* Pursuant to 5 T H C IR . R. 47.5, the court has determined that this opinion should not be published and is not precedent except under the limited circumstances set forth in 5 TH C IR . R. 47.5.4. United States Court of Appeals Fifth Circuit FILED July 15, 2003 Charles R. Fulbruge III Clerk IN THE UNITED STATES COURT OF APPEALS FOR THE FIFTH CIRCUIT

No. 01-11243 JAMES VERNON ALLRIDGE, Petitioner-Appellant, versus JANIE COCKRELL, TEXAS DEPARTMENT OF CRIMINAL JUSTICE, INSTITUTIONAL DIVISION, Respondent-Appellee. Appeal from the United States District Court for the Northern District of Texas (96-CV-271) Before HIGGINBOTHAM, WIENER, and BARKSDALE, Circuit Judges.

WIENER, CIRCUIT JUDGE. * Having been convicted of capital murder in Texas and sentenced to death, Petitioner James Vernon Allridge is before us by virtue of a certificate of appealability (COA) granted by the district court in connection with Allridge§ 2254.

We deny all relief sought.

I. BACKGROUND In 1987, a Texas jury convicted Allri dge of capital murder, after which the trial court imposed a sentence of death. See Allridge v. State , 850 S.W.2d 471, 475 (Tex. Crim. App. 1991). In this § 2254 habeas action, the district court granted Allridge a COA on two claims: “(1) PetitionerÂ’s claim that the trial court violated the Witherspoon [v. Illinoi s , 391 U.S. 510 (1968)] doctrine when it granted th e StateÂ’s challenge for cause against prospective juror Martin Osborn, and (2) PetitionerÂ’s ineffectiveassistance -of-counsel claim.” As AllridgeÂ’s appellate brief is limited to these two claims, the following recitation of the procedural history of this case is limited to issues relevant to them.

A. Trial 1. Voir Dire During voir dire, the State challenged venireman Martin Osborn for cause on the ground that his doubts about the propriety of the death penalty would “substantially impair” the performance of his duties as a juror in accordance with the court’s instructions and his oath. Although the trial court initially denied the State’s motion, it eventually granted the motion after further testimony from Osborn. This testimony will be discussed below at length in addressing Allridge’s Witherspoon arguments. 2. Guilt/innocence phase The trial evidence, as related by the Texas Court of Criminal Appeals (“CCA”), may be summariz ed as follows: On the night of February 3, 1985, Allridge and his older brother, Ronald, left their Fort Worth apartment with the intention of robbing a Circle K convenience sto re. Allridge , 850 S.W.2d at 476. Allridge was carrying a semi-automatic pistol, and Ronald drove Allridge’s car. Id . Allridge had previously worked at the Circle K, was familiar with the store’s procedures, and knew where the combination to the safe was kept. I d . He also knew the clerk on duty, Brian Clendennen, having worked with him before. I d . At about midnight, Ronald dropped Allridge off around the co rner from the targeted store. I d . Clendennen had already closed the store, but admitted A llridge when he asked for change to use the phone. I d .

Clendennen made change, and Allridge “pretended to use the phone and left to rejoin Ronald.” I d . Ronald accused Allridge of “chickening out” and dropped Allridge off at the store again. I d .

Clendennen again let Allridge into the store, but this time Allridge pulled his gun and forced Clendennen into the storeroom.

Id . After tying Clendennen’s hands behind his back, Allridge emptied the sa fe. Id . Allridge heard sounds coming from the storeroom and discovered that Clendennen had moved. Id . He made Clendennen “get back on his knees,” then shot him twice in the back of th e head. I d . Allridge and Ronald left, and Clendennen died from the gunshot wounds the next day. Id . 3. Punishment phase At the punishment phase, the State sought an affirmative finding on the “second special issue,” which addressed “future dangerousness” or whether the defendant would commit criminal acts of violence that would constitute a continuing threat to society.

Id . at 487 (citing T EX . C ODE C RIM . P ROC . art. 37.071). The State presented evidence of several armed robberies committed by Allridge and Ronald in the two months following after their robbery of the Circle K and murder of Clendennen. S ee id . at 487-88. In his defense, Allridge attempted to show that, since childhood, he had been intimidated and dominated by Ronald. I d . at 488. In support, he called a psychologist, Dr. Richard Schmitt, to testify that Allridge was intelligen t and competent and not psychotic or sociopathic. I d . The jury made affirmative findings as to both special issues, and the court sentenced Allridge to death.

B. Direct appeal On direct appeal, Allridge raised 21 points of error. S ee Allridge , 850 S.W.2d at 476. In his first point, Allridge contended that the trial court had violated the doctrine of Witherspoon v. Illinois , 391 U.S. 510 (1968 ), by excluding venireman Osborn for cause on t he ground that his views on the death penalty would adversely affect his impartiality. I d . at 477.

Allridge argued that Osborn ’s answers during voir dire reflected that he could follow the law and not be controlled by his feelings about the death penalty. I d . After recounting Osborn’s testimony, the CCA determined that “it appear[ed] that Osborn was torn between the obligation to honestly comply with his oath as a juror and his strong feelings in opposition to the death penalty.” Id . at 478.

That court concluded: OsbornÂ’s answers that those feelings would influence his assessment of the evidence at punishment and affect his abi lity to comply with his oath support the trial courtÂ’s determination that Osborn was substantially impaired in his ability to perform his duties as a capital juror in accordance with his instructions and oath.

Id . The CCA affirmed AllridgeÂ’s conviction and sentence, id. at 497, and the United States Supreme Court denied AllridgeÂ’s application for a writ of certiorari. A llridge v. Texas , 510 U.S. 831 (1993).

