Petitioner Wydell Evans, a Florida death row inmate, appeals the District Court’s denial of his first federal habeas corpus petition. This appeal presents a single claim for relief — whether Evans was denied the constitutional right to effective assistance of counsel at the penalty phase of his capital trial. Because Evans filed his federal petition after April 24, 1996, this case is governed by 28 U.S.C. § 2254, as amended by the Anti-Terrorism and Effective Death Penalty Act of 1996 (AEDPA). See Guzman v. Sec’y, Dep’t of Corr., 663 F.3d 1336, 1345 (11th Cir.2011). Where, as here, a state court has denied an ineffective assistance of counsel claim on the merits, the standard a petitioner must meet to warrant federal habeas relief “was intended to be, and is, a difficult one.” Johnson v. Sec’y, DOC, 643 F.3d 907, 910 (11th Cir.2011) (citing Harrington v. Richter, — U.S. -, 131 S.Ct. 770, 786, 178 L.Ed.2d 624 (2011)). After carefully reviewing the record in this case and having the benefit of oral argument, we conclude that Evans has satisfied AED-PA’s demanding standard and, therefore, that habeas relief is warranted.
I. BACKGROUND FACTS1
A. Trial
The Florida Supreme Court summarized the facts presented at Evans’s 1999 trial as follows:
On October 21, 1998, two days after being released from prison, Wydell Evans shot and killed his brother’s seventeen-year-old girlfriend, Angel Johnson, during an argument over her alleged unfaithfulness to Evans’[s] brother. At the time of the shooting, Evans was in an automobile with Johnson, Erica Foster, Sammy Hogan, and Lino Odenat. At some point during the argument, Johnson laughed, to which Evans responded, “You think it’s funny? You think it’s funny?” Evans then pulled out a gun and shot Johnson in the chest.
Evans I, 838 So.2d at 1092.
After the shooting, Evans instructed Hogan to drive him to a friend’s house. Along the way, Evans instructed Odenat to get rid of the gun. After stopping at the friend’s house, Evans directed Hogan to take him to a nearby parking lot where *1245Evans tried to wipe his prints from the car. He also threatened to kill Foster and Hogan if they told authorities who shot Johnson. Once Evans was out of the car, Foster and Hogan took Johnson to the hospital where she later died from her gunshot wound.
At the hospital, Foster and Hogan eventually identified Evans as the shooter. He was arrested the next day. During his jury trial in October 1999, the defense theory was that Evans accidentally shot Johnson while handing the gun to her. Evans testified to this version of events during the guilt phase of his trial. He was convicted of one count of first-degree premeditated murder, one count of kidnapping, and one count of aggravated assault.
B. Penalty Phase
On November 3, 1999, a penalty phase was conducted. The state established that Evans had two prior convictions for battery upon a law enforcement officer and a prior conviction for aggravated battery, and that he was on probation at the time of the offense.
Defense counsel presented no mental health mitigation. But defense counsel did present the testimony of several character witnesses, including Lilly Evans (defendant’s mother), Sandra Evans (defendant’s aunt), Minne Jarrett (defendant’s cousin), Linda Key (defendant’s lifelong friend), and Patty Walker (defendant’s cousin). These character witnesses gave positive accounts of Evans, “describing him as a generous man, a good father, a loving and obedient son and grandson, a good friend, and someone who counseled children to stay out of trouble by staying in school.” Evans II, 946 So.2d at 4. Further, Lilly Evans told the jury that while Evans had been an obedient child and good in school, her addiction to crack cocaine had affected his behavior. She also testified that Evans had been her inspiration to stop abusing crack cocaine. Evans testified that his mother’s crack addiction adversely affected him.
The jury recommended the death penalty by a vote of ten to two. Then the trial court held a hearing pursuant to Spencer v. State, 615 So.2d 688 (Fla.1993), and entered a written order sentencing Evans to death. The trial court found two aggravating circumstances: (1). Evans had been convicted of prior violent felonies, Fla. Stat. § 921.141(5)(b); and (2) the crime was committed while Evans was on probation, Fla. Stat. § 921.141(5)(a). Although Evans did not request the trial court to consider any of the statutory mitigators, the court considered each statutory miti-gator in its sentencing order and found that none applied. Based on the character evidence presented during the penalty phase, the trial court found the following nonstatutory mitigating circumstances: “(1) Evans experienced an abused or deprived childhood; (2) he contributed to society; (3) he performed charitable deeds; (4) he counseled youth to avoid crime and stay in school; and (5) he exhibited good behavior in prison.” Evans II, 946 So.2d at 5 n. 3. But the trial court only gave these mitigating circumstances “some” or “little weight.” Ultimately, the trial court concluded that the aggravating circumstances outweighed the mitigating circumstances and sentenced Evans to death for the first-degree murder conviction.2
C. Direct Appeal
The Florida Supreme Court affirmed Evans’s convictions and death sentence on direct appeal. Evans I, 838 So.2d 1090. Three Florida Supreme Court justices concurred in affirming Evans’s convictions, but dissented as to the death penalty, find*1246ing that a death sentence was disproportionate because the homicide “occurred during an argument between Evans and the victim over her alleged unfaithfulness to his brother.” Id. at 1099 (Shaw, J., dissenting joined by Anstead, C.J., and Pariente, J.).3 The Supreme Court denied certiorari review. Evans v. Florida, 540 U.S. 846, 124 S.Ct. 121, 157 L.Ed.2d 84 (2003).
D. State Postconviction Evidentiary Hearing
Evans filed a motion for postconviction relief pursuant to Fla. R.Crim. P. 3.851 in the state trial court raising ten claims, including the penalty phase ineffective assistance of counsel claim now before us. The state trial court held an evidentiary hearing on October 19 and 20 and December 16, 2004. At the evidentiary hearing, Evans presented the testimony of five lay witnesses: Lilly Evans, Sandra Evans, Oren Evans (defendant’s brother), Margaret O’Shaunessy (defendant’s special education teacher), and Barbara McFadden (defendant’s school counselor). He also presented the testimony of three mental health experts, Dr. Richard Carpenter, Dr. Henry Dee, and Dr. Harry McClaren,4 and the testimony of his trial counsel. Evans also testified.
The Florida Supreme Court summarized the lay witness evidence presented at the evidentiary hearing as follows:
Evans presented evidence of a troubled childhood. He presented his school and medical records, which demonstrated that he was hit by a car and was diagnosed with a closedhead injury at the age of three. [FN 9] His mother, Lilly, and his aunt, Sandra, witnessed the accident and testified that he stopped breathing for at least a minute after it occurred. They also testified that his speech and language patterns changed, and he began to stutter. In addition, Evans’[s] life at home was difficult. His father died when he was young, and he grew up with a mother who was addicted to crack cocaine and a stepfather who was abusive. As a result of his stuttering and difficulties at home, Evans was frequently picked on and beaten up by neighborhood children. He eventually retaliated and started fighting frequently. Lilly testified that his temper “became pretty quick.” Evans’[s] brother, Oren, testified that Evans was very aggressive and recalled instances of Evans throwing rocks at a police officer.
FN9. Evans’[s] injury is noted in his school records because Evans’ mother relayed the incident to a school social worker during one of Evans’[s] evaluations.
Evans’[s] school records showed a history of speech and learning disabilities, as well as escalating behavioral problems due to anger and aggression. In elementary school, Evans was placed in a class for children with learning disabilities and received speech therapy. While Evans’[s] IQ tested in the normal range, disparities between his verbal and performance scores on the test indicated severe learning disabilities. Evans received mostly failing grades. In addition to academic problems, Evans was regularly disciplined for being *1247disruptive and, as he got older, was frequently expelled from school for fighting. Evans’[s] violent behavior escalated from being aggressive with other students to being aggressive with a teacher. He was eventually classified as emotionally handicapped and was recommended for the severely emotionally disturbed program in high school, a program for the most violent students. School psychology reports described Evans as having poor impulse control, displaying excessive aggression, and having difficulty controlling anger. At the evidentiary hearing, Evans’[s] high school teacher, Barbara McFadden, and counselor, Margaret O’Shaunessy, also described Evans as angry, violent, impulsive, and having an explosive temper.
Evans II, 946 So.2d at 6-7. All of the lay witnesses testified that they would have testified at Evans’s trial if they had. been asked. But while both Lilly and Sandra Evans testified at the penalty phase, only Lilly was asked before trial to do so.
Three mental health experts also testified at the evidentiary hearing: Dr. Carpenter, Dr. Dee, and Dr. McClaren. All three testified that Evans “had brain damage attributable to his head injury.” Evans II, 946 So.2d at 7. The experts also “agreed that the difference between Evans’[s] verbal and performance scores on his IQ test indicated possible brain damage as further evidenced by his learning disabilities.” Id. But Dr. Carpenter and Dr. Dee parted ways with Dr. McClaren “over whether Evans’[s] brain damage led to any particular behavior.” Id. at 7-8.
Dr. Carpenter and Dr. Dee opined that Evans “suffered from an uncontrollable rage reaction or impulse disorder as a result of the brain damage.” Id. at 8. They agreed that at the time Johnson was shot, Evans’s “behavior demonstrated that he was experiencing an uncontrollable rage reaction, which was further exacerbated by his alcohol use.” Id. In Dr. Carpenter’s opinion, Evans’s “behavior at the time of the shooting was due to alcohol abuse, impulse disorder, and rage reaction secondary to Evans’[s] closed-head injury.”5 Id. Dr. Dee testified that Evans’s “alcohol abuse exacerbated his brain damage, causing Evans to experience an instantaneous rage known as ‘clicking.’ ”6 Id. Both testified that Evans did not form the intent to kill Johnson. Id. Instead, because of Evans’s impulse control disorder, both experts testified that he met the criteria for Florida’s two statutory mental-state miti-gators: (1) “[t]he capital felony was committed while the defendant was under the influence of extreme mental or emotional disturbance,” Fla. Stat. § 921.141(6)(b); and (2), “[t]he capacity of the defendant to appreciate the criminality of his or her conduct or to .conform his or her conduct to the requirements of law was substantially impaired,” id. at § 921.141(6)(f).7 See Evans II, 946 So.2d at 8.
*1248According to the Florida Supreme Court, “[wjhile Dr. McClaren agreed that Evans suffered from a brain dysfunction and some degree of brain damage, he did not agree that Evans’[s] brain dysfunction led him to behave in any particular way.” Id. Dr. McClaren opined that Evans’s brain damage was “probably overshadowed by personality characteristics and the effect of alcohol.” Id. More specifically, Dr. McClaren testified as follows:
There certainly is evidence that he sustained a closed-head injury as a young child and subsequently had difficulties with his behavior and learning in school which progressed to juvenile delinquency and adult criminality. While some minor degree of brain dysfunction is certainly a possibility given his history of head injury and emotional and behavioral problems as a child prior to the onset of criminality!,] his behavior later in life appears to be more associated with a criminal lifestyle than with the effects of some disabling cognitive disorder of such a severity to cause impairments which would be a significant difficulty for him.
