WILSON, Circuit Judge:
Derrick Anthony DeBruee, an inmate on Alabama’s death row, appeals the district court’s denial of his petition for a writ of habeas corpus under 28 U.S.C. § 2254 challenging his death sentence. DeBruee contends that he received ineffective assistance of counsel in violation of the Sixth Amendment to the United States Constitution in both the guilt and the penalty phases of his capital murder trial.
DeBruee was convicted of fatally shooting a customer during the robbery of an AutoZone store which he committed with five other men.1 DeBruee argues that his retained trial attorney, Erskine Mathis, was constitutionally ineffective in failing to cross-examine state witness LuJuan McCants, a co-participant in the robbery who identified DeBruee as the shooter at trial, with McCants’s earlier allegedly contradictory statements. He also argues that Mathis was ineffective in failing to investigate and present evidence about De-Bruce’s mental capacity and background during the penalty phase of his trial.2
I.
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). Because DeBruee 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 [1266]*1266must 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, 562 U.S. 86, -, 131 S.Ct. 770, 786, 178 L.Ed.2d 624 (2011)).
AJEDPA allows federal courts to grant habeas relief only if 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).
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 decision 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, we are mindful that “an unreasonable application of federal law is different from an incorrect application of federal law.” Id. at 410, 120 S.Ct. at 1522. “A state court must be granted a deference and latitude that are not in operation when the case involves review under the Strickland3 standard itself.” Richter, 131 S.Ct. at 785 (footnote added). As a result, “[a] state court’s determination that a claim lacks merit pre-eludes federal habeas relief so long as fairminded jurists could disagree on the correctness of the state court’s decision.” Id. at 786 (quotation marks omitted).
In addition, a state court’s factual determination is entitled to a presumption of correctness. 28 U.S.C. § 2254(e)(1). 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). Ineffective assistance of counsel claims present mixed questions of law and fact not entitled to a presumption of correctness. See Cade v. Haley, 222 F.3d 1298, 1302 (11th Cir.2000) (stating Strickland’s deficient performance and prejudice prongs “present mixed questions of law and fact reviewed de novo on appeal”). Under 28 U.S.C. § 2254(e)(1), we must presume the state court’s factual findings to be correct unless the petitioner rebuts that presumption by clear and convincing evidence. See id. § 2254(e)(1); Parker, 244 F.3d at 835-36. Although the Supreme Court has “not defined the precise relationship between § 2254(d)(2) and § 2254(e)(1),” Burt v. Titlow, — U.S. -, -, 134 S.Ct. 10, 15, 187 L.Ed.2d 348 (2013), it has emphasized “that a state-court factual determination is not unreasonable merely because the federal habeas court would have reached a different conclusion in the first instance.” Id. (quoting Wood v. Allen, 558 U.S. 290, 301, 130 S.Ct. 841, 849, 175 L.Ed.2d 738 (2010)).
If we determine that the state court’s adjudication of DeBruce’s Strickland claims was unreasonable under § 2254(d), we must then undertake a de novo review of the record. See McGahee v. Ala. Dep’t of Corr., 560 F.3d 1252, 1266 (11th Cir.2009).
[1267]*1267To establish that his counsel provided constitutionally ineffective assistance, DeBruce must show both that his counsel’s performance was deficient and that the deficiency prejudiced his defense. See Strickland, 466 U.S. 668, 687, 104 S.Ct. 2052, 2064. Deficient performance for purposes of the Sixth Amendment is representation which falls “below an objective standard of reasonableness” measured by prevailing professional norms. Id. at 688, 104 S.Ct. at 2064. Trial “counsel is strongly presumed to have rendered adequate assistance and made all significant decisions in the exercise of reasonable professional judgment.” Id. at 690, 104 S.Ct. at 2066. “To overcome' that presumption, a defendant must show that counsel failed to act reasonably considering all the circumstances.” Cullen v. Pinholster, — U.S. -, -, 131 S.Ct. 1388, 1403, 179 L.Ed.2d 557 (2011) (quotation marks and alterations omitted). Since it is DeBruce’s burden to overcome this presumption, “the absence of evidence cannot overcome the strong presumption that counsel’s conduct fell within the wide range of reasonable professional assistance.” Titlow, 134 S.Ct. at 17 (quotation marks and alterations omitted). Given the deference owed counsel under Strickland in combination with AEDPA, our review of counsel’s performance is “doubly deferential.” Knowles v. Mirzayance, 556 U.S. 111, 123, 129 S.Ct. 1411, 1420, 173 L.Ed.2d 251 (2009); see also Pinholster, 131 S.Ct. at 1403.
