TJOFLAT, Circuit Judge:

William Emmett LeCroy, Jr. is a federal death-row inmate. In 2004, he was convicted in the Northern District of Georgia of taking a motor vehicle by force from Joann Lee Tiesler, resulting in her death. See 18 U.S.C. § 2119(3) (2010). LeCroy was sentenced to death, and this court affirmed his conviction and sentence on direct appeal. United States v. LeCroy, 441 F.3d 914 (11th Cir.2006), cert. denied 550 U.S. 905, 127 S.Ct. 2096, 167 L.Ed.2d 816 (2007). LeCroy then petitioned the District Court to vacate his sentence pursuant to 28 U.S.C. § 2255 (2010). The court, after holding a three-day evidentiary hearing, denied his petition. United States v. LeCroy, Nos. 2:02-CR-38-RWS-SSC, 2:08-CV-2277-RWS, 2012 WL 1114238 (N.D.Ga. Mar. 30, 2012). LeCroy now appeals that ruling. His claim is that he was denied the effective assistance of counsel during the penalty phase of his trial in violation of the Sixth Amendment.1 See Strickland v. Washington, 466 U.S. 668, 104 S.Ct. 2052, 80 L.Ed.2d 674 (1984). After reviewing the record and the parties' briefs, and after hearing oral argument, we affirm the judgment of the District Court.

I.

The basic facts of the crime for which LeCroy was convicted and sentenced have never been in dispute: on October 7, 2001, LeCroy broke into the home of Joann Tiesler, raped and murdered her, and fled in her car to the Canadian border, where he was arrested two days later. But because evaluating LeCroy's § 2255 claim requires an understanding of his life and background, we begin in Subpart A with a sketch of LeCroy's biography up to the murder, then describe the murder and his arrest. Subpart B deals with LeCroy's attorneys, their investigation of the case, and their strategic choices heading to trial. Subpart C recounts the events at trial, and Subpart D describes the § 2255 proceedings in the District Court.

A.

LeCroy was born in 1970 in Marietta, Georgia, to William Emmett LeCroy, Sr., and Donna Houston. At seventeen, shortly after his parents divorced, he joined the United States Army. Stationed in Hawaii, LeCroy went absent-without-leave and lived on the streets of Honolulu, supporting himself by breaking into homes to steal food. He was arrested in 1989 and discharged from the Army. Then nineteen years old, LeCroy moved back in with his mother—who had since married Sam Houston, a former police partner of LeCroy's father—in Cobb County, Georgia, just northwest of Atlanta.

While living with his mother and stepfather, LeCroy began a sexual relationship with one of Sam Houston's two daughters, Alecia, who was just shy of her fourteenth birthday. The family discovered the relationship in January, 1990, when Alecia's mother—Sam Houston's ex-wife—found a note written by Alecia to a friend detailing her sexual encounters with LeCroy. Alecia's mother reported the incident to the police and urged local authorities to charge LeCroy with statutory rape.

Meanwhile, the Cobb County police were investigating a string of burglaries that occurred between late 1990 and early 1991. The police identified LeCroy as a suspect and arrested him on March 3, 1991, following a traffic stop. Police found a gun in LeCroy's car along with several handwritten notes.2

LeCroy was first convicted in state court for aggravated assault, burglary, child molestation, and statutory rape. As he was serving his sentence for those crimes, he was convicted in federal court for possession of a sawed-off shotgun, which he had obtained in one of his burglaries. He served an additional five years in federal prison for that offense. In total, he was incarcerated for just over ten years.

LeCroy was released from federal prison in August 2001 and subject to a three-year term of supervised release. He moved back in with his mother and Sam Houston, who had since relocated to Blue Ridge, Georgia, in the mountains south of the Tennessee border. LeCroy was required to undergo a psychosexual evaluation as a condition of supervised release, but LeCroy left the evaluation prematurely. His probation officer warned that if he refused the evaluation he risked being sent back to prison, so LeCroy agreed to return to complete the evaluation in late October.

Later, at trial, the Government would contend that LeCroy never intended to make good on this agreement to submit to the evaluation. LeCroy was, according to his increasingly concerned stepfather, spending time alone in his room on the computer. Investigators determined the computer was used to search for survival gear, and to scan and copy Sam Houston's passport. Investigators also found a "need to acquire" list written on the back of the letter scheduling LeCroy's original evaluation; the list included binoculars, boots, gloves, guns, ammunition, food, and water.

LeCroy's mother and stepfather went away for the weekend on October 5, leaving LeCroy alone in their cabin. That weekend there was a series of robberies in the neighborhood, including the theft of medical supplies, a shotgun, and ammunition.

On the evening of October 7, LeCroy broke into Joann Tiesler's home through a bedroom window, armed with a shotgun, a knife, and plastic cable ties. LeCroy made sure to return the open window blinds to their original position, so that from the outside Tiesler's house appeared undisturbed. Tiesler, a nurse, had been away for the weekend visiting her fiancé in Rome, Georgia. She returned home around 6:00 p.m.

As Tiesler entered her home, LeCroy approached from behind and struck her on her head with the butt of his shotgun, causing the gun to discharge in the hallway outside her bedroom. LeCroy bound Tiesler's hands behind her back with the plastic cable ties. LeCroy stripped her and raped her vaginally and anally. After that, he strangled her with an electrical cord, slashed her throat with his knife, and stabbed her five times in the back before wiping the knife off on her shirt. LeCroy left Tiesler's house and drove away in her car. A real-estate agent and one of Tiesler's coworkers discovered her body the next day, naked and bound on her bed.

LeCroy was arrested on October 9, 2001, two days after Tiesler's murder. He was captured at the border between Minnesota and Canada, still driving Tiesler's car. Inside the car, police found a knife covered in Tiesler's blood and plastic cable ties like those used to bind Tiesler's wrists. A note found in the car read, "Please call the police and report this vehicle as stolen. Thanks, The Thief." A second note, written on the back of a map, read "Please please please forgive me Joanne [sic]. You were an angel and I killed you. Now I have to live with that and I can never go home. I am a vagabond and doomed to hell."

B.

On May 15, 2002, a Northern District of Georgia grand jury indicted LeCroy for taking a motor vehicle by force, violence, and intimidation from Joann Tiesler, resulting in her death. See 18 U.S.C. § 2119(3). A superseding indictment, dated August 13, 2002, added special death-eligibility allegations. LeCroy was appointed a team of lawyers from the Northern District of Georgia Federal Defender Program: Paul Kish, Stephanie Kearns, and Brian Mendelsohn.3 Later, a fourth attorney—Daniel Summer—was appointed as local counsel from the Gainesville area, where the trial would be held.4 The attorneys agreed that Kish and Summer would focus on the guilt phase of trial, while Kearns and Mendelsohn would focus on the penalty phase.

