John Lesko v. Secretary Pennsylvania Departm

PRECEDENTIAL UNITED STATES COURT OF APPEALS FOR THE THIRD CIRCUIT ______ No. 15-9005 ______ JOHN C. LESKO, Appellant v. SECRETARY PENNSYLVANIA DEPARTMENT OF CORRECTIONS; SUPERINTENDENT GREENE SCI; SUPERINTENDENT ROCKVIEW SCI; DISTRICT ATTORNEY WESTMORELAND COUNTY ______ On Appeal from the United States District Court for the Western District of Pennsylvania (D. C. No. 2-11-cv-01049) District Judge: Honorable Cathy Bissoon, U.S. District Judge ______ Argued June 23, 2021 Before: KRAUSE, ROTH and FISHER, Circuit Judges. (Filed: May 17, 2022) Samuel J.B. Angell [ARGUED] Timothy P. Kane [ARGUED] Federal Community Defender Office for the Eastern District of Pennsylvania 601 Walnut Street The Curtis Center, Suite 540 West Philadelphia, PA 19106 Counsel for Appellant Thomas R. Grace, I John W. Peck [ARGUED] Elizabeth P. Ranger Westmoreland County Office of District Attorney 2 North Main Street, Suite 206 Greensburg, PA 15601 Counsel for Appellee ______ OPINION OF THE COURT 2 ______ FISHER, Circuit Judge. Over 40 years ago, John Lesko went on a multi-day killing spree with his friend, Michael Travaglia, ending the lives of four individuals in Western Pennsylvania in a tragedy dubbed the “Kill for Thrill” murders by the media. For the last killing, Lesko was convicted of first-degree murder and sentenced to death. After proceeding through many levels of the Pennsylvania state courts and two rounds of federal habeas proceedings, Lesko now asks this Court to grant him relief as to both his conviction and sentence. Like the District Court before us, we conclude that the 28 U.S.C. § 2254 petition fails to entitle him to relief. We therefore will affirm. I. A. Factual History In late December 1979 and early January 1980, John Lesko and Michael Travaglia killed four unrelated individuals without any apparent motive, except for a desire to kill. Their first victim was Peter Levato, then 52 years old. On December 27, 1979, Lesko and Travaglia abducted Levato from the Edison Hotel in downtown Pittsburgh. They entered Levato’s car, held him at gunpoint, and bound his arms and legs. Lesko and Travaglia then drove to a bridge, where they pushed Levato, still bound, over the side and into the water. After hearing Levato scream below, Lesko and Travaglia went to search for him. They found him standing beneath a tree, at which point Travaglia shot Levato once in the chest and twice in the head, killing him. 3 Their next victim was 26-year-old Marlene Newcomer. On January 1, 1980, Newcomer picked up Lesko and Travaglia as they were hitchhiking. Inside the car, Lesko pointed a gun at Newcomer and the pair ordered her to pull over, where she was then handcuffed, put in the back seat, and covered with a blanket. Lesko and Travaglia drove to a convenience store and robbed it. They then returned to the car, where Lesko shot Newcomer, missing with his first shot but shooting twice more, killing her. A day later, Lesko and Travaglia killed their third victim, William Nicholls, age 32. At about 11:00 p.m. on January 2, 1980, Lesko and Travaglia were together at a hot dog shop in downtown Pittsburgh when they ran into 15-year- old Ricky Rutherford, who would later testify for the prosecution. Rutherford had used illegal drugs and drunk alcohol with Travaglia before. At the hot dog shop, Lesko and Travaglia asked Rutherford if he wanted to go partying with them and he said he did. Travaglia then instructed Lesko and Rutherford to wait for him in a nearby alley. The two went to the alley, where Lesko told Rutherford that when they heard a car horn, it would be Travaglia. Five or ten minutes later, a car came down the alleyway and beeped its horn. Williams Nicholls, the victim, was the driver of that car, and Travaglia was sitting in the front passenger seat. Rutherford entered the car and sat in the back seat. Travaglia then pulled out a gun and shot Nicholls in the arm. After Nicholls was shot, Lesko joined Rutherford in the backseat of the car and when asked by Travaglia what the gunshot sounded like, Lesko said that it sounded like a firecracker. Travaglia ordered Nicholls, still in the driver seat, to drive out of the alley and he complied. The group left Pittsburgh and later stopped so Travaglia could take the wheel and move Nicholls to the back. At Lesko’s direction, 4 Rutherford handcuffed Nicholls. Nicholls was then put into the backseat next to Lesko, with Travaglia in the driver’s seat and Rutherford in the passenger seat. In the backseat, Lesko repeatedly punched Nicholls in the face and chest, while calling him a “queer,” asking him if wanted to perform oral sex on him, mocking the sound of his voice, and threatening him with a knife. J.A. 554-55. Travaglia also punched Nicholls, who begged his assailants to stop. Instead, they laughed at him and continued to torture him until he eventually passed out. Lesko and Rutherford then gagged the unconscious victim. The group drove to a frozen lake in the woods. There, Lesko took Nicholls’ wallet, keys, and other personal effects. Travaglia, taking Rutherford with him, went down to find a heavy brick or rock, which Rutherford then carried back to the car. They returned to the car, where Lesko had pulled Nicholls out of the vehicle, his hands still shackled and his legs now bound together by a belt. Travaglia broke a hole in the surface of the frozen lake, and he and Lesko dragged Nicholls down to the lake. Rutherford did not see what happened next, but Travaglia and Lesko returned to the car without Nicholls. Nicholls died at the lake that day.1 Lesko, Travaglia, and Rutherford drove away in Nicholls’ stolen car. During the drive, Travaglia described how they put Nicholls into the frozen waters, where he resurfaced once, coughed, and then sank back down. 1 The record is not entirely clear as to Nicholls’s cause of death—whether he died by drowning or some other cause— but that is of limited importance here. Lesko pleaded guilty to second-degree murder for his involvement in killing Nicholls. See Commonwealth v. Lesko, 467 A.2d 307, 308 (Pa. 1983). 5 The group drove to Travaglia’s father’s house. There, Lesko and Travaglia went inside and stole a gun. They returned to the car and drove away, until Lesko realized that the gun was loaded with only birdshot. They then returned to the house, where Travaglia directed Rutherford to get a box of real bullets from his father’s garage, which he did. They then drove away, Travaglia driving, Lesko in the passenger seat, and Rutherford in the back. After driving around for some time in the early morning of January 3, the group came upon a police officer parked at the side of the road—21-year-old Leonard Miller. Travaglia remarked, “Let’s have some fun with this cop,” and then sped past him in the stolen car, honking the horn. J.A. 569. Officer Miller did not pursue them, so Travaglia tried again, speeding past him, running a red light, and sounding the horn once more. This time, Officer Miller turned on his lights and gave pursuit. Lesko told Rutherford to lay down in the back of the car, “because it might turn into a shooting gallery.” J.A. 569. Rutherford did so. Travaglia pulled the car off to the side of the road, stopped, and rolled down his window. Officer Miller then came up to the side of the car, where Travaglia shot him twice, killing him. Before he died, Officer Miller returned fire, ultimately breaking the passenger window. Travaglia, Lesko, and Rutherford sped away, with Travaglia commenting on how the cop fell when he was shot and Lesko stating that they would have to get rid of the car because someone might notice the broken window. The car broke down a few miles later and the group began walking on foot. They were eventually able to get a ride from an acquaintance of Travaglia. Lesko and Travaglia returned to Pittsburgh, and that evening, they came across Daniel Montgomery at the hot dog shop downtown. Montgomery, who would later testify on 6 behalf of the prosecution, stated that Travaglia asked him to accompany them to a hotel room across the street. There, Travaglia confessed to Montgomery, “I shot a cop,” and Lesko snickered and added, “I wanted to.” J.A. 679-80. Travaglia then gave Montgomery the gun he had used to shoot Officer Miller and asked him to keep it. Montgomery obliged, left the hotel room, and returned to the hot dog shop, where he was arrested by plain clothes police officers who recovered the gun. Lesko and Travaglia were arrested later that night. Both admitted to some role in the killings. Lesko confessed he was present when Officer Miller was shot in a narrative which largely tracked Rutherford’s account. He admitted that Travaglia, Rutherford, and himself, while driving a stolen vehicle, sped by Miller twice to draw his attention, but claimed it was only to lure him away from a convenience store so they could rob it. According to Lesko, he did not know Travaglia would shoot Miller. B. Procedural History Lesko and Travaglia were arraigned for the killing of Officer Miller, the last victim in the four-victim spree. The Commonwealth charged first degree murder and sought the death penalty. By the time trial came for the Miller killing, Lesko and Travaglia had already been convicted of (1) first degree murder for the killing of Levato, for which they were sentenced to life in prison; (2) first degree murder for the killing of Newcomer, for which they were also given a life sentence; and (3) second degree murder for the killing of Nicholls, for which sentencing was to be deferred until after the Miller trial. 7 1. Lesko’s 1981 Trial, Direct Appeal, and Petition for Post- Conviction Relief In 1981, Lesko and Travaglia were tried together in Westmoreland County for the killing of Officer Miller. Rutherford and Montgomery testified for the prosecution and Lesko’s confession to the police was also introduced. The jury also heard about Lesko’s role in killing Nicholls the night before Officer Miller was shot. For the defense, Rabe F. Marsh, III, a local attorney in private practice who had never tried a capital case, was appointed as attorney for Lesko. The jury found Lesko and Travaglia guilty of first- degree murder. The case proceeded to a separate sentencing trial. There, the jury found that the aggravating circumstances surrounding the murder outweighed any mitigating circumstances and imposed the death penalty as to both. The Pennsylvania Supreme Court affirmed on appeal, Commonwealth v. Travaglia, 467 A.2d 288 (Pa. 1983) (“Lesko I”), and relief was denied on collateral review. See Commonwealth v. Lesko, 501 A.2d 200 (Pa. 1985) (“Lesko III”).2 2. Lesko’s First Habeas Petition In 1986, Lesko, still represented by his appointed lawyer, Marsh, filed his initial habeas petition under 28 U.S.C. § 2254. The District Court granted the petition on the ground 2 There have been more than a dozen decisions in Lesko’s case. In his filings in the District Court, Lesko labeled the decisions Lesko I through Lesko XIII. The District Court adopted this convention, so we do as well, noting that not all decisions are pertinent to the resolution of Lesko’s current petition. 8 that Lesko’s right to a fair trial had been violated by the introduction of evidence concerning his participation in the Nicholls killing. However, on appeal, we concluded that this evidence was probative of Lesko’s motive and intent and that its introduction did not violate due process. See Lesko v. Owens, 881 F.2d 44, 60 (3d Cir. 1989) (“Lesko VI”). We thus vacated the District Court’s judgment and remanded for consideration of Lesko’s other claims. On remand, the District Court denied all remaining requests for relief. Lesko appealed once more. Again, we vacated and remanded. We held that he was entitled to: (1) an evidentiary hearing regarding his claim that the introduction of his guilty plea for the second-degree murder of Nicholls violated his due process rights; and (2) a new sentencing trial, because the prosecution had improperly appealed to vengeance and commented on Lesko’s failure to testify. See Lesko v. Lehman, 925 F.2d 1527, 1555 (3d Cir. 1991) (“Lesko VIII”). 3. Lesko’s 1995 Resentencing Hearing and Second Death Sentence The Commonwealth held a new sentencing hearing in 1995. Seeking the death penalty once more, it sought to establish four aggravating factors: (1) and (2) that Lesko had previously been convicted of two other crimes that subjected him to a life sentence or the death penalty, the Levato and Newcomer murders; (3) that the victim in the immediate case was a police officer killed while performing his duties; and (4) that Lesko had a significant history of felony convictions involving the use of violence. The Commonwealth presented evidence showing Lesko’s role in killing Levato, Newcomer, 9 and Nicholls, though the Nicholls killing was offered only to establish Lesko’s intent in killing Miller.3 In mitigation, Lesko—still represented by his appointed lawyer, Marsh, alongside a new lawyer, Brian O’Leary, who joined the team just days before the resentencing—called several witnesses. They brought in a forensic social worker, who presented Lesko’s background; two of Lesko’s family members, his aunt and one brother; a clinical and forensic psychologist who evaluated Lesko; a prison chaplain; and Lesko himself. The forensic social worker, Lois Nardone, testified about her investigation of Lesko, which included interviews with Lesko, his mother, his grandmother, his two brothers, his two sisters, his aunt, two of his teachers, a prison chaplain, and a social worker. She also reviewed relevant records. Based on her research, she gave a disturbing biography. Lesko never knew the identity of his father. His mother, Marion Fedorko, raised Lesko, but was a “terrible” parent. J.A. 2295. She would routinely have sex with strangers in front of Lesko; would go out drinking, leaving the children home alone at night; and did 3 In Lesko’s first habeas proceedings, we remanded for an evidentiary hearing on whether Lesko’s plea in the Nicholls case was voluntary for the purpose of being considered as an aggravating factor in the Miller case. On remand, the magistrate judge who presided over that evidentiary hearing determined—and the District Court agreed—that the plea was involuntary for that purpose, because Lesko “had every reason to reasonably believe that the plea agreement included a ‘non- use’ provision” barring such use. Lesko v. Lehman, No. Civ. A. 86-1238, 1992 WL 717815, at *6 (W.D. Pa. Feb. 20, 1992). Thus, the Nicholls killing was not itself an aggravating factor. 10 not provide habitable housing for Lesko and his siblings. The “deplorable” housing conditions included disconnected utilities, minimal food, and frequent moves—some 15 times before Lesko was even 10 years old. J.A. 2296. Lesko suffered greatly. At age five or six, he was wandering the streets at night with his younger brother, Michael, often stealing food. The children went to school only when they wanted, and when they did, they appeared dirty and sleep deprived. Nardone also testified about Lesko’s traumas. When he was four years old, another boy set him on fire. He sustained first- and second-degree burns and was hospitalized for more than a month. At age six or seven, Lesko was molested while he and his brother were attempting to earn money by shining shoes in a bar. Lesko was removed from his mother’s care when he was nine and spent time in a few different facilities. He had “a very difficult time” adjusting, reacting poorly when apart from his younger brother, Michael, who Lesko viewed as a father figure. J.A. 2300. At times, Lesko had to be restrained, because he could not calm down. He would receive some visits from his grandmother, but none from his mother. Lesko remained institutionalized until just before his fifteenth birthday, when he and Michael were placed in the custody of his grandmother. There too, life was difficult. His grandmother’s daughter—an alcoholic and a “very nasty drunk”—also lived at the home, and she resented Lesko and Michael coming there. J.A. 2302. She picked on Lesko, as the physically weaker of the two children, and verbally abused him. Lesko did not cope well. By age 15 or 16, he was heavily abusing drugs and alcohol, would go for days without sleep, and would be unable to calm himself down without substances. He did not regularly attend school. He pleaded guilty to burglary. Eventually, his probation officer told him that he should join the military or he would end up in 11 jail. Lesko joined the Marines, which led to a short period of stability in his life. But, after boot camp was over and the rules were relaxed, he went “AWOL” and was discharged. J.A. 2305-06. He returned to his grandmother’s home, where his behavior worsened, with substance abuse becoming a daily occurrence and his anger issues worsening. According to Nardone, that anger intensified when Lesko, in 1978 or 1979, learned that his other brother, Joey, was released from an institution because he had been molested there by a social worker. Lesko was filled with rage. Paralleling this narrative which Nardone presented, Lesko’s two testifying family members reiterated his awful upbringing. His aunt, Anna Aikman, testified that the Lesko’s mother’s home was a “pigsty,” often with no lights, gas, or food. J.A. 2452. She confirmed that the children would be left unattended; their mother would bring men home from the bars and direct her children to steal from them; and that Lesko suffered verbal abuse when he was living with his grandmother. His brother, Michael, who also testified, confirmed that they suffered severe neglect from their mother; Lesko got heavily involved with drugs when he lived with his grandmother and suffered verbal abuse there; and Lesko was molested. Lesko’s mitigation team also called Dr. Herbert Levit, a forensic psychologist, who had examined Lesko and performed a variety of psychological tests. Dr. Levit diagnosed Lesko with borderline personality disorder: “a person who is emotionally explosive at times, has erratic behavior, engages in behavior which is contrary to societ[y’s] standards, is frequently immature, and has had difficulty in adjusting.” J.A. 2410-11. He also opined that, at the time of the killings, Lesko was suffering from polysubstance abuse, meaning that he was 12 using drugs and alcohol consistently and excessively such that it impaired his functioning. He said that, despite better than average intelligence, Lesko was emotionally a teenager at the time of the crimes. According to Dr. Levit, Lesko suffered from diminished capacity at that time—the “ability to think clearly, rationally, logically and in a mature manner”—and his ability to abide by the law was also impaired by the influence of alcohol, drugs, and trauma. J.A. 2415-16. By 1995, however, this unstable person was, apparently, a thing of the past. Lesko’s lawyers offered the testimony of a prison chaplain, who had witnessed the past 10 years of Lesko’s life in prison and said that his behavior had been good during that time. He had not used drugs or alcohol or fought with other prisoners. Lesko had even taken an interest in religion and had been a positive influence on other prisoners. Lesko also took the stand in support of his own mitigation case. He discussed being set on fire as child, the time he was molested while shining shoes, his drug and alcohol abuse, and the way he bounced from home to home as a child. He also talked about the murders. Lesko said that, at the time, he felt nothing when killing the victims, because he was under the influence of drugs and alcohol and was upset from learning that his brother Joey had been molested. He claimed that, since being incarcerated, he had changed a great deal and now felt remorse. As in the first sentencing proceeding, the jury once again voted for death. They unanimously found four aggravating factors: that the victim was a police officer who was killed during the performance of his duties; that Lesko had a significant history of violent felony convictions; and that Lesko had previously been convicted of two offenses (murdering Levato and Newcomer) that each carried a 13 sentence of life imprisonment or death. One or more jurors found the following mitigating factors: that Lesko was under the influence of extreme mental or emotional disturbance; Lesko’s service to others on death row; Lesko’s horrible childhood; and that Lesko’s character had changed over the last 15 years of his confinement. The jury determined that the aggravating factors outweighed the mitigating and voted for death. Lesko appealed, but the judgment was affirmed. See Commonwealth v. Lesko, 719 A.2d 217, 227 (Pa. 1998) (“Lesko X”). 