concurring in part and dissenting in part.
I concur with the majority’s opinion except for Part III, which reverses the district court’s denial of habeas relief to Wilson on his ineffective assistance of counsel claim and grants him an evidentiary hearing. I conclude, even taking all of Wilsons allegations regarding ineffective assistance as true, counsel’s performance was not constitutionally inadequate and the district court consequently did not abuse its discretion in denying Wilson an evidentiary hearing on that claim.
I would therefore affirm the district court’s denial of habeas relief to Wilson in all respects.
I. Introduction
Wilson’s ineffective assistance of counsel claim presents two issues for our review: (1) whether Wilson is entitled to habeas relief on the claim, and (2) whether Wilson is entitled to an evidentiary hearing on the claim.
While Wilson presents these issues as analytically distinct, they are not; in deciding the first issue, we necessarily decide the second. The decision whether to grant an evidentiary hearing turns on whether the petitioner has alleged facts which would entitle him to habeas relief on the underlying claim. Schriro v. Landrigan, — U.S. -, 127 S.Ct. 1933, 1940, 167 L.Ed.2d 836 (2007); see also Mayes v. Gibson, 210 F.3d 1284, 1287-88 (10th Cir.2000). “In deciding whether to grant an evidentiary hearing, a federal court must consider whether such a hearing could enable an applicant to prove the petition’s factual allegations, which, if true, would entitle the applicant to federal habeas relief.” Schriro, 127 S.Ct. at 1940. Because of the convergence between the standard for evidentiary hearings and the standard for habeas relief, in the course of explaining why Wilson is not entitled to habeas relief on his ineffectiveness claim, I will also explain why he is not entitled to an evidentiary hearing.
In this opinion, I will first discuss the standards of review applicable to Wilson’s appeal. I will then show why Wilson is not entitled to habeas relief on his claim of ineffective assistance. Finally, I will explain why the district court properly denied Wilson an evidentiary hearing on his claim.
*1126 II. Standard of Review
We apply two standards of review in considering Wilson’s ineffectiveness claim. The district court’s decision to deny habe-as relief on the claim is a legal question we review de novo. Fricke v. Sec’y of the Navy, 509 F.3d 1287, 1289 (10th Cir.2007); Maynard v. Boone, 468 F.3d 665, 669 (10th Cir.2006). But the court’s decision not to grant an evidentiary hearing is reviewed only for an abuse of discretion. Coronado v. Ward, 517 F.3d 1212, 1217 (10th Cir. 2008), petition for cert. filed, — U.S.L.W. -, (U.S. Apr. 23, 2008) (No. 07-11293); Anderson v. Att’y Gen. of Kan., 425 F.3d 853, 858 (10th Cir.2005). These appellate standards of review are employed in conjunction with the standards all federal courts must use when collaterally reviewing state court judgments. Following AEDPA’s clear commands, a federal court may grant a state habeas petitioner relief only if the state court’s judgment involved an unreasonable application of federal law or unreasonable determination of the facts of his case. Miller-El v. Dretke, 545 U.S. 231, 240, 125 S.Ct. 2317, 162 L.Ed.2d 196 (2005); Lockyer v. Andrade, 538 U.S. 63, 70-73, 123 S.Ct. 1166, 155 L.Ed.2d 144 (2003).
The twin layers of our standard of review make sense considering this court’s position in the judicial hierarchy. First, as a federal court reviewing a state court’s final criminal judgment, our sole role is to uphold the United States Constitution; errors of state law are not within our purview. Principles of federalism, comity, and finality compel us to accord proper deference to the state court’s judgment. See, e.g., Williams v. Taylor, 529 U.S. 420, 436, 120 S.Ct. 1479, 146 L.Ed.2d 435 (2000). Second, as a court of appeals reviewing a lower court’s decision, our main role is correcting errors of law and abuses of judicial discretion. See, e.g., Anderson v. Bessemer City, N.C., 470 U.S. 564, 574-75, 105 S.Ct. 1504, 84 L.Ed.2d 518 (1985). We are not a court of first resort, but one of (nearly) last resort.
The majority, without according proper deference to either the state or district court’s judgments, conducts a de novo review of the entire record before us. The majority is wrong for two reasons: (1) it fails to correctly apply the abuse of discretion standard to the district court’s decision denying an evidentiary hearing, and (2) it fails to accord AEDPA deference to the state court’s judgment on the merits.
The Supreme Court has reaffirmed that we review a district court’s decision to deny an evidentiary hearing for an abuse of discretion only. The Court held, “In cases where an applicant for federal habe-as relief is not barred from obtaining an evidentiary hearing by 28 U.S.C. § 2254(e)(2), the decision to grant such a hearing rests in the discretion of the district court.” Schriro, 127 S.Ct. at 1937.1 Prior to the enactment of AEDPA, the decision to grant an evidentiary hearing “was generally left to the sound discretion of district courts.” Id. at 1939. “That basic rule has not changed.” Id. (citing 28 U.S.C. § 2254, Rule 8(a)). Therefore, it is clear that today, as always, we review the district court’s decision to grant or deny a habeas petitioner’s request for an eviden-tiary hearing for abuse of discretion. Id.; see also Coronado, 517 F.3d at 1217; Anderson, 425 F.3d at 858.
*1127It is equally clear that when we review a state court decision on the merits, AED-PA’s deferential standards apply. “Whether [petitioner’s] allegations, if proven, would entitle him to habeas relief is a question governed by [AEDPA].” Mayes, 210 F.3d at 1287-88; Schriro, 127 S.Ct. at 1940 (noting “the deferential standards prescribed by § 2254 control whether to grant habeas relief’). This deference is appropriate whether or not the state court granted petitioner an evidentiary hearing. See 28 U.S.C. § 2254(d) (mandating a deferential standard of review “-with respect to any claim that was adjudicated on the merits in State court proceedings,” without regard to whether an evidentiary hearing was held); see also Schriro, 127 S.Ct. at 1938-39; Hammon v. Ward, 466 F.3d 919, 926 n. 7, 928, 931 (10th Cir.2006). In Schriro, for example, although the state court refused to grant an evidentiary hearing before considering the merits of petitioner’s claims, the Court nevertheless applied § 2254(d)’s deferential standard of review. 127 S.Ct. at 194(M4.
Relying upon Schriro, several other circuits have recently acknowledged deference must be accorded state court judgments even where the state court did not hold an evidentiary hearing. The First Circuit, in a case like this one dealing with ineffective assistance, noted federal courts must accord AEDPA deference to state judgments when deciding whether an evi-dentiary hearing is appropriate. See Teti v. Bender, 507 F.3d 50, 62 (1st Cir.2007), cert. denied, — U.S.-, 128 S.Ct. 1719, 170 L.Ed.2d 525 (2008). The court held, “[Petitioner] must allege more than that he received inadequate assistance; he must allege facts sufficient to overcome AEDPA deference to the state court’s fact-findings and legal conclusion to the contrary.” Id. The Sixth and Ninth Circuits have also noted the deferential standards of § 2254(d) must be taken into account when determining whether petitioner has made out a claim for habeas relief. See Estrada v. Scribner, 512 F.3d 1227, 1235 (9th Cir.2008), cert. denied, — U.S.-, 128 S.Ct. 2973, — L.Ed.2d- (2008); Ivory v. Jackson, 509 F.3d 284, 298 (6th Cir.2007), cert. denied, — U.S.-, 128 S.Ct. 1897, 170 L.Ed.2d 765 (2008).
The majority suggests that because Wilson was diligent in his efforts to obtain an evidentiary hearing in state court, but was denied that hearing, de novo review is appropriate. This is not so. Diligence does not control our standard of review. Rather, the requirement that a petitioner show diligence is merely one of two procedural hurdles a petitioner must cross before receiving an evidentiary hearing in federal court. After the enactment of AEDPA, a federal habeas petitioner must show he was diligent in developing the factual basis for his claim in state court (hurdle number one), and that his allegations, if true, would entitle him to habeas relief (hurdle number two).2 In a case like this, where the state court has decided petitioner’s claim on the merits, this sec*1128ond hurdle is even higher because petitioner must show the state court’s adjudication of his claim involved an unreasonable determination of the facts or unreasonable application of clearly established federal law.3
The majority relies heavily upon our precedents in Bryan v. Mullin, 335 F.3d 1207, 1215 (10th Cir.2003) (en banc), and Miller v. Champion, 161 F.3d 1249, 1254 (10th Cir.1998), to conclude we review Wilson’s ineffective assistance of counsel claim without deference to the state court’s judgment. But to the extent those cases rely on § 2254(e)(2)’s diligence inquiry to determine our standard of review, they are no longer good law. Compare Bryan, 335 F.3d at 1215 (concluding that because petitioner was diligent, “the appropriate question is whether [petitioner] was entitled to a hearing under pre-AEDPA law”), with Schriro, 127 S.Ct. at 1940 (concluding that even though petitioner was diligent, “[b]e-cause the deferential standards prescribed by § 2254 control whether to grant habeas relief, a federal court must take into account those standards in deciding whether an evidentiary hearing is appropriate”). Whether the state court adjudicated petitioner’s claim on the merits — not whether petitioner was diligent — guides our standard of review. See 28 U.S.C. § 2254(d); Schriro, 127 S.Ct. at 1939-44 & n. 1; Turrentine v. Mullin, 390 F.3d 1181, 1188-89 (10th Cir.2004) (quoting 28 U.S.C. § 2254(d)(1) and (d)(2)).
