PER CURIAM:

Brandon Rhode, a Georgia death row inmate, appeals from the district court's denial of his federal habeas corpus petition. The court granted a certificate of appealability ("COA") as to Rhode's claim of ineffective penalty phase investigation and presentation of mitigation evidence by his trial counsel. For the reasons that follow, we affirm the district court's denial of Rhode's petition.

I. BACKGROUND

A. The Crimes

The Georgia Supreme Court provided the following account of the crimes in its opinion affirming Rhode's convictions:

Rhode and his co-perpetrator, Daniel Lucas, burglarized the home of Steven and Gerri Ann Moss on April 23, 1998, fled the scene when Rhode discovered an alarm system, and returned later that day to burglarize the home again. While Rhode and Lucas were ransacking the home searching for valuables, 11-year-old Bryan Moss arrived, observed Rhode and Lucas through a front window, and entered through a back door, armed with a baseball bat. Rhode and Lucas subdued Bryan at gunpoint, sat him in a chair, and began discussing what to do with him. Lucas turned and fired at the boy, inflicting a non-fatal shoulder wound. As [Bryan's sister] Kristin Moss was approaching the house, Lucas took Bryan into a back bedroom. Rhode met Kristin as she arrived, sat her in a chair, and shot her twice with a .357 caliber pistol. Lucas repeatedly shot Bryan with a .25 caliber pistol. Rhode later shot Steven Moss with the .357 caliber pistol when Steven arrived. Finally, Lucas obtained a .22 caliber pistol from Rhode's automobile and shot Bryan and Kristin again.

Chad Derrick Jackson, Rhode's roommate, testified that he observed Rhode and Lucas handing rifles and other items out of Jackson and Rhode's bedroom window and loading them into Rhode's automobile on the evening of the crimes. Jackson further testified that Rhode and Lucas admitted to him the next day that Lucas first shot Bryan in the shoulder, that Lucas then shot Bryan while Rhode simultaneously shot Kristin, that Rhode next shot Steven Moss, and that lastly Lucas shot each victim to ensure their deaths.

Danny Ray Bell, who also lived in the same house as Rhode, testified that Rhode and Lucas spoke to him between the two burglaries and that Bell advised Rhode not to return to burglarize the same home. Bell testified that, at the time of this conversation, Rhode had a .357 caliber pistol in his waistband. According to Bell, when Rhode returned from the second burglary, Rhode said that he had "messed up big time" and needed to dispose of some weapons and other items. Rhode admitted to Bell that Lucas shot a young boy and that Rhode shot a girl and a man.

Several witnesses testified that they saw an automobile similar to Rhode's at or near the victims' home on the day of the murders. A search of Rhode's automobile revealed damage to the front and rear bumpers and a spare tire in the trunk that showed signs of use. A photograph of the crime scene suggested a vehicle had backed into a gas tank at the victims' home. Expert testimony disclosed that paint on a cement block at the victims' home matched the paint on Rhode's automobile, including two layers applied at the factory and a third layer likely applied later. Additional expert testimony indicated that a crime scene imprint could have been made by Rhode's spare tire.

Rhode made a statement admitting he fired two times at Kristin with the .357 caliber pistol, and he led law enforcement officers to two locations where he and Lucas had secreted weapons and other items. Expert testimony matched the found .357 and .25 caliber pistols to bullets retrieved from the crime scene and the victims' bodies.

Rhode v. State, 274 Ga. 377, 552 S.E.2d 855, 858-59 (2001).

B. Procedural History

The jury found Rhode guilty of three counts of malice murder, three counts of felony murder, two counts of burglary, and one count of kidnapping with bodily injury. Id. at 858. Rhode's felony murder convictions were vacated by operation of law. Id. He was sentenced to death after the jury concluded that death sentences were warranted for the murder convictions. Id.

On March 3, 2000, Rhode moved for a new trial. The court denied his motion on December 22, 2000.