4. Lesko’s Second Round of PCRA Proceedings Lesko sought relief under the Pennsylvania Post Conviction Relief Act. After a six-day hearing, the PCRA court, with a different judge presiding than during Lesko’s original trial, granted relief, in part. It held (1) Lesko’s rights were violated at the 1995 resentencing when counsel did not offer expert testimony addressing whether he suffered brain damage; (2) counsel was also ineffective there by failing to adequately investigate and present a range of mitigating evidence; (3) Lesko’s right to testify was violated when counsel prevented him from testifying in his own defense at the guilt phase; (4) the prosecution suppressed evidence that would have allowed Lesko to impeach Montgomery and Rutherford in violation of Brady v. Maryland; and (5) counsel performed ineffectively by failing to adequately impeach Montgomery. The PCRA court ruled that Lesko was entitled to a new guilt- phase trial and sentencing hearing. The Supreme Court reversed and dismissed Lesko’s PCRA petition. Commonwealth v. Lesko, 15 A.3d 345, 356 (Pa. 2011) (“Lesko XIII”). It held that his guilt-phase claims, except his Brady claims, were all time-barred, and the Brady claims failed on the 14 merits. Id. at 360, 372-73. It also denied on the merits Lesko’s sentencing claims. Id. at 417. 5. Lesko’s Second Habeas Petition Lesko filed a second habeas petition—the one now at issue. He raised 22 claims, challenging both his 1981 guilt- phase trial and his 1995 resentencing. The District Court denied relief on each without an evidentiary hearing. See Lesko v. Wetzel, Civ. A. No. 11-1049, 2015 WL 249502 (W.D. Pa. Jan. 20, 2015) (“Lesko XIV”). It did, however, grant a certificate of appealability on Lesko’s claims that (1) the Commonwealth violated his Brady rights by suppressing an agreement between Montgomery and the prosecution, a January 1980 police report, and information from Rutherford’s juvenile file, and Lesko’s counsel performed ineffectively by failing to adequately investigate and cross-examine Montgomery and Rutherford; and (2) Lesko’s right to testify was violated by his trial counsel. See id. at *24, *26. Lesko moved to expand the certificate to cover additional claims. We granted the motion as to two claims—(3) counsel performed ineffectively by failing to properly investigate and present mitigating evidence at resentencing; and (4) Lesko is entitled to relief based on the cumulative effect of the purported guilt- phase and sentencing errors. Third Cir. Ord. of May 3, 2018. II. A. Jurisdiction We have jurisdiction to adjudicate Lesko’s claims pursuant to 28 U.S.C. §§ 1291 and 2253(a). However, this case presents a difficult question about the District Court’s jurisdiction. In Lesko’s present habeas petition, he challenges both his 1981 guilt-phase trial and 1995 resentencing. As to the guilt-phase, Lesko argues (1) his Brady rights were violated 15 and (2) his right to testify was obstructed by his ineffective trial counsel. As to his resentencing, Lesko argues (3) his legal team was ineffective in investigating and presenting a mitigation case and (4) he suffered prejudice from the cumulative effect of their errors. Lesko’s current habeas petition presented the first opportunity in which he could challenge his resentencing. However, he already litigated a petition contesting his guilt- phase proceedings. Under AEDPA, a district court lacks jurisdiction to consider claims presented in a “second or successive habeas corpus application” unless a court of appeals has authorized the petition in accordance with a restrictive criterion.4 28 U.S.C. § 2244(b)(2), (b)(3)(A). Lesko did not seek, let alone obtain, the requisite authorization under this portion of the statute. Thus, the question is whether the guilt phase challenges in his petition constitute a prohibited second or successive petition. If so, the District Court lacked jurisdiction to adjudicate it. In Magwood v. Patterson, 561 U.S. 320 (2010), the Supreme Court interpreted AEDPA’s second or successive provision in a related scenario. There, the defendant was 4 Specifically, a court of appeals must certify that the petition relies on either: (1) “a new rule of constitutional law, made retroactive to cases on collateral review by the Supreme Court, that was previously unavailable[,]” or (2) newly discovered facts which, if proven, would “establish by clear and convincing evidence that . . . no reasonable factfinder would have found the applicant guilty.” 28 U.S.C. § 2244(b)(2)(A)-(B). If the court of appeals finds that these requirements are not satisfied, the petition must be dismissed. Id. § 2244(b)(3)(C). 16 convicted of murder and sentenced to death in state court. Id. at 323. After exhausting state court remedies, he filed a federal habeas petition, ultimately achieving vacatur of his sentence. Id. The state court conducted an entirely new sentencing proceeding, which ended in the same result: a death sentence. Id. at 326. Magwood then filed a second habeas petition, this time attacking his new sentence on the basis that he “did not have fair warning that his offense could be punished by death.” Id. at 327. The State argued that because Magwood could have made a fair-warning argument in his initial habeas petition after his first sentencing, but did not, his petition was second or successive. Id. at 331. The Supreme Court disagreed. It reasoned that “both § 2254(b)’s text and the relief it provides indicate that the phrase ‘second or successive’ must be interpreted with respect to the judgment challenged.” Id. at 332-33. Thus, where “there is a new judgment intervening between the two habeas petitions, . . . an application challenging the resulting new judgment is not ‘second or successive’ at all.” Id. at 341-42 (internal quotation and citation omitted). Lesko’s present habeas petition indeed challenges his new 1995 sentence, but also his undisturbed 1981 conviction. The Magwood Court recognized the possibility of such a scenario. Though the prisoner there did not seek to relitigate his undisturbed conviction, the State argued that the Court’s “reading of § 2244(b) would allow a petitioner who obtains a conditional writ as to his sentence to file a subsequent application challenging not only his resulting, new sentence, but also his original, undisturbed conviction.” Id. at 342. The Court refused to address the question, saying that it would not speculate about what might arise in future cases. Id. Thus, it is an open question whether a petition like Lesko’s, which attacks 17 both a new sentence and an undisturbed conviction, runs afoul of the second or successive rule. Since Magwood was decided, five Circuits have held that the Magwood rule applies with equal force to cases like this one, concluding that a second-in-time5 petition is not second or successive as to an undisturbed conviction because a new sentence creates a new judgment which has not yet been challenged. See In re Gray, 850 F.3d 139, 141-43 (4th Cir. 2017); King v. Morgan, 807 F.3d 154, 158 (6th Cir. 2015); Insignares v. Sec’y, Fla. Dep’t of Corr., 755 F.3d 1273, 1281 (11th Cir. 2014) (per curiam); Wentzell v. Neven, 674 F.3d 1124, 1127-28 (9th Cir. 2012); Johnson v. United States, 623 F.3d 41, 45-46 (2d Cir. 2010). Despite AEDPA’s tightening of habeas relief, these courts have concluded, Magwood compels the conclusion that a prisoner who obtains relief as to his sentence may nonetheless take another bite at the apple in contesting his original conviction. See, e.g., Johnson, 623 F.3d at 46 (“[W]here a first habeas petition results in an amended judgment, a subsequent petition is not successive regardless of whether it challenges the conviction, the sentence, or both.”). Two Circuits, the Seventh and the Tenth, have ruled otherwise. In Dahler v. United States, a pre-Magwood case, the Seventh Circuit held that a habeas petition challenging an error preceding resentencing “must be treated as a collateral attack on the original conviction and sentence, rather than as an initial challenge to the latest sentence.” 259 F.3d 763, 765 (7th Cir. 2001). Because Magwood expressly left this question 5 By “second-in-time,” we mean any numerically second petition. All “second or successive” petitions are “second-in-time,” but not all “second-in-time” petitions are “second or successive” within the meaning of § 2244(b). 18 open, the Seventh Circuit has adhered to its precedent and held that a second-in-time habeas petition is barred as second or successive where it attempts to challenge a prisoner’s undisturbed conviction after he already litigated a habeas petition. See Suggs v. United States, 705 F.3d 279, 281-84 (7th Cir. 2013). For its part, the Tenth Circuit has held that a resentencing does not create a new judgment that resets AEDPA’s one-year statute of limitation, 28 U.S.C. § 2244(d)(1), a related procedural hurdle housed in the same section as the second-or-successive rule. See Prendergast v. Clements, 699 F.3d 1182, 1187 (10th Cir. 2012). Prendergast, however, relied on pre-Magwood precedent from our own Circuit without mentioning Magwood. See id. at 1186-88 (citing Fielder v. Varner, 379 F.3d 113 (3d Cir. 2004)). We have not yet weighed in on this post-Magwood debate about how to interpret § 2244, though we have acknowledged the circuit split. Romansky v. Superintendent Greene SCI, 933 F.3d 293, 299-300 (3d Cir. 2019). Now, after careful consideration, we hold that the majority interpretation of § 2244(b) is correct: a prisoner who receives relief as to his sentence is not barred from raising, in a second-in-time habeas petition, a challenge to an undisturbed conviction. Notwithstanding the troubling implications for comity and finality, we are persuaded the reasoning of Magwood compels this conclusion. The Supreme Court in Magwood held that a new “judgment” resets the count for second-or-successive purposes. See Magwood, 561 U.S. at 341-42. Before AEDPA, the Court had stated on more than one occasion that “[a] judgment of conviction includes both the adjudication of guilt and the sentence,” Deal v. United States, 508 U.S. 129, 132 (1993) superseded by statute, Pub. L. 115–391, § 403(a), 132 Stat. 5221, as recognized in United States v. Davis, 139 S. Ct. 19 2319, 2324 n.1 (2019), emphasizing that a “[f]inal judgment in a criminal case means sentence[;] [t]he sentence is the judgment,” Burton v. Stewart, 549 U.S. 147, 156 (2007) (quoting Berman v. United States, 302 U.S. 211, 212 (1937)). And Magwood makes clear that the Court continues to read it that way in the context of AEDPA. For purposes of a § 2254 petition, the Court explained, a “judgment” is that which “authoriz[es] the prisoner’s confinement,” and custody “is inextricable from the judgment that authorizes it.” Magwood, 561 U.S. at 332-33 (internal quotation omitted). Because both a conviction and sentence are necessary to authorize a prisoner’s confinement,6 and resentencing creates a new judgment authorizing a prisoner’s continued confinement, a petition challenging either component of that new judgment— be it conviction or sentence—is not second-or-successive.7 Our decision in Romansky is not to the contrary. There, we held that a habeas petitioner’s resentencing as to one count 6 True, a prisoner like Lesko whose original sentence is vacated typically remains in custody pending resentencing, but this is not because his conviction alone is sufficient to authorize his confinement. Rather, it is because a federal court has issued a conditional writ and delayed habeas relief “in order to provide the State an opportunity to correct the constitutional violation,” Hilton v. Braunskill, 481 U.S. 770, 775 (1987), and to “replace an invalid judgment with a valid one,” Wilkinson v. Dotson, 544 U.S. 74, 87 (2005) (Scalia, J., concurring). 7 We interpret “judgment” only as that term appears in the federal habeas statutes. We offer no view on what might constitute a new or intervening judgment for purposes of other statutes not presently before us. 20 of conviction “did not impose a new judgment” as to other counts for which both the conviction and sentence remained undisturbed, see Romansky, 933 F.3d at 300-01, and each count of conviction for which a separate sentence was imposed authorized its own confinement, each constituting a distinct “judgment” for purposes of AEDPA. The Fifth Circuit has adopted a similar approach, holding that, because vacating a conviction and sentence for a lesser included offense did not disturb either the conviction or sentence for the greater offense, it did not constitute a new judgment.8 See In re Lampton, 667 F.3d 585, 588-89 (5th Cir. 2012); see also Patterson v. Sec’y Fla. Dep’t of Corr., 849 F.3d 1321, 1325-26 (11th Cir. 2017) (holding that only changes to sentences that authorize new confinements constitute new judgments under Magwood). Romansky thus stands for the proposition that a resentencing as to one count of a conviction that makes no changes to the confinements authorized by the other undisturbed counts does not affect the “judgment” as to those counts and, hence, does not reset the habeas counter. 933 F.3d at 300. What we confront today, however, presents a different question, one we anticipated in Romansky. There, we suggested that where a court “undertake[s] a de novo 8 In contrast, the Second and Ninth Circuits treat the convictions and sentences for multiple counts as “components” of a single judgment such that a change to any resets the habeas counter for all. See Wentzell, 674 F.3d at 1127; Johnson, 623 F.3d at 46. We see no reason, however, to treat separate counts of conviction and sentences imposing separate confinements as components of a single judgment merely because they were handed down at the same time; rather, consistent with Romansky, we will treat these are separate and distinct judgments for purposes of AEDPA. 21 resentencing as to all counts of conviction if any count is vacated on appeal . . . the resentencing might constitute a new judgment as to every count of conviction.” Id. As we have explained here, that reasoning is compelled by Magwood. Resentencing creates a new judgment as to each count of conviction for which a new or altered sentence is imposed, while leaving undisturbed the judgments for any counts of conviction for which neither the sentence nor the conviction is changed. Lesko was resentenced as to all counts of his conviction; as a result, his guilt-phase claims are not barred as second-or-successive. One of Lesko’s guilt-phase claims is not barred for a second reason as well. In his present petition, Lesko contends that his counsel rendered ineffective assistance by preventing him from exercising his Sixth Amendment right to testify at his 1981 guilt-phase trial. That same lawyer, Rabe Marsh, represented Lesko for nearly twenty years, including at Lesko’s first guilt and sentencing trials, first direct appeal, first PCRA proceedings, and first federal habeas proceedings, along with his 1995 resentencing and the ensuing direct appeal. It was only after Marsh withdrew in 1999 and Lesko obtained new counsel that Lesko for the first time raised the issue of Marsh’s ineffectiveness at trial. The PCRA court granted a new trial and sentencing, but the Pennsylvania Supreme Court reversed, denying his ineffective assistance claim—and all but one of his other claims—as untimely. Lesko XIII, 15 A.3d at 359-73. Now, in his second round of federal habeas proceedings, Lesko attempts to advance this claim once more. Recognizing these circumstances to be unique, we requested additional briefing on how the term “second or successive” should be construed in such a situation and whether Marsh, having continuously represented Lesko since the guilt-phase trial, could be expected to argue his own 22 ineffectiveness either in Lesko’s state PCRA or initial federal habeas proceedings. Lesko argued that he could not. We agree. In Martinez v. Ryan, the Supreme Court warned that “[a] prisoner’s inability to present a claim of trial error is of particular concern when the claim is one of ineffective assistance of counsel.” 