Wilson’s claim of ineffective assistance was adjudicated on the merits in Oklahoma state court. “An adjudication on the merits occurs when the state court resolves the case on substantive grounds, rather than procedural grounds.” Valdez v. Cockrell, 274 F.3d 941, 946-47 (5th Cir.2001) (internal quotation marks omitted); see also Harris v. Poppell, 411 F.3d 1189, 1195-96 (10th Cir.2005) (noting “[wjhere there is no indication ... the state court did not reach the merits of a claim, we have held that a state court reaches a decision ‘on the merits’ ” (quotation omitted)). The Oklahoma Court of Criminal Appeals (OCCA) concluded Wilson could not show his trial counsel’s performance was deficient or prejudicial under the Supreme Court’s precedent in Strickland v. Washington, 466 U.S. 668, 104 S.Ct. 2052, 80 L.Ed.2d 674 (1984). Wilson v. State, 983 P.2d 448, 471-72 (Okla.Crim.App.1998). The OCCA’s decision was based on substantive, rather than procedural, grounds. Id. at 472. Despite this clear holding, the majority applies our Tenth Circuit cases to conclude the decision does not constitute an adjudication on the merits because the OCCA did not consider additional non-record evidence submitted by Wilson. But nothing in the Supreme Court’s precedents or AEDPA’s text supports the position that a decision ceases to be an adjudication on the merits simply because the petitioner has submitted additional evidence to the federal habeas court.4 As noted above, § 2254 deference *1129applies whether or not the state court held an evidentiary hearing to allow petitioner to supplement the state court record.
This conclusion is bolstered by our cases holding that the OCCA’s denial of an evi-dentiary hearing under that court’s Rule 3.11 bears on the substantive Strickland analysis. In both Bland v. Sirmons, 459 F.3d 999 (10th Cir.2006), and Welch v. Sirmons, 451 F.3d 675 (10th Cir.2006), we concluded the OCCA had adjudicated a claim on the merits because the court decided petitioner failed to meet the substantive standard for obtaining an evidentiary hearing under Rule 3.11. See Bland, 459 F.3d at 1030 (conducting a deferential review where “the OCCA considered the claim in light of [petitioner’s] request for an evidentiary hearing”); Welch, 451 F.3d at 709 (conducting a deferential review where the OCCA determined an evidentia-ry hearing was not warranted under Rule 3.11); see also Mayes, 210 F.3d at 1287-88. Because the standard for obtaining an evi-dentiary hearing under Rule 3.11 is lower than the standard set forth in Strickland— petitioner need only show a “strong possibility” of ineffective assistance5 — when the OCCA denies an evidentiary hearing under Rule 3.11, it necessarily makes a merits determination petitioner cannot meet the substantive Strickland standard. The OCCA need not use any special words in reaching this conclusion; our cases look to the result of a state court’s decision, not to its reasoning. See, e.g., Aycox v. Lytle, 196 F.3d 1174, 1177 (10th Cir.1999).
Here, although the OCCA did not explain why it denied Wilson an evidentiary hearing, it did conclude “an evidentiary hearing on this claim should be denied.” Wilson, 983 P.2d at 472 n. 8. The OCCA thus concluded the requirements of Rule 3.11 had not been met. Looking to the OCCA’s result, we must conclude the OCCA determined an evidentiary hearing would not have changed the court’s determination that counsel’s performance was adequate under Strickland. See id. at 471-72. This is a state court decision on the merits to which we owe AEDPA deference. See 28 U.S.C. § 2254(d).
In sum, a petitioner’s claim may have been adjudicated on the merits, and therefore subject to AEDPA deference, even if petitioner makes allegations in federal ha-beas court based on new evidence (1) presented in the form of affidavits (as here), or (2) arising out of a federal evidentiary hearing. See, e.g., Schriro, 127 S.Ct. at 1938-39; Matheney v. Anderson, 377 F.3d 740, 747 (7th Cir.2004). The new evidence can be considered by the federal habeas court to determine whether the state court reached an unreasonable determination of clearly established federal law or the facts of petitioner’s case — but not to eliminate AEDPA’s deferential standard of review.
Because Wilson’s claim was adjudicated on the merits in Oklahoma state court, AEDPA’s § 2254(d) applies to Wilson’s federal habeas appeal. To prevail under this deferential standard, Wilson must show the state court’s decision “was contrary to, or involved an unreasonable application of, clearly established Federal law, as determined by the Supreme Court of the United States,” or “was based on an unreasonable determination of the facts in *1130light of the evidence presented in the State court proceeding.” 28 U.S.C. § 2254(d).
As the Supreme Court has emphasized, “an unreasonable application of federal law is different from an incorrect application of federal law.” Williams v. Taylor, 529 U.S. 362, 410, 120 S.Ct. 1495, 146 L.Ed.2d 389 (2000); see also Schriro, 127 S.Ct. at 1939 (noting the unreasonableness standard presents a “substantially higher threshold” than incorrectness). “[0]nly the most serious misapplications of Supreme Court precedent will be a basis for relief under § 2254.” Maynard, 468 F.3d at 671. “[T]he state court decision must be ‘at such tension with governing U.S. Supreme Court precedents, or so inadequately supported by the record, or so arbitrary as to be unreasonable.’ ” Id. (quotation omitted).
:|c * *
The majority incorrectly conducts a de novo review of the state court’s judgment without deference to the court’s legal or factual conclusions. Applying the correct standard of review, this court should ask whether Wilson is entitled to federal habe-as relief because the state court’s judgment was in some way unreasonable; and, if so, whether the district court abused its discretion in denying Wilson an evidentia-ry hearing. In answering these questions, we apply twin layers of review, taking into account our role as a federal court reviewing a state court judgment and as a court of appeals reviewing a lower court’s decision.
Bounded by these important parameters, our review of the district court’s decision to deny habeas relief equates to (1) an independent review (2) of the state court’s judgment (3) in which we ask whether the decision “was contrary to, or involved an unreasonable application of, clearly established Federal law,” 28 U.S.C. § 2254(d)(1), or “was based on an unreasonable determination of the facts,” id. § 2254(d)(2). Further, our review of the district court’s decision to deny an eviden-tiary hearing equates to (1) a deferential review (2) of the district court’s decision (3) in which we ask whether the district court abused its discretion in concluding petitioner’s allegations, if true, would not entitle him to federal habeas relief.
III. Ineffective Assistance of Counsel
I agree with the district court Wilson is not entitled to habeas relief on his ineffective assistance of counsel claim. We evaluate counsel’s efforts under the now-familiar two-part standard announced in Strickland v. Washington, 466 U.S. 668, 104 S.Ct. 2052, 80 L.Ed.2d 674 (1984), and may deem counsel ineffective only if “counsel’s conduct so undermined the proper functioning of the adversarial process that the trial cannot be relied on as having produced a just result.” Strickland, 466 U.S. at 686, 104 S.Ct. 2052. Wilson must meet both prongs of the two-part test to prevail: “First, [Wilson] must show that counsel’s performance was deficient .... Second, [Wilson] must show that the deficient performance prejudiced the defense.” Id. at 687, 104 S.Ct. 2052 (emphasis added). Because “the proper standard for attorney performance is that of reasonably effective assistance ... [Wilson] must show that counsel’s representation fell below an objective standard of reasonableness.” Id. at 687-88, 104 S.Ct. 2052.
The OCCA concluded the performance of Wilson’s trial counsel was reasonable under Strickland because counsel “put forth a mental health expert to rebut the State’s continuing threat contention and to mitigate punishment,” and that expert had access to Wilson’s mother, Wilson’s medical records, school records, and statements from other people who knew Wilson. Wil*1131son, 983 P.2d at 472. The OCCA concluded, “The mere fact more evidence could have been presented is not, in itself, sufficient to show counsel was deficient.” Id. The court also determined Wilson had failed to show prejudice.
The district court concluded the OCCA’s adjudication of Wilson’s claim on the merits did not run afoul of § 2254(d). The court held, “[T]he OCCA’s rejection of this [ineffectiveness] claim on direct appeal was not an unreasonable application of the legal principle announced by the Supreme Court in Strickland to the facts of Petitioner’s case. Petitioner has failed to satisfy the § 2254(d) standard on this portion of his ineffective assistance of counsel claim.” Wilson v. Sirmons, No. 00-147, 2006 WL 2289777, at *43 (N.D.Okla. Aug.8, 2006).
I agree with the district court that the OCCA’s determination of Wilson’s claim was neither “an unreasonable application of[] clearly established Federal law,” or “based on an unreasonable determination of the facts in light of the evidence presented in the State court proceeding.” 28 U.S.C. § 2254(d). The district court thus correctly determined Wilson was not entitled to habeas relief on his ineffectiveness claim.
After reviewing all the evidence before us, including the additional affidavits submitted by Wilson in conjunction with his habeas petition,6 it is clear Wilson is not entitled to relief. First, exercising “a heavy measure of deference to counsel’s judgments,” Strickland, 466 U.S. at 691, 104 S.Ct. 2052, I cannot conclude counsel’s representation failed to meet minimal professional standards. Second, accepting Wilson’s counsel should have known — but apparently did not — about the supplemental mental health evidence Wilson presents in the affidavits, I cannot conclude Wilson was prejudiced by any deficiency. Even if counsel knew a more specific diagnosis were possible with additional testing, counsel knew enough about Wilson’s mental health profile to reasonably pursue a mitigation strategy that de-emphasized mental health in favor of what he considered a more promising approach. Reviewing additional evidence would not have substantially altered the portrait of Wilson counsel presented to the jury.