With the help of two investigators, Susan Miller and Michael Hutcheson, the defense attorneys gathered extensive records of LeCroy's background. They gathered school records, military records, records relating to LeCroy's earlier criminal convictions, prison records, and police reports relating to Tiesler's murder. The attorneys delivered all these records to Doctor Michael Hilton, a forensic psychiatrist the attorneys hired on March 27, 2003, to conduct an evaluation of LeCroy. Paul Kish testified at the § 2255 hearing that the public defender's office "quite often used [Hilton]" to conduct these sorts of evaluations. Collateral Tr. Vol. I at 14. As Paul Mendelsohn later testified, the attorneys never planned to call Doctor Hilton as a witness at trial; his evaluation was, instead, to be a "test run" so that the defense team could see what a psychiatric evaluation of LeCroy would reveal before making any final strategic judgments: "[T]he idea behind Dr. Hilton was to see what would happen, what kind of results we would get in an evaluation if we just did a straight evaluation.... [W]e were going to see what results he came back with, and then with that information in hand, proceed with the rest of the case." Collateral Tr. Vol. II at 265.

Doctor Hilton met with LeCroy in prison for four and a half hours, and prepared a set of reports for defense counsel. One report related Doctor Hilton's conclusion that LeCroy was competent and that he could not present an affirmative defense of not guilty by reason of insanity. Another report summarized Doctor Hilton's psychiatric evaluation of LeCroy, reflecting Doctor Hilton's diagnoses and professional observations.

During the evaluation, LeCroy told Doctor Hilton about his troubled upbringing. His parents' marriage had been an unhappy one: his father—William LeCroy, Sr.—was verbally abusive toward his mother, a "loving, gentle, timid woman." Hilton Report at 3. The two eventually divorced, and one year later LeCroy's mother married Sam Houston.

LeCroy told Doctor Hilton that when he was eight years old he had sexual encounters with a female babysitter he knew as "Tinkerbell." LeCroy and his brother played a "kissing game" with her: they would run up to her, kiss her, and then run away. One night, Tinkerbell came into LeCroy's bedroom and told him that he needed to "know how to do it right," and began kissing him, undressing him, and performing oral sex on him. A week later, Tinkerbell again molested him and the two may have had intercourse. The day after that, however, LeCroy attempted to visit Tinkerbell—who lived in an apartment above the LeCroys' apartment—and discovered her coming down the stairs arm-in-arm with a boyfriend. Tinkerbell gave LeCroy what he called a "malevolent" smile. The LeCroy family moved a week later, and he never saw Tinkerbell again or told his family what happened.

Turning to Tiesler's murder, LeCroy described the crime to Doctor Hilton in lurid detail. After being released from federal prison in August 2001 and moving in with his mother and Sam Houston in Blue Ridge, Georgia, LeCroy began commuting by motorcycle to the town of Marietta, where he worked with his father. He passed Joanne Tiesler's house every day and the two would usually wave to each other. In early October 2001, LeCroy—convinced that the government was determined to put him back in prison—traveled out into the woods, where he had hidden a cache of survival gear.

On his way to check up on his cache in the woods, LeCroy passed Tiesler's house and waved to her. She did not wave back. Later, at his hiding spot in the woods, LeCroy heard the sound of car tires on gravel and turned to see Tiesler driving toward him in a sport-utility vehicle. LeCroy told Doctor Hilton that Tiesler stopped, said "Huh!" through her half-open driver's side window, and then turned her vehicle around and drove away.

LeCroy's brief interaction with Tiesler in the woods apparently disturbed him a great deal. He began to dwell on Tiesler, until his thoughts turned to the possibility that Tiesler was, in fact, Tinkerbell, the babysitter who had sexually abused him as a child. LeCroy began to suspect that Tinkerbell had been a witch and that her sexual abuse had placed him under some kind of spell, and that in turn this spell explained all the troubles and frustrations in his life. He reasoned that if Tiesler—by now synonymous with Tinkerbell—had placed some kind of sexual spell on him, she could also be made to undo that spell. As Doctor Hilton related it, "[LeCroy] started to develop a plan that he needed to reverse the roles on Tinkerbell and do to her what she had done to him." Hilton Report at 10.

Because the contents of Doctor Hilton's report came to play a crucial role both in the defense attorneys' thinking and in LeCroy's subsequent ineffective assistance claim, we reprint here an illustrative excerpt of the report:

Hilton Report at 10-11.

As LeCroy's defense team considered their options in light of Doctor Hilton's report, they were operating in the shadow of Rule 12.2(b) of the Federal Rules of Criminal Procedure, which requires a defendant to notify the Government in advance of trial if the defendant plans to introduce expert testimony on mental health issues at either the guilt or penalty phase of trial.5 Rule 12.2 further provides that if a defendant elects to introduce expert evidence bearing on a mental health issue, the defendant must make the results and reports of the defendant's expert available to the Government, and the defendant himself may be required to submit to an evaluation by a Government expert.6 See Fed.R.Crim.P. 12.2(c).

Having seen Doctor Hilton's report, the attorneys unanimously agreed they were, as Mendelsohn put it, "very scared of [a] government evaluation." Collateral Tr. Vol. II at 159. In short, Mendelsohn said, the defense team "made a decision not to have [LeCroy] evaluated and stuck with that all the way through." Id. at 172. Nonetheless, the attorneys recognized that some information in Doctor Hilton's report would be useful in mitigation: the evidence of childhood sexual abuse, specifically, might both arouse sympathy from the jury and offer a mitigating explanation of LeCroy's crime. The trick was finding a way to introduce the mitigating aspects of LeCroy's background without also opening the door to the wealth of aggravating information in Doctor Hilton's report.

The attorneys developed a plan to introduce Doctor David Lisak as a "teaching expert" on the relationship between childhood sexual abuse and criminality in men. Doctor Lisak, a clinical psychologist and Associate Professor of Psychology at the University of Massachusetts in Boston, was known to the attorneys as perhaps the nation's preeminent expert on childhood sexual abuse of males. Instead of evaluating LeCroy himself—and thereby triggering the reciprocity provisions of Rule 12.2—Doctor Lisak was to review a documentary record of LeCroy's life, drawn from his state prison records, his pre-sentence report from the 1995 federal conviction, his counseling records from prison, and Doctor Hilton's report. In front of the jury, Doctor Lisak would explain that individuals who experience childhood sexual abuse experience significant psychological problems later in life, and that these problems can be especially acute in men who are abused as children, leading to an elevated risk of later criminal behavior.

