Dismissed by published opinion. Judge LUTTIG wrote the opinion, in which Judge WILLIAMS joined. Judge DIANA GRIBBON MOTZ wrote an opinion concurring in the judgment.
OPINION
LUTTIG, Circuit Judge:
A North Carolina jury convicted Ernest Paul McCarver of first-degree murder and robbery with a dangerous weapon, and McCarver was sentenced to death and forty years imprisonment, respectively, for those convictions. After appealing his convictions in state court on direct review and in state habeas proceedings, McCarver filed a petition for writ of habeas corpus in federal district court. See 28 U.S.C. § 2254. The district court dismissed MeCarver’s petition, and he now appeals that dismissal. Because we conclude that McCarver has failed to make a substantial showing of the denial of a constitutional right, we deny his application for a certificate of appealability and dismiss his petition for writ of habeas corpus.
I.
On January 2,1987, Ernest Paul McCar-ver stabbed Woodrow F. Hartley to death. According to the North Carolina Supreme Court:
[McCarver] and [Jimmy] Rape entered through the rear entrance of the K & W Cafeteria shortly after Hartley arrived at 5:00 a.m. [McCarver] walked up to Hartley and talked to him for a few minutes. Rape grabbed Hartley from behind in a headlock and attempted to strangle him. Rape released Hartley, who was then grabbed by [McCarver] in a headlock. When [McCarver] let him go, Hartley fell to the ground. [McCar-ver] took a knife from his pants pocket and stuck it into Hartley’s chest several times. Hartley died within minutes.
Gene Blovsky, an employee of the cafeteria, observed [MeCarver’s] automobile parked near the back door of the cafeteria. He saw [McCarver] emerge from behind a wall; [McCarver] was carrying a knife, which he attempted to hide in his right hand. Next, Blovsky saw Hartley lying on the floor in the hallway with a spot of blood on his wrist. Blovsky saw another man near Hartley, realized what had happened, became frightened, and ran out the door.
.... [After leaving the scene and][b]efore going to their assigned job site, [McCarver] and Rape pawned a 1902 silver dollar, which had been taken from the victim, for seven dollars at a Monroe pawn shop. [McCarver] and Rape were arrested by Monroe police at their assigned job site.
State v. McCarver, 341 N.C. 364, 462 S.E.2d 25, 30-31 (N.C.1995), cert. denied, 517 U.S. 1110, 116 S.Ct. 1332, 134 L.Ed.2d 482 (1996). McCarver believed that Hart-ley was responsible for his probation being revoked. See id. at 30. After his arrest, McCarver confessed to Hartley’s murder.
McCarver was indicted in North Carolina for murder and robbery with a dangerous weapon. He was convicted of first-degree murder and robbery with a dangerous weapon in the April 18, 1988 Criminal Session of Superior Court, Cabarrus County. He was sentenced to death for the murder conviction and to forty years imprisonment for the robbery conviction. On appeal, the North Carolina Supreme Court ordered a new trial on both charges. See State v. McCarver, 329 N.C. 259, 404 S.E.2d 821 (N.C.1991).
A second trial was held during the September 8, 1992 Special Criminal Session of Superior Court, Cabarrus County, and a jury again convicted McCarver of first-degree murder and robbery with a danger[587]*587ous weapon. It is this second trial that is the subject of McCarver’s federal habeas petition.
Before the first trial, McCarver’s appointed counsel requested a competency examination for McCarver at Dorothea Dix Hospital (“Dix”). During Dr. Tanas’ January 1987 examination of McCarver, McCarver completed fill-in-the-blank sentences as follows:
I like to kill.
The happiest time was when I killed that man.
At bedtime I dream about killing.
J.A 364 (Report of Dr. Tanas). A second doctor at Dix, Dr. Lara, also evaluated McCarver and incorporated McCarver’s fill-in-the-blank answers into her report. J.A. 367 (Report of Dr. Lara). The reports were forwarded to the trial court, to defense counsel, and to the district attorney. J.A. 365, 371. Both doctors concluded that McCarver was competent to stand trial, and no competency hearing was held.
In the first trial, defense counsel hired Dr. Parasi to testify as an expert on McCarver’s behalf. Trial counsel gave both Dix reports to Dr. Parasi for the doctor’s evaluation of McCarver, and Dr. Parasi incorporated them into Ms evaluation.
In the second trial, defense counsel hired another expert, Dr. Sultan, a clinical forensic psychologist, to evaluate McCar-ver and to testify on her evaluation. Defense counsel provided her with copies of the Dix reports, as well as Dr. Parasi’s report, which reports she used to evaluate McCarver. Dr. Sultan testified that McCarver was suffering from borderline intellectual functioning, depression, substance abuse disorder, personality disorder, and various other disorders. Defense counsel asked Dr. Sultan about the Dix fill-in-the-blank answers on direct examination, J.A. 65, and the district attorney asked about those same fill-in-the-blank responses on cross-examination, J.A. 150.
