PER CURIAM:
A Texas jury found Juan Castillo guilty of the capital murder of 19 year old Tommy Garcia, Jr. and sentenced him to death. After he was denied relief on direct appeal and in state habeas corpus proceedings, Castillo collaterally attacked his conviction and sentence through a federal habeas corpus petition. The federal district court denied him both habeas relief and a certificate of appealability (“COA”) in an exhaus-five 134 page opinion. Castillo now seeks a COA from this court under 28 U.S'.C. § 2253(c)(2). For the following reasons, we DENY a COA.
BACKGROUND
On the night of December 2 and early morning hours of December 3, 2003, Castillo and his girlfriend, Debra Espinosa, along with Castillo’s friend, Francisco Gonzales, and Gonzales’s girlfriend, Teresa Quintero, developed a plan to rob Tommy Garcia, Jr. Espinosa, who had been intimate with Garcia in the past, was to take Garcia to a secluded spot in a residential neighborhood in San Antonio for another sexual encounter. Castillo and Gonzales, masked and armed with guns, would storm the car and rob Garcia. Espinosa would play along as if she, too, was a victim. Quintero would be the get-away driver.
During the ensuing robbery, Castillo shot and killed Garcia.
Gonzales was arrested fleeing the scene and Espinosa was arrested shortly after. Implicated in the killing, both negotiated agreements to testify against Castillo. In exchange for Gonzales’s testimony, the State agreed to a charge of murder and a sentence of forty years in prison; for Espi-nosa’s testimony, the - State agreed to a charge of aggravated robbery and a sentence of forty years in prison.
Gonzales and Espinosa testified at Castillo’s trial that Castillo took the lead in planning the robbery and they saw him shoot Garcia.
Two of Gonzales’s family members testified that they saw Castillo, Gonzales, and Quintero leave Gonzales’s house together shortly before the murder in a car bor
rowed from one of the witnesses. Later that night, only Quintero came back to the house to return the car. In the days after the murder, both family members overheard, on separate occasions, Castillo incriminate himself as the shooter and talk about how he got rid of the evidence.
Two of Garcia’s friends testified that they were hanging out with Garcia when he received a call from Espinosa asking to meet up. Not long after Garcia left to meet Espinosa, one of the friends received a call from a hysterical Espinosa. She told him that Garcia had been shot.
A fellow jail inmate testified that Castillo admitted to him, while they were in the jail together, that he had murdered Garcia during a robbery.
Witnesses who lived near the murder scene testified that a car like the one Castillo and the others borrowed was seen fleeing the scene right after the shooting. A witness saw Castillo wearing, after the murder, a distinctive necklace that Garcia owned and wore the night he died.
On the strength of this and other evidence, the jury convicted Castillo of capital murder. Before the punishment phase, Castillo communicated to the court a desire to represent himself. The court determined Castillo was making a knowing and voluntary decision and allowed- him to represent himself during the punishment phase. Callahan and Harris, his appointed attorneys, remained as stand-by counsel.
During the punishment phase, the state presented substantial evidence of future dangerousness. Witnesses testified that Castillo brutally beat and threatened the mother of his child; shot at a man in a road rage incident; was arrested in possession of guns and bulletproof vests; committed numerous armed street robberies; bragged about committing home invasion robberies of drug dealers; and boasted about having shot, stabbed, and killed people in the past. Castillo did not cross-examine, nor did he offer evidence in the punishment phase. The jury found evidence of future dangerousness and no mitigating circumstances. The judge imposed the death penalty.
Castillo raised four points of error on direct appeal. His conviction and sentence were affirmed by the Texas Court of Criminal Appeals.
See Castillo v. State,
221 S.W.3d 689 (Tex.Crim.App.2007). Castillo sought a writ of habeas corpus before the state district court, which held an eviden-tiary hearing and denied relief in a 50 page opinion. The Texas Court of Criminal Appeals adopted the district court’s findings and conclusions and denied relief.
See Ex parte Castillo,
No. WR-70,510-01, 2012 WL 3999797 (Tex.Crim.App. Sept. 12, 2012) (per curiam).
Castillo sought a federal writ of habeas corpus. The federal district court denied an evidentiary hearing, denied him habeas relief, and denied a COA.
Castillo v. Stephens,
No. SA-12-CA-924-XR, 2014 WL 6090411 (W.D.Tex. Nov. 12, 2014).
DISCUSSION
The federal law of habeas corpus is “a guard against extreme malfunctions in the state criminal justice systems, not a substitute for ordinary error correction through appeal.”
