Ladd v. Cockrell

RHESA HAWKINS BARKSDALE, Circuit Judge: This appeal is from the denial of habeas relief concerning a Texas capital murder conviction for which a death sentence was imposed. Primarily at issue is whether the State’s request, as permitted by Texas law, for a “jury-shuffle” at the start of jury selection was impermissibly motivated by race. The district court granted a Certificate of Appealability (COA) on two issues: (1) whether Robert Charles Ladd was denied the right to a fair and impartial jury because of the prosecution’s “shuffle” of the venire; and (2) whether Ladd was denied effective assistance of counsel at trial. AFFIRMED. I. On 25 September 1996, firemen responding to a fire in Tyler, Texas, found the body of 38-year-old Vickie Ann Gardner. She was on the floor in her apartment, her wrists bound together in front of her. The fire had been started on or around Gardner’s body, most likely on bedding that had been placed between her legs. An autopsy revealed Gardner died as a result of strangulation and had sustained blunt force trauma to the head. A vaginal smear revealed the presence of spermatozoa. Gardner’s apartment had been ransacked and several items were missing, including: a microwave oven; a combination television and video recorder; and two telephones. The day firemen responded to the fire (25 September), Edwin Wright pawned the missing combination television/video recorder and one of the telephones. Wright testified he received the items from J.T. Robertson. Also that day, other items identified as Gardner’s were recovered from Robertson’s apartment. Robertson testified: at some point between 9:00 and 10:00 p.m. on 24 September, he received the items from Ladd in exchange for five $20 “rocks” of crack cocaine; early the next morning, Ladd returned with additional items, for which Robertson gave Ladd two more $20 “rocks”. Ladd was arrested the same day the items were recovered from the pawn shop and Robertson’s apartment (25 September); various pieces of jewelry on his person when he was arrested were identified as Gardner’s. A fingerprint lifted from *351 the microwave oven that had been missing from Gardner’s apartment matched Ladd’s, as did a palm print lifted from a kitchen cabinet in Gardner’s apartment. Ladd had previously worked at, and been a client of, a rehabilitation center where Gardner was employed. DNA tests indicated Ladd was in the group that could have produced the spermatozoa found in the vaginal smear. On 23 August 1997, Ladd was convicted of capital murder under four separate theories — the murder having taken place during the commission of burglary, robbery, sexual assault, and arson. At the sentencing phase, the State presented 11 witnesses, including testimony that Ladd had previously committed a triple murder (discussed in part II.B. infra) and testimony by two psychiatrists that, in their opinion, Ladd constituted a continuing danger to society. The defense did not present evidence at that phase. The jury answered the special issues as follows: the killing of Gardner was deliberate; there was a probability Ladd would commit acts of criminal violence that would constitute a continuing danger to society; and there was not sufficient mitigating evidence to justify imposing a sentence of life imprisonment. On 27 August 1997, the trial judge sentenced Ladd to death. In October 1999, the Texas Court of Criminal Appeals affirmed Ladd’s conviction and sentence. Ladd v. State, 3 S.W.3d 547 (Tex.Crim.App.1999), cert. denied, 529 U.S. 1070, 120 S.Ct. 1680, 146 L.Ed.2d 487 (2000). Ladd had earlier filed for post-conviction relief in state court; on 11 June 1999, the trial court conducted an evidentiary hearing on his application. On 15 December 1999, the Texas Court of Criminal Appeals adopted the trial court’s proposed findings of fact and conclusions of law and denied the application. Ex Parte Ladd, No. 42,639-01 (Tex.Crim.App.1999). Ladd filed for habeas relief in federal district court in January 2001. The district court rejected, inter alia, Ladd’s jury-shuffle and ineffective assistance claims and denied habeas relief. II. From the numerous COA requests by Ladd, the district court granted a COA on two: (1) the jury-shuffle denied him the right to a fair and impartial jury; and (2) he received ineffective assistance at trial. Along this line, Ladd asserts: (1) the Equal Protection Clause and the right to a fair and impartial jury were violated when the State was granted the shuffle of potential jurors prior to jury selection; and (2) he received ineffective assistance at the sentencing phase. The district court’s legal conclusions are reviewed de novo; its factual findings, for clear error. E.g., United States v. Williams, 264 F.3d 561, 571 (5th Cir.2001). Of course, review is through the strictures imposed by the Antiterrorism and Effective Death Penalty Act of 1996 (AEDPA). The federal habeas statute, as amended by AEDPA, requires a great deal of deference to state court proceedings. A federal court may not grant habeas relief to a state prisoner with respect to any claim that was adjudicated on the merits in State court proceedings unless the adjudication of the claim [in state court]— (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 *352 presented in the State court proceeding. ... 