During McCarver’s capital sentencing proceeding, the State submitted three aggravating circumstances to the jury: (1) that the murder was committed to avoid lawful arrest; (2) that the murder was committed while the defendant was engaged in the commission of a robbery with a dangerous weapon; and (3) that the murder was especially heinous, atrocious, or cruel. The jury found the first two aggravating circumstances to be present in McCarver’s case. J.A. 307-08.
In the same proceeding, McCarver submitted seventeen mitigating circumstances to the sentencing jury. The jury found fourteen of those circumstances to be present. The jury did not find that McCarver had a history of alcohol or substance abuse which began at an early age as a consequence of emotional and sexual abuse, that McCarver expressed remorse for his actions, or that there were other unspecified mitigating circumstances. J.A. 309-12.
At the conclusion of the second trial, the jury again returned a verdict of death for the murder conviction and of forty years imprisonment for the robbery conviction.
McCarver appealed his convictions from the second trial. On September 8, 1995, the North Carolina Supreme Court affirmed McCarver’s death sentence. State v. McCarver, 341 N.C. 364, 462 S.E.2d 25 (N.C.1995), cert. denied, 517 U.S. 1110, 116 S.Ct. 1332, 134 L.Ed.2d 482 (1996).
On October 22, 1996, McCarver filed a motion for appropriate relief (“MAR”) in North Carolina Superior Court. The court granted the State’s motion for summary denial and denied McCarver’s motion for appropriate relief. J.A. 492 (North Carolina Superior Court opimon). The Supreme Court of North Carolina denied McCarver’s petition for certiorari to review the denial of his motion for appropriate relief. J.A. 590 (North Carolina Supreme Court order).
On September 22, 1998, McCarver filed a petition for writ of habeas corpus, pursuant to 28 U.S.C. § 2254, in federal district court. A federal magistrate judge recommended denying the petition, J.A. 691 (report and recommendation), and the district [588]*588court adopted the magistrate judge’s report and recommendation, J.A. 746 (district court opinion). This appeal followed.
II.
McCarver’s sole argument on appeal is that his trial counsel rendered ineffective assistance in failing to take appropriate steps to prevent the jury from learning of McCarver’s responses in the Dix fill-in-the-blank sentences.1 Specifically, McCarver argues that trial counsel erred (1) by failing to file a motion in limine before trial to prevent the district attorney from using the information contained in the two Dix reports, and (2) by providing the Dix reports to Dr. Sultan, thereby ensuring that the information would be subject to cross-examination. According to McCarver, had the jury not learned of McCarver’s fill-in-the-blank responses, the jury would have found the mitigating circumstance that he was remorseful, and had the jury found that he was remorseful, there is a reasonable probability that it would have returned a sentence of life imprisonment instead of death.
The district court held that McCarver’s ineffective assistance of counsel claim was procedurally barred under North Carolina law and, alternatively, that the claim lacked merit. We are satisfied that McCarver’s ineffective assistance of counsel claim was procedurally barred. But, in any event, we proceed to the merits, and we, too, are confident that there is no merit to McCarver’s ineffective assistance claim.2
A.
Federal habeas review of a state prisoner’s claims that are procedurally defaulted under independent and adequate state procedural rules is barred unless the prisoner can show cause for the default and demonstrate actual prejudice as a result of the alleged violation of federal law, or prove that failure to consider the claims will result in a fundamental miscarriage of justice.3 See Coleman v. Thompson, 501 U.S. 722, 750, 111 S.Ct. 2546, 115 L.Ed.2d 640 (1991). A state rule is adequate if it is “firmly established,” James v. Kentucky, 466 U.S. 341, 348, 104 S.Ct. 1830, 80 L.Ed.2d 346 (1984), and regularly and consistently applied by the state court, see Johnson v. Mississippi, 486 U.S. 578, 587, 108 S.Ct. 1981, 100 L.Ed.2d 575 (1988).
The North Carolina Superior Court, in dismissing McCarver’s motion for appropriate relief, rejected his ineffective assistance of counsel claim as procedurally barred under N.C.G.S. § 15A-1419(a)(3). Section 15A-1419(a)(3) provides for denial of appropriate relief when “[u]pon a previous appeal the defendant was in a position [589]*589to adequately raise the ground or issue underlying the present motion but did not do so.” The Superior Court specifically held that “[t]his claim could have been raised on direct appeal because the transcript and record contain information upon which defendant could have asserted such a claim.” J.A. 521, 523 (North Carolina Superior Court opinion).
