Donnie Cleveland Lance, a Georgia prisoner convicted and sentenced to death for the murders of his ex-wife and her boyfriend, appeals the denial of his petition for a writ of habeas corpus, 28 U.S.C. § 2254. Lance contends that we should vacate his sentence on the grounds that his trial counsel provided ineffective assistance when he failed to introduce mitigating mental health testimony and character evidence during the penalty phase of Lance’s trial and when counsel failed to obtain funds to hire expert witnesses. We disagree. The Supreme Court of Georgia reasonably concluded that Lance did not suffer prejudice when counsel failed to introduce mental health testimony. Counsel also made a strategic decision not to introduce character evidence during the penalty phase that we decline to second guess. And the Supreme Court of Georgia reasonably concluded that counsel’s failure to obtain funds to hire expert witnesses did not prejudice Lance. We affirm.
I. BACKGROUND
We divide this background in three parts. We first recount the facts of the crime. We then summarize the preparation for, and disposition of, Lance’s trial, sentencing,- and direct appeal. We conclude with a summary of the state and federal habeas proceedings.
A. The Crime
Donnie Cleveland Lance murdered his ex-wife, Sabrina “Joy” Lance, and her boy-friend, Dwight “Butch” Wood, Jr., in the early morning of November 9, 1997, at Butch Wood’s home. Lance v. State (“Lance I”), 275 Ga. 11, 560 S.E.2d 663, 669-70 (2002). The Supreme Court of Georgia described the events surrounding the murders as follows:
Shortly before midnight on November 8, 1997, Lance called Joy Lance’s father, asked to speak to her, and learned that she was not at home. Shortly afterward, a passing police officer noticed Lance’s automobile leaving his driveway. Lance arrived at Butch Wood’s home, kicked in the front door, shot Butch Wood on the front and the back of his body with a *567shotgun, and then beat Joy Lance to death by repeatedly striking her in‘the face with the butt of the shotgun, which broke into pieces during the attack. Joy Lance’s face was rendered utterly unrecognizable. Later that morning, Lance told his friend, Joe Moore, that Joy Lance (whom Lance referred to in a derogatory manner) would not be coming to clean Lance’s house that day; that Butch Wood’s father could not “buy him out of Hell”; and that both Joy Lance and Butch Wood were dead. Lance later told a fellow inmate that he “felt' stupid” that he had called Joy Lance’s father before the murders, and Lance bragged to the inmate that “he hit Joy so hard that one of her eyeballs stuck to the wall.”
Hall v. Lance (“Lance II”), 286 Ga. 365, 687 S.E.2d 809, 811 (2010).
■ Lance had long abused Joy. Id. In the past, he had kidnapped her, and he had beaten “her with his fist, a belt, and a handgun.” Id. He had strangled her, electrocuted “her with a car battery,” and threatened “her with a flammable liquid, handguns, and a chainsaw.” Id, “He had repeatedly threatened to kill her himself, and he had once inquired of a relative about what it might cost to hire someone to kill her and Butch Wood.” Id. In 1993, Lance, accompanied by Joe Moore, “kicked in the door of Butch Wood’s home ... armed with a shotgun, loaded a shell into the chamber of the shotgun, and then fled only after a child in the home identified and spoke to Joe Moore.” Id.
B. Trial, Conviction, Sentence, and Direct Appeal
Lance hired J. Richardson Brannon to represent him at trial. An experienced criminal attorney, Brannon had tried around 160 criminal cases to verdict before Lance hired him. Three paralegals and a crime-scene investigator named Andy Pennington assisted Brannon in his preparation for trial. Lance and his family initially paid Brannon $50,000 to represent him, but after the exhaustion of that initial sum, the court declared Lance indigent and retained Brannon as court-appointed counsel.