Harrington v. Richter,
562 U.S. 86, 102-03, 131 S.Ct. 770, 786, 178 L.Ed.2d 624 (2011) (internal quotation marks omitted). Federal courts respect the conscientious labor of state courts and promote comity, federalism, and finality through the faithful application of the Antiterrorism and Effective Death Penalty Act of 1996 (“AEDPA”).
See Jimenez v. Quarterman,
555 U.S. 113, 121, 129 S.Ct. 681, 686, 172 L.Ed.2d 475 (2009).
AEDPA requires proof that the state court’s adjudication of a habeas claim “re-
suited in a decision that was contrary to, or involved an unreasonable application of, clearly established Federal law, as determined by the Supreme Court of the United States” or “resulted in a decision that was based on an unreasonable determination of the facts in light of the evidence presented in the State court proceeding.” 28 U.S.C. § 2254(d). A state court’s application of federal law may be reasonable for the purposes of § 2254(d) even though another court has applied or would have applied the same law to the same facts in a different manner.
Harrington,
562 U.S. at 101-02, 131 S.Ct. at 785-86.. Rather, a petitioner is only eligible for federal habeas relief if he shows that there is “no possibility fairminded jurists could disagree that the state court’s decision” is in conflict with the Supreme Court’s determination of federal law.
Id.
at 102, 131 S.Ct. at 786.
AEDPA bars appeal to this court without a COA. If the federal district court “reject[ed] the [petitioner's constitutional claims on the merits, this court will issue a COA only if the petitioner demonstrates that reasonable jurists could debate whether the district court’s resolution of his constitutional claims was correct, or could conclude that the issues presented ‘deserve encouragement to proceed further.’ ”
Williams v. Stephens,
761 F.3d 561, 566 (5th Cir.2014) (quoting
Miller-El v. Cockrell,
537 U.S. 322, 336, 123 S.Ct. 1029, 1039, 154 L.Ed.2d 931 (2003)),
cert. denied, —U
.S. -, 135 S.Ct. 1735, 191 L.Ed.2d 706 (2015). Where the district court rejected those claims on procedural grounds, a COA should issue when the prisoner shows that reasonable jurists “would find it debatable whether the petition states a valid claim of the denial of a constitutional right ... and whether the district court was correct in its procedural ruling.”
Slack v. McDaniel,
529 U.S. 473, 484, 120 S.Ct. 1595, 1604, 146 L.Ed.2d 542 (2000). Where the petitioner has been
sentenced to death, any doubts about whether a COA should issue “must be resolved” in the petitioner’s favor.
Williams,
761 F.3d at 566 (citing
Pippin v. Dretke,
434 F.3d 782, 787 (5th Cir.2005)). However, as the Supreme Court recently cautioned: “[T]he provisions of AEDPA apply with full force even when reviewing a conviction and sentence imposing the death penalty.”
White v. Wheeler,
—U.S.-, 136 S.Ct. 456, 462, 193 L.Ed.2d 384 (2015) (per curiam).
In sum, Castillo is not entitled to a COA unless.reasonable jurists could debate the district court’s conclusion that the state court reasonably applied federal law as determined by Supreme Court precedents to Castillo’s claims.
Castillo seeks a COA on 17 different grounds, fifteen of which assert that his trial counsel were unconstitutionally ineffective under the standard set forth in
Strickland v. Washington,
466 U.S. 668, 104 S.Ct. 2052, 80 L.Ed.2d 674 (1984). We begin there and note at the outset two things applicable to all of Castillo’s ineffective assistance of counsel (IAC) claims.
First, Castillo was appointed two attorneys. His first-chair attorney (Callahan) had been practicing criminal law for approximately 25 years when he was appointed to represent Castillo. Prior to that, he had represented 40 to 50 people, charged with capital murder and tried 10 death penalty cases. Castillo’s second-chair attorney (Harris) had been practicing criminal law for over 35 years, including 29 years in the Bexar County District Attorney’s Office. As a prosecutor, he had tried 20 death penalty cases. The record reveals that both attorneys are professional and very familiar with capital case procedures.
Second, at the habeas hearing, Castillo presented no evidence besides the testimo
ny of Callahan and Harris. Although Callahan and Harris were called by Castillo’s attorney, they did not concede ineffectiveness and vigorously contested it at many points. In short, the state court habeas record contains almost no evidence of what
might have been
presented at trial but for the ineffectiveness of Callahan and Harris.
To show that his trial counsel were constitutionally ineffective in the
state
habeas proceedings, Castillo was required to demonstrate that 1) his counsel’s performance was so deficient that they were no longer “functioning as the counsel guaranteed the defendant by the Sixth Amendment” and 2) he was prejudiced by these deficiencies because “there is a reasonable probability that, but for [the] errors, the result of the proceeding would have been different.”