The Supreme Court has “declined to articulate specific guidelines for appropriate attorney conduct and instead ha[s] emphasized that ‘[t]he proper measure of attorney performance remains simply reasonableness under prevailing professional norms.’ ” Wiggins v. Smith, 539 U.S. 510, 521, 123 S.Ct. 2527, 156 L.Ed.2d 471 (2003) (quoting Strickland, 466 U.S. at 688, 104 S.Ct. 2052). There is a strong presumption “an attorney acted in an objectively reasonable manner and that an attorney’s challenged conduct might have been part of a sound trial strategy.” Bullock v. Carver, 297 F.3d 1036, 1046 (10th Cir.2002) (emphasis omitted). “There are countless ways to provide effective assistance in any given case. Even the best criminal defense attorneys would not defend a particular client in the same way.” Strickland, 466 U.S. at 689, 104 S.Ct. 2052.
To the extent counsel conducts a less-than-complete investigation to uncover potentially mitigating evidence, counsel’s investigation remains constitutionally acceptable “precisely to the extent that reasonable professional judgments support the limitations on investigation.” Strickland, 466 U.S. at 691, 104 S.Ct. 2052. “In other words, counsel has a duty to make reasonable investigations or to *1132make a reasonable decision that makes particular investigations unnecessary.” Wiggins, 539 U.S. at 521, 123 S.Ct. 2527. Although we of course defer to counsel’s strategic judgments, Strickland, 466 U.S. at 689, 104 S.Ct. 2052, strategic judgments are not reasonable unless backed by a reasonable investigation. Fisher v. Gibson, 282 F.3d 1283, 1296 (10th Cir.2002) (citing Strickland, 466 U.S. at 691, 104 S.Ct. 2052).
Keeping these principles in mind, I turn to whether trial counsel’s performance at the mitigation phase was deficient and prejudicial under Strickland.
A. Deficient
Wilson contends his counsel’s investigation was unreasonable because counsel did not obtain a diagnosis of schizophrenia, which would have been possible had the defense expert, Dr. Reynolds, done additional testing and interviewed additional witnesses. I disagree, and for three reasons conclude counsel exercised reasonable professional judgment in this case.
First, the record does not indicate counsel knew or should have known further investigation was necessary. If counsel had no reason to think the information he had was incomplete and that additional diagnoses were possible, counsel reasonably developed his trial strategy based on the information the expert examination had already produced. Glaringly absent in the record is any statement from Wilson’s counsel that he did not have enough time to obtain a further diagnosis. Wilson’s counsel did, in fact, submit an affidavit— but only about the dual jury issue. The record before us lacks the single most accessible and helpful source of information to make an informed and non-speculative conclusion about trial counsel’s performance.7
Second, counsel obtained sufficient information about Wilson’s mental health to make a reasonable decision about trial strategy, which he acted on at the mitigation phase. Once counsel decided to focus his mitigation strategy on Wilson’s high intelligence and capacity for reform, it was reasonable for counsel not to pursue further leads of mental illness.
Finally, we have no reason to believe on this record that the amount of time counsel allotted for the investigation unreasonably limited the information counsel was able to discover. Trial counsel developed a theory of mitigation based on knowledge of Wilson’s personal history. As I discuss below, the record supports the conclusion counsel adequately put forward a defense based on his theory of the case.
During the penalty phase of a death penalty case, effective assistance requires counsel to make reasonable efforts to determine whether a defendant’s mental health presents a plausible argument against imposing the death penalty. See Smith v. Mullin, 379 F.3d 919, 942 (10th Cir.2004) (describing mental health evidence as “of vital importance to the jury’s decision at the punishment phase”). “In assessing the reasonableness of an attorney’s investigation,” we consider “the quantum of evidence already known to counsel, [and] also whether the known evidence would lead a reasonable attorney to investigate further.” Wiggins, 539 U.S. at 527, 123 S.Ct. 2527.
Counsel conducted a constitutionally sufficient inquiry into Wilson’s mental health, *1133and the record does not reveal that counsel had additional information that should have led to further testing or further witness interviews.
1. Counsel Conducted Adequate Testing and Interviews
Wilson argues trial counsel faltered by failing to order further mental health testing and by failing to personally interview several family members. Neither claim has constitutional merit.
Further Testing
When investigating a defendant’s mental health, counsel by necessity often relies on expert assistance. See, e.g., American Bar Association, American Bar Association Guidelines for the Appointment and Performance of Counsel in Death Penalty Cases 11.4.1 (1989) [hereinafter ABA Guidelines ] (“Counsel should secure the assistance of experts where it is necessary or appropriate” for the defense.). Wilson concedes counsel engaged a qualified clinical psychologist, Dr. Reynolds, to evaluate him. Dr. Reynolds met with Wilson on three separate occasions, during which time he administered multiple psychological tests. Reynolds also met with Wilson’s mother and had access to Wilson’s medical records, school records, and statements from people who knew Wilson well, including a teacher, a fellow church member, and a long-time family friend. Counsel and Dr. Reynolds discussed these findings before trial.
In the Supreme Court cases relied upon by the majority, trial counsel did not perform nearly as well. In Rompilla v. Beard, for example, counsel failed to investigate the defendant’s prior convictions for rape and assault, despite knowing the state intended to introduce those convictions as aggravating factors at sentencing. 545 U.S. 374, 383-87, 125 S.Ct. 2456, 162 L.Ed.2d 360 (2005). As the Court noted, “looking at a file the prosecution says it will use is a sure bet.” Id. at 389, 125 S.Ct. 2456. In Wiggins v. Smith, moreover, counsel was deficient for failing to expand his investigation beyond readily available materials. Counsel’s knowledge of the defendant’s life history rested exclusively on the court-created presentence report and foster care records supplied by the city of Baltimore. 539 U.S. at 523, 123 S.Ct. 2527. Although funds were made available for counsel to retain a forensic social worker to investigate the defendant’s background and prepare a report, counsel chose not to commission such a report. Id. at 524., 123 S.Ct. 2527 This paltry investigation stood in stark contrast to the “standard practice in Maryland in capital cases at the time.” Id.
Our Tenth Circuit cases have also faulted counsel for doing far less. See, e.g., Anderson v. Sirmons, 476 F.3d 1131, 1143 (10th Cir.2007) (finding ineffective assistance when defendant was never evaluated by an expert); Smith, 379 F.3d at 939 (finding ineffective assistance when counsel was unaware “mental state or mental illness could be introduced as mitigation”).
In determining what would lead a reasonable attorney to order additional rounds of mental health tests, we may expect counsel to rely on the opinion of a mental health expert. See Bell v. Thompson, 545 U.S. 794, 809-10, 125 S.Ct. 2825, 162 L.Ed.2d 693 (2005) (suggesting defendant “would have faced an uphill battle” to convince a court the mental health investigation should have continued despite an expert opinion defendant was not mentally ill); Clark v. Mitchell, 425 F.3d 270, 285 (6th Cir.2005) (“It was not unreasonable for ... counsel, untrained in the field of mental health, to rely on the opinions of these professionals.”). Even Strickland counseled that when determining the appropriate scope of an investigation, counsel should rely on the information already ob*1134tained from the defendant and others. 466 U.S. at 691,104 S.Ct. 2052.
Wilson’s counsel appropriately relied on Dr. Reynolds to decide how many rounds of mental health testing should be conducted. The record does not reveal that Dr. Reynolds ever advised counsel further testing beyond the initial round was necessary or advisable. Nor did Dr. Reynolds advise further investigation might yield any definitive diagnoses. Only after trial did Dr. Reynolds indicate he needed additional information to support a schizophrenia diagnosis — apparently because of an invalid MMPI-2 test. Even in his post-trial affidavit, however, Dr. Reynolds does not say he told counsel about the invalid test or advised counsel that further testing or investigation would be helpful. Counsel in his affidavit does not indicate he received such advice, and we do not have any report Dr. Reynolds prepared for counsel prior to giving his testimony.8
Even if Dr. Reynolds had conducted more testing, there is no reason to think counsel’s picture of Wilson would have changed. Dr. Reynolds’s affidavit, written after he performed a new battery of tests on Wilson after trial, says only it was “possible ” Wilson “could have been delusional at the time of the crime.” Aplt. Add. 2 (emphasis added). Dr. Reynolds noted there was also a “possibility” Wilson suffered from “delusions or hallucinations.” Id. at 2. These generalized possibilities do not show, in any way, that if Wilson’s counsel had given Dr. Reynolds more time to conduct more tests counsel’s picture of Wilson would have changed in the least. The majority’s insistence that further testing was necessary is pure speculation.
“ ‘An ambiguous or silent record is not sufficient to disprove the strong and continuing presumption’ that counsel’s performance was reasonable and that counsel made all significant decisions in the exercise of reasonable professional judgment.” Sallahdin v. Mullin, 380 F.3d 1242, 1251-52 (10th Cir.2004) (quoting Chandler v. United States, 218 F.3d 1305, 1314 n. 15 (11th Cir.2000)). Without any evidence that counsel disregarded the expert’s professional judgment, I conclude counsel reasonably believed he had fulfilled his obligation to investigate Wilson’s mental health by hiring an expert and considering the expert’s conclusions. Compare Burger v. Kemp, 483 U.S. 776, 793, 107 S.Ct. 3114, 97 L.Ed.2d 638 (1987) (finding no ineffective assistance when record left ambiguity about reasonableness of counsel’s decision), with Hooper v. Mullin, 314 F.3d 1162, 1171 (10th Cir.2002) (finding ineffective assistance when counsel disregarded expert report explicitly recommending further diagnostic investigation).9 Under the circumstances here, and because of counsel’s reasonable strategic choices discussed *1135below, counsel’s failure to pursue further diagnosis was not unreasonable.