McCarver does not deny that he failed to raise his ineffective assistance claim on direct appeal. He also does not deny that section 15A-1419(a)(3) is generally an independent and adequate state procedural bar. See Appellant’s Br. at 44 (“In Williams v. French, ... this Court stated that N.C.G.S. § 15A-1419(a)(3), in general, is an independent and adequate state ground. Mr. McCarver does not disagree with this holding.”); see also Williams v. French, 146 F.3d 203, 209 (4th Cir.1998) (“North Carolina General Statutes § 15A-1419(a)[ ] is an independent and adequate state ground.”), cert. denied, 525 U.S. 1155, 119 S.Ct. 1061, 143 L.Ed.2d 66 (1999). McCarver argues instead that in the special case of ineffective assistance of counsel claims, the bar is not consistently and regularly applied, and is therefore not adequate. In the alternative, McCarver argues that he can show cause for his default and actual prejudice arising from the default of his ineffective assistance claim in state court. We are satisfied that the state procedural bar is adequate, and that, alternatively, McCarver cannot show cause for his default and actual prejudice resulting from the default.
1.
First, as to McCarver’s claim that the state procedural bar is not consistently and regularly applied to ineffective assistance of counsel claims, we reject McCarver’s attempt to treat ineffective assistance claims as categorically different from other kinds of claims that can be barred under section 15A-1419(a)(3). McCarver argues that because ineffective assistance claims are ordinarily more appropriately raised in collateral proceedings, these claims are not consistently procedurally barred in collateral proceedings in North Carolina. Although McCarver is undoubtedly correct that many ineffective assistance claims could not be brought on direct appeal, section 15A-1419 is not a general rule that any claim not brought on direct appeal is forfeited on state collateral review. Instead, the rule requires North Carolina courts to determine whether the particular claim at issue could have been brought on direct review.
Here, the North Carolina Superior Court did find that McCarver could have raised his ineffective assistance claim on direct review. And it was because of this specific finding that the Superior Court held that McCarver’s particular ineffective assistance of counsel claims were procedurally barred. The question we must ask, then, is whether the particular procedural bar is applied consistently to cases that are ‘procedurally analogous — here, cases in which the particular claim raised could have been raised previously but was not. It would be incorrect to ask, as McCarver would, whether ineffective assistance claims that cannot be brought on direct review are barred under section 15A-1419, because by the very terms of section 15A-1419(a)(3) — requiring that the claim could have previously been brought — such claims would not be procedurally barred.
In this case, for McCarver to make a colorable showing that section 15A-1419(a)(3) is not consistently and regularly applied to ineffective assistance claims, he would need to cite a non-negligible number of cases in which ineffective assistance claims could have been brought on direct review but were not, and in which the collateral review court nonetheless failed to bar the claim under section 15A-1419(a)(3) because the claim was an ineffective assistance claim. To the contrary, North Carolina courts have held that ineffective assistance claims that are not brought on direct review can be procedurally defaulted under section 15A-1419(a)(3). See, e.g., Williams v. French, 146 F.3d at 218. Indeed, the very case [590]*590McCarver cites for his concession that N.C.G.S. § 15A-1419(a)(3) is “in general” an independent and adequate state ground, see Appellant’s Br. at 44 (referencing Williams v. French), is a case in which we refused to review the defendant’s ineffective assistance of counsel claim because it was procedurally barred under section 15A-1419(a)(3). See Williams v. French, 146 F.3d at 217-18. In Williams, we did not even discuss the possibility that there is a general ineffective assistance of counsel claim exception to our holding that section 15A-1419(a)(3) is an independent and adequate state procedural bar.
McCarver’s attempt to find support in our decision in Smith v. Dixon, 14 F.3d 956 (4th Cir.1994) (en banc), cert. denied, 513 U.S. 841, 115 S.Ct. 129, 130 L.Ed.2d 72 (1994), for his position that N.C.G.S. § 15A-1419(a)(3) is not regularly and consistently applied to ineffective assistance of counsel claims, is unavailing. In Smith, we did not say that section 15A-1419(a)(3) is not regularly and consistently applied to ineffective assistance claims. Rather, we made the unremarkable statement that “allegations of ineffective assistance of counsel generally are properly raised on collateral review.” Id. at 966; see also id. at 969 (noting that although there are exceptions, “[t]he accepted practice is to raise claims of ineffective assistance of eounsel in post-conviction proceedings, rather than direct appeal,” due to the need for evidentiary development of these appeals; and stating that “allegations of ineffective assistance usually would not be subject to summary denial under § 15A-1419(a)” (emphases added)). More importantly for McCarver’s case, but ultimately detrimental to his position, in holding that Smith’s particular Sixth Amendment claim (“failure of the Supreme Court of North Carolina to provide standards for review deprived him of his Sixth Amendment right to counsel because it rendered counsel unable to adequately prepare or present argument”) was properly barred under section 15A-1419(a)(3), we recognized that there are exceptions to the general rule in North Carolina that ineffective assistance of counsel claims cannot be raised on direct appeal. See id. at 969 (citing State v. Dockery, 78 N.C.App. 190, 336 S.E.2d 719, 721 (1985), as “noting that although there are exceptions, ‘[t]he accepted practice is to raise claims of ineffective assistance of counsel in post-conviction proceedings, rather than direct appeal,’ due to the need for evidentiary development of these claims” (emphasis added)). And McCar-ver’s case, according to the North Carolina Superior Court, is such an exception for which the state procedural bar would be applicable and adequate.4
[591]*5912.