Doctor Lisak's testimony would thereby provide a backdrop for the jury, against which the defense lawyers hoped to introduce evidence of LeCroy's childhood abuse. Doctor Gary Ganahl, a psychological consultant for the Georgia Department of Corrections who evaluated LeCroy in prison following a suicide attempt, had been told by LeCroy about physical abuse at the hands of his father and sexual abuse by a female babysitter. Similarly. Doctor Marti Carlson—a clinical psychologist from the federal Bureau of Prisons—had also seen LeCroy when he was incarcerated and heard him report physical abuse by his father and sexual abuse by a female babysitter. Finally, the defense retained Jan Vogelsang, a clinical social worker, to conduct wide-ranging social and psychological evaluation of LeCroy and his family. Vogelsang would testify that aspects of LeCroy's childhood put him at special risk of criminality as an adult, including a history of mental illness in his family, a troubled relationship with his father, and a generally dysfunctional family environment.

By combining Doctor Lisak's "teaching" testimony with the factual testimony of Vogelsang and doctors Ganahl and Carlson, the defense team hoped to replicate the helpful aspects of Doctor Hilton's report—that is, the inference that LeCroy's crime was the product of an abusive and dysfunctional childhood, or as Kearns put it more bluntly: "That Mr. LeCroy was damaged goods because of what he suffered in his childhood and that that might explain why he would do something so awful," Collateral Tr. Vol. I at 96—without triggering Rule 12.2's requirement that LeCroy be evaluated by a Government expert. Kish explained that the strategy was to "have the teaching expert [i.e., Doctor Lisak] talk with the jury about what happens when a person has been the subject of childhood sexual abuse, and in that fashion avoid having to make Mr. LeCroy available for a government examination." Collateral Tr. Vol. I at 45-46. Mendelsohn explained that the lawyers themselves could then "draw inferences from what we got out of Dr. Lisak and the facts that we put into the record to then argue it to the jury." Collateral Tr. Vol. II at 159.

The defense team's teaching-expert strategy became the subject of intense pre-trial wrangling with the Government. On October 17, 2003, LeCroy gave notice pursuant to Federal Rule of Criminal Procedure 12.2(b) that he intended to introduce an expert regarding mental health. The Government moved for a psychiatric examination of LeCroy, and Magistrate Judge Susan S. Cole granted that motion on November 10, 2003. LeCroy appealed, and defense counsel met ex parte with District Judge Richard W. Story to contest the Government's right to an independent evaluation. Judge Story ordered a compromise: the Government would appoint a team of "firewalled" attorneys to address only the mental-health issues in LeCroy's trial, and these firewalled attorneys would not disclose any mental-health evidence to the primary prosecution team. Judge Story ordered the firewalled attorneys and the defense team to confer regarding the need for the Government to conduct an independent evaluation.

The firewalled attorneys ultimately argued that they did need an independent mental evaluation of LeCroy, and Judge Story agreed after a hearing on December 22, 2003, that the Government was entitled to such an evaluation under Rule 12.2. The defense complied with an order to give a list of mental-health experts they planned to use to the firewalled attorneys, but argued that there was no need for the Government to conduct a full examination of LeCroy because their own expert—Doctor Lisak—would be testifying only as a teaching expert, without himself having conducted an evaluation of LeCroy. The Government objected, and Judge Story agreed that because LeCroy planned to present expert evidence regarding childhood trauma, "thereby inviting the jury to make conclusions about how Defendant's childhood trauma may have impacted him and could impact him in the future," that the Government was entitled to its own evaluation by its own expert, Doctor Julie Medlin. That examination never happened, however, because LeCroy—following his lawyers' advice—invoked his Fifth Amendment right against self incrimination7 and refused to submit to the Government evaluation.

Judge Story nevertheless reserved a final ruling on whether Doctor Lisak would be allowed to testify as a teaching witness. Before trial, the arrangement was that Doctor Medlin would review the mental-health evidence otherwise available through discovery and prepare a written mental-health opinion under seal. At the close of the guilt phase of trial, the report would be released to defense counsel and to the firewalled prosecutors, at which point the defense team could make a decision about whether to introduce Doctor Lisak's testimony at sentencing, in which case the Medlin report would be released to the primary prosecution team.

C.

LeCroy's trial began on February 17, 2004. Faced with overwhelming evidence that LeCroy had killed Tiesler, the defense attorneys seized upon a botched-burglary defense at the guilt phase of the trial: they argued that LeCroy merely planned to rob Tiesler's home, but panicked when she caught him in the act and murdered her on impulse. As Kish put it to the jury in his guilt-phase closing argument, the whole episode was not a carjacking but "an irrational criminal act in the middle of a burglary."

The botched-burglary defense was jurisdictional: the case was only in federal court because the Government had charged LeCroy with carjacking, which in turn required the Government to prove that LeCroy killed Tiesler with the intent of taking her car. See 18 U.S.C. § 2119; see also Holloway v. United States, 526 U.S. 1, 12, 119 S.Ct. 966, 972, 143 L.Ed.2d 1 (1999) ("The intent requirement of § 2119 is satisfied when the Government proves that at the moment the defendant demanded or took control over the driver's automobile the defendant possessed the intent to seriously harm or kill the driver if necessary to steal the car.") (emphasis added); United States v. Applewhaite, 195 F.3d 679, 686 (3d. Cir.1999). If the defense could instead convince the jury that the crime had been a robbery gone bad—that LeCroy had panicked upon being discovered in Tiesler's house and impulsively killed her, and only later formed the intent to take her car as a means of escape—then the federal jurisdictional hook would be absent and LeCroy would evade a federal conviction.8 LeCroy's jurisdictional argument was unsuccessful, and the jury found LeCroy guilty on March 1, 2004, but the botched-burglary theory of the case would continue to play a role in the defense team's calculations as the case proceeded to sentencing.

At the sentencing phase of the trial, the Government's case-in-chief consisted of victim-impact evidence from Tiesler's family and friends, testimony relating to LeCroy's convictions in the early 1990s, and testimony regarding LeCroy's conduct while incarcerated. Of special relevance was testimony from two Lumpkin County law-enforcement officials who had witnessed LeCroy escape from his cell in the Lumpkin County Detention Center. Officer Aaron Welch, a Lumpkin County detention officer, testified that in 2003 LeCroy (and other inmates) were using a catwalk in the Detention Center's drop ceiling as a "virtual highway" between male and female cells. Officer Christopher Holman of the Lumpkin County Sheriff's Office testified that in one incident two female inmates were found under a bed in LeCroy's cell. In a separate incident, Holman inspected LeCroy's cell and found that LeCroy had created a hole in the shower wall that was large enough for a man to sneak through. In the crawlspace, they found a note written on the wall: "So well, have a great day explaining to the Marshals about me. Thanks for the food, ... smokes, and women, LeCroy. P.S. Had a great time here but sorry I won't miss you."

After the Government rested, LeCroy called a variety of mitigation witnesses to rebut the suggestion that LeCroy had been a threat to others in prison or that he had attempted to escape. Some of the witnesses had been incarcerated with LeCroy and testified that he had been pleasant and peaceful. A retired Bureau of Prisons official, Donald Romine, testified as an expert on the security of federal prisons and said that a person convicted of LeCroy's crimes would be held in a maximum security facility from which escape would be especially difficult.