Brannon then filed a motion for funds to hire expert witnesses, which he amended three times. The original motion and the first two amendments; filed in late 1998, requested funds to hire experts and a private investigator but did not specify the kinds of experts needed, their names, the fees they charged, or any other information. At a pre-trial hearing, Brannon requested funds to hire an expert on jury selection, a private investigator, a DNA serologist, a forensic pathologist, a ballistics expert, a criminologist, and an expert on shoe print analysis., He requested the jury expert by name and gave the court information about the hourly expenses of the requested private investigator, DNA serologist, and the forensic pathologist. Brannon stated that, of all the experts he requested, a forensic pathologist was “imperative” to establish “time of death” and “manner of death.” A month after the hearing, Brannon filed a third amendment to the motion for funds to hire expert witnesses. This amendment proffered the names, credentials, and fees of the experts requested,
The trial court initially denied the request for funds to. hire experts, but reversed course a month before trial and granted $4,000 to hire an investigator. Brannon used these funds to pay Pennington, a private investigator, and did not hire any other experts or present any other expert testimony during the guilt or penalty phases of the trial.
By contrast, the state introduced testimony from six expert witnesses at trial: *568Terry Cooper, an agent with the Georgia Bureau of Investigation, who testified about the crime scene and the shoe print he removed from the door at Butch Wood’s home; David Cochran, the chief crime scene investigator for the Jackson County, Georgia, Sheriffs Department, who testified about investigating the crime; Charles Moss, a fingerprint expert who testified that he was unable to retrieve prints from the shotgun shell casings involved in the crime; Bernadette Davy, a firearms expert, who testified about the shotgun shell casings found at the scene and the kinds of wood used to manufacture the butts of shotguns; Larry Peterson, a microanalyst who testified about the shoeprint found on Butch Wood’s door and other evidence found at the crime scene; and Frederick Heilman, an associate medical examiner for the Georgia Bureau of Investigation who testified about the causes of death pf Joy Lance and Butch Wood. Brannon extensively cross-examined each of these expert witnesses, except the fingerprint expert.
The defense theory of the case was innocence. Brannon attempted to establish an alibi defense based on the time of death and Lance’s whereabouts on November 8-9. Lance’s uncle testified that he was with Lance into the late evening of November 8 and then after midnight on November 9 until 5:00 a.m. Other witnesses corroborated this timeline and testified that Lance behaved normally immediately before and after the time when the murder occurred. Two children who were neighbors of Butch Wood also testified that they heard gunshots and a scream sometime after lunch on November 9, more than twelve hours later than when the crime allegedly occurred.
Pennington, the private investigator hired by Brannon, also testified as an expert crime scene technician. Pennington testified that the ballistics report from the crime scene suggested the possibility that the shooter used weapons in addition to the shotgun. He also testified that the absence of footprints on the stairs leading to the house was suspicious and that the lack of latent fingerprints on the shotgun shells suggested “[a] good burglar” committed the crime.
The jury found Lance guilty of the murders of Joy Lance and Butch Wood, of burglary, and of possession of a firearm during the commission of a crime. Lance I, 560 S.E.2d at 669. During the penalty phase of the trial, the state presented testimony from Joy Lance’s and Butch Wood’s relatives and from David Cochran, a crime scene investigator for the Georgia Bureau of Investigation. Brannon presented no mitigating evidence during the penalty phase. The jury sentenced Lance to death for the murders. Id. at 670.
The Supreme Court of Georgia affirmed Lance’s conviction and sentence on direct appeal. Id. at 670, 677-79. Lance argued that the trial court erred when it denied Lance’s motion for funds to hire expert witnesses. Id. at 671. But the Supreme Court of Georgia ruled that “Lance’s request for the contested funds was too unspecific, uncertain, and conclusory” to overturn his conviction. Id. Lance’s conviction became final when the Supreme Court of the United States denied his petition for a writ of certiorari. Lance v. Georgia, 537 U.S. 1050, 123 S.Ct. 620, 154 L.Ed.2d 525 (2002).