Further Interviews
I also disagree with the conclusion counsel was unreasonable for failing to probe deeper into Wilson’s mental health history through additional witness interviews. Wilson argues counsel’s investigation was insufficient because counsel failed to uncover the information contained in the post-conviction affidavits of Wilson’s mother, girlfriend, brother, and sister. Dr. Reynolds’s post-conviction affidavit additionally suggests this information would have been helpful to a diagnosis of schizophrenia.
Strickland imposes only the obligation to conduct a reasonable investigation — not a perfect one. 466 U.S. at 688, 104 S.Ct. 2052. The record indicates counsel gathered information from a number of people who knew Wilson. For example, counsel provided Dr. Reynolds with statements from a close friend of Wilson’s family, a fellow church-member, and two of Wilson’s former teachers. All testified on Wilson’s behalf at trial.
Counsel also provided Dr. Reynolds with statements from two additional individuals who knew Wilson, but the record does not reveal the substance of their information. And counsel had access to information from Wilson’s mother through Dr. Reynolds, who personally interviewed her. Wilson’s mother testified at trial and apparently provided counsel with names of individuals who might testify on his behalf. The record does not reveal that counsel failed to pursue any of these leads. Compare Burger, 483 U.S. at 794, 107 S.Ct. 3114 (finding no ineffective assistance where counsel interviewed “all potential witnesses who had been called to his attention”), with Mayes, 210 F.3d at 1290 (finding ineffective assistance when counsel “never contacted any mitigation witnesses”).
From all of these witnesses, counsel obtained significant mitigating information about Wilson’s life demonstrating a constitutionally adequate investigation. That counsel used this information to present a constitutionally adequate mitigation defense is shown by the following two points.
First, counsel presented a full picture of Wilson to the jury. Counsel proffered the following evidence: (1) Wilson’s friends and family knew him as a kind, caring, church-going person; (2) Wilson’s father, a drug addict and alcoholic, abandoned the family when Wilson was young; (3) Wilson’s mother was a strong positive influence; (4) a Sunday school teacher close to Wilson died of cancer, which extremely upset Wilson; (5) Wilson was exposed to significant gang activity growing up, and in one incident Wilson was shot in the leg; (6) Wilson’s home, where he was living with his mother, was torched by a rival gang; (7) Wilson’s brother was in jail; and (8) Wilson lived with his sister in North Carolina for some time and did very well when removed from the environment of his violent neighborhood. The bulk of the affidavit testimony Wilson offers in his habe-as petition simply repackages the information counsel actually presented to the jury. This repetition suggests counsel did a reasonably thorough job of uncovering the major contours of Wilson’s family and social history. This is certainly not a case where counsel sat idly by, thinking investigation would be futile.
Second, all of the witnesses who provided mitigation evidence at trial were also in a position to observe the kind of strange behavior Wilson now asserts his counsel should have uncovered through additional interviews. Yet the record does not show that anyone during counsel’s investigation mentioned Wilson’s extreme conduct or otherwise provided information to counsel that should have led to such inquiries. *1136Because counsel had information from a wide range of people with knowledge of Wilson’s family and personal history, none of whom mentioned a serious mental health issue beyond that diagnosed by Dr. Reynolds, it was reasonable for counsel not to interview additional witnesses. Counsel had no reason to think additional witnesses would offer helpful non-cumulative testimony. See Rompilla, 545 U.S. at 389, 125 S.Ct. 2456 (“Questioning a few more family members and searching for old records can promise less than looking for a needle in a haystack, when a lawyer truly has reason to doubt there is any needle there.”).
To the extent counsel was obligated to seek out any behavioral evidence of mental illness, counsel fulfilled his obligation by engaging Dr. Reynolds and providing him with access to Wilson’s mother, other witnesses, and Wilson’s records. Counsel could reasonably have expected Dr. Reynolds to obtain any information he needed about Wilson’s behavior from Wilson’s mother, who lived with Wilson and was very close to him. If Dr. Reynolds thought further interviews would be helpful, he could have suggested them to counsel, but we have no information that Dr. Reynolds did so. That Dr. Reynolds later found information provided by additional family members helpful in corroborating his own conclusions does not make the initial investigation unreasonable; reasonableness is evaluated based on the information available to counsel at the time. Strickland, 466 U.S. at 689, 104 S.Ct. 2052.
Wilson’s mother’s post-trial affidavit claims counsel never personally interviewed her prior to trial. Failure to interview a witness prior to her testimony at trial can in some circumstances constitute ineffective assistance of counsel. See Hooper, 314 F.3d at 1171. In this case, however, counsel’s failure to personally interview the mother was not unreasonable. Counsel had access to her testimony through Dr. Reynolds and reasonably relied on Reynolds’s interviews. It is well-settled counsel may rely on the efforts of co-counsel, investigators, and experts in preparing for trial, e.g., Clark, 425 F.3d at 286, and that the failure to conduct personal interviews is not necessarily deficient performance, ABA Guidelines 11.4.1(3) (interviewing potential witnesses). In fact, ABA Guideline 11.4.1(3) instructs capital defense counsel to “conduct interviews of potential witnesses in the presence of a third person” or rely on “an investigator or mitigation specialist” so there is someone to call “as a defense witness at trial.” This is precisely the course Wilson’s trial counsel followed.10
Perhaps the most cautious of counsel would always interview each member of the defendant’s family as well as other close contacts, but the Supreme Court has declined to make prudence the measure of constitutionally effective counsel. “[I]n *1137considering claims of ineffective assistance of counsel, we address not what is prudent or appropriate, but only what is constitutionally compelled.” Burger, 483 U.S. at 794, 107 S.Ct. 3114 (internal quotations omitted). “[CJounsel need not interview every possible witness to have performed proficiently.” Young v. Sirmons, 486 F.3d 655, 680 (10th Cir.2007).
Counsel need only conduct a reasonable investigation. I cannot say that counsel’s investigative efforts in this case — which included collecting information from Wilson’s mother, teachers, and others close to the family — were unreasonable, particularly because counsel’s investigation evidently uncovered the bulk of the mitigating information Wilson’s new affidavits offer. On this record, counsel had no reason to think additional interviews would provide helpful noncumulative evidence, and counsel’s failure to conduct additional interviews was not constitutionally ineffective.
Finally, the affidavit evidence the majority claims would have been helpful in the mitigation case is, at best, a two-edged sword.11 See Bryan, 335 F.3d at 1216 & n. 21; Cannon v. Gibson, 259 F.3d 1253, 1277-78 (10th Cir.2001). The majority claims the affidavit evidence could have supported a finding of schizophrenia, which might have elicited sympathy from the jury. Perhaps. On the other hand, though, the jury could have drawn a negative picture about Wilson’s gang involvement and the murder itself. Wilson’s brother describes his own gang involvement and how Wilson became involved along with him. He said he and Wilson “both put on our gang face when we went out the front door” and talked of Wilson being targeted by “rival gangs.” Aplt. Add. 5. The brother also said Wilson “knew that the gang considered him a police rat and his reputation was permanently ruined. I never heard anything specific, but I am sure there was teasing and serious pressure to do this crime.” Id.
Wilson’s sister said she believed Wilson “was being pulled into the gang scene by at least his ninth grade year of high school. I remember [Wilson] got real suspicious and paranoid after he joined the gang.... I know that [Wilson] was present when some of his fellow gang members were killed.” Aplt. Add. 6. She also corroborated the idea the gang thought Wilson was a rat:
I knew that [Wilson] had been arrested on another offense prior to his trial.... [Wilson] fumbled with the decision of what to say to police because the gang saw it as ratting out one of their own.... [Wilson] told me that the [sic ] he was in danger with the gang. The gang threatened him and made it clear he had to prove he was with them. I believe [Wilson] was pressured and coerced into the Quick Trip crime.
Id.
Finally, Wilson’s girlfriend refers to Wilson having trouble in prison because of a rumor he “snitched” on two Bloods until “word was out [Wilson] was back on the same side.” Aplt. Add. 4.
The affidavits present a disturbing picture of the murder in this case — Wilson *1138was in trouble with his gang because of the perception he had “snitched” in a prior case, and Wilson got involved in the crime to prove his loyalty. If Wilson’s counsel were aware of this background, he had good reason not to give the prosecutor the opportunity to confirm it by allowing Wilson’s girlfriend, sister, and brother to testify. This negative affidavit information is further indication Wilson was not prejudiced by counsel’s failure to interview these witnesses or to present their testimony at trial.
* * *
In sum, I cannot conclude that after counsel interviewed a number of people who knew Wilson and obtained an initial mental health diagnosis from Dr. Reynolds, it was unreasonable for counsel not to pursue the mental health investigation further. Counsel conducted a reasonably extensive inquiry into Wilson’s background. Nothing in the record indicates reasonably competent counsel could not have concluded he had all the relevant information available. I therefore agree with the district court’s conclusion that the OCCA’s adjudication of Wilson’s ineffectiveness claim was reasonable.