Strickland,
466 U.S. at 687-88, 694, 104 S.Ct. at 2064, 2068. “Judicial scrutiny of counsel’s performance [under this standard] must be highly deferential.”
Id.
at 689, 104 S.Ct. at 2065. But in
federal
habeas proceedings another layer of deference is added and the “pivotal question is whether the state court’s application of the
Strickland
standard was unreasonable.”
Harrington,
562 U.S. at 101, 131 S.Ct. at 785. Any “reasonable argument that counsel satisfied” the “highly deferential”
Strickland
standard will bar federal habeas relief.
Id.
at 105, 131 S.Ct. at 788.
I. Claims regarding ineffective assistance of counsel
A. Trial counsel “abandonment”
Castillo asserts that he was denied effective assistance because Callahan and Hairris only visited him in jail three times. Because of this “abandonment,” they were unable to properly investigate his case or form a competent defense theory.
The state habeas court found that Castillo was not abandoned at all. Callahan and Harris communicated extensively with him. In addition to jail visits, they saw him at court settings and hearings. Most important, Castillo and his lawyers exchanged at least 59 letters discussing the case status and potential strategies, theories, and witnesses. The state court concluded that there was no deficient performance because Callahan and Harris communicated adequately with Castillo via letter, visited the jail when appropriate, and saw Castillo at numerous court appearances. The state court also concluded Castillo presented no evidence of prejudice; Castillo points to nothing in his trial or sentencing that would have gone differently if his counsel had visited him more often in jail.
The federal district court concluded this was a reasonable application of the
Strickland,
standard.
See, e.g., Murray v. Maggio,
736 F.2d 279, 282-83 (5th Cir.1984) (“[B]revity of consultation time between a defendant and his counsel, alone, cannot support a claim of ineffective assistance of counsel.”). Reasonable jurists could not debate the district court’s conclusion. Castillo is not granted a COA on this claim.
B. Counsel’s failure to interview a single prosecution witness
Castillo asserts that neither attorney interviewed a single prosecution wit
ness during their pre-trial investigation. This left them unprepared to cross-examine the witnesses and they ended up bolstering the prosecution’s case when their ignorant questions elicited damning information.
The state habeas court found that counsel’s performance was not deficient. Callahan and Harris reviewed the prosecution’s entire file and evidence and were not surprised by any of the state’s case at trial. They requested to speak with the prosecution accomplice witnesses, but the witnesses refused on advice of their counsel. Since none of the forensic evidence inculpated Castillo, counsel made the reasonable decision not to spend time interviewing those witnesses. Counsel dedicated significant time to trying to track down affirmative witnesses for Castillo’s case (both alibi and mitigation), and this was a strategic choice about how best to utilize their finite investigative time and resources. Counsel’s cross-examinations, though not always totally effective, were not unprepared and attempted to impugn the credibility of and confuse the prosecution witnesses. The state habeas court also found that Castillo was not prejudiced because he did not show what additional evidence could have been elicited by pretrial interviews of the non-accomplice witnesses.
The federal district court concluded this was a reasonable application of the
Strickland
standard.
See Bobby v. Van Hook,
558 U.S. 4, 11, 130 S.Ct. 13, 19, 175 L.Ed.2d 255 (2009) (per curiam) (acceptable under
Strickland
to forego witness interviews that are cumulative and “dis-tractive from more important duties”);
Mattheson v. King,
751 F.2d 1432, 1438 & n. 12 (5th Cir.1985) (inability to show “additional facts” that would have been uncovered dooms an IAC claim for failing to interview witnesses). Reasonable jurists could not debate the district court’s conclusion. Castillo is not granted a COA on this claim.
C. Counsel did not begin preparing until the last minute
Castillo asserts that his trial counsel did not begin preparing the case until the last possible minute and thus were unprepared and failed to adequately cross-examine the accomplice witnesses.
The state habeas court found that though many of counsel’s trial notes were not created until shortly before the trial, the evidence demonstrates that they began preparing the case in earnest once they were appointed. The 59 letters exchanged by counsel and Castillo belie this claim of last-minute work. Reducing their notes to writing shortly before trial was not deficient performance. Further, the state court found that Castillo failed to demonstrate prejudice, that is, what evidence would have been presented if Callahan and Harris had started their preparation earlier or conducted it more diligently.
The federal district court concluded this was a reasonable application of the
Strickland
standard.
See Van Hook,
558 U.S. at 9-10, 130 S.Ct. at 18. Reasonable jurists could not debate the district court’s conclusion. Castillo is not granted a COA on this claim.