566 U.S. 1, 12 (2012). And we, along with several of our sister circuits, have recognized that lawyers cannot be expected to argue their own ineffectiveness. See United States v. Cocivera, 104 F.3d 566, 570 (3d Cir. 1996); see also Harris v. Comm’r, Ala. Dep’t of Corr., 874 F.3d 682, 690 (11th Cir. 2017) (observing that a lawyer’s “personal interest in not being found to have performed ineffectively . . . conflicts with the interests of a client asserting a claim based on his lawyer’s ineffectiveness”) (internal quotation marks omitted); Combs v. Coyle, 205 F.3d 269, 276 (6th Cir. 2000) (“[C]ounsel cannot be expected to raise his own ineffectiveness on appeal . . . .”); United States v. Del Muro, 87 F.3d 1078, 1080 (9th Cir. 1996) (agreeing that “forcing trial counsel to prove his own ineffectiveness” created a conflict of interest); Ciak v. United States, 59 F.3d 296, 303 (2d Cir. 1995) (“We cannot expect ineffective assistance claims to be raised on direct appeal—and therefore we should not penalize a petitioner for failing to raise them—when a petitioner’s counsel on direct appeal also represented him at trial.”). Here, attorney Marsh was operating under a conflict of interest which effectively prevented him in either the state PCRA or initial federal habeas proceedings from raising a claim that he interfered with his client’s right to testify. “Advancing such a claim would have required [counsel] to denigrate [his] own performance”—something he “cannot reasonably be expected to” do, as it would “threaten[] [his] professional reputation and livelihood.” Christeson v. Roper, 23 574 U.S. 373, 378 (2015); see also Maples v. Thomas, 565 U.S. 266, 285 n.8 (2012) (explaining that a “significant conflict of interest” arises when a lawyer’s “interest in avoiding damage to [his] own reputation” is at odds with the petitioner’s “strongest argument”); Model Rules of Prof’l Conduct 1.7 cmt. 10 (“[I]f the probity of a lawyer’s own conduct in a transaction is in serious question, it may be difficult or impossible for the lawyer to give a client detached advice.”). Our conclusion that attorney Marsh could not have been expected to raise a claim as to his own ineffectiveness at trial has significant implications for our second or successive analysis. The Supreme Court has held that “the § 2244(b) restrictions simply do not apply to” certain claims based on a prisoner’s inability to raise them in an initial habeas petition. Stewart v. Martinez-Villareal, 523 U.S. 637, 642 (1998). Likewise, we have previously stated that “a subsequent petition . . . is clearly not a ‘second or successive’ petition within the meaning of § 2244 if the claim had not arisen or could not have been raised at the time of the prior petition.” Benchoff, 404 F.3d at 817. That is the case here: because Marsh was operating under a conflict of interest when he filed Lesko’s first habeas petition, the ineffectiveness claim “could not have been raised at [that] time,” id, and Lesko’s first opportunity to raise his counsel’s ineffectiveness was in his second-in-time petition. Put differently, if Lesko’s petition were barred as second-or-successive, he would, as a practical matter, have never had a chance to bring that claim. Yet “[t]he right to the effective assistance of counsel at trial is a bedrock principle in our justice system.” Martinez, 566 U.S. at 12. We therefore decline to interpret § 2244(b) in a way that allows for an ineffective assistance counsel claim to completely evade federal habeas review. Instead, we hold that a second-in-time habeas petition is not second or successive to the extent it raises 24 an ineffective assistance of counsel claim that the prisoner lacked opportunity to raise because the same counsel represented him both at trial and in his first round of habeas proceedings.9 As Lesko’s current habeas petition is not second or successive with respect to his guilt-phase claims, the District Court had jurisdiction to decide it, and we may proceed to address the merits of his claims. B. Lesko’s claim that his Brady rights were violated We now turn to the merits of those guilt-phase claims. Lesko first argues that the Commonwealth violated his Brady rights by withholding evidence that he could have used to impeach Montgomery and Rutherford. A Brady violation occurs when the defendant demonstrates that (1) the evidence was favorable to the accused, either because it was exculpatory or had impeachment value; (2) the government withheld the evidence, either willfully or inadvertently; and (3) the evidence was material. Dennis v. Sec’y, PA Dep’t of Corr., 834 F.3d 263, 284–85 (3d Cir. 2016) (en banc). This claim involves three pieces of evidence that the Commonwealth did not produce before trial: The first is a copy of an “agreement and statement of intent” between the Westmoreland County District Attorney 9 We stress the narrowness of this holding. We simply hold that in the rare circumstance where a conflict of interest— like ripeness, Stewart, 523 U.S. at 644-45, or a procedural bar, Slack v. McDaniel, 529 U.S. 473, 487-88 (2000)—makes it impossible to bring a claim of trial ineffectiveness in a first habeas petition, that claim is not second or successive when raised in a subsequent petition with new counsel. 25 and Montgomery in which the district attorney agreed that he would not prosecute Montgomery for any crimes against property that were alleged to have occurred in December 1979 or January 1980. In exchange for this promise, Montgomery agreed to advise and testify as a witness about the deaths of Miller, Newcomer, and Levato. The second piece of evidence is a police report prepared by Trooper Michael K. Steffee of the Pennsylvania State Police on January 8, 1980, recording his interview with Montgomery. This interview occurred before Montgomery began to cooperate with authorities. At this time, Montgomery told Steffee that after he joined Lesko and Travaglia in the hotel room, Travaglia asked if he wanted a gun, and “[t]hat is when I should have known something was wrong.” J.A. 4734. Montgomery went on to say, “I don’t know anything about the cop getting shot or any armed robberies.” Id. The last is Rutherford’s juvenile file. Lesko believes that three pieces of this file are particularly important: (1) notes reporting that on November 27, 1980, and December 29, 1980, Rutherford was allowed to leave the Juvenile Detention Center for five hours to spend time with his family; (2) a copy of a letter from one of the prosecutors to Rutherford’s attorney stating that “a disposition of this case prior to the completion of the first trial in the matter of Michael Travaglia and John Lesko would be too risky from a prosecution standpoint,” id. at 4755; (3) an “Application to Transfer Case to Juvenile Court,” in which Rutherford’s counsel claimed that transfer was necessary because Rutherford’s “medical and psychiatric condition require that, if convicted, he not be confined to custody along with adults and that he be given adequate opportunity for medical and psychiatric treatment.” Id. at 4762. 26 Lesko argues that, had the prosecution disclosed this evidence, he would have been able to mount more effective cross-examinations of Montgomery and Rutherford, which would have helped his case in both the guilt and sentencing phases. The PCRA Court concluded that the Commonwealth violated Brady with respect to these items. The Court explained that this evidence would have called into question testimony from Montgomery and Rutherford and undermined the case that Lesko possessed the requisite intent for first- degree murder. On appeal, the Pennsylvania Supreme Court determined that Lesko’s claims concerning Montgomery’s agreement and Rutherford’s juvenile file were time-barred, see Lesko XIII, 15 A.3d at 371, and that the Steffee report was not material, see id. at 372. The Supreme Court acknowledged that Montgomery’s earlier statement that he knew nothing about a police officer’s being shot would have been useful to impeach his testimony that Travaglia had admitted to shooting an officer and that Lesko had said he wanted to. However, the Court concluded that “Montgomery’s earlier non-cooperation with police would not have made his in-court testimony disappear, nor would it have altered the overall volume of abundant, independent evidence offered at the 1981 trial establishing Lesko’s course of conduct and intent.” Id. The District Court, addressing the claims in the § 2254 proceedings, agreed that the withheld evidence was not material. Starting with Rutherford’s file, the Court found the fact Rutherford received two short furloughs to be insignificant. It also concluded that the attorney’s statement about psychiatric conditions was mere boilerplate and that Lesko had not shown that Rutherford had any such condition. See Lesko XIV, 2015 WL 249502, at *18. The Court further ruled that the prosecutor’s letter did not say anything that 27 Lesko did not already know because Rutherford testified at some length that he was facing charges for murder and that the district attorney had agreed that his case would be disposed of in juvenile court if he testified truthfully in Lesko’s case. See id. The Court also determined that Officer Steffee’s report was cumulative of two other reports of police interviews with Montgomery before he began cooperating in which he also did not say that Lesko had made any incriminating comments. See id. at *19. Likewise, the Court concluded that Montgomery’s agreement was not material because his deal with the prosecution was discussed in a hearing that was transcribed and provided to Lesko, see id., and that Montgomery’s testimony was of limited importance at trial, further reducing the value of any evidence that could have been used to impeach him, see id. at 20. Here, Lesko argues that the crucial issue at trial was his intent—i.e., that the prosecution in large part sought to prove his intent through testimony from Montgomery and Rutherford, and that the withheld evidence was material because it would have caused the jury to question that testimony. More specifically, Lesko contends that “[h]ad the Steffee Report and the written agreement been disclosed, defense counsel could have argued that Montgomery’s testimony was fashioned to ensure the benefit of an extremely favorable immunity deal, because Montgomery’s initial account of his interactions with Lesko—before the deal was worked out—was devoid of inculpatory evidence.” Br. at 47. And “[t]he defense could have made similar arguments about Rutherford’s testimony had the full benefits of his deal been disclosed, i.e., that he literally walked away from two murder charges shortly after Lesko’s trial.” Id. at 54. Lesko also argues that “the juvenile file could have been used as evidence of 28 Rutherford’s psychiatric illness to challenge his credibility and his ability to accurately report events.” Id. The Commonwealth’s argument largely tracks the District Court’s analysis. It contends that any inquiry into Rutherford’s psychiatric condition would not have been fruitful, that there is no evidence his brief furloughs affected his testimony, that the Steffee report is cumulative, and that the impeachment evidence concerning Montgomery is of limited consequence. Br. at 39. The Commonwealth also argues that “it should have been clear to [Lesko] that it is common practice to keep pending charges open until the terms of the bargain have been fulfilled, in this case, Rutherford’s testimony,” Br. at 46, and that there was no secret deal to drop the charges after Lesko’s trial, see id. at 46–47. The key question is whether this undisclosed evidence was material. The Pennsylvania Supreme Court concluded that Officer Steffee’s report was not material and that decision is entitled to AEDPA deference. See Harrington, 562 U.S. at 100. Because the Supreme Court disposed of the other parts of the claim—i.e., the written non-prosecution agreement with Montgomery and the contents of Rutherford’s juvenile file— on purely procedural grounds, we will review those aspects of the claim de novo. See Thomas 570 F.3d at 114. “The ‘touchstone of materiality is a reasonable probability of a different result.’” Dennis, 834 F.3d at 285 (quoting Kyles v. Whitley, 514 U.S. 419, 434 (1995)). “The materiality of Brady material depends almost entirely on the value of the evidence relative to the other evidence mustered by the state.” Johnson v. Folino, 705 F.3d 117, 129 (3d Cir. 2013) (quoting Rocha v. Thaler, 619 F.3d 387, 396 (5th Cir. 2010)). Nevertheless, “[t]he question is not whether the defendant would more likely than not have received a different 29 verdict with the evidence, but whether in its absence he received a fair trial, understood as a trial resulting in a verdict worthy of confidence.” Kyles, 514 U.S. at 434. We consider materiality based on the collective value of the withheld evidence. See Dennis, 834 F.3d at 312. At the outset, the guilt-phase evidence against Lesko was reasonably strong, albeit largely circumstantial. Under Pennsylvania law, the elements of first-degree murder are “that a human being was unlawfully killed, that the accused was responsible for the killing, and that the accused acted with a specific intent to kill.” Commonwealth v. Pagan, 950 A.2d 270, 279 (Pa. 2008). Since Travaglia was the one to pull the trigger, the Commonwealth pursued an accomplice-liability theory against Lesko. “To establish Lesko’s accomplice culpability for first degree murder, the Commonwealth was required to prove that petitioner, with the intent of promoting or facilitating the commission of first degree murder, aided, agreed, or attempted to aid Travaglia in planning or committing the murder of Officer Miller.” Lesko VI, 881 F.2d at 53; see also Pagan, 950 A.2d at 279. Intent can be proved solely through circumstantial evidence. See Pagan, 950 A.2d at 279. The most compelling evidence of Lesko’s intent— which is what Lesko seeks to attack here—came from the surrounding context. As we previously stated, in explaining the importance of the evidence concerning the Nicholls killing— The jury could only have fairly evaluated the Commonwealth’s theory regarding Lesko’s state of mind by hearing evidence tending to show that Travaglia and Lesko had jointly embarked that evening on a crime spree, that they had already committed a homicide likely to command the 30 death penalty, and that they had in their possession powerful evidence of their guilt of that homicide. Moreover, to be in a position to evaluate Lesko’s state of mind during the critical moments during the Miller encounter, the jury needed to hear sufficient details about these matters to be able to appreciate the nature of the evening’s joint undertaking, the relationship and mood of the participants, and the extent of the criminal exposure of those participants in their apprehension by Miller. Lesko VI, 881 F.2d at 54. This evidence—as well as the fact that Lesko and Travaglia stopped to get another gun and Lesko determined that birdshot was insufficient for their purposes— strongly points toward Lesko’s guilt. That evidence was reinforced by two statements from Lesko. First, Rutherford testified that, when the car chase with Miller began, Lesko warned him to “lay down in the back, because it might turn into a shooting gallery.” J.A. at 589. This statement, of course, tended to show that Lesko anticipated that the episode with Miller would turn violent. Second, Montgomery testified that, after Travaglia told him that he had shot a cop, Lesko said, “I wanted to.” Id. at 680. This comment, although made after the fact, also supported the notion that the killing was a joint enterprise. Lesko argues that the three pieces of Brady evidence were material because they would have helped him impeach Montgomery and Rutherford and thus cause the jury to disbelieve their testimony. We will discuss the significance of each piece of evidence in turn, then consider the importance of the evidence in combination. See generally Kyles, 514 U.S. at 437 n.10 (“We evaluate the tendency and force of the 31 undisclosed evidence item by item. . . We evaluate its cumulative effect for purposes of materiality separately and at the end of the discussion[.]”). 1. Montgomery’s Non-Prosecution Agreement Is Not Material In this agreement, the Commonwealth promised that, in exchange for Montgomery’s testimony in the Lesko trial, it would not prosecute him for his involvement in property crimes that occurred in December 1979 or January 1980. Lesko argues the existence of this deal could damage Montgomery’s credibility. See, e.g., Giglio v. United States, 405 U.S. 150, 154–55 (1972); Simmons v. Beard, 590 F.3d 223, 236 (3d Cir. 2009). The force of this evidence is weakened by the fact that Lesko possessed a copy of a transcribed hearing involving Montgomery and the assistant district attorney that covered the same ground. In that hearing, Montgomery affirmed that his attorney had explained to him “the extent of the agreement and statement of intent,” which “pertain[ed] to crimes and offense generally described as offenses against property, such as burglary, robbery and theft and similar crimes.” J.A. at 4444– 45. Montgomery went on to acknowledge that he was required to answer the prosecution’s questions truthfully, and then gave his account of his interaction with Lesko and Travaglia, which was similar to his testimony at trial.10 10 Montgomery also testified that Lesko had told him the day before the shooting that he and Travaglia had a contract to kill a cop, which Montgomery said “scare[d] the hell out of [him],” J.A. at 4461, and that Lesko stated, after the Officer Miller killing, “I would shoot my own mother if the price was 32 Thus, the suppressed evidence is cumulative of impeachment evidence that the Commonwealth did produce. See Johnson, 705 F.3d at 129 (“Suppressed evidence that would be cumulative of other evidence . . . is generally not considered material for Brady purposes.”). Although the agreement is somewhat more precise than the hearing testimony, the marginal value of this added specificity does not undermine the fairness of the trial. See, e.g., Landano v. Rafferty, 856 F.2d 569, 574 (3d Cir. 1988) (considering impeachment evidence immaterial under Brady where the “marginal effect in diminishing [the witness’s] perceived credibility would have been negligible”). Moreover, Montgomery’s testimony, while helpful to the prosecution, was not vital. Rutherford’s testimony was much more important in establishing Lesko’s intent than Lesko’s after-the-fact statement to Montgomery. Accordingly, the evidence about the agreement and the statement of intent is not material. See, e.g., Smith v. Cain, 565 U.S. 73, 76 (2012) (“We have observed that evidence impeaching an eyewitness may not be material if the State’s other evidence is strong enough to sustain confidence in the verdict.”). 2. Officer Steffee’s Report is Not Material In this document, Officer Steffee wrote that, when questioned by police on January 8, 1980, Montgomery said, “I don’t know anything about the cop getting shot or any armed robberies.” J.A. 4734. This, of course, contradicts Montgomery’s testimony at trial that Travaglia had told him that he had shot a cop and that Lesko had said he wanted to, and thus would have been useful impeachment evidence. right.” Id. at 4466. For reasons that are not provided in the record, Montgomery did not mention these statements at trial. 