To establish that his counsel’s deficiency prejudiced his defense, DeBruce “must show that there is a reasonable probability that, but for [his] counsel’s unprofessional errors, the result of the proceeding would have been different.” Strickland, 466 U.S. at 694, 104 S.Ct. at 2068. A reasonable probability is one that sufficiently undermines confidence in the outcome. Id. But “[t]he likelihood of a different result must be substantial, not just conceivable.” Richter, 131 S.Ct. at 792. That said, 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.” Strickland, 466 U.S. at 693, 104 S.Ct. at 2068. The Supreme Court has explained 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 determination about whether there is a reasonable probability that DeBruce would have received a different sentence, a reviewing court must “consider the totality 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, 558 U.S. 30, 41, 130 S.Ct. 447, 453-54, 175 L.Ed.2d 398 (2009) (quotation marks and alteration omitted).
II.
First, we address DeBruce’s arguments pertaining to the guilt phase of his trial, then those pertaining to the sentencing phase.
A. Guilt-Phase Ineffective Assistance of Counsel
Testimony at trial established that five armed men were involved in the robbery. During DeBruce’s trial, McCants was the only witness who testified that DeBruce fired the gunshot that killed Douglas Bat-[1268]*1268tie, a customer in the store.4 According to McCants’s trial testimony, Battle entered the store while the robbery was in progress. McCants saw Battle and ordered him to get down on the floor. Battle threw his wallet at McCants, but did not get onto the ground. DeBruce approached Battle from behind and hit him with his pistol, causing Battle to fall to the ground. DeBruce and Battle then engaged in a verbal altercation. After all of the robbers other than DeBruce left the store, McCants heard a gunshot from inside, and then saw DeBruce run out. McCants testified that while DeBruce and McCants were riding in the same car escaping from the crime scene, DeBruce admitted to McCants that he shot Battle. McCants also testified that all of the robbers had agreed prior to the robbery that if they encountered resistance from anyone in the store, the lead robber, Charles Burton, “would take care of it.” Finally, McCants identified the handguns belonging to several of the robbers, and testified that the handgun DeBruce carried on the day of Battle’s murder was silver, while claiming that his own handgun was black.
DeBruce argues that these aspects of McCants’s trial testimony were contradicted by statements McCants made to police officers one week after the incident. During that interrogation, McCants initially denied that he knew who fired the fatal shot; denied that the robbers had agreed to use force against anyone who resisted the robbery; and stated that he and De-Bruce left the scene of the murder in separate cars, and that he did not speak to DeBruce after the murder. Further, McCants claimed during the interrogation that he had carried a silver handgun, the same color as the murder weapon, and that he could not remember the guns carried by any of the robbers other than by himself and DeBruce.
DeBruce argues that Mathis performed deficiently in failing to cross-examine McCants with the discrepancies between his interrogation statements and his trial testimony. The Alabama Court of Criminal Appeals concluded that Mathis’s cross-examination was not deficient because he impeached McCants with his plea agreement and advanced, via the testimony of DeBruce’s niece, the defense’s theory that McCants was the shooter. DeBruce, 890 So.2d at 1087.
We cannot say that the state court’s conclusion regarding Mathis’s performance was an unreasonable application of or contrary to Strickland. See 28 U.S.C. § 2254(d). In addition to the above-referenced statements, McCants did identify DeBruce as the shooter during his interrogation, albeit only after the police claimed that DeBruce and one of the other robbers had named McCants as the gunman. Further, the central aspects of McCants’s interrogation statements were supportive of his trial testimony. In particular, [1269]*1269McCants stated during both his interrogation and his trial testimony that he saw DeBruce attack Battle from behind, that DeBruce and Battle were in a verbal altercation, and that DeBruce was the only robber remaining in the store when the gunshot was fired.5
DeBruce’s counsel, Mathis, reviewed the interrogation video but did not use McCants’s interrogation statements to impeach him.6 Rather, the record reflects that Mathis viewed the interrogation as primarily corroborative of McCants’s trial testimony at the time of trial. When the prosecutor notified Mathis shortly before McCants testified that the State would use the interrogation statements to rehabilitate McCants’s credibility if impeached, Mathis insisted that he would not attempt to argue that McCants had fabricated his account of the robbery in order to avoid allowing the prosecution to introduce McCants’s interrogation statements.