McCarver argues in the alternative that even if we find North Carolina’s procedural default rule to be applicable to his ineffective assistance of counsel claim, he can demonstrate cause to excuse his default in not raising the claim on direct review and actual prejudice resulting from the default. To demonstrate cause, McCarver needs to make “a showing that the factual or legal basis for a claim was not reasonably available to counsel.”5 McCleskey v. Zant, 499 U.S. 467, 494, 111 S.Ct. 1454, 113 L.Ed.2d 517 (1991) (quoting Murray v. Carrier, 477 U.S. 478, 488, 106 S.Ct. 2689, 91 L.Ed.2d 397 (1986)) (internal quotation marks omitted).
McCarver argues not merely that the factual basis for his ineffective assistance of counsel claim was not reasonably available to appellate counsel, but that “it would have been literally impossible ” for his appellate counsel to have raised the claim on direct appeal because the record was not sufficient to support a reasonable claim of ineffective assistance. Appellant’s Br. at 46. Specifically, McCarver posits that he could only have made his ineffective assistance claim if he had had the opportunity to supplement the record through an evidentiary hearing. McCar-ver alleges that, at such an evidentiary hearing, he could have presented the testimony of trial counsel who could have explained why counsel failed to prevent the information from entering the record, the testimony of a criminal law expert who could have explained why counsel was ineffective, and the testimony of a mental health professional who could have explained that the Dix examination fill-in-the-blank questions were beyond the scope of a normal competency evaluation. Although there is no question that with an evidentiary hearing McCarver may have gathered or created more evidence, we are satisfied that the North Carolina Superior Court did not err in concluding that McCarver’s appellate counsel could have brought McCarver’s ineffective assistance of counsel claim on direct appeal in North Carolina state court.
McCarver’s ineffective assistance claim, in essence, is that counsel should have prevented the damaging Dix report from being entered into the trial record through all available means. As noted, the North Carolina Superior Court found that McCarver could have raised this claim on direct review. In reviewing the North Carolina court’s findings that support its application of the state procedural bar, we accord the state’s findings a presumption of correctness, see Williams, 146 F.3d at 209 (“Findings of the state court supporting its decision to apply the state procedural default rule are entitled to a presumption of correctness in determining whether cause exists to excuse a procedural default.” (citing 28 U.S.C. § 2254(d))). But even absent this presumption, we believe the trial record was more than sufficient to enable McCarver’s counsel to advance on direct appeal the ineffective assistance claim that McCarver is advancing now. At the time of McCar-ver’s direct appeal, the trial record included not merely the damaging fill-in-the-blank statements themselves, but also the fact that it was defense counsel who asked specifically about those statements first on direct examination of Dr. Sultan. The record also indicated that trial counsel had turned over Dr. Parasi’s report and testimony, which included the fill-in-the-blank responses, to Dr. Sultan. J.A. 20-23. The record also showed that trial counsel filed no motion in limine and that no competency hearing was ever held. With this information in the record, McCarver’s appellate counsel could easily have made the claim on direct review that trial counsel erred by providing the reports to Dr. Sultan. To be sure, in a hearing on collateral review, McCarver likely could have [592]*592found a criminal law expert or other expert to comment on trial counsel’s performance. By focusing on the evidence that was available instead of the evidence that could have been generated, however, cf. Smith v. Murray, 477 U.S. 527, 537, 106 S.Ct. 2661, 91 L.Ed.2d 434 (1986) (“[T]he question is not whether subsequent legal developments have made counsel’s task easier, but whether at the time of the default the claim was ‘available’ at all.”), it is plain that McCarver could have made the very ineffective assistance of counsel claim on direct review that he does today — and just as effectively. Therefore, we do not find that the lack of non-record evidence available to appellate counsel constitutes cause to excuse McCarver’s procedural default.