Id. at 8 n. 16. While Dr. McClaren agreed with the other experts that alcohol was a significant factor in Evans’s behavior at the time of the shooting, he was uncertain about the quantity of alcohol that Evans had consumed before the shooting. See id. at 8-9. He noted that Evans reported not drinking during the car ride and had indi-eated that he was under control and otherwise knew what he was doing. Id. at 9. As a result, Dr. McClaren concluded that Evans did not meet the criteria for the two statutory mental health mitigators, Fla. Stat. § 921.141(6)(b) & (f).8 Id. at 9.
Evans’s trial counsel also testified at the evidentiary hearing. He testified that he met with Evans two times between the time of his initial appointment and the start of trial for a total of approximately four hours. Trial counsel testified that he did not investigate Evans’s “mental health because his conversations with Evans gave him no reason to believe that Evans was insane or incompetent.” Id. Further, trial counsel affirmed that the presentence investigation report (PSI) prepared for sentencing “indicated that his mental health was perfect and that he had only seen a mental heath expert when he was young.” Id.
Trial counsel explained that he presented an accidental shooting defense during the guilt phase because Evans insisted on it. Id. Although trial counsel was aware there was evidence that Evans had been drinking leading up to the shooting, Evans told him he was not drunk and knew what he was doing. Id. In short, trial counsel testified that he had no reason to question Evans’s veracity and planned to exploit inconsistencies in the evidence to create reasonable doubt. Id. Further, trial counsel testified that the accident theory *1249was supported by Evans’s statements and there was no evidence to support a heat of passion defense. Id.
With respect to the penalty phase, trial counsel testified that he started preparing more than five months prior to trial. He met with Lilly for approximately thirty-minutes and asked her to collect good character witnesses because Evans could face the death penalty. According to Lilly’s testimony, trial counsel told her he only wanted good character information. Trial counsel testified that Lilly told him that Evans had been an obedient child and had received good grades in school. Trial counsel asked Lilly to provide him names of potential character witnesses, whom he later contacted by mail. Lilly testified that trial counsel did not explain mitigation to her; that trial counsel did not address other matters during their meeting; that she never met with a defense investigator; that she only met with trial counsel this one time prior to trial; and that trial counsel otherwise did not have time to discuss the case with her; but that she did speak with counsel at trial prior to her testimony. Based on his interview with Lilly, trial counsel “adopted the strategy of showing Evans’[s] redeeming qualities, such as how Evans spent time with his children and devoted time to his grandmother, at the penalty phase.” Evans II, 946 So.2d at 9. Thus, without the benefit of Evans’s school or medical records, which trial counsel never obtained,9 or any awareness of Evans’s closedhead injury, counsel chose to present a positive character “strategy because he was aware that Evans’s criminal record indicated a propensity for violence, and he did not want to open the door to the State introducing evidence of Evans’[s] acts of violence during his teenage years.” Evans II, 946 So.2d at 9-10. Moreover, trial counsel testified he did not have any evidence to support statutory mitigation and did not believe that Evans may have met the two statutory mental health miti-gators. Id. at 10. Trial counsel was not aware that Evans had any problems in school. Id.
Following the evidentiary hearing, the state trial court denied relief on February 15, 2005 in a written order. With respect to Evans’s claim that trial counsel was ineffective at the penalty phase (also referred to by Evans, the trial court, and the Florida Supreme Court as trial counsel’s failure to challenge the state’s case at penalty phase), the trial court found neither that trial counsel was deficient nor that Evans had demonstrated prejudice.
E. Postconviction Appeal to Florida Supreme Court
On direct appeal from the trial court’s denial of Evans’s Rule. 3.851 motion, the Florida Supreme Court considered three ineffective assistance of counsel claims, including the penalty phase ineffective assistance of counsel claim at issue in this appeal.10 Evans II, 946 So.2d at 6, 10-15. *1250After correctly citing to Strickland v. Washington, 466 U.S. 668, 104 S.Ct. 2052, 80 L.Ed.2d 674 (1984), as the governing standard, Evans II, 946 So.2d at 10, the Florida Supreme Court affirmed the trial court’s denial of Evans’s penalty phase ineffective assistance of counsel claim. Id. at 11-13. In doing so, the Florida Supreme Court stated:11
Evans next claims that trial counsel was ineffective for failing to adequately investigate his background for mental health mitigation to present at the penalty phase. Evans claims that defense counsel was deficient for failing to discover a wealth of mitigating information, including Evans’[s] brain injury and lifelong emotional and behavioral problems, and for failing to present anything other than positive character evidence at the penalty phase. Evans alleges that he was prejudiced by counsel’s deficient investigation because discovery and presentation of his brain injury and impulse disorder would have led to an instruction on both mental health statutory miti-gators.
We decline to address whether counsel’s investigation was deficient because we affirm the trial court’s conclusion that Evans has failed to demonstrate that he was prejudiced by counsel’s failure to present the mitigation evidence presented at the evidentiary hearing. “[Bjecause the Strickland standard requires establishment of both [deficient performance and prejudice] prongs, when a defendant fails to make a showing as to one prong, it is not necessary to delve into whether he has made a showing as to the other prong.”
Evans has failed to establish prejudice because the mitigation evidence he presented at the evidentiary hearing would likely have been more harmful than helpful. “An ineffective assistance claim does not arise from the failure to present mitigation evidence where that evidence presents a double-edged sword.” Reed v. State, 875 So.2d 415, 437 (Fla. 2004). While the testimony presented at the evidentiary hearing established that Evans suffered from mental health problems, it also displayed a long history of behavioral problems and escalating violence throughout his school career. Presenting this evidence at the penalty phase would have resulted in the jury hearing about Evans’[s] aggression towards students and teachers, his aggression towards police officers, his pride in being known as a “jack-boy” because he robs drug dealers, and his habit of carrying a gun. It is just as likely that this evidence would have been more “aggravating” than mitigating. See Reed, 875 So.2d at 436-37 (denying ineffective assistance claim because “even if [defense] counsel had ... investigated further, the testimony that could have been present*1251ed was just as likely to have resulted in aggravation against rather than mitigation for [the defendant]”).
Moreover, this mitigation evidence presented at the evidentiary hearing is inconsistent with Evans’[s] own testimony at the guilt phase. The evidence presented at the evidentiary hearing regarding Evans’[s] uncontrollable rage disorder is inconsistent with his testimony that he had a “clear recollection” of the shooting because he was focused and in control.
Based on the foregoing, Evans has not established that there is a reasonable probability that his sentence would have been different had counsel discovered and presented the mitigation evidence Evans presented at the evidentiary hearing. See Strickland, 466 U.S. at 694, 104 S.Ct. [at 2068] (establishing that appropriate test for prejudice is whether defendant has demonstrated reasonable probability that outcome would have been different but for counsel’s unprofessional errors and defining reasonable probability as “probability sufficient to undermine confidence in the outcome”). Therefore, we are confident in the outcome and affirm the trial court’s denial of this claim.
Evans II, 946 So.2d at 12-18 (some citations omitted).
F. Federal Habeas Proceedings
Evans timely filed a § 2254 petition in the District Court for the Middle District of Florida on May 29, 2007, raising fifteen claims. After briefing by the parties, but without an evidentiary hearing, the District Court denied the petition on September 29, 2010. The District Court granted Evans a COA as to his claim that trial counsel was ineffective during the penalty phase.
II. STANDARDS OF REVIEW
A district court’s grant or denial of a habeas corpus petition is reviewed de novo. Ward v. Hall, 592 F.3d 1144, 1155 (11th Cir.2010). AEDPA “imposes a highly deferential standard for evaluating state-court rulings” and “demands that state-court decisions be given the benefit of the doubt.” Renico v. Lett, — U.S. -, -, 130 S.Ct. 1855, 1862, 176 L.Ed.2d 678 (2010) (quotation marks omitted). Because Evans seeks habeas relief under 28 U.S.C. § 2254 based upon a claim that was adjudicated on the merits by the state courts, we are restricted in our ability to grant relief by § 2254(d). See Cave v. Sec’y, Dep’t of Corr., 638 F.3d 739, 742-43 (11th Cir.2011). To grant Evans’s ha-beas petition, we must find not only that Evans’s constitutional claims are meritorious, but also that the state court’s resolution of those claims:
(1) resulted in a decision that was contrary to, or involved an unreasonable application of, clearly established Federal law, as determined by the Supreme Court of the United States; or
(2) resulted in a decision that was based on an unreasonable determination of the facts in light of the evidence presented in the State court proceeding.
28 U.S.C. § 2254(d); see also Cave, 638 F.3d at 742-43. The Supreme Court has recognized that § 2254(d)(l)’s “contrary to” and “unreasonable application” clauses each have independent meaning. Bell v. Cone, 535 U.S. 685, 694, 122 S.Ct. 1843, 1850, 152 L.Ed.2d 914 (2002).
A state court’s decision is “contrary to” clearly established Supreme Court precedent in either of two respects: (1) “if the state court applies a rule that contradicts the governing law set forth in [Supreme Court] cases,” or (2) “if the state court confronts a set of facts that are materially indistinguishable from a deci*1252sion of [the Supreme] Court and nevertheless arrives at a result different from [the Supreme Court’s] precedent.” Williams v. Taylor, 529 U.S. 362, 405-06, 120 S.Ct. 1495, 1519-20, 146 L.Ed.2d 389 (2000).
To determine whether a state court decision is an “unreasonable application” of clearly established federal law, the Supreme Court recently stated:
The pivotal question is whether the state court’s application of the [relevant constitutional] standard was unreasonable .... For purposes of § 2254(d)(1), an unreasonable application of federal law is different from an incorrect application of federal law. A state court must be granted a deference and latitude that are not in operation when the case involves review under the [relevant constitutional] standard itself.
A state court’s determination that a claim lacks merit precludes federal ha-beas relief so long as fairminded jurists could disagree on the correctness of the state court’s decision. And as the [Supreme Court] has explained, evaluating whether a rule application was unreasonable requires considering the rule’s specificity. The more general the rule, the more leeway courts have in reaching outcomes in case-by-case determinations.
Harrington, 131 S.Ct. at 785-86 (2011) (citations, quotation marks, and emphasis omitted).