28 U.S.C. § 2254(d) (emphasis added). In determining what constitutes an “unreasonable application” of law under § 2254(d)(1), “ ‘[ujnreasonable’ does not mean merely ‘incorrect’: an application of clearly established Supreme Court precedent must be incorrect and unreasonable to warrant federal habeas relief’. Foster v. Johnson, 293 F.3d 766, 776 (5th Cir.2002) (emphasis in original) (citing Williams v. Taylor, 529 U.S. 362, 410-12, 120 S.Ct. 1495, 146 L.Ed.2d 389 (2000)): Any state court factual findings are presumed correct, and the unreasonableness, if any, of such findings must be established by clear and convincing evidence. 28 U.S.C. § 2254(e)(1). A. The venire consisted of more than 200 persons. Ladd notes that he is black; the victim, white. Relying on Batson v. Kentucky, 476 U.S. 79, 106 S.Ct. 1712, 90 L.Ed.2d 69 (1986) (peremptory strike to remove potential juror on basis of race violates Fourteenth Amendment), Ladd contends: Batson’s ban on the use of race as a ground for striking potential jurors extends to a jury-shuffle, as employed at his trial; and Batson was violated because the State requested and obtained a shuffle of the venire in which seven of the first 34 potential jurors were black. In his objection at trial to the requested shuffle, although Ladd did not mention Batson or the Equal Protection Clause, he did object on the ground that the racial make-up of the venire would be altered. The State does not claim procedural default; and, on direct appeal, the Court of Criminal Appeals addressed Ladd’s Bat-son claim: it “[a]ssum[ed] arguendo that Batson extends to jury shuffles”. That court denied Ladd’s claim on the merits. The district court held Batson not applicable to a jury-shuffle. Accordingly, at issue is whether the Court of Criminal Appeals unreasonably declined to find a Batson violation under this set of facts. Texas procedure allows either side to request a shuffle of the venire before voir dire: The trial judge, on the demand of the defendant or his attorney, or of the State’s counsel, shall cause a sufficient number of jurors from which a jury may be selected to try the ease to be randomly selected from the members of the general panel drawn or assigned as jurors in the case. The clerk shall randomly select the jurors by computer or other process of random selection.... Tex.Code Crim. Proc. Art. 35.11 (emphasis added). The purpose of the shuffle is to prevent any perceived unfairness in the original order in which the prospective jurors are seated; the parties have a chance to view the venire and decide whether to request a shuffle. E.g., Davis v. State, 782 S.W.2d 211, 213-14 (Tex.Crim.App.1989), cert. denied, 495 U.S. 940, 110 S.Ct. 2193, 109 L.Ed.2d 520 (1990). In fact, the opportunity to request a shuffle has been viewed as a protection for the defendant. See, e.g., Jones v. State, 833 S.W.2d 146, 147-48 (Tex.Crim.App.1992). The jury may be shuffled only once; it does not matter which side requested it. Id. at 148. 1. As it did in district court, the State maintains that Ladd’s Batson claim is barred by the non-retroactivity doctrine of Teague v. Lane, 489 U.S. 288, 109 S.Ct. 1060, 103 L.Ed.2d 334 (1989). The district court did not address the Teague issue. Again, the Court of Criminal Appeals assumed that Batson applies to a jury-shuffle. *353 Under Teague, federal courts may not create new constitutional rules of criminal procedure on habeas review unless those rights apply retroactively. According to the State, it would not be possible to grant relief without applying a new constitutional rule of criminal procedure (that Batson extends to jury-shuffles). Where, as here, the “the State [argues] that the defendant seeks the benefit of a new rule of constitutional law, [we] must apply Teague before considering the merits of the claim”. Caspari v. Bohlen, 510 U.S. 383, 389, 114 S.Ct. 948, 127 L.Ed.2d 236 (1994) (emphasis in original). It is not immediately clear that Teague bars Ladd’s claim. Teague prevents only new rules; applying Batson to a jury-shuffle