. 413 (2013). However, we need not and thus do not resolve whether this claim is exhausted or unexhausted, because we conclude that reasonable jurists could not debate Nelson’s entitlement to relief on his IATC- Mental Health claim in either circumstance. i. If Exhausted If, as the district court found, this claim is the same as the ineffective assistance of trial counsel at sentencing claim that Nelson raised on state habeas, considerable deference is owed to the state court’s denial of the claim. We could only grant a COA if reasonable jurists would debate whether the state court’s decision “involved an unreasonable application of[] clearly established Federal law” or “an unreasonable determination of the facts in light of the evidence presented in the State court proceeding.” 28 U.S.C. § 2254(d); see Ward, 777 F.3d at 256. Additionally, if this claim was addressed by the state court on the merits, Nelson is barred under Cullen v. Pinholster from presenting any new evidence not before the state court to bolster this claim. 563 U.S. 170, 185 (2011) (“If a claim has been adjudicated on the merits by a state court, a federal habeas petitioner must overcome the limitation of § 2254(d)(1) on the record that was before that state court.”); see also Escamilla, 749 F.3d at 395 (5th Cir. 2014) (“[O]nce a claim is considered and denied on the merits by the state habeas court,” a petitioner’s allegation that his state habeas counsel was ineffective for failing to provide further evidence in support “may not function as an exception to Pinholsters rule that bars a federal habeas court from considering evidence not presented to the state habeas court.” (citations omitted)). In his state habeas proceedings, Nelson made the conclusory allegation that his trial attorneys “failed to investigate [his] background, history, family, and friends, and, as a result, failed to discover relevant and important mitigation evidence that would have made a difference in his punishment.” He referenced the double-edged nature of Dr. McGarrahan’s testimony, noting that she informed the jury that he “has a number of risk factors besides ADHD including a mother working two jobs, an absent father, verbal abuse, witnessing domestic violence, and minority status,” but also testified that he was “predisposed to severe aggression and violence from a very early age” and demonstrated “underlying problems with empathy and attachment.” Nelson’s state habeas argument concluded by declaring that Nelson “has many family members, friends[,] and former teachers that could have testified on his behalf during the punishment phase of trial but did not do so.” Notably, Nelson’s petition to the state court lacked any claim of severe PTSD that he now emphasizes in his federal petition. The state habeas court, based on its review of the punishment phase testimony and affidavits prepared by Nelson’s trial counsel, concluded that trial counsel “called numerous witnesses whose testimony shed light on [Nelson's] life history and allowed the jury to decide whether the choices and lifestyles of others during [his] childhood affected [him] as an adult and whether the evidence was sufficiently mitigating to avoid a death sentence.” The state court further found that Nelson “fail[ed] to identify a single undiscovered or uncalled witness . . . or to demonstrate how such witness’ testimony would have benefited him.” The state court noted that trial counsel made diligent efforts to contact and speak with potential mitigation witnesses, including “visit[ing] Oklahoma several times in order to speak with and locate witnesses” and “personally beg[ging] [Nelson's] mother to attend the trial[] to testify on [his] behalf.” If Nelson’s IATC-Mental Health claim is the same as this IATC claim that he exhausted in state court, we conclude that no reasonable jurist would debate that the state court’s denial of this claim was reasonable. See 28 U.S.C. § 2254(d); see Ward, 777 F.3d at 256. ii. If Unexhausted Recognizing the substantial limitations on our review of an exhausted claim, Nelson argues that his IATC-Mental Health claim is unexhausted because it is not the same as the ineffectiveness claim that he brought on state habeas. “For claims that are not adjudicated on the merits in the state court . . . we do not apply the deferential scheme laid out in § 2254(d) and instead apply a de novo standard of review.” Ward, 777 F.3d at 256 (citations and internal quotation marks omitted). Further, Pinholster‘s bar on new evidence would not apply to an unexhausted claim. See Pinholster, 563 U.S. at 186 (“[N]ot all federal habeas claims by state prisoners fall within the scope of § 2254(d) [limiting the federal court to the record that was before the state court], which applies only claims adjudicated on the merits in State court proceedings.” (internal quotation marks omitted)). If a petitioner has not exhausted the available state remedies for his claim, that claim is procedurally defaulted and a federal court ordinarily cannot consider it on habeas review. See Coleman v. Thompson, 501 U.S. 722, 731 (1991). However, “merits-review of a procedurally barred claim is permitted when the petitioner is able to demonstrate cause for the default and actual prejudice as a result of the alleged violation of federal law.” Segundo, 831 F.3d at 350 (citation and internal quotation marks omitted). Nelson argues that he can demonstrate cause for his asserted procedural default of this claim under Martinez v. Ryan and Trevino v. Thaler. In these cases, the Supreme Court established that “[i]nadequate assistance of counsel at initial- review collateral proceedings may establish cause for a prisoner’s procedural default of a claim of ineffective assistance at trial.” Martinez, 566 U.S. at 9 (so holding for jurisdictions where ineffective assistance of trial counsel claims cannot be brought on direct appeal); see Trevino, 569 U.S. at 428 (extending Martinez to jurisdictions such as Texas that “do not offer most defendants a meaningful opportunity to present a claim of ineffective assistance of trial counsel on direct appeal,” even if they do not expressly prohibit it). The petitioner must demonstrate that state habeas counsel was ineffective under the standard established in Strickland and, further, that the underlying ineffective assistance of trial counsel claim on which he ultimately seeks relief is “substantial.” Martinez, 566 U.S. at 14. Here, at the COA stage, Nelson would have to show that reasonable jurists could debate that he can make such a showing for there to be “cause” under Martinez for the procedural default. As discussed further below, we find that Nelson cannot make this showing because reasonable jurists could not debate the substantiality of Nelson’s underlying IATC-Mental Health claim. No. 17-70012 2. Substantiality of the Claim Reasonable jurists could not debate the substantiality of Nelson’s underlying IATC-Mental Health claim. As noted, a petitioner alleging ineffective assistance of trial counsel must demonstrate that counsel’s performance was deficient and that this deficiency prejudiced the defense. Wiggins v. Smith, 539 U.S. 510, 521 (2003) (citing Strickland v. Washington, 466 U.S. 668, 687). When the alleged ineffective performance is a failure to investigate, we ask whether “reasonable professional judgments support the limitations on investigation.” Id. at 528 (citing Strickland, 466 U.S. at 691). In Wiggins, as here, the petitioner alleged that counsel’s deficiency “stem[med] from counsel’s decision to limit the scope of their investigation into potential mitigating evidence.” Wiggins, 539 U.S. at 521. The petitioner’s death penalty counsel in Wiggins relied on the pre-sentencing report and foster care records as their exclusive sources of information about their client’s personal history, despite indications therein that he had suffered a traumatic childhood worth investigating. Id. at 523-24. The Supreme Court noted that death penalty counsel has an “obligation to conduct a thorough investigation of the defendant’s background,” (citing Williams, 529 U.S. at 396), and held that counsel was ineffective for unreasonably limiting their investigation to these two sources that provided only a cursory understanding of the petitioner’s history. This is not a case, like Wiggins, in which counsel “abandoned their investigation of petitioner’s background after only a rudimentary knowledge of [defendant's] history from a narrow set of sources.” 