C. State postconviction proceedings In 1994, represented by a new attorney, Allridge filed a state postconviction application, raising a newly-discovered-evidence claim. He argued that the “new” evidence consisted of statistical studies showing a strong correlation between the Jehovah’s Witnesses religion, of which Allridge was a practitioner, and the commission of crimes. Alternatively, Allridge contended that his trial c ounsel had performed ineffectively by failing to develop this exculpatory and mitigating evidence at trial. Allridge also filed first a nd second amended applications raising additional arguments. One was that the State had made improper jury arguments when it stated that Allridge’s counsel had acted unet hically by having Dr. Schmitt testify about Allridge without having conducted written psychological tests in person.

The next fall, the state trial court issued findings of fact and conclusions of law recommending that Allridge§ 2254 habeas proceedings.

D. Federal habeas corpus proceedings Following his appointment, newly appointed federal habeas counsel filed Allridge§ 2254 petition. In it Allridge contended that (1) the trial court violated the Witherspoon rule when it granted the StateÂ’s challenge to venireman Osborn for c ause; (2) the court violated AllridgeÂ’s due process rights when it overruled his challenges for cause to three veniremen who allegedly would refuse to consider particular kinds of mitigating evidence at the punishment phase; and (3) his trial counsel had performed ineffectively at the punishment phase by (a) fa iling to present expert testimony about the nexus between Allri d geÂ’s so-called indoctrination in the Jehovah Â’s Witnesses (which he refers to as cult-like), and his criminal behavior, and (b ) allowing his psychological expert, Dr. Schm itt, to base his testimony on the results of written tests that Allridge filled out in his jail cell without observation or supervision.

The State filed an answer, contending that all of AllridgeÂ’s claims were meritless. I d . The State also argued that AllridgeÂ’s second ineffective-assistance claim had not been exhausted in the state courts and was thus procedurally defaulted. Allridge did not address procedural default in reply.

The magistrate judge issued a repor t recommending that Allridge§ 2254 petition be denied, concluding that, by upholding the StateÂ’s challenge to venireman Osborn, the trial court implicitly found that his ability to act as a j uror would be substantially impaired by his qualms about capital punishment.

This finding was entitled to a pres umption of correctness, which Allridge had failed to rebut by clear and convincing evidence. The magistrate judge also concluded that AllridgeÂ’s claim t hat trial counsel performed ineffectively by failing to ensure that the written psychological tests were supervised by Dr. Schmitt had not been developed in state court and was thus procedurally defaulted; and that Allridge had not shown cause and prejudice to excuse this default. The magistrate judge concluded that, in any event, the ineff ectiveness claim was meritless because Allridge could not demonst rate prejudice: There was no evidence that anyone other than Allridge had completed the tests and there was extensive additional evidence of AllridgeÂ’s future dang erousness. The magistrate judge also recommended that AllridgeÂ’s other claims be denied as meritless.

Allridge filed lengthy objections to this report and recommendation. In a footnote, Allridge argued that the second ineffectiveness claim was exhaus ted because the CCA adopted the s tate habeas trial courtÂ’s finding that defense counsel had m ade reasonabl e strategic choices about how to present all possible mitigating factor s in consultation with Dr. Schmitt and rendered effective assistance under the totality of the circumstances. The State filed a responding brief that addressed these objections.

The district court adopted the magistrate judge§ 2254 petition. The court issued its own findings and conclus ions, which closely resembled those of the magistrate judge.

Within 10 days following the entry of judgment, Allridge filed a F ED . R. C IV . P. 59(e) motion to alter or amend the judgment. He continued to maintain that his ineffective-assistance claim regarding the written tests had been exhausted because, in his state application, he had included a general allegation that counsel was ineffective in presenting psychiatric mitigating evidence. The court denied AllridgeÂ’s Rule 59(e) motion, stressing that even if the claim at issue were not procedurally defaulted, it was meritless.

Allridge timely filed a notice of appeal, which also functioned as a motion for a certificate of appealability (“COA”).

As noted, the district court granted Allridge a COA on his Witherspoon claim regarding venireman Osborn an d on his claim of ineffective assistance of counsel. As to that latter claim, the district court did not specify that it was granting COA as to both of AllridgeÂ’s ineffective assistance claims or that the question of procedural default remained at issue.

II ANALYSIS Under 28 U.S.C. § 2253(c)(2), our review is limited to the issues on which the district court granted COA. K iser v. Johnson , 163 F.3d 326, 327 (5th Cir. 1999). Although a habeas appellant may obtain review of issues not certified by the district court if he expressly requests from us, and we grant, a COA on such issues, see Ott v. Johnson , 192 F.3d 510, 512 n.6 (5th Cir. 1999) , Allridge does not seek review of non-certifi ed issues. Furthermore, A llridge has abandoned his claim that counsel per formed ineffectively by failing to present statistical evidence regarding JehovahÂ’s Witnesses and crime; he does not brief this issue on appeal. See Dowthitt v. Johnson , 230 F.3d 733, 742 n.6 (5th Cir. 2000).

Allridge filed his § 2254 petition on April 15, 1996, just before the April 24, 1996, effective date of the Antiterrorism and Effective Death Penalty Act (§ 2254(d), we must accord a presumption of correctness to all f indings of fact if they are supported by the record. I d .; see former § 2254 (d)(1)-(d)(8) (listing eight exceptions to this rule). The pre-AEDPA standards do not require a federal court to defer to the state courtsÂ’ legal conclusions. S ee Valdez v. Cockrell , 274 F.3d 941, 949 (5th Cir. 2001), cert. denied , 123 S. Ct. 106 (2002) .

