MARCUS, Circuit Judge:

Petitioner Christopher Brooks was convicted of murdering a female acquaintance in the course of a rape, robbery, and burglary and was sentenced to death in Alabama state court. The issue presented by this appeal is whether Brooks's direct appellate counsel rendered ineffective assistance by inadequately litigating his trial counsel's ineffective assistance at the penalty phase of his trial. Trial counsel presented only one witness — Brooks's mother — in mitigation. Brooks says that trial counsel should have sought out and presented more mitigation evidence: in particular, witnesses testifying to his good character, evidence of his moderate alcoholism, and the fact that he was intoxicated on the night of the crime. He faults his direct appellate counsel for two alleged deficiencies. First, his direct appellate counsel raised the ineffectiveness-of-trial-counsel claim with regard to the failure to introduce good-character evidence, but she did not seek out and failed to present any character witnesses herself. This meant that Brooks was unable to show that his trial counsel's decision not to present character witnesses prejudiced him at trial. Second, his direct appellate counsel did not even raise the claim that his trial counsel should have discovered and presented evidence of his alcoholism and intoxication on the night of the murder.

The district court denied Brooks's 28 U.S.C. § 2254 petition, finding that the Alabama Court of Criminal Appeals reasonably applied Strickland v. Washington, 466 U.S. 668, 104 S.Ct. 2052, 80 L.Ed.2d 674 (1984), in rejecting this claim. The Court of Criminal Appeals held, in relevant part, that Brooks's trial counsel's failings did not prejudice him during the penalty phase because there was no reasonable probability that the balance of aggravating and mitigating factors, even including all of the mitigating evidence Brooks now says should have been presented, would have weighed in favor of life imprisonment rather than death. Consequently, his appellate counsel could not have prejudiced Brooks during his direct appeal. After thorough review, we conclude that this determination was reasonable and, therefore, affirm.

I.

A.

In 1993, Brooks was convicted of murder during the course of a rape, robbery, and burglary, see Ala.Code § 13A-5-40(a)(2)-(4), for killing Jo Campbell. The Alabama Court of Criminal Appeals summarized the underlying facts of the case:

Brooks v. State, 695 So.2d 176, 178-79 (Ala.Crim.App.1996) ("Brooks I") (footnote omitted). "The record also reflects that after searching Brooks's apartment, the police recovered the keys to the victim's automobile; pawn tickets; the victim's AT & T answering machine; and receipts from purchases that had been made with the victim's Shell Oil Company credit card." Ex parte Brooks, 695 So.2d 184, 187 (Ala.1997) ("Brooks II").

At trial, the state also established several other facts. Another man, Robert Leeper, a friend of Brooks who had traveled with him to Birmingham, was with Brooks and the victim at the time. The police officer who found Campbell's body testified that he found both a portion of a baseball bat and a barbell with one end broken off in the apartment. Dr. Gary Simmons, the medical examiner who had performed the victim's autopsy, also testified. He said that the cause of death was blunt force trauma to the head and neck. In addition, his examination revealed an abrasion on the victim's vagina. The prosecution's theory of the case was that Brooks, not Leeper, committed the murder, since the evidence demonstrated that he had intercourse with Campbell and his bloody fingerprint was found on the doorknob. The jury convicted Brooks of three counts of capital murder for killing the victim during the course of a rape, during the course of a robbery, and during the course of a burglary. Brooks v. State, 929 So.2d 491, 494 (Ala.Crim.App.2005) ("Brooks III").

At the penalty phase of the trial, the state put on several witnesses to establish the aggravating circumstance that the murder was especially heinous, atrocious, or cruel. See Ala.Code § 13A-5-49(8). The main witness was Simmons, who returned to the stand to describe the severe injuries to Campbell's body. In addition, the state admitted into evidence photographs of those injuries. Campbell had several lacerations and bruises to her face, extensive fractures at the base of her skull, evidence of internal bleeding into her lungs, contusions to the brain, hemorrhaging in the eye, injuries to the neck either from blunt force trauma or strangulation, a broken nose, a fractured jawbone, and several missing teeth, some of which had lodged in her windpipe. In Simmons's opinion, the injuries resulted from being struck five to eight times with a heavy blunt object. Simmons also testified that the hemorrhaging and brain swelling indicated that the victim was alive throughout at least part of the attack, and that she could have lived for anywhere from five minutes up to several hours after the attack, although he could not know whether or for how long she remained conscious.