The defense attorneys had, by this time, decided not to call Doctor Lisak as a teaching expert. They had been given a copy of Doctor Medlin's report at the conclusion of the guilt phase of trial, and concluded after reviewing the report that calling Doctor Lisak would only initiate an unhelpful battle of experts between Doctor Lisak and Doctor Medlin. Before presenting other mental health evidence, the defense team sought clarification from the Judge Story about what questions could be put to LeCroy's former psychiatrists9 about LeCroy's mental health. Judge Story ruled that the witnesses were permitted to testify about fact matters related to them in meetings with LeCroy, but could not testify about psychological testing, results, diagnoses, or opinions without triggering Doctor Medlin's testimony in rebuttal.

LeCroy subsequently called Doctor Gary Ganahl, who testified that he met with LeCroy in 1992 after LeCroy attempted suicide in state prison. Doctor Ganahl said that LeCroy related a troubled childhood, including sexual abuse by a female babysitter. Doctor Marti Carlson, who saw LeCroy in federal prison, also testified and also related that LeCroy had told her about having a troubled childhood and about being molested as a child by a female babysitter.

LeCroy also called a series of friends and family to testify about his good character. His mother, Donna Houston, took the stand, but given her emotional state, was unable to go forward with her testimony. She then said that she had informed Jan Vogelsang about LeCroy's early life and her marriage to LeCroy's father and had nothing to add. She did ask the jury for mercy.

Vogelsang testified at length about her interviews with LeCroy's friends and family, which she compiled as part of what she called a "biopsychosocial" assessment.10 Beginning with LeCroy's father, LeCroy, Sr., Vogelsang testified that he grew up in "a family that was without limits and without boundaries; and this was more in the area of gambling, the areas of money and the areas of the ready availability of weapons." Several family members demonstrated suicidal behavior and the children were poorly supervised. Within the extended family, "cousins continued to have sex with each other, to engage in sexual behaviors that sometimes were almost right in front of their parents with nothing being said." The male members of LeCroy, Sr.'s family were particularly dismissive of women, saying that "if it weren't for sex, women would have a bounty on their heads."

With his own family, LeCroy, Sr. was highly controlling and often abusive. He was particularly controlling of LeCroy's mother, frequently interrogating her about her whereabouts and the possibility that she was unfaithful to him. Their divorce was, Vogelsang said, particularly terrible: LeCroy, Sr. threatened to rape and kill LeCroy's mother and kill her coworker's. After an episode in which LeCroy, Sr. put his gun to Donna Houston's forehead, he gave his gun to LeCroy to keep him from killing her.

According to Vogelsang, LeCroy joined the Army to escape his fractured home life. Early in his military career, however, LeCroy broke his ankle, scuttling his dream of becoming a paratrooper. Stationed in Hawaii, he began drinking and taking drugs before going absent-without-leave, getting arrested, and taking a discharge. After returning home, LeCroy began a sexual relationship with his step-sister Alecia after his brother, Chad, began a sexual relationship with their other step-sister, Priscilla. Finally, Vogelsang listed a set of factors she believed made it more likely for someone of LeCroy's background to commit serious crimes.

On cross-examination, Vogelsang said her report covered LeCroy's background up to 1991 and did not include the time of Tiesler's murder. Vogelsang testified that LeCroy did well in school, that no member of LeCroy's immediate family remembered a female babysitter named Tinkerbell or knew that LeCroy had been molested, and that after his release from federal prison in 2001, LeCroy's mother and father opened their homes to him and tried to facilitate his reentry to society.

Kearns delivered the closing argument in mitigation. About sexual abuse, Kearns said:

Trial Tr. at 2715-16.

The jury returned a death sentence on March 10, 2004.11 LeCroy filed a motion for a new trial on March 17, 2004, and the District Court denied that motion on October 12 of the same year. On October 25, LeCroy appealed his conviction and sentence to this court, and we affirmed on March 2, 2006. See LeCroy, 441 F.3d at 931.

D.

On April 22, 2008, LeCroy petitioned the District Court to vacate his death sentence pursuant to § 2255, asserting—as relevant to this appeal—that he had been denied the effective assistance of counsel at trial.12 The claim LeCroy presented has two elements.

Strickland, 466 U.S. at 687, 104 S.Ct. at 2064.

The court held an evidentiary hearing on LeCroy's claims on January 11-13, 2010 and on February 8, 2010. LeCroy presented the testimony of his attorneys Kish, Kearns, and Mendelsohn, in addition to doctors Hilton, Carlson, Lisak, and Medlin. The court found that LeCroy had failed to carry his burden of proof on either of Strickland's elements, performance or prejudice, and accordingly denied him § 2255 relief. Regarding performance, the District Court credited testimony of LeCroy's attorneys that they had settled on the jurisdictional defense at the guilt phase of trial and on the teaching expert strategy at the sentencing phase, and that they had adapted the teaching expert strategy in response to District Court rulings during the course of the trial.

In particular, the District Court found that it was not ineffective to not call Doctor Hilton as a witness. Contrary to LeCroy's assertions, the court found that the attorneys were aware of the contents of Doctor Hilton's report and alert to both the mitigating and aggravating aspects of it. They had, the court found, made a reasonable strategic judgment that the risk of aggravation outweighed the possible benefits of mitigation, and this judgment was entitled to Strickland deference.

The same considerations made it reasonable to not have Doctor Lisak personally evaluate LeCroy for fear that such an evaluation would both expose Doctor Lisak to damaging cross-examination and also trigger an evaluation of LeCroy by a Government expert. Nor was it unreasonable for the defense team to ultimately forego Doctor Lisak's testimony—even as a teaching expert—once they saw Doctor Medlin's report and decided that they did not want to engage in a battle of experts in front of the jury. The District Court found that these decisions were reasonable strategic judgments made on the basis of a thorough investigation and consideration of the lawyers' options at trial.

Finally, the District Court concluded that Kearns was not ineffective in her closing argument. Though LeCroy might wish that she made the connection between his childhood abuse and his murder of Tiesler more vigorously or used different language, the court concluded that Kearns met minimum constitutional requirements by drawing the jury's attention to evidence that LeCroy had been abused and that this abuse played a role in his criminality as an adult.

Regarding prejudice, the District Court found that LeCroy had not been prejudiced by his attorneys choices regarding expert mental-health testimony because the testimony they elected not to offer was at least as aggravating as it was mitigating, and that therefore there was no reasonable probability that its introduction would have persuaded the jury to reach a different verdict. Indeed, it might have made the jury more certain that LeCroy deserved the death penalty.

Following the District Court's denial of his § 2255 petition, LeCroy lodged this appeal.

II.