539 U.S. at 524. Nor did Nelson’s counsel, as in the other cases Nelson relies on, fail to “even take the first step of interviewing witnesses or requesting records,” Porter v. McCollum, 558 U.S. 30 (2009); “fail[] to conduct an investigation that would have uncovered extensive records graphically describing [defendant's] nightmarish childhood . . . because they incorrectly thought that state law barred access to such records,” Williams, 529 U.S. at 395; fail to review easily available prior conviction records that were informed the prosecution would rely on as aggravating evidence, Rompilla v. Beard, 545 U.S. 374, 387-89 (2005); fail to hire a mitigation specialist or, by their “own admission . . . conduct any mitigation investigation” Canales v. Stephens, 765 F.3d 551, 569 (5th Cir. 2014) (emphasis in original); or “only skim[] the records” on the defendant’s background and fail to discuss the mitigation issue with the psychologist hired for guilt phase or contact witnesses who had “first[-]hand knowledge of his troubled childhood,” Walbey v. Quarterman, 309 F. App’x 795, 796 (5th Cir. 2009). To the contrary, Nelson’s trial counsel hired a mitigation specialist who “generated a detailed Psychosocial History” for Nelson; obtained and reviewed Nelson’s voluminous school, juvenile, medical, criminal, jail, and mental health records; interviewed approximately twenty of Nelson’s family and friends and tried to contact others who refused to help or would not answer calls; retained a forensic psychologist to evaluate Nelson; and met with Nelson on numerous occasions to “keep him informed and afford him every opportunity to assist counsel in preparing his defense.” Nelson cites no authority that indicates that his counsel’s extensive and manifold mitigation investigation fell below the objective standard of reasonableness. See Wiggins, 539 U.S. at 533 (“Strickland does not require counsel to investigate every conceivable line of mitigating evidence”). Moreover, reasonable jurists cannot debate Nelson’s core complaint that counsel should have hired another psychological expert other than Dr. McGarrahan to investigate how childhood trauma shaped his destructive choices. Dr. McGarrahan, a forensic psychologist who specialized in evaluating individuals in the criminal justice system, met with Nelson twice to interview him and perform psychological testing for “approximately six to eight hours.” Additionally, Dr. McGarrahan spoke with Nelson’s mother and reviewed “several thousand pages of records” provided by trial counsel, including documents from the underlying capital murder offense, past criminal history, jail and juvenile detention mental health and disciplinary records, educational records, and his medical and mental health records from early childhood. Dr. McGarrahan assessed that Nelson had “significant psychiatric issues . . . that began at a very early age . . . a history of severe ADHD, antisocial personality disorder, and some substance abuse history.” She did not diagnose Nelson with PTSD or indicate that his psychological damage could be remedied so as to render him no longer dangerous. We have consistently found that death penalty counsel is not ineffective if they rely on a medical expert’s assessment of the defendant’s mental functioning to inform their punishment phase strategy “instead of pushing ahead with [their] own investigation or hiring new experts who may have reached a different diagnosis.” Smith v. Cockrell, 311 F.3d 661, 676 (5th Cir. 2002), abrogated on other grounds by Tennard v. Dretke, 542 U.S. 274 (2004) (“this court has refused to find that counsel violated the Strickland standard by failing to locate a different expert after the original expert concluded that the defendant was not mentally retarded”); see Segundo, 831 F.3d at 352 (“Given trial counsel’s investigation and reliance on reasonable expert evaluations, Segundo cannot overcome the strong presumption that counsel’s representation fell within the wide range of reasonable professional assistance.”); see also Fairbank v. Ayers, 650 F.3d 1243, 1252 (9th Cir. 2011) (“[C]ounsel in this case provided the defense expert with the information necessary to form an expert opinion, and the expert did, in fact, investigate the potential defense. Later disagreement by other experts as to the conclusions does not demonstrate a violation of Strickland.”). The fact that habeas counsel located another expert, Dr. Bradley, who reached different and arguably more sympathetic conclusions than Dr. McGarrahan when Dr. Bradley interviewed Nelson five years later, does not render trial counsel ineffective for relying on Dr. McGarrahan’s assessments. To the extent Nelson claims that counsel was ineffective for presenting testimony from McGarrahan at all, or for generally failing to present a persuasive picture of his mental health and background, we also do not believe reasonable jurists could debate that he has failed to demonstrate a substantial claim. Once we determine that the investigation underlying a mitigating strategy was reasonable, counsels’ decisions on what evidence to present and how deserve considerable deference. See Strickland, 466 U.S. at 690 (“strategic choices made after thorough investigation of law and facts relevant to plausible options are virtually unchallengeable”); Tenny v. Cockrell, 420 F. Supp. 2d 617 (W.D. Tex. 2004) (“If the investigation into mitigating evidence was reasonable under prevailing professional norms, the strategy developed from the results of the investigation deserve deference.”). The record demonstrates that counsel presented a detailed and significant mitigation case, aided by McGarrahan’s assessment of how childhood neglect and mistreatment likely left Nelson with significant psychological damage that set him on his violent path. Dr. McGarrahan testified that “research shows that . . . emotional unavailability or emotional neglect of an infant is worse psychologically than physical abuse” and told the jury that she believed that Nelson was exposed to this type of harm from an early age. She emphasized that Nelson’s childhood behavior indicated that he had to “cry out for help” through violence because he had important needs that “went unmet,” and asserted that the degree of his psychological damage indicated that this mistreatment was severe. Specifically, McGarrahan cited a number of risk factors that she believed led to Nelson’s “psychologically abnormal development,” including an overworked mother, a father who was either abusive or absent throughout Nelson’s life, and Nelson’s exposure to violent domestic abuse. She concluded that there were “absolutely” choices made by other people in Nelson’s formative years that shaped the direction of his life and that, by the time Nelson could make choices for himself, he was already “wired” to be “predisposed to severe aggression and violence” because of what he had experienced since infancy. Dr. McGarrahan’s testimony tied together the descriptions of Nelson’s absentee mother, abusive father, and other childhood struggles offered by his other mitigation witnesses, including Nelson’s mother, brother, sister, uncle, his mother’s ex-boyfriend, and a behavioral health counselor who treated him when he was young. Nelson argues that counsel’s conduct in calling Dr. McGarrahan was ineffective because she testified that Nelson had “many, many psychopathic characteristics” after informing counsel she would have to admit as much if asked. However, “a conscious and informed decision on trial tactics and strategy cannot be the basis of constitutionally ineffective assistance of counsel unless it is so ill chosen that it permeates the entire trial with obvious unfairness.” Richards v. Quarterman, 566 F.3d 553 (5th Cir. 2009) (citations omitted)). Dr. McGarrahan admittedly made no attempt to downplay Nelson’s violent and destructive tendencies, declaring for instance that “once we are at where we are now, there’s certainly no