Rhode then appealed to the Georgia Supreme Court, which affirmed his convictions and death sentences on October 1, 2001. The court denied Rhode's motion for reconsideration on October 22, 2001. Rhode petitioned for a writ of certiorari, which the U.S. Supreme Court denied on June 17, 2002.

On April 4, 2003, Rhode filed a state habeas corpus petition to challenge his convictions and sentences. After conducting an evidentiary hearing, the state habeas court denied relief on all claims. In denying relief, the court adopted as its own the State's proposed order and dated it March 14, 2006. Rhode then filed an Application for Certificate of Probable Cause to Appeal, which the Georgia Supreme Court denied on April 24, 2007. Ex. 84.

On June 25, 2007, Rhode filed pursuant to 28 U.S.C. § 2254 a federal habeas corpus petition in the Middle District of Georgia. The district court denied Rhode's petition but issued a COA.

II. STANDARDS OF REVIEW

Rhode's petition is governed by the Antiterrorism and Effective Death Penalty Act's ("AEDPA") "highly deferential standard for reviewing state court judgments." McNair v. Campbell, 416 F.3d 1291, 1297 (11th Cir. 2005) (citation and quotation marks omitted). Under AEDPA, a federal court may not grant habeas relief with respect to any claim adjudicated on the merits in state court unless the state court's adjudication

(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.

28 U.S.C. § 2254(d). The statutory phrase "clearly established Federal law" "refers to the holdings, as opposed to the dicta, of [the U.S. Supreme] Court's decisions as of the time of the relevant state-court decision." Williams v. Taylor, 529 U.S. 362, 412, 120 S.Ct. 1495, 1523, 146 L.Ed.2d 389 (2000) (majority opinion of O'Connor, J.). "[A] determination of a factual issue made by a State court shall be presumed to be correct. The [petitioner] shall have the burden of rebutting the presumption of correctness by clear and convincing evidence." 28 U.S.C. § 2254(e)(1).

"When examining a district court's denial of a § 2254 habeas petition, we review questions of law and mixed questions of law and fact de novo, and findings of fact for clear error." Grossman v. McDonough, 466 F.3d 1325, 1335 (11th Cir. 2006). "An ineffective assistance of counsel claim is a mixed question of law and fact subject to de novo review." McNair v. Campbell, 416 F.3d 1291, 1297 (11th Cir. 2005).

The U.S. Supreme Court established in Strickland v. Washington, 466 U.S. 668, 104 S.Ct. 2052, 80 L.Ed.2d 674 (1984), the legal principles governing ineffective assistance claims. "An ineffective assistance claim has two components: A petitioner must show that counsel's performance was deficient, and that the deficiency prejudiced the defense." Wiggins v. Smith, 539 U.S. 510, 521, 123 S.Ct. 2527, 2535, 156 L.Ed.2d 471 (2003) (citing Strickland, 466 U.S. at 687, 104 S.Ct. 2052, 80 L.Ed.2d 674).

We have explained that:

The petitioner satisfies the test's performance prong by proving that counsel's performance failed to meet the standard of reasonableness under prevailing professional norms. Our evaluation of counsel's performance is highly deferential; we must indulge a strong presumption that counsel's performance was reasonable and that counsel made all significant decisions in the exercise of reasonable professional judgment. We review counsel's performance from counsel's perspective at the time, to avoid the distorting effects of hindsight. Our review is objective, in that we consider whether there was any reasonable justification for the attorney's conduct. Thus, the petitioner must establish that no competent counsel would have taken the action that his counsel did take.

The petitioner satisfies the Strickland test's prejudice prong by showing that there is a reasonable probability that, but for counsel's unprofessional errors, the result of the proceeding would have been different. A reasonable probability is a probability sufficient to undermine confidence in the outcome.