D. Failure to conduct an adequate mitigation investigation
Castillo asserts that his attorneys failed to prepare for the punishment phase of his trial by locating and introducing mitigation evidence. In particular, he argues that counsel did not seek appointment of a mitigation expert or investigator and did not prepare evidence of his troubled upbringing and limited education, • In support, Castillo offered the state court an
affidavit from his mother, detailing the circumstances of his youth.
The state court found no deficient representation in preparing mitigation evidence for the punishment phase. Callahan and Harris got stipulations from the prosecution to allow introduction of Castillo’s educational records and part of his federal pre-sentencing report, which detailed how his father abandoned the family. They had the court appoint a mental health expert to testify regarding dangerousness and mitigation. They attempted to contact Castillo’s mother and other family members to testify, but were, unsuccessful in locating them. Significantly, Castillo was very little help in locating his mother or other family members; he was adamant that he did not want his mother to testify and claimed just months before trial not to know how to contact her.
See Sonnier v. Quarterman,
476 F.3d 349, 362 (5th Cir. 2007) (“[Wjhen a defendant blocks his attorney’s efforts to defend him, including forbidding his attorney from interviewing his family members for purposes of soliciting their testimony as mitigating evidence during the punishment phase of the trial, he cannot later claim ineffective assistance of counsel”) (quoting
Roberts v. Dretke,
356 F.3d 632, 638 (5th Cir.2004)). The court concluded this level of preparation was not deficient.
Moreover, the state court concluded that Castillo was not prejudiced, because Castillo failed to show what additional mitigation evidence could have been discovered through a more thorough investigation. As Castillo admits in his COA application, much of the information discussed in his mother’s affidavit was contained in the mental health expert’s report and likely would have been included in the expert’s testimony.
Rejection of this claim is reinforced by the fact, emphasized by the state court, that Castillo represented himself at punishment and presented no evidence on his own behalf. He decided not to present the educational records, the pre-sentencing report, and the' mental health expert’s testimony. Castillo’s then-standby counsel cannot be faulted for Castillo’s own trial choices after he elected self-representation.
See Nixon v. Epps,
405 F.3d 318, 325-26 (5th Cir.2005) (“A defendant cannot block his counsel from attempting one line of defense at trial, and then on appeal assert that counsel was ineffective for failing to introduce evidence supporting that defense.”).
The federal district court concluded that the state courts reasonably applied the
Strickland
standard. , Castillo is not granted a COA on this claim.
E. Defective, opening statement was constitutionally ineffective
Castillo asserts that Harris’s opening statement was constitutionally ineffective because it failed to “present a cogent defense theory that emphasized the known strengths of Castillo’s case.”
In the state habeas court, Castillo argued both that the opening statement failed to emphasize the evidentiary strengths of his case
and
that it was inconsistent with the closing argument. The state habeas court found that the opening statement was not inconsistent with the closing argument, but did not address the theory that the statement was ineffective by failing to call the jury’s attention to certain pieces of evidence. Before the federal district court, Castillo shifted his criticism to counsel’s failure to focus on the strengths of Castillo’s case. The federal district court found that the state court’s conclusion that the opening statement was not constitutionally ineffective was a reasonable application of the
Strickland
standard.
Reasonable jurists could not debate the federal district court’s conclusion, despite the state court’s not fully addressing Castillo’s argument about the opening statement. State courts need not explain their habeas decisions to be entitled to AEDPA deference.
See Harrington,
562 U.S. at 98, 131 S.Ct. at 784. Federal courts reviewing state court convictions under § 2254 must focus “on the ultimate legal conclusion that the state court reached and not whether the state court considered and discussed every angle of evidence.”
Neal v. Puckett,
286 F.3d 230, 246 (5th Cir.2002) (en banc) (per curiam).
The opening statement emphasized Castillo’s most critical defense theory: that the accomplice witnesses were not credible. Castillo’s application before this court recognizes this fact: “This case had one chance for a defense victory — to destroy the credibility of the accomplices.” This choice of approach was. reasonable strategy under
Strickland. See Yarborough v. Gentry,
540 U.S. 1, 7-9, 124 S.Ct. 1, 5-6, 157 L.Ed.2d 1 (2003) (per curiam) (“When counsel focuses on some issues to the exclusion of others, there is a strong presumption that he did so for tactical reasons rather than through sheer neglect.”). Moreover, the opening and closing did not conflict. The closing statement suggested that the jury might vote for a lesser included offense of murder if they found Castillo killed Garcia. In light of the evidence and the capital murder charge, this was not an unreasonable strategy.