In addition, a state court’s factual determination is entitled to a presumption of correctness. 28 U.S.C. § 2254(e)(1). But AEDPA’s “statutory presumption of correctness applies only to findings of fact made by the state court, not to mixed determinations of law and fact.” Parker v. Head, 244 F.3d 831, 836 (11th Cir.2001); see also Hallford v. Culliver, 459 F.3d 1193, 1199 (11th Cir.2006). “An ineffective assistance of counsel claim is a mixed question of law and fact.... ” Williams v. Allen, 542 F.3d 1326, 1336 (11th Cir.2008).
Moreover, because “the standards created by Strickland [for evaluating the merits of an ineffective assistance of counsel claim] and § 2254(d) are both ‘highly deferential,’ ” Harrington, 131 S.Ct. at 788 (quoting Strickland, 466 U.S. at 689, 104 S.Ct. at 2065), “when the two apply in tandem, review is ‘doubly’ so.” Id. (quoting Knowles v. Mirzayance, 556 U.S. 111, 123, 129 S.Ct. 1411, 1420, 173 L.Ed.2d 251 (2009)); see also, Cullen v. Pinholster, — U.S.-,-, 131 S.Ct. 1388, 1403, 179 L.Ed.2d 557 (2011) (stating double deference standard and reversing grant of habeas relief for failure to pay adequate deference to state court adjudication of ineffective assistance of counsel claim); Premo v. Moore, — U.S.-,-, 131 S.Ct. 733, 740, 178 L.Ed.2d 649 (2011) (same); Yarborough v. Gentry, 540 U.S. 1, 6, 124 S.Ct. 1, 4, 157 L.Ed.2d 1 (2003) (same). When § 2254(d) applies to a Strickland claim, “the question is not whether counsel’s actions were reasonable. The question is whether there is any reasonable argument that counsel satisfied Strickland’s deferential standard.” Harrington, 131 S.Ct. at 788.
Notwithstanding double deference under AEDPA and Strickland, “[t]he writ of ha-beas corpus stands as a safeguard against imprisonment of those held in violation of the law.” Id. at 780. Even under AED-PA, we “must be vigilant and independent in reviewing petitions for the writ.” Id. Indeed, post-AEDPA, the Supreme Court and this Court have each granted habeas relief based upon penalty phase ineffective assistance of counsel claims. See, e.g., Porter v. McCollum, — U.S.-,-, 130 S.Ct. 447, 455-56, 175 L.Ed.2d 398 (2009) (granting habeas relief under AED-PA based on Strickland); Rompilla v. Beard, 545 U.S. 374, 380-81, 393, 125 S.Ct. *12532456, 2462, 2469, 162 L.Ed.2d 360 (2005) (same); Wiggins v. Smith, 539 U.S. 510, 520, 534-38, 123 S.Ct. 2527, 2534, 2542-44, 156 L.Ed.2d 471 (2003) (same); Williams, 529 U.S. at 367, 120 S.Ct. at 1499 (same); Cooper v. Sec’y, Dep’t of Corr., 646 F.3d 1328, 1356 (11th Cir.2011) (same); Johnson, 643 F.3d at 910-11 (same); Ferrell v. Hall, 640 F.3d 1199, 1203 (11th Cir.2011); Williams, 542 F.3d at 1345 (same).
III. DISCUSSION
Evans contends that he was denied his Sixth Amendment right to the effective assistance of counsel when his trial lawyer failed to investigate and to present substantial mitigating evidence. Thus, the merits of Evans’s claim are squarely governed by the Supreme Court’s holding in Strickland. See Williams, 529 U.S. at 390-91, 120 S.Ct. at 1511-12 (holding that Strickland “clearly established” Supreme Court precedent for evaluating ineffective assistance of counsel claims under AED-PA). We have summarized the Strickland standard as follows:
Under Strickland [a petitioner] must make two showings. First, he must show that his counsel’s performance was deficient, which means that it “fell below an objective standard of reasonableness” and was “outside the wide range of professionally competent assistance.” In deciding whether there was deficient performance, courts must review counsel’s actions in a “highly deferential” manner and “must indulge a strong presumption that counsel’s conduct falls within the wide range of reasonable professional assistance.” To overcome Strickland’s presumption of reasonableness, [a petitioner] must show that “no competent counsel would have taken the action that his counsel did take.”
The second showing required under Strickland is prejudice: the petitioner must also show that, but for his counsel’s deficient performance, there is a reasonable probability that the result of the proceeding would have been different — that is, our confidence in the outcome must be undermined by counsel’s deficient performance....
Johnson, 643 F.3d at 928-29 (citations omitted). The question of prejudice differs somewhat between the guilt and penalty phases, as the Supreme Court explained in Strickland:
When a defendant challenges a conviction, the question is whether there is a reasonable probability that, absent the errors, the factfinder would have had a reasonable doubt respecting guilt. When a defendant challenges a death sentence ... the question is whether there is a reasonable probability that, absent the errors, the sentencer — including an appellate court, to the extent it independently reweighs the evidence — would have concluded that the balance of aggravating and mitigating circumstances did not warrant death.
Strickland, 466 U.S. at 695, 104 S.Ct. at 2068-69.
To show prejudice at either phase of a capital trial, “a defendant need not show that counsel’s deficient conduct more likely than not altered the outcome in the case.” Id. at 693, 104 S.Ct. at 2068. This is so because an ineffective assistance of counsel “claim asserts the absence of one of the crucial assurances that the result of the proceeding is reliable, so finality concerns are somewhat weaker and the appropriate standard of prejudice should be somewhat lower.” Id. at 694, 104 S.Ct. at 2068. “The result of a proceeding can be rendered unreliable, and hence the proceeding itself unfair, even if the errors of counsel cannot be shown by a preponderance of the evidence to have determined the outcome.” Id.
In making the prejudice determination, a reviewing court “must consider the *1254totality of the available mitigation evidence — both that adduced at trial, and the evidence adduced in the habeas proceeding — and reweigh it against the evidence in aggravation.” Porter v. McCollum, 130 S.Ct. at 453-54 (quotation marks omitted). In Sears v. Upton, — U.S.-, 130 S.Ct. 3259, 177 L.Ed.2d 1025 (2010), the Supreme Court recognized that, even where some mitigation evidence was presented by counsel, it has:
never held that counsel’s effort to present some mitigation evidence should foreclose an inquiry into whether a facially deficient mitigation investigation might have prejudiced the defendant. To the contrary, we have consistently explained that the Strickland inquiry requires precisely the type of probing and fact-specific analysis that the state trial court failed to undertake below.
Id. at 3266. Taken together, Porter and Sears make clear that Strickland requires a “probing and fact-specific analysis” in evaluating the totality of the available evidence. Id. The Supreme Court noted that such a prejudice inquiry “will necessarily require a court to ‘speculate’ as to the effect of the new evidence....” Id. With these general principles in mind, we turn to Evans’s claim of penalty phase ineffective assistance of counsel.
A. Deficient Performance
As a threshold matter, we reject Evans’s argument that we should review Strickland’s deficiency prong de novo because the Florida Supreme Court based its decision only on Strickland’s prejudice prong. The state trial court rejected Evans’s penalty phase ineffective assistance of counsel claim after it determined that he had not shown deficient performance or prejudice. While it is true that the Florida Supreme Court, having determined that Evans was not prejudiced by counsel’s failure to present mitigating evidence, declined to address whether counsel’s penalty phase investigation was deficient, it did not disapprove of the trial court’s conclusion regarding deficiency. See Evans II, 946 So.2d at 12. Thus, both Strickland prongs are due AEDPA deference. See Hammond v. Hall, 586 F.3d 1289, 1332 (11th Cir.2009) (holding that, when “a state trial court rejects a claim on one prong of the ineffective assistance of counsel test and the state supreme court, without disapproving that holding, affirms on the other prong, both of those state court decisions are due AEDPA deference”). Nevertheless, we do not view this case as presenting a close question about whether the trial court’s decision on deficiency was objectively reasonable. In light of trial counsel’s failure to discharge his constitutional duty to conduct a thorough background investigation — which was well- and long-established at the time of Evans’s 1999 trial — no fairminded jurist could agree with the state court’s conclusion that Evans’s counsel was not deficient.
We begin our AEDPA analysis of Strickland’s deficiency prong with the state trial court because the Florida Supreme Court declined to address Strickland’s performance prong. See Hammond, 586 F.3d at 1332. The trial court correctly identified Strickland as the governing standard and concluded that trial counsel’s actions in investigating mental health mitigation were not deficient. In doing so, the trial court determined that counsel “conducted a reasonable investigation, under the circumstances.” Specifically, the trial court relied upon the following: (1) that trial counsel testified at the evidentiary hearing that he did not think Evans was “crazy”; (2) that Evans had told trial counsel the shooting was an accident and that he was a good student; (3) that trial counsel was aware of Evans’s violent background; and (4) that trial counsel had “sent letters [to] various people, asking them if they would be character witnesses for the defense.” In *1255addition, as part of his penalty phase investigation, counsel had spoken with Evans, his mother, and an aunt. Counsel was also aware of a presentence investigation in which Evans stated his mental health was perfect. It fairly appears from the trial court’s order, as well as our review of all the evidentiary hearing testimony, that trial counsel focused his investigation on developing evidence of Evans’s “redeeming qualities.” Moreover, the trial court found that counsel “conducted a reasonable investigation into [Evans’s] background and made a strategic decision to focus on the positive aspects of [Evans’s] character — not open the door to all the prior acts of violence perpetrated by [Evans] since his teenage years.”
Even under the double deference accorded to trial counsel’s performance under AEDPA and Strickland, we hold that the state trial court’s determination that counsel was not deficient is either contrary to, or involved an unreasonable application of, clearly established federal law. In this case, trial counsel’s failure to conduct a constitutionally adequate background investigation meant that he never knew about significant mitigating evidence, including Evans’s closedhead injury at age three, resulting brain damage, speech and language difficulties, learning disabilities, and impulse control problems, all of which were documented by school records and school psychologists when he was seven years old. The state trial court’s contrary conclusion unreasonably discounts counsel’s duty to conduct a thorough background investigation as recognized by the Supreme Court and then-prevailing professional norms.