C. State and Federal Habeas Proceedings
In May 2003, Lance filed a petition for a writ of habeas corpus in the Butts County Superior Court. The superior court held an evidentiary hearing, at which Lance presented evidence that he réceived ineffective assistance of counsel because Brannon *569failed to investigate or present evidence of Lance’s mental impairments during the penalty phase of Lance’s trial. The Supreme Court of Georgia described the evidence on that issue as follows:
Lance presented testimony in the habe-as court from three experts in neuropsy-chology. Thomas Hyde, M.D., Ph.D., testified that he administered over 100 neurological tests to Lance. Yet, as his testimony establishes, only one of those tests indicated brain dysfunction. Dr. Hyde concluded that Lance had “significant damage to the frontal and temporal lobes” resulting from multiple blows to the head and from alcohol abuse. He testified that persons with frontal lobe dysfunction “often decompensate under periods of extreme emotional distress.” He also testified that such persons are unlikely to be able to plan and commit murder without leaving evidence but,instead, are more often “involved in crimes of impulse.” Dr. Hyde concluded that Lance’s mental state might have had an “impact” on Lance’s “ability to appreciate the criminality of his conduct or to conform his conduct to the requirements of law,” but he also acknowledged that other “reasonable” neurologists might disagree with his conclusions in Lance’s case. The second of Lance’s three experts in neuropsychology, Ricardo Weinstein, Ph.D., commented generally on Lance’s “psychosocial history” as follows:
[I]t’s a relatively unusual case in terms of his upbringing, fairly normal upbringing from an intact family, no major history of dysfunction, no history of child abuse, neglect, things of that nature, no history of significant mental illness in the family.
However, Dr. Weinstein concluded that Lancei as a result of multiple head injuries, the exposure to toxic fumes, the ingestion of gasoline, and a history of “heavy alcohol use starting at the age of 19,” suffered from “generalized and diffuse brain dysfunction” and “clear compromises in the frontal lobe functions.” Dr. Weinstein concluded that Lance was not insane or mentally retarded, that he understood “that certain behaviors are unacceptable,” but that his “brain dysfunction ... negatively impacted] his ability to conform his conduct to the requirements of the law.” In particular, Dr. Weinstein concluded that Lance would have difficulty in planning and in impulse control and that the combined effects of Lance’s brain dysfunction and his alcohol intoxication on the night of the murders would have rendered “his capacity to think in a logical, well-directed manner ... equivalent or similar to an individual that suffers from mental retardation.” Finally, Lance presented testimony from a third expert, David Pickar, M.D., who concluded that Lance, as a result of multiple head traumas and alcohol abuse, suffered from “impaired intellectual and frontal lobe function” that resulted in impairments of his ability to plan and to control his impulses.
Lance II, 687 S.E.2d at 814-15 (alterations in original).
The state presented the testimony of Dr. Daniel A. Martell, a neuropsychologist, who testified that Lance had an IQ of 79 and suffered from dementia:
[Martell] concluded that Lance functioned within “a range that’s higher than mild mental retardation but lower than average.” Dr. Martell added, however, that he had administered an additional test to determine what Lance’s IQ had been before any possible brain injuries and that the test showed Lance’s earlier IQ to fall within the “exact same ranges” as found by the various experts who testified in the habeas court. Dr. Martell *570testified that some of Lance’s test results indicated frontal lobe dysfunction, but. Dr. Martell further testified as follows:
His weaknesses with regard to frontal lobe have to do with a tendency to perseverate or repeat himself and mild to moderate impairment in problem-solving abilities in certain contexts like adapting to changing problems but not others like planning an effective strategy for solving a. problem. However, his ability to inhibit unwanted or impulsive behaviors appears to be relatively intact. And I think that’s important in my analysis with regard to the issue of the crime itself because these data do not suggest to me that he is, in fact, impulsive or unable to control his impulses.
Dr. Martell concluded that Lance’s frontal lobe dysfunction would not have prevented him from planning the murders and would not have made him so impulsive that he could not prevent himself from committing the murders. As we noted above, Dr. Martell also stated that Lance’s symptoms were so subtle that a typical court-ordered evaluation might not have given any indication of problems. Dr. Martell summarized his opinion by stating, “In my opinion, [Lance’s diagnosis is] not significant to the crime.”
Lance II, 687 S.E.2d at 815 (second alteration in original).