In reviewing a district court denial of a § 2255 petition, we review the court's legal conclusions de novo and its factual findings for clear error. Devine v. United States, 520 F.3d 1286, 1287 (11th Cir.2008) (per curiam). "A claim of ineffective assistance of counsel is a mixed question of law and fact that we review de novo." Id. We give substantial deference to the factfinder on credibility determinations. Id. We may affirm on any ground supported by the record. Lucas v. W.W. Grainger, Inc., 257 F.3d 1249, 1256 (11th Cir.2001).

Claims of ineffective assistance of counsel require the petitioner to show both that his attorneys' performance was deficient and that their deficient performance prejudiced his defense. Strickland, 466 U.S. at 687, 104 S.Ct. at 2064. In evaluating performance, "counsel is strongly presumed to have rendered adequate assistance and made all significant decisions in the exercise of reasonable professional judgment." Id. at 690, 104 S.Ct. at 2066. It is petitioner's burden to "establish that counsel preformed outside the wide range of reasonable professional assistance" by making "errors so serious that [counsel] failed to function as the kind of counsel guaranteed by the Sixth Amendment." Butcher v. United States, 368 F.3d 1290, 1293 (11th Cir.2004) (citing Strickland, 466 U.S. at 687-89, 104 S.Ct. at 2064-65). Showing prejudice requires petitioner to establish "a reasonable probability that, but for counsel's unprofessional errors, the result of the proceeding would have been different." Strickland, 466 U.S. at 694, 104 S.Ct. at 2068. "The prejudice prong requires a petitioner to demonstrate that seriously deficient performance of his attorney prejudiced the defense." Butcher, 368 F.3d at 1293 (citing Strickland, 466 U.S. at 687, 104 S.Ct. at 2064).

Where, as here, the petitioner challenges the scope of his attorneys' investigation and the reasonableness of their strategic choices, a further word about attorney performance is warranted. Strickland makes plain that a reviewing court's objective "is not to grade counsel's performance." 466 U.S. at 697, 104 S.Ct. at 2069. We do not measure counsel against what we imagine some hypothetical "best" lawyer would do, in part to avoid "the distorting effects of hindsight" and in part to avoid judicial interference with "the constitutionally protected independence of counsel," lest we "restrict the wide latitude counsel must have in making tactical decisions." Id. at 689, 104 S.Ct. at 2065. We instead "reconstruct the circumstances of counsel's challenged conduct" and "evaluate the conduct from counsel's perspective at the time." Id.

Underpinning Strickland, then, is the assumption that "[t]here 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." Id. at 689-90, 104 S.Ct. at 2065-66. Crucially, Strickland permits attorneys to choose between viable avenues of defense, and attorneys are not ineffective for making a reasonable choice to take one avenue to the exclusion of another, or for selecting a reasonable course without considering some other, equally reasonable course. "If a defense lawyer pursued course A, it is immaterial that some other reasonable courses of defense (that the lawyer did not think of at all) existed and that the lawyer's pursuit of course A was not a deliberate choice between course A, course B, and so on. The lawyer's strategy was course A. And, our inquiry is limited to whether that strategy, that is, course A, might have been a reasonable one." Chandler v. United States, 218 F.3d 1305, 1315 at n.16 (11th Cir.2000). With these standards in mind, we turn to LeCroy's specific claims in this case.

III.

LeCroy makes three allegations of ineffective assistance of counsel on appeal: (1) that his attorneys failed to investigate mental health mitigation evidence and then present it during the sentencing phase of the trial; (2) that his attorneys failed to object to jury instructions on the issue of LeCroy's future dangerousness and escape risk; and (3) that his attorneys failed to request a jury instruction that the balancing of aggravating and mitigating factors be conducted according to the reasonable doubt standard. See Pet'r's Br. at v.13 Applying Strickland's deferential lens, we examine each claim in turn.

A.

LeCroy claims that his lawyers' performance was constitutionally deficient because they failed "to investigate and present mental health evidence" at the sentencing phase of his trial. Pet'r's Br. at 23. There are two variants of this claim: one is that the attorneys failed to meet with Doctor Hilton to discuss his report on LeCroy's mental health and subsequently failed to call him as a witness. Had they done so, the jury would have heard mitigating testimony about how LeCroy's mental health issues contributed to the crime. The second variant is that the attorneys failed to implement their preferred strategy of calling Doctor Lisak as a teaching expert, both because they unreasonably abandoned calling Doctor Lisak as a witness in the face of anticipated rebuttal testimony from Doctor Medlin and because lawyers Mendelsohn and Kearns miscommunicated during their closing arguments and failed to make the mitigating argument each expected the other to make.

1.

We do not agree with LeCroy that his attorneys performed ineffectively by not meeting in person with Doctor Hilton after reviewing his reports. Even LeCroy concedes that Doctor Hilton's written report was "extensive," Pet'r's Br. at 27, and it is plain that the defense attorneys reviewed and understood the contents of the report. See, e.g., Collateral Tr. Vol. II at 264 (Mendelsohn responding to questions about the potentially damaging contents of Doctor Hilton's report). The District Court was right to conclude that Doctor Hilton's report "was so detailed that a reasonable attorney did not need to speak with him to understand his diagnoses and the fact that he thought `magical thinking' was a cause of the event." LeCroy, 2012 WL 1114238, at *52. Nor does LeCroy identify any "red flags" in Doctor Hilton's reports that might have prompted a reasonable attorney to initiate a meeting or explore other avenues of investigation. Cf. Ferrell v. Hall, 640 F.3d 1199, 1233-34 (11th Cir.2011). LeCroy's argument seems to be, instead, that attorneys are per se ineffective if they commission and review an expert report without taking the additional step of meeting in person with the report's author. That is not what Strickland requires. LeCroy's attorneys made a thorough investigation of his background and employed an expert to evaluate the viability of a mental health defense, and were not required to sit down with Doctor Hilton to rehash conclusions that were already plain from the written report. See Bobby v. Van Hook, 558 U.S. 4, 11-12, 130 S.Ct. 13, 19, 175 L.Ed.2d 255 (2009) (explaining that Strickland does not require attorneys to "dig deeper" when substantial mitigation evidence is already in hand following a reasonably thorough investigation); Reed v. Sec'y, Florida Dep't of Corrs., 593 F.3d 1217, 1242-43 (11th Cir.2010) (same).

But even supposing LeCroy's attorneys performed ineffectively when they decided not to meet with Doctor Hilton in person, that decision did not prejudice LeCroy's defense. LeCroy's theory of prejudice is twofold: he argues first that by failing to meet with Doctor Hilton the defense team made strategic decisions on the basis of an incomplete understanding of Doctor Hilton's findings, see Strickland 466 U.S. at 690-91, 104 S.Ct. at 2066 ("[S]trategic choices made after less than complete investigation are reasonable precisely to the extent that reasonable professional judgments support the limitations on investigation."), and second that if the attorneys had met with Doctor Hilton they would have also called him as a witness at trial, thereby exposing the jury to powerful mitigation evidence concerning the role LeCroy's mental illnesses played in the crime. But the record does not support either theory of prejudice.