As early as 1986, and long before Evans’s 1999 trial, the Supreme Court clearly established that counsel in capital cases have an “obligation to conduct a thorough investigation of the defendant’s background.” Williams, 529 U.S. at 368-70, 120 S.Ct. at 1500-01 (finding counsel performed deficiently by failing to conduct constitutionally adequate penalty phase investigation in 1986 trial); see also Porter v. McCollum, 130 S.Ct. at 448-49, 452-53 (same except 1988 trial); Rompilla, 545 U.S. at 378, 387-89, 125 S.Ct. 2456 (same); Wiggins, 539 U.S. at 515-16, 524-29, 123 S.Ct. at 2532, 2536-39 (same except 1989 trial); see also Ferrell, 640 F.3d at 1234 (holding that the state court unreasonably applied Strickland when it did not find ineffective assistance in connection with a 1988 trial where counsel, among other things, “failed to investigate [the defendant’s] upbringing ... which would have uncovered evidence [of his] impoverished and abused childhood”). The Court has further instructed that under Strickland, “our principal concern ... is not whether counsel should have presented” mitigating evidence, but rather “whether the investigation supporting counsel’s decision not to introduce mitigating evidence of [the defendant’s] background was itself reasonable.” Wiggins, 539 U.S. at 522-23, 123 S.Ct. at 2536 (emphasis omitted). Of course, we acknowledge that “the Strickland test of necessity requires a case-by-case examination of the evidence,” Cullen, 131 S.Ct. at 1407 n. 17 (quotation marks omitted), and that “Strickland itself rejected the notion that the same investigation will be required in every case.” Id. at 1406-07. Thus, “in each case we must determine whether counsel conducted a reasonable background investigation” or “made a reasonable decision that made conducting a background investigation unnecessary.” Johnson, 643 F.3d at 931 (quotation marks omitted). As we have explained:
The question under Strickland is not whether ... trial counsel’s overall performance at the sentence stage was exemplary or even average, but whether he conducted an adequate background investigation or reasonably decided to *1256end the background investigation when he did. See Ferrell, 640 F.3d at 1226 (“ ‘[Strategic choices made after thorough investigation of law and facts relevant to plausible options are virtually unchallengeable,’ but those made after ‘less than complete investigation’ are reasonable only to the extent that reasonable professional judgment supports the limitations on investigation.”) (quoting Strickland, 466 U.S. at 690-691, 104 S.Ct. at 2066).
Id. at 931-32.
Further, although a specific mitigation “theory might be reasonable, in the abstract, [that fact] does not obviate the need to analyze whether counsel’s failure to conduct an adequate mitigation investigation before arriving at this particular theory prejudiced” the defendant. Sears, 130 S.Ct. at 3265. “The ‘reasonableness’ of counsel’s theory [is] beside the point [because a defendant] might be prejudiced by his counsel’s failures, whether his haphazard choice was reasonable or not.” Id. As the Supreme Court reiterated in Sears:
This point is plain in Williams: We rejected any suggestion that a decision to focus on one potentially reasonable trial strategy — in that case, petitioner’s voluntary confession — was “justified by a tactical decision” when “counsel did not fulfill their obligation to conduct a thorough investigation of the defendant’s background.” 529 U.S. at 396 [,120 S.Ct. at 1515]. A “tactical decision” is a precursor to concluding that counsel has developed a “reasonable” mitigation theory in a particular case.
Id.; see also Wiggins, 539 U.S. at 522, 123 S.Ct. at 2535 (explaining that “counsel’s failure to uncover and present voluminous mitigating evidence at sentencing could not be justified as a tactical decision ... because counsel had not ‘fulfilled] their obligation to conduct a thorough investigation of the defendant’s background’” (quoting Williams, 529 U.S. at 396, 120 S.Ct. at 1515)).
Here, trial counsel conducted limited family interviews, relying principally upon a short thirty-minute interview of Evans’s mother together with form letters sent to possible character witnesses. Counsel did not collect readily available basic background materials, such as Evans’s school and medical records. No competent counsel in 1999 would have failed to collect and review such information especially where, as here, the defendant was a lifelong resident of the county in which the trial occurred. Similarly, no competent counsel in 1999 would have conducted only a circumscribed interview of the defendant’s mother and solicited witnesses with form letters. See AJBA Guidelines for the Appointment and Performance of Counsel in Death Penalty Cases (1989),12 cited in Wiggins, 539 U.S. at 524, 123 S.Ct. at 2536-37. For example, ABA *1257Guideline § 11.4.1 from 1989 — reflecting prevailing professional norms ten years prior to Evans’s trial — provides that counsel’s “investigation should comprise efforts to discover all reasonably available mitigating evidence and evidence to rebut any aggravating evidence that may be introduced by the prosecutor.” ABA Guideline § 11.4.1(C).
Nor would competent counsel in 1999 fail to explore, at least preliminarily, all pertinent avenues of mitigation going beyond only good character evidence. The ABA guideline advised counsel to “explore the existence of other potential sources of information relating to the offense, the client’s mental state, and the presence or absence of any aggravating factors under the applicable death penalty statute and any mitigating factors.” Id. § 11.4.1(D)(2)(B).
In addition, the ABA guidelines recognize the critical importance of investigating the defendant’s background by collecting the defendant’s medical and school records. Id. § 11.4.1(D)(2)(C). Specifically, the ABA guidelines advise counsel to “[cjollect information relevant to the sentencing phase of trial including, but not limited to: medical history, (mental and physical illness or injury, alcohol and drug use, birth trauma and developmental delays); educational history (achievement, performance and behavior) special educational needs (including cognitive limitations and learning disabilities).... ” Id. (emphasis added). Because Evans’s trial counsel did not obtain any of his school or medical records, trial counsel clearly did not discharge this duty.
Neither did Evans’s counsel identify and adequately interview collateral witnesses. See id. § 11.4.1(D)(3)(B) (recommending that counsel should interview “witnesses familiar with aspects of the client’s life history that might affect the ... possible mitigating reasons for the offense(s), and/or other mitigating evidence to show why the client should not be sentenced to death”). It is plain from the state eviden-tiary hearing that trial counsel only sought to investigate positive character evidence. While it is true that counsel met with Evans’s mother Lilly prior to trial and asked her to supply the names of character witnesses to testify at the penalty phase, it is abundantly clear from the evidentiary hearing that counsel limited his investigation to, and was only interested in, positive character evidence.
Trial counsel’s interaction with Lilly Evans exemplifies the limited nature of his investigation and underscores the corresponding unreasonableness of his so-called strategic decision to curtail his mitigation investigation. Lilly testified that she met with trial counsel only once prior to trial, and then only for thirty minutes. During this brief meeting, counsel advised Lilly that he wanted her to collect as many character witnesses as possible because Evans could get the death penalty. By character witnesses, Lilly testified that she understood counsel wanted only “people that could say some good things about” Evans. She further testified that she related good character evidence to Evans’s counsel. According to Lilly, she never met with a defense investigator and never met again with Evans’s counsel until the day of trial. Lilly further testified that counsel never explored other issues about Evans’s background with her.
That counsel’s background investigation was limited to good character evidence is further demonstrated by the form letters he sent to penalty phase character witnesses identified by Lilly.13 During the *1258evidentiary hearing, trial counsel affirmed that he used these letters to prepare for the penalty phase. Further, counsel testified that the questions in the letter were intended to establish that Evans was a good person. Indeed, trial counsel confirmed that was his understanding of what “non-statutory mitigation was.”
Of course, the range of constitutionally relevant mitigating evidence is far wider than just good character evidence, which is what trial counsel understood it be limited to.14 In light of this misunderstanding of mitigation, it is not surprising that the form letters he sent to investigate mitigation did not explore the full range of constitutionally relevant background information.
Because counsel conducted no mitigation investigation beyond Evans’s good character, he failed to uncover Evans’s closed-head injury at age three, resulting brain damage, learning disabilities, emotional handicaps, and impulse control problems, all of which were significant enough at age seven to warrant psychological assessment by public school authorities. This further supports our conclusion that the state trial court’s failure to find counsel’s performance deficient was unreasonable under AEDPA. See Ferrell, 640 F.3d at 1229-30 (“Because Ferrell’s counsel failed to conduct a mitigation investigation beyond Ferrell’s character, they failed to uncover additional, obvious evidence of serious mental illness, further compounding our conclusion that the [state court’s] failure to find their mitigation investigation deficient was unreasonable.”).
Furthermore, counsel unreasonably failed to follow up on the limited background information he did have. For example, the Florida Supreme Court noted that attorney “[Counsel] testified that the presentence investigation reports (PSI) from Evans’[s] prior convictions indicated that his mental health was perfect and that he had only seen a mental heath expert when he was young.” Evans II, 946 So.2d at 9 (emphasis added). Thus, *1259while the state courts emphasized that Evans had reported “perfect” mental health in his PSI reports, the PSIs also reported that he had “seen a mental health expert when he was young.” Id. This information should have been a “red flag” to trial counsel, alerting him to the need to conduct a follow-up investigation. See Rompilla, 545 U.S. at 392, 125 S.Ct. at 2468-69. No competent counsel in 1999 would have failed to follow-up on such information. Prompted by this red flag, competent counsel would have then obtained and reviewed school and medical records, for example. And armed with these records, competent counsel would have investigated Evans’s mental health and consequently discovered his history of brain damage.
Moreover, the state trial court all but ignored counsel’s obligation to undertake a thorough and adequate investigation. The trial court side-stepped, and therefore unreasonably applied Wiggins and its progeny, by placing undue emphasis on trial counsel’s decision to portray Evans as a “good guy,” despite the undisputed evidence that trial counsel reached that decision without first collecting readily available school and medical records or adequately interviewing family members. Indeed, prior to trial, the only mitigation investigation counsel conducted was meeting twice with Evans for at total of approximately four hours,15 meeting once with Lilly Evans for about thirty minutes, and mailing form letters to potential character witnesses.
While the state trial court indicated that trial counsel’s penalty phase investigation also included speaking with Evans’s aunt, the record clearly indicates that Aunt Sandra Evans became a penalty phase witness by happenstance. Sandra testified during the evidentiary hearing that she became a character witness at the penalty phase only after she fortuitously attended closing arguments during the guilt phase. Although Sandra testified that she had lived in the same house as Evans and knew him his entire life, she had never been contacted by trial counsel about testifying prior to trial. Further, Sandra testified that counsel spoke with her less than five minutes just prior to her penalty phase testimony and that they did not discuss mitigation. Instead, Sandra testified that defense counsel told her to say only good things about Evans at the penalty phase. Had counsel explored the full range of mitigation with Sandra, instead of limiting his investigation to good character evidence, he would have discovered substantial mitigation. Sandra witnessed Evans being struck by a car at the age of three. She testified that after Evans came home from the hospital he “was just totally different.” In her words, “[h]e wasn’t like the same kid I used to know.” Sandra testified that counsel never asked her about these matters but she would have testified to these facts at the penalty phase had she been asked. Thus, trial counsel was deficient because he failed to adequately utilize the family members to whom he had spoken. See Ferrell, 640 F.3d at 1230-32 (finding that trial counsel’s performance was deficient, in part, because he failed to elicit significant and powerful mitigating evi*1260dence from witnesses who were willing to testify if counsel had asked them about the defendant’s background).