Newland v. Hall, 527 F.3d 1162, 1184 (11th Cir. 2008) (citations and quotation marks omitted). The prejudice prong does not focus only on the outcome; rather, to establish prejudice, the petitioner must show that counsel's deficient representation rendered the result of the trial fundamentally unfair or unreliable. See Lockhart v. Fretwell, 506 U.S. 364, 369, 113 S.Ct. 838, 842, 122 L.Ed.2d 180 (1993) ("[A]n analysis focusing solely on mere outcome determination, without attention to whether the result of the proceeding was fundamentally unfair or unreliable, is defective.").

III. DISCUSSION

Rhode argues that the district court erred by denying federal habeas relief. First, he argues that counsel rendered ineffective assistance in the investigation of mitigation evidence for the penalty phase of his trial. Second, he argues that counsel's presentation of mitigating evidence at the penalty phase was ineffective. Finally, he argues that the state habeas court's decision to reject his ineffective assistance claims is contrary to clearly established U.S. Supreme Court precedent. All of these arguments fail. We address each in turn.

A. Counsel's Strategy

Before assessing Rhode's ineffective assistance claims, "we must determine the strategy actually pursued by counsel." Blankenship v. Hall, 542 F.3d 1253, 1273 (11th Cir. 2008). Rhode asserts that counsel decided from the beginning that this was a mitigation-only case. The state habeas court found, however, that counsel thought that the penalty phase strategy would involve both mitigation and residual doubt. Ex. 80 at 8.

The state habeas court's finding is not unreasonable. Counsel testified at the evidentiary hearing that the defense believed that the jury, while deliberating during the guilt phase, was "grappling with residual doubt on some portions of the case, and that [its residual doubt] would carry over to the penalty phase." Ex. 59 at 3106. Counsel's trial tactics confirm that belief.

Counsel conceded that Rhode was at the crime scene, but counsel otherwise "fought everything in that case." Ex. 59 at 3063. During the guilt phase, counsel tried to show that Rhode's co-defendant Lucas murdered the three victims while Rhode fired only one shot after Rhode "turned his head and closed his eyes." Ex. 12 at 2009. Rhode's penalty phase testimony supported the defense's account of the murders and may have addressed any residual doubt among the jurors. In addition to providing mitigating testimony about his childhood, Rhode testified that it was Lucas' idea to go to the Moss home on the day of the murders, that he remembered "freezing up" when the shooting started, and that he did not know where the one shot he fired went. Ex. 17 at 3997, 4010-21. During closing argument of the sentencing phase Rhode's attorney addressed the jury directly on the possibility of residual doubt. Counsel's decision to "f[i]ght everything" belies Rhode's assertion that counsel knew that this was a mitigation-only case.

Even if this were a mitigation-only case, we would still conclude that counsel's penalty phase investigation and presentation were not ineffective. We have explained that "even when trial counsel's investigation is less complete than collateral counsel's, trial counsel has not performed deficiently when a reasonable lawyer could have decided, in the circumstances, not to investigate[]" further. Housel v. Head, 238 F.3d 1289, 1295 (11th Cir. 2001). We have also explained that "counsel [is not] required to present all mitigation evidence, even if the additional mitigation evidence would not have been incompatible with counsel's strategy." Chandler v. United States, 218 F.3d 1305, 1319 (11th Cir. 2000) (en banc). As we explain further below, the state habeas court did not make an unreasonable determination of the facts or contravene clearly established U.S. Supreme Court precedent when it rejected Rhode's claim that counsel rendered ineffective penalty phase investigation and presentation.

B. State Habeas Court Order

Rhode complains that the state habeas court adopted verbatim the State's proposed order as its own. Because the court adopted the proposed order verbatim, Rhode asserts, "the order uncritically incorporates [the State's] selective use of evidence and mischaracterizations of the evidentiary record." Appellant's Br. 23. He characterizes the order as an "artifact of [the State's] having drafted [it] with the specific intent, not of producing a fair and impartial assessment of the facts and law, but of deliberately glossing over or camouflaging significant attorney errors in order to ensure that those errors are shielded from any meaningful review." Appellant's Br. 24. Rhode argues that because the state habeas court's "partisan final order" is based on . . . "unreasonable determinations of fact," the district court erroneously denied federal habeas relief. Appellant's Br. 24.