33 Again, however, the Commonwealth produced similar materials to Lesko: police reports from January 4, 1980, and January 5, 1980, in which Montgomery described how Travaglia had given him the gun without mentioning a shooting. Lesko argues that Officer Steffee’s report would have been somewhat more useful on cross-examination than the other two reports because, unlike the other reports, it would have allowed the defense to cross Montgomery about a contradiction rather than a mere omission. We cannot agree that Officer Steffee’s report is material because in its absence, the trial still “result[ed] in a verdict worthy of confidence.” Kyles, 514 U.S. at 434. The trajectory of Montgomery’s statements is clear: he initially denied knowing anything about Officer Miller’s shooting then, after obtaining counsel and entering into a cooperation agreement, changed course and testified that Lesko and Travaglia made incriminating statements.11 The defense could have made this point without Officer Steffee’s report; counsel could have asked, for instance, “When the police interviewed you on January 4 (or 5), 1980, you didn’t tell them that Lesko stated, ‘I wanted to,’ did you? You added these details to your statement after you entered into an agreement with the district attorney, right?” Those questions would have made the same point—that Montgomery should not be believed because he 11 At the PCRA hearing, Montgomery readily admitted to having changed his statements. When asked whether his statement that Lesko and Travaglia had not told him about killing anyone was true, he flatly responded, “No sir, that’s not so, that they didn’t tell me that, because, yes, sir, it was told.” J.A. 3691. 34 tailored his testimony to curry favor with the prosecution—as any questions based on Officer Steffee’s report. See generally United States v. Georgiou, 777 F.3d 125, 140 (3d Cir. 2015) (“To the extent Appellant argues that additional information about the intensity or duration of Waltzer’s substance abuse may have impacted the trial’s outcome, he explains neither why he did not probe these issues more fully on cross examination, nor why such new information would have changed the trial’s outcome when the substance abuse evidence that was set out at trial did not.”). In short, the added value of Officer Steffee’s Report is too insignificant to render it material, a conclusion which is reinforced by the relatively minor importance of Montgomery’s testimony in the overall scheme of trial. 3. The Three Items in Rutherford’s Juvenile File Are Not Material Lesko next argues that, had the prosecution produced Rutherford’s juvenile file, he could have impeached him based on (1) his receiving two five-hour furloughs, (2) his attorney’s stating in a legal filing that he had a psychiatric condition, and (3) the prosecutor’s stating that the juvenile case should not be disposed until after Lesko’s trial. The first two items are not compelling. First, the jury was informed that Rutherford was cooperating with the Commonwealth and that, in return for his cooperation, his case would proceed in Juvenile Court. It is unlikely that the fact that he also received two short furloughs would have caused the jury to view his testimony any differently. See United States v. Walker, 657 F.3d 160, 186 (3d Cir. 2011) (concluding that evidence of further non-prosecution was not material on the ground that it would have “little, if any, probative value because it is impeachment by the same avenue already taken 35 by the defendants, namely Rhoades’s motivation for testifying against the Walkers as part of a bargained-for reduction in criminal penalties” (quotation marks omitted)). Second, the pertinent statement from Rutherford’s Application to Transfer Case to Juvenile Court, in full, provides that “the applicant’s medical and psychiatric condition require that, if convicted, he not be confined to custody along with adults and that he be given adequate opportunity for medical and psychiatric treatment.” J.A. 4762. This was one of four grounds that counsel claimed justified transferring the proceedings to Juvenile Court; the filing did not further elaborate upon any psychiatric condition. This document would have been sufficient to permit Lesko’s counsel to have asked Rutherford about his psychiatric condition on cross-examination. In general, counsel must have a “good faith basis” to ask a question on cross, see Commonwealth v. May, 887 A.2d 750, 763 (Pa. 2005), and this filing meets that standard. See generally Cogley v. Duncan, 32 A.3d 1288, 1292 (Pa. Super. Ct. 2011 (“The general rule is that admissions of fact in pleadings are admissible, but that the pleader’s conclusions of law are not admissions of facts in issue.”) (citation omitted); cf. J.A. at 4150 (prosecutor’s arguing at the PCRA hearing that “I would object to that, Your Honor, because I don’t know how he would cross examine him on an application that’s made by his lawyer and is not signed by the witness himself.”). Nevertheless, evidence of a witness’s mental health will be material only if it “undermines [the witness’s] reliability []or calls into question his ability to perceive, remember and narrate perceptions accurately.” Georgiou, 777 F.3d at 141 (citation and quotation marks omitted). Lesko has possessed this document since at least 2002 but has never presented any 36 evidence that Rutherford suffered from a psychiatric condition that would affect his ability to testify accurately. It is therefore pure speculation to conclude that this document, either on its own or as a way of prompting counsel to investigate the matter further, would have served to undermine Rutherford’s credibility. This is not enough to show materiality. See Riley v. Taylor, 277 F.3d 261, 302 (3d Cir. 2001) (holding that evidence was not material because its exculpatory nature was too speculative); United States v. Brown, 250 F.3d 811, 817– 818 (3d Cir. 2001) (concluding that evidence was not material in part because it showed only “the possibility of an alibi defense” rather than “demonstat[ing] the [the petitioner] had an alibi”). Lesko’s argument about the letter from the prosecutor to Rutherford’s counsel is somewhat more complicated. The Commonwealth interprets the letter to say only that Rutherford’s case in Juvenile Court would remain open until after Lesko’s case had concluded. That fact was squarely addressed at trial, so having another document making the same point could not have affected the verdict. See, e.g., Walker, 657 F.3d at 186 (noting that cumulative impeachment evidence is not probative). Lesko also argues that the letter reflects an agreement that all charges against Rutherford would be dismissed after Lesko’s trial. We disagree. The letter purports to confirm a telephone conversation between Rutherford’s attorney (John Murtagh) and the prosecutor; the prosecutor states that it was his “[f]eeling that a disposition of this case prior to completion of [Lesko’s trial] would be too risky from a prosecution standpoint.” J.A. 4755. The relevant section of the Juvenile Act (both now and at the time the prosecutor wrote the letter) refers to the outcomes of cases involving delinquent children as “dispositions.” See 42 Pa. Cons. Stat. § 6352 (in section titled 37 “Disposition of delinquent child,” stating that “[i]f the child is found to be a delinquent child the court may make any of the following orders of disposition.”). Authorized “dispositions” range from returning the child home with new conditions to committing the child to an institution. See id. Thus, the most natural reading of the letter is that it refers simply to resolving Rutherford’s case, not dismissing the charges, as Lesko argues here. This interpretation is also supported by the other evidence in the record. The letter followed a phone call between the prosecutor and Rutherford’s attorney, Murtagh. In an affidavit, Murtagh later stated that “[b]ased on [the prosecutor’s] communication with me, I understand that the Commonwealth believed that, if Mr. Rutherford’s case was resolved before he testified against Mr. Lesko and Mr. Travaglia, there was a substantial chance he would not testify against them.” April 3, 2002, affidavit. This statement is consistent with the Commonwealth’s interpretation of the letter, not Lesko’s. Murtagh’s testimony at the PCRA hearing was similar. Likewise, at the 1995 resentencing, Lesko’s attorney asked Rutherford, “And you received no punishment for these two murders because you made a deal with the District Attorney, didn’t you?” J.A. 2165. Rutherford responded, “The deal was that I would be tried as a juvenile. There was no other deal.” Id. In sum, Lesko has not shown that there was any agreement to dismiss the charges against Rutherford. Accordingly, this Brady claim also fails. See Simon v. Gov’t of the V.I., 929 F.3d 118, 127 (3d Cir. 2019) (“[F]avorable treatment alone is insufficient to state a Brady claim.”); Shabazz v. Artuz, 336 F.3d 154, 165 (2d Cir. 2003) (“The government is free to reward witnesses for their cooperation with favorable treatment in pending criminal cases without 38 disclosing to the defendant its intention to do so, provided that it does not promise anything to the witnesses prior to their testimony.”).12 4. The Cumulative Prejudice Would Not Have Affected The Verdict Viewing the evidence in combination does not change our analysis in a meaningful way. Even if the prosecution had produced Montgomery’s agreement and Officer Steffee’s report, it is unlikely that the jury would have doubted Montgomery’s statement or changed its view of Lesko’s guilt. The same is true as to Rutherford’s juvenile file. His furloughs and his attorney’s statement concerning possible psychiatric issues are of only trivial value as impeachment evidence. Moreover, the natural interpretation of the letter between the prosecutor and Rutherford’s attorney memorialized only an agreement to wait to resolve the juvenile case until after Lesko’s trial—information that was already shared with the jury. Accordingly, Lesko has failed to show that this evidence is material even when viewed collectively. 5. The Suppressed Evidence Did Not Affect The Sentencing Phase Lesko also argues that this evidence was material at the sentencing phase. Again, we disagree. If anything, Lesko’s argument is weaker in this context. The jury had already found that Lesko was guilty of first-degree murder. At sentencing, the 12 As previously noted, the fact that the Commonwealth had promised to resolve Rutherford’s case in Juvenile Court after Lesko’s trial was cumulative of evidence already presented to the jury. Thus, to the extent that this letter merely restates that point, it was not material. See Johnson, 705 F.3d at 129. 39 jury was focused on the statutorily prescribed aggravating and mitigating factors, and it sentenced Lesko to death because it concluded that the aggravating factors (that he committed several previous violent felonies and killed two other individuals before Officer Miller’s murder) outweighed the mitigating factors (that he was under the influence of extreme mental or emotional disturbance, had an awful childhood, and had behaved well in prison). It is our opinion that this relatively weak impeachment evidence would not have affected the jury’s overall balance of those factors.13 13 In its COA grant, the District Court authorized Lesko to appeal the denial of his claim that counsel had performed ineffectively by failing to properly cross-examine Montgomery and Rutherford. However, Lesko has not presented any argument in support of that claim in his brief and has therefore forfeited the claim. See, e.g., In re Wettach, 811 F.3d 99, 115 (3d Cir. 2016). Regardless, the claim lacks merit. It is primarily directed toward counsel’s alleged deficient performance in failing to cross-examine Montgomery more fully. However, it is easy to see why counsel would have kept his cross-examination of Montgomery brief: on direct examination, Montgomery did not include the much more damaging comments he made during the hearing about Lesko’s saying that he had a contract to kill a police officer and that he would kill his mother if the price were right. Further, there was some indication that Montgomery had participated in prior robberies with Lesko and Travaglia, and counsel could also have reasonably wished to avoid informing the jury about those incidents. 40 C. Lesko’s Claim That Trial Counsel Was Ineffective By Violating His Right To Testify On his second guilt-phase claim, Lesko contends his counsel prevented him from testifying at his trial. The possibility that Lesko would testify was first raised to the court near the end of trial, when Lesko’s lawyer, Marsh, stated: I would like to reiterate on the record what I advised the Court this morning and off the record this afternoon; that it was my advice to John Lesko that if Michael Travaglia did not testify that he should not testify. I have many good reasons for it, both legal and tactical, and I am not sure whether my client is going to follow my advice. And I will need to spend some time with him to find out what he wants to do. Hopefully, he will follow my advice. J.A. 1222-23. Travaglia conferred with his counsel and decided not to testify. Marsh then said: I have discussed with my client the possibility of his taking the witness stand in his own behalf, so he knows he has that right. He told me he does wish to take the witness stand in his own behalf. And I advised him that that is the wrong thing to do in this case for legal and tactical reasons. He has told me he will follow my advice reluctantly, but he will follow my advice. J.A. 1227. Lesko did not testify. Lesko claims that his lawyer failed to properly advise him that it was his decision whether to testify. Before the PCRA court, Lesko testified that counsel “never discussed with me that I had the right to testify,” even though Lesko expressed 41 his desire to do so. J.A. 4241. Lesko said he wanted to refute Rutherford’s testimony that Lesko had said that the car was about to become a shooting gallery shortly before Officer Miller was killed, and also Montgomery’s testimony that Lesko had said that he wanted to shoot the officer. According to Lesko, Marsh closed the defense case without his consent and without giving him a chance to take the stand. Marsh offered little response to these allegations. In the PCRA proceedings, he testified that he had no “independent recollection” of Lesko’s desire to testify. J.A. 2874. The PCRA court granted relief to Lesko, concluding that Lesko’s recounting of how he wanted to testify was credible. The court found that Lesko was prejudiced by counsel’s alleged interference because he could not “rebut the testimony of Mr. Montgomery and Mr. Rutherford that [Lesko] made statements that demonstrated his intent to commit first degree murder.” J.A. 168. The Commonwealth appealed. The Pennsylvania Supreme Court held that the claim was untimely and did not address its merits. See Lesko XIII, 15 A.3d at 359-60. Lesko then asserted the claim in his federal habeas proceedings. The District Court assumed counsel had performed deficiently by preventing Lesko from testifying but concluded that he had not been prejudiced. It reasoned that even if Lesko took the stand to deny making the statements Rutherford and Montgomery attributed to him, he would not have been able to refute the other evidence establishing his intent, “including the details of the horrific crime spree the defendants embarked upon when they abducted Nicholls, then murdered him, then stole the handgun that would subsequently be used to kill Officer Miller, then stole bullets that could be used with the gun, and then goaded Officer Miller into chasing them.” Lesko XIV, 2015 WL 249502, at *25. The Court further determined that it was 42 unlikely that Lesko’s testimony about his side of the story would have meaningfully advanced his defense. That is, because Lesko’s statement to police after his arrest was admitted as evidence, the jury already knew that he claimed the plan had been only to draw Officer Miller away from the convenience store to rob it; the jury simply did not credit that account. See id. at *26. Before us, Lesko once again claims that his counsel rendered ineffective assistance by interfering with his right to testify. Under Strickland v. Washington, 466 U.S. 668 (1984), Lesko must show that counsel performed deficiently and that he was prejudiced by that deficiency. Palmer v. Hendricks, 592 F.3d 386, 394 (3d Cir. 2010). Because the Pennsylvania Supreme Court rejected this claim as time-barred and did not address its merits, AEDPA’s deferential standards do not apply. See id. at 400. However, the PCRA court’s factual findings—that Lesko told Marsh he wanted to testify but Marsh overrode Lesko’s decision—are still presumed to be correct under § 2254(e)(1). See Nara v. Frank, 488 F.3d 187, 201-02 (3d Cir. 2007). It is an open legal question whether the Commonwealth, as opposed to a habeas petitioner, may rebut the presumption of correctness. See id. at 202. Today, however, we need not decide the question. Even if we were to accept the PCRA court’s findings and conclude that counsel performed deficiently by interfering with his client’s right to testify, Lesko cannot prevail. He fails to establish prejudice. See Palmer, 592 F.3d at 394 (courts may resolve a Strickland claim by concluding that prejudice has not been established). In evaluating prejudice, we consider Lesko’s proposed testimony in the context of the other evidence presented at trial. See id. at 399. Lesko argues that if he was allowed to testify, 43 there is a reasonable probability that the result would have been different, because he “would have spoken to the central dispute at trial, would have explained his state of mind and his relatively minor role in the offense, and would have undercut the credibility of the principal witnesses against him,” Rutherford and Montgomery. Br. 59. There is no question that a defendant’s own testimony is significant. We have stated that it “is very likely to be highly important” and, “‘as a general matter, it is only the most extraordinary of trials in which a denial of the defendant’s right to testify can be said to be harmless beyond a reasonable doubt.’” Palmer, 592 F.3d at 399 (alteration omitted) (quoting Martinez v. Ylst, 951 F.2d 1153, 1157 (9th Cir. 1991)). At the same time, a finding of prejudice is hardly foreordained. In Palmer, we recognized the importance of the defendant’s testimony but nevertheless concluded that the defendant failed to establish prejudice. See id. at 394. There, counsel failed to advise the defendant that he controlled the decision whether to testify. See id. The defendant later asserted that had he been properly advised, he would have told the jury his side of what happened when a bar fight broke out and he fired a gun, killing one person and injuring another. See id. at 388, 390. That conclusory assertion, however, was insufficient to establish prejudice. A mere “desire to tell his side of the story” was not enough on its own. Id. at 395. Nor was his “conclusory invocation of the words ‘self-defense”—an “unadorned legal conclusion”—without detailed factual analysis of how the testimony would have swayed a juror. Id. We also rejected the contention that prejudice should be presumed or a reversal automatic when a defendant’s right to testify has been violated. See id. at 397. Rather, we applied the Strickland standard, which requires that a defendant “affirmatively prove prejudice.” Id. at 398 (quoting Strickland, 466 U.S. at 693). 