The state trial court’s conclusion that trial counsel made a strategic decision to present only good character evidence, despite counsel’s lack of investigation, smacks of the kind of “post-hoc rationalization” rejected in Wiggins. Wiggins, 539 U.S. at 526-527, 123 S.Ct. at 2538. Notably, on the deficiency prong, Evans’s counsel did far less to investigate mitigation than trial counsel in Wiggins. In Wiggins, trial counsel did not put on any mental health mitigation or life history evidence, such as evidence of petitioner’s physical and sexual abuse, because counsel decided before trial to focus their efforts on retrying the factual case and disputing Wiggins’s direct responsibility for the crime. See id. at 516-17, 123 S.Ct. at 2533. Although trial counsel in Wiggins retained a mental health expert who evaluated the defendant, counsel did not compile a life history with the assistance of a forensic social worker. Acknowledging these facts, the state court denied relief, stating, “when the decision not to investigate ... is a matter of trial tactics, there is no ineffective assistance of counsel.” Id. at 517-18, 123 S.Ct. at 2533 (quotation marks omitted).
The state appellate court affirmed the denial of relief, concluding that trial counsel had made “a deliberate, tactical decision to concentrate their effort at convincing the jury” that Wiggins was not directly responsible. Id. at 518, 123 S.Ct. at 2533 (quotation marks omitted). The state court specifically found that trial counsel “knew of Wiggins’[s] unfortunate childhood” because they had available to them the PSI, as well as “more detailed social service records that recorded incidences of physical and sexual abuse, an alcoholic mother, placements in foster care, and borderline mental retardation.” Id. (quotation marks and citations omitted). While the state court acknowledged that the information counsel had was not as graphic or as detailed as the information developed by a forensic social worker in the postcon-viction hearing, the state court found “counsel did investigate and were aware of appellant’s background.” Id. (quotation marks omitted). As a result, the state court concluded that counsel made a reasoned choice to proceed with what they perceived was their best defense. Id. at 518, 123 S.Ct. at 2533-34.
Notwithstanding the state court’s conclusion about the strategy of Wiggins’s counsel, the Supreme Court found that the state court unreasonably applied Strickland and made an unreasonable determination of the facts, within the meaning of AEDPA. Id. at 527, 123 S.Ct. at 2538. In doing so, the Court emphasized that: (1) “petitioner’s claim stems from counsel’s decision to limit the scope of their investigation into potential mitigating evidence;” and (2) that trial counsel attempted to justify their limited investigation as reflecting a tactical decision. Id. at 521, 123 S.Ct. at 2535. In rejecting the contention that trial counsel made a reasonable tactical decision, which is substantially the same as the argument advanced by the state in Evans’s case, the Supreme Court in Wiggins emphasized an aspect of Strickland that bears repeating:
[Strategic choices made after thorough investigation of law and facts relevant to plausible options are virtually unchallengeable; and strategic choices made after less than complete investigation are reasonable precisely to the extent that reasonable professional judgments support the limitations on investigation. In other words, counsel has a duty to make reasonable investigations or to make a reasonable decision that makes particular investigations unnecessary. In any ineffectiveness case, a particular *1261decision not to investigate must be directly assessed for reasonableness in all the circumstances, applying a heavy measure of deference to counsel’s judgments.
Id. at 521-22, 123 S.Ct. at 2535 (quoting Strickland, 466 U.S. at 690-91, 104 S.Ct. at 2066) (emphasis added). The Court in Wiggins ultimately concluded, “[w]hen viewed in this light, the ‘strategic decision’ the state courts and respondents all invoke to justify counsel’s limited pursuit of mitigating evidence resembles more a post hoc rationalization of counsel’s conduct than an accurate description of their deliberations prior to sentencing.” Id. at 526-527, 123 S.Ct. at 2538.
We believe the Supreme Court’s reasoning and conclusion apply equally to Evans’s case. Even assuming that defense counsel limited the scope of his investigation for strategic reasons, Wiggins make clear that “Strickland does not establish that a cursory investigation automatically justifies a tactical decision with respect to sentencing strategy.” Id. at 527, 123 S.Ct. at 2538. In Evans’s case, the state trial court assumed that, because counsel had obtained some information with respect to Evans’s background, (ie., talking to Evans and his mother and sending out form letters), trial counsel was in a position to make a tactical decision not to conduct further investigation and instead to present a good character defense. This was an unreasonable application of Strickland.
Further, the state trial court focused on whether trial counsel should “have presented [mitigation evidence],” instead of considering “whether the investigation supporting counsel’s decision not to introduce mitigating evidence of [the defendant’s] background was itself reasonable.” Id. at 523, 123 S.Ct. at 2536 (emphasis omitted). That the state trial court’s focus was misplaced is reflected throughout its order. For example, after summarizing the positive character mitigation that was presented at the penalty phase, the state trial court concluded that Evans “has not shown that his attorney was ineffective for failing to further investigate his mental health and to' present this information in mitigation.” The trial court then noted trial counsel was “aware of Evans’[s] prior criminal record :... [and] was also aware that [Evans] had stated his mental health was perfect.” The court went on to find that trial counsel made a “reasonable investigation into the Defendant’s background and made a strategic decision to focus on the positive aspects of the Defendant’s character — and not open the door to all the prior acts of violence perpetrated by the Defendant since his teenage years.” Moreover, the state trial court’s discussion of deficiency focuses almost entirely on whether trial counsel should have presented the mitigation he did not know of, instead of on counsel’s decision not to investigate. This is an unreasonable application of Strickland’s deficiency prong.
In sum, Evans’s trial counsel’s performance was deficient because he terminated his investigation too early, before having completed the kind of thorough investigation contemplated by Wiggins and then-prevailing professional norms. See Sears, 130 S.Ct. at 3264 (agreeing with state court’s determination that the “the cursory nature of counsel’s investigation into mitigation evidence — ‘limited to one day or less, talking to witnesses selected by [defendant’s] mother’ — was ‘on its face ... constitutionally inadequate’ ” (citation omitted)). Had trial counsel collected readily available medical records, or interviewed family members about • Evans’s medical or psychological history, he would have learned about the closedhead injury from the car accident. Similarly, had counsel obtained Evans’s school records or interviewed any of his special education teachers and counselors who lived in the *1262community, he would have also learned about the closedhead injury because it was documented in his school records along with his learning disability, emotional handicap, and impulse control problems. “Whether or not those omissions were sufficiently prejudicial to have affected the outcome of sentencing, they clearly demonstrate that trial counsel did not fulfill their obligation to conduct a thorough investigation of the defendant’s background.” Williams, 529 U.S. at 396, 120 S.Ct. at 1514-15.
Here, counsel’s decision to focus his investigation on Evans’s good character was not an informed decision based upon a constitutionally adequate investigation. As in Wiggins, and in breach of well-defined norms in existence at the time of Evans’s 1999 penalty phase, Evans’s trial counsel “abandoned [his] investigation of [Evans’s] background after having acquired only rudimentary knowledge of his history from a narrow set of sources.” Wiggins, 539 U.S. at 524, 123 S.Ct. 2527; see also Johnson, 643 F.3d at 934-35 (finding, under AEDPA, that trial counsel’s failure to adequately investigate defendant’s background was deficient performance and that state court’s contrary conclusion was an unreasonable application of Strickland); Williams, 542 F.3d at 1340-1342 (holding,, under AEDPA, that trial counsel was deficient when they tried to obtain firsthand knowledge about the defendant’s background only from the defendant’s mother and that state court’s contrary conclusion was an unreasonable application of Strickland).
Although counsel’s decision to focus on good character evidence might be “reasonable, in the abstract, [it] does not obviate the need to analyze whether counsel’s failure to conduct an adequate mitigation investigation before arriving at this particular theory prejudiced” the defendant. Sears, 130 S.Ct. at 3265. While we accept that the state trial court made a factual determination that trial counsel made a strategic decision to limit the scope of his investigation, the reasonableness of trial counsel’s decision is a question of law. Collier v. Turpin, 177 F.3d 1184, 1199 (11th Cir.1999) (quotation marks and citations omitted). For all these reasons, we hold that the state trial court’s conclusion that trial counsel made a reasonable strategic decision to limit his background investigation to good character evidence was contrary to, or involved an unreasonable application of, clearly established Supreme Court precedent. See generally, Porter v. McCollum, — U.S. -, 130 S.Ct. 447, 175 L.Ed.2d 398; Rompilla, 545 U.S. 374, 125 S.Ct. 2456; Wiggins, 539 U.S. 510, 123 S.Ct. 2527; Williams, 529 U.S. 362, 120 S.Ct. 1495.
B. Prejudice
Because the Florida Supreme Court adjudicated the merits of Evans’s ineffective assistance of counsel claim on Strickland’s prejudice prong, its decision as to prejudice is entitled to deference under AED-PA. See Hammond, 586 F.3d at 1332. Further, since the Florida Supreme Court correctly identified Strickland as the governing Supreme Court principle, we ask whether the state court’s adjudication of prejudice “involved an unreasonable application” of Strickland when it concluded that Evans was not prejudiced. See Williams, 529 U.S. at 413, 120 S.Ct. at 1523 (“Under the ‘unreasonable application’ clause, a federal habeas court may grant the writ if the state court identifies the correct governing legal principle from this Court’s decisions but unreasonably applies that principle to the facts of the prisoner’s case”). After carefully reviewing the Florida Supreme Court’s written opinion, we hold that it unreasonably applied Strickland. Specifically, Porter v. McCollum makes plain that a state court unreasonably applies Strickland’s preju*1263dice prong when it does “not consider or unreasonably discounts] the mitigation evidence adduced in the [state] postconviction hearing.” 130 S.Ct. at 454.
An understanding of the facts in Porter v. McCollum, especially the reasoning of the Florida Supreme Court with respect to its Strickland prejudice analysis in Porter v. State, 788 So.2d 917 (Fla.2001), is critical to understanding the Supreme Court’s holding in the case. In Porter v. State, the defendant was convicted of two counts of first-degree murder and sentenced to death on one of those counts. 788 So.2d at 919-20. After Porter’s conviction and death sentence became final on direct review, he filed a motion for postconviction relief in the state trial court alleging ineffective assistance of trial counsel under Strickland. Id. at 920. Porter claimed that his trial counsel was ineffective because he failed to investigate and present mitigating evidence during the penalty phase. Id. at 921. The state trial court conducted an evidentiary hearing and was presented with evidence describing Porter’s abusive childhood, his heroic military service and the trauma he suffered while serving, and his long-term substance abuse. Id. at 921-24. The evidence included conflicting testimony from the state’s and defendant’s mental health experts regarding two statutory mental-state mitigating circumstances: (1) substantial impairment, Fla. Stat. § 921.141(6)(b); and (2) extreme mental or emotional disturbance, id. at § 921.141(6)(f). Id. at 923. The state trial court found the defense expert’s testimony that the statutory mental-state mitigators were present to be speculative and not supported by the evidence, and credited the state expert’s contrary testimony. Id. The state trial court denied relief and the Florida Supreme Court affirmed, holding that the defendant failed to meet the prejudice requirement of Strickland. Id. at 925. In doing so, the Florida Supreme Court stated:
At the conclusion of the postconviction evidentiary hearing in this case, the trial court had before it two conflicting expert opinions over the existence of mitigation. Based upon our ease law, it was then for the trial court to resolve the conflict by the weight the trial court afforded one expert’s opinion as compared to the other. The trial court did this and resolved the conflict by determining that the greatest weight was to be afforded the State’s expert. We accept this finding by the trial court because it was based upon competent, substantial evidence.