2. Counsel’s Strategic Decision Not to Emphasize Mental Illness Made Further Investigation Unnecessary
The trial record reflects that counsel made a strategic decision to de-emphasize Wilson’s mental illness and instead to focus his mitigation case on Wilson’s intelligence and capacity to become a productive member of society. Strickland requires that we defer to counsel’s strategic judgments that fall within the “wide range of professionally competent assistance.” 466 U.S. at 690, 104 S.Ct. 2052. Although the majority criticizes my focus on counsel’s strategic judgments as outside the scope of the parties’ briefing, the issue of strategy is inextricably intertwined with the rest of the Strickland analysis, including the proper scope of investigation and witness choices. See Wiggins, 539 U.S. at 521, 123 S.Ct. 2527 (“[Sjtrategic choices ... are reasonable precisely to the extent that reasonable professional judgments support the limitations on investigation.” (quoting Strickland, 466 U.S. at 690-91, 104 S.Ct. 2052)); Williams, 529 U.S. at 373, 120 S.Ct. 1495 (concluding “counsel’s failure to contact a potentially persuasive character witness was likewise not a conscious strategic choice,” but rather deficient representation under Strickland); Strickland, 466 U.S. at 690, 104 S.Ct. 2052 (holding “strategic choices made after thorough investigation of law and facts relevant to plausible options are virtually unchallengeable”).
In any event, this court may affirm the district court “on any ground adequately supported by the record.” Elkins v. Comfort, 392 F.3d 1159, 1162 (10th Cir.2004); see also Richie v. Mullin, 417 F.3d 1117, 1128 n. 3 (10th Cir.2005) (Hartz, J., concurring) (“Comity with the state courts is a particularly compelling reason to consider grounds for affirmance not argued on appeal.”).
I would conclude on this record counsel made a reasonable strategic judgment backed by a reasonable investigation.
Reasonable Strategic Judgment
Counsel’s mitigation emphasis during the penalty phase of trial is a classically strategic decision. See, e.g., Bell, 545 U.S. at 810, 125 S.Ct. 2825 (characterizing as strategic the decision, after a mental health evaluation, “not to pursue a mitigation strategy based on mental illness”); Young, 486 F.3d at 682 (noting counsel “chose to pursue an alternative theory [of mitigation], making exactly the type of strategic decision the Supreme Court and *1139this court have held is not ineffective assistance of counsel”).
Counsel chose to emphasize Wilson’s intelligence and rehabilitative potential in the following manner: (1) Dr. Reynolds testified that because of Wilson’s high intelligence and the excellent positive influence of his mother, Wilson’s mental disorders could probably be successfully treated and Wilson would likely mature into a contributing member of society; (2) Dr. Reynolds testified Wilson excelled in structured environments and because of his intelligence could be particularly beneficial to others; (3) other character witnesses testified Wilson was an intelligent, helpful, and caring person who could turn his life around; (4) counsel argued in closing: “There is some reason for him to live.... I submit to you that an intelligent person who has the capacity to do good can be of benefit to society.”; and (5) counsel suggested in closing that even serving a life prison term Wilson could, because of his high intelligence and positive influences, mentor other young men in prison to help them become productive members of society when released. Trial Tr. (Feb. 20, 1997), at 42-44. Wilson has failed to demonstrate counsel’s strategic focus was unreasonable.
Wilson argues counsel should have discovered and used a diagnosis of schizophrenia to convey to the jury that Wilson was a mentally ill man deserving of sympathy and pity. But emphasizing Wilson’s mental health could have undercut counsel’s chosen strategy of focusing on Wilson’s ability to grow into a useful role model for other young men in trouble. A schizophrenia diagnosis could have made Wilson’s mental health problems appear more intractable and unbeatable, and added ammunition to the prosecution’s case that Wilson was a dangerously ill person.
As with the evidence of Wilson’s gang involvement, emphasizing Wilson’s mental health issues was a two-edged sword. As the majority believes, the jury may have felt some sympathy for Wilson based on a diagnosis of schizophrenia. But, equally as likely, this diagnosis may have supported the prosecution’s portrait of Wilson as a dangerous and continuing threat to society. See, e.g., Bryan, 335 F.3d at 1222 (recognizing counsel’s legitimate concern that testimony about defendant’s mental health “might play into the prosecution’s case that [defendant] was a continuing threat to society”); Cannon, 259 F.3d at 1277-78 (noting certain neuropsychological evidence defendant’s counsel did not present to the jury was “far less beneficial than asserted by [defendant] ... and could have strengthened the prosecution’s argument”). In this case, counsel could reasonably conclude that additional mental health evidence would not help Wilson’s case and might actually harm it.
The majority presents a false dichotomy with regard to Dr. Reynolds’s more recent mental health evaluation. The majority argues trial counsel had to either (1) go all the way in basing a mitigation defense on a schizophrenia diagnosis, or (2) not present any evidence of Wilson’s mental health at all. Although the majority presents an all- or-nothing choice, trial counsel could have quite reasonably chosen a middle path: presenting enough mental health evidence to obtain the sympathy of some members of the jury without having to persuade more skeptical members Wilson’s mental illness was severe. Neither Supreme Court precedent nor the ABA Guidelines foreclose such a middle ground.
Commentators have recognized a mitigation defense based purely on the defendant’s mental health can be risky. One commentator, for example, noted counsel may decide to limit mitigation evidence because it “purportedly undermines residual doubt, because it has a double-edged *1140effect of inspiring jury fears, or because it opens the door to unrevealed criminal history.” Leona D. Jochnowitz, Missed Mitigation: Counsel’s Evolving Duty to Assess and Present Mitigation at Death Penalty Sentencing, 43 Crim. L. Bull. 1 (2007). Another commentator suggested trial counsel faced a barrier of “juror cynicism toward mental health issues in criminal cases.” Russell Stetler, Mental Disabilities and Mitigation (rev.3/13/01), http:// www.nynd-fpd.org/articles.htm (follow “Mental Health”; then follow “Mental Health Mitigation”). See also Scott E. Sundby, The Jury as Critic: An Empirical Look at How Capital Juries Perceive Expert and Lay Testimony, 83 Va. L.Rev. 1109 (1997) (finding capital jurors were skeptical of mental health experts supplied by the defense); James M. Doyle, The Lawyers’ Art: ‘Representation’ in Capital Cases, 8 Yale J.L. & Human. 417 (1996) (noting difficulties of presenting mental illness as mitigation evidence). In the end, as Professor Sundby found: “Severe mental illness in particular, although appearing to be a compelling mitigating circumstance, raises a number of collateral issues that may lead the jury to vote for a sentence of death rather than life.” 83 Va. L.Rev. at 1165.
Wilson also suggests his trial counsel was ineffective for not asking Dr. Reynolds to testify more completely about his diagnosis. Wilson claims,
[Dr. Reynolds’s] testimony was left completely out of context, and on cross-examination the prosecutor was able to turn him into a sounding board for the prosecutor’s diagnosis of [Wilson] as a psychopath. Despite the fact that Dr. Reynolds had said [Wilson] was not a psychopath, on redirect counsel never gave Dr. Reynolds an opportunity to explain the meaning of the term psychopath, or to explain why his diagnosis indicated that [Wilson] was not a psychopath. This left the prosecutor free on closing argument to ridicule [Wilson] as a psychopath.
Aplt. Br. 71-72.
First of all, faulting counsel’s questioning during witness examination is particularly vulnerable to the kind of hindsight second-guessing that Strickland warned against. 466 U.S. at 689, 104 S.Ct. 2052. Undoubtedly, all witness examinations fall short of perfection, but this does not mean counsel’s examination “fell below an objective standard of reasonableness.” Id. at 688, 104 S.Ct. 2052. The failure to ask questions another counsel might have asked does not prove ineffective assistance.
Secondly, it was reasonable for counsel not to revisit a point he wished to avoid. See, e.g., Bland v. Sirmons, 459 F.3d 999, 1032 (10th Cir.2006) (holding counsel was not deficient “for failing to question [the expert] in such a way as to reveal the full extent of [defendant’s] mental and substance-abuse problems” when expert’s testimony was reasonably well-developed). Dr. Reynolds had already testified on cross-examination that Wilson was not a psychopath. Counsel’s decision to de-em-phasize Wilson’s mental illness supports his decision not to ask Dr. Reynolds for a more specific diagnosis, to describe Wilson’s illness in greater detail, or to describe in more detail the characteristics of a psychopath.
Once counsel decided a mitigation argument based on Wilson’s mental illness would not strengthen his case, he had no reason to obtain a definitive diagnosis of schizophrenia from Dr. Reynolds. Wilson’s chaotic family life, coupled with his mental health problems, meant he did not do well on the Tulsa streets. But in the structured environment of prison — as with his sister’s home in North Carolina — Wilson could succeed. Wilson’s high IQ, religious faith, and well-mannered behavior *1141could make Mm an asset to other troubled youths.
The Supreme Court has concluded counsel was not ineffective for limiting the investigation into mitigating evidence under similar circumstances. See Bell, 545 U.S. at 810, 125 S.Ct. 2825 (suggesting reasonable counsel might end mental health investigation after initial evaluations were completed and counsel selected mitigation strategy not emphasizing mental illness); Burger, 483 U.S. at 795, 107 S.Ct. 3114 (holding counsel not ineffective for limiting investigation when counsel reasonably determined that additional evidence of a “tragic childhood” would not “minimize[] the risk of the death penalty”); Darden v. Wainwright, 477 U.S. 168, 186, 106 S.Ct. 2464, 91 L.Ed.2d 144 (1986) (holding counsel not ineffective for limiting mitigation strategy to “simple plea for mercy” when character evidence could be harmful). This court has reached similar conclusions. Compare Bryan, 335 F.3d at 1222 & n. 20 (holding counsel not ineffective for failing to present mental health evidence that “would just more nearly accentuate the position of the State, that [defendant] was prone to- be and could be a danger to society”), with Hooper, 314 F.3d at 1170-71 (counsel had obligation to fully develop mental health evidence once he decided to rely on it at trial). Counsel’s determination that additional mental health evidence would not strengthen his case that Wilson could be rehabilitated was thus a “reasonable decision that makes particular investigations unnecessary.” Wiggins, 539 U.S. at 521, 123 S.Ct. 2527.