Rhode seems to be pointing out that the state habeas court adopted verbatim the State's proposed order simply to emphasize his position that the findings of the state habeas court were based on a State centered interpretation of the facts. However, the record clearly reflects that both Rhode and the State had the opportunity to present the state habeas court with their version of the facts. See Ex. 73, 75. Despite the fact that the state habeas court adopted the State's facts verbatim, these findings of fact are still entitled to deference from this court unless Rhode can show the facts to be clearly erroneous. As discussed further below, the record supports the findings of fact of the state habeas court.

C. Investigation

Rhode argues that counsel inadequately and untimely prepared for the penalty phase because counsel failed to oversee an adequate investigation. He further argues that the state habeas court made an unreasonable determination of the facts when it found that counsel was personally involved in the mitigation investigation. Rhode argues that the investigation was untimely because counsel spent the equivalent of one standard work week (40.16 hours) on the case in the first year of representation and because the defense lost opportunities to develop critical evidence. These arguments are without merit.

1. Performance Prong

a. Inadequate Investigation

Rhode's argument that counsel's investigation was inadequate is without merit. Rhode's counsel included two experienced death penalty lawyers, Frank Ford and Jack Nebl. Ford, who was lead counsel, had practiced criminal law for about nine years by the time he was appointed to the case in June 1998. Ex. 59 at 3046-50. At that time, Ford had already handled several death penalty cases and had attended death penalty seminars annually for six or seven years. Ex. 59 at 3050-54. He asked that Nebl be appointed as co-counsel because Nebl had previous death penalty experience. Ex. 59 at 3056-57. Their extensive experience is important because "[o]ur strong reluctance to second guess strategic decisions is even greater where those decisions were made by experienced criminal defense counsel." Provenzano v. Singletary, 148 F.3d 1327, 1332 (11th Cir. 1998).

We agree with the district court that the state habeas court did not unreasonably find that counsel's investigation was adequate. Ford contacted Rhode's mother, Patches Rhode, almost immediately after his appointment to the case, seeking Rhode's school and medical records. Ex. 59 at 3121. He reviewed tapes of witness statements and the crime scene. Ex. 59 at 3164. He also obtained funding for and hired Dr. Dan Grant, a defense psychologist, and Cheryl Abernathy, a mitigation investigator who had handled over twenty-five death penalty cases. Ex. 59 at 3059-61, 3069, 3119-24, 3205.

Abernathy met with Rhode at least five times to collect information about his life. Ex. 61 at 3896-3901, Ex. 62 at 3903-18. She traveled to Louisiana and Mississippi to interview at least ten potential mitigation witnesses. Ex. 59 at 3215, Ex. 62 at 4004. She prepared her assistant Christopher DiPietro for his own trip to the two states to interview seventeen more potential mitigation witnesses. Ex. 59 at 3224-27, Ex. 62 at 4008. DiPietro shared the information he gathered with Abernathy; she, in turn, provided counsel with all of the information that she and DiPietro had gathered. Ex. 59 at 3207-08. She and DiPietro also interviewed several potential witnesses in Macon, Georgia. They discussed with counsel the results of those interviews. Ex. 59 at 3240-48.

Through the investigation, Abernathy sought records that might contain mitigation evidence, including those from various hospitals, sheriffs' offices, and schools. She forwarded to counsel all of the records that she received. Ex. 59 at 3207-08. And she had numerous telephone conversations with lead counsel about those records. Ex. 59 at 3228-29, 3233.