A. Exclusion of venireman Osborn Allridge contends that venireman OsbornÂ’s dismissal was improper under Adams v. Texas , 448 U.S. 38 (1980), because the CCA relied on impermissible reasons for u pholding the dismissal.

Allridge argues that, in its opinion on direct appeal, the CCA made “implicit ” findings of fact that Osborn’s testimony that his feelings about the death penalty would “temper” his views of the evidence meant that those feelings would “influence” and “affect” his assessment of the evidence. Allridge insists that we must defer to those appellate findings of fact b ecause they are support ed by the record, even if we believe that the trial court gave the word “temper” a different meaning and based its ruling on a different ground. A llridge urges that, under Adams , the findings of the CCA that Osborn’s feelings would merely “influence” and “affect” his view of the evidence were constitut ionally insufficient to support his dismissal. Allridge argues that the Adams violation is clear, emphasizing that Osborn never said that he could not participate in returning a verdict that would require the judge to impose the death penalty.

A llridge acknowledges that the CCA and the federal district court cited other grounds for disqualification that might have been supported by the record. He maintains, however, that these other grounds cannot be used to uphold the exclusion of Osborn because they are based on facts that did not constitutionally authorize the exclusion.

The State counter s that Allridge is improperly raising his contention regarding the state appellate court§ 2254(d) standard of review should be applied specifically to the trial courtÂ’s findings. 1. The applicable law In Witherspoon , a direct appeal f rom a criminal conviction, the Supreme Court held that a death sentence cannot be carried out if it followed the exclusion of a venireman soley because he voiced general objections to the death penalty or expressed conscientious or religi ous scruples against its infliction. W itherspoon , 391 U.S. at 521-22. W itherspoon concerned (1) an Illino is capitalsentencing system in which the jury had broad discretion to impose the death penalty, and (2) an Illinois statute that permitted the prosecution to challenge for cause any venireman who had “conscientious sc ruples against capital punishment.” S ee id . at 512, 519.

In Adams , also a direct appeal, the Supreme Court held that a ve nireman may be excused if his § 12.31(b) stated that: Prospective jurors shall be informed that a sentence of life imprisonment or death is mandatory on conviction of a capital felony. A prospective juror shall be disqualified from serving as a juror unless he states under oath that the mand a tory penalty of death or imprisonment for life will not affect his deliberations on any issue of fact.

Id . at 42 (emphasis added). The S upreme Court in A dams concluded that this oath was applied to exclude prospective jurors on grounds impermissible under Witherspoon . I d . at 49. The Court reasoned that “it is apparent that a Texas juror’s views about the death penalty might i nfluence the manner in which he performs his role without exceeding the ‘guided jury discretion,’ . . . permitted him under Texas law.” I d . at 46-47 (emphasis added). The provision improperly excluded potential jurors “who stated that they would be ‘affected’ by the possibility of the death penalty, but who apparently meant only that the potentially lethal consequences of their decision would invest their deliberations with greater seriousness and gravity or would involve them emotionally.” I d . at 49. “[N]either nervousness, emotional involvement, nor inability to deny or confirm any effect whatsoever is equivalent to an unwillingness or an inability on the part of the jurors to follow the court’s instructions and obey their oaths[.]” I d . at 50. “[T]o exclude all jurors who would be in the slightest way affected by the prospect of the death penalty or by their views about such a penalty would . . . deprive the defendant of [an] impartial jury” under the Sixth Amendment. I d . (emphasis added). A State may, however, “bar from jury service those whose beliefs about capital punishment would lead them to ignore the law or violate their oaths.” Id .

In the context of a § 2254 habeas proceeding, the Supreme Court in Witt reiterated the holding of Adams that § 2254(d), under which the finding of the trial judge is “presumed correct” unless one of the reasons enumerated in the statute is present. I d . at 426-27, 430, 431. “[W]here the record does not indicate the [constitutional] standard applied by a state trial judge, he is presumed to have applied the correct one.” I d . at 431; s ee McFadden v. Johnson , 166 F.3d 757, 758 (5th Cir. 1999).

To rebut this presumption, the peti tioner must adduce “clear and convincing evidence that the factual dete rmination by the State court was erroneous.” W itt , 469 U.S. at 435; Kelly v. Lynaugh , 862 F.2d 1126, 1134 (5th Cir. 1988). In Witt , the Supreme Court made a number of observations about the trial judge’s duties in addressing a challenge for cause for bias and the very nature of that function. [D]eterminations of juror bias cannot be reduced to q uestion-and-answer sessions which obtain results in the manne r of a catechism . . . . [M]any veniremen simply cannot be asked enough questions to reach the point where their bias has been made ‘unmistakably cle ar’. . . . Despite this lack of clarity in the printed record, however, there will be situations where the trial judge is l eft with the definite impression that a prospective juror would be unable to faithfully and impartially apply the law. Witt , 469 U.S. at 424-26. The trial judge’s “predominant function in determining juror bias involves credibility findings whose basis cannot be easily discerned from the appellate record.” I d . at 429.

Accordingly, the trial judge is not required to § 2254(d) to state appellate court findings In the face of W itt § 2254(d), we address AllridgeÂ’s contention below and conclude that its basic premise is flawed.

Allridge does not cite a single Supreme Court or Fifth Circuit decision holding that, in a Witherspoon habeas challenge, a federal court should defer solely to a s tate appellate court§ 2254(d). S ee id . at 545-46. The Supreme Cou rt emphasized that the state appellate court had even held a § 2254(d), on the claim.

Id . at 546. Mata thus offers no guidance in the circumstances of AllridgeÂ’s case.