Instead, Mathis introduced the testimony of DeBruce’s niece stating that McCants told her one week after the robbery that he, McCants, had committed the shooting, and cross-examined McCants about this statement and his plea agreement. Mathis used this testimony to argue that McCants was the shooter and that he had falsely accused DeBruce in exchange for a lesser sentence. The interrogation statements, which were consistent with McCants’s identification of DeBruce as the shooter, would have conflicted with Mathis’s attempt to portray McCants as the actual shooter through the testimony of DeBruce’s niece. Based on this record, we must reject DeBruce’s argument that the Alabama Court of Criminal Appeals’ conclusion that Mathis was deficient in failing to cross-examine McCants with the transcript of his interrogation was an unreasonable application of Strickland,.'7 See Strickland, 466 U.S. at 690-91, 104 S.Ct. at 2066 (discussing deference owed to counsel’s strategic decisions based on adequate investigation).
B. Penalty-Phase Ineffective Assistance
Also, DeBruce argues that Mathis was ineffective in failing to investigate sev[1270]*1270eral issues that, he contends, would have led Mathis to discover mitigating evidence that was never presented at his sentencing, including that DeBruce experienced violence in his neighborhood and severe abuse at the hands of his sister, that he dropped out of school in the seventh grade, suffers from mental impairments and low intellectual functioning, and has a painful intestinal disorder.
To prepare for the sentencing phase of DeBruce’s trial, Mathis spoke to two people: DeBruce and his mother, Jessie De-Bruce. At sentencing, Mathis presented one witness, Jessie DeBruce, who testified that DeBruce completed high school, attended college at the University of Alabama, and that although he had an impoverished childhood, it was otherwise unremarkable. She also made passing mention of DeBruce’s treatment for a mental disorder, although this statement was not accompanied by any explanation of the disorder or its effects. DeBruce was also permitted to make a plea to the jury to spare his life.
DeBruce argues that had Mathis conducted a reasonably adequate mitigation investigation the jury should have heard the evidence presented at his state collateral hearing, which consisted of the testimony of three expert medical witnesses and two of DeBruce’s sisters. Among the medical experts, two psychologists presented testimony that DeBruce suffers from lingering emotional damage and social impairments associated with having been raised in a violent community, and that he dropped out of school at the age of sixteen having advanced only to the seventh grade. Further, the psychologists testified that DeBruce suffers from brain damage and that his intelligence is in the borderline range, with IQ scores between 76 and 79, which was not disputed by the psychologist who testified for the state at the state collateral hearing.8 They also testified that DeBruce suffers from blackout episodes consistent with seizures accompanied by periods of non-responsive staring and loss of memory. One of the psychologists noted that DeBruce was raised in an impoverished family of eleven children with an alcoholic father who was verbally abusive and uninvolved in his chil-drens’ upbringing. DeBruce was frequently attacked by gangs in his housing project, often on the way to and from school. DeBruce developed substance abuse problems during his teenage years and was pressured to join gangs. Finally, a gastroenterologist testified that DeBruce was diagnosed with an intestinal disease that causes severe abdominal pain and frequent vomiting, although the diagnosis was not made until after DeBruce’s trial and conviction.
Also during the state collateral hearing, DeBruce’s sisters testified that DeBruce witnessed stabbings, shootings, and other violence in the housing project where he grew up, and that DeBruce was frequently attacked by gangs. His sister Jacqueline testified that DeBruce was very close to his older brother, who was incarcerated when DeBruce was eight years old, and that DeBruce had little relationship with his father. Jacqueline also testified that DeBruce’s older sister, Juanita, was primarily responsible for raising DeBruce and the other children, and that Juanita disciplined DeBruce by beating him and withholding meals. Juanita beat DeBruce with switches and extension cords, and [1271]*1271even threatened DeBruce and the other children with a knife. These beatings were often followed by leaving DeBruce in a closet for several hours. Jacqueline explained that sometimes the beatings would occur because DeBruce did not obey an instruction, and at other times Juanita would beat DeBruce because he did not understand his homework. Jacqueline testified that these beatings occurred virtually every day. DeBruce did not resist the beatings but complied with whatever Juanita told him to do, including remaining in the closet until she told him he could leave. As a child, DeBruce frequently cried himself to sleep, and drew pictures of Juanita holding a knife in her hand. Jacqueline also testified that DeBruce suffered from chronic health problems, including severe stomach pain and frequent seizures that paralyzed him for five to ten minutes at a time and that required him to be hospitalized on several occasions. DeBruce’s sister Linda also testified that DeBruce took care of her after she had a stroke at the age of twenty-six that rendered her unable to walk, bathe, put on clothes, or feed herself. Both Jaqueline and Linda testified that they were not contacted by Mathis before DeBruce’s trial.