Although "[p]revailing norms of practice as reflected in American Bar Association standards . . . are [only] guides to determining what is reasonable," Strickland, 466 U.S. at 688, 104 S.Ct. at 2065, counsel's investigation here met those prevailing norms. One of those norms is that "[c]ounsel should conduct interviews of potential witnesses in the presence of a third person so that there is someone to call as a defense witness at trial. . . . [or] have an investigator or mitigation specialist conduct the interviews." American Bar Association Guidelines for the Appointment and Performance of Defense Counsel in Death Penalty Cases, 31 HOFSTRA L. REV. 913, 1020 (2003). Since Rhode's counsel hired investigators who interviewed potential witnesses and shared all of their information with counsel, we cannot say that counsel performed deficiently by delegating the mitigation investigation to them.

And, contrary to Rhode's assertion, the state habeas court did not make an unreasonable determination of the facts when it found that counsel was personally involved in the mitigation investigation. Just because counsel did not personally gather all of the mitigation evidence does not mean that counsel was not personally involved in the investigation. Counsel evaluated all of the evidence that the investigators gathered; consulted them; and, with their help, decided which witnesses should be called to testify at the penalty phase. Ex. 59 at 3088, 3207-08. Thus, the record supports the state habeas court's finding that counsel was personally involved in the mitigation investigation.

b. Untimely Investigation

Rhode's argument that counsel's investigation was untimely is also without merit. Contrary to his intimation, the effectiveness of counsel's representation at sentencing is not an exact derivative of the amount of time counsel spends investigating mitigation evidence. See Conklin v. Schofield, 366 F.3d 1191, 1202 (11th Cir. 2004). Rather, "the time and effort [that counsel spent] in preparing to defend [Rhode] in the guilt phase of [this] capital case continues to count at the sentencing phase." Chandler, 218 F.3d at 1320 n. 27. In contrast to Rhode's argument that counsel only spent one standard work week on his case in the first year, Ford's billing records reflect that he alone spent over one hundred and twenty five hours working on Rhode's case in the last eight months prior to trial. See Ex. 59 at 3168-81. The extensive investigation here confirms counsel's initial belief that the defense strategy would involve mitigation, though not exclusively. See Ex. 59 at 3063.

Rhode notes that Abernathy testified at the state habeas evidentiary hearing that she would have wanted more time to investigate. Ex. 59 at 3261, 3268-69. Her testimony is not, however, a sufficient basis to grant federal habeas relief because the record does not establish that she communicated that thought to Rhode's lawyer. Ex. 59 at 3065-67. The state habeas court noted Ford's testimony that he would have filed an ex parte motion asking for more time and funds if Abernathy had told him that she needed more time to investigate. Ex. 59 at 3066-67; Ex. 80 at 16. The record supports the state habeas court's finding that Abernathy did not tell counsel that she wanted more time to investigate. We agree with the state habeas court that counsel did not ineffectively investigate the mitigation evidence for the penalty phase and that counsel did not, in light of the chosen defense strategy, perform deficiently.

2. Prejudice Prong

Even if counsel performed deficiently by not personally gathering mitigation evidence or by not initially spending much time on the case, counsel's performance was not prejudicial. Counsel reviewed all of the information presented to them, extensively discussed it with the investigators, and, as explained further below, could make strategic decisions based on it. Thus, Rhode fails to show that counsel's investigation rendered "the result of the proceeding . . . fundamentally unfair or unreliable." Lockhart, 506 U.S. at 369, 113 S.Ct. at 842.

The district court properly denied federal habeas relief as to Rhode's ineffective preparation claim.

D. Presentation

Rhode argues that counsel ineffectively presented mitigation evidence by (1) unreasonably failing to inquire meaningfully into or present any mental health related mitigating factors other than Rhode's ability to adapt to prison despite having promised the jury such evidence, (2) unreasonably failing to use records of Rhode's prior psychiatric hospitalization at age thirteen or to contact any staff from the hospital, and (3) failing to review or use his juvenile probation records.