44 Palmer is instructive. Lesko has repeatedly asserted that it was critical for him to testify that he did not say the words that Rutherford and Montgomery attributed to him. However, a “stated desire to tell his side of the story ‘falls far short of satisfying Strickland’s prejudice element.’” Id. at 400 (quoting Sayre v. Anderson, 238 F.3d 631, 635 (5th Cir. 2001)). Lesko does go farther than the Palmer petitioner, saying he would have denied Rutherford’s and Montgomery’s statements, which the Commonwealth relied upon to show he possessed the intent to commit first-degree murder. But ultimately, he fails to show prejudice. At the outset, is it questionable that the jury would believe Lesko’s self-interested denials rather than the testimony of Rutherford and Montgomery, especially because Lesko apparently did not plan to contest the veracity of any other part of their detailed factual narratives. Moreover, the record indicates that Lesko would not have been a persuasive witness. Brian O’Leary, co-counsel at Lesko’s resentencing, who spent hours interviewing Lesko and saw him testify at his 1995 resentencing, has since described Lesko as having “a flatness to his personality I had never encountered in a human being before.” J.A. 3127. And, according to O’Leary, the “pivotal moment” where the mitigation case was lost was when Lesko crumbled under cross-examination. J.A. 3127. A lawyer’s judgment about how effective his client’s hypothetical testimony may have been is meaningful in evaluating prejudice. See El-Tabech v. Hopkins, 997 F.2d 386, 390 (8th Cir. 1993). So too is the fact that a defendant projects flatness when describing something so weighty as a violent homicide. See Matylinsky v. Budge, 577 F.3d 1083, 1097 (9th Cir. 2009) (holding that petitioner failed to establish prejudice because, in part, he had a “flat” affect and would have 45 described the killing of his wife in a “matter-of-fact” and “disinterested” manner). Putting Lesko on the stand also came with other risks. Even if he could effectively rebut Rutherford’s and Montgomery’s statements by testifying that he did not know Travaglia intended to shoot Officer Miller, Lesko still would have faced hazardous cross-examination about the Nicholls killing. Lesko’s confession to police showed that his role in it was even greater—and more horrifying—than Rutherford knew. While Rutherford’s testimony and Lesko’s confession would congruently establish that Lesko beat, mocked, and tortured Nicholls and that Nicholls perished at the lake after Lesko and Travaglia dragged him away, cross-examining Lesko with the details of his confession would paint the final moments of Nicholls’ life in even more gruesome detail. Lesko admitted to police that he had knocked Nicholls unconscious but knew the man to still be alive when they dragged him down to the lake. Lesko confessed that he knew the plan was to kill Nicholls at the lake. Lesko admitted that he attached a 150- pound brick to Nicholls by tying it to his torso before throwing him head-first into the frozen waters. Lesko watched Nicholls resurface once and then disappear back into the water, never to return. A painstaking blow-by-blow of those details would likely dispense of any sympathy the jury might have felt for Lesko. Moreover, Lesko’s purposeful conduct in Nicholls’ torture and murder would also directly aid the prosecution in establishing Lesko’s premeditated intent when he helped steal Travaglia’s father’s gun, returned for lethal ammunition to replace the birdshot inside it, and then participated in the goading and killing of Officer Miller. Nicholls’ drawn-out murder—which occurred just hours before Miller was murdered—would only buttress the Commonwealth’s case that Lesko knew and intended exactly what would transpire. 46 Lesko’s testimony may have opened the door to other damning evidence as well. Though Lesko had been involved in the Levato and Newcomer killings, evidence of those crimes had been excluded as impermissible character evidence. But if Lesko took the stand, he might have said something which would have opened the door to this evidence coming in for the permissible purpose of impeachment. Commonwealth v. Nypaver, 69 A.3d 708, 716 (Pa. Super. Ct. 2013) (“A litigant opens the door to inadmissible evidence” if his testimony “creates a false impression refuted by the otherwise prohibited evidence.”). This was a potentially catastrophic risk which we properly consider in the prejudice analysis. See Smith v. Dickhaut, 836 F.3d 97, 107 (1st Cir. 2016); Medley v. Runnels, 506 F.3d 857, 861 (9th Cir. 2007). Finally, even assuming that the jury would have believed Lesko over both Rutherford and Montgomery and that Lesko would have somehow dodged the perils of cross- examination, there still would be substantial evidence to convict him of intentional, premeditated murder. The jury knew that Lesko and Travaglia had jointly engaged in a multi- day killing spree, had just stolen a gun which Lesko took pains to ensure was equipped with the bullets necessary for a fatal shooting, and had just intentionally killed Nicholls when they made multiple efforts to get Officer Miller to chase them. This is not a case where trial evidence was scant, see Nichols v. Butler, 953 F.2d 1550, 1554 (11th Cir. 1992), nor was it one where the defendant’s testimony would have been an affirmative denial of involvement in the crime, see Owens v. United States, 483 F.3d 48, 59-60 (1st Cir. 2007). “Strickland’s second prong requires a petitioner to show that the errors were ‘sufficiently serious as to deprive the defendant of a fair trial, a trial whose result is reliable.’” Palmer, 592 F.3d at 394 (quoting Strickland, 466 U.S. at 687). On balance, the jury’s 47 decision to convict was not rendered unreliable by any violation of Lesko’s right to testify. D. Lesko’s Claim That Counsel Was Ineffective At His Resentencing Lesko also contests his 1995 resentencing, claiming ineffective assistance of counsel. At resentencing, Lesko was once again represented by Attorney Marsh—who had previously represented Lesko in his initial guilt and sentencing proceedings—as well as Brian O’Leary. According to O’Leary, Marsh called him less than five days before jury selection for the resentencing and asked him to be his co- counsel. Despite never having handled a criminal matter or a trial, O’Leary agreed. He soon realized that their “backs were against the wall”: trial was just days away, and Lesko’s counsel still did not have an expert report from their mental health expert. J.A. 3036. O’Leary was in “panic” at the level of preparation undertaken. J.A. 3036. He believed that Marsh had failed to adequately prepare a mitigation case. O’Leary interviewed Lesko and learned that “none of the people or institutions that [he] had come to learn about in [his] interview with [Lesko] had been contacted” directly by Marsh. J.A. 3045. Marsh also failed to obtain Lesko’s records from Allegheny County Children and Youth Services, and appeared “burnt out,” with O’Leary describing the case as being like a “ball and chain” for Marsh who just wished to be “relieved of [its] burden.” J.A. 3125. O’Leary sought to right the ship. He contacted several of Lesko’s family members and other key characters, but the limited time before trial did not allow for the extent of interviewing and preparation O’Leary desired. He also subpoenaed Lesko’s CYS records, but they did not arrive until trial was already underway. 48 Marsh had made some efforts to present a mitigation case, however. In 1992, he hired Alfonso Associates to complete a psychosocial assessment of Lesko. As with other parts of pre-resentencing preparation, this was not without problems. According to the founder of Alfonso Associates, Marsh failed to communicate and failed to pay the firm’s bill, so in late 1992 or early 1993, they stopped working on Lesko’s case. Once work resumed in October 1994, Alfonso’s founder was “taken aback by the short time that [they] had,” and decided to assign case worker Lois Nardone, who had a master’s degree and some familiarity with the case, to handle it. J.A. 3770. Nardone testified that she “was uncomfortable with the amount of time” she had to work up the case, believing it “insufficient.” J.A. 3897. During the four months she had to prepare a social history report of Lesko, she also had little communication with Marsh. And, because Alfonso previously had trouble getting paid, the firm limited the amount of work Nardone could do. Nardone later admitted that, based on her constraints in time and funding, she “did an insufficient investigation.” J.A. 3958. Marsh also hired clinical and forensic psychologist Dr. Herbert Levit to examine Lesko. After performing multiple psychological tests, Dr. Levit diagnosed Lesko with borderline personality disorder: “a person who is emotionally explosive at times, has erratic behavior, engages in behavior which is contrary to society’s standards, is frequently immature, and has had difficulty in adjusting.” J.A. 2410-11. He further opined that, at the time of the crimes, Lesko was suffering from polysubstance dependence—he was using drugs and alcohol consistently and excessively, to the point that his functioning was impaired. According to Dr. Levit, Lesko, despite being 21 at the time of the murders, “was emotionally a teenager at best.” J.A. 2414. He suffered from diminished capacity—the 49 “ability to think clearly, rationally, logically and in a mature manner.” J.A. 2415. Dr. Levit concluded that Lesko was “sufficiently under the influence of alcohol, drugs[,] and the trauma of having discovered what happened to his [molested] brother, which triggered off some of the feelings as to what happened to himself, so that his ability to conform to [the law] was justly impaired.” J.A. 2415-16. Thus, at resentencing, Marsh and O’Leary presented a multi-faceted mitigation case that included: (1) a sympathetic narrative of Lesko’s life, as retold by investigator Nardone; (2) family member testimony buttressing that narrative; (3) expert testimony from a competent psychologist; and (4) Lesko’s own words. The jury found four aggravating factors and four mitigating factors, and ultimately voted for death. Lesko claims his counsel performed ineffectively by failing to prepare and present his mitigation case in three overlapping and self-reinforcing ways. First, he says Marsh and O’Leary failed to identify and advance an argument that he suffered organic brain damage. Second, he says they failed to interview or adequately prepare fact witnesses who knew him. Third, he says they failed to timely acquire and use his CYS records. Lesko also alleges that he suffered prejudice from the cumulative effect of these errors. The PCRA court granted relief, concluding that Lesko’s counsel failed to develop and present mitigating evidence, and that “there is a reasonable probability that the outcome of the proceedings would have been different if [his] counsel had adequately prepared.” J.A. 163. The Pennsylvania Supreme Court reversed. It concluded that counsel was not deficient in the mitigation case they did put on. Describing that mitigation case, the Court observed that Lesko’s lawyers presented persuasive testimony from Nardone, who detailed Lesko’s life 50 and trauma, and from Dr. Levit, who had been qualified as an expert witness in several hundred court proceedings. See Lesko XIII, 15 A.3d at 381–82. On balance, the Court found as to the presentation of a mitigation case that counsel “undertook a reasonable investigation and presented a compelling and partly successful case in mitigation,” as evidenced by the jury finding multiple mitigating factors. Id. at 381. As to organic brain damage, the Court held, any qualms with Marsh’s failure to discover it were properly directed at Dr. Levit, and counsel was not ineffective for relying on a qualified expert who missed a diagnosis. See id. at 382. The Court further concluded that even if counsel had performed deficiently, Lesko had not been prejudiced: Faced with the aggravating circumstances where the defendant has been found guilty of multiple murders occurring within a one-week period, including the cold-blooded murder of an on-duty police officer, and the case in mitigation already successfully presented, we simply cannot conclude that Strickland relief can be premised upon the additional mitigation evidence the PCRA court found would have carried the day. Id. at 385.14 14 The decision was not unanimous. Justice Saylor, concurring in the judgment, found that counsel performed deficiently but that Lesko had not been prejudiced. See id. at 417-18 (Saylor, J., concurring). Justice Todd dissented, arguing that Lesko was entitled to a new sentencing hearing because counsel performed ineffectively by failing to “retain a neuropsychologist for the purpose of evaluating Lesko and 51 In Lesko’s habeas proceedings, the District Court ruled that the Pennsylvania Supreme Court’s analysis was neither contrary to nor an unreasonable application of federal law. “At the very most, . . . Lesko only show[ed] that the Pennsylvania Supreme Court arguably reached an incorrect result (although this Court does not think so, particularly with respect to its holding that Lesko was not prejudiced by counsel’s alleged deficient performance).” Lesko XIV, 2015 WL 249502, at *38. Focusing primarily on prejudice, the District Court held that, given “the powerful aggravating circumstances presented by the prosecution, the mitigating evidence that his defense counsel did present, and the mitigating circumstances that the jury did find,” Lesko could not show that the Pennsylvania Supreme Court’s application of federal law was objectively unreasonable. Id. Recognizing that under AEDPA, our “‘review must be doubly deferential’ in order to afford ‘both the state court and the defense attorney the benefit of the doubt,’” Woods v. Donale, 575 U.S. 312, 316-17 (2015) (per curiam) (quoting Burt v. Titlow, 571 U.S. 12, 15 (2013)) (internal quotation marks omitted), we agree with the District Court. To prevail on an ineffective assistance of counsel claim, Lesko had to show that his lawyers performed deficiently and that he was prejudiced as a result. See Strickland, 466 U.S. at 687. Deficient performance is shown by proving that the representation “fell below an objective standard of reasonableness” under “prevailing professional norms.” Id. at 688. Prejudice requires a showing of a reasonable probability that, had counsel performed properly, at least one juror would presenting testimony on organic brain damage.” Id. at 426 (Todd, J., concurring in part and dissenting in part). 52 have found the mitigating factors to outweigh the aggravating factors. See Jermyn v. Horn, 266 F.3d 257, 309 (3d Cir. 2001). But we review here not merely with this high standard in mind, but also cognizant of the deference owed to the Pennsylvania Supreme Court’s determinations under AEDPA. See Harrington v. Richter, 562 U.S. 86, 101 (2011). So Lesko must “show that the state court’s ruling on [his ineffectiveness claim] was so lacking in justification that there was an error beyond any possibility for fairminded disagreement.” Burt, 571 U.S. at 19-20 (alterations omitted) (quoting Harrington, 562 U.S. at 103). This standard presents a “formidable barrier to federal habeas relief,” and here, at least as to prejudice if not also performance, that barrier has not been cleared. Id. 1. Deficient Performance In assessing deficient performance, we must keep in mind the claim at issue: alleged deficiency in failing to develop and present a mitigation case. Lawyers have “a duty to make reasonable investigations or to make a reasonable decision that makes particular investigations unnecessary.” Strickland, 466 U.S. at 691. In the death penalty context, prevailing professional standards call for counsel to make “efforts to discover all reasonably available mitigating evidence and evidence to rebut any aggravating evidence that may be introduced by the prosecutor.” Wiggins v. Smith, 539 U.S. 510, 524 (2003) (quoting ABA Guidelines for the Appointment and Performance of Counsel in Death Penalty Cases, Guideline 11.4.1(C) (1989) (emphasis in original)). The state supreme court held that Lesko’s team did not run afoul of these duties, as this was not “a case where counsel conducted minimal 53 investigation and failed to uncover evidence that was immediately available.” Lesko XIII, 15 A.3d at 381.15 Instead, Attorney Marsh put on a significant mitigation case and was successful in persuading the jury to find four mitigating factors. He presented testimony from Dr. Levit, who interviewed Lesko and his family members and conducted psychological testing of Lesko. Dr. Levit diagnosed Lesko with borderline personality disorder and stated his opinion that Lesko was suffering from polysubstance abuse at the time of the crimes, all of which led the jury to find as a mitigating factor that Lesko was under the influence of extreme emotional or mental disturbance pursuant to 42 Pa. C.S. § 9711(e)(2). Marsh also presented an account of Lesko’s childhood, including extensive and compelling testimony from Nardone as to the conditions under which Lesko lived and the abuse he suffered; this persuaded the jury to find that Lesko’s horrible childhood was another mitigating factor pursuant to 42 Pa. C.S. § 9711(e)(8). In addition, he provided evidence of Lesko’s change in character and service to others during his incarceration, which persuaded jurors to find two additional mitigating factors under 42 Pa. C.S. § 9711(e)(8). Lesko contends that his legal team was constitutionally required to do more. First, he argues that counsel should have 15 “[A] state court conclusion that counsel rendered effective assistance is not a finding of fact binding on the federal court”; rather, “it is a mixed question of law and fact.” Strickland, 466 U.S. at 698. While “state court findings of fact made in the course of deciding an ineffectiveness claim are subject to the deference requirement” of § 2254(e)(1), id., the Pennsylvania Supreme Court here did not make any such findings related to counsel’s actions before and at resentencing. 