PER CURIAM:

Paul Edward Mansur, Denver City, TX, Cathryn Stewart Crawford, TX, Defender Service, Austin, TX, for Petitioner–Appellant. Fredericka Searle Sargent, Assistant Attorney General, for Respondent–Appellee.
Appeals from the United States District Court for the Southern District of Texas.
Before HIGGINBOTHAM, DAVIS, and ELROD, Circuit Judges.

PER CURIAM:

Derrick Dewayne Charles appeals the district court's denial of his 28 U.S.C. § 2254 application challenging his death sentence for the murder of three people. The district court denied his application, but granted a Certificate of Appealability (“COA”) on his claim that he was prejudiced by his counsel's deficient performance at trial. We AFFIRM.

I.

In 2002, Texas charged Charles with capital murder for killing three victims during the same criminal transaction. Charles pleaded guilty and the case proceeded to the punishment phase, in which the jury deliberated the special issues of Charles's future dangerousness and whether sufficient circumstances mitigated against a death sentence.

In seeking a death sentence, the state emphasized Charles's prior lawlessness, which the district court described as follows:

While a juvenile, the police arrested Charles several times. His prior crimes included trespass of an apartment, assault, fighting with police officers, several incidents of burglary, theft, and threatening teachers. While in Texas Youth Commission (“TYC”) custody his numerous rule violations resulted more than one hundred write-ups and confinement in lock-down twenty-two times. At TYC he bragged about selling dope, threatened staff, fought with other students, and received a “chronic serious offender” classification. Upon his release, he failed to attend weekly parole appointments. After serving time in the Texas Department of Criminal Justice (“TDCJ”) for burglary of a habitation, Charles only met with his parole officer once.
Charles v. Thaler, No. 4:09–0592, 2011 WL 5040438, at *4 (S.D.Tex. Oct. 24, 2011).

In response, defense counsel focused on proving that Charles would not be a future societal threat. The district court summarized the testimony presented:

Trial counsel began the defense's case for a life sentence during the cross-examination of the State's witnesses. Trial counsel's cross-examination revealed that Charles was young, was remorseful for the murders, only acted aggressively when provoked by others, had committed only minor infractions while in TYC custody, had never assaulted correctional officers, and was generally a good inmate. Also, trial counsel elicited testimony that one victim's sister forgave Charles for the murders.

Trial counsel affirmatively presented punishment phase testimony through four witnesses. Trial counsel called Dr. Walter Quijano, a psychologist who often testified in capital murder trials, to describe how prison security measures and classification systems would lessen Charles'[s] future dangerousness. Trial counsel supported Dr. Quijano's testimony with testimony from a correctional officer who explained that, since his arrest for capital murder, Charles had only committed two minor infractions of jail rules. The defense also solicited testimony to counter the State's negative evidence from two juvenile corrections officers from TYC who described Charles as a likeable youth and good student whose behavior improved while in custody. Trial counsel focused closing arguments on showing that Charles would not be a future danger to society.
Charles, 2011 WL 5040438, at *7–8.

Pursuant to the jury's answers to the special issues based on this evidence, the trial court sentenced Charles to death. The Texas Court of Criminal Appeals affirmed Charles's conviction and sentence on direct appeal. Charles v. Texas, No. 74,694, 2005 WL 283598 (Tex.Crim.App. Feb. 2, 2005) (unpublished).

While his direct appeal was pending, Charles filed an application for state habeas relief, raising an ineffective-assistance-of-counsel claim under the Sixth Amendment. Charles argued that his trial counsel inadequately investigated and presented mitigating evidence at sentencing. More specifically, Charles cited the failure of his trial counsel to investigate and present mitigating evidence regarding his potential drug use at the time of the crime, troubled family background, and mental health history. Charles argued that trial counsel should have discovered this mitigating evidence (1) from his stepfather, Leroy Phillips, who trial counsel did not call as a witness, and (2) from Charles's medical records from his hospitalization at Gulf Pines Hospital as an adolescent, which trial counsel knew about but never found.

At the state habeas proceedings, Phillips submitted an affidavit describing testimony he would have given if trial counsel had called him as a witness at Charles's trial. The district court summarized the affidavit as follows:

Charles—who was born prematurely and had two seizures as an infant—had a history of depression. His mother's own undiagnosed schizophrenia made her an inattentive parent who sometimes failed to provide adequate food and clothing. Instability, poverty, discord, and turbulence marked Charles'[s] home life. His mother's own mental problems allegedly prevented her from seeing to fruition Charles'[s] treatment for depression.
Charles, 2011 WL 5040438, at *15.