does not necessarily constitute a new rule. A new rule “breaks new ground” or “imposes a new obligation”, Teague, 489 U.S. at 301, 109 S.Ct. 1060, and is not “dictated by existing law at the time [the defendant’s] conviction became final”, Sawyer v. Smith, 497 U.S. 227, 237, 110 S.Ct. 2822, 111 L.Ed.2d 193 (1990). Although “new rule” has been defined quite broadly, see, e.g., Gray v. Netherland, 518 U.S. 152, 116 S.Ct. 2074, 135 L.Ed.2d 457 (1996), applying an existing rule to a different circumstance does not create a new rule where the application is logical and foreseeable. See Stringer v. Black, 503 U.S. 222, 112 S.Ct. 1130, 117 L.Ed.2d 367 (1992). There is considerable force to the contention that the State was already obligated under Batson to select the jury without using racial bias. Thus, to determine whether applying Batson to a jury-shuffle is logical and foreseeable, we must at least describe Ladd’s claim. In addition, even if prohibiting the illegitimate use of race in a jury-shuffle request were a new rule, doing so might be justified under the second of the two exceptions to Teague. Courts may create new constitutional law that will be retroactively applied: (1) where the new rule “places certain kinds of primary; private individual conduct beyond the power of the criminal law mak[er]”; and (2) where the new rule adopts a procedure that is “implicit in the concept of ordered liberty”. Teague, 489 U.S. at 307, 109 S.Ct. 1060. The second exception is limited in scope to those procedures “without which the likelihood of an accurate conviction is seriously diminished”. Id. at 313, 109 S.Ct. 1060. Needless to say, a fair and impartial jury is implicit in the concept of ordered liberty, see, e.g., Irvin v. Dowd, 366 U.S. 717, 721-22, 81 S.Ct. 1639, 6 L.Ed.2d 751 (1961), as is the right to equal protection in the jury selection context, see, e.g., Holland v. Illinois, 493 U.S. 474, 494, 110 S.Ct. 803, 107 L.Ed.2d 905 (1990) (exclusion of racial groups from jury service “at war with our basic concepts of a democratic society and a representative government”) (internal citations omitted). Thus, to determine whether Ladd’s claim is Teague-barred, we must determine the extent to which ordered liberty is infringed, if at all, by the use of race in the decision to request a shuffle. 2. Accordingly, for deciding whether Teag-ue bars this Batson claim, we must consider whether Batson should be extended to a jury-shuffle. This is an issue of first impression. The Supreme Court’s “Fourteenth Amendment jurisprudence evinces a commitment to eliminate unnecessary and excessive governmental use and reinforcement of racial stereotypes”. Bush v. Vera, 517 U.S. 952, 985, 116 S.Ct. 1941, 135 L.Ed.2d 248 (1996). Batson prohibits purposeful racial discrimination by the prosecution in selecting the jury for three reasons: (1) it violates the defendant’s right *354 to equal protection by denying him a fair and impartial jury of his peers; (2) it denies the potential juror participation in jury service on account of his race; and (3) it undermines public confidence in the fairness of our system of justice. 476 U.S. at 86-87, 106 S.Ct. 1712. For these reasons, Batson stated it was “clear that the Constitution prohibits all forms of purposeful racial discrimination in selection of jurors”. Id. at 88, 106 S.Ct. 1712 (emphasis added). If, as Ladd claims, the prosecution requested a jury-shuffle to decrease the chances that prospective jurors of a certain race would have the opportunity to serve, it appears the prosecution may have violated Batson’s ban on the use of race to select jurors. On the one hand, the purposes behind Batson seem implicated by jury-shuffling to disperse potential jurors of a particular race who are near the front of the venire. Allowing this would hinder efforts to achieve a fair and impartial jury, sanction an attempt to exclude potential jurors on account of race, and diminish public confidence in the fairness of criminal trials and convictions. Further, we have previously suggested the requirement that jury selection be race-neutral is not limited to the use of a peremptory challenge. See McGinnis v. Johnson, 181 F.3d 686 (5th Cir.), cert. denied, 528 U.S. 1125, 120 S.Ct. 955, 145. L.Ed.2d 829 (2000) (applying Batson to jurors’ being excused); Wilson v. Butler, 813 F.2d 664 (5th Cir.1987) (suggesting Batson may apply to division of the venire). On the other hand, Batson need not be read so broadly, and. a shuffle request does not mirror the use of a peremptory strike. Unlike a peremptory strike, a jury-shuffle does not exclude any venire member from serving on a jury on the basis of race. As the district