See Florida v. Nixon,
543 U.S. 175, 191-92, 125 S.Ct. 551, 563, 160 L.Ed.2d 565 (2004) (unanimous) (conceding guilt to focus on avoiding death penalty not per se ineffective; concession was a reasonable strategy in light of strong evidence against defendant).
In sum, Castillo is not granted a COA on this claim.
F. Bolstering the prosecution’s case through ineffective cross-examination
Castillo asserts that Callahan and Harris ineffectively bolstered the prosecution’s case through their cross-examination. He reports specific instances where prosecution witnesses gave unfavorable answers to their questions. Castillo’s habeas petition contains several “model” cross-examinations of key witnesses to demonstrate what more effective and prepared cross-examinations may have looked like.
The state habeas court concluded that Callahan’s and Harris’s cross-examinations were not constitutionally deficient. They used cross-examination to elicit evidence and many statements favorable to Castillo. In particular, they were able to stress the motivations of various witnesses to lie and the lack of physical evidence linking Castillo to the crime. Even where their cross-examination strategy was unsuccessful, it was still reasonable. A few cherry-picked examples of mistakes or bad answers over the course of a long trial do not amount to constitutionally ineffective lawyering. Speculating about the effect of tinkering with the cross-examination questions is exactly the sort of hindsight that
Strickland
warns against. 466 U.S. at 689, 104 S.Ct. at 2065.
The district court found that the state court reasonably applied the
Strickland
standard to this claim and added that even if the cross-examinations were ineffective, there was no showing how a different cross-examination would have created a reasonable probability of a different outcome. Reasonable jurists could not debate the district court’s conclusion. The Constitution, demands reasonable competence, not perfect advocacy.
See Maryland v. Kulbicki,
— U.S.-, 136 S.Ct. 2, 5, —
L.Ed.2d-(2015) (per curiam). Castillo is not granted a COA on this claim.
G. Attorneys’ concession of guilt
Castillo next asserts that Callahan conceded guilt in closing arguments based on a misunderstanding of the law. Castillo argues that Callahan told the jury that Castillo only attempted to commit robbery but did not actually rob Garcia. Thus, if the jury found he killed Garcia, they should only find him guilty of murder. This was essentially an admission that Castillo was guilty of capital murder. Castillo also argues that Callahan’s other possible theory — he told the accomplices they were going to rob Garcia in order to lure him to the scene, but Castillo only intended to kill Garcia, not to rob him — was also a concession of capital murder guilt.
The state habeas court found that Callahan never conceded guilt. Recognizing that the jury might find strong evidence of Castillo’s culpability, Callahan offered the jury an alternative that would spare Castillo’s life. He told the jury that
if
they believed that Castillo killed Garcia, they should find him guilty of murder instead of capital murder. He told the jury to consider the evidence that the medical examiner found Garcia in possession of a large amount of money as evidence that no robbery was planned or occurred. During the trial, Castillo’s counsel had elicited other evidence rebutting the prosecution’s claim that the intention was to rob Garcia or that Garcia was actually robbed. Evidence had also been elicited that Garcia had been intimate with Espinosa, Castillo’s girlfriend, in the past and that Castillo and Garcia had a prior altercation and didn’t like each other. Callahan testified, without contradiction, at the habeas hearing that his strategy was to appeal to this evidence. The state court found this strategy and Callahan’s performance to be reasonable and not deficient.
The district court concluded that the state court reasonably applied the
Strickland
standard. Reasonable jurists could not debate the district court’s conclusion. Decisions to concede guilt or argue for a lesser-included offense are matters of strategy.
See Haynes v. Cain, 298
F.3d 375, 382-83 (5th Cir.2002) (en banc). In light of the evidence, it was a reasonable strategy to assert that no robbery occurred or was planned. This was reasonable in light of the circumstances of this death penalty case, where there was
very
substantial evidence of future dangerousness.
H. Failure to prepare the lone defense witness
Castillo asserts that Callahan and Harris failed to prepare their lone witness and were ineffective for even putting him on the stand. Castillo contends that Ralph Pedrigone, a jail house snitch who said he overheard Gonzales (one of the accomplice witnesses) admit to killing Garcia, had nothing substantial to present and was easily discredited through his extensive criminal record.
The state court concluded there was no deficient performance in the preparation and presentation of Pedrigone’s testimony. Callahan did not interview or prepare Pe-drigone until the day-before he testified, but this was not for lack of trying: he had been trying to get Pedrigone transferred to Bexar County for some time but there was trouble with the transfer. Furthermore, Pedrigone’s testimony was simple; Castillo offered no evidence that Callahan was deficient for not preparing him more. The state court finally concluded there was no prejudice because Castillo did not es
tablish how there was a reasonable probability that more preparation would have affected the outcome of his trial.