Neither does Goode offer such guidance. In G oode , a habeas petitioner had argued § 2254 proceedings, the Eleventh Circuit Court of Appeals assumed a rguendo that the Florida Supreme Cour t§ 2254(d) , but concluded that the state-court finding was § 2254(d) standard of review could have been applied. We pause here to note that Allridge fails to compare his own case with the circumstances of Witt , wherein the Supreme Court closely scrutinized the state trial court§ 2254(d) standard. The petitioner in Witt had raised a Witherspoon challenge in his direct appeal to the Florida Supreme Court, see Witt , 469 U.S. at 415, yet the United States Supreme Court apparently saw no reason to address the Florida Supreme Cou rtÂ’s “findings” as to the Witherspoon challenge when the record included a detailed voir dire transcript an d a decision by the state trial court itself.

Allridge has cited no legal authority to suggest why his case should be treated any differently.

We shall, therefore, review the state trial court§ 2254(d) standard of review. 3 3. Voir dire and the dismissal of Osborn When, during the voir dire questioning, Osborn was asked whether anything in his background would lead him to believe that he could not serve as a juror in AllridgeÂ’s case, Osborn answered, “Nothing specific. I have a problem with the death penalty.” He then explained that he had “spent time in Vietnam as a contractor and saw enough of that, that I donÂ’t know that I could ma ke that decision, in all ho nesty.” Osborn described his feelings as “strong” and stated that, “even tho ugh we are not as a jury directly saying yes [with respect to the decision to impose the d eath penalty], itÂ’s going to be the death penalty; indirectly, because of the questions and the answers, we are the responsible group, and I donÂ’t know that I could, in all honesty, make that kind of a decision.” The following exchange then occurred: Q.: . . . Let me ask you this: Do you think that your feelings are so strong, or you classified them as strong, do you think these strong feelings would substantially impair your abi lity as a juror to follow the oath that you would take? A. It would tend to bias my opinion, I think.

Q. . . .

L et me ask you, Mr. Osborn, if say you were on this jur y and as the foreperson, say you were elected foreman, could yo u sign a verdict that sentenced James Vernon Allridge to death? A. I donÂ’t think I could. Honestly.

Q. . . . Would you say that you had conscientious scruples against the infliction of punishment that resulted in death? A. I don’t know if it would be conscientious or not, but [indicating] it is a gut feeling that I don’t know whether I could or not. The prosecutor then pointed out that jurors were required to take an oath to render a true verdict according to the law and the ev idence and asked Osborn whether he could “assess the death penalty.” Osborn answered, “I honestly don’t think I could” and that to take the oath would “creat e an awful lot of internal conflict.” He also stated that he would feel “very uncomfortable” making a sentencing decision if he were the last undecided juror.

Osborn then answered several questions from AllridgeÂ’s attorney about the general civic responsibilities of jurors, after which the prosecutor asked a specific question of Osborn: Q. . . . [I]f you were selected as a juror in this case , would you be able to follow your civic duty and sit down, set your feelings aside I am not saying ignore them, I am not saying deny their existence but sit down and g i ve fair consideration to the evidence and then answer the questions from the evidence as your oath requires? A. I can make that decision, you know; whether I could actually sign to do what was necess ary, I donÂ’t know.

Osborn subsequently stated that he “would answer them as honestly as I possibly could, but they are always going to be tempered by my basic instincts t hat I think it’s wrong for one person to take another person’s life.” After a short argument session, the trial court denied the prosecutor’s challenge to Osborn for cause.

The prosecutor then resumed his questioning of Osborn. Noting that the State was required to prove the special sentencing issues beyond a reasonable doubt, the prosecutor asked Osborn, “Before you c ould vote yes to any one of those questions, would you require there to be absolutely no doubt in your mind at all?” Osborn replied, “I think I would have to have, you know, little or no doubt,” but then stated, “[n]ot having been put in that position before, I don’t know.” The prosecutor also asked Osborn whether he could “in [his] own mind imagine evidence that could be brought to [him] that could convince [him] that the answer to [special issue] number two should be yes[.]” Osborn initially answered, “I don’t know if I can quantify that,” but when a nearly identical question was posed to him, he answered, “I can’t honestly think of anything, you know, that would make me make that decision.” The prosecutor next asked, “None whatsoever?” to which Osborn replied, “I don’t know.” Several minutes later, the following exchanged occurred: Q. Are your feelings are your strong feelings about the death penalty such that you are more inclined to be biased for Mr. Allridge as we start this case in that you know we are seeking the death penalty? A. I think it is going to temper any decision that I make. It’s got to bias it. I can’t say that it’s that I am starting off with a preconceived concept of guilt or innocence, but that has got to be, you know I mean, had I not known it beforehand, it would have come out eventually, but . . . .

Q. Yes. A. Yeah. It does tend to bias me, you know . . . .

Q. Against the death penalty and for the saving of a life? A. Basically, yes. The prosecutor also asked whether Osborn’s assertion that his decision would be “temper[ed]” by his feelings “would substantially impair [him] from si tting as a juror in this particular case” or “in any capital murder case.” Osborn replied, “I think it would be a consideration. I mean like I said, I don’t know. But I think it would, here again, tend to temper my decisions in how I perceive the evidence knowing what the consequences could be.” Again asking Osborn about the oath to render a “true verdict according to the law of Texas” and whether he could “honestly take that oath and then not do violence to [his] strong feelings” about the death penalty, Osborn stated, “I don’t think I could.” Then this final exchange occurred: Q. Is it fair to say that any verdict you reached at the s econd phase of this trial might not be based solely on the law from the Judge and the facts that you ’ve heard; might it be tempered by your strong feelings about the death penalty? A. That’s entirely possible.