Although LeCroy writes vaguely that through an in-person meeting the lawyers could have "explored with [Doctor Hilton] the full ramifications of the diagnoses he found" and "appreciate the full context of his conclusions," Pet'r's Br. at 27-28,15 LeCroy never says what, exactly, was left to explore or appreciate that was not already in Doctor Hilton's report. LeCroy puts great emphasis on Doctor Hilton's testimony at the § 2255 hearing that "mental illness played a `direct role' in the crime and that the murder would not have `occurred but for' the mental illnesses [Hilton] had diagnosed." Pet'r's Br. at 27; see also Collateral Tr. Vol. II at 238. But one need not be a mental health expert to comprehend from the written report that Doctor Hilton attributed the crime to LeCroy's mental health problems. Indeed, the report plainly lays out Doctor Hilton's theory that LeCroy's "unusual and unexplainable encounter with Ms. Tiesler" in the woods "inflamed some of his unresolved issues relative to his sexual abuse" by Tinkerbell, and that "[h]is schizotypal tendencies of magical thinking and his previous experiences in witchcraft led him to believe that Ms. Tiesler was indeed his former babysitter (Tinkerbell), that witchcraft was involved and that a spell needed to be broken." Hilton Report at 17. Contrary to LeCroy's assertions, then, there was nothing more to be learned by meeting face-to-face with Doctor Hilton, and so the attorneys' failure to do so could not have prejudiced LeCroy's defense.

Nor was the defense prejudiced by the decision not to call Doctor Hilton to testify at trial, a decision LeCroy says is "at the core of the Defendant's claim, because the decision not to call [Hilton] was made without full knowledge of the consequences of this decision, that is without knowledge of the full extent of Dr. Hilton's testimony." Pet'r's Reply Br. at 13 n.1. It is important to note at the outset that the decision not to call Doctor Hilton as a witness was itself a strategic one deserving Strickland deference: indeed, the defense team never intended to use Doctor Hilton as a witness at trial. He was, Mendelsohn explained, a "test run" to "see what would happen": "[W]e were going to see what results he came back with and then with that information in hand, proceed with the rest of the case." Collateral Tr. Vol. III at 265.

A "test run" was strategically important because the defense team knew that offering the testimony of a mental health expert who had evaluated LeCroy would trigger the Government's reciprocal rights under Rule 12.2 to review the defense team's expert reports and to have LeCroy evaluated by an expert of the Government's choosing. Avoiding such an evaluation was a fundamental strategic goal of the defense team: as Kish explained, "[A]n evaluation by a government expert would have resulted in very harmful information. Real frankly, we didn't trust the government experts. And we felt that when the government would have such information, it would make it exceptionally difficult to raise the parts of [LeCroy's] background that we thought were worthy of presenting to the jury when they had to make their decision ... during the sentencing phase of the case." Collateral Tr. Vol. I at 45-46.

Contrary to LeCroy's assertion, then, the decision not to use Doctor Hilton as a witness at trial was imminently reasonable: it allowed the defense team to get a complete picture of LeCroy's background and mental health without having LeCroy evaluated by their preferred expert witness, Doctor Lisak. Doctor Lisak would in turn be able to testify as a "teaching witness" at trial—connecting childhood sexual abuse to criminal acts in general terms—without being subject to damaging cross-examination by the Government about LeCroy's first-hand account of the crime. This strategy had the added virtue of not undercutting the jurisdictional defense the attorneys planned for the guilt phase of trial: an expert who evaluated LeCroy would be cross-examined about the substantial planning and premeditation preceding Tiesler's murder, whereas a "teaching witness" would not.

Doctor Hilton's actual findings only underscore the reasonableness of the attorneys' decision not to use him as a witness. The attorneys reviewed Doctor Hilton's report and recognized correctly that the results were a mixed bag for the defense. "There were parts of Dr. Hilton's report that would have been very helpful," Kearns testified. "The problem was that there was the risk that the actual diagnosis would not be helpful," Collateral Tr. Vol. I. at 104-05, not to mention Hilton's vivid reconstruction of the crime and the report's baseline inconsistency with the botched-burglary defense.

The first problem was LeCroy's graphic and disturbing play-by-play review of the murder. An expert, like Doctor Hilton, who had evaluated LeCroy and learned the first-hand details of the offense would have been subject to lurid cross-examination about the brutality of the murder and the extent to which Tiesler suffered in the minutes before her death. At the § 2255 evidentiary hearing, the Government's attorney illustrated what this might have looked like during his cross-examination of Doctor Hilton:

Collateral Tr. Vol. II at 239-43 (emphasis added). We cannot fault LeCroy's defense attorneys for wanting to avoid this kind of testimony, even if LeCroy is right, as an analytical matter, that the crime scene and autopsy reports had already introduced the jury to many of the same details. LeCroy attempts to take this point a step forward and argue that the first-hand recollection of the crime was "no worse than the evidence that the defense was already facing," Pet'r's Reply Br. at 11, but that is at least debatable: certainly a reasonable attorney could believe that hearing this sort of narrative testimony, with its emphasis on the victim's awareness of her suffering, might inflame the jury in a way that a clinical coroner's report would not. Strickland deference is especially important in this setting because defense attorneys are in the courtroom and able to see the jury and make first-hand assessments of how jurors might react to testimony. Appellate judges, by contrast, review only a cold record months or sometimes years after the events at trial. LeCroy's attorneys made a tactical choice that the mitigating value of expert mental health testimony would be outweighed by the aggravating information the Government would emphasize, and on this record we cannot say their choice was unreasonable for Strickland purposes.

A second problem with Doctor Hilton's findings was that they wholly undercut the botched-burglary defense LeCroy had put on in the guilt phase of his trial. The defense had argued that LeCroy panicked when Tiesler interrupted him midway through the burglary of her home and killed her impulsively, in what they called "an irrational criminal act in the middle of a burglary." Testimony from an expert like Doctor Hilton would have shown the jury that none of that was true and that the defense lawyers had always known it to be untrue: LeCroy instead had sought Tiesler out specifically and laid in wait for her to come home. See Hilton Report at 10-11. This revelation might have seriously undercut the defense lawyers' credibility in the eyes of the jury for the remainder of trial.

Third and finally, the defense team considered Doctor Hilton's clinical diagnoses—that LeCroy suffered from borderline personality disorder and anti-social personality disorder—"not helpful," to use Kearns' term. Collateral Tr. Vol. I at 104-05. Doctor Hilton reported that LeCroy had "homicidal ideations toward a few of the other inmates" over what LeCroy called "trivial things." Hilton Report at 12. At the § 2255 hearing, Hilton testified on cross-examination that he believed LeCroy would represent a danger if he ever escaped custody, that his future dangerousness might extend to others incarcerated with him, and that—because LeCroy understands now that Tiesler was not Tinkerbell—he might mistake a different woman for Tinkerbell in the future and kill again. See Collateral Tr. Vol. II at 225-27.