Brooks's trial counsel put on only one witness in mitigation, Brooks's mother, Norma Silva. Her testimony consisted mostly of basic, biographical information, including the fact that Brooks was twenty at the time of the murder, that Brooks's father had no role in raising him, and that he had only one prior criminal conviction, a petty larceny misdemeanor.1

In closing argument, defense counsel argued that there was enough residual doubt so that, even though the jury had convicted Brooks, it should not recommend sentencing him to death. As counsel put it, "[W]e can't come back in a month or six years, ten years, when you read about an electrocution in the paper and say, `I lie awake at night and I wonder what happened.' It will be too late." Counsel repeatedly stressed that no one could definitively know what had occurred on the night of the murder. Finally, he emphasized that Brooks had no serious criminal history, which was a mitigating circumstance that favored sentencing Brooks to life imprisonment.

The jury recommended the death sentence by a vote of eleven to one. Brooks III, 929 So.2d at 494. The trial court found two aggravating circumstances applied: first, the commission of the murder occurred during the course of a rape, robbery, burglary, or kidnapping, Ala.Code § 13A-5-49(4); and, second, the crime was especially heinous, atrocious, or cruel compared to other capital offenses, Ala.Code § 13A-5-49(8). As for mitigating circumstances, the trial court found, based in part on Silva's testimony, that: Brooks had no significant history of prior criminal activity, Ala.Code § 13A-5-51(1); and he was only twenty at the time of the crime, Ala. Code § 13A-5-51(7). The court found no nonstatutory mitigating circumstances. The trial court weighed the aggravating and mitigating circumstances, found that the former outweighed the latter, and sentenced Brooks to death.

B.

A new lawyer — Virginia Vinson — was appointed to replace trial counsel. Vinson is the attorney whose alleged ineffectiveness is the subject of the current appeal. Vinson moved for a new trial, in part based on trial counsel's ineffectiveness at adequately investigating and presenting mitigating character evidence. She did not, however, argue that trial counsel was ineffective for failing to present evidence regarding Brooks's intoxication and alcoholism.

The trial court conducted an evidentiary hearing at which Brooks, his trial counsel, Kenneth Gomany and James Scott Boudreaux, and the prosecutor, Roger Brown, testified. Both Gomany and Boudreaux were experienced lawyers who had tried capital cases before Brooks's trial. Gomany and Boudreaux testified that they did not hire an investigator or a psychologist to help develop mitigation evidence but instead conducted their own investigation. Based on his interactions with Brooks, Gomany saw no signs that Brooks suffered from psychological or psychiatric conditions, and neither attorney uncovered any significant mitigating circumstances. Brooks had pointed out possible character witnesses during trial, but Boudreaux believed that "interject[ing] character as an issue ... would make the proceedings ludicrous," since the horrific nature of the crime could not be countered by character witnesses, and bringing in character evidence would have opened the door to cross-examination on Brooks's prior criminal history.

Boudreaux also explained the decision not to present character witnesses as part of an informal deal struck after the guilt phase of the trial. The prosecutor agreed to forego putting on "some very damaging potential testimony by the victim's family" as long as Brooks's trial counsel did not introduce evidence of his good character. Boudreaux believed that victim-impact testimony from Campbell's family, who were present "in mass during the trial, crying, sniffling, filling the courtroom with themselves and their supporters," would have "drive[n] even more nails into the coffin in this case." Brown, who also testified, confirmed that he agreed that he wouldn't "go into any of the what a wonderful person Ms. Campbell was unless the defense puts somebody on the witness stand to sob and talk about how wonderful the defendant is."