The federal district court concluded this was a reasonable application of the
Strickland
standard. Pedrigone testified to the fact that Gonzales said he killed Garcia. It was a reasonable strategy to try to shift blame for Garcia’s murder onto the accomplice witnesses. That the prosecution was able to devalue this testimony through skillful cross-examination does not change the reasonableness of the calculation to present Pedrigone’s testimony.
See Strickland,,
466 U.S. at 689, 104 S.Ct. at 2065 (“A fair assessment of attorney performance requires that every effort be made ... to evaluate the conduct from counsel’s perspective at the time.”). Reasonable jurists could not debate the district court’s conclusion. Castillo is not granted a COA on this claim.
I. Ineffective jury selection
Castillo asserts that Callahan and Harris were ineffective during jury selection. Castillo argues that Callahan should have agreed to the prosecution’s offer to strike for cause one venireperson (Doris Cedillo) who overheard a brief comment by the trial judge that they were starting late due to a mix-up at the jail. Castillo also argues that Callahan should have struck another venireperson (Arthur Carter) who made a wrong statement of the law during voir dire.
The state court found that there was no deficient performance in jury selection. There was no indication in the record how the judge’s fleeting comment would prejudice the juror against Castillo. It was not implied that Castillo was the source of the mix-up or delay. Furthermore, the juror who overheard the brief comment gave answers indicating she might be reluctant to impose the death penalty, making her a potentially favorable defense juror. The juror who briefly misunderstood the law was immediately corrected and there was no indication that he did not understand the correction. Without further explanation why these jurors should have been struck, Castillo’s claim falls within
Strickland’s
presumption of reasonableness and strategy. See
id.
(“[C]ourt[s] must indulge a strong presumption that counsel’s conduct falls within the wide range of reasonable professional assistance.”).
The district court found that the state court reasonably applied the
Strickland
standard. Reasonable jurists could not debate the district court’s conclusion. Castillo is not granted a COA on this claim.
J. Attorneys did not properly represent him in connection with his desire to represent himself
Castillo asserts that Callahan and Harris rendered ineffective assistance of counsel regarding his desire to represent himself. He has waived this claim, however, because the entirety of his briefing on it is as follows: “This issue is discussed in this brief as a separate claim [Castillo’s
Faretta
claims]. For brevity, this issue incorporates by reference the facts and law set out in the separate claim.” Because the separate claim addresses alleged errors by the trial judge, it does not address what Castillo believed was the ineffective assistance of his counsel.
See Loden v. McCarty, 778
F.3d 484, 501 (5th Cir.) (inadequately briefed arguments are waived),
cert. denied,
—U.S.-, 136 S.Ct. 402, 193 L.Ed.2d 315 (2015). (We analyze the free-standing
Faretta
claim below.)
We note, however, that the federal district court found that the state court made reasonable factual determinations in light of the evidence presented and reasonably
applied the law of
Strickland
and
Faretta.
Reasonable jurists could not debate the district court’s conclusions. In any event, Castillo is not granted a COA on this claim.
K. Punishment phase ineffective assistance of counsel
Castillo asserts several ineffective assistance of counsel claims based on the punishment phase. Though these have some pretrial dimensions, they also fault Callahan and Harris for mistakes made during the punishment phase. As noted above, Castillo represented himself during the punishment phase, and neither requested assistance from stand-by counsel nor attempted to present punishment evidence of his own or challenge the state’s evidence.
The Supreme Court has strongly and sensibly suggested that no ineffective assistance of counsel claims can lie where a defendant undertakes self-representation.
See Faretta v. California,
422 U.S. 806, 834 n. 46, 95 S.Ct. 2525, 2541 n. 46, 45 L.Ed.2d 562 (1975) (“[W]hatever else may or may not be open to him on appeal, a defendant who elects to represent himself cannot thereafter complain that the quality of his own defense amount to a denial of ‘effective assistance of counsel.’ ”) (quoting
United States v. Dougherty,
473 F.2d 1113, 1124-26 (D.C.Cir.1972)). As to stand-by counsel, the Court has said only that their role should be limited to participation either outside the jury’s presence or “with the defendant’s express or tacit consent.”
McKaskle v. Wiggins,
465 U.S. 168, 188, 104 S.Ct. 944, 956, 79 L.Ed.2d 122 (1984). With no federal law clearly established by the Supreme Court delineating the duties of stand-by counsel, it was not unreasonable for the state habeas court to find that in this case no punishment phase ineffective assistance of counsel claims could lie.