Instead of focusing entirely on serious mental health problems, counsel reasonably pursued a strategy where Dr. Reynolds’s limited mental health testimony would bolster a rehabilitation defense. Although counsel’s chosen strategy was ultimately unsuccessful, a “fair assessment of attorney performance requires that every effort be made to eliminate the distorting effects of hindsight, to reconstruct the circumstances of counsel’s challenged conduct, and to evaluate the conduct from counsel’s perspective at the time.” Strickland, 466 U.S. at 689, 104 S.Ct. 2052. If I consider counsel’s decision without the benefit of hindsight, I cannot conclude the argument Wilson suggests based on his allegations of schizophrenia was so vastly superior to the argument counsel actually made that counsel’s choice was not “within the wide range of reasonable professional assistance.” See id.
Supported by Reasonable Investigation
An investigation is sufficient to support a strategic decision if it uncovers for counsel all available options for presenting and arguing mitigating evidence. See Strickland, 466 U.S. at 690, 104 S.Ct. 2052 (explaining “strategic choices made after thorough investigation of law and facts relevant to plausible options are virtually unchallengeable”); cf. Wiggins, 539 U.S. at 524-25, 123 S.Ct. 2527 (finding ineffective assistance where counsel did not know the extent of potential mitigating evidence); Anderson, 476 F.3d at 1145 (finding ineffective assistance where counsel never investigated the defendant’s family background and mental health).
In Wiggins, counsel was ineffective for not uncovering evidence of the extensive physical, sexual, and emotional abuse suffered by the defendant. 539 U.S. at 524-25, 123 S.Ct. 2527. Although counsel knew something about the defendant’s troubled childhood and alluded to a difficult childhood in opening statements, counsel failed to obtain a complete social history of the defendant, even though it was easily obtainable. Id. at 525-26, 123 S.Ct. 2527.
Unlike Wiggins, Wilson’s is not a case where counsel’s investigation touched on mitigating evidence but failed to unearth the full extent of it. Counsel unearthed *1142enough information about Wilson’s mental health to reasonably formulate a trial strategy, and the new information Wilson now offers would not have changed that strategy. Cf. id. at 535, 123 S.Ct. 2527. Even if counsel had known Wilson suffered from “schizophrenia, paranoid type,” rather than simply “a severe personality disturbance,” Wilson has not shown this knowledge would have changed counsel’s strategic decision to de-emphasize Wilson’s mental health problems.
In this case, counsel had enough information to consider all reasonable options prior to trial. Dr. Reynolds’s evaluation revealed a great deal about Wilson’s mental health. Before trial, Dr. Reynolds told counsel: (1) Wilson had an IQ in the superior range; (2) Wilson had no organic brain damage; (3) Wilson exhibited some indications of several disorders including (i) generalized anxiety disorder, (ii) bipolar disorder, (iii) post-traumatic stress disorder, (iv) paranoid personality disorder, and (v) narcissistic personality disorder with passive-aggressive and schizotypal personality features; and (4) Wilson sometimes exhibited a lack of contact with reality and paranoid suspicious behaviors. Counsel knew enough to establish during Dr. Reynolds’s direct examination Wilson had “a severe mental disorder” and testing indicated “a severe personality disturbance.” Trial Tr. (Feb. 19, 1997), at 57. Dr. Reynolds also told the jury Wilson “has some very unusual, bizarre types of thinking. That would suggest that at times he’s not or has not periodically been in touch with reality. That he basically does not necessarily function at times in a normal state but that he has a great deal of emotional pathology.” Id. Counsel therefore knew Wilson had significant mental health problems.
Given the indications of mental illness in Dr. Reynolds’s initial analysis, I cannot conclude counsel had insufficient knowledge to make an informed decision about whether to focus his mitigation strategy on Wilson’s mental health. Counsel had evidence available to make the argument Wilson suggests he should have — that because Wilson was mentally ill he did not deserve to die. Instead, counsel chose a different focus, which is the essence of strategy. Whether or not counsel knew a more specific diagnosis was possible with further testing, counsel certainly knew mental illness was an available argument in mitigation and was entitled to de-emphasize that argument in favor of what he considered the more promising emphasis.
In sum, because counsel’s decision to focus on Wilson’s rehabilitative potential was supported by reasonable investigation and strategic judgment, his decision not to pursue further diagnosis of mental illness was similarly reasonable. I therefore find nothing wrong with the OCCA’s view of the matter under Strickland.
3. Counsel Allowed Sufficient Time to Prepare
Wilson contends counsel’s mental health investigation was unreasonable because counsel did not contact Dr. Reynolds until three weeks before trial and did not meet with Dr. Reynolds until two days before he testified. “No particular set of detailed rules for counsel’s conduct can satisfactorily take account of the variety of circumstances faced by defense counsel or the range of legitimate decisions regarding how best to represent a criminal defendant.” Strickland, 466 U.S. at 688-89, 104 S.Ct. 2052. As a court of appeals, we are in no position to micromanage defense counsel’s representation by establishing investigation deadlines.
To be sure, insufficient preparation of the mitigation case can constitute ineffective assistance of counsel. See Williams, *1143529 U.S. at 395, 120 S.Ct. 1495; Anderson, 476 F.3d at 1143-44. But this is so only if the investigation actually fails to uncover evidence due to time limitations, not simply because of the amount of time allotted. In Williams, the Supreme Court noted the mitigation investigation began only a week before trial, and in Anderson, this court noted the mitigation investigation was undertaken only in the month before trial. Timing was only part of the problem, though. The investigations in those cases were unreasonable not simply because they had been undertaken late, but because they failed to uncover significant mitigating evidence. See id.
The cases from the Ninth Circuit relied upon by the majority also reveal that counsel’s inadequate timing must result in detriment to the client to constitute ineffective assistance. See Bloom v. Calderon, 132 F.3d 1267, 1277 (9th Cir.1997) (quoting Hendricks v. Calderon, 70 F.3d 1032, 1038 (9th Cir.1995)). “The complete lack of effort by [defendant’s] trial counsel to obtain a psychiatric expert until days before trial, combined with counsel’s failure to adequately prepare his expert and then present him as a trial witness, was constitutionally deficient performance.” Id. (emphasis added). Although the majority would read out the causality implied by the phrase “combined with,” this causality is essential to the constitutional analysis. See id. (“Because counsel did not acquire the services of this key witness until days before trial, a hurried and inaccurate report resulted.”); see also Wallace v. Stewart, 184 F.3d 1112, 1116 (9th Cir. 1999) (finding that “had [medical] experts known the details of [defendant’s] family background, the substance and tone of the sentencing hearings would have been significantly different”). Counsel cannot be deemed deficient based on the timing of counsel’s investigation alone.
The record in this case does not indicate counsel allowed insufficient time to conduct his investigation. The majority suggests three weeks is insufficient to complete an adequate expert investigation.12 Yet in this very case, Dr. Reynolds had sufficient time to meet with Wilson multiple times, review life history files, conduct multiple tests, and interview Wilson’s mother. Nor do I see a reason why meeting with an expert two days before his testimony should in all cases be unreasonable. See Rompilla, 545 U.S. at 381, 125 S.Ct. 2456 (“A standard of reasonableness applied as if one stood in counsel’s shoes spawns few hard-edged rules.... ”). In this case, if Dr. Reynolds’s findings required action on counsel’s part, two days provided counsel sufficient time to adjust strategy, ask for a continuance, or otherwise respond accordingly.
The record gives no indication that the time counsel allowed for investigating Wilson’s mental health in any way limited the development or presentation of mitigating evidence. Wilson argues that if counsel had allowed more time for the investigation, Dr. Reynolds, by conducting additional tests and interviews, could have obtained a diagnosis of schizophrenia on top of the other mental health problems he had already identified.13 But Dr. Reyn*1144olds’s affidavit does not state he advised counsel more time was needed. Nor does the affidavit state counsel rebuffed any request for more time to conduct additional tests or collect additional data from Wilson’s family or friends. Not surprisingly, trial counsel did not consider additional tests and interviews necessary. If counsel had considered further investigation necessary, he could have sought a continuance or conducted what further investigation was possible in the time remaining.
As the above discussion demonstrates, Dr. Reynolds’s investigation revealed enough information to convince a reasonable counsel the investigation was adequate and to ensure counsel’s strategic decisions at trial were reasonably informed. The Constitution requires nothing more.
B. Prejudice
Even if counsel’s performance were deficient under Strickland’s first prong, Wilson must also establish that any deficiency prejudiced his defense. Strickland, 466 U.S. at 691, 104 S.Ct. 2052 (“An error by counsel, even if professionally unreasonable, does not warrant setting aside the judgment of a criminal proceeding if the error had no effect on the judgment.”). We must therefore determine “whether there is a reasonable probability that, absent the errors, the [jury] ... would have concluded that the balance of aggravating and mitigating circumstances did not warrant death.” Strickland, 466 U.S. at 695, 104 S.Ct. 2052; accord Boltz v. Mullin, 415 F.3d 1215, 1222 (10th Cir.2005) (citing Strickland). “In making this determination, we consider the strength of the State’s case, the aggravating circumstances the jury found, the mitigating evidence defense counsel did present, and the additional mitigating evidence the defense might have presented.” Neill v. Gibson, 278 F.3d 1044, 1062 (10th Cir.2001).