cure.” Nelson’s trial counsel, however, strategically framed this characterization: eliciting Dr. McGarrahan’s testimony that the decisions that caused Nelson to reach the “point of no return” were “essentially his mother’s and his father’s,” not his own choices. Though this would do nothing to convince a jury to answer in Nelson’s favor on the first special question, whether he would “commit criminal acts of violence that would constitute a continuing threat to society,” it arguably could have worked in Nelson’s favor when the jury was evaluating the third special question, whether Nelson’s “character and background, and [] personal moral culpability” provided mitigating circumstances to warrant a life instead of death sentence. Trial counsel’s mitigation notes and closing argument indicate that this trade-off was indeed a conscious, strategic decision. In her pre-trail notes to counsel, Nelson’s mitigation specialist wrote that “in light of current jail events . . . . [o]f course [Dr. McGarrahan] agrees with us that future dangerousness cannot be refuted.” (emphasis added). Consistently, trial counsel stated at closing of future dangerousness: There’s certainly been enough of that for you to find if that’s what you want to find. Okay. Our own expert pointed that out . . . as we have tried to present this case, we have not tried to hide a fact from you. I’ve not tried to keep something from you. Evidently and, we believe, reasonably, Nelson’s trial counsel determined that they would lose the jury’s trust if they attempted to maintain that Nelson was not a present and future danger. Instead, they built a defense around presenting him as someone whom the jury should pity because he did not stand a chance of growing up differently because of childhood abuse and neglect: You looked over at him, I know you did, when the verdict was read and he didn’t cry and showed no emotion. . . . He can’t cry because crying quit doing anything for him when he was about four years old. That’s why he set the bed on fire. Every decision that’s ever been made for Steven Nelson has been the wrong decision. He’s made a lot of them. But the first ones, the ones that Dr. McGarrahan told you about that put him on the track for permanent derailment, those were the ones that were beyond his control. And if that’s not mitigating, there is not mitigation in a death penalty case. . . . He will never be any better. He was a train wreck waiting to happen. He didn’t ask to be in that position. We do not find it debatable that “under the circumstances, the challenged action might be considered sound trial strategy,” and accordingly find that Nelson cannot raise a substantial claim that trial counsels’ decision to present Dr. McGarrahan’s expert testimony as part of their mitigation strategy fell outside the bounds of professional reasonableness. Strickland, 466 U.S. at 689 (citations and internal quotation marks omitted). Accordingly, Nelson’s IATC-Mental Health claim is neither debatable on the merits, nor so substantial as to permit him to overcome procedural default. 3. Funding Under § 3599(f) In addition to seeking a COA on this claim, Nelson directly appeals the district court’s denial of funding under 18 U.S.C. § 3599(f) to hire both a psychiatric expert, and an expert in life-long incarceration, to further evaluate Nelson in support of this claim. As relevant here, § 3599(f) provides that capital defendants seeking habeas review are entitled to funding for “reasonably necessary” investigative and expert services. We review the district court’s denial of motions for funding under this section for abuse of discretion. While this appeal was pending, the Supreme Court decided Ayestas v. Davis, 138 S. Ct. 1080 (2018). In Ayestas, the Supreme Court determined that this circuit’s requirement that petitioners demonstrate a “substantial need” for services requested under § 3599 was impermissibly more demanding than the “reasonably necessary” standard established in the statute. Id. at 1092. “What the statutory phrase calls for,” the Supreme Court held in Ayestas, “is a determination by the district court, in the exercise of its discretion, as to whether a reasonable attorney would regard the services as sufficiently important, guided by the considerations we set out more fully below.” Id. at 1093. The Supreme Court then identified three factors that a district court must consider when evaluating whether a reasonable attorney would seek such services: “[1] the potential merits of the claims that the applicant wants to pursue, [2] the likelihood that the services will generate useful and admissible evidence, and [3] the prospect that the applicant will be able to clear any procedural hurdles standing in the way.” Id. at 1094. Nelson requests that we vacate the district court judgment and remand for that court to apply this newly-articulated Ayestas standard to his requests for investigative funding for his IATC-Mental Health claim. As we have just determined, however, Nelson has not raised a substantial claim that he can overcome the applicable procedural hurdles to this claim, nor can he demonstrate that the IATC-Mental Health claim has potential merit. No evidence Nelson could uncover with the aid of further investigative funding would affect our determination, detailed above, that counsel’s investigation of these issues was reasonable based on what they knew at the time. See Wiggins, 539 U.S. at 523 (reasonableness of counsels’ investigation is “a context-dependent consideration of the challenged conduct as seen ‘from counsel’s perspective at the time’” (citing Strickland, 466 U.S. at 689)). Because Nelson therefore could not demonstrate that he is entitled to funding for the requested services to bolster his IATC-Mental Health claim, remand is unnecessary. See Ayestas II, 933 F.3d at 388 (remand for the district court to reconsider funding under the Supreme Court’s annunciated standard in Ayestas not required “if the judgment is sustainable for any reason” (quoting Af-Cap Inc. v. Rep. of Congo, 462 F.3d 417, 425 (5th Cir. 2006))). We therefore affirm the district court’s denial of funding under § 3599(f) for a psychiatric expert and an expert on life-long incarceration. B. Failure to Adequately Investigate Holden’s Death Nelson also contends that his trial counsel were ineffective at the punishment phase because they failed to adequately investigate and present a defense to the State’s punishment phase evidence that Nelson killed Jonathan Holden, another inmate at the Tarrant County Jail, while Nelson was awaiting trial. Specifically, Nelson argues that counsel insufficiently cross-examined Rick Seely, the State’s eyewitness, and failed to present additional evidence that Holden was suicidal.^[2] 1. Procedural Hurdles As with Nelson’s IATC-Mental Health argument, the district court found that this was simply new evidence in support of the same ineffective assistance of counsel at sentencing claim that Nelson brought on state habeas. Accordingly, it held that Nelson could not raise these new examples of counsel’s alleged ineffectiveness because these facts were not before the state court when it denied this claim. See Pinholster, 563 U.S. 170, 185. Nelson, as with the claim discussed above, contends that this is a different, unexhausted claim, and he invokes Martinez and Trevino to attempt to demonstrate cause and prejudice for the procedural default. As with his IATC-Mental Health claim, however, we find that reasonable jurists could not debate whether Nelson is entitled to relief regardless of whether this claim is exhausted or unexhausted. If exhausted, these new examples of alleged ineffectiveness are barred from consideration under Pinholster. If unexhausted, Nelson cannot show cause and prejudice for his failure to raise this claim in state court because no reasonable jurists could debate that this underlying ineffective assistance of counsel claim is not substantial. 