Q. Might that those strong feelings change how you might view the evidence knowing what the result, the severe consequences of yes votes? A. I think so. At this point, the prosecutor resubmitted the challenge, and the trial court granted it. The court nev e rtheless allowed Allridge’s counsel to ask Osborn a few more questions. Defense counsel asked Osborn whether he c ould listen to the evidence at the guilt phase and decide whether the case had been proven beyond a reasonable doubt. Osborn stated, “I could make that decision, yes, but knowing the possible consequences, here again, that decision is also going to be tempered by the possible consequences later on.” Finally: Q. Well, I guess I have some trouble with temper. That is probably true for everybody.

It may be tempered the other way for some jurors.

Could you listen to the evidence presented both at the first stage and second stage, hav i ng found somebody guilty based on the evid ence, and answer those questions under your o ath and answer them based on what you thought the evidence showed? A. Not without involving my feelings for what I was doing.

At this point the trial court cut off the questioning, stating, “The Court has listened to the answers of Mr. Osborn and observed his demeanor and hi s manner in answering the questions. I think that I have made the determination in that regard, so I will grant the State’s challenge[.]” We conclude that, under the standards of review set forth by the Supreme Court in Witt , Osborn’s testimony supported the trial court’s sustaini ng of the State’s challenge for cause. 4 As there is nothing in t he record to suggest that the trial court applied any particular constitutional standard, it must be presumed that it applied the correct one. W itt , 469 U.S. at 431. That standard, as noted a bove, is whether the juror’s views would “substantially impair the performance of his duties as a juror and in accordance with his instructions and his oath.” Id . at 424. As the Court noted in Witt , “there will be situations where the trial judge is left with the definite impression that a prospecti ve juror would be unable to faithfully and impartially apply the law,” even when there is a “lack of clarity” in the record. W itt , 469 U.S. at 425-26. The juror’s bias need not be “proved with unmistakable clarity.” I d . at 424. In addition, Osborn’s failure to state explicitly th at his feelings about the death penalty would “substantially impair” his performance is not dispositive, because “[r]elevant voir dire questions . . . need not be framed exclusively in the language of” Adams . Id . at 433-34.

Allridge has not adduced clear and convincing evidence that the trial court’s sustaining of the prosecu tion’s challenge to Osborn for cause was erroneous. “[T]he question is not whether a reviewing court might disagree with the trial court’s findings, but whether those findings are fairly supported by the record.” W itt , 469 U.S. at 434 (citing Marshall v. Lonberger , 459 U.S. 422, 432 (1983)). Osborn made a number of statements that support the trial court’s finding. He twice suggested that in following his oath as a juror, his feelings would bias his opinion. He repeatedly expressed uncertainty and even doubt about whether he could make decisions that would result in imposition of the death penalty. He indicated that he could not think of anything that would cause him to vote “yes” as to the special issues. Finally, Osborn conceded that he did not think that he could take the oath and not do violence to his strong feelings, and that it was entirely possible that a verdict he reached at the close of the punishment phase might not be based solely on the law from the Judge and the facts that he had heard.

We acknowledge that the record in this case is not as clearcut as those in other cases in which we have rejected habeas claims under Witt . 5 Still, a trial court’s findings on a Witherspoon challenge are based on “determinations of demeanor and credibility that are peculiarly within a trial judge’s province.” Witt , 469 U.S. at 428. The trial court in Allridge’s case initially denie d the State’s challenge for cause with respect to Osborn; however, after hearing additional t estimony, the court sustained the challenge. The court emphasized that it had “listened to the answers of Mr. Osborn and observed his demeanor and his manner in answering his questions.” In the final analysis, we are sa tisfied that the district court’s ruling on Allridge’s Witherspoon claim was correct and free of reversible error. We therefore affirm the court’s ruling based largely on the standards of Witt . 6 B. Ineffective assistance of counsel Allridge only briefly sets forth his claim that his trial attorney performed ineffectively by failing to supervise personally a psychological test that was given to Allridge at jail. 7 He emphasizes that at the punishment phase, his psychological expert, Dr. Schmitt, 8 testified that: (1) Allridge had engaged in a crime spree with his brother Ronald, a classic sociopath, because Ronald had dominated him; (2) Allridge believed that his brother would kill him if he did not participate in the crimes; and (3) Allridge would not be violent in prison if removed from Ronald’s influence.

Allridge asserts that this testimony was based in part on written psychological tests administered to Allridge, which, as revealed in the State’s cross-examination of Dr. Schmitt, Allridge completed by himself, without Dr. Schmitt’s presence or supervision. Allridge contends that Dr. Schmitt’s mitigation testimony was t hus discredited and made virtually useless. He blames this failure on counsel’s abrogation of his duty to investigate and maintains that prejudice resulted because the failure doomed any chance of persuading the jury that Allridge did not present a future danger to society. 1. Procedural default A llridge does not address the State’s contention, and t h e district court’s conclusion, that this particular ineffectiveassistance claim was procedurally defaulted when Allridge failed to raise it in the state courts. Before we address the question of procedu ral default, however, we must first determine whether the issue is even properly before us. The district court granted Allridge a COA on his ineffective assistance of counsel claim without elaborating further. At least two circuits have held that, “[a]bsent an explicit statement by the district court, in cases where a district court grants a COA with respect t o the merits of a constitutional claim but the COA is silent with respect to procedural claims that must be resolved if the panel is to reach the merits, [the court of appeals] will assume that the COA also encompasses any procedural claims that must be addressed on appeal.” J ones v. Smith , 231 F.3d 1227, 1231 (9th Cir. 2000); McCoy v. United States , 266 F.3d 1245, 1248 (11th Cir. 2001) (quoting Jones ), cert. denied , 536 U.S. 906 (2002). We agree with this analysis and conclude that, under this standard, we should address the district court’s procedural-default ruling.