In assessing the reasonableness of Mathis’s decision to limit his mitigation investigation to speaking with DeBruce and his mother and not to further investigate DeBruce’s mental health and background, we analyze whether Mathis “conducted an adequate background investigation or reasonably decided to end the background investigation when [he] did.” Cooper v. Sec’y, 646 F.3d 1328, 1351 (11th Cir.2011) (internal quotation marks omitted); see Wiggins v. Smith, 539 U.S. 510, 533, 123 S.Ct. 2527, 2541, 156 L.Ed.2d 471 (2003) (“A decision not to investigate ... must be directly assessed for reasonableness in all the circumstances.” (internal quotation marks omitted)); see also Strickland, 466 U.S. at 690-91, 104 S.Ct. at 2066 (“[Strategic choices made after less than complete investigation are reasonable precisely to the extent that reasonable professional judgments support 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.”) DeBruce argues that a pre-trial report created by a social worker at Mathis’s request, which assessed De-Bruce’s competence to stand trial, should have alerted Mathis of the need to make further inquiry into DeBruce’s mental health and background.9 The report conveyed that DeBruce, who was then twenty years old, had attempted suicide on four occasions. He had refused special education, and had dropped out of school at the age of sixteen, having advanced only to the seventh grade. The report stated that DeBruce’s intelligence was in the “low average” range, but he denied receiving professional treatment or having been hospitalized for psyehiat-[1272]*1272ric reasons. The report also noted that DeBruce “has a history of problems since childhood including marijuana and alcohol abuse.”
Mathis testified during the state collateral hearing that he took on DeBruce’s case within three to four weeks of trial.10 During the limited time that Mathis had to prepare for DeBruce’s trial, he maintained his involvement in other cases that were being tried in the same period, and did not hire an investigator because he did not have the funds to pay for one. Mathis expressly denied that he had any strategic reason not to conduct a background investigation or to follow up on information in the competency report, but that he did not do so because “we just didn’t have time.” After DeBruce’s family retained Mathis, Mathis associated with William DelGrosso to assist in the defense. Mathis testified that DelGrosso began his involvement in the case even later than Mathis and that DelGrosso’s assistance in investigating for either the guilt or penalty phases consisted of helping Mathis sort through a banker’s box of discovery material. The Alabama Court of Criminal Appeals noted that “DelGrosso did not testify or execute an affidavit for purposes of the postconviction proceedings. We do not know the extent of his involvement in the case or what investigation he conducted in preparation for trial. Neither was Mathis questioned about what investigations that he and DelGrosso conducted.” DeBruce, 890 So.2d at 1085.
While it is true that DelGrosso did not testify at the state postconviction hearing, the state court record is not silent about DelGrosso’s involvement in this case or about the investigation conducted by Mathis and DelGrosso. During the state postconviction hearing, Mathis was specifically questioned about whether he used the services of a social worker, investigator or other specialist to help him investigate potential mitigation. Mathis explained that he typically used an investigator or subpoenas to help him collect background records, such as school and medical records. And Mathis usually relied on that same investigator to interview siblings, pastors, teachers, and other mitigation witnesses. But in DeBruce’s case, Mathis did not work with an investigator to help him do this critical investigation work because there was insufficient time. Nor did Mathis himself have time to do the work the investigator would have done. As a result, he testified “[w]e didn’t do it.” And there can be no doubt that Mathis was not only speaking for himself, but also testifying about the work DelGrosso did on the case when he Mathis said “[w]e didn’t do it.” This is so because the record is clear that the only person Mathis had helping him was DelGrosso. Significantly, Mathis repeatedly denied having any strategic reason for not doing the mitigation work that an investigator would have performed. Rather, Mathis offered a simple explanation: “We got in it too late.” This post-conviction evidence, combined with the minimal mitigation evidence actually presented during the original sentencing proceeding, suggests trial counsel’s failure to investigate and present available and compelling mitigating evidence was the result of inattention, not reasonable professional judgment.
[1273]*1273Moreover, our review of the entire state court record leads us to conclude that Mathis was lead counsel and DelGrosso played only a minor role in the investigation, preparation and presentation of De-Bruce’s case. This conclusion is supported by the following facts in the state court record: (1) Mathis was retained by De-Bruce’s mother; (2) Mathis testified that DelGrosso entered the case much later and only after Mathis “realized how big a task [he] had and asked [DelGrosso]” for help; (3) DelGrosso entered his notice of appearance on January 28, 1992, less than two weeks before the-trial started; (4) Mathis’s name alone appears on all of DeBruce’s pretrial motions; (5) Mathis, not DelGros-so, cross examined all of the state’s guilt phase witnesses except for one; (6) Mathis presented DeBruce’s case-in-chief during the guilt phase; (7) Mathis presented the testimony of the only penalty phase witness; (8) Mathis alone appeared on De-Bruce’s behalf at the sentencing hearing before the trial judge; and (9) Mathis alone appeared on DeBruce’s behalf at the motion for new trial hearing.