Based on this testimony, the trial court found that Gomany and Boudreaux were not ineffective. The Alabama Court of Criminal Appeals affirmed Brooks's conviction and sentence. Brooks I, 695 So.2d at 184. As for trial counsel's alleged ineffectiveness, the Court of Criminal Appeals held that Brooks failed to satisfy either prong of Strickland v. Washington's test, in particular noting that "the appellant did not inform the trial court at the hearing on his motion for new trial of who these witnesses were and what they would testify to." Brooks I, 695 So.2d at 181.

The Alabama Supreme Court affirmed the Court of Criminal Appeals's decision, rejecting the ineffectiveness-of-trial-counsel claim on essentially the same rationale. See Brooks II, 695 So.2d at 192. The United States Supreme Court subsequently denied Brooks's petition for certiorari. Brooks v. Alabama, 522 U.S. 893, 118 S.Ct. 233, 139 L.Ed.2d 164 (1997).

C.

Brooks obtained new counsel for his Rule 32 petition, which he filed in September 1998. His petition raised several claims that his direct appellate counsel (Vinson) had ineffectively litigated his direct appeal. Two of those claims are the subject of this appeal. First, Vinson failed to adequately investigate and present the claim that trial counsel was ineffective during the penalty phase. Vinson raised trial counsel's ineffectiveness based on the failure to put on character witnesses, but she did not seek out and present character witnesses herself. Second, Vinson failed to even raise the claim that trial counsel was ineffective for not offering Brooks's alcoholism and intoxication during the murder as a mitigating circumstance.

1.

As part of his Rule 32 petition, Brooks presented testimony from several witnesses whom his trial counsel could have called during the penalty phase of his trial. Roy Harbor, whose daughter had dated Brooks in the summer of 1992, testified that Brooks was "very well mannered, ... mature, and very outgoing, a nice kid." He had never known Brooks to be violent. Shirley Harbor, Roy Harbor's wife, was also deposed and said that Brooks was "very polite," never misbehaved or acted violently, and treated her daughter "very well." The other four witnesses — Rosemary Brooks, the defendant's aunt; Ronald Brooks, the defendant's uncle; Norma Silva, Brooks's mother; and Edgar Silva, Brooks's stepfather — all had the same impression of Brooks's character: in his youth, Brooks had been nice, mild-mannered, polite, and nonviolent.

Edgar Silva was the only one to express a concern regarding Brooks's character: "[T]he only thing I could really think that I didn't like about him, he liked to drink sometimes." Nonetheless, he also remembered Brooks as nonviolent and "even-tempered" as a child, not becoming provoked even when another child pushed or kicked him.

2.

The same court and judge that presided over Brooks's trial also presided over his Rule 32 proceedings. At two evidentiary hearings, Brooks put on an expert witness, Dr. William Beidleman, to testify to his alcoholism and the fact that he was intoxicated on the night of the crime. Based on interviews with Brooks, a number of tests, and his review of documentary materials, including Brooks's past arrests for alcohol-related offenses, Beidleman opined that Brooks exhibited symptoms of alcoholism. Brooks scored 12 on the Michigan Alcoholism Screening Test, which classified him as a moderate alcoholic. In conversations with Beidleman, Brooks had recounted that he began drinking heavily in the sixth or seventh grade; that his drinking increased to the point where he drank every day by the age of eighteen; and that he had experienced three to four alcohol-related blackouts by that time.

As for the events of December 30, 1992, Beidleman had reviewed additional evidence, including police records of their interview with Leeper and a statement taken from Angela Turman, an acquaintance of Brooks and Leeper. Leeper said that the two men had gone to Birmingham, that he had gotten drunk, and that he and Brooks returned to Campbell's apartment, where Leeper fell asleep and did not awake or hear anything during the night. Leeper never stated expressly that Brooks was drunk, although he did say that he and Brooks drank a case of Michelob beer. Turman told police that Brooks had told her that he and Leeper both had gotten drunk, returned to the apartment, and passed out. In his interview with Beidleman, Brooks claimed that he had consensual sex with Campbell on the night of her murder. There is no indication that Brooks expressly told Beidleman that he was intoxicated on the night of the offense. Nevertheless, Beidleman opined that, based on all the evidence, Brooks "was indeed acutely intoxicated" at the time of the murder. On cross-examination, Beidleman admitted that much of his evidence came from Brooks's own statements or those of individuals who received their information from Brooks.