2. Substantiality of the Claim As noted, “[j]udicial scrutiny of counsel’s performance must be highly deferential,” and we “indulge a strong presumption that counsel’s conduct falls within the wide range of reasonable professional assistance.” Strickland, 466 U.S. 689. On the whole, the record demonstrates that Nelson’s trial counsel made substantial efforts to discredit and rebut the State’s position that Nelson murdered Holden, both by cross-examining the State’s key witnesses and offering their own competing expert testimony. The State called Rick Seely, another inmate at the Tarrant County facility with Nelson, as the only eyewitness to Holden’s death. Seely told the jury that Holden had angered Nelson and other inmates by muttering the N- word under his breath. Later that morning, Nelson was released out into the common area surrounding the jail cells for his designated recreation time. According to Seely, Nelson, after jabbing at several other inmates including Holden through the bars of their cells with a broom handle, told Holden that he, Nelson, wanted Holden to get himself transferred out of the area. Nelson instructed Holden to press the button in his cell to call the guards and tell them he was going to kill himself. Seely stated that Nelson then “coaxed” Holden to stage a suicide attempt, convinced Holden to come over to the bars of his door, and wrapped a blanket around his neck. Nelson then pulled on the ends of the blanket from outside the cell for several minutes until Holden died. Seely testified that Nelson then tied the blanket to the top horizontal rail on the jail bars so that it would look like Holden had hanged himself. In cross-examining Seely, Nelson’s counsel noted a potential inconsistency in his testimony and highlighted Seely’s own violent felony convictions. Additionally, trial counsel questioned Seely’s motives for his testimony, probing whether he hoped for special treatment in exchange for his cooperation and openly expressing skepticism that he was providing evidence “out of the goodness of [his] heart.” Though Nelson asserts that counsel failed to press other potential inconsistencies in Seely’s testimony, the fact that Nelson has identified in hindsight another un-probed weakness in Seely’s testimony does not render his trial counsel’s cross-examination unreasonable. See United States v. Bernard, 762 F.3d 467, 472-73 (5th Cir. 2014) (rejecting petitioners’ claim that counsel was ineffective at the punishment phase for failing to “more effectively attack[]” witnesses they “ vigorously cross-examined”). The State also called Sergeant John Campos, an employee at the jail. Campos stated that he was on duty the day Holden died and found him hanging from a blanket tied to the cell door. Campos testified that the knots were unusually loose and simple compared to suicide hangings. On cross- examination, Nelson’s counsel asked Campos if he knew that Holden was on suicide watch, and elicited Campos’s confirmation that his initial belief on finding Holden was that he had hanged himself. The State also presented a forensic scientist who testified that Nelson could not be excluded as a contributor to the DNA mixture found under Holden’s fingernails, and Dr. Lloyd White, the medical examiner who performed Holden’s autopsy, who testified that he believed Holden’s injuries and ultimate death resulted from “ligature strangling due to assault by another person.”^[3] Nelson contends that his trial counsel should have more thoroughly cross-examined Seely and presented additional evidence that Holden was suicidal and killed himself. We find that reasonable jurists cannot debate the sufficiency of counsels’ performance in either respect. During Seely’s cross- examination, Nelson’s counsel noted a potential inconsistency in his testimony and highlighted Seely’s own violent felony convictions. Additionally, trial counsel questioned Seely’s motives for his testimony, probing whether he hoped for special treatment in exchange for his cooperation and openly expressing skepticism that he was providing evidence “out of the goodness of [his] heart. Though Nelson asserts that counsel failed to press other potential inconsistencies in Seely’s testimony, the fact that Nelson has identified in hindsight another un-probed weakness in Seely’s testimony does not render his trial counsel’s cross-examination unreasonable. See United States v. Bernard, 762 F.3d 467, 472-73 (5th Cir. 2014) (rejecting petitioner’s claim that counsel was ineffective at the punishment phase for failing to “more effectively attack[]” witnesses they “ vigorously cross-examined”). Trial counsel also called Dr. John Plunkett as an expert witness for the defense, an independent medical examiner who reviewed the records of Holden’s death. Dr. Plunkett testified that there were no injuries to Holden’s head, neck, or back to indicate he was pulled up against the jail door and forcibly strangled, calling Seely’s account into question. Dr. Plunkett informed the jury that, based on Holden’s position when he was found, he would have suffered cardiac arrest if he had merely “slouch[ed] down or lean[ed] forward” into the tied-off blanket for approximately five minutes or could have “simply stood up and got out of it.” Ultimately, Dr. Plunkett opined that he could not definitively conclude whether Holden killed himself or was killed by another person, but could conclude with confidence that “if someone else assisted [Holden] in his death,” Holden “must have been an active participant.” At closing, Nelson’s counsel reiterated that Dr. Plunkett’s testimony “lends to the obvious story . . . [that] Holden had to have been some sort of active participant” in his death, and stated that “there’s absolutely no injury on Jonathan Holden’s body that would support” Seely’s testimony that Nelson strangled him for four minutes or more. Nelson asserts that counsel should also have informed the jury that Holden had attempted suicide just weeks before his arrest and had already injured himself while incarcerated, and also should have called as a witness Charles Bailey, another inmate at the jail, who allegedly would have testified that he believed Holden’s death was a suicide. “[C]omplaints of uncalled witnesses are not favored in federal habeas corpus review because the presentation of testimonial evidence is a matter of trial strategy and because allegations of what a witness would have stated are largely speculative.” Day v. Quarterman, 566 F.3d 527, 538 (5th Cir. 2009) (citations omitted). “To prevail . . . the petitioner must name the witness, demonstrate that the witness was available to testify and would have done so, set out the content of the witness’s proposed testimony, and show that the testimony would have been favorable to a particular defense.” Id. (citing Alexander v. McCotter, 775 F.2d 595, 602 (5th Cir.1985)). Here, Nelson’s trial counsel did raise the fact that Holden had been on suicide watch in their cross-examination of Campos, and Nelson does not explain who else counsel should have called as a witness to present additional evidence of Holden’s prior self-injury. Regarding Charles Bailey, Nelson similarly makes no showing that he was available to testify, nor does he make any proffer of what Bailey’s testimony would have been. Instead, Nelson only provides the bare assertion that Bailey would have “corroborated Mr. Nelson’s claim that Mr. Holden killed himself.” In fact, Bailey’s prior statements only intimate that Bailey once believed Holden’s death resembled a suicide, and certainly do not convey that the sum total of Bailey’s potential testimony would have been favorable to Nelson or that Bailey, who specifically stated that he blocked his view from his cell because he didn’t “want to be a witness to nothing,” would have testified. Ultimately, reasonable jurists cannot debate that Nelson cannot raise a substantial claim that his trial counsel’s methods of vigorously challenging the State’s evidence that Nelson killed Holden did not “fall[] within the wide range of reasonable professional assistance.” Strickland, 466 U.S. at 688. C. Failure to Investigate Involvement of Alleged Co-conspirators In his final argument that his trial counsel was ineffective during the punishment phase, Nelson alleges that counsel failed to properly investigate and present potential evidence that Claude “Twist” Jefferson and Anthony “AG” Springs were involved in Dobson’s murder. We refer to this as Nelson’s “IATC-Participation” claim. Specifically, Nelson