The procedural default doctrine precludes federal habeas review when the last reasoned state-court opinion addressing a claim explicitly rejects it on a state procedural ground. Y lst v. Nunnemaker , 501 U.S. 797 , 801, 803 (1991). When the state court has relie d on an independent and adequate state procedural rule, federal habeas review is pr ecluded unless the petitioner demonstrates either (1) cause and prejudice or (2) that a failure to address the claim will result in a fundamental miscarriage of justice. Coleman v. Thompson , 501 U.S. 722 , 750 (1991). Ordinarily, a habeas petition must be dismissed if any issue has not been exhausted in the state courts. R ose v. Lundy , 455 U.S. 509, 513-19 (1982). “When . . . state remedies are rendered unavailab le by the petitioner’s own procedural default, federal courts are barred from reviewing those claims.” S ones v. Hargett , 61 F.3d 410, 416 (5th Cir. 1995). “‘[I]f the petitioner failed to exhaust state remedies and the court to which petitioner would be required to present his claims in order to meet the exhaustion requirement would now find the claims procedurally barred, . . . [then] there is procedural default for the purposes of federal habeas. . . .’” I d . (quoting Coleman , 501 U.S. at 735 n.1 ). A second state po stconviction application by Allridge would almost certainly be barred by Texas courts as an abuse of the wr it, and this bar would operate as an adequate and independent state procedural ground for procedural-def a ult purposes. F inley v. Johnson , 243 F.3d 215, 219 (5th Cir. 2001). In the dist rict court, Allridge denied that this particular ineffectiveness claim was unexhausted. He argued that the CCA had adopted the state habeas trial court’s finding that defense counsel made reasonable strategic choices about how to present all possible mitigating factors in consultation with Dr. Sch mitt and thereby rendered effectiv e assistance under the totality of the circumstances (citing Vela v. Estelle , 708 F.2d 954, 959 (5th Cir. 1983)). Although it has not disputed that he did not raise an ineffective-assistance claim specifically addressing his trial counsel’s failure personall y to supervise the psychological test given to Allridge or to discover that the expert had failed to do so; neither has Allridge claimed that the Texas courts did not address this specific claim. Instead, the state habeas court did express the general conclusion that, after a thorough investigation into all possible mitigating factors available to counsel at time of trial and in consultation with Dr. Schmitt, the defense made a reasoned, strategic, and effective presentation based on their professional assessment of those factors.

To exhaust, a petitioner “must have fairly presented the substance of his claim to the state c ourts.” N obles v. Johnson , 127 F.3d 409, 420 (5th Cir. 1997) (cit ing P icard v. Connor , 404 U.S. 270, 275-76 (1971)). “‘It is not enough that all the facts necessary to support the federal claim were before the state courts or that a somewhat similar state-law claim was made.’” Wilder v. Cockrell , 274 F.3d 255, 259 (5th Cir. 2001) (quoting Anderson v. Harless , 459 U.S. 4, 6 (1982)).

Allridge cites Vela for the proposition that he successfully exhausted his current ineffective-assistance claim. In Vela , the petitioner asserted three error s in his state habeas petition as grounds for a finding that counsel was ineffective, and later in his federal petition he urged several additional grounds supporting his claim. V ela , 708 F.2d at 957-58. We determined that Vela’s state habeas petition asserted ineffective assistance on the basis of counsel’s entire performance and that Vela’s three assertions of error were merely “singling out for comment certain strik ingly prejudicial errors.” Id . at 959. We determined further that the state court conducted its own ind ependent analysis of counsel’s performance based on a review of the re cord as a whole. I d .

Concluding that “[c]haracterizing these allegations as ‘unexhausted claims’ would require us to find that the state habeas court failed in its duty to evaluate counsel’s performance on the basis of the record as a whole[,]” we held that Vela had exhausted h is state remedies. Id . at 960.

In Thomas v. Collins , 919 F.2d 333, 334 (5th Cir. 1990), however, the petitioner asserted ineffective assistance of counsel in the state court, arguing specifically that his appellate counsel failed to notify him timely of his right to petition for discretionary revie w. Id . In federal court, the petitioner presented an entirely new claim regarding his counselÂ’s ineffectiveness, arguing that counsel failed to assert on appeal a claim that the trial judge had failed to comply with T EX . C ODE . C RIM .

P ROC . art. 1.15. I d . The petitioner suggested that exhaustion was met because his case was similar to V ela . I d . at 334-35. We distinguished Vela , determining that the “record does not support a finding that the state court ever reviewed counsel’s performance in li ght of the Article 1.15 violation.” I d . at 335. We thus affirme d the district court’s dismissal for failure to exhaust.

Id .