court stated: It is counsel’s ability through the use of the peremptory challenge to effect the deliberate and complete exclusion of the African-American venireperson from the venire which mandates the articulation of [a] race-neutral explanation. By contrast, the jury shuffle acts randomly, and merely alters each venireperson’s place in line, the result of which may either lessen or increase the chance that any one of them will be chosen for the petit jury. Ladd v. Cockrell, No. 1:99-CV-822, slip op. at 13 (E.D. Tex. 24 Oct. 2001). Moreover, a jury-shuffle precedes jury-selection. Post-shuffle, each side still has the opportunity to exercise its peremptory challenges; and, if one side does so in a discriminatory manner, the other side can object under Batson. In that regard, a shuffle request where several members of one race are near the front could be later used as evidence of discriminatory intent for challenging peremptory strikes. See Henry v. Texas, No. 05-00-01869-CR, 2002 WL 449700, at *3 (Tex. Ct.App. 25 Mar. 2002) (unpublished). Finally, the shuffle provides one opportunity for either side to eliminate an extraordinary or unusual grouping of an observable trait. Therefore, it may further, rather than undermine, the public’s confidence in the proceeding. In sum, as long as a requested shuffle is done once randomly (pursuant to the statute), and is not repeated (contrary to the statute), especially to obtain a certain grouping near the front of the venire, the potential harm seems far less severe than that imposed by the discriminatory use of a peremptory strike. As our discussion indicates, reasonable arguments support both positions on whether Batson applies to jury-shuffling. Along this line, that reasonable jurists can disagree on whether Batson extends to *355 jury-shuffling suggests that Ladd’s claim is Teague-bsareá. See Fisher v. Texas, 169 F.3d 295, 305 (5th Cir.1999) (internal quotation marks and citation omitted) (“Unless reasonable jurists ... at the time [the] conviction became final would have felt compelled by existing precedent to rule in [defendant’s] favor, we are barred [by Teague ] from doing so now.” (emphasis added)). Moreover, the second (“implicit in the concept of ordered liberty”) exception to Teague does not appear applicable. An examination of the mechanics of jury-shuffling has demonstrated that it cannot infringe the rights to a fair and impartial jury and to equal protection as significantly as can the use of a peremptory strike. Thus, applying Batson to jury-shuffling likely does not constitute one of those procedures “without which the likelihood of an accurate conviction is seriously diminished”. Teague, 489 U.S. at 313, 109 S.Ct. 1060. In any event, two things are certain. First, because the Texas Court of Criminal Appeals assumed Batson applies to jury-shuffling, some of the primary reasons for a Teague-bar, such as comity, are not in play. See Teague, 489 U.S. at 308, 109 S.Ct. 1060. Second, as discussed below, the holding by the Texas court that there was no Batson violation is not unreasonable for AEDPA purposes. Therefore, especially in the light of the reasonable positions on both sides for whether Batson applies to a shuffle, we need not decide the Teague issue. 3. Assuming arguendo that Ladd’s Batson claim is not Teague-barred and that Batson applies to jury-shuffling, Ladd fails to show the Court of Criminal Appeals was unreasonable in holding there was no Batson violation. To prove a violation, the burden is on the “defendant who alleges [discrimination] ... to prove the existence of purposeful discrimination”. Batson, 476 U.S. at 93, 106 S.Ct. 1712 (internal quotation marks and citation omitted). Once the [defendant] has made out a prima facie case of racial discrimination (step one), the burden of production shifts to the proponent of the strike to come forward with a race-neutral explanation (step two). If a race-neutral explanation is tendered, the trial court must then decide (step three) whether the [defendant] has proved purposeful racial discrimination. Purkett v. Elem, 514 U.S. 765, 767, 115 S.Ct. 1769, 131 L.Ed.2d 834 (1995). In objecting to the State’s shuffle request, Ladd stated: “Seven of the first several [potential jurors] are black panel members, and this will change the racial makeup”. Out of “an abundance of caution”, the trial judge asked the State to articulate its reasons for the shuffle-request. Responding that the request “had nothing to do with race”, the State offered the following justifications: the first section of venire members consisted of (1) a higher concentration of individuals with criminal histories (“we have fourteen of the