contends that counsel was ineffective for failing to follow-up on known weaknesses in the other men’s alibis or even interview these men directly. Though Nelson did not take the stand during the punishment phase, he testified during the guilt phase (despite his counsels’ advice to the contrary) that he was not present in the church during the assaults on Dobson and Elliott. Instead, he asserted, he served as a look-out while Springs and Jefferson entered the church to rob the people inside. In anticipation of this defense, the State presented alibi witnesses for both Springs and Jefferson. These witnesses testified that Springs and Jefferson were indeed with Nelson later that afternoon when he used the victims’ stolen credit cards, but testified that, when the murder was committed earlier that day, Springs was with his girlfriend Kelsey Bursey in Venus forty-five minutes away and Jefferson was in class at the University of Texas. At closing, the State argued that “one person cause[d] the devastation and the horror and the terror that took place in that church . . . . One person committed this act, not the other two people he wants to incriminate because he thinks he can con you all into believing something that’s not true.” In response, Nelson’s counsel urged the jury to believe that all three men were involved, expressing doubt about Springs and Jefferson’s alibis and encouraging the jury to conclude that the State’s “lone actor theory doesn’t make much sense.” In finding Nelson guilty of Dobson’s murder, the jury did not necessarily reject Nelson’s narrative that others were involved and perhaps even committed the murder. Consistent with Texas’s law of parties, the jury received the following instruction: If you find from the evidence beyond a reasonable doubt that . . . STEVEN LEWAYNE NELSON, did then and there intentionally cause the death of an individual, CLINTON DOBSON . . . [and was] in the course of committing or attempting to commit the offense of robbery . . . then you will find the Defendant guilty of the offense of capital murder . . . -OR- If you find from the evidence beyond a reasonable doubt that the defendant, STEVEN LAWAYNE NELSON, entered into a conspiracy, if any, with CLAUDE JEFFERSON or ANTHONY SPRINGS . . . to commit the felony offense of robbery, and that . . . in an attempt to carry out the agreement, if any, CLAUDE JEFFERSON or ANTHONY SPRINGS . . . intentionally cause[d] the death of an individual, CLINTON DOBSON . . . and that such offense was committed in the furtherance of the robbery, and was an offense that STEVEN LEWAYNE NELSON should have anticipated as the result of carrying out of the agreement, if any, then you will find the defendant, STEVEN LEWAYNE NELSON, guilty of the offense of capital murder, though he may have had no specific intent to commit the offense of capital murder. On the verdict form, the jury declared the Nelson was “guilty of the offense of Capital Murder” without identifying which theory it relied on. It is therefore not clear from the conviction whether the jurors had unanimously accepted the State’s narrative that Nelson alone murdered Dobson and assaulted Elliot. The extent of Nelson’s role in the murder was critical at the punishment phase. Nelson could be sentenced to death only if the jury determined that he “actually caused the death of the deceased or did not actually cause the death of the deceased but intended to kill the deceased or another or anticipated that a human life would be taken.” Tex. CODE CRIM. PRO. Art. 37.071. Whether Nelson participated in a robbery in which another murdered Dobson or single- handedly murdered Dobson himself could also substantially impact the jury’s answer to the two other special questions: “whether there is a probability that the defendant would commit criminal acts of violence that would constitute a continuing threat to society” and “[whether] all evidence admitted at the guilt or innocence stage and the punishment stage . . . militates for or mitigates against the imposition of the death penalty.” Id. In closing arguments at the punishment phase, Nelson’s trial counsel asked the jurors to consider whether “you, in the back of your mind, affirmatively believe that there was only one person there? Do you really think that’s the case?” The State, in contrast, stated emphatically that “there wasn’t anyone else there. This is the killer right here.” 1. Procedural Hurdles Like Nelson’s other IATC claims, we must first examine whether his IATC-Participation claim is exhausted (as the district court held and the State argues), or unexhausted (as Nelson argues). As discussed, Nelson’s state habeas counsel, John Stickels, alleged that trial counsel “failed to investigate [his] background, history, family, and friends, and, as a result, failed to discover relevant and important mitigation evidence,” and declared that Nelson “has many family members, friends[,] and former teachers that could have testified on his behalf during the punishment phase of trial but did not do so.” This claim, and the state court’s discussion thereof, addressed whether trial counsel’s investigation into Nelson’s character and background was deficient. It did not touch on Nelson’s allegations in this IATC-Participation claim that undiscovered evidence indicating that he played a minimal role in the capital murder itself could have been presented to the jury. The Supreme Court in Pinholster specifically noted that “we do not decide where to draw the line between new claims and claims adjudicated on the merits.” 563 U.S. at 186, n.10; see also id. at 216, n.7 (Sotomayor, J., dissenting) (“The majority declines, however, to provide any guidance to the lower courts on how to distinguish claims adjudicated on the merits from new claims.”). Our circuit has found that, while “merely putting a claim in a stronger evidentiary posture is not enough,” new evidence that “fundamentally alters the legal claim” or “place[s] the claim in a ‘significantly different legal posture’” can render it a new claim that was not adjudicated on the merits by the state court. Ward, 777 F.3d at 258, 259. We believe reasonable jurists could debate whether Nelson’s IATC-Participation allegations “fundamentally alter” his IATC claim, and so constitute a different and unexhausted claim. As he does for his other IATC claims, Nelson contends that Stickels’ ineffectiveness in failing to bring this claim permits him, under Martinez and Trevino, to overcome the procedural default of this claim. The district court briefly addressed this argument, rejecting what it deemed Nelson’s “conclusory allegations that Stickels’ representation was deficient.” In so doing, however, the district court relied in part on its finding that Stickels’ alleged ineffectiveness for failing to bring Nelson’s IATC-Participation claim could not be considered because Stickels in fact raised this claim when he alleged that trial counsel “failed to investigate [Nelson's] background, history, family, and friends.” If Stickels did not raise Nelson’s IATC-Participation claim, the correct inquiry here is whether reasonable jurists could debate that Stickels provided ineffective assistance in failing to do so. We conclude that they could. Counsel can be found ineffective if they failed to “raise or properly brief or argue certain issues.” Sharp v. Puckett, 930 F.2d 450 (1991) (citing Penson v. Ohio, 488 U.S. 75 (1988)). As in the typical Strickland context, “our review is deferential, presuming that ‘counsel’s conduct falls within the wide range of reasonable professional assistance.’” United States v. Williamson, 183 F.3d 458, 462 (5th Cir. 1999) (quoting Strickland, 466 U.S. at 688). Counsel “need not (and should not) raise every nonfrivolous claim.” Smith v. Robbins, 528 U.S. 259, 288 (2000) (citing Jones v. Barnes, 463 U.S. 745 (1983)). However, “a reasonable attorney has an obligation to research relevant facts and law, or make an informed decision that certain avenues will not prove fruitful.” Williamson, 183 F.3d at 462 (citing Strickland, 466 U.S. at 688); see also Wiggins, 539 U.S. at 523 (counsel performs deficiently when the “investigation supporting counsel’s decision not to” pursue particular