Juries may interpret evidence of mental illness in mitigation in several ways. They may decide not to impose the death penalty because mental illness helps to explain why the defendant behaved the way he did and makes the defendant less culpable for his crimes. Or they may decide not to impose the death penalty because mental illness makes the defendant a more humanized, sympathetic figure. See Smith, 379 F.3d at 943 (“We have previously emphasized that mitigation evidence affords an opportunity to humanize and explain.” (internal marks omitted)). Wilson has not shown a reasonable probability that the additional evidence he offers would have changed the jury’s balance of aggravating and mitigating circumstances under either approach.
Notably, none of the evidence Wilson offers on habeas review would add to a jury’s understanding of how Wilson’s mental illness explains his role in the murder. In Smith, we found prejudice when counsel failed to offer mental health evidence explaining how the defendant’s childhood brain injury caused a loss of emotional control, which could have resulted in murder when the defendant was unable to control his anger. Id.; see also Williams, 529 U.S. at 398, 120 S.Ct. 1495 (finding prejudice when omitted evidence would have corroborated “the view that in each case [defendant’s] violent behavior was a *1145compulsive reaction rather than the product of cold-blooded premeditation”).
Wilson offers no plausible explanation of the role his mental health could - have played in the murder. Even after diagnosing Wilson as possibly schizophrenic, Dr. Reynolds suggests only an inconclusive “possibility]” that Wilson “could have been delusional at the time of the crime.” Aplt. Add. 2. As evidence of this delusion, Dr. Reynolds suggests that Wilson “must have been delusional to believe that he would not be easily identified” when he robbed the store where he worked. Id. Of course, the better explanation for Wilson’s behavior that night is Wilson and his three accomplices planned to kill the only witness who could identify him. None of the additional affidavit evidence Wilson offers corroborates the idea Wilson suffered from delusions the night of the murder or at any other time that would help a jury understand why Wilson behaved the way he did.14 This is not the kind of additional explanatory information that could have altered the outcome in this case, particularly where the jury already knew the basic details about Wilson’s mental health.
To counteract the prosecution’s aggravating factors, counsel presented a constitutionally adequate case of mitigation on Wilson’s behalf. The jury heard the various categories of mitigating evidence the ABA has identified as appropriate in capital cases. See ABA Guidelines 11.4.1(2) (identifying categories of potential mitigating evidence). This evidence included: (1) medical history, including signs of mental illness; (2) family and social history, including substance abuse and abandonment by Wilson’s father and life in a violent neighborhood affected by gang violence; (3) traumatic events, such as the sudden death of Wilson’s Sunday school teacher, being shot in the leg, and having his house burned down; (4) religious influences, including the faith of Wilson’s mother and his church activities; and (5) educational history, including academic achievement and well-mannered behavior in the structured school environment.
With regard to Wilson’s mental health, the jury heard evidence that Wilson had “a severe mental disorder” and a “severe personality disturbance.” Trial Tr. (Feb. 19, 1997), at 57. The jurors knew Wilson “has some very unusual bizarre types of thinking” and “has not periodically been in touch with reality.” Id. They also heard Wilson was not a psychopath, even though he had some characteristics that could indicate psychopathy.
Had the jury heard all of this alongside a specific diagnosis of paranoid schizophrenia, I cannot conclude to a reasonable probability the jury would have weighed the aggravating and mitigating factors differently. Further diagnosis and discussion of Wilson’s condition would only have given the prosecutor more opportunity to focus on the dangerous characteristics associated with Wilson’s mental illness. Dr. *1146Reynolds was able to deny that Wilson was a psychopath, but he could not have denied the dangerous aspects of schizophrenia if asked about them by the prosecutor.
This case is a far cry from Anderson v. Sirmons, 476 F.3d at 1148, where the mitigation evidence left the jury with a “pitifully incomplete” picture of the defendant, or Wiggins v. Smith, 539 U.S. at 537, 123 S.Ct. 2527, where counsel failed to present any life history evidence in mitigation, relying instead on the fact defendant had no prior convictions. The additional evidence Wilson now offers “would barely have altered the sentencing profile presented to” the jury. Strickland, 466 U.S. at 700, 104 S.Ct. 2052; see also Clark, 425 F.3d at 286 (“[T]o establish prejudice, the new evidence ... must differ in a substantial way — in strength and subject matter— from the evidence actually presented at sentencing.”).
Based on Dr. Reynolds’s testimony alone, the jury was able to consider Wilson’s mental disabilities when determining his sentence. Dr. Reynolds’s descriptions of Wilson’s behavior meant the same thing to a lay jury as a clinical diagnosis, and may actually have been more useful to them. See Clark, 425 F.3d at 286 (finding no prejudice when, although expert failed to specifically name the cause of defendant’s mental deficiencies, he described their effects). Although counsel could have presented a succession of witnesses to corroborate the bizarre thinking Dr. Reynolds described, the additional testimony would only have repeated what the jury heard through Dr. Reynolds. See Bland, 459 F.3d at 1031 (finding no prejudice when counsel failed to present additional, redundant testimony about defendant’s drug use).
It is not even clear from the record that Dr. Reynolds’s testimony at trial would have changed had he performed additional testing. The description of Wilson’s illness Dr. Reynolds gave at trial reasonably encompasses the diagnosis Dr. Reynolds says he arrived at after trial. At trial, Dr. Reynolds testified Wilson had “very unusual, bizarre types of thinking. That would suggest that at times he’s not or has not periodically been in touch with reality.” Trial Tr. (Feb. 19, 1997), at 57. That description is similar to a diagnosis of schizophrenia. The American Psychiatric Association describes the characteristic symptoms of schizophrenia as including, among other things, “bizarre delusions,” “disorganized thinking and behavior,” and hallucinations. American Psychiatric Association, Diagnostic and Statistical Manual of Mental Disorders DSM-IV-TR 299-300, 313 (4th ed.2000); see also id. (“The essential feature of the Paranoid Type of Schizophrenia is the presence of prominent delusions or auditory hallucinations in the context of a relative preservation of cognitive functioning and affect.”).
Having Dr. Reynolds change his testimony or add additional diagnoses to it would not likely have changed the jury’s decision. The failure to present additional diagnoses to a jury will not prejudice the defendant when the jury has already heard some evidence about the defendant’s mental health. See Malicoat v. Mullin, 426 F.3d 1241, 1261 (10th Cir.2005) (finding additional diagnosis regarding history of seizures would not have influenced outcome when jury heard evidence of abuse defendant suffered as child); Knighton v. Mullin, 293 F.3d 1165, 1178-79 (10th Cir.2002) (finding additional diagnosis of “significant organic brain damage present, with a likely psychotic condition with auditory and visual hallucinations” would not have influenced outcome when jury heard evidence of depression, substance abuse, suicide attempts, and memory loss); Humphreys v. Gibson, 261 F.3d 1016, 1021 (10th Cir.2001) (finding additional diagnoses of addiction, organic brain damage, *1147and brain seizures would not have influenced outcome when jury heard evidence of depression, alcohol abuse, and a personality disorder).
Multiple aggravating circumstances lessen the likelihood of prejudice even further. See McCracken v. Gibson, 268 F.3d 970, 978-80 (10th Cir.2001) (finding additional mental health diagnoses would not have influenced outcome when jury found six aggravating factors and diagnoses may have supported aggravating factor of continuing threat). Wilson’s jury found three aggravating factors: (1) the murder was especially heinous, atrocious or cruel; (2) the murder was committed for the purpose of avoiding or preventing a lawful arrest or prosecution; and (3) a probability existed that the defendant would commit criminal acts of violence that would constitute a continuing threat to society. Wilson, 2006 WL 2289777, at *3. Because the jury heard evidence about Wilson’s mental health and still found these three aggravating circumstances, there is no reasonable probability that a further diagnosis indicating possible schizophrenia would have influenced the sentencing outcome.
In sum, the jury heard extensive mitigating evidence, and Wilson’s newly proffered evidence adds little to that calculus. I cannot conclude there is a reasonable probability that had Dr. Reynolds explained Wilson’s diagnosis in more detail rather than simply describing its effects, the jury “would have struck a different balance between the mitigating and aggravating factors.” Anderson, 476 F.3d at 1148. The OCCA’s decision was not an unreasonable application of clearly established federal law.
IV. Evidentiary Hearing
The district court did not abuse its discretion in denying Wilson an evidentiary hearing on his claim of ineffective assistance of trial counsel. Because my review of the record reveals Wilson is not entitled to relief on his claim, the district court was within its discretion in denying Wilson an evidentiary hearing.
A district court abuses its discretion only if its decision is “arbitrary, capricious, whimsical, or manifestly unreasonable.” United States v. Atencio, 435 F.3d 1222, 1235 (10th Cir.2006) (quotation omitted). “In deciding whether to grant an eviden-tiary hearing, a federal court must consider whether such a hearing could enable an applicant to prove the petition’s factual allegations, which, if true, would entitle the applicant to federal habeas relief.” Schriro v. Landrigan, — U.S. -, 127 S.Ct. 1933, 1940, 167 L.Ed.2d 836 (2007); accord Mayes v. Gibson, 210 F.3d 1284, 1287 (10th Cir.2000).
Where, as here, the state court has adjudicated a petitioner’s claim on the merits, the petitioner must pass a high bar to show his entitlement to federal habeas relief. See Schriro, 127 S.Ct. at 1940; Mayes, 210 F.3d at 1287-88. Taking petitioner’s allegations as true, petitioner must show the state court’s decision “was contrary to, or involved an unreasonable application of, clearly established Federal law, as determined by the Supreme Court of the United States” or “was based on an unreasonable determination of the facts in light of the evidence presented in the State court proceeding.” 28 U.S.C. § 2254(d).