Ultimately, Charles's trial counsel decided not to call Phillips as a witness because his testimony, in the opinion of trial counsel, would not be helpful. According to trial counsel, Phillips gave Charles's bloody clothing from the crime to police, turned in Charles for prior crimes, suffered an assault from Charles, and attempted to teach Charles right from wrong throughout his life.

The Gulf Pines Hospital records, which state habeas counsel discovered by contacting Charles's psychologist, Dr. Lawrence Ginsberg, outlined Charles's adolescent mental and behavioral problems. When Charles was ten years old, Dr. Ginsberg evaluated Charles and recommended hospitalization at Gulf Pines. Upon Charles's first admission in 1993, the hospital provisionally diagnosed him with Oppositional Defiant Disorder, Depressive Disorder, and Rule Out Seizure Disorder. A psychological report suggested that further testing could rule out neurological problems. Admission records noted that ten-year-old Charles had been violent towards peers, easily lost his temper, had witnessed domestic violence, easily became angered, opposed authority, and started treatment at the hospital due to “serious dysfunctionality at home and at school.”

In addition to his behavioral problems, the hospital admission records showed that Charles had risk of suicide, mental anguish, and a medical history that included seizures as an infant, premature birth, and occasional stuttering. A psychosocial assessment noted that Charles came from a “very deprived background,” did not have many opportunities for leisure and recreation, and lived in a bad area. Although the psychosocial assessment suspected “some other significant problems at home,” it identified Charles's family as supportive, “genuinely concerned about [him],” and hopeful that he could go back to school without “fighting and being mean.”

After Charles's initial admission to Gulf Pines in 1993, Dr. Ginsberg assumed responsibility for Charles's treatment and recommended that he stay two or three weeks in the hospital. Instead of following Dr. Ginsberg's recommendation, however, Charles's mother removed him from Gulf Pines after two days and scheduled additional treatment through outpatient therapy. Charles's outpatient therapy continued intermittently until 1995, when his parents again admitted him to Gulf Pines due to depression, decreased amounts of sleep, irritability, and episodes of violence, which included threats and fighting at school, explosive outbursts, and generally disruptive behavior:

On the day of admission, [Charles] had been expelled from school for the fourth time. He threatened to flatten the tires of a teacher's car. He had been fighting and had been assaultive in school. He was using profane language in school and had been very disruptive. Mother notes decreased amounts of sleep, irritability, labile mood, crying spells, impaired memory, and impaired concentration. The patient was charged by the police with assault by threat.

Records from Charles's second visit to Gulf Pines chronicled a range of behavioral and mental issues similar to those of his first visit, but provide more detail into Charles's potential mental problems. Specifically, the records explained that Charles tested “in the intellectually deficientrange of intelligence,” likely had “some organic damage that is contributing to his behavior problems,” and exhibited “some signs of neurological problems.” Results from neurological testing, however, showed that Charles was within normal limits.

Charles remained at Gulf Pines for three weeks during his second visit. During this time, Charles exhibited poor behavior. He swore at and stole from peers, destroyed property, refused to participate in and disrupted therapy sessions, and was described as “silly, vulgar, immature, superficial, [and] non-working.” Upon his discharge, however, the records described Charles as stable and as demonstrating an improved mood, listing his parents as “very supportive” and his mother as “healthy [and] willing to follow direction.”

After reviewing this undiscovered mitigating evidence, the state habeas court denied Charles's petition for habeas relief, reasoning that Charles's trial counsel did not perform deficiently and that trial counsel made “a reasonable, plausible trial decision, made after thorough investigation, to concentrate on the future dangerousness special issue.”

In its factual findings, the state habeas court described the efforts trial counsel took to investigate mitigating evidence:

[Trial counsel] interviewed witnesses, talked to [Charles's] family about possible mitigating evidence, obtained discovery, reviewed the State's file, employed an investigator, vigorously cross-examined the State's witnesses, made appropriate trial objections, presented mitigation evidence at punishment, ... were familiar with the facts of the case and the applicable law, and talked to the applicant about the offense, potential witnesses, the pending trial, and his background and life.
With regard to the efforts trial counsel took to find the Gulf Pines records specifically, the state habeas court found that Charles's trial counsel “exercised due diligence in attempting to locate [the records].” Trial counsel employed a investigator who “had always been able to previously locate records or witnesses who were thought to be unavailable” to locate the Gulf Pines records, and the investigator was unable to locate the records “after a thorough search.”