Id. at 923.
After losing in the state courts, Porter then sought and was granted federal habe-as relief in the District Court based upon ineffective assistance of counsel at the penalty phase. See Porter v. Crosby, No. 6:03-cv-1465-Orl-31KRS, 2007 WL 1747316, *23-31 (M.D.Fla. June 18, 2007). But this Court reversed because it did not find anything unreasonable about the Florida Supreme Court’s decision in Porter v. Crosby, 840 So.2d 981 (Fla.2003). See Porter v. Att’y Gen., 552 F.3d 1260, 1272-75 (11th Cir.2008). The Supreme Court reversed this Court’s judgment in part and held that the Florida Supreme Court unreasonably applied Strickland’s prejudice prong. See Porter v. McCollum, 130 S.Ct. at 454-455. The Supreme Court stated its reasoning as follows:
The Florida Supreme Court’s decision that Porter was not prejudiced by his counsel’s failure to conduct a thorough— or even cursory — investigation is unreasonable. The Florida Supreme Court either did not consider or unreasonably discounted the mitigation evidence adduced in the postconviction hearing. Under Florida law, mental health evidence that does not rise to the level of establishing a statutory mitigating circumstance may nonetheless be consid*1264ered by the sentencing judge and jury as mitigating. Indeed, the Constitution requires that “the sentencer in capital cases must be permitted to consider any relevant mitigating factor.” Yet neither the postconviction trial court nor the Florida Supreme Court gave any consideration for the purpose of nonstatutory mitigation to [the defense expert’s] testimony regarding the existence of a brain abnormality and cognitive defects. While the State’s experts identified perceived problems with the tests that [the defense expert] used and the conclusions that he drew from them, it was not reasonable to discount entirely the effect that his testimony might have had on the jury or the sentencing judge.
Id. (emphasis added) (footnotes and citations omitted).
We believe that Porter v. McCollum compels the conclusion that the Florida Supreme Court, in Evans’s case, “did not consider or unreasonably discounted the non-statutory mitigation evidence adduced in the postconviction hearing.” Id. at 454. While we acknowledge that Dr. McClaren may have disagreed with the defense expert’s conclusions regarding the extent of Evans’s brain damage and the conclusions that were drawn from his medical and school records (i.e., the defense’s contention that it supported the existence of statutory mental-state mitigators), “it was not reasonable to discount entirely the effect that his testimony might have had on the jury or the sentencing judge.” Id. at 455.
Much of Evans’s mitigation evidence presented at the evidentiary hearing was not rebutted. All three experts agreed that Evans had been hit by a car, suffered brain damage, and that the evidence from his school records and teachers showed that he was emotionally handicapped, severely learning disabled, and had difficulty controlling his impulses long before the onset of his criminality and alcohol abuse at age twelve. Indeed, it is not disputed that Evans was hit by a car when he was three years old, stopped breathing, was rushed to the hospital, and was later diagnosed with a closedhead injury and probable concussion. It is not disputed that following the accident, Evans had problems with his speech and behavior. Nor is it disputed that, by age seven at the latest, Evans’s cognitive deficits and impulse control problems manifested themselves to such a degree as to warrant psychological testing by school authorities. As a result, Evans was thereafter classified as severely learning disabled, emotionally handicapped, and unable to adequately control his impulsive behavior.
Nevertheless, Dr. McClaren disagreed with the opinions of Evans’s experts that this evidence rose to the level of statutory mental-state mitigation. Specifically, Dr. McClaren questioned whether Evans’s adult conduct and criminality, particularly at the time of the offense, were caused by these factors. Dr. McClaren testified on this point at the evidentiary hearing as follows:
There certainly is evidence that [Evans] sustained a closed head injury as a young child and subsequently had difficulties with his behavior and learning in school which progressed to juvenile delinquency and adult criminality. While some minor degree of brain dysfunction is certainly a possibility given his history of head injury and emotional and behavioral problems as a child prior to the onset of criminalityf,] his behavior later in life appears to be more associated with a criminal lifestyle than with the effects of some disabling cognitive disorder of such a severity to cause impairments which would be a significant difficulty for him.
Evans II, 946 So.2d at 9 n. 16.
We are not persuaded by the District Court’s attempt to distinguish Porter v. *1265McCollum from Evans’s case. The District Court reasoned that, unlike Porter v. McCollum, the Florida Supreme Court “did address the new mitigating evidence in its opinion, and found that although the evidence established [Evans] suffered from mental health problems, the evidence also showed a history of escalating violence.” But in our view, this analysis misapprehends the Supreme Court’s holding in Porter v. McCollum.
In Porter v. McCollum, the Supreme Court explained that the state court’s prejudice analysis was unreasonable because it “either did not consider or unreasonably discounted the mitigation evidence adduced in the postconviction hearing.” 130 S.Ct. at 454. Given the context of this statement, we interpret this language to mean that a state court unreasonably applies Strickland’s prejudice prong when it fails to consider, or unreasonably discounts to irrelevance, significant mitigation that does not rise to the level of statutory mitigation and its probable effect on the sentencer.
In other words, the error recognized in Porter v. McCollum cannot be merely that the Florida Supreme Court literally failed to address or discuss the new mitigating evidence in its opinion. Even a cursory examination of the Florida Supreme Court’s written opinion in Porter v. State reveals that that court explicitly addressed Porter’s new mitigating evidence by reviewing the evidence adduced at the state court evidentiary hearing at length. See 788 So.2d at 921-25. Indeed, the Florida Supreme Court gave a detailed description of the mitigation presented by Porter at his state court evidentiary hearing. See id. at 922-25. Despite this lengthy exposition of the mitigation, however, the Supreme Court clearly held that the Florida Supreme Court’s application of Strickland’s prejudice analysis was unreasonable. See Porter v. McCollum, 130 S.Ct. at 454-55. This is because Porter v. McCollum stands for something different than the notion that a state court must mention all of the mitigation presented by a defendant in state postconviction proceedings in order to properly conduct Strickland’s prejudice analysis. Rather, Porter v. McCollum confirms that Strickland’s prejudice analysis must focus on the obligation of the judge and jury to consider mitigation, as well as the important role of the jury in the determination of life and death. This principle is reflected in three aspects of the Court’s analysis in Porter v. McCollum.
First, the Supreme Court in Porter v. McCollum emphasized the important role of the jury in Florida’s capital punishment statute in relation to Strickland’s prejudice standard:
Under Strickland, a defendant is prejudiced by his counsel’s deficient performance if “there is a reasonable probability that, but for counsel’s unprofessional errors, the result of the proceeding would have been different.” In Florida, the sentencing judge makes the determination as to the existence and weight of aggravating and mitigating circumstances and the punishment, Fla. Stat. § 921.141(3), but he must give the jury verdict of life or death “great weight.” Porter must show that but for his counsel’s deficiency, there is a reasonable probability he would have received a different sentence. To assess that probability, we consider “the totality of the available mitigation evidence — both that adduced at trial, and the evidence adduced in the habeas proceeding” — and “reweigfh] it against the evidence in aggravation.”
Id. at 453-54 (some citations omitted). Applying this standard, the Supreme *1266Court’s analysis emphasized the effect of trial counsel’s errors on the sentencing jury. See id. at 454 (“The judge and jury at Porter’s original sentencing heard almost nothing that would humanize Porter or allow them to accurately gauge his moral culpability.”); id. (“Had Porter’s counsel been effective, the judge and jury would have learned of the ‘kind of troubled history we have declared relevant to assessing a defendant’s moral culpability.’ ” (quoting Wiggins, at 535, 123 S.Ct. at 2542)); id. (stating, the judge and jury did not hear evidence which “might well have influenced the jury’s appraisal of [Porter’s] moral culpability” (quoting Williams, 529 U.S. at 398, 120 S.Ct. at 1515)); id. at 455 (“[T]he relevance of Porter’s extensive combat experience is not only that he served honorably ... but also that the jury might find mitigating the intense stress and mental and emotional toll that combat took on Porter.”)
Second, what is important under Strickland’s prejudice analysis is the jury’s appraisal of the evidence, including credibility.16 “Under Florida law, mental health evidence that does not rise to the level of establishing a statutory mitigating circumstance may nonetheless be considered by the sentencing judge and jury as mitigating.” Id. at 454 (citing Hoskins v. State, 965 So.2d 1, 17-18 (Fla.2007)). Also, “the Constitution requires that ‘the sentencer in capital cases must be permitted to consider any relevant mitigating factor,’ ” id. at 454-55 (quoting Eddings v. Oklahoma, 455 U.S. 104, 112, 102 S.Ct. 869, 875, 71 L.Ed.2d 1 (1982)), and give that mitigating evidence effect.17 As a *1267consequence, although a reviewing court may resolve a conflict in expert testimony regarding the existence of statutory mitigating evidence against a defendant, it may not fail to give, for example, “any consideration for the purpose of nonstatu-tory mitigation to [expert] testimony regarding the existence of a brain abnormality or cognitive deficit.” Id. at 455. Said another way, where there is a conflict in the evidence about the import of mitigating evidence that may require credibility determinations, it is “not reasonable to discount entirely the effect that [expert] testimony might have had on the jury.” Id.
Thus, a reviewing court applying Strickland’s prejudice standard must consider the jury’s perspective. The question is whether there is a reasonable probability of a different outcome from the sentences See Strickland, 466 U.S. at 695, 104 S.Ct. at 2068-69 (“[T]he 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.”). This necessarily includes considering the probability that a reasonable jury, after hearing conflicting expert testimony, “would have concluded that the balance of aggravating and mitigating circumstanced did not warrant death.” Id. at 695, 104 S.Ct. at 2069.
Third, and finally, while Porter v. McCollum clearly stands for the proposition that a defendant’s heroic or distinguished military service for this country is inherently a significant type of mitigating evidence, and therefore cannot be ignored in Strickland’s prejudice analysis, the reasoning of Porter v. McCollum is not limited to war heroes for several reasons. To begin, nothing in Porter v. McCollum itself indicates the Supreme Court meant for its holding to be so limited. Indeed, Porter v. McCollum illustrates how courts must apply Strickland’s prejudice standard. It did not break any new legal ground. Rather, it is an example of an unreasonable application of an old rule established by Strickland.