Finally, even if McCarver could demonstrate cause, we reject his argument that he was actually prejudiced by the failure of trial counsel to prevent the Dix fill-in-the-blank responses from being entered into the record. To establish prejudice, McCarver must show “not merely that the errors at his trial created a possibility of prejudice, but that they worked to his actual and substantial disadvantage, infecting his entire trial with error of constitutional dimensions.” United States v. Frady, 456 U.S. 152, 170, 102 S.Ct. 1584, 71 L.Ed.2d 816 (1982). McCarver’s theory of prejudice does not demonstrate that any error worked to his actual and substantial disadvantage. McCarver argues that had trial counsel attempted to keep the Dix reports out of the record, counsel would have been successful; had the Dix report fill-in-the-blank responses been kept out of the record, the jury would have found the mitigating circumstance that McCarver was remorseful; and had the jury found that McCarver showed remorse, then there is a reasonable probability that it would not have sentenced him to death.
Even assuming that counsel would have been successful if he had tried to keep the Dix reports out of the record, McCarver’s actual prejudice theory is simply too tenuous to withstand scrutiny. First, we are confident that trial counsel’s provision of the Dix reports to Dr. Sultan served to add to the expert’s credibility with the jury. Dr. Sultan’s use of the prior reports added to the substantive and temporal depth of her evaluation of McCarver. And, in particular, she was actually able to use positively McCarver’s answers on the Dix fill-in-the-blank to convey more fully to the jury her psychological profile of McCarver. Dr. Sultan testified that McCarver was misled by his cell mates who told him to provide crazy answers so that he would seem incompetent. J.A. 67-68. She also emphasized that after being reminded that he needed to be truthful, he finished the sentences differently on a subsequent fill-in-the-blank test. J.A. 70 (e.g., “The best time I ever had was when I was with Tina.”). Therefore, because Dr. Sultan had access to the Dix reports, including the specific fill-in-the-blank responses, the jury was provided with sufficient testimony by the defense expert not merely to discount McCaryer’s responses as not being sincere reflections of his thoughts about the murder, but also to believe that Dr. Sultan had a deep, credible understanding of McCarver as a troubled individual who could easily be misled by others.
Second, we recognize that Dr. Sultan’s credibility was invaluable not only in ameliorating any prejudicial impact of the Dix responses, but, more importantly, in helping McCarver prove the existence of many of the fourteen mitigating factors that the jury did find. That is, because Dr. Sultan was able to provide the jury with a detailed psychological picture of McCarver, by incorporating evaluations and reports made years before, the jury had reason to believe that Dr. Sultan’s testimony related to all of McCarver’s psychological problems was credible. And, Dr. Sultan’s testimony addressed at least six of the seventeen mitigating factors the jury did find, including McCarver’s capacity to appreciate his behavior, his level of intellectual functioning, and his suffering from depression. Therefore, it is reasonable to as[593]*593sume that even if the mere reference to the Dix reports caused the jury not to find remorse, the use of the Dix reports by Dr. Sultan may have aided the jury in finding the presence of many of the other mitigating circumstances. In other words, if Dr. Sultan had not consulted the prior studies and thereby increased the scope of her evaluation, the jury may have discredited her testimony regarding some if not six of the mitigating circumstances, thus counteracting any value gained from the possibility of the jury finding remorse. And, importantly, if the jury discredited Dr. Sultan’s testimony, it may have also discredited Dr. Sultan’s recitation of McCar-ver’s statement that he was “sorry” for what he did. J.A. 207.
Third, in addition to not finding remorse, the jury also failed to find two other mitigating circumstances to be present in McCarver’s case. The jury, even having found only two aggravating circumstances and fourteen mitigating circumstances, still recommended that MeCarver be sentenced to death. Given that the jury found only two aggravating circumstances to outweigh the fourteen mitigating circumstances, we are unpersuaded that had the jury found the presence of a fifteenth mitigating circumstance, remorse, it would be reasonable to believe that the jury would not have imposed the death penalty.
Even though we appreciate that it is not necessarily the case, it is at least relevant that the jury may very well have found through the testimony of various witnesses that MeCarver did express remorse, but accorded it no mitigating value anyway. This is because the mitigating circumstance question posed to the jury — “Consider whether Ernest Paul MeCarver has expressed remorse for his actions and whether you deem this to have mitigating value.” J.A. 311 (emphasis added)' — is actually a two-part question. It asks first, whether the jury believed that MeCarver expressed remorse for his actions, and second, whether the jury would then accord that remorse any mitigating value. The jury answered the two-part question with a single “no.” Therefore, it could well be the case that the jury did find that MeCar-ver had expressed remorse, answering the first question, but that it accorded that remorse no mitigating value in answering the second question. MeCarver has provided us with no reason to believe that a reasonable jury would accord the statements MeCarver allegedly made to various witnesses with any mitigating value at all, despite the statements’ apparent expression of remorse.