The state rebutted Beidleman's testimony with its own expert, Dr. Glenn King. King explained that Campbell's murderer had engaged in "goal-directed behaviors" — he attempted to wipe down part of the apartment, concealed the body, and stole things from the victim. He conceded that Brooks could have been intoxicated, but opined that Brooks remained able to form the intent to kill. King also testified that the state had administered the Wechsler Adult Intelligence Scale to Brooks, and that Brooks was of average intelligence, almost exactly at the 50th percentile.

3.

The Alabama trial court denied Brooks's Rule 32 petition. It rejected his ineffective-assistance-of-direct-appellate-counsel claims on Strickland's performance prong because Brooks had failed to demonstrate trial counsel's ineffectiveness during the penalty phase. Alternatively, the court also held that Brooks had failed to satisfy Strickland's prejudice prong.

On appeal, the Alabama Court of Criminal Appeals affirmed, beginning with its finding that trial counsel performed competently. Brooks III, 929 So.2d at 511. Moreover, on the issue of whether trial counsel prejudiced Brooks, the Court of Criminal Appeals concluded:

Id. at 512 (citation omitted). Although the court did not expressly extend this analysis to direct appellate counsel, the implicit conclusion must have been that, if trial counsel performed competently and did not prejudice Brooks, then there was no way that appellate counsel's failings in litigating the ineffectiveness-of-trial-counsel claim could have prejudiced Brooks during his direct appeal. Thus, the Court of Criminal Appeals denied Brooks's Strickland claim.

Since the Alabama Supreme Court denied certiorari review of this case, Brooks III is the controlling state-court decision that we review pursuant to 28 U.S.C. § 2254. See Newland v. Hall, 527 F.3d 1162, 1199 (11th Cir.2008).

D.

Brooks petitioned for a writ of habeas corpus from the United States District Court for the Northern District of Alabama in 2005. The district court denied Brooks's § 2254 petition. As for direct appellate counsel's failure to claim that trial counsel was ineffective in not offering intoxication as a mitigating circumstance, the district court determined that the state courts' rejection of this claim was reasonable since competent trial counsel could have differed regarding whether to present evidence of Brooks's intoxication.

As for direct appellate counsel's failure to develop additional good-character mitigating evidence, the district court found the state court's decision was "not unreasonable" because there was no evidence that the failure to present character witnesses resulted in prejudice to Brooks during the penalty phase of his trial. We subsequently granted Brooks a certificate of appealability ("COA") "only with respect to Petitioner's claim that his trial counsel were ineffective in developing and presenting mitigating evidence during the penalty phase of his trial and that his appellate counsel were ineffective in presenting this issue on appeal."

II.

A.

We review de novo the district court's denial of habeas relief pursuant to 28 U.S.C. § 2254. Wellons v. Warden, 695 F.3d 1202, 1206 (11th Cir.2012). The district court's legal conclusions, and its resolution of mixed questions of law and fact, also receive de novo review. Id. Since Brooks's habeas petition was filed after April 24, 1996, the effective date of the Antiterrorism and Effective Death Penalty Act of 1996 ("AEDPA"), Pub.L. No. 104-132, 110 Stat. 1214, the following provision governs his challenge:

28 U.S.C. § 2254(d).

"Under AEDPA, our review of a final state habeas decision is greatly circumscribed and is highly deferential to the state courts." Hill v. Humphrey, 662 F.3d 1335, 1343 (11th Cir.2011) (en banc) (internal quotation marks omitted). The application of § 2254(d)(1)'s language is well-settled:

Williams v. Taylor, 529 U.S. 362, 412-13, 120 S.Ct. 1495, 146 L.Ed.2d 389 (2000).