Despite all of this, the Alabama Court of Criminal Appeals concluded that Mathis did not fail to conduct a reasonable investigation because Mathis testified during the state collateral hearing that the information he received did not lead him to question DeBruce’s competence to stand trial or to consider defending DeBruce based on a lack of mental capacity.11 However, DeBruce’s fitness to be tried and decisions about whether to present a mental health defense during the guilt phase are separate issues from the decision whether to investigate and present mitigating evidence during sentencing. See Williams v. Taylor, 529 U.S. 362, 398, 120 S.Ct. 1495, 1516, 146 L.Ed.2d 389 (2000) (noting that evidence of defendant’s mental capacity and abusive background “may alter the jury’s selection of penalty, even if it does not undermine or rebut the prosecution’s death-eligibility case”); Hardwick v. Crosby, 320 F.3d 1127, 1163 (11th Cir.2003) (“Regarding mental health mitigating evidence, our court has distinguished between its use during the guilt phase to establish competency to stand trial and presenting mental health mitigating evidence at the penalty phase.”); Blanco v. Singletary, 943 F.2d 1477, 1503 (11th Cir.1991) (“[T]here is a great difference between failing to present evidence sufficient to establish incompetency at trial and failing to pursue mental health mitigating evidence at all. One can be competent to stand trial and yet suffer from mental health problems that the sentencing jury and judge should have had an opportunity to consider.”). The Alabama Court of Criminal Appeals’ determination that Mathis acted strategically in failing to conduct a mitigation investigation is based on its erroneous conflation of the issues of guilt-phase mental health defenses and competence to stand trial with the separate issue of whether to conduct a mitigation investigation. We have held that questions regarding whether an attorney’s decision is “strategic” or “tactical” is a question of fact, and whether that decision is reasonable is a question of law. See Hardwick, 320 F.3d at 1163. Moreover, a strategic decision cannot be reasonable if it is based on a misunderstanding of the law. Id. Here, even if we accept the state court’s factual determination that Mathis made a strategic decision not to investigate mitigation evidence based on the results of the pre-trial report governing DeBruce’s competency to stand trial, that decision [1274]*1274could not have been reasonable as it would have been based on a failure to understand the law. See Williams, 529 U.S. at 398, 120 S.Ct. at 1515; Hardwick, 320 F.3d at 1163; Blanco, 943 F.2d at 1503.12 Because no lawyer could reasonably have made a strategic decision to forego the pursuit of mitigation evidence based on the results of the pre-trial report governing competency to stand trial, the Alabama Court of Criminal Appeals’ conclusion to the contrary constitutes an unreasonable application of Strickland’s performance prong.
Moreover, Mathis’s decision not to follow up on the information in the competency report was all the more unreasonable insofar as the report contradicted what Mathis had learned about DeBruce’s background from DeBruce’s mother and from DeBruce’s youthful offender investigation report. While DeBruce’s mother testified at the sentencing phase of De-Bruce’s trial that her son had completed high school and attended the University of Alabama, the competency report and DeBruce’s school records clearly indicated that he had dropped out of school in the seventh grade at the age of sixteen after being held back several times. The youthful offender investigation report filed on January 7, 1992, also reflected that DeBruce had reported to the probation officer that he was in a “gifted” program in high school and that he had attended one semester of college at the University of Alabama. Although the information furnished by the competency report obviously conflicted with this account of De-Bruce’s background, Mathis never confronted DeBruce or his mother with these inconsistencies in order to discover the truth, nor did he follow up on the competency report’s references to DeBruce’s four attempts to commit suicide before the age of twenty. As a result, Mathis permitted DeBruce’s mother to present grossly inaccurate testimony during her son’s sentencing hearing that DeBruce had been a successful student and had attended the University of Alabama. See Williams v. Allen, 542 F.3d 1326, 1340 (11th Cir.2008) (holding counsel’s investigation deficient where “counsel obtained an incomplete and misleading understanding of [the defendant’s] life history” by relying upon the defendant’s mother’s account despite the availability of several other family members). See also Porter v. McCollum, 558 U.S. 30, 40, 130 S.Ct. 447, 175 L.Ed.2d 398 (2009) (per curiam) (holding that counsel performed deficiently by failing to investigate mitigating mental health information presented in the defendant’s competency report, and that this deficiency could not be excused by defendant’s “fatalistic and uncooperative” representations that discouraged counsel from investigating in this area); Wiggins, 539 U.S. at 523-24, 123 S.Ct. at 2536 (holding counsel’s performance deficient where counsel failed to follow up on references in the defendant’s pre-sentence report to the defendant’s “misery as a youth” and his own description of his background as “disgusting”).