Ultimately, the state habeas court concluded that Charles failed to show that his trial counsel's performance was factually similar to the situations in Rompilla v. Beard, 545 U.S. 374, 125 S.Ct. 2456, 162 L.Ed.2d 360 (2005), Wiggins v. Smith, 539 U.S. 510, 123 S.Ct. 2527, 156 L.Ed.2d 471 (2003), and Williams v. Taylor, 529 U.S. 362, 120 S.Ct. 1495, 146 L.Ed.2d 389 (2000), where counsel were found ineffective. The court reasoned:

Trial counsel cannot be considered ineffective for not presenting [Charles's] Gulf Pines Hospital records whose slight mitigating value, if any, would have been lost amidst the harmful information contained in the records about the applicant's anger, hostile behavior, and oppositional defiance; the applicant fails to show that the results of the proceeding would have been different if such records [had] been presented.

Charles appealed to the Texas Court of Criminal Appeals, which also denied Charles's petition for habeas relief. After its own review of the record, the Court of Criminal Appeals adopted all of the lower court's findings and conclusions, except that it did not accept the lower court's finding that trial counsel “exercised due diligence in attempting to locate [Charles's] Gulf Pines Hospital records.” Ex parte Charles, No. 67,171–01, 2008 WL 556015, at *1 (Tex.Crim.App. Feb. 27, 2008). The Court of Criminal Appeals did not explain why it agreed with the lower state habeas court in denying Charles relief, yet disagreed with the lower court's finding on the diligence of trial counsel regarding the Gulf Pines records. Charles then filed a federal petition for a writ of habeas corpus, governed by the Antiterrorism and Effective Death Penalty Act of 1996 (“AEDPA”).

After an extensive review of the state habeas proceedings, the district court denied Charles's petition, concluding that the state habeas court's determination was not unreasonable under the standard of review imposed by AEDPA. See Charles, 2011 WL 5040438, at *27. Nonetheless, the district court stated that “[w]hile settled precedent forecloses relief on Charles'[s] Strickland claim, the Court finds his arguments deserve ‘encouragement to proceed further’ ” and certified “appellate review of the issues raised by Charles'[s] petition.” Id. (citation omitted).

II.

“In reviewing a district court's decision on a § 2254 application, we review the district court's findings of fact for clear error and its conclusions of law de novo.” Miller v. Thaler, 714 F.3d 897, 901 (5th Cir.2013) (quoting Harrison v. Quarterman, 496 F.3d 419, 423 (5th Cir.2007)).

Charles's petition is also subject to the deferential standards of AEDPA, as codified in 28 U.S.C. § 2254(d). Id. Section 2254(d) bars federal habeas relief for claims previously adjudicated on the merits by a state court, subject to two exceptions. See Premo v. Moore, ––– U.S. ––––, 131 S.Ct. 733, 739, 178 L.Ed.2d 649 (2011). AEDPA permits federal habeas relief where the state court adjudication either (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.” § 2254(d). Moreover, a factual determination made in state court “shall be presumed to be correct,” § 2254(e)(1), and federal courts must apply § 2254(d)(1) using “the record that was before the state court.” Cullen v. Pinholster, ––– U.S. ––––, 131 S.Ct. 1388, 1398, 179 L.Ed.2d 557 (2011).

“Section 2254(d) sets forth a ‘highly deferential standard for evaluating state-court rulings, which demands that state-court decisions be given the benefit of the doubt.’ ” Miller, 714 F.3d at 901 (quoting Pinholster, 131 S.Ct. at 1398). This deferential standard “stops short of imposing a complete bar on federal court relitigation of claims already rejected in state proceedings” and requires the state prisoner to show that the state court's ruling “was so lacking in justification that there was an error well understood and comprehended in existing law beyond any possibility.” Harrington v. Richter, ––– U.S. ––––, 131 S.Ct. 770, 786–87, 178 L.Ed.2d 624 (2011).

When reviewing a state habeas court's decision under AEDPA's deferential standard of review, we review “only the ultimate legal determination by the state court—not every link in its reasoning.” Trottie v. Stephens, 720 F.3d 231, 241 (5th Cir.2013); see also Green v. Thaler, 699 F.3d 404, 414 (5th Cir.2012) (“[W]e only review the state court's actual decision, not the written opinion on which it is based.”); Neal v. Puckett, 286 F.3d 230, 246 (5th Cir.2002) (en banc) (“[O]ur focus on the ‘unreasonable application’ test under Section 2254(d) should be on the ultimate legal conclusion that the state court reached and not on whether the state court considered and discussed every angle of the evidence.”).