"Claims of ineffective assistance of appellate counsel are governed by the same standards applied to trial counsel under Strickland." Philmore v. McNeil, 575 F.3d 1251, 1264 (11th Cir.2009) (per curiam) (citing Heath v. Jones, 941 F.2d 1126, 1130 (11th Cir.1991)); see also Smith v. Robbins, 528 U.S. 259, 285, 120 S.Ct. 746, 145 L.Ed.2d 756 (2000). Under the first prong, Brooks must show that his direct appellate counsel's performance "fell below an objective standard of reasonableness." Strickland, 466 U.S. at 688, 104 S.Ct. 2052. Under Strickland's second prong, Brooks must show that there "is a reasonable probability that, but for counsel's unprofessional errors, the result of the proceeding would have been different." Id. at 694, 104 S.Ct. 2052. "A reasonable probability is a probability sufficient to undermine confidence in the outcome." Id. This inquiry requires us to consider "the merits of the omitted claim," Philmore, 575 F.3d at 1264-65, and we will find counsel's performance prejudicial if "the neglected claim would have a reasonable probability of success on appeal," Heath, 941 F.2d at 1132. When, as here, the claim concerns the failure to present mitigating evidence, "we reweigh the evidence in aggravation against the totality of available mitigating evidence," Wiggins v. Smith, 539 U.S. 510, 534, 123 S.Ct. 2527, 156 L.Ed.2d 471 (2003), to determine whether the petitioner suffered prejudice.

Both prongs of the Strickland inquiry present mixed questions of law and fact and therefore receive de novo review upon appeal. Cade v. Haley, 222 F.3d 1298, 1302 (11th Cir.2000). However, our review of Brooks's Strickland claims is also subject to AEDPA deference. Thus, we do not merely evaluate whether, under Strickland's prejudice prong, there was a reasonable probability that the result of Brooks's proceeding would have been different. To grant relief, we must conclude that no reasonable jurist could decide otherwise. If fairminded jurists could disagree about the correctness of the state court's decision, then the state court's application of Strickland was not unreasonable, and AEDPA precludes the grant of habeas relief. Evans v. Sec'y, Dep't of Corrs., 703 F.3d 1316, 1326-27 (11th Cir. 2013) (en banc); cf. Harrington v. Richter, ___ U.S. ___, 131 S.Ct. 770, 786, 178 L.Ed.2d 624 (2011).

There is no dispute that the Alabama courts correctly identified the relevant Supreme Court law: Strickland v. Washington. See Brooks III, 929 So.2d at 497. Therefore, Brooks must show that the Alabama Court of Criminal Appeals's decision was an unreasonable application of Strickland. This he cannot do.

B.

The essence of Brooks's claim is that his direct appellate counsel ineffectively presented or neglected to present two winning ineffective-assistance-of-trial-counsel arguments during his direct appeal: first, that his trial counsel did not prepare and put on character witnesses as mitigating evidence; and, second, that his trial counsel did not discover and put on mitigating evidence that Brooks was an alcoholic who was intoxicated on the night of the offense.

We begin and end our analysis with Strickland's prejudice prong. Because we must reweigh the totality of the aggravating and mitigating evidence to determine whether Brooks suffered prejudice, Wiggins, 539 U.S. at 534, 123 S.Ct. 2527, we evaluate his two ineffectiveness claims — concerning good-character evidence and intoxication evidence — together. Since Brooks has failed to establish prejudice, we need not, and do not address direct appellate counsel's performance. See Strickland, 466 U.S. at 697, 104 S.Ct. 2052; Holladay v. Haley, 209 F.3d 1243, 1248 (11th Cir.2000) ("[T]he court need not address the performance prong if the defendant cannot meet the prejudice prong, or vice versa." (citation omitted)).

In the Rule 32 collateral appeal, the Alabama Court of Criminal Appeals concluded that, even if all of Brooks's additional mitigating evidence had been presented during the penalty phase, there was not a reasonable probability that it would have altered the outcome of Brooks's trial. See Brooks III, 929 So.2d at 512. In other words, the state court determined that Brooks's trial counsel's performance did not prejudice him. If Brooks's trial counsel did not prejudice him, then it necessarily followed that he suffered no prejudice to his direct appeal, since he never could have succeeded in establishing trial counsel's ineffectiveness. Since the state court's resolution of this claim hinged on whether Brooks suffered prejudice at the penalty phase, the question we must answer under AEDPA's deferential standard is this: could a reasonable jurist conclude that there was not a reasonable probability that the totality of Brooks's mitigating evidence would have altered the outcome of the penalty phase? The answer to this question is yes, and, therefore, the Alabama Court of Criminal Appeals's determination that he suffered no prejudice was reasonable.