By failing to take even the first step of asking DeBruce and his mother about the information in the report, Mathis “chose to abandon [his] investigation at an unreasonable juncture, making a fully informed decision with respect to sentencing strategy impossible.” Wiggins, 539 U.S. at 527-28, 123 S.Ct. at 2538. As we have held before, “[a]lthough counsel need not investigate every evidentiary lead, he must gather [1275]*1275enough knowledge of the potential mitigation evidence to arrive at an informed judgment in making that decision.” Jackson v. Herring, 42 F.3d 1350, 1367 (11th Cir.1995) (internal quotation marks omitted); see Williams, 542 F.3d at 1340 (holding counsel deficient for ignoring “red flags [that] would have prompted a reasonable attorney to conduct additional investigation”); Fortenberry v. Haley, 297 F.3d 1213, 1230 (11th Cir.2002) (Per Curiam) (“Absent any viable strategic reason, however, the failure to present available mitigating evidence renders assistance constitutionally ineffective.”).
Here, rather than taking the simple step of questioning DeBruce and his mother about the inconsistency between his mother’s account of DeBruce’s childhood and the information in the social worker’s pretrial report, trial counsel unreasonably ignored this disparity, failing to make an informed choice not to investigate De-Bruce’s mental health and background. Furthermore, counsel’s testimony during the state collateral proceedings made clear that his failure to conduct adequate investigation was not the result of a strategic decision, but rather a result of the lack of time as he scrambled to prepare for DeBruce’s capital trial with less than four weeks’ time while also preparing for and conducting other trials. As a result of this rush and inattention trial counsel obtained a misleadingly “benign conception of [De-Bruce’s] upbringing and mental capacity,” Rompilla v. Beard, 545 U.S. 374, 391, 125 S.Ct. 2456, 2459, 162 L.Ed.2d 360 (2005), which led him to present the sentencing jury with an incomplete and inaccurate “life profile of [the] defendant.” Williams, 542 F.3d at 1339 (alteration in original); see id. (“[W]e have found deficient performance in cases where an attorney’s efforts to speak with available witnesses were insufficient to formulate an accurate life profile of the defendant.” (internal quotation marks omitted)).
“[Strategic choices made after less than complete investigation are reasonable only to the extent that reasonable professional judgments support the limitations on investigation,” Wiggins, 539 U.S. at 533, 123 S.Ct. at 2530 (internal quotation marks omitted), and here, there is no indication that Mathis’ failure to develop the troubling leads in DeBruce’s competency report was supported by a professional judgment not to pursue a mitigation investigation. Accordingly, the Alabama Court of Criminal Appeals conclusion that Mathis was not deficient in failing to question DeBruce and his mother about these overt contradictions constituted an unreasonable application of Strickland and its progeny.
Having established that counsel performed deficiently in failing to investigate DeBruce’s mental health and background, DeBruce next must show that he suffered prejudice. See Strickland, 466 U.S. at 687, 104 S.Ct. at 2064. The “relevant question for determining” whether the defendant was prejudiced at sentencing by his counsel’s deficiency “is whether ‘the entire postconviction record, viewed as a whole and cumulative of mitigation evidence presented originally, raised a reasonable probability that the result of the sentencing proceeding would have been different if competent counsel had presented and explained the significance of all the available evidence.’ ” Fortenberry, 297 F.3d at 1229 (quoting Williams, 529 U.S. at 399, 120 S.Ct. at 1516). “A reasonable probability is a probability sufficient to undermine confidence in the outcome” of the petitioner’s sentencing. Williams, 529 U.S. at 391, 120 S.Ct. at 1512.
We have explained that “the primary purpose of the penalty phase is to insure that the sentence is individualized by focusing on the particularized characteristics of the defendant.” Brownlee v. Haley, 306 [1276]*1276F.3d 1043, 1074 (11th Cir.2002) (internal quotation marks omitted). To ensure that the penalty phase achieves this purpose we have held that a petitioner is prejudiced where the mitigation evidence omitted by counsel’s deficient investigation “paints a vastly different picture of [the petitioner’s] background than that created by” the actual penalty-phase testimony. Williams, 542 F.3d at 1342. “By failing to provide such evidence to the jury, though readily available, trial counsel’s deficient performance prejudice^ a petitioner’s] ability to receive an individualized sentence.” Brownlee, 306 F.3d at 1074 (internal quotation marks omitted).