The circumstances of Allridge’s postconviction proceeding approximate those in Thomas more closely than they approx imate those in Vela . Allridge did not raise a claim that counsel’s entire performance was ineffective. Neither is there any indication that the state trial court conducted an independent review of the record as a whole. Although we conclude that the district co urt’s order granting COA included a certification regarding the question of procedural default, we also conclude that (1) Allridge’s particular ineffective-assistance claim was unexhausted and thus was procedurally defaulted, and (2) Allridge has not shown cause or prejudice to excuse the default. We nevertheless address the merits of that claim in the alternative. 2. Merits of ineffective-assistance claim It is universally recognized that, to prevail on a claim of ineffective assistance of counsel, a petitioner must show (1) that his counsel’s performance was deficient in that it fell below an objective standard of reasonableness and (2) that the deficient performance prejudiced his defense. S trickland v. Washington , 466 U.S. 668, 689-94 (1984). When we assess whether an attorney’s performance was deficient, we “must indulge a strong presumption that counsel’s conduct falls wi thin the wide range of reasonable professional assistance.” I d . at 689; Andrews v. Collins , 21 F.3d 612, 621 (5th Cir. 1994). To show S trickland prejudice, a petitioner must demonstrate that counsel’s errors were so serious as to “render[ ] the result of th e trial unreliable or the proceeding fundamentally unfair.” Lockhart v. Fretwell , 506 U.S. 364, 369 (1993). “In the context of a claim that counsel rendered ineffective assis tance by failing to present evidence at the punishment phase of a capital murder trial, the inquiry is whether there is a reasonable probabi lity that, had the evidence been presented, it would have altered the punishment verdict.” Harris v. Cockrell , 313 F.3d 238, 243 (5th Cir. 2002), petition for cert. filed , (U.S. Mar. 17, 2003) (No. 02-1433). Failure to establish either deficient performance or prejudice d efeats the claim.

Strickland , 466 U.S. at 697.

To the ext ent that Allridge classifies his ineffectiveassistance claims as one involving a failure to investigate, “[a] def ense counsel’s failure to engage in an appropriate investigation of potential mitigating evidence in the pun ishment phase can support a claim of ineffe ctive assistance of counsel.” Smith v. Cockrell , 311 F.3d 661, 668-69 (5th Cir. 2002) (citing Williams v. Taylor , 529 U.S. 362, 390 (2000)). “[W]e focus on w hether the investigation supporting counsel’s decision ... w as itself reasonable .” Wiggins v. Smith , 2003 WL 21467222 *8 (U.S.) (emphasis in original). “‘A defendant who alleges a failure to investigate on the part of his counsel must allege with specificity what the investigation would have revealed and how it would have altered the outcome of the trial.’” Moawad v. Anderson , 143 F.3d 942, 948 (5th Cir. 1998) (citation omitted). At the punishment phase, trial co unsel for Allridge called clinical psychologist Dr. Schmitt as an exper t witness. Dr. Schmitt had conducted a two-hour clinical interview of Allridge at the jail and had administered and expla ined some psychological tests, which he left with Allridge and which were returned to Dr. Schmitt after Allridge had completed them. 9 Dr. Schmitt had also interviewed Allridge’s parents and examined Allridge’s artwork.

Dr. Schmitt testified that the results of all the tests showed that Allridge was not a sociopath, that he had demonstrated the ability to maintain relationships and to exhibit loyalty to other persons, and that he is an individual who ex hibits remorse. Dr. Schmitt expressed the o pinion that AllridgeÂ’s brother, Ronald, has a classic sociopathic personality, and testified that Allridge had been picked on, physically intimidated , and beaten on a regular basis by Ronald. Dr. Schmitt maintained that, aside from his relationship with Ronald, Allridge is basically a no n -violent person.

Dr. Schmitt admittedly was not present when the written tests were completed and did not have personal knowledge that they were in fact filled out by Allridge himself. Dr. Schmitt nevertheless testified that he had every reason to believe that the tests were filled out by Allridge, and that the handwriting on the Sentence Completion portion was very similar to the handwriting on AllridgeÂ’s artwork. Dr. Schmitt also confirmed that everything that he told the jury about Ronald was based on information related to him by either Allridge or his parents.

Allridge has shown neither cause (deficient performance) nor prejudice. Although the cross-examination of Dr. Schmitt confirmed that he did not remain in the presence of Allridge after submitting the written tests to him, the State presented no evidence to suggest that any one other than Allridge had completed them. In addition, Dr. Schmitt’s testimony about Allridge was based not merely on the written test results but also on his two-hour interview with Allridge, his intervie ws with Allridge’s parents, and his analysis of Allridge’s artwork. Allridge exaggerates when he contends that his attorney’s failure personally to supervise the written tests, or to investigate an d determine before trial that Dr. Schmitt had not done so either, “destroyed the defense’s case in the sentencing ph ase.” The record confirms that counsel’s investigation in regard to Dr. Sc hmitt was itself reasonable and that it was sufficient to support counsel’s decision to put on this mitigating evidence. S ee Wiggins v. Smith , 2003 WL 21467222 *8 (U.S.). He has not overcome S trickland ’s presumption that counsel’s performance fell “within the wide ran ge of reasonable professional assistance.” See Strickland , 466 U.S. at 689.

CounselÂ’s investigation was more than adequate to support the factual decision to have Dr. Schmitt testify; and doing so was not deficient.

Neither has Allridge demonstrated S trickland prejudice. On direct appeal, Allridge challenged the sufficiency of the evidence to support the jury’s affirmative finding on the second special issue. See Allridge , 850 S.W.2d at 487. In rejecting Allridge’s challenge, the CCA emphasized that, when Allridge, Ronald, and two accomplices drove around Tarrant County committing robberies on the night of March 24, 1985, Allridge personally committed three robberies while the other three men waited in the car; and that the evening culminated with Ronald’s shooting and killing a customer at a Whataburger restaurant while Allridge waited in the car. That court also stressed that, during the period of two months following the instant Circle K robbery-murder, Allridge personally committed several armed robberies in which Ronald was not involved at all; that the morning after the Whataburger robbery-murder, Allridge apparently us ed the proceeds of the prior evening’s robberies to pay his rent and to pay his attorney in a matter unrelated to the armed robberies; that the facts of the Circle K robbery-murder showed Allridge’s cool calculation; and that the robberies afterwards showed his “remorselessness.” Id . at 487-89. The CCA noted Dr. Schmitt ’ s testimony regarding Ronald’s dominance of Allridge, but did not even mention his discussion of the written tests. I d . at 488, 489. The court discounted the testimony regarding Ronald’s domination on the ground that Allridge had committed several armed robberies in which Ronald was not involved.