Any of these considerations, standing alone, would give a good attorney pause. Cumulatively, they present an ironclad case for the reasonableness of the defense team's decision not to call Doctor Hilton as a witness. Mendelsohn's testimony summarized the attorneys' thinking before trial:

Collateral Tr. Vol. II at 264. We agree with the District Court that this strategic judgment was a reasonable one.

In any event, the decision not to call Doctor Hilton as a witness did not prejudice the defense because—as the District Court put it—"LeCroy's mental-illness evidence was the ultimate double-edged sword." LeCroy, 2012 WL 1114238 at *68. For one thing, none of the evidence went to LeCroy's competency or presented an affirmative defense; LeCroy understood that killing Tiesler was wrong, and he did it anyway. Collateral Tr. Vol. II at 264. Second, "the testimony that could have been presented was just as likely to have resulted in aggravation against rather than mitigation for" LeCroy. See Reed, 593 F.3d at 1238 (citation omitted). Doctor Hilton's testimony would have underscored all of the Government's aggravating factors: that LeCroy targeted Tiesler, that he laid in wait for her inside her home, and that he committed the murder in an especially heinous manner. Doctor Hilton found, moreover, that LeCroy could be a danger to others in the future, either to fellow inmates or to the public if he ever escaped—and LeCroy had tried to escape from prison before. Third, Doctor Hilton's testimony would have undercut the credibility of the defense lawyers in the eyes of the jury once it became clear that Doctor Hilton's conclusions were orthogonal to the botched-burglary defense. Considered in total, then, the District Court was correct in concluding that there was no reasonable probability that the jury would have reached a different sentence with Doctor Hilton's testimony before them. Indeed, Doctor Hilton's testimony might have made things worse. See LeCroy, 2012 WL 1114238 at *68-69.

2.

LeCroy's second argument is that—whatever the merits of the botched-burglary defense and teaching-witness gambit as strategic questions—his attorneys failed to actually implement their strategy at trial. Had the attorneys pulled off the teaching-expert plan, LeCroy allows that "there might, but only might, be some merit" to the District Court's conclusion that the attorneys were not ineffective. Pet'r's Reply Br. at 1. But LeCroy says that "simply did not happen." Id. Instead, LeCroy says his attorneys unreasonably decided not to call their teaching expert, Doctor Lisak, because of an irrational fear of the Government's rebuttal witness, and then compounded the error in their closing argument by failing to adequately connect LeCroy's sexual abuse to the crime. See id. at 1-4. On this account, LeCroy received at best only a fragmentary mental health defense at trial: "What was missing was any connection of the sexual abuse to Defendant's crime, which could only be provided by expert testimony, which counsel unreasonably abandoned." Id. at 4-5.

We do not agree that LeCroy's attorneys were incompetent in their execution of their trial strategy. For purposes of our discussion, we divide LeCroy's claim on this point into three alleged defects in his attorneys' representation: (1) that they ought to have had Doctor Lisak personally evaluate LeCroy; (2) that they ought to have called Doctor Lisak as a witness at trial; and (3) that they ought to have done a better job in their closing argument connecting LeCroy's childhood sexual abuse to Tiesler's murder. We conclude that none of these claims warrant § 2255 relief.

First, all the problems that would have attended calling Doctor Hilton as a witness at trial, discussed in Part III.A.1, supra, applied with equal force to any expert who evaluated LeCroy. As Kearns explained at the § 2255 hearing, "[I]f we had Dr. Lisak examine Mr. LeCroy and then intended to use Dr. Lisak as a witness, then we would have to make Bill LeCroy available to the government and their expert for an evaluation," Collateral Tr. Vol. I at 93, in addition to disclosing Doctor Lisak's findings to the Government. Id. at 113. Just as we do not think it was unreasonable not to call Doctor Hilton as a witness, we do not think it was unreasonable for the defense team to try and insulate their preferred teaching expert from damaging cross examination or protect their client from a hostile Government evaluation. These choices represented reasonable tactical decisions by competent attorneys.

The decision not to call Doctor Lisak, even as a teaching expert, was also a reasoned tactical decision on the defense team's part. At the close of the guilt phase and before the jury returned a verdict, the lawyers were provided a copy of Doctor Medlin's report, which she had prepared from examining LeCroy's documentary records from school, the military, and prison. In her report, Doctor Medlin rejected out of hand the causal link between childhood sexual abuse and violent criminality as an adult, and she further drew attention to LeCroy's inconsistent reporting of his abuse at the hands of Tinkerbell. Doctor Medlin suggested that some inmates fabricate tales of abuse to garner sympathy in the criminal justice system, and that such fabrication might explain why LeCroy sometimes reported being abused and sometimes did not. At the § 2255 hearing, Kearns explained "what was so threatening about Dr. Medlin's report. It was the complete apparent unwillingness to accept any impact of childhood sexual abuse on an adult man. I mean, it was more that she seemed—we seemed to have two experts that were polar opposites, one [i.e., Lisak] that said this is a very damaging, this trauma is very damaging and has very serious consequences, and then an expert who seemed to be unwilling to accept it." Collateral Tr. Vol. I at 125. Not wanting to engage in a battle of the experts, and still fearful that calling Doctor Lisak would trigger a Government evaluation,16 the defense elected not to call Doctor Lisak and instead try to connect the abuse to the crime on their own in their closing argument.

LeCroy contends that this decision was manifestly unreasonable because Doctor Medlin's report was of such poor quality that, had Doctor Lisak been called to testify, he could have easily dismantled a report he told the attorneys was "nonsense," a "hatchet job by somebody who really didn't know what she was talking about." See Pet'r's Reply Br. at 14 (quoting Collateral Tr. Vol. I at 166). But as Mendelsohn explained, the attorneys understood that they "weren't dealing with experts, we were dealing with lay people; and in the end it may just be a battle of the experts. No matter how incredible she may be on an objective expert level, you know, she could come off polished enough to a lay jury that we would end up having her be believable." Collateral Tr. Vol. II at 277-78. The District Court, having seen Doctor Medlin testify at the § 2255 hearing, expressly found that "the Defense's concerns were real. Even though Dr. Lisak and Dr. Hilton disagree with Dr. Medlin, her experience with sexual abuse perpetrators and citation of studies may well have convinced a lay jury that LeCroy only admitted sexual abuse in situations in which he wanted sympathy or benefits." LeCroy, 2012 WL 1114238 at *55. We are especially hesitant to disturb this conclusion insofar as we give special deference to the factfinder on questions of credibility. See Devine, 520 F.3d at 1287.