strategy “was itself [unreasonable"). "[C]ourts are ‘not required to condone unreasonable decisions parading under the umbrella of strategy, or to fabricate tactical decisions on behalf of counsel when it appears on the face of the record that counsel made no strategic decision at all.’” Richards, 566 F.3d at 564 (quoting Moore v. Johnson, 194 F.3d 586, 604 (5th Cir. 1999)). Here, reasonable jurists could debate whether Stickels was ineffective for failing to do the investigation necessary to make an informed decision on whether to consider an IATC-Participation claim. Stickels hired a mitigation specialist to assist him, Gerald Byington, and both spent substantial time reviewing Nelson’s case file and considering Nelson’s trial team’s mitigation investigation. However, there is no indication that they considered whether Nelson’s trial team adequately investigated and presented the argument that Springs and Jefferson were involved in the crime. As Byington summarized trial counsels’ mitigation strategy: [I]t appears there were two major themes presented by the defense. One of these themes was the presentation of medical/DNA evidence related to the death of Mr. Holden . . . The evidence presented by the defense appears to have been an effort to provide reasonable doubt that Mr. Nelson was in fact responsible for Mr. Holden’s death. The second theme of the defense’s punishment case appeared to focus on the numerous developmental problems and circumstances of Mr. Nelson’s life. Nowhere in his report, however, does Byington mention that trial counsel also attempted at the punishment phase to contend that Nelson was not the sole or even primary assailant. Nor did Byington or Stickels appear to evaluate the extent of trial counsels’ investigation of Nelson’s alleged co-conspirators that they did—or failed to do—in preparation for this argument. It is also undisputed that neither Stickels nor Byington did any independent research into Springs’ or Jefferson’s involvement to determine whether there was information that trial counsel should have uncovered. Reasonable jurists could debate that Stickels failed to do the investigation necessary to make an informed decision about whether pursuing a IATC-Participation claim on state habeas could prove fruitful. As with “a counseled appeal after conviction . . . the key is whether the failure to raise an issue worked to the prejudice of the defendant.” Sharp, 930 F.2d at 453. In other words, Nelson can demonstrate prejudice if there is merit to his underlying IATC-Participation claim. Here, reasonable jurists could debate whether Nelson was prejudiced by Stickels’ failure to bring this claim on state habeas because, as discussed below, reasonable jurists could debate the merits of this underlying claim. 2. Substantiality of the Claim As with any other IATC claim, the underlying IATC-Participation claim (which, if viable, may allow a claim that state habeas counsel potential ineffectiveness prejudiced Nelson, thereby excusing procedural default) requires a showing of two elements: (1) deficient performance; and (2) prejudice. Strickland, 466 U.S. at 687. i. Trial Counsels’ Performance Nelson contends that his trial counsel was ineffective for conducting a deficient investigation into Springs’ and Jefferson’s involvement to support their defense theory that Nelson was not the sole assailant. “In assessing the reasonableness of an attorney’s investigation . . . a court must consider . . . whether the known evidence would lead a reasonable attorney to investigate further.” Wiggins, 539 U.S. at 527 (2003). Nelson emphasizes several leads that, he asserts, should have alerted competent counsel to investigate further. The State initially arrested both Springs and Nelson for the murder, after receiving information from two of the men’s acquaintances connecting them both to the murder. These acquaintances told the police, as memorialized in the incident reports, that both Nelson and Springs made inappropriate and suspicious comments when a news report of Dobson’s death showed on television, that Springs tried to sell them Dobson’s iPhone that he had in his possession, and that Springs and Nelson would on other occasions “go out and commit robberies and burglaries together.” Further, counsel was aware of images in the police file showing that Springs, unlike Nelson, had “extensive bruising and swelling on [the] knuckles of both hands” days after Dobson’s murder and Elliot’s beating, and had provided only a weak explanation of how he had sustained these injuries. Finally, Nelson asserts that trial counsel knew that there were weaknesses in Springs’ alibi. Kelsey Bursey, who testified that Springs was with her in Venus, Texas, before and during the murder, was his girlfriend and mother of his child, and therefore not an unbiased source. In fact, the police officer who initially interviewed her after she arrived at the station to tell them that Springs had been with her and not been in Arlington where Dobson was killed wrote in his incident report that he “believed Springs was involved in this offense and further believed [Bursey] may be attempting to cover up his behavior by supplying him an alibi.” Further, though the State presented Springs’ phone records as additional evidence to demonstrate that he was in Venus when the murder was committed, the defense pointed out on cross- examination that these phone records did not provide any information about where Springs was between 10:18 p.m. the night before until 12:13 p.m. the day of the murder. Despite counsels’ awareness of these leads, Nelson notes that trial counsel never interviewed Springs. Failure to interview important potential witnesses can constitute ineffective assistance. See, e.g., Richards, 566 F.3d at 570-71; Bryant v. Scott, 28 F.3d 1411, 1415 (5th Cir. 1994). Nelson adds that counsel also did not take other steps to probe the veracity of Springs’ alibi or investigate the cause of the bruising on his fists. Nelson argues that Jefferson’s alibi was similarly questionable and that, despite this and Nelson’s insistence that Jefferson was involved, counsel did not take even a basic step to verify it. Though Jefferson alleged that he was in Chemistry class at the University of Texas taking a test, his instructor stated that she had not administered any quiz or exam on that day. Correspondence with the school also revealed that Jefferson did not complete that semester, and stopped going to class entirely less than a month later. Jefferson’s professor provided a sign-in sheet for the day in question and also stated that there was security camera footage on file with the school that would show the students as they entered the classroom. The record shows that Nelson’s trial counsel reviewed the sign-in sheet and observed that, though there were initials written next to Jefferson’s name, the handwriting looked markedly similar to the initials just below his, as if that other person had signed in for him. Despite noting this, however, trial counsel did not contact a handwriting expert, obtain other samples of Jefferson’s handwriting, or otherwise seek evidence that Jefferson had not signed his own initials. Instead, the only step they took was to ask Jefferson’s aunt on cross-examination if it looked from the handwriting like someone else had signed in for her nephew. She responded that it did not look like that to her. Nelson’s counsel also did not obtain the security camera footage of students entering the classroom. Federal habeas counsel submitted evidence from the university’s technology department that this recording would have been available if trial counsel had sought it shortly before Nelson’s trial, but had been erased before Nelson’s federal proceedings began. As with Springs, counsel also did not interview Jefferson, either about his uncertain alibi or the attack at the church. The reasonableness of pretrial investigation should be considered in light of the chosen trial strategy. Cf. Moore, 194 F.3d at 608 (“counsel’s pretrial investigation into extraneous conduct was inadequate in light of the chosen alibi defense” that required defendant to testify and thus open the door to the prosecution to present such evidence (emphasis