It is also true that the Court in Porter v. McCollum not only concerned itself with the Florida Supreme Court’s failure to consider and give effect to Porter’s military service, but was equally concerned with the fact that Porter’s jury never heard about other constitutionally relevant mitigating evidence, such as Porter’s “childhood history of physical abuse, and ... his brain abnormality, difficulty reading and writing, and limited schooling.” 130 S.Ct. at 454. The Court emphasized that “evidence about the defendant’s background and character is relevant because of the belief, long held by this society, that defendants who commit criminal acts that are attributable to a disadvantaged background ... may be less culpable.” Id. (alteration omitted) (quoting Penny v. Lynaugh, 492 U.S. 302, 319, 109 S.Ct. 2934, 2947, 106 L.Ed.2d 256 (1989), abrogated on other grounds by Atkins v. Virginia, 536 U.S. 304, 122 S.Ct. 2242, 153 L.Ed.2d 335 (2002)); id. at 448 (noting that Porter’s “commanding officer’s moving description of [Porter’s combat service] was only a fraction of the mitigating evidence that his counsel failed to discover or present during the penalty phase of his trial” (emphasis added)).
Also, independent of Porter’s heroic military service, the Court found “it was not reasonable to discount entirely the effect that [the defense expert’s] testimony might have had on the jury or the sentencing *1268judge,” notwithstanding the state court’s conclusion that the expert evidence did not rise to the level of statutory mitigation. Id. at 454.
And finally in this regard, after Porter v. McCollum was decided, the Supreme Court relied on it to support its reversal in Sears v. Upton. Sears is the most recent example of the Supreme Court’s finding “deficiency and prejudice in other cases in which counsel presented what could be described as a superficially reasonable mitigation theory.” 130 S.Ct. at 3266 (citing Porter v. McCollum). In Sears, the Supreme Court reversed the Georgia Supreme Court for misapplying Strickland’s prejudice inquiry and determining that the defendant was not prejudiced by trial counsel’s facially inadequate mitigation investigation. See id. at 3264-65. The mitigation that was developed during Sears’s state postconviction hearing did not include heroic military service. Rather, Sears’s unpresented mitigation evidence showed he was the victim of childhood abuse and, like Evans, struggled in school from a young age with substantial behavioral problems. See id. at 3262. Also just like Evans, by the time Sears reached high school he was “described as severely learning disabled and as severely behaviorally handicapped.” Id. at 3262.
Unlike the Florida courts in Evans’s case, this Court has the benefit of the Supreme Court’s decision in Porter v. McCollum to guide our analysis and understanding of Strickland’s prejudice inquiry. Again, we do not interpret Porter v. McCollum as creating new law but rather as illustrating how Strickland’s prejudice test must be properly applied. Porter v. McCollum compels the conclusion that the state court unreasonably applied Strickland’s prejudice prong in Evans’s case.
Like Porter, the state courts in Evans’s case failed to give any meaningful consideration to the unrebutted testimony of brain damage, learning disability, and emotional handicap by the experts and lay witnesses for “the purpose of nonstatuto-ry” mitigation. Even assuming the jury would have completely agreed with Dr. McClaren’s opinion that these conditions did not directly contribute to Evans’s inability to control his impulses at the time of the offense, and therefore establish statutory mental health mitigation, this evidence is relevant mitigation that could have been considered by the sentencing judge and jury. As a result, it should have been weighed by the Florida Supreme Court in its prejudice analysis. Yet, that court failed to do so because, as in Porter v. McCollum, it unreasonably discounted this evidence to “irrelevance” insofar as it assumed the evidence was more harmful than helpful. In doing so, the Florida Supreme Court ignored the possibility that competent counsel could have turned the negative evidence into support for his mitigation theory. That Evans was disruptive and otherwise a difficult student with escalating patterns of violence is consistent with evidence of brain damage and an impulse control disorder. Indeed, that was the testimony of Evans’s post-conviction mental health experts. Just as evidence that Porter went AWOL following combat was consistent with his theory of mitigation, so too was evidence of Evans’ disruptive behavior in school. See Porter v. McCollum, 130 S.Ct. at 455; cf. Sears, 130 S.Ct. at 3263 (“[T]he fact that Sears’ brother is a convicted drug dealer and user, and introduced Sears to a life of crime, actually would have been consistent with a mitigation theory portraying Sears as an individual with diminished judgment and reasoning skills, who may have desired to follow in the footsteps of an older brother who had shut him out of his life.” (citation omitted)). To conclude otherwise represents, like Porter, a “failure to engage” with what Evans went through as a result *1269of his closedhead injury, resulting brain damage, learning disabilities, and impulse control problems. Porter v. McCollum, 130 S.Ct. at 455.
We recognize that evidence of Evans’s teenage alcohol abuse and violence may have been a double edged sword, though only at a certain chronological point. But we cannot agree that Evans’s closedhead injury at the age of three, resulting brain damage, and academic and behavioral difficulties in school are more harmful than helpful, as the state courts concluded. Just the opposite is true. We emphasize that undisputed brain damage resulting from a traumatic brain injury is inherently mitigating. See, e.g., Jefferson v. Hall, 570 F.3d 1283, 1311-15 (11th Cir.2009) (Carnes, J., dissenting) (discussing the significance of a capital defendant’s head injury and brain damage in the context of a penalty phase ineffective assistance of counsel claim), cert, granted, judgment vacated sub. nom. by Jefferson v. Upton, — U.S. -, -, 130 S.Ct. 2217, 176 L.Ed.2d 1032 (2010).
We emphasize as well the early onset of Evans’s mental health problems: Evans’s difficulty in controlling his impulsive behavior, as well as his learning disabilities, were documented by school records and psychological testing no later than age seven. Such undisputed mitigation evidence must be given some credit in the Strickland prejudice analysis, notwithstanding Evans’s later teenage alcohol abuse and criminality. See, e.g., Cooper, 646 F.3d at 1355 (“We acknowledge that evidence of alcoholism and drug abuse is often a two-edged sword which can harm a capital defendant as easily as it can help him at sentencing. However, we credit [defendant’s] evidence of alcohol abuse beginning at age 11 as mitigation, as it was used as a way to escape his horrible background.” (quotation marks and citations omitted)); cf. Roper v. Simmons, 543 U.S. 551, 568-70, 125 S.Ct. 1183, 1194-95, 161 L.Ed.2d 1 (2005) (recognizing, generally, the differences between juvenile and adult offenders in terms of moral culpability); id. at 570, 125 S.Ct. at 1195 (“The susceptibility of juveniles to immature and irresponsible behavior means ‘their irresponsible conduct is not as morally reprehensible as that of an adult.’ ” (quoting Thompson v. Oklahoma, 487 U.S. 815, 835, 108 S.Ct. 2687, 2699, 101 L.Ed.2d 702 (1988) (plurality opinion))); id. (“[Juveniles’] vulnerability and comparative lack of control over their immediate surroundings mean juveniles have a greater claim than adults to be forgiven for failing to escape negative influences in their whole environment.”).
In sum, it was unreasonable for the Florida Supreme Court to discount Evans’s postconviction evidence of mitigation to irrelevance in its Strickland analysis. Its decision therefore is not entitled to AEDPA deference because it was an unreasonable application of Strickland’s prejudice test.
We conclude that Evans has satisfied Strickland’s demanding standard of deficiency and prejudice. His trial counsel presented no mental health mitigating evidence, statutory or otherwise. Trial counsel failed to investigate or obtain Evans’s school or medical records, and conducted unreasonably limited family interviews, despite the fact that Evans had lived in the community his whole life. Cf. Rompilla, 545 U.S. at 384-86, 390-93, 125 S.Ct. at 2464-65, 2467-69 (holding under AEDPA, that trial counsel was ineffective at the penalty phase where counsel failed to obtain or examine defendant’s prior conviction file, which was a public document and “readily available” at the courthouse where the defendant was tried).
As a result of counsel’s deficient performance, trial counsel was unaware of Evans’s closedhead injury, brain damage, impulse control disorder, and learning dis*1270abilities. There was ample nonstatutory mitigating evidence that counsel could have presented which would have painted a remarkably different picture of Evans, but that the jury never heard. Much of this evidence was not disputed by even the state’s expert. Further, it is not difficult to imagine that competent counsel could have used this evidence to argue for a life sentence, even if it did not rise to the level of statutory mitigation. For example, evidence of Evans’s brain damage and impulse control disorder, either alone or in combination with his alcohol use and abuse, are consistent with a theory of defense that the killing was a spontaneous act, and therefore not deserving of death. See Evans I, 838 So.2d at 1099 (Shaw, J., dissenting joined by Anstead, C.J., and Pariente, J.) (“The victim made the fatal mistake of laughing which incensed Evans to the point that he pulled out a gun and shot her in the chest. There is nothing in the record to show that this was a planned or premeditated killing. In fact, the record reflects that the shooting was a spontaneous response to what Evans obviously felt was inappropriate laughter.”).
In light of all this, we hold “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.” Strickland, 466 U.S. at 695, 104 S.Ct. at 2069. Had Evans’s jury been confronted with the considerable mitigating evidence that was presented during the postconviction proceedings, “there is a reasonable probability that it would have returned with a different sentence.” Wiggins, 539 U.S. at 536, 123 S.Ct. at 2543.
IV. CONCLUSION
For all the reasons above, we vacate the order of the District Court with respect to Evans’s claim of ineffective assistance of counsel at the penalty phase. Except for this claim, we affirm the remainder of the District Court’s order. This case is remanded to the District Court with instructions to grant the writ of habeas corpus as to Evans’s penalty phase ineffective assistance of counsel claim consistent with this opinion.
. Except as otherwise cited, the facts are taken from the opinions of the Florida Supreme Court: Evans v. State, 838 So.2d 1090 (Fla. 2002) (Evans I); Evans v. State, 946 So.2d 1 (Fla.2006) (Evans II).
. The trial court also sentenced Evans to life imprisonment for the kidnapping conviction and to 108.15 months in prison for the aggravated assault conviction.
. As Justice Shaw explained:
The victim made the fatal mistake of laughing which incensed Evans to the point that he pulled out a gun and shot her in the chest. There is nothing in the record to show that this was a planned or premeditated killing. In fact, the record reflects that the shooting was a spontaneous response to what Evans obviously felt was inappropriate laughter.
Evans I, 838 So.2d at 1099 (Shaw, J., dissenting joined by Anstead, C.J., and Pariente, J.).
. Dr. McClaren had been retained by the state as a rebuttal witness, but was called as a witness by the defense during its case-in-chief.