For the reasons stated above, we conclude that there is no categorical, ineffective assistance of counsel claim exception to our previous holding that section 15A-1419(a)(3) is an independent and adequate state procedural bar, and, therefore, that McCarver’s claim was proeedurally defaulted in state court, pursuant to an independent and adequate state procedural bar. Additionally, we are satisfied that MeCarver has demonstrated neither cause to excuse his procedural default nor actual prejudice resulting from the default.
B.
Although we conclude that McCarver’s ineffective assistance of counsel claim is proeedurally barred, we nevertheless consider the merits of that claim. As discussed, MeCarver argues that trial counsel was ineffective in two respects: his failure to file a motion in limine to ensure that the Dix reports could not be introduced, and his decision to give the Dix reports to McCarver’s own expert, Dr. Sultan, thereby ensuring that the district attorney would be permitted to ask about the reports on cross-examination. For the reasons that follow, we unhesitatingly conclude that McCarver’s trial counsel’s performance was not constitutionally ineffective.
Under Strickland v. Washington, 466 U.S. 668, 104 S.Ct. 2052, 80 L.Ed.2d 674 (1984), MeCarver must demonstrate that his counsel’s performance not only fell below an objective standard of reasonable[594]*594ness, see id. at 687-88, 104 S.Ct. 2052, but also that he was actually prejudiced by that ineffectiveness, see id. at 693-94, 104 S.Ct. 2052. The North Carolina Superior Court found that, in addition to being procedurally barred, McCarver’s ineffective assistance claim was without merit.6 J.A. 520-23. The district court, adopting the magistrate judge’s report and recommendation, agreed with the North Carolina court, holding that, even if McCarver’s claim had not been procedurally defaulted in state court, he did not demonstrate that counsel’s performance fell below an objective standard of reasonableness.
We address separately each of McCar-ver’s arguments on the merits, first, his claim that trial counsel erred by giving the Dix reports to Dr. Sultan; second, his argument that trial counsel should have filed a motion in limine to prevent the Dix reports from becoming part of the record; and, finally, his contention that trial counsel’s alleged ineffectiveness actually prejudiced him.
In evaluating trial counsel’s performance, we must be highly deferential to counsel’s strategic decisions and not allow hindsight to influence our assessment of counsel’s performance. See Strickland, 466 U.S. at 689, 104 S.Ct. 2052. Even were we not bound by this highly deferential standard, however, we would be satisfied that trial counsel’s decision to fully inform McCarver’s own expert, Dr. Sultan, of McCarver’s previous psychological evaluations did not fall below an objective standard of reasonableness. In fact, we believe that trial counsel’s performance would have bordered on the ineffective if he had not given the Dix reports to Dr. Sultan before she testified.7
Trial counsel’s decision to provide an expert with all available truthful information is not only imminently reasonable, but we would expect that an expert psychologist witness would always be provided with all available information on a patient. Particularly in a case in which the expert witness is not a long standing treating psychologist of the patient, we assume that it would be in the defendant’s best interest to have that expert obtain as much data and as many prior evaluations of that defendant, from as many sources as possible, as are available, so that the expert can form a full and credible picture of the defendant’s psychological make-up. Because of our understanding that generally it is in the best interests of the client for counsel to provide available background psychological information to an expert psychologist witness, we are not at all surprised that McCarver is unable to cite any [595]*595cases in which trial counsel was considered ineffective for providing accurate information to an expert witness.
That it is a sound and reasonable trial strategy to provide all available information to an expert psychologist witness is confirmed by the facts of this case. By making the reports available to Dr. Sultan, trial counsel ensured not only that Dr. Sultan’s evaluation would be as complete and as comprehensive as possible, J.A. 30-31 (Dr. Sultan testified: “It’s important when I’m trying to formulate an opinion about somebody for me to know what else has been said about that person. My opinion is really only as reliable as it is contrasted or compared with the overall impressions, the cumulative impressions that have been gathered by other professionals in the field over time. And so in part what I’m doing is double checking my notion of a person.”), but also that Dr. Sultan was able to incorporate the otherwise potentially damaging statements into her assessment of McCarver, ameliorating their impact while at the same time bolstering her own credibility as a defense witness, see supra Part II.A.2. Therefore, we conclude that McCarver’s trial counsel’s decision to give the reports to Dr. Sultan was, at the very least, not an objectively unreasonable trial strategy and that such did not constitute constitutionally ineffective assistance of counsel.8
2.