In raising these arguments, Rhode faults counsel for failing to call certain witnesses and not inquiring into certain issues with others, and not introducing evidence that could have been presented. But "[w]hich witnesses, if any, to call, and when to call them, is the epitome of a strategic decision, and it is one that we will seldom, if ever, second guess." Waters v. Thomas, 46 F.3d 1506, 1512 (11th Cir. 1995) (en banc). Furthermore, "[i]t is well-settled in this Circuit that a petitioner cannot establish an ineffective assistance claim simply by pointing to additional evidence that could have been presented." Van Poyck v. Florida Dep't of Corr., 290 F.3d 1318, 1324 (11th Cir. 2002) (per curiam).

Here, counsel had nine lay witnesses, Dr. Grant, and Rhode himself testify during the penalty phase. Counsel testified at the state habeas evidentiary hearing that the defense team chose these specific witnesses to avoid presenting duplicate testimony. Ex. 59 at 3127. The defense team, counsel testified, tried to figure out what each witness could contribute uniquely and informed each what he or she would be asked while on the witness stand. Ex. 59 at 3127. Through the defense's eleven witnesses, counsel developed the following mitigation themes: (1) Rhode could adapt to prison and not be a future danger so there was no need to put him to death; (2) Rhode had a bad childhood; (3) Rhode was constantly failed by the adults in his life; (4) Rhode started to abuse drugs when he was just a child; (5) Rhode was a follower led by Lucas, his co-defendant who was more responsible for the crimes; (6) Rhode was not evil and had done good deeds in the past; and (7) many innocent people would be penalized if Rhodes were sentenced to death. For the reasons explained further below, Rhode's argument that counsel ineffectively presented mitigation evidence is without merit.

1. Performance Prong

a. Mental Health

Rhode faults counsel for failing to deliver on the defense's promise to the jury that an expert witness, Dr. Grant, would testify "as to [Rhode's] psychological state . . . and [would] have a definite opinion as to how [Rhode] became involved in this type of life style and eventually became involved with [the crimes]." Ex. 17 at 3642. Contrary to Rhode's assertion, however, Dr. Grant testified accordingly. Dr. Grant told the jury that Rhode's drug use interfered with his ability to learn to deal with people and that the lack of structure at home caused many of his problems. He further testified that Rhode would be able to adapt to prison and not be a danger to anyone because prison provides a structured environment with strict limits and few choices. He explained that if drugs and alcohol were removed from Rhode's life, their absence would "eliminate a lot of . . . problems." Ex. 17 at 4080.

Rhode faults counsel for not hiring an expert like Dr. David Fassler, a psychiatrist retained by Rhode who testified at the state habeas evidentiary hearing. See Ex. 47 at 212-61. Dr. Fassler testified that Rhode's development of severe polysubstance addiction occurred through his exposure from infancy to drug and alcohol abusing family members at every turn. Dr. Fassler concluded that Rhode had no real choice in the development of his addiction, which extended through the time of the crime. He also offered expert testimony about Rhode's problems with judgment and impulse control. He testified that Rhode's organic brain damage impaired his ability to control his impulses and behavior. Ex. 47 at 241-42. Rhode contends that Dr. Fassler's testimony is the kind of testimony that counsel promised to present during the penalty phase but never delivered.

Dr. Fassler's testimony, however, merely expands upon that of Dr. Grant, who also found, but as part of the defense strategy did not tell the jury, that Rhode had organic brain damage. The emergence of an additional expert witness at a state collateral proceeding does not mean that trial counsel was ineffective. See Hendrix v. Sec'y, Fla. Dep't of Corr., 527 F.3d 1149, 1154 (11th Cir. 2008). "[C]laims based on such witnesses are made seemingly without regard to the trial counsel's actual investigation and the basis for his strategic decisions." Id. Counsel reasonably believed that the jury would see Rhode's impulsive behavior, which more than one expert believed was triggered by his organic brain damage, as aggravating.