first forty-two individuals that show some criminal history or same address”); (2) not as many people wearing coats and ties; (3) only a small number of elderly professional people; and (4) a probation officer the State wanted to avoid having to strike. Where, as here, the State “tendered] a race-neutral explanation”, the question of the defendant’s prima facie case is moot, and our review begins at step two. United States v. Williams, 264 F.3d 561, 571 (2001). The State’s proffered race-neutral explanation is a legal issue. Id. At this second step, we “do[] not demand an ex *356 planation that is persuasive, or even plausible”. Purkett, 514 U.S. at 767-68, 115 S.Ct. 1769. The issue is whether the explanation is facially valid: “Unless a discriminatory intent is inherent in the prosecutor’s explanation, the reason offered will be deemed race neutral”. Id. at 768, 115 S.Ct. 1769. In other words, a neutral explanation is based on something other than race. Hernandez v. New York, 500 U.S. 352, 360, 111 S.Ct. 1859, 114 L.Ed.2d 395 (1991). There is no question that the State’s explanations were race-neutral. Ladd contends: the explanation that there were more with criminal histories among the first 42 potential jurors should “set off alarms that these were code words for blacks”; and the explanation that there were too few jurors with coats and ties is “just another way of saying there were too many poor blacks”. We disagree. To accept Ladd’s assertion that there is an inherent connection between race and criminal history or race and clothing would be to embrace the very stereotypes he condemns. Further, even assuming Ladd could prove a correlation between race and criminal history or race and poverty, disparate impact is “not ... conclusive in the preliminary race-neutrality step”. Hernandez, 500 U.S. at 361, 111 S.Ct. 1859 (Spanish-speaking ability a race-neutral justification at step two). See also United States v. Webster, 162 F.3d 308 (5th Cir.), cert. denied, 528 U.S. 829, 120 S.Ct. 83, 145 L.Ed.2d 70 (1999) (having relatives with criminal records race-neutral). The prosecutor’s asserted reasons were race-neutral. At step three, the decisive question is normally whether a proffered race-neutral explanation can be believed. This ultimate conclusion of discriminatory intent is a finding of fact. Hernandez, 500 U.S. at 363, 111 S.Ct. 1859. The burden of persuasion continues to lie with the party making the claim of purposeful discrimination. United States v. Montgomery, 210 F.3d 446, 453 (5th Cir.2000). We would tend to reverse if the State’s reasons were “fantastic or inconsistent with its treatment of similar non-minority jurors”. Williams, 264 F.3d at 572. Where its reasons are believable, however, the inquiry is one of credibility. Id. Obviously, this question of fact “turns heavily on demeanor and other issues not discernible from a cold record, such that deference to the trial court is highly warranted” (even if AEDPA did not apply). Id. The trial court overruled Ladd’s objection and allowed the shuffle; the Court of Criminal Appeals affirmed on direct appeal. Under AEDPA, as previously stated, we presume the state court’s factual findings are correct unless they would result in a decision that is unreasonable in light of the evidence presented; the unreasonableness, if any, must be established by clear and convincing evidence. See 28 U.S.C. § 2254(d)(2), (e)(1). The State’s asserted justifications are plausible, not fantastic, and there is no evidence that black venire members were treated any differently than whites. All were shuffled, not just the black members; also, the similarly situated white members (those near the front of the venire) were in the same position as the seven blacks referred to by Ladd. In addition, when asked, the State was able to back up its assertions; at Ladd’s request, it identified 13 of the first 42 potential jurors (of more than 200) who had “some criminal history or same address”. The State’s other explanations appear similarly credible. Ladd simply has not established by any evidence, much less the requisite clear and convincing evidence, that the jury-shuffle request was based on race. In sum, assuming Teague does not bar a Batson claim and Batson applies, Ladd does not show under AEDPA that the *357 Court of Criminal Appeals was unreasonable in holding there was no Batson violation. B. In his state and federal habeas petitions, Ladd claimed ineffective assistance of counsel. Ladd asserts trial counsel was ineffective for failure to fully investigate mitigating evidence and to present it at the sentencing phase. The Court of Criminal Appeals, adopting the trial court’s findings