The totality of the new mitigating evidence Brooks presented in his Rule 32 proceeding consisted of the following. Regarding Brooks's good character, six witnesses testified to their impression that, as a child and as a young man, Brooks was nice, polite, and nonviolent. Four of Brooks's family members — his mother, his stepfather, his aunt, and his uncle — said that, during Brooks's childhood, he had been nice, was a good athlete and a people person, and did not get into fights. An ex-girlfriend's parents also remembered Brooks as "very well mannered, ... mature, and very outgoing, a nice kid," and said that he had treated their daughter "very well" and was never violent. As for the evidence of alcoholism and intoxication, Brooks presented an expert witness, Dr. Beidleman, who testified that Brooks was a moderate alcoholic and that, in Beidleman's opinion, Brooks had been "acutely intoxicated" at the time of the murder.

A reasonable jurist could conclude that evidence that Brooks had been a nice, polite, and nonviolent person would not sway a jury from its recommendation of death, especially in light of the aggravating circumstances in the case that directly and powerfully contradicted any generic impression of Brooks's good nature. The trial court found two aggravating circumstances: first, the commission of the murder occurred during the course of a rape, robbery, burglary, or kidnapping; and, second, the crime was especially heinous, atrocious, or cruel compared to other capital offenses. The evidence adduced in the guilt phase, and the jury's verdict in finding Brooks guilty on all three counts of capital murder, clearly established the first aggravating circumstance. See Brooks III, 929 So.2d at 494.

The state's presentation of medical and blood spatter experts during the penalty-phase proceeding established the second circumstance in gruesome detail. The state admitted into evidence photographs of the victim's injuries, which Dr. Simmons, the medical examiner, explained at length. The evidence indicated that the victim was killed by five to eight blows from a heavy object, which resulted in broken bones or fractures to her skull, jawbone, and nose, internal bleeding in her lungs, brain contusions, and neck injuries either from blunt force trauma or strangulation, among other injuries. The hemorrhaging indicated that the victim was alive during part of the attack, and, based on the swelling of the brain, Simmons believed that the victim could have lived for anywhere from five minutes up to several hours after the attack.

In light of the extensive evidence regarding the horrific nature of this crime, it was reasonable for the Alabama Court of Criminal Appeals to conclude that the penalty-phase outcome would not be affected by Brooks's acquaintances' and relatives' impression of him as a nice and polite young man — an impression that the evidence of this murder squarely contradicted. Indeed, a reasonable jurist could conclude that the horrific facts of the murder would have completely eviscerated the effect of Brooks's character witnesses. In a similar case, Dill v. Allen, a petitioner claimed that his trial counsel was ineffective in failing to present character witnesses — mostly family members — who would have testified to his nonviolent nature, his care for his family, and his good nature. See 488 F.3d 1344, 1360 (11th Cir.2007). Yet, faced with evidence that merely showed "petitioner's good character and stability," the panel in Dill could not find that the state court unreasonably determined that Dill had not suffered prejudice. Id. at 1362. The panel characterized his mitigating evidence as "weak" and was "unpersuaded by his argument that such testimony would have surmounted the three aggravating factors found by the trial court." Id. at 1363 (footnote omitted).

Indeed, this case is more clear-cut than Dill: had Brooks's attorneys introduced additional mitigating evidence, particularly in an attempt to establish his good character, then the prosecution would have come forward with victim-impact evidence that would have further added to the state's already formidable case. The effect of this additional aggravating evidence could have been damaging to Brooks; at the very least, the state court reasonably could have so found. In weighing the totality of the aggravating circumstances against the totality of the mitigating circumstances, then, we recognize that Brooks's good-character evidence would not have been all-upside and no-downside. The presentation of Brooks's character witnesses would have resulted directly in the prosecution's presentation of additional evidence weighing in favor of death. Given these circumstances, it was even more reasonable for the Alabama Court of Criminal Appeals to conclude that Brooks had suffered no prejudice from trial counsel's decision not to pursue the mitigation strategy he now believes would have been best.2