added)). Here, counsels’ alleged deficiencies in investigating Springs’ and Jefferson’s alibis and involvement were compounded by counsels’ strategy: to convince the jury that these two men were involved in the murder, but without evidence to back up the theory. As noted, counsel pointed out on Bursey’s cross-examination that Springs’ cell phone data could not confirm his location during the murder, but had not interviewed Springs to see if they could learn anything else connecting him to the murder or otherwise undermining Bursey’s account of his actions. Counsel also attempted to cast doubt on Jefferson’s alibi that he was in Chemistry class, unsuccessfully cross-examining his aunt and proposing at closing: “Could it be that, gee, a college-age kid who runs around with other knuckleheads doesn’t show up for class? Is that that hard to believe?” Nelson provided an affidavit from one juror who stated that he found against Nelson in part because Nelson “tried to pin it on other people, but there was no evidence to support that,” illustrating how counsels’ failure to seek this evidence weakened Nelson’s defense in light of their strategy. Because Nelson’s counsel sought to convince the jury that Springs and Jefferson were involved but arguably failed to take reasonable investigative steps in developing evidence in support of this argument, we believe reasonable jurists could debate that his trial counsel’s performance in this regard was deficient. ii. Prejudice from Trial Counsel’s Alleged Deficiency and § 3599(f) Funding Whether a failure to investigate prejudiced a defendant depends on what evidence a reasonable investigation would have uncovered. Wiggins, 539 U.S. at 534-35. Nelson acknowledges that he would have to conduct further investigation to identify what evidence of Springs’ and Jefferson’s involvement trial counsel may have been able to uncover, and sought funding under § 3599(f) for this purpose. The district court denied funding after concluding, based on the evidence presented to the jury, that Nelson did indeed commit the crime alone so no evidence of another’s participation would exist. Of course, Nelson seeks to conduct the requested investigation precisely to locate the evidence that he alleges exists and could have been uncovered to disprove this version of events. Though the district court should not permit Nelson to conduct a “fishing expedition,” neither should it presume that it can glean the full story based solely on the evidence before it when the petitioner’s very claim is that the available evidence was lacking due to deficient investigation. See Ayestas II, 933 F.3d at 388 (emphasis in original) (noting that a district court may abuse its discretion by denying funding after considering only “existing as opposed to potential evidence”). In order to determine whether Nelson should receive the funding that would be necessary to develop his argument that trial counsels’ alleged deficiency prejudiced him, the district court would need to properly consider his motion under the standard articulated by the Supreme Court in Ayestas in the first instance. See Ayestas, 138 S. Ct. at 1092-95 (noting district court’s discretion in assessing funding requests under the “reasonably necessary” standard). Here, however, we find ourselves in something of a Catch-22. We cannot determine whether Nelson was prejudiced without knowing what evidence could have been uncovered, and should not make this determination based solely on the record before us when he may be entitled to investigative funding to support this claim under Ayestas. However, we also cannot vacate the district court’s determination that this claim was procedurally barred and without merit—findings that necessarily preclude awarding funding—without first granting a COA, vesting us with jurisdiction to examine the merits of such claims. We are, finally, reticent to proceed to a thorough merits determination of this claim without the benefit of full briefing on the merits after the COA stage. Acknowledging, then, that we lack and will continue to lack the evidence needed to assess whether Nelson was prejudiced by this deficiency, we nevertheless assess that this claim “deserve[s] encouragement to proceed further,” see Miller-El, 537 U.S. at 327, because without further proceedings beyond the COA stage, we are unable to fully evaluate the district court’s rulings that ultimately precluded funding. We are also mindful that “any doubts as to whether a COA should issue must be resolved in [the petitioner's] favor” in a death penalty case. Ramirez v. Dretke, 393 F.3d 691, 694 (5th Cir. 2005) (alteration in original) (internal quotation marks and citation omitted). Accordingly, we grant a COA on Nelson’s IATC-Participation claim limited to the question of counsel’s performance and whether the claim is procedurally barred. Depending on our resolution of these issues, we may then find it necessary to remand for the district court to apply Ayestas to determine whether there is a “likelihood that the services [requested] will generate useful and admissible evidence” on the prejudice prong. See 138 S. Ct. at 1094. IV. Batson Claim Nelson next seeks a COA on his claim that the State unconstitutionally used race to select an all-white jury, in violation of Batson v. Kentucky, 476 U.S. 79 (1986). The district court found, and the parties do not dispute, that Nelson properly exhausted these claims in state court on direct appeal. Nelson, 2015 WL 1757144, at *10-11. Trial courts employ a three-step inquiry to assess a contemporaneous Batson objection: First, a defendant must make a prima facie showing that a peremptory challenge has been exercised on the basis of race; second, if that showing has been made, the prosecution must offer a race-neutral basis for striking the juror in question; and third, in light of the parties’ submissions, the trial court must determine whether the defendant has shown purposeful discrimination. Foster v. Chatman, 136 S. Ct. 1737, 1747 (2016). “[The] trial court’s ruling on the issue of discriminatory intent must be sustained unless it is clearly erroneous,” even setting aside the required deference we owe to the state court under AEDPA. Snyder v. Louisiana, 552 U.S. 472, 477 (2008) (finding on direct appeal that the state court’s rejection of defendant’s Batson claim in that case “fail[ed] even under th[is] highly deferential standard of review”). Incorporating AEDPA deference, Nelson would have to prove that the TCCA was unreasonable when it concluded that the trial court did not clearly err in finding that the State’s provided race-neutral explanations for striking two black jurors, Martima Mays and Talmadge Spivey, were not pretext. See Miller-El v. Dretke (“Miller-El II”), 545 U.S. 231, 240 (2005) (citing § 2254(d)(2)).^[4] At the COA stage, we consider whether reasonable jurists would debate that Nelson can make this showing. Nelson’s trial counsel objected during voir dire to the State’s use of peremptory strikes to eliminate Mays, Spivey, and other minority jurors. Based on the fact that the State used a “disproportionate number of strikes”—about a third of its total number—on minority jurors, the trial court determined that Nelson made a prima facie case of racial discrimination. The State then provided race-neutral rationales. As to Mays, the State articulated: She served on a jury that resulted in a mistrial. She also, with regard to several questions on her questionnaire[5], wrote, ["]I have not thought about it["], in regard to her feelings on the death penalty. She believed that the death penalty should never be invoked. She again writes, ["]I’ve not thought about it["] for two more questions dealing with the death penalty, but that she would not lose any sleep over the fact that she did not get picked. She also believed that the death penalty was not at the top of her list for possible punishment for a crime. She hesitated during questioning with regard to Question No. 2 with the parties issue. Nelson argues in his federal habeas proceedings that these reasons were mere pretext. He notes that several white jurors also expressed discomfort with the death penalty but were not struck by the State. “If a prosecution’s