Cf. Wood v. Quarterman,
491 F.3d 196, 204 (5th Cir.2007) (“The record in this case reflects that [the defendant] clearly and repeatedly instructed his . trial counsel to sit idly throughout the sentencing phase of his trial. On these facts, the state court’s decision holding that [the defendant] could not show that counsel performed deficiently was not an unreasonable application of
Strickland.”).
Even if we consider Castillo’s punishment phase ineffective assistance of counsel claims, however, none deserves a COA.
1. Forensic psychologist
Castillo’s contends that Callahan and Harris never sought appointment of a mitigation expert and failed to prepare the forensic psychologist to testify in the punishment phase. This claim also alleges that Callahan did not adequately explain to Castillo what the forensic psychologist would testify to or prepare any advance questions for Castillo to use.
On the contrary, the state habeas court found that Callahan retained a forensic psychologist who was present and prepared to testify at the punishment phase of trial regarding future dangerousness and the facts of Castillo’s background. Castillo made the decision not to call the expert, and he had been provided the expert’s reports and an explanation by Callahan in several letters. Thus, the state court concluded that there was no deficient performance by Castillo’s attorneys, either before trial or during punishment.
The federal district court found this to be a reasonable application of the
Strickland
standard. Reasonable jurists could not debate the district court’s conclusion. Castillo is not granted a COA on this claim.
2. Failure to interview his family
Castillo asserts that Callahan and Harris failed to interview and use his fami
ly members at punishment. Much of this is premised on their failure to communicate with Castillo’s mother or contact family members regarding testifying. According to the motion, Castillo’s mother made numerous efforts to contact Callahan, but he never returned her calls. Castillo argues that
Rompilla v. Beard, 545 U.S. 374, 125
S.Ct. 2456, 162 L.Ed.2d 360 (2005), requires attorneys to “contact family members to get them to testify ,.. despite what the capital defendant wants.”
The state habeas court found that Callahan and Harris made reasonable attempts to contact Castillo’s family and mother despite Castillo’s lack of cooperation. Further, Callahan had access to much of the evidence that Castillo’s mother likely would have furnished in his educational records, the federal PSR, and the proposed testimony of the forensic psychologist. But Castillo himself failed to present this evidence during the punishment phase. The state court found that Castillo’s counsel did not render deficient performance. The state court also concluded there was no prejudice. Aside from his mother’s proposed testimony, none of the family members testified at the habeas hearing or submitted affidavits to the ha-beas court.
The district court found this a reasonable application of the
Strickland
standard and also stated that there was likely no prejudice since the mitigating evidence Castillo proposed was fairly weak and the evidence of Castillo’s future dangerousness was very strong. Reasonable jurists could not debate the district court’s conclusion.
Finally, Castillo misapprehends the holding in
Rompilla.
In that case, the Supreme Court discussed counsel’s responsibility to pursue a mitigation investigation despite, among other things, the habeas petitioner’s active obstruction before concluding that there was “room for debate” on this issue — in other words, reasonable jurists could disagree. 545 U.S. at 383, 125 S.Ct. at 2463. The Court instead decided
Rompilla
on other grounds.
Id.
3. Failure to object to sleeping jurors
Castillo’s eleventh claim is that Callahan and Harris were ineffective when they failed to object to jurors who were sleeping. Castillo argues that Callahan referenced sleeping jurors during his closing argument, and references to a juror sleeping and nodding off appear in his trial notes. Furthermore, Castillo’s habeas application contains notes which he purportedly passed to Callahan which reference one of the jurors sleeping and says he thinks an alternate should step in.
The state habeas court, which was also the trial court, had no independent recollection of observing any juror sleeping — a relevant fact in the AEDPA analysis.
See Schriro v. Landrigan,
550 U.S. 465, 476, 127 S.Ct. 1933, 1941, 167 L.Ed.2d 836 (2007) (calling the state habeas judge “ideally situated” to make a sentencing assessment because it was the same judge who presided over sentencing). Callahan testified without contradiction that his closing argument reference to jurors sleeping or not paying attention was an attempt to seize the conscience of jurors who had not been paying perfect attention during the trial. Callahan’s notes only refer to a juror sleeping during the state’s punishment case, while Castillo’s personal notes do not indicate when he observed a juror sleeping, and he did not take the stand to explain them.
The state court accordingly found that “there is no evidence concerning when the jurors slept, if at all, or during which
portions of the trial. Mr. Callahan cannot recall if the jurors were sleeping or simply not paying' attention.” • At bottom, the state court found the record “insufficient to establish the possibility that any juror slept during the defendant’s cross-examination of the state’s witnesses or during the defense’s case in chief.” Because there was insufficient evidence to establish that jurors slept during the guilt-innocence phase, the state court concluded that Callahan and Harris were not constitutionally ineffective. .No ineffective assistance claims can lie against Callahan and Harris for failing to object to sleeping jurors during the punishment phase when they were merely stand-by counsel.