III.

Under Strickland, we apply a two-pronged approach to ineffective-assistance-of-counsel claims. See, e.g., Trottie v. Stephens, 720 F.3d 231, 241 (5th Cir.2013). Strickland 's first prong requires the petitioner to show that trial counsel's representation was deficient—that is, it “fell below an objective standard of reasonableness.” 466 U.S. at 688, 104 S.Ct. 2052. Under the second prong, a petitioner must demonstrate prejudice: a “reasonable probability that, but for counsel's unprofessional errors, the result of the proceeding would have been different.” Id. at 694, 104 S.Ct. 2052. A failure to establish either element is fatal to a petitioner's claim. Id. at 697, 104 S.Ct. 2052.

In applying AEDPA deference to an ineffective-assistance-of-counsel claim, “the pivotal question is whether the state court's application of the Strickland standard was unreasonable.” Richter, 131 S.Ct. at 785. The Supreme Court has explained:

This is different from asking whether defense counsel's performance fell below Strickland 's standard. Were that the inquiry, the analysis would be no different than if, for example, this Court were adjudicating a Strickland claim on direct review of a criminal conviction in a United States district court. Under AEDPA, though, it is a necessary premise that the two questions are different. For purposes of § 2254(d)(1), “an unreasonable application of federal law is different from an incorrect application of federal law.” Williams, [529 U.S.] at 410 . A state court must be granted a deference and latitude that are not in operation when the case involves review under the Strickland standard itself.
Id. Hence, while “[s]urmounting Strickland's high bar is never an easy task,” “[e]stablishing that a state court's application of Strickland was unreasonable under § 2254(d) is all the more difficult.” Id. at 788. Both the Strickland standard and the AEDPA standard are “highly deferential,” and “when the two apply in tandem, review is ‘doubly’ so.” Id. (citation omitted).

A.

We now address whether the state habeas court unreasonably applied Strickland 's deficiency prong. Charles asserts that the state habeas court unreasonably applied Strickland because it failed to address whether trial counsel unreasonably narrowed the scope of his sentencing investigation. Specifically, Charles contends that, had trial counsel conducted a proper investigation, he would have discovered additional mitigating evidence regarding Charles's mental health history, troubled family background, and his stay at Gulf Pines Hospital. In light of his trial counsel's failure to discover this evidence, Charles argues that he has a “paradigmatic Wiggins case” and that trial counsel's decision not to investigate further was uninformed. As recognized by the district court and the parties, Charles's most troubling ineffective assistance allegation centers on the failure of trial counsel to discover the records from Gulf Pines Hospital, which chronicled Charles's adolescent mental and behavioral problems.

In evaluating trial counsels' performance for deficiency, Strickland requires a “strong presumption that counsel's conduct falls within the wide range of reasonable professional assistance.” 466 U.S. at 689, 104 S.Ct. 2052. Ultimately, the defendant must show that the errors were so egregious as to deprive the defendant of the “counsel” guaranteed by the Sixth Amendment. Richter, 131 S.Ct. at 787. In short, judicial scrutiny of counsel's performance is highly deferential. Id. at 788;see also Strickland, 466 U.S. at 689, 104 S.Ct. 2052. But despite the wide latitude afforded trial counsel in evaluating their actions, it is not boundless.

In investigating potential mitigating evidence, counsel must either (1) undertake a reasonable investigation or (2) make an informed strategic decision that investigation is unnecessary. Higgins v. Cain, 720 F.3d 255, 265 (5th Cir.2013) (explaining that under Strickland, counsel must “research relevant facts and law, or make an informed decision that certain avenues will not prove fruitful”); see also Strickland, 466 U.S. at 690–91, 104 S.Ct. 2052 (“[S]trategic choices made after less than complete investigation are reasonable precisely to the extent that reasonable professional judgments support the limitations on investigation.”).

Because, as the parties agree, trial counsel did not discover the Gulf Pines Hospital records and similar information regarding Charles's mental health history, we address the latter inquiry: whether the state habeas court unreasonably applied Strickland in reviewing Charles's petition.