proffered reasons for striking a black panelist applies just as well to an otherwise-similar nonblack who is permitted to serve, that is evidence tending to prove purposeful discrimination.” Miller-El II, 545 U.S. at 241. The district court considered each of Nelson’s proffered comparators in turn and determined that the trial court could reasonably have found their responses on this issue materially distinguishable. From our review of the voir dire, we agree. Further, as the district court noted, the State contemporaneously provided another reason for treating Mays differently: she had previously “served on a jury that resulted in a mistrial” because the jury could not reach a unanimous verdict. In fact, this was the first reason the State cited for striking her. Nelson does not state that any of the other jurors who expressed hesitation with the death penalty and were not struck had also served on hung juries in the past. Nelson additionally contends that the State mischaracterized Mays’ testimony when it claimed that she “believe[d] that the death penalty should never be invoked.” Nelson points out that Mays’ testimony conveyed hesitancy for the death penalty but that she affirmed that she could impose this disfavored punishment in certain cases. Nelson relies on Foster to argue that this “misrepresentation[] of the record” demonstrates that the State’s proffered rationale was mere pretext. 136 S. Ct. at 1754. Foster, however, was a case where “much of the reasoning provided by [the prosecution for striking the jurors] ha[d] no grounding in fact,” and “the shifting explanations, the misrepresentations of the record, and the persistent focus on race in the prosecution’s file,” combined with a comparative juror that further indicated pretext, all demonstrated that the prosecution was concealing racially discriminatory motives for striking jurors. Id. at 1749, 1754. Foster does not support a conclusion that the State’s exaggeration of Mays’s position here warrants a similar finding of discriminatory intent. Because reasonable jurists would not debate the state court’s reasonableness in denying Nelson’s claim that Mays was struck on account of race, we deny a COA on this claim. The State offered the following reasons at voir dire for striking Spivey, the second juror who Nelson discusses in his petition for a COA: He slept through [the judge's instructions at the initial meeting] and most of our time downstairs in the Central Jury Room. He denied arrests on his questionnaire. He actually had two, one in 1998 and one in 2010. He checked he did not want to serve on a jury because he did not believe the Defendant could get a fair trial. He also indicated he did not like jury service because he didn’t want to sit around all day and that he works a lot of forced overtime, so he did not think he wanted to be on the panel. And he had problems sitting in judgment of other people. Nelson contends on appeal that this was a mischaracterization of Spivey’s testimony. Specifically, he contests the State’s statement that Spivey “did not want to serve on a jury because he did not believe the Defendant could get a fair trial.” Spivey’s actual testimony was as follows: [Prosecution:] I want to refer you to something that you filled out on your questionnaire . . . . ["]Do you want to serve as a juror in this case,["] and you checked no and told us that you would give your reasons in the interview. [Spivey:] I counted the number of African-American males in the actual pool, and I believe it was like eight. . . . I don’t believe that’s a jury of a man’s peers. . . . I mean, it may look bad but it might turn out to be all right. And it looked bad to me that day. I believe that man don’t stand a chance. [Prosecution:] . . . do you feel like you can’t give us, as the State, a fair trial? [Spivey:] I can give you a fair trial. It’s just that it looked bad. It’s like having a nice steak served to you on a garbage can. The steak looked good, but you don’t want to eat on that garbage can. Nelson emphasizes that “Mr. Spivey stated that he could be fair; he expressed only a concern with whether the rest of the jury would be racially representative.” As he does when discussing Mays, Nelson relies on Foster to argue that this mischaracterization demonstrates pretext. We find this argument similarly unpersuasive. Though the State’s description of Spivey’s testimony was imprecise, this does not make it unreasonable for the state court to accept the State’s rationale as genuine, especially in light of the numerous other race- neutral reasons given for striking Spivey from the jury. Accordingly, we further deny a COA on Nelson’s claim that Spivey was struck in violation of Batson. V. Ineffective Assistance of Trial Counsel in Raising Batson Objection Nelson also seeks a COA on his related claim that his trial counsel was ineffective in arguing the Batson objection at voir dire. Trial counsel raised Baton objections in response to the State’s decisions to strike five minority jurors, and argued that the disproportionate number of strikes used to remove minorities jurists from the venire demonstrated a prima facie case. The trial court agreed, and asked the State to provide its race-neutral reasons, as discussed above. Consistent with the third step of a Batson challenge, the district court then advised defense counsel that “I think it now becomes your burden to show purposeful discrimination.” Nelson argues that his counsel was ineffective because, instead of offering comparators, pointing out the State’s misrepresentations of jurors’ testimony, or otherwise arguing that that these proffered reasons were mere pretext, counsel only noted that the State did not challenge three of the five for cause and then stated “I’ll let the record speak for itself.” Nelson did not exhaust this claim in state court, and cannot demonstrate cause and prejudice for his failure to do so because the underlying claim is not substantial. See Martinez, 566 U.S. at 14. As discussed above, Nelson cannot demonstrate that the peremptory strikes in question were motivated by the jurors’ race. Accordingly, Nelson cannot raise a substantial claim that he was prejudiced by his trial counsel’s alleged failure to adequately argue its Batson objections at the third step. See Eagle v. Linaham, 279 F.3d 926, 943 (11th Cir. 2001) (finding that there is prejudice from counsel’s failure to argue Batson claim when that claim “would have had a reasonable probability of success”); see also United States v. Kimler, 167 F.3d 889, 893 (5th Cir. 1999) (“An attorney’s failure to raise a meritless argument thus cannot form the basis of a successful ineffective assistance of counsel claim because the result of the proceeding would not have been different had the attorney raised the issue”). We deny a COA on this claim as well. VI. Denial of Motion to Stay and Abate Finally, Nelson appeals the district court’s denial of a stay and abatement of his federal proceedings to permit him to exhaust his ineffective assistance of counsel claims and a new claim that the State presented false testimony. See generally Rhines v. Weber, 544 U.S. 269 (2005). The district court did not address the substance of this motion, but held simply that “[i]n light of the court’s rulings in this memoranda opinion and order” in which it denied relief on all claims, “[n]o legitimate purpose would be served by granting the relief sought.” Because our determination on the merits of Nelson’s appeal, which we defer pending merits briefing, could affect the correctness of this ruling, we also defer consideration of the court’s denial of this motion until that time. *** For these reasons, we GRANT a COA on Nelson’s claim that trial counsel was ineffective for failing to investigate Springs’ and Jefferson’s alleged participation in Dobson’s murder. We defer consideration of the denial of funding in support of this claim until our decision on the merits of Nelson’s appeal. We also defer consideration of the denial of a stay to allow Nelson to exhaust the claim pending consideration of the foregoing. With respect to all other claims, a COA is DENIED, and the district court’s denial of investigative funding in support of these claims AFFIRMED.