Nor would the evidence of alcoholism and Brooks's intoxication on the night of the offense necessarily have compelled a reasonable jurist to conclude that the failure to present that evidence in mitigation prejudiced Brooks. As we've already explained, Brooks faced a formidable obstacle in the two aggravating factors presented by the state. To obtain relief now, he must demonstrate that being nice and polite, while on the other hand suffering from alcoholism and being intoxicated on the night of the offense, outweighs the brutal nature of the victim's death and the fact that he raped her and fled with many of her possessions.

Yet Brooks's best presentation of the alcoholism and intoxication evidence was hardly compelling and, in fact, was called into substantial question by both the evidence at trial and the state's rebuttal expert. The most that the jury and sentencing judge could have gleaned from Dr. Beidleman's testimony was that Brooks was a moderate alcoholic, and that he was intoxicated on the night of the murder. Moreover, Beidleman did not even testify that Brooks was so intoxicated that he was entirely incapable of forming the necessary intent to render him culpable. Rather, Beidleman opined that Brooks's "ability to plan and think rationally was profoundly impaired, if his alcohol intoxication and ingestion is accurate as we've described.... [H]e would have [had] difficulty engaging in goal directed behavior," and would have exhibited "impulsivity, poor planning, poor judgment, and poor cognitive organization, poor thinking skills."

In fact, there were strong reasons not to credit Beidleman's account. For one thing, he admitted on cross-examination that most of his information came from Brooks himself or from people who received their information from Brooks. More importantly, the extrinsic evidence presented during the guilt phase and during the penalty phase directly contradicted Beidleman's opinion that Brooks had been so intoxicated that he could not engage in goal-directed behavior. In his interview with Beidleman, Brooks admitted to having had sexual intercourse with the victim, which the physical evidence confirmed. Furthermore, the evidence presented during the guilt phase indicated that Brooks had the presence of mind to steal the victim's car, her credit card, personal checks, and a paycheck. See Brooks I, 695 So.2d at 179. Brooks also had taken several personal items from Campbell's apartment and pawned them, id., and a search of his apartment turned up Campbell's answering machine, the keys to her car, and receipts of purchases made with her credit card, Brooks II, 695 So.2d at 187.

In addition, had Brooks offered Beidleman as an expert witness, the state would have rebutted that presentation with its own expert, Dr. King. King would have given the jury further reason to doubt Beidleman's depiction of Brooks. As King recounted, the trial testimony had shown that the victim's body had been hidden underneath the bed, the bed had been made, and some of the blood at the scene had been wiped down — all of which indicated an effort to eliminate or conceal evidence of the crime. When Brooks was arrested by the police, he gave a twenty-two-page statement recounting the night of Campbell's murder. These circumstances led King to conclude that, "if [Brooks] were intoxicated at the time of the offense, which I think is a possibility, it certainly did not rise to the level where it impaired his memory to such an extent that he could not recall what took place," and that Brooks retained "the ability to form an intent" and engage in goal-directed behavior. King's testimony would also have added a new fact to the weight of evidence in aggravation: the state had administered an intelligence test to Brooks, which revealed that he was of average intelligence. As panels of this Court have suggested before, a petitioner's "average to above-average intelligence" could undercut the rest of his mitigation case. See Suggs v. McNeil, 609 F.3d 1218, 1231 (11th Cir.2010); Grayson v. Thompson, 257 F.3d 1194, 1227 (11th Cir.2001).