Rhode also faults counsel for not introducing Dr. Jerold Lower's report. Dr. Lower, a State psychologist who evaluated Rhode, reported that although Rhode did not suffer from a major disorder of thought or mood that rendered him insane at the time of the crimes, Ex. 50 at 1079-84, Rhode's most serious psychiatric problem was his substance abuse and his resulting organic brain damage. Ex. 50 at 1081.

The state habeas court reasonably concluded, however, that counsel strategically decided not to introduce Dr. Lower's report. The report would have exposed to the jury how lightly Rhode made of his drug abuse. It would also have exposed Rhode's view of himself as antisocial and unable to learn from punishing experiences. Ex. 80 at 34-35.

Keeping Lower's report away from the jury was reasonable because counsel, through Dr. Grant, tried to show that Rhode could adapt to prison. Since Rhode's counsel could have reasonably believed that the jury would consider the report aggravating rather than mitigating, we cannot say that counsel performed deficiently by not presenting it.

b. Prior Records

Rhode further faults counsel for failing to use Rhode's prior psychiatric hospitalization records or juvenile probation records and not contacting anyone from New Orleans Adolescent Hospital ("NOAH"), including psychiatric social worker Frances Wellington, who had treated Rhode and could have offered mitigation testimony. Rhode argues that counsel's failure to use that evidence prejudiced the defense because the evidence would have rebutted Patches Rhode's intimation that her son had wilfully rejected treatment.

The record supports the state habeas court's finding that counsel had copies of the NOAH records and juvenile probation records before trial and gave them to Dr. Grant for his review. The record also shows, for example, that counsel obtained and reviewed a June 1993 psychological assessment by Dr. Jorge H. Durana, a NOAH clinical psychologist. Dr. Durana evaluated Rhode after Rhode was arrested as a juvenile on five felony charges and two counts of misdemeanor trespassing. Dr. Durana found that Rhode had a significant history of antisocial behavior and a disregard for rules and the well being of others. Ex. 62 at 3947. Rhode's NOAH records, in fact, indicate that he was disciplined during his treatment for breaking major rules. They show that he was frequently uncooperative, undermining, and generally in denial. Ex. 62 at 3943, 3951.

Counsel's decision not to introduce Rhode's NOAH and juvenile probation records at trial was strategically reasonable because the jury could have seen them as aggravating and inconsistent with counsel's argument that Rhode could adapt to prison. Counsel did not perform deficiently by failing to present Wellington's testimony because the emergence of an additional expert witness at a state collateral proceeding does not mean that trial counsel was ineffective. Hendrix, 527 F.3d at 1154.

2. Prejudice Prong

Even if counsel performed deficiently by failing to present the above mitigating evidence, Rhode still fails to show prejudice. First, his experienced counsel had the opportunity to evaluate the evidence that Rhode argues counsel should have presented during the penalty phase. Counsel strategically determined, however, that the jury could view that evidence as aggravating. We are strongly reluctant to second guess these types of determinations by counsel. Provenzano, 148 F.3d at 1332.

Second, much of the evidence that Rhode faults counsel for not presenting to the jury is potentially aggravating or cumulative. Counsel is not required to present cumulative evidence or evidence incompatible with the defense strategy. Van Poyck, 290 F.3d at 1324 n. 7; Chandler, 218 F.3d at 1319. Dr. Fassler's testimony would have merely expanded upon Dr. Grant's, which the penalty phase jury heard. Dr. Lower's report, Dr. Durana's psychological assessment, and Rhode's NOAH and juvenile probation records would have, at worst, been perceived as inconsistent with counsel's argument that the jury should consider a penalty other than death and that Rhode could adapt to prison. At best, the evidence would have been cumulative, providing more information about Rhode's bad childhood and early exposure to drugs and alcohol. Because the evidence that Rhode faults counsel for failing to present is either inconsistent with the defense strategy or cumulative, he cannot establish ineffective assistance in counsel's failure to present it.