In addition, Lance presented evidence that Brannon rendered ineffective assistance when Brannon failed to introduce mitigating character evidence during the penalty phase of the trial. Friends and family testified that Lance was a loving father, that his children loved him, and that he had a good character. But Brannon testified that he chose not to introduce this evidence because to do so would have allowed the state to cross-examine the character witnesses about aggravating character evidence.
Lance also presented evidence that Brannon' rendered ineffective assistance during the guilt phase of the trial because Brannon’s request for funds to hire expert witnesses was. deficient and the failure to present this expert testimony prejudiced Lance. Brannon explained that experts “were needed in this case, particularly since it’s a death penalty case.” But he also testified that his motions for funds to hire expert witnesses were sufficiently detailed for the trial court to grant them. Although Lance’s habeas counsel acknowledged that Brannon “made a request repeatedly for expert assistance in the case and pointed out the specific categories that [Brannon] thought experts were critical to,” he argued that these motions were “vague and entirely unspecific.”
The superior court granted Lance’s petition in part and vacated his death sentence on the ground that Lance had received ineffective assistance of counsel during the penalty phase of his trial. The superior court found that Brannon’s failure to investigate and introduce evidence of Lance’s mental health history was unreasonable. The superior court also found that, had Brannon introduced evidence of Lance’s mental health history, “such an investigation ... would have provided significant mitigating evidence for the jury to consider.”
The Supreme Court of Georgia reversed and reinstated Lance’s death sentence. Lance II, 687 S.E.2d at 811-12, Although the Supreme Court of Georgia agreed with the superior court that Brannon’s failure to investigate Lance’s mental health history was deficient performance, it disagreed that Lance suffered prejudice. Id. at 812. The Supreme Court of Georgia explained that even if Brannon had investigated *571Lance’s mental health background, Bran-non would not have sought “a psychological evaluation of Lance,” because such an investigation would have revealed only mild mental impairment. Id. at 813. In addition, the court reasoned that “the trial court [would not] have abused its discretion ] if it had been asked by trial counsel for funds for a psychological evaluation of Lance, [but] determin[ed] that this information failed to show that the assistance of a psychologist was critical to Lance’s defense.” Id. (citation omitted).
In the alternative, the Supreme Court of Georgia held that even if Lance had presented the expert testimony that he presented during his habeas proceedings, there was not a reasonable probability that the testimony would have changed the outcome of the trial. The court explained that the evidence established only mild mental impairments, and “[a]gainst this somewhat mitigating evidence, the jury would have weighed Lance’s long history of horrific abuse against Joy Lance,” the horrific nature of the crime, and evidence about Lance’s statements and demeanor after the crime, such as his declaration that Butch Wood was in “Hell” and “his boast to an inmate that ‘he hit Joy so hard that one of her eyeballs stuck to the wall,’ ” Id. at 815-16.
The Supreme Court of Georgia also denied Lance’s claim that his trial counsel rendered ineffective assistance in his application for funds for forensic experts. Id. at 816. The court explained that his trial counsel was not deficient even though the court had described the motions on direct appeal as conclusory. Id. In the alternative, the court ruled that Brannon’s failure to request funds for several expert witnesses did not prejudice Lance. Id. at 817. Lance argued that his trial counsel should have obtained three additional experts: (1) “an expert to testify that the repeated blows to Joy Lance’s face with the butt of the shotgun likely would have resulted in the perpetrator being spattered with blood and brain matter”; (2) “an expert to testify that there were no shoe prints at the crime scene other than the one on the front door”; and (3) “an expert in polygraph science to testify that the results of the polygraph examination taken by Joe Moore were ‘inconclusive.’” Id. (footnote omitted). But according to the Supreme Court of Georgia, the absence of this testimony did not prejudice Lance. Testimony regarding “spattered ... blood and brain matter” was unnecessary because it would “have been obvious to the jury” and consistent with the evidence that showed Lance “had initially walked away from the crime scene rather than driving away in his automobile.” Id. The absence of expert testimony about shoe prints “was not a matter that was subject to varying scientific opinions.” Id. And, the absence of expert testimony regarding Joe Moore’s polygraph did not prejudice Lance because “Moore’s volunteered [polygraph] testimony was ruled inadmissible, and the jury was instructed to disregard it.” Id. (citation omitted).