54 obtained helpful mitigation testimony from more people who knew him, specifically, eyewitnesses to his traumatic upbringing, including siblings who did not testify, a neighbor, CYS caseworkers, and a priest who supervised him when he was institutionalized. He also asserts his counsel should have performed more thorough examinations of those who did testify. The Pennsylvania Supreme Court determined that “defense counsel’s ‘decision not to seek more’ mitigating evidence from the defendant’s background ‘than was already in hand’ fell ‘well within the range of professionally reasonable judgments.’” Lesko XIII, 15 A.3d at 386 (quoting Bobby v. Van Hook, 558 U.S. 4, 11–12 (2009) (per curiam)). This determination was not unreasonable. As noted above, Attorney Marsh had retained investigator Nardone, who interviewed many key figures in Lesko’s life: his mother, grandmother, sisters, aunt, two teachers, a social worker, a prison chaplain, and Lesko himself. Marsh testified that he relied on Nardone to investigate Lesko’s background, which was appropriate. See Rhode v. Hall, 582 F.3d 1273, 1283 (11th Cir. 2009) (“Since [defendant’s] counsel hired investigators who interviewed potential witnesses and shared all of their information with counsel, we cannot say that counsel performed deficiently by delegating the mitigation investigation to them.”). Contrary to Lesko’s contentions, it appears counsel made a reasoned judgment in selecting certain individuals from Lesko’s life to testify, while allowing Nardone to be the conduit through which the broader arc of his life was told. Informed, “strategic choices” like this “are virtually unchallengeable,” Strickland, 466 U.S. at 690, and an attorney’s decision not to call a witness “is precisely the sort of strategic trial decision that Strickland protects from second- guessing.” Sanders v. Trickey, 875 F.2d 205, 212 (8th Cir. 55 1989). Even Attorney O’Leary, who has been highly critical of Marsh’s performance, reinforced that Marsh’s witness plan was logical. See J.A. 3061 (“I remember thinking that if [Nardone] testified well, and she was persuasive, and the jury was with her, that that would be adequate.”). The decision not to present additional witnesses is often rooted in strategic judgment. “[T]here comes a point at which evidence from more distant relatives can reasonably be expected to be only cumulative, and the search for it distractive from more important duties.” Bobby, 558 U.S. at 11. There would also be risks in having Lesko’s troubled upbringing told through family members who themselves might have mental and credibility concerns, rather than through a professional, who might appear more objective, organized, and composed. Family members and inexperienced lay witnesses could testify poorly or lose credibility upon cross-examination. Indeed, Lesko’s own testimony exemplified these risks. See J.A. 3127 (Attorney O’Leary describing Lesko, as a witness, having “a flatness to his personality I had never encountered in a human being before.”). The jury might also wonder why other members of the family—also having experienced trauma— were able to conform their behavior to the law while Lesko was not. While there are reasons to second-guess counsel’s effectiveness, his witness strategy was not one of them. Thus, the Pennsylvania Supreme Court’s ruling that counsel performed adequately in investigating and presenting mitigation witnesses was not unreasonable. Next, Lesko, like our dissenting colleague, points out that Lesko’s team at resentencing—which included experienced forensic psychologist Dr. Levit—failed to discover and present mitigating evidence showing that Lesko suffered from physiological brain damage. His likely brain damage was subsequently identified when Lesko’s new 56 counsel before the PCRA court retained Dr. Barry Crown, a practitioner of clinical and forensic psychology and neuropsychology. Dr. Crown opined that information in Dr. Levit’s report should have prompted Lesko’s team to conduct neuropsychological testing to evaluate whether he had brain damage. For instance, Lesko’s scores fluctuated widely in the tests that Dr. Levit did perform, a “very strong” indicator of an organic impairment, according to Dr. Crown. J.A. 4039. Dr. Crown also stated that other test results—including those showing that Lesko’s executive functions were inhibited, that his performance was on a primitive level, and that he was diagnosed with borderline personality disorder—were “red flags for brain damage.” J.A. 4044. Dr. Crown himself concluded to a reasonable degree of neuropsychological certainty that Lesko had brain damage—a materially different diagnosis than Dr. Levit’s, and at the PCRA hearing, counsel for the Commonwealth all but conceded that Dr. Crown was correct. J.A. 2981 (“I think he’s established that there is evidence of organic brain damage and it should be tested. I think we’re beyond that now.”). This evidence is significant, Lesko maintains, because it could have persuaded jurors to find as yet another mitigating factor that Lesko’s capacity to appreciate the criminality of his conduct or to conform his conduct to the requirements of the law was substantially impaired pursuant to 42 Pa. C.S. § 9711(e)(3). The problem for Lesko’s argument, however, is that Marsh was not ineffective in relying on a well-qualified psychological expert in Dr. Levit, who simply missed the diagnosis, and that was the basis the Pennsylvania Supreme Court to conclude that his failure to discover and advance a brain damage theory at resentencing did not violate Strickland. We cannot say that conclusion is “so lacking in justification that there was an error beyond any possibility for fairminded 57 disagreement.” Burt, 571 U.S. at 20 (alterations omitted) (quoting Harrington, 562 U.S. at 103). As our sister Circuits have persuasively stated, “An expert’s failure to diagnose a mental condition does not constitute ineffective assistance of counsel,” because a defendant “has no constitutional guarantee of effective assistance of experts.” Earp v. Cullen, 623 F.3d 1065, 1077 (9th Cir. 2010) (expert’s failure to conclude that defendant had organic brain damage is not ineffective assistance of counsel). There is no question that lawyers can and often do reasonably rely on experts, especially in navigating a field in which they have no training. Nor can lawyers be expected to catch every possible gap in an expert’s analysis in a specialized discipline. See Clark v. Mitchell, 425 F.3d 270, 285 (6th Cir. 2005) (concluding that “[i]t was not unreasonable for [the petitioner’s] counsel, untrained in the field of mental health, to rely on the opinions of [the psychological] professionals” counsel retained). Where, as here, an expert’s report contains no deficiency evident to an attorney untrained in the field, “forcing lawyers to second- guess their experts . . . would effectively eliminate the legitimate role experts play in guiding and narrowing an attorney’s investigation.” Hendricks v. Calderon, 70 F.3d 1032, 1039 (9th Cir. 1995). It also “would raise the Sixth Amendment hurdle well above the floor of minimal competence, requiring attorneys to have the specialized knowledge to evaluate an expert’s conclusions before relying upon them in making strategic choices.” Id.16 16 The dissent would hold that Attorney Marsh’s performance was constitutionally deficient for its failure to develop this mitigating evidence. In brief, the dissent’s argument is that the materials Attorney Marsh received from 58 Lesko contends that Marsh had reason to know that brain damage was a possibility and thus, his reliance on Dr. Levit was unreasonable. At the outset, we note that there was no obvious signal which would indicate to counsel that Lesko needed a brain evaluation beyond what their unquestionably qualified psychological expert provided. In fact, before Dr. Levit was even retained, Lesko had been examined by two mental health experts, and neither indicated any organic brain damage. Lesko retorts that Marsh received an article from Professor Welsh White in 1992, which discussed the linkage between child abuse and brain damage. Professor White, who had previously assisted Marsh, also sent him a sample motion to request neuropsychological testing. According to Lesko, this information should have triggered counsel to dig further. However, counsel did retain Dr. Levit, and reasonably relied on him to address these matters. As Marsh explained: “I had to rely on psychologists for that, to diagnose brain damage, or to indicate to me that there was a possibility of brain damage.” J.A. 2883. Nothing in the materials Professor White provided suggests that a psychologist like Dr. Levit cannot perform neuropsychological testing. To the contrary, in the sample affidavit therein, it was a clinical psychologist who concluded that the defendant may have brain damage. Thus, Marsh’s dependency on Dr. Levit to competently evaluate and diagnose Professor White, which linked child abuse to brain damage, should have prompted Marsh to second guess Dr. Levit’s conclusions and to review Lesko’s CYS records. The dissent would also hold that the Pennsylvania Supreme Court’s application of Strickland to conclude his reliance on Dr. Levit was not deficient was itself incorrect and unreasonable. 59 Lesko, while perhaps not peak advocacy, does not amount to deficient performance, given the “strong presumption that counsel’s conduct falls within the wide range of reasonable professional assistance.” Strickland, 466 U.S. at 689; see Campbell v. Coyle, 260 F.3d 531, 555 (6th Cir. 2001) (“[I]t was objectively reasonable for [petitioner’s] trial counsel to rely upon [a psychological expert’s] diagnosis and, further, trial counsel’s failure to independently diagnose PTSD [where the expert did not] was not unreasonable.”). Nor can we question counsel’s decision to hire Dr. Levit over someone like Dr. Crown. “The selection of an expert witness is a paradigmatic example of the type of strategic choice that, when made after thorough investigation of the law and facts, is virtually unchallengeable.” Hinton v. Alabama, 571 U.S. 263, 275 (2014) (per curiam) (alterations, quotation marks omitted) (declining to “launch federal courts into examination of the relative qualifications of experts hired and experts that might have been hired”). Lesko attempts to shift blame for the missed diagnosis onto his lawyer’s tardiness. He contends Marsh hampered Dr. Levit’s evaluation by retaining him only days before sentencing, which resulted in the report not being ready until after resentencing proceedings had begun (though not before the mitigation case was to be presented) and which would have prevented the completion of additional neuropsychological testing in time for resentencing had Dr. Levit ordered it. Cf. Escamilla v. Stephens, 749 F.3d 380, 392 (5th Cir. 2014) (concluding that counsel may have performed deficiently in part because they “failed to obtain a psychological evaluation for their client until after trial began”); Collins v. Sec’y of Pa. Dep’t of Corr., 742 F.3d 528, 550 (3d Cir. 2014). While this may very well be true, the problem is that Dr. Levit never opined that Lesko had brain damage, never recommended 60 further testing, and never represented that his analysis was rendered unreliable by a lack of time. Thus, even if counsel’s tardiness itself fell below prevailing professional standards,17 it did not cause Dr. Levit to render the expert opinion he did, nor did it render Marsh’s reliance on that opinion unreasonable.18 For the reasons we have explained, assessed against the “doubly deferential” standard under which we review an ineffective assistance of counsel claim, Titlow, 571 U.S. at 15, we cannot say that Marsh’s failure to procure a neuropsychologist notwithstanding the findings of his well- qualified expert was constitutionally deficient. Lesko’s final argument gives us pause, however; the record plainly shows that counsel, without excuse, failed to timely acquire and use Lesko’s CYS records to advance his mitigation defense. It appears that the need to obtain these records never occurred to Attorney Marsh, or if it did, that he did not prudently act on it. Although Attorney O’Leary 17 To be clear, we do not hold that the late-stage retention of Dr. Levit was necessarily deficient. As the Commonwealth has pointed out, some defense lawyers wait to hire an expert (assuming procedural rules so permit) for strategic reasons, to limit the time available to the Commonwealth to hire its own expert. Commonwealth’s Br. at 65. Reasonable strategic decisions are not to be second- guessed under Strickland, even where those decisions may prove to be clearly faulty in hindsight. Abdul-Salaam, 895 F.3d at 266. 18 To the extent Lesko argues that this delay precluded Dr. Levit from conducting a thorough review of Lesko’s CYS records, we address that argument below. 61 ultimately realized the importance of this evidence, he had joined the team just days before the resentencing and was unable to acquire the records before the proceedings began. This left counsel with insufficient time to integrate the records into their mitigation case and prevented Dr. Levit from having the opportunity to review those records in a timely fashion. Lesko relies particularly on Dr. Levit’s testimony at the 1999 PCRA hearing, years after his report. After being confronted with Dr. Crown’s findings, Dr. Levit testified that if he had reviewed Lesko’s CYS records before preparing his report he would have diagnosed Lesko with post-traumatic stress disorder, failure to thrive syndrome, and substantial impairment of his ability to appreciate the criminality of his conduct and to conform his conduct to the requirements of the law. Lesko’s lawyers owed it to him to obtain his CYS records. “Supreme Court precedent makes clear . . . [that] defense counsel has a duty to obtain administrative records . . . as part of the ‘obligation to conduct a thorough investigation of the defendant’s background.’” Blystone v. Horn, 664 F.3d 397, 422 (3d Cir. 2011) (quoting Williams v. Taylor, 529 U.S. 362, 396 (2000)). “[J]uvenile records” are particularly important, because they provide “the kind of rudimentary background information that there can be no strategic reason not to investigate.” Abdul-Salaam v. Sec’y of Pa. Dep’t of Corr., 895 F.3d 254, 268 (3d Cir. 2018) (internal quotation marks omitted). The Supreme Court has agreed that counsel who “failed to conduct an investigation that would have uncovered extensive records graphically describing [a habeas petitioner’s] nightmarish childhood” performed deficiently. Williams, 529 U.S. at 395. Here, Lesko’s lawyers had no strategic reason nor “reasoned judgment” for failing to timely subpoena his CYS records. Blystone, 664 F.3d at 423. And 62 once they did arrive, Marsh confessed that he probably spent about only “an hour or so” reviewing the records, and that he did not give them to Nardone or Dr. Levit. J.A. 2832; 2834-35, 3917. O’Leary was “not sure [he] even got to go through them.” J.A. 3048. In short, counsel’s inexplicable failure to obtain the CYS records in a timely fashion (and resulting inability to use them or to allow Dr. Levit to review them) raises doubts as to his performance. But, as explained below, it did not cast doubt on the significant and partially successful mitigation case that counsel did put on, and we need not second guess the Pennsylvania Supreme Court’s determination as to deficient performance where, as here, Lesko has not shown that he suffered prejudice as a result.19 19 The dissent would hold that Marsh’s performance was ineffective “[b]ecause of [his] delay in obtaining and reviewing records and in consulting Dr. Levit and because of Marsh’s failure to follow up on the mitigating issue of brain damage.” But this fails to give deference to the Pennsylvania Supreme Court’s decision as required under AEDPA. See § 2254(d)(1). Indeed, the dissent’s analysis reads like de novo review of Lesko’s ineffective assistance of counsel claim. This is improper for a federal habeas court. To obtain habeas relief, “a state prisoner must show that the state court's ruling on the claim being presented in federal court was so lacking in justification that there was an error well understood and comprehended in existing law beyond any possibility for fairminded disagreement.” Harrington, 562 U.S. at 103. The Pennsylvania Supreme Court’s ruling undoubtedly meets this standard. Moreover, the dissent elides the two inquiries for an ineffective assistance of counsel claim. It is not enough for 63 2. Prejudice In assessing whether Lesko established prejudice from counsel’s ineffectiveness, the Court considers “whether all of counsel’s unprofessional errors combined undermine our confidence in the result.” Frey v. Fulcomer, 974 F.2d 348, 361 n.12 (3d Cir. 1992). The Pennsylvania Supreme Court concluded that Lesko had not so shown, given the severity of the aggravating circumstances when weighed against the mitigating factors. As the District Court concluded, this represents a reasonable application of Strickland. As the state supreme court emphasized, the aggravating factors reflected an exceptionally brutal set of facts. The jury heard that Lesko had been involved in three other killings before the one at issue, two of which served as aggravating factors. Unlike during the guilt phase of the trial, the prosecution was permitted to read to the jury Lesko’s confessions to killing Nicholls and Newcomer. His statements were chilling and devoid of remorse. He explained how Newcomer, a “fat, gold or blond hair[ed]” woman, kindly Lesko to show that Marsh’s performance was deficient. He must also show that Marsh’s performance prejudiced his defense. Wiggins, 539 U.S. at 534. This means that a “defendant must show that 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. As discussed in detail below, see discussion infra Part II.D.2, Lesko has not shown prejudice, especially in light of the substantial aggravating evidence weighing against him. See Wiggins, 539 U.S. at 534 (“In assessing prejudice, we reweigh the evidence in aggravation against the totality of available mitigating evidence.”). 