Finally, Rhode testified during the penalty phase. He told the jury, "I accept what I've done. But in all fairness, I can't blame anyone else. You know, no one. It's my responsibility." Ex. 17 at 4004. By "[choosing] to testify on his own behalf, he ran the risk that the jury might conclude the opposite of his testimony is true." Atkins v. Singletary, 965 F.2d 952, 961 n. 7 (11th Cir. 1992). For these reasons, Rhode fails to show "a reasonable probability that, but for counsel's [supposed] errors, the result of the proceeding would have been different." Newland, 527 F.3d at 1184.

The district court properly denied federal habeas relief as to Rhode's ineffective presentation claim.

E. "Contrary to"

Rhode argues that the state habeas court's decision is contrary to Williams v. Taylor, 529 U.S. 362, 120 S.Ct. 1495, 146 L.Ed.2d 389 (2000) and Rompilla v. Beard, 545 U.S. 374, 125 S.Ct. 2456, 162 L.Ed.2d 360 (2005). In so arguing, he asserts that his jury appeared to struggle with its verdicts during both the guilt and penalty phases. He argues that the jury's struggle suggests that had counsel presented more mitigating evidence, there is a strong likelihood that "at least one juror would have struck a different balance." Wiggins, 539 U.S. at 537, 123 S.Ct. at 2543. Rhode argues that the state habeas court overlooked the jury's struggle when it found that counsel's performance did not prejudice the defense. In overlooking that struggle, he argues, the state habeas court failed to assess whether the entire record raised a reasonable probability that the outcome would be different.

A state court decision is "contrary to" the U.S. Supreme Court's clearly established precedent "if the state court applies a rule that contradicts the governing law set forth in [the U.S. Supreme Court's] cases," Williams, 529 U.S. at 405, 120 S.Ct. at 1519, or "if the state court confronts a set of facts that are materially indistinguishable from a decision of [the U.S. Supreme] Court and nevertheless arrives at a result different from [the Court's] precedent," id. at 406, 120 S.Ct. at 1519-20. The phrase "clearly established Federal law" "refers to the [Court's] holdings," not dicta. Id. at 412, 120 S.Ct. at 1523.

The state habeas court's decision here is not contrary to clearly established U.S. Supreme Court precedent because Rhode's case is not materially indistinguishable from Williams and Rompilla. Unlike counsel in those two cases, Rhode's counsel did not, as explained above, perform deficiently. Unlike Williams' counsel, who did not adequately investigate for mitigating evidence, Rhode's counsel strategically decided not to present certain mitigating evidence after a thorough investigation. Cf. Williams, 529 U.S. at 396, 120 S.Ct. at 1514 ("[T]he failure to introduce the comparatively voluminous amount of evidence that did speak in Williams' favor was not justified by a tactical decision to focus on Williams' voluntary confession."). And, unlike Rompilla's counsel, Rhode's counsel did not fail to examine any file that the prosecutor warned would be used at trial. Cf. Rompilla, 545 U.S. at 377, 125 S.Ct. at 2460 ("hold[ing] that . . . [a capital defendant's] lawyer is bound to make reasonable efforts to obtain and review materials that counsel knows the prosecutor will probably rely on as evidence of aggravation at the sentencing phase of trial").

The state habeas court's decision is not contrary to clearly established U.S. Supreme Court precedent.

IV. CONCLUSION

The district court did not err by denying federal habeas relief as to Rhode's claim that counsel rendered ineffective penalty phase investigation and presentation. Rhode has failed to show that the state habeas court's decision is contrary to, or an unreasonable application of, clearly established federal law, or resulted in a decision based on an unreasonable determination of the facts in light of the evidence presented.

AFFIRMED.