Lance then filed a federal petition for a writ of habeas corpus, which the district court denied. The district court granted Lance a certificate of appealability about whether his trial counsel was ineffective in “preparing for and presenting [Lance’s] case in mitigation during the penalty phase of his trial.” Lance appealed and filed a motion to expand his certificate. We granted it on one issue: whether Brannon “rendered ineffective assistance when he failed to properly request funds for an investigator and expert witnesses.”
II. STANDARD OF REVIEW
“We review de novo the denial of a petition for a writ of habeas corpus.” Wil*572liamson v. Fla. Dep’t of Corr., 805 F.3d 1009, 1016 (11th Cir. 2015). The Antiter-rorism and Effective Death Penalty Act of 1996 imposes a “highly deferential standard for evaluating state-court rulings” that “demands that state-court decisions be given the benefit of the doubt.” Rutherford v. Crosby, 385 F.3d 1300, 1306-07 (11th Cir. 2004) (citations omitted). We will not disturb the decision of the state court unless it “was contrary to, or involved an unreasonable application of, clearly established Federal law, as determined by the Supreme Court of the United States” or “was based on an unreasonable determination of the facts in light of the evidence presented in the State court proceeding.” 28 U.S.C. § 2254(d); accord McClain v. Hall, 552 F.3d 1245, 1250 (11th Cir. 2008).
III. DISCUSSION
Lance argues that he is entitled to relief because Brannon rendered ineffective assistance of counsel and the decision of the Supreme Court of Georgia that Brannon did not do so was unreasonable, but we disagree. “It is by now hornbook law that to succeed on a Sixth Amendment ineffective-assistance claim, a petitioner must show that: (1) ‘counsel’s representation fell below an objective standard of reasonableness,’ and (2) ‘there is a reasonable probability that, but for counsel’s unprofessional errors, the result of the proceeding would have been different.’ ” Jones v. Sec’y, Fla. Dep’t of Corr., 834 F.3d 1299, 1312 (11th Cir. 2016) (quoting Strickland v. Washington, 466 U.S. 668, 688, 694, 104 S.Ct. 2052, 80 L.Ed.2d 674 (1984)). “To establish deficient performance, the petitioner must show that his attorney ‘made errors so serious that he was not functioning as the “counsel” guaranteed the defendant by the Sixth Amendment.’” Id. (alteration adopted) (citation omitted). “On the issue of prejudice, ... a reasonable probability of a different result means ‘a probability sufficient to undermine confidence in the outcome.’ ” Id. (citation omitted). “When a petitioner challenges his conviction, ‘the question is whether there is a reasonable probability that, absent the errors, the factfinder would have had a reasonable doubt respecting guilt.’” Id. (citation omitted). “When a capital petitioner challenges his death sentence, ‘the question is whether there is a reasonable probability that, absent the errors, the sentencer .., would have concluded that the balance of aggravating and mitigating circumstances did not warrant death.’ ” Id. (citation omitted). “The standards created by Strickland and [section] 2254(d) are both highly deferential, and when the two apply in tandem, review is doubly so.” Harrington v. Richter, 562 U.S. 86, 105, 131 S.Ct. 770, 178 L.Ed.2d 624 (2011) (citations and internal quotation marks omitted).
We divide our discussion in three parts. First, we explain that the Supreme Court of Georgia reasonably determined that Lance did not suffer prejudice when Bran-non failed to present mental health testimony in the penalty phase. Second, we explain that Brannon made a strategic decision not to introduce mitigating character evidence. Third, we explain that Brannon’s failure to obtain funds to hire expert witnesses was not prejudicial.
A. Inadequate Mental Health Investigation and Testimony
Although the parties dispute whether the Supreme Court of Georgia reasonably determined that Brannon’s failure to investigate and present expert mental health testimony during the penalty phase of the trial was deficient performance, we need not decide this question. We need only decide that the determination of *573the Supreme Court of Georgia that this deficiency did not affect the outcome of the case was reasonable.