Indeed, we have repeatedly stressed that evidence of intoxication or alcoholism is a double-edged sword that itself could harm a petitioner's case. See Housel v. Head, 238 F.3d 1289, 1296 (11th Cir.2001). A panel of this Court has gone as far as saying that a "history of excessive alcohol and drug use" is not even "evidence in mitigation of the death penalty," and that "admission of some of this evidence might have been harmful to [petitioner's] case." Waldrop v. Jones, 77 F.3d 1308, 1313 (11th Cir.1996); see also Suggs, 609 F.3d at 1231 ("As we have repeatedly recognized, evidence of drug and alcohol use is often a two-edged sword that provides an independent basis for moral judgment by the jury." (internal quotation marks and citations omitted)). In this case in particular, the evidence of alcoholism or intoxication might well have seemed inconsistent with trial counsel's penalty-phase argument, which focused on any residual doubt that the jury had regarding Brooks's guilt. While an intoxication-mitigation strategy attempts to lessen the defendant's culpability for an act he concededly committed, a residual-doubt strategy depends on the defendant maintaining his innocence. See Hubbard v. Haley, 317 F.3d 1245, 1260-61 (11th Cir.2003). That Brooks's intoxication evidence could have blunted the force of his residual-doubt argument is merely another way in which the new mitigating evidence could have hurt Brooks as easily as it could have helped him.

In contrast to this case, past petitioners who have overcome AEDPA deference on a Strickland claim have made far more compelling showings that their lawyers missed powerful mitigating evidence. In Porter v. McCollum, for example, defense counsel failed to conduct a thorough investigation of the petitioner's background and did not uncover the following mitigation evidence: Porter served heroically in two of the most critical and horrific battles of the Korean War; he struggled to readjust to normal life upon returning from the conflict; he suffered from physical abuse as a child; and he had a brain abnormality, difficulty reading and writing, and limited schooling. 558 U.S. 30, 41, 130 S.Ct. 447, 175 L.Ed.2d 398 (2009).3 The Supreme Court found the Florida Supreme Court's finding that Porter suffered no prejudice to be unreasonable; "the relevance of Porter's extensive combat experience [wa]s not only that he served honorably under extreme hardship and gruesome conditions, but also that the jury might [have] f[ound] mitigating the intense stress and mental and emotional toll that combat took on Porter." Id. at 43-44, 130 S.Ct. 447. In Wiggins, the petitioner showed that his trial counsel had missed evidence of severe physical and sexual abuse that he suffered at the hands of his mother and later his foster parents. See 539 U.S. at 516-17, 123 S.Ct. 2527 (Wiggins's mother left the children home alone for days, forcing them to beg for food and to eat paint chips and garbage; she beat the children and had sex with men while the children slept in the same bed; in one incident, she forced Wiggins to touch a hot stove burner, resulting in hospitalization; Wiggins's foster father and siblings molested and raped him). The Court found this mitigating evidence to be "powerful," id. at 534, 123 S.Ct. 2527, and noted that it "contained little of the double edge we have found to justify limited investigation in other cases," id. at 535, 123 S.Ct. 2527. Finally, in Collier v. Turpin, a case that Brooks cites as a comparison, the petitioner presented much more substantial evidence of good character, including evidence based on specific acts rather than more generalized impressions. See 177 F.3d 1184, 1200 n. 20 (11th Cir.1999) (Collier worked from an early age to support himself, his aunt, and his uncle; he was in the army, then returned home to take care of his aunt while she was dying of cancer; he was a devoted husband and father and worked hard to provide for his family; shortly before committing his crime, he lost his job; Collier heroically attempted to save the life of a motorist who had gotten into an accident).

In sum, Brooks has not met his burden of showing that the balance of aggravating and mitigating factors would lead all reasonable jurists to conclude that he suffered prejudice during the penalty phase of his trial. On one side of the balance, the state presented extensive aggravating evidence during the penalty phase of Brooks's trial, and it would have presented both additional victim-impact evidence and Dr. King's testimony. On the other side of the balance, Brooks could have presented six witnesses, who would have said that he was nice, polite, and nonviolent as a young man; and Dr. Beidleman, who would have testified that Brooks was a moderate alcoholic and was intoxicated on the night of the offense. In light of the strength of the state's aggravating evidence, a reasonable jurist could conclude that there was no reasonable probability that Brooks's best possible mitigation presentation would have altered the outcome of his penalty phase proceeding. It logically follows that a reasonable jurist could conclude that Brooks suffered no prejudice during his direct appeal on account of direct appellate counsel's alleged failings in arguing this ineffectiveness-of-trial-counsel claim. Consequently, we affirm the district court's denial of habeas relief.

AFFIRMED.