Here, because Mathis’s investigation left him unaware of DeBruce’s upbringing, the sentencing jury heard nothing of the daily beatings that DeBruce suffered as a child at the hands of his older sister, his resistance to joining gangs despite their assaults and intimidation, the pervasive violence in his neighborhood that caused him to witness the stabbing of a neighbor and his brother being shot, his one or more suicide attempts, DeBruce’s efforts to nurse his sister while she recovered from an incapacitating stroke, DeBruce’s alcoholic and disengaged father, or his struggles in school and his low-average intelligence. Although “a troubled history that includes ‘severe privation,’ ‘abuse,’ ‘physical torment,’ and an ‘alcoholic, absentee [parent]” like DeBruce’s “is the kind of troubled history that the [Supreme] Court has ‘declared relevant to assessing a defendant’s moral culpability,’ ” Johnson v. Sec’y, Dept. of Corr., 643 F.3d 907, 935 (11th Cir.2011) (quoting Wiggins, 539 U.S. at 535, 123 S.Ct. 2527), and this information could have been discovered by following up on DeBruce’s competency report, the jury learned nothing of it. Instead, the sentencing jury heard DeBruee’s mother’s falsely embellished testimony that he completed high school, attended college, and that although he had an impoverished childhood, it was otherwise unremarkable. Though the jurors also heard his mother make passing mention of his treatment for a mental disorder, this was not accompanied by any explanation, whatsoever, of the disorder or its effects. This generalized description omitted the “particularized characteristics” of DeBruce’s heavily disadvantaged background and upbringing, Brownlee, 306 F.3d at 1074 (internal quotation marks omitted), and did not even begin to explain the significance of DeBruce’s mental impairments, all of which “might well have influenced the jury’s appraisal of [DeBruce’s] moral culpability,” Wiggins, 539 U.S. at 538, 123 S.Ct. at 2544 (internal quotation marks omitted). Indeed, far from presenting the jury with “an accurate life profile of [DeBruce],” Williams, 542 F.3d at 1339 (internal quotation marks omitted), the profile DeBruce’s sentencing jury received was not only incomplete, but also filled with the misinformation that DeBruce had been a successful student who had attended college.
Accordingly, “[t]his is not a case in which the new evidence would barely have altered the sentencing profile presented to the sentencing judge[,]” Porter, 558 U.S. at 41, 130 S.Ct. at 454 (internal quotation marks omitted); rather, it is one in which adequate investigation would have enabled counsel to correct a positively misleading sentencing profile, and to present “a vastly different picture ... than that created” by the actual trial testimony. Williams, 542 F.3d at 1342.
The Alabama Court of Criminal Appeals determined that counsel’s failure to present this mitigating evidence did not prejudice DeBruce because two aggravating factors were proven against him, and because of its view of the mental health evidence as “conflicting,” and of the facts of the murder itself. DeBruce, 890 So.2d at 1093. However, the Alabama court’s decision to [1277]*1277discount the significance of DeBruce’s mental health mitigating evidence as “conflicting” lacks support from the record. There was no inconsistency in the testimony by DeBruce’s experts at the state collateral hearing, and although the state’s psychologist testified that DeBruce “suffers from no serious mental illness or mental defect” and exhibited some signs of malingering, the state’s expert also conceded that DeBruce “was giving reasonable effort” during testing and that the IQ testing by DeBruce’s experts was accurate and would qualify as a mitigating circumstance. Thus, although there was inconsistency in the testimony by the state’s expert, this inconsistency favors DeBruce, and does not justify discounting DeBruce’s mitigating evidence.