Lance argues that, had Brannon performed a reasonable investigation, Bran-non would have discovered “red flags” in Lance’s background that would have led him to introduce mitigating evidence during the penalty phase of the trial. According to Lance, “a basic investigation would have revealed” that Lance was shot in the head, that Lance “had been hospitalized for mental health treatment” for depression, and that Lance abused alcohol. Lance argues that discovery of this evidence “necessarily would have led [Brannon] to [request] a comprehensive mental health investigation.” Such an investigation would have led in turn to the introduction of the expert testimony of doctors, such as Wein-stein, Pickar, and Hyde, that Lance suffered from “borderline retardation,” dementia, and frontal lobe dysfunction, which impaired Lance’s ability to control his behavior. Lance contends that, in the light of this evidence, it was unreasonable for the Supreme Court of Georgia to conclude that Lance did not suffer prejudice.
The Supreme Court of Georgia made two alternative holdings on prejudice, and we conclude that its second holding was not unreasonable. The Supreme Court of Georgia held that even if Brannon had introduced the mental health testimony presented at the state habeas hearing, that evidence would not have changed the outcome of the case. Lance II, 687 S.E.2d at 815. The court explained that the evidence presented on habeas review “showed merely that Lance functioned, when sober, in the lower range of normal intelligence”; had memory issues; suffered from mild depression; was “somewhat impulsive”; and had some trouble problem solving. Id.
This conclusion was not unreasonable because much of the evidence that Lance introduced in the superior court of his mental impairments was not necessarily mitigating. We have often acknowledged that juries may infer that a defendant’s alcohol abuse or impulsive behavior that is triggered by organic brain damage is aggravating. See Evans v. Sec’y, Dep’t of Corr., 703 F.3d 1316, 1329 (11th Cir. 2013) (en banc) (“We have held too that evidence of substance abuse ‘can do as much or more harm than good in the eyes of the jury.’ ” (citation omitted)); Rhode v. Hall, 582 F.3d 1273, 1285-86 (11th Cir. 2009) (“Counsel reasonably believed that the jury would see Rhode’s impulsive behavior, which more than one expert believed was triggered by his organic brain damage, as aggravating.”). And although Lance insists that the Supreme Court of Georgia erred because it “never even mentioned the word ‘dementia’ in its decision,” the Georgia Supreme Court did acknowledge “new evidence of subtle neurological impairments.” Lance II, 687 S.E.2d at 816. This characterization of the evidence was not objectively unreasonable.
Indeed, “[o]ur analysis of the prejudice prong ... must also take into account the aggravating circumstance’s associated with [Lanee]’s case ....” Dobbs v. Turpin, 142 F.3d 1383, 1390 (11th Cir. 1998). “At the end of the day, we are required to ‘reweigh the evidence in aggravation against the totality of available mitigating evidence.’ ” Boyd v. Allen, 592 F.3d 1274, 1301 (11th Cir. 2010) (quoting Wiggins v. Smith, 539 U.S. 510, 534, 123 S.Ct. 2527, 156 L.Ed.2d 471 (2003)). In Boyd, for example, we explained that although trial counsel’s investigation overlooked mitigating evidence of childhood abuse that “undeniably would have been relevant to Boyd’s mitigation case,” we determined “that the evidence of abuse would not ultimately have affected weighing the aggravators and the miti-gators.” Id. at 1299. The petitioner in Boyd *574had participated in a gruesome double murder that culminated in Boyd and his accomplice beating and shooting the victims. Id. at 1279-81. Boyd later “bragged about the killings and about how cold blooded he was.” Id. at 1284. In the light of these circumstances, “we conclude[d] that the totality of mitigating evidence ... pales when compared to the brutal nature and extent of the aggravating evidence.” Id. at 1302. As in Boyd, the aggravating factors of Lance’s crime are substantial. He had a long history of abusing Joy Lance, he beat her during the crime until her face was “utterly unrecognizable,” he made derogatory statements about her and Butch Wood, and Lance showed little remorse after the crime. Lance II, 687 S.E.2d at 811. And Lance’s new mitigating evidence fails to convince us that the Georgia Supreme Court unreasonably determined that Lance was not prejudiced by his defense counsel’s performance.