Id . at 489. T he evidence supporting a finding of Allridge§ 2254 petition.

PETITION DENIED.

2 The State§ 2254 petition that the CCA erred when it stated that Osborn§ 2254(d).

3 Even if we were to credit Allridge’s arguments that we should defer to the CCA’s “implied” findings of fact, we would see that Allridge has somewhat mi srepresented those findings. It is true that the CCA stated that Osborn’s feelings would “influence” his assessment of the evidence and “affect” his ability to comply with his oath. S ee Allridge , 850 S.W.2d at 478. The court was merely explaining, however, that these findings “support[ed]” the trial court’s determination that Osborn was “substantially impaired” in his performance in his duties as a juror; it did not state that these factors, standing alone, supported the trial court’s finding. Id . Moreover, the CCA’s ultimate “factual finding” was that Osborn was “ substantially impaired,” the key phrase from Adams .

4 In arguing at length that we must defer to the CCA§ 2254 actions involving Witherspoon challenges, we have repeatedly applied those standards since the issuance of Witt . S ee , e.g. , Soria v. Johnson , 207 F.3d 232, 245-47 (5th Cir. 2000); McFadden , 166 F.3d at 758-61; Fuller v. Johnson , 114 F.3d 491, 498-501 (5th Cir. 1997); Mann v. Scott , 41 F.3d 968, 980-82 (5th Cir. 1994); Nethery v. Collins , 993 F.2d 1154, 1159-60 (5th Cir. 1993); Drew v. Collins , 964 F.2d 411, 416-17 (5th Cir. 1992); Granviel v. Lynaugh , 881 F.2d 185, 187-89 (5th Cir. 1989); Ellis v. Lynaugh , 873 F.2d 830, 832-37 (5th Cir. 1989); K elly , 862 F.2d at 1133-35; Bell v. Lynaugh , 828 F.2d 1085, 1092-93 (5th Cir. 1987); Riles v. McCotter , 799 F.2d 947, 949-50 (5th Cir. 1986); Smith v. McCotter , 798 F.2d 129, 132-34 (5th Cir. 1986)

5 C f. McFadden , 166 F.3d at 759-60 (venireman agreed that he would “automatically” vote against the death penalty “[r]egardless of the facts and circumstances of the case”); M ann , 41 F.3d at 980-81 & n.9 (veniremans flatly stated that they could not take “oath” to base answers to punishment-pha se issues solely on evidence); Nethery , 993 F.2d at 1160 (at least one venireman would vote “no” on special issues, regardless of the evidence); Drew , 964 F.2d at 416-17 (one venireman would hold State to burden of proof higher than reasonable-doubt standard, and another would vote “no” as to future-dangerousness special issue unless evidence showed that defendant wo uld commit future murders); E llis , 873 F.2d at 834-36 (venireman could not take oath if it required him to answer “yes” to both special issues); K elly , 862 F.2d at 1134 & n.15 (venire man would answer “no” to both special issues “[n]o matter what the evidence i s”); B ell , 828 F.2d at 1092 (venireman repeatedly stated that she could not impose death penalty “under any circumstances”); Smith , 798 F.2d a t 133 (venireman would “ignore the law” or “violate [his] oath” in certain circumstances). But see Riles , 799 F.2d at 949 & n.2 (rejecting Witt challenge to venireman whose feelings about the death penalty would “influence” his “way of thinking” and who, if he had a “choice,” would “choose something less than death”)

6 We need not specifically address Allridge§ 2254(d), as discussed in Witt , does not require us to address alternative grounds that might have been cited by reviewing courts. In any event, AllridgeÂ’s grievances regarding these alternative grounds are without merit. He complains, for instance, t hat the prosecution asked Osborn questions about whether , as a jury foreman, he c ould sign a verdict that would result in the defendantÂ’s execution, when Texas law states that any juror can refuse to serve as jury foreman. It is true that whether a venireman could impartially sign a verdict is “immaterial to jury service under Witherspoon .” S ee Alderman v. Austin , 663 F.2d 558, 563-64 (5th Cir. 1981). Aside from answers to the questions regarding their abilities as foremen, however, the excluded veniremen in Alderman had “evidenced no ‘unambiguousÂ’ intent to oppose capital punishment eith er in principle or in the trial.” Id . at 563. In contrast, the record of OsbornÂ’s voir dire is not nearly so clear. 7 There are minor inconsistencies in AllridgeÂ’s catego rization of this claim at different times. Although he ref ers to trial counselÂ’s own failure to supervise the psychological test, we perceive the gravamen of AllridgeÂ’s inadequate investigation claim to be that counsel should have discovered before trial that his psychological expert, who gave the tests, had not remained present while the written pro tions were being completed; and that this led to the expertÂ’s being discredited on cross-examination

8 Dr. Schmitt’s name appears as “Schmidt” in the trial transcript and in some other pleadings, but is spelled “Schmitt” in his affidavit

9 The tests consisted of the Minnesota Multiphasic Personality Inventory (“MMPI”), a “commonly used personality test,” a nd a Sentence Completion Test, a test that permits the psychologist to determine people’s attitudes and ways of thinking about a wide variety of different subjects