and conclusions, acknowledged that, for effective assistance, defense counsel in a capital case has a duty to ensure reasonable efforts are made to investigate potential mitigating evidence. It determined: counsel did so; the investigation of evidence did not fall below an objective standard of reasonableness or violate any constitutional standards for effective assistance; and Ladd received effective assistance of counsel at the sentencing phase. The district court similarly rejected the ineffective assistance claim. Although the court concluded certain conduct by defense counsel at the sentencing phase constituted deficient performance, it held there was no reasonable probability that, but for such performance, the result would have been different. Because an ineffective assistance claim is a mixed question of law and fact, we review de novo. E.g., Crane v. Johnson, 178 F.3d 309, 312 (5th Cir.), cert. denied, 528 U.S. 947, 120 S.Ct. 369, 145 L.Ed.2d 285 (1999). Again, we may reverse only if the Texas court’s decision is “contrary to, or an unreasonable application of, clearly established federal law” 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). Ladd does not contest the state court’s “determination of the facts”; therefore, at issue is only whether the state court’s decision is “contrary to, or involved an unreasonable application of’ federal law. 28 U.S.C. § 2254(d)(1). For determining under § 2254(d)(1) whether “clearly established Federal law” has been unreasonably applied, in play are the Sixth Amendment, entitling defendants to effective assistance of counsel, and Strickland v. Washington, 466 U.S. 668, 104 S.Ct. 2052, 80 L.Ed.2d 674 (1984) and its progeny. See Neal v. Puckett, 286 F.3d 230 (5th Cir.2002). For the well-known Strickland test for ineffective assistance, Ladd must show: (1) counsel’s performance was deficient; and (2) that performance prejudiced Ladd such that, absent such performance, there is a reasonable probability that the result would have been different. Strickland, 466 U.S. at 687, 104 S.Ct. 2052. 1. Ladd contends counsel was deficient for failing to investigate and obtain mitigating evidence and for failing to present it during the sentencing phase. Ladd points to counsel’s not obtaining Ladd’s prior records from the Texas Department of Corrections (which Ladd claims could have demonstrated his non-violent behavior in prison), and not locating earlier juvenile records (which Ladd claims could have, inter alia, helped explain his life as a child). To establish deficient performance, Ladd must show counsel “made errors so serious that counsel was not functioning as the ‘counsel’ guaranteed ... by the Sixth Amendment”. Id. at 687, 104 S.Ct. 2052. Counsel’s performance is considered deficient if it “falls below an objective standard of reasonableness”, as measured by professional norms. Id. at 688, 104 S.Ct. 2052. A court must, however, be “highly deferential” of counsel’s performance and make every effort to “eliminate the distort *358 ing effects of hindsight”, id. at 689, 104 S.Ct. 2052; must “indulge a strong presumption that counsel’s conduct falls within the wide range of reasonable professional assistance”, id., 104 S.Ct. 2052; and “will not find ineffective assistance of counsel merely because [it] disagreed] with counsel’s trial strategy”, Crane v. Johnson, 178 F.3d at 312. It goes without saying that the Sixth Amendment requires counsel to conduct a reasonably thorough, independent inquiry into the defenses that might be offered in mitigation of punishment. E.g., Baldwin v. Maggio, 704 F.2d 1325, 1332-33 (5th Cir.), cert. denied, 467 U.S. 1220, 104 S.Ct. 2669, 81 L.Ed.2d 374 (1984). “In assessing counsel’s performance [at the sentencing phase], we look to such factors as what counsel did to prepare for sentencing, what mitigating evidence he had accumulated, what additional ‘leads’ he had, and what results he might reasonably have expected from these leads.” Neal, 286 F.3d at 237 (5th Cir.2002). Applying these factors to Ladd’s contention that counsel should have presented evidence of his non-violent disciplinary infractions in prison, we conclude that, in the light of AEDPA, the Court of Criminal Appeals was not unreasonable in deciding that, although counsel was aware of this evidence and chose not to present it, his performance was not deficient. On the other hand, as discussed below, its deciding that the failure to obtain Ladd’s juvenile records was not deficient performance may have been unreasonable under AED-PA. With respect to the first factor (preparation for the sentencing phase), counsel relied on the State’s broad subpoena to the Texas Department of Corrections (“any and all records, files or documents”) and did not independently subpoena documents. All records the State received in response to the subpoena were produced to Ladd’s counsel. His counsel had prepared a subpoena for Ladd’s prison records but chose not to serve it after seeing the records provided by the prosecution; Ladd’s counsel had subpoenaed documents in prior cases and felt the State received more documents in response to its subpoenas than did the defendant. In addition to reviewing the documents provided by the State, Ladd’s counsel made the following attempts to obtain mitigating evidence: (1) counsel sent his investigator to Dallas, Texas, to try to contact family members; (2) counsel or his investigator spoke to all potential witnesses Ladd identified; and (3) counsel used a psychiatrist to advise the defense on Ladd’s future dangerousness and assist with development of defense theories. Concerning the second factor (mitigating evidence accumulated), it appears Ladd’s counsel obtained little if any. Counsel did not present any evidence at the sentencing phase. Although counsel took the above-described steps to gain mitigating evidence, he concluded the defense had nothing to dispute the State’s sentencing phase evidence. Counsel did have evidence of Ladd’s prior non-violent incidents of misconduct while in prison, but chose not to present such evidence, because counsel viewed it as aggravating. The Court of Criminal Appeals determined Ladd’s “counsel had no reliable, truthful, honest, and credible source of evidence” to counter the State’s sentencing evidence. With respect to the third factor (leads defense counsel had), neither Ladd nor his family ever advised counsel that Ladd had been sent to a juvenile facility (Gatesville State School) for an arson conviction as a child, or that Ladd, while at that facility, had been given a psychological evaluation, a prescription for a major tranquilizer, and an IQ test. On the other hand, because *359 the subject came up at trial, counsel was aware that Ladd had been arrested as a juvenile and that the prosecution’s file contained no juvenile records. Finally, regarding the fourth factor (results counsel might reasonably have expected), counsel might reasonably have expected to find mitigating evidence in Ladd’s juvenile record. Ladd asserts that the records counsel failed to obtain (they were obtained by his new habeas counsel) establish five points that should have been presented as mitigating; the State responds the records would not have assisted the defense. The question, however, is not what following the lead would have revealed, but rather “what results [counsel] might reasonably have expected” from the lead. Neal, 286 F.3d at 237 (emphasis added). Juvenile records may contain mitigating evidence (e.g., childhood abuse, childhood trauma, mental problems). See, e.g., Williams v. Taylor, 529 U.S. 362, 395, 120 S.Ct. 1495, 146 L.Ed.2d 389 (2000) (juvenile records showed defendant had been severely abused as a child and spent time in foster homes, including an abusive foster home, while his parents were imprisoned for criminal neglect). Without having viewed Ladd’s juvenile record, his counsel could not reasonably have presumed that Ladd’s record did not contain such evidence. Relying on a district attorney’s “open file” policy is not, as Ladd urges, per se unreasonable. See Strickler v. Greene, 527 U.S. 263, 282 n. 23, 119 S.Ct. 1936, 144 L.Ed.2d 286 (1999); Williams v. Head, 185 F.3d 1223, 1243 (11th Cir.), cert. denied, 530 U.S. 1246, 120 S.Ct. 2696, 147 L.Ed.2d 967 (2000). Here, however, where counsel became aware of a juvenile arrest, the total lack of juvenile records should have been noticed and investigated. In this regard, we reject Ladd’s contention that the failure to present any mitigating evidence is per se unreasonable, because mitigating evidence may not exist. See Thomas v. Gilmore, 144 F.3d 513, 516 (7th Cir.), cert. denied, 525 U.S. 1123, 119 S.Ct. 907, 142 L.Ed.2d 905 (1999); McCleskey v. Kemp, 753 F.2d 877, 900 (11th Cir.), aff'd, 481 U.S. 279, 107 S.Ct. 1756, 95 L.Ed.2d 262 (1987). The failure to obtain juvenile records, however, seems significant when viewed against this absence, especially because this was a capital case. Thus, although counsel took steps to prepare for the sentencing phase, his failure to follow a lead which could have led to the discovery of mitigating evidence, coupled with his failure to present any such evidence in this capital murder case, could be considered objectively unreasonable. As noted, the Texas court’s decision that counsel’s performance was not deficient may have been unreasonable under AED-PA. In any event, we need not decide this question. Regardless of whether the decision was unreasonable for AEDPA purposes, we hold, as discussed below, that the Texas court’s decision that there was no prejudice to Ladd was not unreasonable under AEDPA. 