Because we are satisfied that trial counsel’s decision to turn over the Dix reports to the defense expert was a reasonable trial strategy, McCarver’s first allegation of error — that counsel should have filed a motion in limine to keep the reports out of the trial record — is without merit. That is, after counsel reasonably decided to give the reports to Dr. Sultan and thereby expose the reports to cross-examination,9 counsel could not also have filed a motion in limine to keep the same reports out of the record, and there would be no logical reason to have done so.
However, even if we were to conclude, as McCarver argues, that trial counsel should have attempted to exclude the Dix reports, the North Carolina Superior Court held that McCarver did “not demonstrate[ ] the existence of a solid foundation upon which .trial counsel could have demanded the exclusion of the evidence concerning the two reports.” J.A. 520.
McCarver argues primarily that there are two grounds on which trial counsel could have successfully moved the trial court to exclude the Dix reports from the record. First, he argues that under North Carolina law, the Dix reports were confidential and could not be made part of the record. Second, McCarver argues, in reliance upon Powell v. Texas, that the Dix reports could have been excluded from the record because the examinations performed at Dix went beyond the scope of a normal competency evaluation. We reject both claims.
(a)
McCarver cites N.C.G.S. § 15A-1002(d) for the proposition that the Dix [596]*596reports were not part of the public record and therefore that trial counsel could have invoked the provision to keep the reports confidential. Section 15A-1002(d) reads:
Any report made to the court pursuant to this section shall be forwarded to the clerk of the superior court in a sealed envelope addressed to the attention of a presiding judge.... A copy of the full report shall be forwarded to defense counsel, ... if the question of defendant’s capacity to proceed is raised at any time, a copy of the full report must be forwarded to the district attorney. Until such report becomes a public record, the full report to the court shall be kept under such conditions as are directed by the court, and its contents shall not be revealed except as directed by the court.
(emphasis added). McCarver argues that, because no hearing was ever held on McCarver’s competency, his capacity to proceed was not raised, and therefore copies of the full report should not have been sent to the district attorney. McCarver suggests therefore that trial counsel could have made a motion to exclude the reports from the record on the ground that the reports had been sent to the district attorney unlawfully.
However, section 15A-1002(d) is not an absolute confidentiality rule. To the contrary, the court is given the discretion to reveal the contents of the reports (“the full report to the court shall be kept under such conditions as are directed by the court, and its contents shall not be revealed except as directed by the court”), and the North Carolina Superior Court found that “the [Dr. Lara] report demonstrates that disclosure of the information in the reports to the district attorney was authorized by court order.” J.A. 513. Although there is no such court order in the record before us, the Dix report produced by Dr. Lara, which includes the relevant fill-in-the-blank responses, does indicate that such an order was made, J.A. 367 (“Court order instructed forensic report to be delivered to the patient’s attorneys, District Attorney’s Office, and to the judge.”), and we therefore do not find that the superior court’s factual conclusion was unreasonable. Accordingly, we do not find that the North Carolina court’s conclusion that the prosecution lawfully possessed the Dix reports prior to trial was in error, and McCarver’s counsel thus was not constitutionally ineffective for failing to make a motion to exclude the reports on the ground that the prosecution improperly possessed the reports or that they remained confidential.
(b)
MeCarver’s reliance on Powell v. Texas, 492 U.S. 680, 109 S.Ct. 3146, 106 L.Ed.2d 551 (1989), as a ground on which trial counsel could have moved the trial court to exclude the Dix reports from the record, is likewise misplaced. In Powell, the Supreme Court held that the Sixth Amendment was violated where the prosecution examined the defendant for future dangerousness during his competency evaluation, and defendant’s counsel neither knew of the future dangerousness testing nor was present. Presumably, McCarver is arguing that the competency reports could not be used in the sentencing stage of the trial because at the time the competency evaluations were performed, trial counsel was not present and did not know that the evaluations would include questions related to possible aggravating factors.10
However, McCarver’s ease is distinguishable from Powell in at least three ways. First, defense counsel, not the State, as in Powell, ordered the tests and there is no evidence that counsel was unaware of the kind of testing that would [597]*597take place. Second, there was no aggravating factor of future dangerousness in McCarver’s sentencing proceeding, as there was in Powell’s.
Third, in Poivell, the defendant was examined, at the State’s request and by order of the court, particularly for future dangerousness, in addition to competency. We are unconvinced that in McCarver’s case mere fill-in-the-blank sentences lie outside the scope of what a competency evaluation would entail. What McCarver fails to recognize in treating McCarver’s responses to the fill-in-the-blank sentences as obviously pertaining to the mitigating factor of remorse and not to competence is that the fill-in-the-blank sentences, before McCarver responded, were merely innocuous half sentences that could have as easily been left blank by McCarver or answered in the way he eventually did — “My favorite time was tvhen I was with Tina.” McCar-ver does not allege that the doctors asked him whether he ever dreams about killing, or any similar question that would necessarily yield responses that might be relevant to mitigating or aggravating factors, and we find no such questions in the record.11
For the reasons above, then, we conclude that the North Carolina Superior Court’s finding, that McCarver did “not demonstrate[ ] the existence of a solid foundation upon which trial counsel could have demanded the exclusion of the evidence concerning the two reports,” was not unreasonable. And, therefore, given that the district attorney already possessed the reports and could ask Dr. Sultan about them anyway, trial counsel’s decision to give the Dix reports to Dr. Sultan was entirely reasonable.