To answer that question, Wiggins, 539 U.S. 510, 123 S.Ct. 2527, provides helpful guidance. In Wiggins, the petitioner's attorneys failed to investigate and present mitigating evidence of the petitioner's dysfunctional family and social history, despite having some information available to them in a pre-sentence investigation report and social service records. Id. at 516, 534, 123 S.Ct. 2527. The Court concluded that counsel's failure to investigate Wiggins's life history fell short of professional standards, basing its decision on a multi-faceted, “reasonableness in all the circumstances” approach: (1) counsel cut short their investigation after “having acquired only rudimentary knowledge of [Wiggins's] history from a narrow set of sources,” (2) counsel “uncovered no evidence in their investigation to suggest that ... further investigation would have been fruitless,” and (3) the failure to investigate “resulted from inattention, not strategic judgment.” Id. at 521–27, 123 S.Ct. 2527.

The Supreme Court has also stated that, under a Strickland analysis, trial counsel must not ignore “pertinent avenues of investigation,” Porter v. McCollum, 558 U.S. 30, 40, 130 S.Ct. 447, 175 L.Ed.2d 398 (2009), or even a single, particularly promising investigation lead, Rompilla, 545 U.S. at 383–84, 125 S.Ct. 2456. In Porter, for example, the Supreme Court held that counsel performed deficiently when he “did not even take the first step of interviewing witnesses or requesting records” and “ignored pertinent avenues for investigation of which he should have been aware.” 558 U.S. at 40, 130 S.Ct. 447. In Rompilla, the Supreme Court faulted defense counsel for failing to look at a file to investigate a prior conviction that he knew the prosecutor intended to use against his client. 545 U.S. at 383–84, 125 S.Ct. 2456. The Court explained that counsel did not “look at any part of that file, including the transcript, until warned by the prosecution a second time.” Id. at 384, 125 S.Ct. 2456. Had counsel looked, he would have discovered “a range of mitigation leads that no other source had opened up.” Id. at 390, 125 S.Ct. 2456.

A careful comparison to Wiggins, Porter, and Rompilla suggests that the state habeas court did not unreasonably apply Strickland's deficiency prong by concluding that Charles's trial counsel performed an adequate mitigation investigation.

1.

Unlike counsel in Wiggins, Charles's trial counsel did not abandon his mitigation investigation after a cursory review of “a narrow set of sources.” 539 U.S. at 524, 123 S.Ct. 2527. The record shows that trial counsel started to prepare a punishment-phase defense soon after his appointment and reviewed Charles's school, Texas Youth Commission, and Texas Department of Criminal Justice records. Trial counsel spoke with Charles, his parents, and other family members with the specific purpose of discovering mitigating evidence. In addition, trial counsel hired experts to assist with the mitigation investigation. An investigator interviewed witnesses, including Charles's family members, parents, teachers, and schoolmates. A psychologist, Dr. Jerome Brown, interviewed Charles and his mother and stepfather, and reviewed Charles's school and medical records.

Moreover, aside from the Gulf Pines records or the affidavit from Charles's stepfather, Leroy Phillips, Charles fails to identify the sources from which trial counsel could have obtained additional mitigating evidence. See Trottie, 720 F.3d at 243 (“[A] defendant who alleges a failure to investigate on the part of his counsel must allege with specificity what the investigation would have revealed and how it would have altered the outcome of the trial.” (quoting Druery v. Thaler, 647 F.3d 535, 541 (5th Cir.2011))). With regard to Phillips, the state habeas record indicates that counsel knew the information that Phillips possessed but made a strategic decision not to call him as a witness due to the harm that might result from cross-examination. Phillips gave Charles's bloody clothing from the crime to police, turned in Charles for prior crimes, suffered an assault from Charles, and attempted to teach Charles right from wrong throughout his life. To the extent Charles argues that Phillips should have testified, “[c]laims that counsel failed to call witnesses are not favored on federal habeas review because the presentation of witnesses is generally a matter of trial strategy and speculation about what witnesses would have said on the stand is too uncertain.” Woodfox v. Cain, 609 F.3d 774, 808 (5th Cir.2010).

2.

Whereas trial counsel in Wiggins had “no evidence” that further investigation would have been fruitless, Charles's trial counsel possessed some evidence that the Gulf Pines records would not be helpful to Charles's mitigation argument. At the state habeas proceedings, trial counsel testified that Charles's parents said they admitted Charles to Gulf Pines because he was “acting out” and that the visit “really didn't have anything to do with mental illness.” A review of Charles's hospital records confirms that Charles's family, particularly in regard to his 1995 admission, wanted help dealing with Charles's violent behavior.