The federal district court found that the state court made a reasonable determination of the facts in light of the evidence presented, and reasonably applied the
Strickland
standard, to both the trial and punishment phase ineffective assistance claims.
Reasonable jurists could not debate the district' court’s conclusion on either the factual or legal determination during the guilt-innocence phase. Castillo has no ineffective assistance claims for the punishment phase. Castillo is not granted a COA on this claim.
L. “Cumulative” ineffective assistance of counsel
Castillo’s fifteenth claim is that he was abandoned at almost every turn during trial and sentencing and these errors added up to a conviction and sentence “not worthy of confidence.”
The state court found that the totality of the alleged deficiencies in representation did not add up to ineffective assistance.
The federal district court concluded this was a reasonable application of the
Strickland,
standard — ineffective assistance could not result from the cumulative total of reasonable decisions and actions. Reasonable jurists could not debate the district court’s conclusion.
See, e.g., Miller v. Johnson,
200 F.3d 274, 286 n. 6 (5th Cir. 2000) (denying a COA on a cumulative ineffective assistance theory where no single error by itself rose to the level of ineffectiveness). Castillo is not granted a COA on this claim.
II. Claims regarding self-representation
Castillo contends that the trial court committed error by 1) denying his right to self-representation under the Sixth Amendment before trial when he contacted the court seeking to replace his attorney and 2) allowing him later to represent himself when he was obviously depressed and would not have adequate time to prepare.
The state habeas court adjudicated this claim on the merits, but the Texas Court of Criminal Appeals found that it was pro-eedurally barred as a record-based claim that should have been brought during direct appeal.
The federal district court concluded that the claims were procedurally barred. Reasonable jurists would not be able to debate that the “district court was correct in its procedural ruling.”
Slack,
529 U.S. at 484, 120 S.Ct. at 1604. Castillo did not raise the self-representation issues on direct appeal. Whether the trial court erred in alternatively forbidding and then allowing Castillo to represent himself are record-based claims that should have been raised on direct appeal. Failure to raise a record-based claim on direct review in the Texas courts is a regularly applied ground for preventing state habeas consideration of that claim. It is also an adequate and independent state ground barring review in federal habeas proceedings.
See, e.g.,
Dorsey v. Quarterman,
494 F.3d 527, 532 (5th Cir.2007). Castillo does not attempt to argue the cause and prejudice necessary to defeat that bar, nor does he contend that a miscarriage of justice would result if the claim is not considered on its merits.
See id.
(citing
Coleman v. Thompson,
501 U.S. 722, 750, 111 S.Ct. 2546, 2565, 115 L.Ed.2d 640 (1991)).
We also note that, in the alternative, the district court conducted a lengthy analysis of the merits of Castillo’s self-representation claims. The district court found that the state court made reasonable factual determinations in light of the evidence presented and reasonably applied the law of
Faretta.
Reasonable jurists could not debate the district court’s conclusions. In any event, Castillo is not granted a COA on this claim.
III. Perjured testimony
Castillo’s final claim is that his due process rights were violated when he was convicted on the basis of perjured testimony, regardless whether the prosecutor knew the testimony was false. Gerardo Gutierrez, at one time Castillo’s fellow county jail inmate, provided an affidavit that he had testified falsely that Castillo admitted robbing and shooting the victim.
This claim was not presented to the state courts. The federal district court found the claim procedurally defaulted under Texas’ abuse of the writ doctrine. Alternatively, the court found this claim meritless because no federal due process violation occurs when false testimony is used to secure a conviction unless the government
knowingly
used false testimony.
See, e.g., Kinsel v. Cain,
647 F.3d 265, 271-72 (5th Cir.2011). Castillo does not allege the state knew this testimony to be false.
The federal district court held correctly that under Texas’ regularly and strictly applied abuse of the writ rule, such a claim would now be procedurally barred from state habeas review.
See, e.g., Nickleson v. Stephens,
803 F.3d 748, 754 (5th Cir. 2015). Reasonable jurists could not debate the district court’s procedural ruling.
In his federal habeas petition and before this court in his initial COA application, Castillo, arguing only the merits, has made no attempt
to
demonstrate cause or prejudice to defeat the procedural bar. Castillo’s inadequate briefing has waived any challenge to the procedural default or claim to cause or prejudice.
See, e.g., Canales v. Stephens,
765 F.3d 551, 577-78 (5th Cir.2014).
CONCLUSION
For the foregoing reasons, Castillo’s application for a COA is DENIED.