64 stopped her car to help him get out of a fog on New Year’s Day. J.A. 1849. Lesko described how he handcuffed the woman; how he shot his gun at her once, realizing he missed when Newcomer started moaning in what Lesko said was an attempt to fake a heart attack; and then proceeded to shoot again, making sure no pulse remained after his second attempt. Lesko’s description of the Nicholls killing was even worse. When asked by the police how Nicholls responded to being kidnapped and shot by Travaglia, Lesko said Nicholls was acting “[l]ike a faggot.” J.A. 1909. He also spoke flippantly about his role in the killing. Lesko explained that he had punched Nicholls and knocked him unconscious, and then, when asked how he knew Nicholls had still been alive at that time, responded, “Because I never killed anybody yet with a punch.” J.A. 1905. Lesko confessed that he “knew [he was] going to kill [Nicholls] along with” the help of Travaglia and Rutherford at the lake. J.A. 1903. While Rutherford’s testimony for the prosecution could offer only a limited account of how Nicholls died, Lesko did not spare any of the gruesome details in his confession. He said that he and Travaglia attached a 150-pound brick to Nicholls, “put his head through the ice . . . and just threw him down.” J.A. 1906-07. Nicholls bobbed up once, went back down, and then never reappeared. The prosecution was also able to present evidence that Lesko’s claims of newfound remorse may have been disingenuous. On cross-examination, Lesko admitted that, back at the time of his first trial, he had told a newspaper reporter that he “had a better chance of getting hit by a car than getting the electric chair,” and that he “would be out of jail in ten years.” J.A. 2573. If nothing else, Lesko’s own words undermined any effort to engender sympathy for him. The aggravation case was damning. 65 Meanwhile, as the Pennsylvania Supreme Court explained, Marsh, despite his shortcomings, did present a meaningful mitigation case. He presented Dr. Levit’s testimony pointedly explaining how Lesko suffered from diminished capacity, and the effects that could have. He presented evidence of Lesko’s horrific childhood, including testimony from Ms. Nardone and two family members detailing how Lesko had been molested and set on fire. He presented evidence that Lesko’s judgment had been impaired by alcohol and drugs and that, when he committed the crimes, he was deeply upset by news that his younger brother had also been molested. Lesko insists this was not enough, and that a much stronger case could have been made absent counsel’s deficiencies. He argues that the CYS records likely would have painted an even more vivid picture of the childhood neglect and trauma Lesko suffered and the failure of CYS to intervene or otherwise protect Lesko and his siblings, to the extent jurors did not already recognize that failure. He also contends, in terms of cumulative prejudice, that if counsel had promptly obtained the CYS records and shared them with Dr. Levit, Dr. Levit might have ordered further testing to evaluate the possibility of brain damage, strengthening Lesko’s argument that he was “under the influence of extreme mental or emotional disturbance,” 42 Pa. Cons. Stat. § 9711(e)(2) (listing such disturbance as a mitigating circumstance), and likely would have allowed counsel to claim that Lesko’s capacity “to appreciate the criminality of his conduct or to conform his conduct to the requirements of law was substantially impaired,” id. § 9711(e)(3). In addition, he argues, with a timely brain damage diagnosis, counsel might have foregone Lesko’s own testimony, in which, as his lawyers subsequently said, he made a poor witness. 66 The Pennsylvania Supreme Court assessed this new mitigation evidence alongside the extensive aggravating evidence. It concluded that the strength of the latter was fatal: “[T]here is not a reasonable probability that a life sentence would have been returned if only the mitigation evidence presented at trial had been supplemented by the mitigation evidence presented at the PCRA hearing, particularly given the strength of the aggravating circumstances detailed above.” Lesko XIII, 15 A.3d at 386. Given the uniquely compelling aggravating circumstances to Lesko’s crimes, and the mixed bag of additional mitigating evidence—some persuasive but much of it cumulative—it was not unreasonable for the state court to conclude that no reasonable juror would have voted for life. “Strickland places the burden on the defendant, not the State, to show a ‘reasonable probability’ that the result would have been different,” a difficult burden to carry when “powerful” aggravating evidence has been presented. Wong v. Belmontes, 558 U.S. 15, 27-28 (2009) (per curiam) (holding that additional mitigation evidence would not sway a juror to vote for life, where the defendant brutally bludgeoned one victim with a dumbbell bar 15 to 20 times and had previously killed another execution style). As the Wong Court reasoned, “[i]t is hard to imagine expert testimony and additional facts about [the petitioner’s] difficult childhood outweighing the facts of [a brutal] murder.” Id. at 27-28. Envisioning a swayed juror “becomes even harder” “when the evidence that [the petitioner] had committed another [heinous] murder . . . is added to the mix.” Id. at 28. Luring, taunting, and killing a police officer who is actively serving the community is, in itself, a heinous crime that cries out for justice, and indeed, the victim’s status as an on-duty officer is an aggravating circumstance. And, here, we “add into the mix” three additional murders—each chilling and horrific in its own 67 way—in a senseless killing spree that epitomizes a total disregard for the value of life. As in other cases in which capital sentences have survived both direct and habeas challenges, the aggravating case here was truly “devastating.” Woodford v. Visciotti, 537 U.S. 19, 26-27 (2002) (holding that a state court decision rejecting a Strickland claim was not unreasonable, where the aggravating circumstances of an execution-style murder coupled with another attempted execution style-murder and prior stabbing offenses were “so severe” and “overwhelming”). This is especially true given the role of the federal court in reviewing a state court’s decision under AEDPA. As the Supreme Court has reminded, “‘it is not enough to convince a federal habeas court that, in its independent judgment, the state-court decision applied Strickland incorrectly.” Id. at 27 (quoting Bell v. Cone, 535 U.S. 685, 699 (2002)). “The federal habeas scheme leaves primary responsibility with the state courts for these judgments, and authorizes federal-court intervention only when a state-court decision is objectively unreasonable.” Id. And here, “[w]hether or not we would reach the same conclusion as [the Pennsylvania Supreme Court]” if we were reviewing de novo, “‘we think at the very least that the state court’s contrary assessment was not ‘unreasonable.’” Id. (quoting Bell, 535 U.S. at 701). We simply cannot say, on the record before us, that the state court’s no-prejudice determination amounts to an “‘extreme malfunction[] in the state criminal justice system,’” and proving only “‘ordinary error’ or even . . . ‘a strong case for relief’” is not enough for us to grant Lesko’s petition. Mays v. Hines, 141 S. Ct. 1145, 1149 (2021) (per curiam) (quoting Harrington, 562 U.S. at 102). 68 III. For these reasons, we will affirm the District Court’s denial of habeas relief. 69 FISHER, Circuit Judge, concurring. I write separately to highlight the tension between the conclusion we believe is required by the Supreme Court’s textual analysis in Magwood and Congress’s likely intent in drafting AEDPA. It is “our goal when interpreting a statute . . . to effectuate Congress’s intent.” Hagans v. Comm’r of Soc. Sec., 694 F.3d 287, 295 (3d Cir. 2012). And while the statutory text and precedent compel our holding that Lesko’s guilt-phase claims are not second or successive, that reading of § 2244(b) in my view runs counter to “AEDPA’s goal of promoting ‘comity, finality, and federalism’ by giving state courts ‘the first opportunity to review [the] claim, and to correct any constitutional violation in the first instance.’” Jimenez v. Quarterman, 555 U.S. 113, 121 (2009) (quoting Carey v. Saffold, 536 U.S. 214, 220 (2002)). In short, our interpretation of § 2244(b) allows a petitioner who already had a full and fair opportunity to attack his underlying conviction a second bite at the apple because a discrete sentencing claim was meritorious. This result will obtain even where the prisoner failed to initially raise the issue, and even where evidence and witnesses are long gone. Piecemeal litigation may increase at the cost of judicial economy and efficiency. Federalism concerns also come into play, including the need to respect the sovereignty of state courts in adjudicating cases which implicate federal constitutional rights. See Coleman v. Thompson, 501 U.S. 722, 730–31 (1991) (stressing the importance of respecting State interests). Construing AEDPA’s second or successive rule as we have may revive assumptions that state remedies are inadequate and federal review is superior—a belief Congress sought to lay to rest in passing AEDPA. My reading of congressional intent is reinforced by comparing § 2244(b) with its predecessor, the judge-made 1 abuse of the writ doctrine. As the Seventh Circuit explained, that doctrine would likely have prevented a prisoner from contesting an undisturbed conviction after resentencing. See Suggs v. United States, 705 F.3d 279, 285 (7th Cir. 2013); see also Magwood v. Patterson, 561 U.S. 320, 354 (2010) (Kennedy, J., dissenting). While AEDPA did not directly codify the doctrine, Congress borrowed principles from it when creating the second or successive rule, and the pre- AEDPA doctrine remains relevant to interpreting the statute. See Benchoff v. Colleran, 404 F.3d 812, 813 (3d Cir. 2005) (“[W]e will consult the abuse of the writ jurisprudence, which predated the passage of § 2244, concluding that the doctrine retains vitality as a tool for interpreting the term ‘second or successive’ under § 2244.”). Accordingly, I note that, compared to pre-AEDPA law, AEDPA “placed more, rather than fewer, restrictions on the power of federal courts to grant writs of habeas corpus to state prisoners.” Miller-El v. Cockrell, 537 U.S. 322, 337 (2003). Indeed, “the new substantive standards governing the allowance of second or successive applications are more rigorous than the pre-AEDPA standard developed by the courts.” In re Minarik, 166 F.3d 591, 595 (3d Cir. 1999). It is thus peculiar for AEDPA to be more accepting of successive petitions than the abuse of the writ doctrine. While it is true that Magwood warns against putting too much weight on this comparative reasoning, 561 U.S. at 333, it nonetheless reinforces our view that if Congress intended to restrict rather than expand successive petitions, it should amend AEDPA to leave no doubt. 2 ROTH, Circuit Judge, dissenting in part. I join the Majority in all Sections except Section II.D on Ineffective Assistance of Counsel, including 1. Deficient Performance and 2. Prejudice. As to Section II.D, I respectfully dissent. John Lesko brought this habeas petition, challenging his conviction and sentence for first degree murder. He contends, among other things, that his attorney was ineffective for his failure to present available mitigating evidence of Lesko’s organic brain damage which significantly impaired his ability to appreciate the criminality of his conduct. Lesko committed vicious acts. Nevertheless, under the United States Constitution, when he was put on trial for those acts, he was entitled to effective counsel1 and to the assistance of a competent mental health practitioner who could conduct an appropriate examination.2 Lesko’s attorney, however, committed glaring errors. I cannot overlook the prejudicial disservice that this attorney rendered over nearly 20 years from his initial appointment in 1980 through Lesko’s 1995 resentencing and his direct appeal. 1 See Strickland v. Washington, 466 U.S. 668, 688 (1984). 2 See Ake v. Oklahoma, 479 U.S. 68, 80 (1985). 1 I. INEFFECTIVE ASSISTANCE OF COUNSEL Because a death sentence is based on the “uniquely individual human being[],”3 my analysis of Lesko’s sentence begins by examining his life. I am basing my account on the facts that are available in Lesko’s Children and Youth Services (CYS) records and that were known by Lesko’s CYS case workers and his immediate family at the time of trial and of his sentencings. In doing so, I am presenting facts that demonstrate that, at the time of Officer Miller’s murder, Lesko was suffering from organic brain damage that significantly impaired his ability to appreciate the criminality of his conduct or to conform his conduct to the requirements of law.4 Attorney Marsh, however, did not have neuropsychological testing of Lesko done. Despite the fact that he knew of the existence of the CYS records in 1980, he did not subpoena or review them until he did a cursory one hour review as the 1995 resentencing hearing was about to begin. Nor did he have them examined to determine the need for psychological evaluation. Moreover, although he had a sample motion to obtain neuropsychological testing and had been informed of the importance of it, he never had that testing done. As the Majority has described, John Lesko grew up in a miserable and chaotic environment. For his first nine years, he 3 Woodson v. North Carolina, 428 U.S. 280, 304 (1976); see also Wayne Batchis, The Right’s First Amendment: The Politics of Free Speech & the Return of Conservative Libertarianism 51 (2016) (“[T]he judicial enterprise is a human enterprise, not a mechanistic one.” (emphasis in original)). 4 Appx. 141. 2 lived with his mother Mary Anne Fedorko and his four younger siblings, Michael, Matilda, Kimberly, and Joseph. He does not know who his father is. Because Fedorko neither paid her bills nor maintained her home, the family was frequently evicted. They had moved about fifteen times before Lesko was nine years old. The utilities were turned off most of the time. Despite having no heat in the winter, Lesko and his siblings often wore only t- shirts and simple pants. In the summer, “[t]he heat was oppressive and the odor almost unbearable.”5 The home was filthy. Sometimes there was no running water. CYS records note in 1965 that the family’s living conditions were “uninhabitable.”6 The windows, doors, and floors were falling apart. There was garbage, bugs, urine, and feces from animals and people everywhere; “one’s feet actually stuck to the floor in the muck.”7 Some rooms were so filled with debris that they were unusable. “[T]here [were] piles of dirty diapers in the middle of the living room floor.”8 When Fedorko gave birth to Joseph in 1966, she did not want to bring him home from the hospital because she feared “the rats may harm” him.9 Lesko at that time played with dead rats. The CYS case worker also noted that Lesko had an unusual gait; she was sure that something was wrong with him. The filthy home was made worse by Fedorko’s neglect as a parent. Fedorko would leave the children alone at home 5 Appx. 4600. 6 Appx. 4594. 7 Appx. 4619–20. 8 Appx. 4648. 9 Appx. 4607. 3 for days at a time while she went out drinking.10 The children skipped school, roamed the streets at night, and stole for food. When the children did go to school, they were dirty, sleep- deprived, and “smell[ed] of urine.”11 Fedorko sometimes brought Lesko and his brother Michael to bars with her and taught them to steal from drunk customers. Lesko began drinking alcohol by age eight. The children’s schools often complained about their health and hygiene, including lice. Fedorko insisted that the schools were just “pick[ing] on” them.12 CYS received complaints about Fedorko’s failure to care for the children when they were sick, injured, or malnourished. The children often had diarrhea because of the filth. Lesko had frequent ear infections and high fevers. CYS workers once had to tell Fedorko to take Lesko to the hospital to treat a severe ear infection. When Joseph was just a few months old, he developed “enormous sores on [his] penis” because Fedorko neglected to seek follow-up care for his circumcision and “left him lay in a filthy crib all day.”13 When Fedorko was home, she was “completely overwhelmed”14 and abused the children. Along with their filth and bug bites, the children had bruises from Fedorko beating them. In a fit of anger, Fedorko once threw Kimberly (then just a baby) across the room to her aunt Joann. Sexual abuse was documented as well. A forensic social worker 10 Appx. 2295:10–15. 11 Appx. 4616. 12 Appx. 4623. 13 Appx. 4648. 14 Appx. 4620. 4 testified at Lesko’s resentencing that Fedorko often brought home strange men from her drinking binges and had sex with them on the couch in front of the children. Michael testified that she made him and Lesko put their hands and a soda bottle into her vagina. Lesko once witnessed Michael and Matilda having sex. The abuse did not stop at home. When he was four, a teenager set Lesko on fire, hospitalizing him for a month. When he was six or seven, a customer sexually molested Lesko while he and Michael were shining shoes for money at a bar. Despite many visits from CYS, no one stepped in to help. A caseworker in March 1967 described Lesko as “emotionally flat.” She explained, “It is [Lesko’s] seeming withdrawal and other signs of disturbance that concern us.”15 Still, CYS did nothing. Former CYS employees later admitted that the Commonwealth had failed Lesko and his siblings. When Lesko was nine, Fedorko surrendered the children to shelters. Lesko’s upbringing caused early psychological problems that worsened throughout his life. He “suffered severe trauma” and had “persistent nightmares” and nervousness from being set on fire.16 He also developed separation anxiety. While in the hospital at age four, Lesko was so distressed about being away from his mother that he tried to run away; he was then confined in a restraining cage. 15 Appx. 152. 16 Appx. 4573–74. 