Moreover, in applying Strickland in cases where counsel has failed to introduce any of the available mitigating evidence of the defendant’s mental impairment and history of abuse that should have been discovered by adequate investigation, the Supreme Court has repeatedly rejected arguments that the defendant is not prejudiced by the complete omission of this type of evidence. For example, in Williams v. Taylor, the Supreme Court held that prejudice was established where counsel had introduced the testimony of four witnesses during the sentencing phase, but none of these witnesses described the “abuse and privation” that the defendant suffered during childhood or his serious mental impairments. 529 U.S. at 398, 120 S.Ct. 1495; see id. at 370, 120 S.Ct. 1495 (describing penalty-phase testimony). Similarly, in Wiggins v. Smith, where trial counsel presented no background evidence at all and the jury heard only one mitigating factor at sentencing, 539 U.S. at 515-16, 123 S.Ct. at 2532-33, counsel’s failure to present any evidence of the defendant’s abusive background prejudiced the defendant, see id. at 537-38, 123 S.Ct. at 2543-44. And in Porter v. McCollum, the Court again found prejudice where counsel completely failed to present testimony about the defendant’s abusive background and mental defects, notwithstanding that, as here, one witness did testify on the defendant’s behalf and gave positive information about his background. Porter, 558 U.S. at 41-44, 130 S.Ct. 447.13 As in each of these cases, the penalty-phase testimony provided by the sole witness called at sentencing gave no indication that DeBruce had suffered abuse during his childhood and had been routinely exposed to violence, although this is the type of background evidence that the Supreme Court has repeatedly “declared relevant to assessing a defendant’s moral culpability.” Wiggins, 539 U.S. at 535, 123 S.Ct. at 2542. Moreover, according to the state court’s own characterization of the sentencing proceeding, the testimony adduced at trial consisted almost entirely of DeBruce and his mother “saying that he was sorry and begging the jury to spare his life.” DeBruce, 890 So.2d at 1093. Although DeBruce and his mother pled for mercy at his sentencing, it is unlikely that their pleas resonated with the jury in the absence of the omitted mitigating evidence showing why DeBruce’s background and mental capacity may have warranted clemency. See Rompilla, 545 U.S. at 393, 125 S.Ct. at 2469 (noting that undiscovered “mitigating evidence, taken as a whole, might well have influenced the jury’s appraisal of [a defendant’s] culpability” [1278]*1278where the original sentencing strategy included a plea for mercy) (internal quotation marks omitted). Indeed, because Mathis’s investigation failed to uncover any of the mitigating circumstances that existed in his background, counsel was left without an evidentiary basis on which to make an argument for clemency at the end of sentencing. Thus, when counsel tried to argue that DeBruce’s crime had to be understood in light of his background, the government’s objection that the argument was improper because no evidence of that background had been introduced at sentencing was sustained. But had counsel “presented and explained the significance of’ the regular abuse DeBruce suffered throughout his childhood, his exposure to violence in his community and his limited mental capacity, there would have been a compelling basis on which to argue for clemency in light of DeBruce’s age and life experiences. See Williams, 529 U.S. at 369, 120 S.Ct. 1495 (discussing impact of counsel’s inadequate investigation on ability to present a closing argument).14 Finally, the State does not point to any additional aggravating evidence that would have been introduced had counsel presented testimony about DeBruce’s background and mental health.15 See Wiggins, 539 U.S. at 513, 123 S.Ct. at 2531 (holding that prejudice was enhanced where the State lacked additional aggravating evidence “that could have been introduced ... to offset this powerful mitigating narrative” of childhood abuse and mental impairment).
Accordingly, we find that the Alabama Court of Criminal Appeals unreasonably applied Strickland in holding that the omitted evidence in this case had no reasonable probability of reducing DeBruce’s sentence. Because of trial counsel’s deficient performance, DeBruce’s jury was given almost no reason to spare his life as demonstrated by the paucity of mitigating evidence actually presented during the penalty phase and the fact that the state trial court found only one mitigating circumstance, DeBruce’s age. See DeBruce, 890 So.2d at 1090; see Porter, 558 U.S. at 41, 130 S.Ct. at 454 (finding prejudice where “[t]he judge and jury at [the petitioner’s] original sentencing heard almost nothing that would humanize [the petitioner] or allow them to accurately gauge his moral culpability”). We also emphasize this is not a case where the state courts found, or the State has argued in its briefing, that the prejudicial effect of presenting the omitted mitigating evidence would have been offset by the harmful effect of the evidence itself or would have opened the door to evidence that was more harmful than helpful. See DeYoung v. Schofield, 609 F.3d 1260, 1291 (11th Cir.2010) (discounting the possibility of prejudice where the new mitigating evidence “would have opened the door to harmful testimony which may well have eliminated any mitigating weight in the overall equation”).16 [1279]*1279Because the omitted mitigating evidence, when compared to what little was actually introduced at trial, undermines confidence in the outcome of DeBruce’s sentencing, DeBruce suffered prejudice at the penalty phase of his capital murder trial. See Strickland, 466 U.S. at 694, 104 S.Ct. at 2068.
AFFIRMED IN PART, REVERSED IN PART, AND REMANDED.