Most notably, the report of Dr. Brown indicated that trial counsel would have little success finding mitigating evidence using a psychological theory. Dr. Brown's report stated that

Unfortunately, the information now available does not provide any evidence that might be considered mitigating in this examiner's opinion. Beyond a possible diagnosis of Attention Deficit Disorder, he does not reveal any evidence of mental illness or other types of mental disorder or mental defect that might be offered on his behalf. The possibility of mental retardation was investigated but this was contra-indicated by school records, the psychological test results, and the reports from his parents concerning his adaptive skills.... There was nothing else offered in his history by the defendant or by his parents that would be of use in his defense, including the possibility of brain tissue trauma, childhood abuse, deprivation, or traumatic life events. Because of this, you were advised that any testimony I might offer on the defendant's behalf would be minimal or unhelpful at best, and that cross[-]examination might prove damaging to the defendant.

Hence, in retrospect, the Gulf Pines records contained information that counsel may have used in presenting mitigating evidence for Charles, but that fact was not altogether clear: previous attempts to diagnose Charles with mental illness had failed and trial counsel had reason to believe the records would not be helpful—or worse, harmful. See Strickland, 466 U.S. at 691, 104 S.Ct. 2052 (“[W]hen a defendant has given counsel reason to believe that pursuing certain investigations would be fruitless or even harmful, counsel's failure to pursue those investigations may not later be challenged as unreasonable.”).

3.

Finally, unlike Wiggins, the failure of Charles's trial counsel to discover the Gulf Pines records did not result from pure inattention, and this is not a case like Porter, where counsel wholly ignored multiple avenues of investigation. 558 U.S. at 40, 130 S.Ct. 447. In contrast, Charles's claim centers on the Gulf Pines records, a single mitigation lead. Furthermore, although Rompilla supports the proposition that even missing a single pertinent lead may satisfy Strickland, that case dealt with a “readily available” file that the prosecution tipped-off to defense counsel. Rompilla, 545 U.S. at 385, 125 S.Ct. 2456. Here, by contrast, Charles's trial counsel was aware of the Gulf Pines records but, like the state, gave up looking for the records because the hospital had shut down. Accordingly, for these reasons, the state habeas court's denial of Charles's claim was not an unreasonable application of Strickland's deficiency prong.

B.

Even assuming arguendo that the state habeas court unreasonably applied Strickland's deficiency prong, we are not persuaded that the state habeas court unreasonably applied Strickland's prejudice prong, although it presents a closer question.

The prejudice inquiry under Strickland requires evaluating whether there is a “ reasonable probability that, but for counsel's unprofessional errors, the result of the proceeding would have been different.” Strickland, 466 U.S. at 694, 104 S.Ct. 2052 (emphasis added). Here, the state habeas court omitted the “reasonable probability” modifier and concluded that “[Charles] fails to show that the results of the proceeding would have been different if [the Gulf Pines Hospital records] [had] been presented.” As a threshold matter, Charles argues that, because the state court omitted the “reasonably probability” modifier, we should not apply AEDPA deference to the state habeas court's decision on the issue of prejudice.

A state-court decision is contrary to clearly established federal law—and therefore not entitled to deference under AEDPA—when it uses a preponderance-of-the-evidence test to determine prejudice, rather than the less onerous “reasonable probability” test promulgated by Strickland:

If a state court were to reject a prisoner's claim of ineffective assistance of counsel on the grounds that the prisoner had not established by a preponderance of the evidence that the result of his criminal proceeding would have been different, that decision would be “diametrically different,” “opposite in character or nature,” and “mutually opposed” to our clearly established precedent because we held in Strickland that the prisoner need only demonstrate a “reasonable probability that ... the result of the proceeding would have been different.”
Williams, 529 U.S. at 405–06, 120 S.Ct. 1495 (alteration in original) (citing Strickland, 466 U.S. at 694, 104 S.Ct. 2052). Therefore, because the state court omitted the “reasonable probability” modifier in its decision, we must determine whether the court employed a preponderance-of-the-evidence standard—which is contrary to clearly established federal law—or whether the court used a shorthand method to refer to the correct reasonable probability standard.