Lance urges us to follow a trio of decisions—Rompilla v. Beard, 545 U.S. 374, 125 S.Ct. 2466, 162 L.Ed.2d 360 (2005), Williams v. Taylor, 529 U.S. 362, 120 S.Ct. 1495, 146 L.Ed.2d 389 (2000), and Porter v. McCollum, 558 U.S. 30, 130 (S.Ct. 447, 175 L.Ed.2d 398 2009)—but each decision involved undiscovered evidence that is substantially more mitigating than the evidence Lance introduced on state habeas review. Had trial counsel in Rompilla performed an adequate investigation, he would have discovered that the defendant was raised in a “slum environment,” suffered from schizophrenia, and had a third-grade level of cognition. Rompilla, 545 U.S. at 390-91, 125 S.Ct. 2456. Moreover, the Supreme Court was not bound by the same deferential standard of review that we are. Id. at 390, 125 S.Ct. 2456 (conducting a de novo review). Nor is the mitigating evidence here like the evidence uncovered in Williams. Unlike the evidence that Lance argues Brannon would have uncovered, had Williams’ trial counsel performed an adequate investigation, he “would have uncovered extensive records graphically describing Williams’ nightmarish childhood.” Williams, 529 U.S. at 395, 120 S.Ct. 1495. And in Porter, an adequate investigation would have revealed the defendant’s heroic military service, “his struggles to regain normality upon his return from war,” a childhood of abuse, and.a brain abnormality. Porter, 558 U.S. at 41, 130 S.Ct. 447.
Lance erroneously contends that the Supreme Court of Georgia applied an incorrect prejudice standard, because, according to Lance, it asked “whether the sentencing jury ‘might’ have considered the mitigating evidence and nonetheless imposed the death penalty,” when the correct inquiry is “whether the mitigating evidence might have caused the jury to impose a life sentence in lieu of the death penalty.” But this latter standard was the standard that the Supreme Court of Georgia applied; it asked whether, “in reasonable 'probability [the mitigating evidence would] have changed the outcome of the sentencing phase if it had been presented at Lance’s trial." Lance II, 687 S.E.2d at 816.
Lance also argues that the Supreme Court of Georgia improperly “brushed aside” the factual findings of the superior court, but we disagree. The Supreme Court of Georgia accepted the factual findings of the superior court, but determined the legal question of prejudice de- novo, id. at 812, 815, which Georgia law requires. Humphrey v. Morrow, 289 Ga. 864, 717 S.E.2d 168, 172 (2011). The Supreme Court of Georgia reasonably concluded that Lance did not suffer prejudice when Bran-non failed to introduce mental health testimony.
*575 B. Mitigating Character Evidence
Lance next argues that Brannon rendered ineffective assistance when Bran-non failed to introduce mitigating character evidence during the penalty phase of the trial. Although this claim appears to be procedurally defaulted because the Supreme Court of Georgia held that it was abandoned, Lance II, 687 S.E.2d at 819, neither party addresses this preliminary question, so we deny Lance’s claim on its merits. Valle v. Sec’y for Dep’t of Corr, 459 F.3d 1206, 1213 (11th Cir. 2006) (“Here, it is unnecessary to address the issue of the procedural bar, because even assuming the claim is preserved, Valle is not entitled to habeas relief .... ”). Lance argues that, had Brannon investigated Lance’s background, Brannon would have introduced character evidence that painted Lance as a “quiet, peaceful man,” who was “normally a kind, dependable, and compassionate person.” Brannon “also could have identified witnesses to testify that [Lance] loved his son and daughter dearly, and they loved him in return.” Lance argues that Brannon’s failure to do so constituted ineffective assistance. We disagree.