LeCroy also argues—with respect to both Doctor Hilton and Doctor Lisak— that his attorneys were operating under an irrational fear of a Government evaluation. LeCroy claims that his attorneys could have agreed to a Government evaluation "and, if it turned out badly, still attempted the `teaching expert' ploy." Pet'r's Br. at 31. But without knowing what a Government evaluation would show, LeCroy claims his attorneys could not have made an informed strategic decision because they were operating "without knowledge of what the real stakes were." Id.

This argument misses the mark. First, it was not unreasonable for the defense team to infer, based on Doctor Hilton's evaluation, that an evaluation by a Government expert would come to similar—or worse—conclusions. Kearns testified that she was "very fearful" of a Government evaluation "based on [her], at that point in time, 30 years of experience doing criminal defense work," during which her experiences with Government psychiatric evaluations had "not been good." Collateral Tr. Vol. I at 94, 108, 110. Or, as Mendelsohn put it, "I don't have a lot of faith that the government experts would be coming out trying to help Mr. LeCroy." Collateral Tr. Vol. II at 269. LeCroy certainly offers no reason to think that the lawyers' judgment was unreasonable.

Second, and perhaps more importantly, LeCroy's argument here inverts the burden of proof, which on a § 2255 petition belongs to the petitioner. If LeCroy's claim is that a Government evaluation would have been less damaging than Doctor Hilton's evaluation—and that, accordingly, the defense team ought to have been more willing to roll the dice and see what the Government would come up with— then to carry that argument LeCroy would actually need to show that the Government evaluation would be favorable. Otherwise, LeCroy is asking us to disregard the burden of proof and speculate about what might have been, drawing an inference in his favor that the record simply does not support. Here, where LeCroy is either unwilling or unable to demonstrate that the Government's evaluation would in fact have been favorable—as opposed to conceivably being favorable—he has failed to carry his burden in showing prejudice.

Finally, LeCroy takes issue with his attorneys' closing argument at the sentencing phase, claiming they never adequately connected his childhood abuse to the crime. This Court explained in Lawhorn v. Allen, 519 F.3d 1272 (11th Cir. 2008), that deficient performance in the closing-argument context can be shown by "an attorney's failure to use the closing argument to focus the jury's attention on his client's character or any mitigating factors of the offender's circumstances, and by his failure to ask the jury to spare his client's life." Id. at 1295. Here, Kearns indisputably asked the jury to show mercy and spare LeCroy's life, so LeCroy's claim is that Kearns failed to "drive the point home regarding the effect of sexual and other abuse suffered by the Defendant," Pet'r's Br. at 29-30, and that the link she drew in her closing argument—that LeCroy's self-esteem suffered as a consequence of Tinkerbell's abuse, and that this contributed to Tiesler's murder—was "hardly a compelling case for a sentence other than death." Id. at 30. But as the District Court correctly noted, LeCroy's point here is not that Kearns failed to draw the jury's attention to the available mitigating evidence. Plainly, she did that. His claim is, at best, that he wishes she had done it better. That does not give rise to an ineffectiveness claim of the sort we found in Lawhorn. In that case, we said an attorney had performed deficiently on closing where the attorney waived his closing argument entirely "based on a complete misunderstanding of a clear rule of law." 519 F.3d at 1295-96. By contrast, Kearns in her closing argument reminded the jury that LeCroy had been abused and explained how it affected him as an adult. She might have done more, but Strickland requires competence, not perfection.

B.

LeCroy's second category of alleged ineffectiveness is his attorneys' failure to object to the District Court's instruction at the conclusion of the sentencing phase that the jury could consider the nonstatutory aggravating factor of future dangerousness to the public if the jury found beyond a reasonable doubt that defendant posed a "risk" of escape, as opposed to finding a "likelihood" of escape. See Pet'r's Br. at 42. The word "risk" is, LeCroy says, "so elastic and ill-defined that it could include the mere possibility of an escape, no matter how fanciful." Id. at 43. By failing to object to this instruction, the attorneys allegedly failed to perfect a meritorious legal issue on appeal.

We disagree. First, LeCroy's attorneys were not ineffective in failing to object to the court's wording because the instruction did not misstate the law. We have said that district courts have "broad discretion in formulating a jury charge so long as the charge as a whole accurately reflects the law and facts," United States v. Turner, 871 F.2d 1574, 1578 (11th Cir.1989) (citations omitted), and the jury instruction at issue here was not an inaccurate statement of law. Rather than inviting the jury to entertain a "fanciful" possibility of escape, as LeCroy suggests, Pet'r's Br. at 43, the court instructed the jury that they could only consider LeCroy's future dangerousness "if each of you finds beyond a reasonable doubt that Mr. LeCroy does pose a risk of escape."

When we considered this instruction on direct appeal in LeCroy's case, we said:

LeCroy, 441 F.3d at 931. LeCroy is right that, because trial counsel did not object to this instruction, we conducted only plain error review. But he offers no reason to suggest that the analysis would change under an abuse-of-discretion standard, and indeed cites no authority at all for the proposition that "likelihood" was the magic, mandatory word.17 Our reasoning in LeCroy's direct appeal also suggests that even if counsel had objected to the instruction, we would not have reversed his conviction given that "the jury found every other aggravating factor alleged by the government." Id. LeCroy, therefore, cannot show prejudice from his attorneys' failure to object.

C.

LeCroy's final claim is that his attorneys were ineffective for failing to request an instruction that the jury, in conducting its balancing of aggravating and mitigating factors, was required to find beyond a reasonable doubt that the aggravating factors outweighed the mitigating ones. LeCroy's theory is that, under cases like Ring v. Arizona, 536 U.S. 584, 122 S.Ct. 2428, 153 L.Ed.2d 556 (2002) and Apprendi v. New Jersey, 530 U.S. 466, 120 S.Ct. 2348, 147 L.Ed.2d 435 (2000), any fact essential to the imposition of a particular sentence must be found beyond a reasonable doubt, and that the ultimate balance between aggravating and mitigating sentencing factors is effectively a factual question. Pet'r's Br. at 44-45. A competent attorney would, LeCroy says, have asked for such a jury instruction.

The problem with this argument is that it invites us to evaluate counsel's conduct retrospectively, rather than at the time of trial. Contra Strickland, 466 U.S. at 689, 104 S.Ct. at 2065. At the time of LeCroy's trial in 2004, no court had found that the jury had to be instructed that it conduct its balancing inquiry against a reasonable doubt standard. To be sure, subsequent cases advanced the possibility that Apprendi and the like applied to a jury's balancing deliberations, see, e.g., United States v. Gabrion, 648 F.3d 307, 325-29 (6th Cir.2011), aff'd en banc, 719 F.3d 511 (6th Cir.2013), but LeCroy points to no authority that would have given LeCroy's attorneys notice at the time of trial that such an instruction was even arguably required.18 Even supposing LeCroy's argument had merit, then, he directs it to the wrong target. Strickland is concerned with prevailing professional norms at the time of trial, not with subsequent developments in the law.

IV.

For the foregoing reasons, the judgment of the District Court is AFFIRMED.