Although no Fifth Circuit or Supreme Court precedent controls that question, the Seventh Circuit's decision in Sussman v. Jenkins frames a similar fact pattern. 636 F.3d 329 (2011). In Sussman, a state court omitted the “reasonable probability” language from its Strickland prejudice determination. Id. at 359. In concluding that the state court's decision was nevertheless entitled to deference under AEDPA, the Seventh Circuit reasoned that (1) the state-court decision correctly cited another case that incorporated the correct ineffective assistance standard under Strickland and (2) it was “clear from the [state] court's analysis that it did not believe that the [undiscovered evidence] had a reasonable probability of altering the jury's verdict.” Id. at 359–60.

We find the Seventh Circuit's analysis in Sussman persuasive and therefore apply its reasoning to this case. Here, similar to Sussman, the state habeas court cited a number of Supreme Court cases applying the correct Strickland prejudice standard, including Wiggins, 539 U.S. 510, 123 S.Ct. 2527,Rompilla, 545 U.S. 374, 125 S.Ct. 2456, and Williams, 529 U.S. 362, 120 S.Ct. 1495. This indicates that the state habeas court omitted the “reasonable probability” modifier not due to its incorrect understanding of the prejudice standard, but as a shorthand method to refer to the correct standard. See Visciotti, 537 U.S. 19, 24, 123 S.Ct. 357 (2002) ( “[R]eadiness to attribute error is inconsistent with the presumption that state courts know and follow the law.”).

Furthermore, the state habeas court made clear that it did not believe the Gulf Pines records had a reasonable probability of altering the jury's verdict. Indeed, if anything, the state habeas court believed the Gulf Pines records would have provided no value to Charles's sentencing arguments:

Trial counsel cannot be considered ineffective for not presenting [Charles's] Gulf Pine[s] Hospital records whose slight mitigating value, if any, would have been lost amidst the harmful information contained in the records about [his] anger, hostile behavior, and oppositional defiance....
Accordingly, we apply AEDPA deference to the state habeas court's prejudice conclusion.

In reviewing whether the state habeas court unreasonably applied Strickland 's prejudice prong, we must “consider all the relevant evidence that the jury would have had before it if [trial counsel] had pursued the different path.” Wong v. Belmontes, 558 U.S. 15, 20, 130 S.Ct. 383, 175 L.Ed.2d 328 (2009). In Kitchens v. Johnson, for example, the defendant alleged that his trial counsel performed deficiently by failing to present hospital records detailing not only the defendant's history of suicide attempts and depression, but also his history of drug and alcohol abuse. 190 F.3d 698, 703 (5th Cir.1999). In concluding that the defendant suffered no actual prejudice, we highlighted the “double-edged nature” of the unpresentedevidence: “[I]f counsel had introduced the hospital records, the jury may have better understood [Kitchen's] mental state, but would have seen a long history of drug and alcohol abuse.” Id.

Likewise here, if Charles's trial counsel had discovered and presented the Gulf Pines records, the jury would have received a better picture of Charles's mental history, but also would have seen, for example, instances where Charles was violent towards peers, was charged with assault by threat, lost his temper, and was suspended from school due to his hostile behavior. According to the district court, although the Gulf Pines records showed Charles's potential mental-health problems, they also “portended of future societal danger”:

The [Gulf Pines] records contain two diagnoses: oppositional defiant disorder and depression. Aside from ascribing a psychological label to his bad behavior, Charles does not explain how the jury would find mitigating components in his oppositional defiant disorder. The records do not identify what caused him to have such a confrontational attitude, but more-than-amply show how it caused him to lash out at others. The hospital reports of Charles'[s] defiant and aggressive personality may well have engulfed at trial any mitigating features of the depression diagnosis. Worse, as discussed below, the Gulf Pines Hospital records show that Charles'[s] aggression and violence were not episodic, but a repeated theme in his life.
Charles, 2011 WL 5040438, at *24.

In response, Charles argues that the jury had already heard about Charles's violent tendencies and that, therefore, the additional indications of his violence in the Gulf Pines records would not have hurt his sentencing arguments. To substantiate this argument, Charles lists a number of instances where the jury heard about Charles's unruly and criminal behavior as an adolescent, including where Charles was arrested for criminal trespass, hit his uncle in the head and subsequently resisted arrest, failed to follow probation supervision rules, and threatened and showed extreme disrespect for authority and others.