The decision not to introduce mitigating character evidence was a reasonable strategic decision. Brannon testified that he chose not to introduce mitigating character evidence because it would have opened the door to the introduction of aggravating character evidence. We have repeatedly held that this kind of decision—to call or not call certain witnesses—is the “epitome of a strategic decision ... that we will seldom, if ever, second guess,” Waters v. Thomas, 46 F.3d 1506, 1512 (11th Cir. 1995); accord Ledford v. Warden, Ga. Diagnostic & Classification Prison, 818 F.3d 600, 647 (11th Cir. 2016); Conklin v. Schofield, 366 F.3d 1191, 1204 (11th Cir. 2004), and we decline to do so here.
C. Motion to Obtain Funds for Expert Witnesses
Lance argues that Brannon rendered ineffective assistance because the motions for funds to hire expert witness fell below the standard set by Georgia law and that this deficiency caused Lance to suffer prejudice, but we disagree. Although the Supreme Court of Georgia held that the motions for funds were not deficient, we consider only its holding that the motions did not cause prejudice.
Lance argues that, had Brannon obtained funds to hire expert witnesses, he would have presented the testimony of a forensic pathologist, a crime scene expert, a polygraph expert, and a fingerprint expert. He contends that there is a reasonable probability that this additional testimony would have changed the outcome of the case. A forensic pathologist, according to Lance, would have “uncovered ... the lack of physical evidence” in the case, testified to inconsistencies in Agent Cooper’s testimony on the times of death, and explained that' it could have been “ ‘virtually impossible for the person administering the blows [to Joy Lance] to escape from the scene with little or no blood on her/ him.’ ” A crime scene expert, according to Lance, would have testified that the Jackson County Sheriffs Office “failed to look for footprints, tire marks, or other evidence on the ground around Wood’s home.” A polygraph expert, according to Lance, would have discredited the testimony of Joe Moore, who took a polygraph test and implicated Lance in the murders. And a fingerprint expert, according to Lance, would “have testified that [Lance’s] fingerprints were never found in or around the crime scene.”
The Supreme Court of Georgia held that this additional testimony would not have changed the outcome of the trial. Lance II, 687 S.E.2d at 816-17.The court explained *576“there is, even now, no substantial dispute among the experts regarding the time of death but, instead, that there is merely a dispute over the manner in which the time of death was established.” Id. at 816. The court reasoned that the lack of blood spatter would “have been obvious to the jury,” and it was also “consistent with Lance having disposed of any bloody clothes at the same time he obviously disposed of his distinctive shoes.” Id. at 817. In addition, “the absence . of shoe prints was not a matter that was subject to varying scientific opinions.” Id. A polygraph expert was unnecessary, according to the Supreme Court of Georgia, because “Moore’s volunteered testimony was ruled inadmissible, and the jury was instructed to disregard it.” Id. (citation omitted). Finally, the court explained that the lack of fingerprint evidence was a “matter of common sense,” not varying scientific opinions. Id.
The decision of the Supreme Court of Georgia was not an unreasonable application of federal law. The court weighed the new evidence presented by Lance during state habeas proceedings and concluded that the new evidence would not have changed the outcome of Lance’s trial. Lance II, 687 S.E.2d at 816-17. This analysis is the analysis Strickland commands. The state habeas court compared the “totality of the evidence before the ... jury” with the new evidence presented by Lance and concluded that the new evidence had “an isolated, trivial effect” on the whole “evidentiary picture.” Strickland, 466 U.S. at 695, 104 S.Ct. 2052. There was little testimony introduced that went beyond ruminations about common sense facts, and no testimony that fundamentally undermined the state’s case. As the district court correctly explained, Lance’s claim amounts to a “quibble! ]” with the conclusion of the Supreme Court of Georgia, not that the conclusion was truly unreasonable. None of the evidence presented by Lance would have had a “pervasive effect on the inferences” drawn by the jury. Strickland, 466 U.S. at 695-96, 104 S.Ct. 2052.
IV. CONCLUSION
We AFFIRM the denial of Lance’s petition for a writ of habeas corpus.