In this case, the district court concluded, “As the disposition of Petitioner’s habeas corpus petition does not require reference to any materials beyond those that are available and currently before the Court, the Court finds that there is no need for an evidentiary hearing in this case. There are no disputed factual questions remaining that could possibly entitle Petitioner to habeas corpus relief.” Wilson, 2006 WL 2289777, at *47.
*1148Considering both the state trial court records and the additional affidavits supplied with Wilson’s petition for habeas corpus, I cannot say the district court abused its discretion in concluding an evidentiary hearing could not help Wilson’s habeas case. Rather, I agree with the district court that an evidentiary hearing would be of no use to Wilson. See Schriro, 127 S.Ct. at 1940 (“[I]f the record refutes the applicant’s factual allegations or otherwise precludes habeas relief, a district court is not required to hold an evidentiary hearing.”).
I also note Wilson has not identified a factual dispute requiring an evidentiary hearing. Oklahoma’s and Wilson’s accounts of counsel’s preparation for trial appear to be identical, and the state does not dispute any of the information Wilson offers in his affidavits. Accordingly, the only dispute in this case is a matter of law based on the record before us. Because we need only determine whether the facts contained in the record amount to ineffective assistance of counsel under Strickland, no evidentiary hearing is required. See, e.g., Anderson v. Att’y Gen. of Kan., 425 F.3d 853, 860 (10th Cir.2005) (“The purpose of an evidentiary hearing is to resolve conflicting evidence.”).
The district court therefore correctly rejected Wilson’s request for an evidentiary hearing.
V. Conclusion
Because I conclude the OCCA correctly determined Wilson’s trial counsel was not constitutionally ineffective under Strickland, and because the district court did not abuse its discretion in denying an eviden-tiary hearing, I respectfully dissent from Part III of the majority’s opinion.
. I agree with the majority 28 U.S.C. § 2254(e)(2) does not bar Wilson’s request for an evidentiary hearing because Wilson diligently sought to develop the factual basis for his ineffective assistance of counsel claim in state court. Because Wilson diligently requested, and was denied, the opportunity to develop the state court record, § 2254(e)(2) does not bar his request for an evidentiary hearing in federal court. See Williams, 529 U.S. at 432, 120 S.Ct. 1479; Barkell v. Crouse, 468 F.3d 684, 695-96 (10th Cir.2006).
. See 28 U.S.C. § 2254(e)(2) ("If the applicant has failed to develop the factual basis of a claim in State court proceedings, the court shall not hold an evidentiary hearing on the claim unless the applicant shows that — (A) the claim relies on — (i) a new rule of constitutional law, made retroactive to cases on collateral review by the Supreme Court, that was previously unavailable; or (ii) a factual predicate that could not have been previously discovered through the exercise of due diligence; and (B) the facts underlying the claim would be sufficienl to establish by clear and convincing evidence that but for constitutional error, no reasonable factfinder would have found the applicant guilty of the underlying offense.”); Schriro, 127 S.Ct. at 1940 ("In deciding whether to grant an evidentiary hearing, a federal court must consider whether such a hearing could enable an applicant to prove the petition’s factual allegations, which, if true, would entitle the applicant to federal habeas relief.”)
. See 28 U.S.C. § 2254(d) (“An application for a writ of habeas corpus on behalf of a person in custody pursuant to the judgment of a State court shall not be granted with respect to any claim that was adjudicated on the merits in State court proceedings unless the adjudication of the claim — (1) resulted in a decision that was contrary to, or involved an unreasonable application of, clearly established Federal law, as determined by the Supreme Court of the United States; or (2) resulted in a decision that was based on an unreasonable determination of the facts in light of the evidence presented in the State court proceeding.”).
. I acknowledge there has been a circuit split regarding the correct standard of review in cases where the petitioner presents a federal court with material not considered by the state court. See LeCroy v. Secy Fla. Dep't of Corrs., 421 F.3d 1237, 1262-63 & n. 30 (11th Cir.2005) (describing split). In my opinion. *1129Schriro sheds considerable light on the correct standard.
. See Okla. Stat., tit. 22, ch. 18, Rule 3.1 l(B)(3)(b)(i) (1997) ("This Court will utilize the following procedure in adjudicating applications regarding ineffective assistance of trial counsel based on evidence not in the record: ... [T]he application, and affidavits must contain sufficient information to show this Court by clear and convincing evidence there is a strong possibility trial counsel was ineffective for failing to utilize or identify the complained-of evidence.”); Dewberry v. State, 954 P.2d 774, 775-77 (Okla.Crim.App.1998).
. Wilson presents six affidavits to this court. However, it appears only five were presented to the OCCA. The new (sixth) affidavit is from Wilson's state trial counsel.
. The absence of any admissions in counsel’s affidavit to errors at trial stands in stark contrast to other death-penalty appeals in which counsel confesses his performance was deficient. See, e.g., Fisher, 282 F.3d at 1293-98 (noting counsel admitted he failed to, among other things, discover crucial evidence, interview important witnesses, and investigate his client’s alibi).
. Wilson does not argue Dr. Reynolds performed incompetently as a psychologist. I note, however, the Fourth, Seventh, and Ninth Circuits have explicitly rejected a constitutional right to effective assistance of an expert witness, except to the extent such claim implicates effective assistance of counsel in hiring and preparing the witness. See Wilson v. Greene, 155 F.3d 396, 401 (4th Cir.1998); Harris v. Vasquez, 949 F.2d 1497, 1520 (9th Cir.1990); Silagy v. Peters, 905 F.2d 986, 1013 (7th Cir.1990).
. Other circuits have recognized the importance of expert recommendations in counsel's decision-making concerning mental health investigations. See Martinez v. Quarterman, 481 F.3d 249, 257 (5th Cir.2007) (finding no ineffective assistance when nothing in expert report suggested a link between the defendant's epilepsy and his violent behavior); Hedrick v. True, 443 F.3d 342, 350-51 (4th Cir.2006) (finding no ineffective assistance when expert did not indicate further investigation of defendant's childhood experiences would be helpful); Clark, 425 F.3d at 282 (finding no ineffective assistance when expert review did not indicate further investigation was necessary).
. In any event, Wilson has not shown he was prejudiced by counsel's failure to interview his mother. Counsel's examination of the mother at trial effectively portrayed the sympathetic side of Wilson’s life and the effect his death would have on his family. To the extent her affidavit contains mental health information not presented at trial, there is no reason to believe trial counsel would have elicited it in an interview. I find it instructive Dr. Reynolds apparently did not discover this information when he spoke with Wilson’s mother during an interview whose purpose was to bolster the psychological testing already conducted. Wilson has not shown counsel would have discovered this information by conducting another interview himself. Accordingly, Wilson has made no showing that the mother’s testimony would have changed and therefore no prejudice. See Hedrick, 443 F.3d at 354 (finding no prejudice where defendant did not show witness’s testimony would have differed); Crisp v. Duckworth, 743 F.2d 580, 584 (7th Cir.1984) (same).
. What Wilson asserts is substantively new testimony involves the following: (1) Wilson telling his girlfriend his father was dead, when his father was alive but had abandoned the family years earlier; (2) Wilson introducing himself as Tom and using a different voice and facial expressions; (3) Wilson telling his girlfriend he heard voices; (4) Wilson, when he was a child, trying to convince his school his mother was white; (5) Wilson acting paranoid and suspicious; (6) Wilson having frequent violent dreams; (7) Wilson experiencing periodic depression, detachment, and memory gaps; and (8) Wilson speaking in a disconnected manner. Aplt. Add. 2-6.
. Of course, in an ideal world defense counsel would in all cases begin a mitigation investigation as soon as a representation begins, not mere weeks in advance. See ABA Guidelines 11.4.1. The record in this case does not reflect when counsel’s efforts to investigate mitigating evidence began. Although counsel engaged Dr. Reynolds as an expert witness three weeks before trial, counsel provided Dr. Reynolds with information showing counsel had begun developing a mitigation defense even earlier.
. After the initial tests, Dr. Reynolds had concluded: "Mr. Wilson was functioning in the Superior Range of intelligence with an IQ score of 126. There was no evidence of neurological or organic brain damage impair*1144ment. The MCMI-III indicated an Axis I diagnosis of Generalized Anxiety Disorder, Bipolar Disorder (severe w/o psychotic features) and Posttraumatic Stress Disorder. It further indicated an Axis II diagnosis of Paranoid Personality Disorder and Narcissistic Personality Disorder with Passive-Aggressive and Schizotypal Personality Features.” Aplt. Add. 2.
. To the contrary, all the evidence points to lucid thinking and planning. In particular, Wilson coordinated his actions with three co-defendants, and the video tape of the robbery shows Wilson first chatting with the victim as he and his co-defendants perused the store. Wilson and his co-defendants waited until the store was free of customers before attacking the victim and forcing him into a back room. Wilson restrained the victim while his co-defendants left the store to retrieve the murder weapon. When his co-defendants returned, Wilson served customers and attempted to remove the store’s safe when the store was empty. Wilson’s words and actions that night offer no evidence of delusional thinking. Furthermore, none of the purported evidence of delusions Wilson offers in the post-conviction affidavits from his mother, sister, brother, and girlfriend are. temporally tied to the night of the murder. Nor does the record reveal any other indication Wilson had been behaving strangely around the time he committed this crime.