2. For the second (prejudice) prong for ineffective assistance, Ladd must show a “reasonable probability that the result of the proceeding would have been different but for counsel’s unprofessional errors”. Crane, 178 F.3d at 312. For prejudice in the context of failure to present mitigating evidence, Ladd must show that, but for counsel’s error, his sentence would have been “significantly less harsh”. United States v. Franks, 230 F.3d 811, 814-15 (5th Cir.2000). *360 Ladd claims five types of information that could have been ascertained had counsel subpoenaed his juvenile records: (1) his troubled childhood and lack of supervision; (2) his mental retardation diagnosis as a child; (3) his low score on an IQ test; (4) his being put on (and doing well with) a psychomotor inhibitor as a child; and (5) his good behavior in institutional settings. According to Ladd, failure to present this evidence was prejudicial, because its presentation could have resulted in a lighter sentence. First, some of the evidence to which Ladd refers is “double-edged” (mitigating and aggravating). Although the evidence of Ladd’s inadequate supervision as a child might permit an inference that he is not as morally culpable for his behavior, it also might suggest Ladd, as a product of his environment, is likely to continue to be dangerous in the future. Moreover, although presenting evidence that Ladd behaved well in juvenile detention may have been somewhat mitigating (assuming the evidence would reflect good behavior), it would have emphasized yet another arson with sexual overtones in which Ladd had been involved. (Ladd was sent to juvenile detention for setting a fire in his girlfriend’s bedroom.) Thus, because some of the evidence to which Ladd refers is “double-edged”, it is uncertain whether reasonable counsel would have used the evidence had it been available; in any event, it is unlikely to have had a significant mitigating effect had counsel presented it. In addition, the rest of the evidence is, at best, minimally mitigating. The evidence of low IQ and/or mental retardation as a child was undermined by evidence that Ladd had later obtained a GED in prison and, as an adult, had obtained a second, higher IQ score. The evidence that he had been treated with medication was mitigating, but its effect was lessened by the doctor’s withdrawing the medication after a month. Finally, and most significantly, the evidence of Ladd’s future dangerousness was overwhelming. When that is the case, it is virtually impossible to establish prejudice. E.g., Strickland, 466 U.S. at 698, 104 S.Ct. 2052 (no prejudice due to State’s overwhelming evidence on aggravating factors supporting death penalty); Jones v. Johnson, 171 F.3d 270, 277 (5th Cir.), cert. denied, 527 U.S. 1059, 120 S.Ct. 29, 144 L.Ed.2d 832 (1999) (no prejudice due to brutal facts of murder); Russell v. Lynaugh, 892 F.2d 1205, 1213 (5th Cir.1989) (no ineffective assistance “[g]iven the weakness of such testimony when juxtaposed with the overwhelming evidence of guilt, the horrifying nature of the crime, and the abundant impeachment material available to the state”). There is no question that the crime was extremely horrific and the evidence of guilt overwhelming. Moreover, as noted, the prosecution presented evidence during the sentencing phase that Ladd had previously committed a triple murder: he sexually assaulted an 18-year-old mother; stabbed her to death; and, as in the murder in the case at hand, set a fire between her legs (her two young children died from asphyxiation). That Ladd killed Gardner some 16 years after having committed a similar crime vividly demonstrates his future dangerousness. (As noted, two psychiatrists testified that, in their opinion, there was a probability that Ladd would be dangerous in the future; Ladd did not testify.) In the light of this overwhelming evidence, the modest mitigating effect any evidence cited by Ladd could have had becomes irrelevant. In short, there is no reasonable probability that Ladd’s sentence would have been different had Ladd’s counsel obtained, and decided to *361 present, information contained in Ladd’s juvenile records. The Texas court was well-within the bounds of AEDPA reasonableness in ruling that Ladd suffered no prejudice. Therefore, the district court properly rejected Ladd’s ineffective assistance claim. III. For the foregoing reasons, the denial of habeas relief is AFFIRMED.