Charles is correct that the Gulf Pines records contain information regarding Charles's violent tendencies similar to that already presented to the jury—that is, the Gulf Pines records merely confirm and provide more evidence of the aggressive, disrespectful, and violent behavioral tendencies that the state attempted to show the jury at sentencing. Therefore, assessing Strickland 's prejudice inquiry boils down to an assessment of the degree to which the harmful information in the Gulf Pines records would have harmed Charles's case and the degree to which the helpful information would have helped it.

That is a difficult question. But the difficulty associated with answering Strickland 's prejudice prong in this case is precisely why it is hard to portray the state habeas court's decision against Charles as unreasonable. Charles has not shown that the state court's ruling was “so lacking in justification that there was an error well understood and comprehended in existing law beyond any possibility for fairminded disagreement.” Richter, 131 S.Ct. at 786–87. Therefore, applying the deferential standard of review imposed by AEDPA, we conclude that the state habeas court did not unreasonably apply Strickland 's prejudice prong and that relief is not warranted under § 2254(d)(1).

IV.

To obtain relief under § 2254(d)(2), Charles must prove that the state habeas court's decision rested on “an unreasonable determination of the facts in light of the evidence.” “Section 2254 also requires that determinations of fact issued by state courts are ‘presumed to be correct,’ and that they not be disturbed unless an applicant rebuts the presumption with clear and convincing evidence.” Lewis v. Thaler, 701 F.3d 783, 791 (5th Cir.2012), cert. denied,––– U.S. ––––, 133 S.Ct. 1739, 185 L.Ed.2d 798 (2013) (quoting 28 U.S.C. § 2254(e)(1)). To satisfy this burden, Charles makes several arguments, all of which either rely on an incorrect understanding of the state habeas court's findings or parse the record in a manner inconsistent with the AEDPA's deferential standard of review. Accordingly, we will not grant relief on any of Charles's claims under § 2254(d)(2).

A.

Charles's first argument relates to trial counsel's understanding of his visits to Gulf Pines Hospital, prior to discovery of the records from those visits. He argues that the state habeas court unreasonably determined that, at the time of trial, trial counsel understood that Charles's hospitalization at Gulf Pines was unrelated to mental illness and lasted for two days when, after trial, Charles's trial counsel admitted knowing that Charles's hospitalization was related to depression and lasted for fifteen days.

Contrary to what Charles implies in his brief, the state habeas court's found only that “the applicant's family informed [trial] counsel that [Charles's] stay at Gulf Pines was a result of the problems he caused at home, and that the applicant's hospitalization had nothing to do with mental illness.” The finding in question did not mention the length of Charles's stay, and the state habeas court's finding did not relate to what trial counsel actually knew—only to what Charles's family told him.

Moreover, Charles makes this argument in response to the district court's conclusion that trial counsel “had little reason to think that the Gulf Pines Hospital records might benefit the defense” when Charles's family had told trial counsel that the visits related to Charles's “acting out” and nothing more. At issue, however, is whether the state habeas court's decision rested on an unreasonable determination of the facts, and Charles does not show, other than a conclusory allegation, that the state habeas decision rested on whether trial counsel believed that Charles's stay at Gulf Pines related to mental illness.

B.

Charles's second argument relates to trial counsel's general knowledge of the case. Specifically, Charles's argues that the state habeas court unreasonably determined that trial counsel was “familiar with the facts of the case” when trial counsel failed to learn about various aspects of Charles's background contained in the Gulf Pines records.

Charles's argument fails when the state habeas court's finding is viewed in context, which shows the efforts counsel undertook to understand and present Charles's case:

The Court finds, based on the appellate record and the credible affidavit of trial counsel ..., that counsel prepared and filed pre-trial motions, interviewed witnesses, talked to the applicant's family about possible mitigating evidence, obtained discovery, reviewed the State's file, employed an investigator, vigorously cross-examined the State's witnesses, made appropriate trial objections, presented mitigation evidence at punishment ... were familiar with the facts of the case and the applicable law, and talked to the applicant about the offense, potential witnesses, the pending trial, and his background and life.
In the context of these other findings, which Charles does not challenge, the state habeas court's determination that counsel knew the facts of the case does not appear unreasonable. The fact that trial counsel did not know every detail regarding Charles's background—including the contents of the undiscovered Gulf Pines records—does not mean that trial counsel was not “familiar with the facts of the case” more generally.

V.

We conclude that the state habeas court's decision was not contrary to, or an unreasonable application of, clearly established federal law and that the state habeas court's decision was not based on an unreasonable determination of the facts in light of the evidence. Accordingly, we AFFIRM.