. The Florida Supreme Court stated: "Dr. Carpenter's ‘clinical impression’ or diagnosis of Evans was in accordance with Axis I of the Diagnostic and Statistical Manual of Mental Disorders (4th ed. 2000) (DSM IV) test: 'Axis I: Alcohol abuse; Impulse Disorder; Rage Reaction Secondary to Closedhead Injury.' ” Evans II, 946 So.2d at 8 n. 13.
. Dr. Dee’s report provided the following "summary impression” of Evans:
Looking at the lifelong histoiy that Evans has, with learning disabilities, a lack of behavioral and impulse control, which included frequent rages, often when they were frankly inexplicable to others, attentional problems, over activity, etc., would strongly seem to support the diagnosis of cerebral damage early in childhood, probably resulting from the accident [at] age 3.
Evans II, 946 So.2d at 8 n. 14.
.Florida also recognizes a third mental-state statutory mitigating circumstance, but that *1248mitigator is not relevant to Evans’s case. See Fla. Stat. § 921.141(6)(3) ("[t]he defendant acted under extreme duress or under the substantial domination of another person.”)
. Dr. McClaren offered a diagnosis of Evans under the DSM-IV that differed from that of Dr. Dee and Dr. Carpenter:
Dr. McClaren reported that Evans "appears to meet ... AXIS I criteria for Alcohol Abuse in Remission in a Controlled Environment and AXIS II diagnosis of Antisocial Personality Disorder.” Dr. McClaren further noted that Evans "probably meets the AXIS II diagnosis of Personality Disorder, Not Otherwise Specified with Narcissistic and Paranoid Traits.” Dr. McClaren also reported that he could not rule out the "possibility” that Evans "suffers from a Cognitive Disorder, Not Otherwise Specified, associated with the after affects of a childhood concussion and other blows to the head ... given his documented history of head injury as a child, as well as ... fights, football, and boxing.”
Evans II, 946 So.2d at 8 n. 16.
. Although trial counsel was aware that Evans had been a life long resident of the jurisdiction where the trial was to occur, counsel did not obtain Evans’s school or medical records.
. The Florida Supreme Court considered the following three ineffective assistance of counsel claims: (1) “trial counsel was ineffective during the guilt phase for failing to investigate, prepare, and present the defense of diminished capacity”; (2) "trial counsel was ineffective at the penalty phase for failing to adequately challenge the State’s case”; and (3) trial "counsel was ineffective during the penalty phase for failing to request jury instructions on the statutory mental health miti-gators based on evidence that Evans was drinking at the time of the murder.” Evans II, 946 So.2d at 6 (footnote omitted). Evans was only granted a certificate of appealability (COA) as to claim two [Doc. 23 at 83]; thus, we do not consider these other ineffective assistance of counsel issues. See Walls v. Buss, 658 F.3d 1274, 1278 n. 3 (11th Cir.2011) (stating we will not consider issues not included in the COA).
*1250We note that claim two had two parts: (1) “counsel was ineffective for failing to request the appointment of co-counsel to assist” with the penally phase; and (b) "counsel was ineffective for failing to adequately investigate Evans’[s] background and discover available mental health mitigation.” Evans II, 946 So.2d at 6. Because Evans does not argue the co-counsel issue, we consider it abandoned and will not address it. See United States v. Curtis, 380 F.3d 1308, 1310 (11th Cir.2004) (stating "issues not raised by a defendant in his initial brief on appeal are deemed waived”).
. We quote extensively from the Florida Supreme Court’s decision because its factfind-ing, arguments and theories must be the focus of our inquiry under AEDPA. The Supreme Court has made clear that "a habeas court must determine what arguments or theories supported ... the state’s court decision; and then it must ask whether it is possible fair-minded jurists could disagree that those arguments or theories are inconsistent with the holding in a prior decision of [the Supreme Court].” Harrington, 131 S.Ct. at 786; see also Johnson, 643 F.3d at 910.
. We recognize that the standard for effective assistance of counsel is necessarily a general one: “representation that does not fall 'below an objective standard of reasonableness’ in light of 'prevailing professional norms.’ ” Bobby v. Van Hook,-U.S.-, -, 130 S.Ct. 13, 16, 175 L.Ed.2d 255 (2009) (quoting Strickland, 466 U.S. at 687-88, 104 S.Ct. at 2064-65). "Restatements of professional standards ... can be useful as 'guides' to what reasonableness entails ... to the extent they describe the professional norms prevailing when the representation took place.” Id. Thus, we look to the 1989 ABA Guidelines for the Appointment and Performance of Counsel in Death Penalty Cases as a guide to what reasonably diligent counsel would do, not a set of inexorable commands of what every capital attorney must do. See id. In doing so, we recognize that the Supreme Court has relied upon ABA guidelines to support its conclusion, post-AEDPA, that counsel was ineffective in failing to investigate and present mitigating evidence. See, e.g., Rompilla, 545 U.S. at 387-89, 125 S.Ct. at 2465-67 (referencing 1982 ABA guidelines in context of 1988 trial); Wiggins, 539 U.S. at 522-25, 123 S.Ct. at 2535-37 (referencing *12571980, 1982, and 1989 ABA guidelines in context of 1989 trial); Williams, 529 U.S. at 396, 120 S.Ct. at 1514-15 (referencing 1980 ABA guidelines in context of 1980 trial).
. Those form letters stated:
*1258Dear [X]:
Your name and address was provided to me by Lilly Evans, as a possible character witness for her son, Wydell Jody Evans, with regard to charges of First Degree Murder.
Would you please indicate, in your own words, the following:
1. How long you have known Wydell Jody Evans;
2. How to [sic] you know Wydell Jody Evans — as a
a. friend;
b. family member;
c. co-worker;
d. employee;
f. employer.
3. In one or two sentences please indicate what you think of Wydell Jody Evans as a person and whether or not you feel he committed the crime for which he has been charged.
. See Lockett v. Ohio, 438 U.S. 586, 604-05, 98 S.Ct. 2954, 2965, 57 L.Ed.2d 973 (1978) (recognizing that Eighth and Fourteenth Amendments require that capital sentencer not be precluded from considering, "as a mitigating factor, any aspect of a defendant's character or record and any of the circumstances of the offense that the defendant proffers as a basis for a sentence less than death” (emphasis in original)); see also Tennard v. Dretke, 542 U.S. 274, 285-87, 124 S.Ct. 2562, 2570-72, 159 L.Ed.2d 384 (2004) (stating that "impaired intellectual functioning is inherently mitigating” and that defendant need not establish nexus between mental capacity and crime for evidence to be relevant to mitigation); Payne v. Tennessee, 501 U.S. 808, 822, 111 S.Ct. 2597, 2607, 115 L.Ed.2d 720 (1991) ("We have held that a State cannot preclude the sentencer from considering ‘any relevant mitigating evidence' that the defendant proffers in support of a sentence less than death.... [Virtually no limits are placed on the relevant mitigating evidence a capital defendant may introduce concerning his own circumstances” (quoting Eddings v. Oklahoma, 455 U.S. 104, 114, 102 S.Ct. 869, 877, 71 L.Ed.2d 1 (1982))).
. Trial counsel testified at the evidentiary hearing that he only met with Evans for a total of four hours between the time of his appointment and the first day of trial. He met with Evans once on December 23, 1998 for two hours and again just before trial on October 23, 1999 for another two hours. Although we do not categorically find that four hours is an insufficient amount of time to meet with a capital defendant for the purpose of preparing both phases of a capital trial, such limited contact between an attorney and an incarcerated capital defendant prior to trial supports our conclusion that counsel’s penalty phase investigation was unreasonable. This is especially true given trial counsel's other limited and cursory efforts to investigate and prepare for the penalty phase.
. The Supreme Court's decision in Kyles v. Whitley, 514 U.S. 419, 115 S.Ct. 1555, 131 L.Ed.2d 490 (1995), confirms the idea that the jury’s appraisal of credibility must be considered in the prejudice analysis. Kyles involved a prosecutor's duty to turn over favorable evidence to the accused under Brady v. Maryland, 373 U.S. 83, 83 S.Ct. 1194, 10 L.Ed.2d 215 (1963). In this context, Kyles rejected the notion that a habeas court should evaluate the “materiality” of withheld evidence exclusively through the eyes of the postconviction judge, by stating:
Justice Scalia [dissenting] suggests that we should 'gauge' Burns's credibility by observing that the state judge presiding over Kyles’s postconviction proceeding did not find Burns's testimony in that proceeding to be convincing, and by noting that Burns has since been convicted for killing Beanie. Of course, neither observation could possibly have affected the jury’s appraisal of Burns’s credibility at the time of Kyles's trials.
Kyles, 514 U.S. 419, 449 n. 19, 115 S.Ct. 1555, 1573 n. 19 (emphasis added). While Brady's standard is formulated somewhat differently than Strickland's standard, as the Supreme Court has recognized, Brady's materiality test was adopted from Strickland's prejudice standard. See United States v. Bagley, 473 U.S. 667, 682, 105 S.Ct. 3375, 3383, 87 L.Ed.2d 481 (1985) (adopting Strickland’s "reasonable probability test” for prejudice as Brady's materiality standard).
. It is black letter law that sentencing judges and juries in capital cases must give independent weight to evidence of the defendant's character, record, and background, as well as any circumstances of the offense that might justify a penalty less severe than death. See, e.g., Lockett, 438 U.S. at 605, 98 S.Ct. at 2965 (plurality opinion) (holding it unconstitutional to exclude evidence of defendant's age, minor role in offense, and absence of intent to cause death); see also Hitchcock v. Dugger, 481 U.S. 393, 398-99, 107 S.Ct. 1821, 1824, 95 L.Ed.2d 347 (1987) (reversing the Eleventh Circuit and holding Florida's death penalty scheme was unconstitutional to the extent it excluded evidence of defendant’s family background and capacity for rehabilitation); Abdul-Kabir v. Quarterman, 550 U.S. 233, 246, 127 S.Ct. 1654, 1664, 167 L.Ed.2d 585 (2007) (holding it unconstitutional to preclude jury from giving meaningful effect to all mitigating evidence that could provide basis for refusing to impose death penalty); Brewer v. Quarterman, 550 U.S. 286, 294-296, 127 S.Ct. 1706, 1713-14, 167 L.Ed.2d 622 (2007) (holding it unconstitutional to preclude jury from giving meaningful consideration to mitigating evidence of defendant's mental illness, substance *1267abuse, and troubled childhood); Skipper v. South Carolina, 476 U.S. 1, 8-9, 106 S.Ct. 1669, 1673, 90 L.Ed.2d 1 (1986) (holding it unconstitutional to exclude evidence of defendant's good behavior during post-arrest incarceration).