3.
We are satisfied, for the reasons stated in Part A.2., supra, that even if McCarver could demonstrate that counsel’s performance was constitutionally ineffective, he has not demonstrated actual prejudice resulting from counsel’s decision to give the Dix reports to the expert defense witness.
Accordingly, for the reasons stated, even if McCarver were able to bring his ineffective assistance of counsel claim in this court, despite its procedural default in state proceedings, we conclude that McCarver has demonstrated neither that trial counsel’s performance fell below an objective standard of reasonableness nor that actual prejudice resulted from trial counsel’s alleged constitutional ineffectiveness.
III.
McCarver also argues that he should have been afforded an evidentiary hearing by the district court on his ineffective assistance of counsel claim. The state court held no evidentiary hearing on his ineffective assistance claim, and the federal district court likewise rejected McCarver’s request for such a hearing. We conclude that the district court did not err in declining to conduct an evidentiary hearing on McCarver’s claim. Although McCarver may be correct that an evidentiary hearing would not be prohibited under 28 U.S.C. § 2254(e)(2), we do not believe an evidentiary hearing was required in his case, see Cardwell v. Greene, 152 F.3d 331, 337-38 (4th Cir.), cert. denied, 525 U.S. 1037, 119 S.Ct. 587, 142 L.Ed.2d 491 (1998); Mc[598]*598Donald v. Johnson, 139 F.3d 1056, 1059-60 (5th Cir.1998) (“[Ejven if[the petitioner’s] claim is not precluded by § 2254(e)(2), that does not mean he is entitled to an eviden-tiary hearing — only that he may be.”).
The district court is permitted to hold an evidentiary hearing only when the petitioner “alleges additional facts that, if true, would entitle him to relief.” Card-well, 152 F.3d at 331 (citation omitted) (internal quotation marks omitted). McCarver alleges that, at an evidentiary hearing, he could establish that trial counsel acted out of lack of knowledge of the issues, not due to a trial strategy; that Dr. Sultan could have conducted her evaluation without the Dix reports; that a legal expert believed trial counsel’s performance to be ineffective; and that the competency examinations went beyond the scope of a routine competency evaluation. See Appellant’s Br. at 48. However, we do not believe that any of the above “evidence” would aid McCarver.
First, regardless whether trial counsel was fully informed when he chose to provide the Dix reports to Dr. Sultan, his choice was a strategy — to inform fully his expert witness by providing truthful, relevant information to her. Second, whether Dr. Sultan could have evaluated McCarver without the reports is irrelevant to the inquiry whether the reports did aid her in making the evaluation, as she has already testified they did, and the reports did make her more credible than if she had merely testified based on her own temporally limited encounters with McCarver. Third, McCarver has not explicated how the testimony of a “legal expert” assessing trial counsel’s performance would aid a federal court in this particular case in making the legal determination whether trial counsel was constitutionally ineffective. If a habeas petitioner were required only to produce a “legal expert” to discuss ineffectiveness in order to obtain an evi-dentiary hearing, presumably all habeas petitioners would be entitled to such a hearing. Without a compelling reason why a criminal law expert on ineffectiveness would aid the court’s inquiry in McCarver’s case, we do not believe such an alleged “fact” warrants the holding of an evidentiary hearing. Finally, with respect to the scope of the competency evaluations, because McCarver’s counsel ordered the evaluations, even if they did go beyond the scope of normal competency examinations, there would be no ground for McCarver’s counsel to have moved to exclude the reports. That is, McCarver’s is not a case in which the prosecution ordered future dangerousness tests to use against him at sentencing.
Additionally, because, as discussed previously, we conclude that McCarver has not demonstrated actual prejudice resulting from any alleged constitutional errors, we also conclude that any facts McCarver could prove to bolster his ineffective assistance of counsel claim if he were afforded an evidentiary hearing, would not entitle him to relief. Therefore, the district court did not err in refusing to grant McCarver such a hearing.
CONCLUSION
For the foregoing reasons, we conclude that McCarver has failed to make a substantial showing of the violation of a constitutional right. His application for a certificate of appealability is denied and his petition for writ of habeas corpus is dismissed.
DISMISSED