5 At age five, Lesko showed behavioral issues and struggled to follow social norms. During a 1965 CYS visit, Lesko “r[an] wildly throughout the house and jump[ed] in and out of the window onto the roof.”17 Neighbors complained that he and his siblings often destroyed their property. At school, Lesko’s teacher said he had “the most disgusting habits she had ever seen in a boy.”18 He “was masturbating openly and very frequently,” “smashes and dirties everything he touches or handles,” “picks his nose and ears,” and ate “as if he has never had any training at all.”19 Lesko’s classmates ostracized him.20 Lesko also developed an unhealthy bond with his brother Michael. Lesko looked up to Michael as a father figure. Despite being younger, Michael was physically larger and could marginally better interact with other children at school Lesko became extremely susceptible to Michael’s influence. As Michael himself explained, Michael was the “leader” and “could manipulate [Lesko] into [doing] whatever [Michael] wanted to do.”21 Both Lesko’s abandonment issues and his attachment to Michael increased over time. These abandonment and attachment issues became more acute when Lesko was sent to the shelters. Lesko and his siblings were separated and moved around. Lesko struggled to adjust without Michael and his mother. Then, when Lesko was fourteen, he and Michael moved in with their grandmother, Anna Ridge, and her alcoholic daughter, Joann. Joann verbally 17 Appx. 4598. 18 Appx. 4611. 19 Appx. 4611. 20 Appx. 4611. 21 Appx. 2465:4–14. 6 abused Lesko, calling him illiterate and an animal and his mother a whore.22 When Fedorko sometimes visited him, Lesko hoped that she would take him home to live with her again, but she never did. While in the shelters, he blamed the Commonwealth for keeping him away from his mother, but when he lived with Ridge he did not know who to blame. Lesko’s anger would keep him awake at night. Under Michael’s guidance, Lesko used drugs and dropped out of school in eleventh grade. At Michael’s instigation, Lesko began stealing and shoplifting and got in trouble with the law. His probation officer urged Lesko to join the Marines. He did so and found some success. After boot camp ended, however, the structure was relaxed, and he went AWOL. He was discharged from the Marines and moved back to Ridge’s home.23 In the next two years, Lesko suffered further crises. In January 1979, Michael was convicted of burglary and sentenced to eighteen months in prison. In October 1979, Lesko sought treatment for his rage and abandonment issues, but his anger worsened when he learned that his brother Joseph’s social worker had been molesting Joseph for years. Then, Ridge kicked Lesko out of her home for arguing with Joann and her. In October 1979, Lesko—without a home, with his brother Michael in prison, and with his mental health deteriorating—met Michael Travaglia. Like Lesko’s brother Michael, Travaglia exercised influence over Lesko. In Lesko’s 22 Appx. 2303:1–6. 23 Appx. 2305:7–2306:8. 7 words, he and Travaglia became “like brothers.”24 Travaglia filled a void. For two months, Lesko and Travaglia spent almost every day together, often sharing hotel rooms, using drugs (usually supplied by Travaglia), and committing robberies. During one of these drug binges in December 1979, the robberies escalated into murders. From the start, Lesko consistently said that Travaglia had orchestrated their crime sprees, just as Lesko’s brother Michael had instigated their childhood stealing and shoplifting. The above account of Lesko’s childhood and adolescence appears, for the most part, in his CYS records. Those records are of utmost importance. In particular, the records contain information that is significant to a neuropsychological investigation into whether Lesko had suffered organic brain damage and, if so, the impact of that damage on his behavior. There are notations of insomnia; hyperactivity; headaches; blackouts; episodic dyscontrol (a form of organically impaired impulse control); inadequate nutrition; his mother’s history of heavy drinking, indicating the possibility of Fetal Alcohol Syndrome; early ingestion of alcohol and other toxins; scars all over his body; and his unusual gait. A review of these references would have indicated to an effective attorney that psychological review was necessary and, as a result of that review, that a neuropsychological assessment for organic brain damage was needed.25 At this point then, we need to consider what defense an effective lawyer should have provided to Lesko when he faced 24 Appx. 1848:25. 25 Appx. 139. 8 the charge of accomplice to first-degree murder in the death of Officer Miller. Courts have repeatedly stressed the need for skilled and experienced lawyers in capital cases.26 “[D]eath penalty cases [are] so specialized that defense counsel have duties and functions definably different from those of counsel in ordinary criminal cases.”27 Rabe F. Marsh, III, represented Lesko for nearly twenty years. Throughout this period, he failed to perform the necessary duties and functions that he owed to Lesko. The most glaring inadequacy was the failure to obtain and review the CYS records in a timely manner and to cull from those records the information that demonstrated the need for psychiatric and neuropsychological examination and testing.28 In 1989, prior to Lesko’s second sentencing in 1995, the American Bar Association had established Guidelines for Death Penalty Cases which required counsel to make “efforts to discover all reasonably available mitigating evidence and evidence to rebut any aggravating evidence that may be 26 See, e.g., Phillips v. White, 851 F.3d 567, 578 (6th Cir. 2017) (finding counsel’s performance deficient where counsel “admitted in court that he had no experience with death-penalty litigation . . . .”); King v. Strickland, 748 F.2d 1462, 1464 (11th Cir. 1984) (similar); see also In re Sterling-Suarez, 323 F.3d 1, 4–7 (1st Cir. 2003) (Torruella, J., dissenting). 27 ABA, GUIDELINES FOR THE APPOINTMENT & PERFORMANCE OF DEFENSE COUNSEL IN DEATH PENALTY CASES 1.1 commentary (2003). 28 Marsh had never defended a first-degree murder charge before but that does not give him an excuse to be ineffective. 9 introduced by the prosecutor,”29 and specifically required counsel to collect a “medical history,” “educational history,” “family and social history,” and “prior adult and juvenile record.”30 Marsh failed to do this although he knew of this duty. In October 1992, three years before the 1995 resentencing, Marsh consulted with Professor Welsh White at the University of Pittsburgh Law School. Professor White gave Marsh literature on the topic of the relationship between child abuse and brain damage. He also gave him a sample motion to obtain psychological and neuropsychological testing.31 The Court of Common Pleas found in the 1999 Pennsylvania PCRA hearing that “Marsh recognized that child abuse and neglect were significant factors in [Lesko’s] argument that a life sentence should be imposed, rather than death, and he agreed that evidence of brain damage would be a ‘very important matter to bring to the attention of the jury.’”32 Marsh, however, did nothing with the material. He never consulted a neuropsychologist – the type of specialist who could test for brain damage. On the eve of the 1995 resentencing, after jury selection had begun, Marsh retained a clinical psychologist, Dr. Herbert Levit. Dr. Levit was not qualified to do a neuropsychological assessment of Lesko’s cognitive functioning. Nor did Marsh advise Dr. Levit that 29 Wiggins v. Smith, 539 U.S. 5190, 524 (2003) (quotation marks and citation omitted) (emphasis added), 30 ABA GUIDELINES FOR THE APPOINTMENT AND PERFORMANCE OF COUNSEL IN DEATH PENALTY CASES 11.4.1(C) (1989). 31 Appx. 136. 32 Appx. 136. 10 brain damage was an area of relevant inquiry.33 In the limited review that he had time to make before his 1995 testimony, Dr. Levit diagnosed Borderline Personality Disorder and Polysubstance Abuse.34 He testified to this effect at the 1995 resentencing. In 1999, Lesko’s conviction and death sentence were reviewed by the Pennsylvania Court of Common Pleas in a Post-Conviction Relief proceeding. New counsel was appointed for Lesko. After a review of the CYS records, new counsel had neuropsychological testing done by Dr. Barry Crown. Dr. Crown testified at the PCRA hearing that both Lesko’s institutional records and Dr. Levit’s evaluation contained indicia of brain damage.35 Dr. Crown also noted that Dr. Levit’s diagnosis of Borderline Personality Disorder raised the possibility of organic brain damage because of the overlap of the symptoms of the two conditions.36 As a result of the complete battery of neuropsychological tests that Dr. Crown performed on Lesko in October 1999, Dr. Crown “concluded to a reasonable degree of neuropsychological certainty that [Lesko] is brain damaged and was brain damaged at the time of [Officer Miller’s murder].”37 Dr. Crown determined that the type of brain damage from which Lesko suffers “constitutes an extreme mental and emotional disturbance (42 Pa.C.S. § 9711(e)(2) 33 Appx. 137. 34 Appx. 137. 35 Appx. 138. 36 Appx. 140. 37 Appx. 140–41. 11 mitigating circumstance) and a significant impairment in his ability to appreciate the criminality of his conduct or conform his conduct to the requirements of law (42 Pa.C.S. § 9711(e)(3) mitigating circumstance).”38 The prosecution presented no expert testimony at the PCRA hearing in opposition to that of Dr. Crown. Dr. Levit did a complete review of the institutional records in 1999 for the PCRA proceeding. He now recognizes the deficiencies in his original diagnosis and in his 1995 testimony. With the additional information, information that attorney Marsh had never supplied to him, Dr. Levit states that he would have recommended neuropsychological testing to confirm the likelihood of brain damage. Dr. Levit’s revised 1999 diagnoses and findings include Post-Traumatic Stress Disorder, failure to thrive syndrome, substantial impairment in Lesko’s capacity to appreciate the criminality of his conduct and conform his conduct to the requirements of the law, probable brain damage, and the failure of social services to rescue Lesko and his siblings from the abject environment in which they were raised.39 This is the testimony that Dr. Levit would have given in 1995 if Marsh had performed as effective counsel and had given the necessary records to Dr. Levit for examination prior to his 1995 testimony. Marsh’s failure to obtain the CYS records sooner and his failure to have neuropsychological testing done can be attributed to his complete lack of understanding of how such material can be used as mitigating evidence. At the 1999 38 Appx. 141. 39 Appx. 141. 12 PCRA hearing, in response to the question, “What is your understanding of the relationship between Mr. Lesko’s background and potential mental health issues?” Marsh replied: The background information, in my mind, wasn’t for mental health purposes. It was to generate sympathy of the jury and to explain why he may have acted the way he did. I didn’t consider that having a rough childhood had a whole lot of bearing on what a psychologist could determine by scientific testing.40 This personal opinion does not excuse Marsh from preparing Lesko’s defense or from performing the necessary duties and functions that he owed to Lesko. It does perhaps explain Marsh’s failure to adequately present mitigating evidence. It is that failure, however, that renders Marsh’s representation of Lesko ineffective. Based on the 1999 testimony of Drs. Crown and Levit, the Court of Common Pleas held that “the testimony of a neuropsychologist at the sentencing hearing with regard to [Lesko’s] organic brain damage would have added significant additional and relevant information for the jury to consider as it weighed mitigating factors against the aggravating factors.”41 40 Appx. 144. 41 Appx. 142. 13 I agree with the Court of Common Pleas. I believe that Marsh was clearly ineffective in his case preparation, specifically in his failure to obtain the records in a timely fashion, to thoroughly review them and to arrange for relevant expert testing and for testimony by a qualified and informed expert witness on Lesko’s brain damage, his extreme mental and emotional disturbance, and his impaired ability to appreciate the criminality of his conduct or to conform his conduct to the requirements of law. To be sure, Marsh retained Dr. Levit, who performed at least a cursory review of relevant records, and presented some evidence of Lesko’s mental disturbance. That is insufficient under these circumstances. When Marsh was informed by Professor White that an entirely distinct avenue of mitigation evidence—evidence of Lesko’s organic brain damage—would be viable, Marsh needed to do more than just present evidence of Lesko’s “rough childhood.” A defense lawyer representing a defendant in a death penalty case has a duty to develop this type of mitigating evidence from the records that are available to him.42 Marsh failed to do so. II. PREJUDICE The Court of Common Pleas went on to determine that there is a reasonable probability that the outcome of the proceedings would have been different if counsel had adequately prepared the mitigation case.43 I agree with that holding. The demonstration of actual physical brain damage, as opposed to just psychological problems caused by a “rough 42 Williams v. Taylor, 529 U.S. 362, 393, 397 (2000). 43 Appx. 163. 14 childhood,” would be a strong argument against the imposition of the death penalty. Marsh had to persuade just one juror to vote for life in prison as opposed to execution.44 I believe that the fact that in January 1980 Lesko’s brain damage had caused him “extreme mental and emotional disturbance” and was “a significant impairment in his ability to appreciate the criminality of his conduct or conform his conduct to the requirements of law” would have had a strong mitigating effect on the jury’s deliberations. In 1999, Dr, Levit, after a sufficient review of the relevant records, changed his 1995 diagnosis and testified at the PCRA hearing that Lesko’s capacity to appreciate the criminality of his conduct and conform his conduct to the requirements of the law, was substantially impaired. Effective counsel for Lesko would have presented this testimony from Drs. Levit and Crown at the resentencing hearing in 1995. It should not have been delayed until 1999. Moreover, the prosecution at the PCRA hearing did not contest either Dr. Crown’s opinion or the revised diagnosis of Dr. Levit. Under the ABA Guidelines for the Appointment and Performance of Counsel in Death Penalty Cases,45 an effective attorney has a duty to discover all reasonably available mitigating evidence regarding the defendant. Under this standard, Marsh was obliged to demonstrate to the jury Lesko’s organic brain damage and the resulting inability to conform his conduct to the requirements of the law. There are many who 44 Jermyn v. Horn, 266 F.3d 267, 309 (3d Cir. 2001); Commonwealth v. Brown, 872 A.2d 1139, 1150-51 (Pa. 2005). 45 11.4.1(C) (1989). 15 believe that the death penalty is not an appropriate punishment for a defendant who cannot appreciate the criminality of his conduct. Only one juror needed to agree that the death penalty was inappropriate for Lesko. Only one juror needed to vote for life imprisonment for that to be the sentence that Lesko received. The probability of this occurring I find to be compelling. For the above reasons, I believe that Lesko was prejudiced by Attorney Marsh’s inadequate preparation of mitigating evidence to present at the sentencing hearing. III. UNREASONABLE APPLICATION OF FEDERAL LAW The Pennsylvania Supreme Court did, of course, overturn the award of a new sentencing hearing by the Court of Common Pleas. However, I believe that the Pennsylvania Supreme Court’s review of the decision of the Court of Common Pleas was an unreasonable application of federal law both as to ineffectiveness and as to prejudice.46 As to ineffectiveness, the court determined that “counsel undertook a reasonable investigation.”47 While it is true that Marsh consulted with Professor White, Marsh failed to utilize the relevant mitigating evidence that Professor White explained to him or to obtain the expert evidence that was necessary to present to the jury. Yet, Marsh has admitted that, three years before the resentencing, as a result of his consultation with 46 See Williams, 529 U.S. at 399. 47 Commonwealth v. Lesko, 15 A.3d 345, 381 (Pa. 2011). 16 Professor White, he knew of the significance of child abuse and neglect and “agreed that evidence of brain damage would be a ‘very important matter to bring to the attention of the jury.’” Moreover, Marsh subpoenaed the CYS records on February 7, 1995, one day before the last day of jury selection and two days before the first day of the Commonwealth’s case- in-chief. Marsh spent “[p]robably an hour or so . . . go[ing] through” the hundreds of pages of CYS records “in the Judge’s ante-room.”48 He requested the records at the last minute, reviewed them for just one hour, and did not recall if he had a strategic reason for failing to request the records sooner or if/how he determined whether records had “value.” However, contrary to this position, Marsh said at the PCRA hearing that he believed he “was obligated to put as much information before the jury as [he] could” and admitted that many records did have value.49 The Pennsylvania Supreme Court held that this record review was cumulative and cited Strickland. That is not the appropriate legal standard for this case. Evidence of Lesko’s organic brain damage is not cumulative to his psychological problems; it is a disability that is distinct from any evidence of his psychological or emotional issues. Moreover, it is a cause of mitigation, in and of itself. Under the Pennsylvania Code, a result of organic brain damage, the inability to appreciate the criminality of one’s conduct or conform one’s conduct to the requirements of law, is a separate consideration for mitigation.50 Accordingly, the Pennsylvania Supreme Court 48 Appx. 2832:13–16. 49 Appx. 2831:8–10. 50 42 Pa.C.S. § 9711(e)(3) 17 erred in deciding that the CYS evidence related to Lesko’s brain damage and the 1999 opinions drawn from those records by Drs. Crown and Levit was cumulative of evidence of his psychological issues. In my view, the appropriate test here of the adequacy of pretrial preparation is Wiggins51 and Williams,52 not the generalized language of Strickland.53 Under Wiggins and Williams, Marsh’s delays in investigating and his failure to timely review the records of Lesko’s “nightmarish childhood” and to order neuropsychological testing to detect any physical brain damage caused during that childhood rendered his representation of Lesko clearly ineffective. In short, Marsh’s failure to investigate and explore the distinct avenue of mitigation evidence related to brain damage suffered by Lesko amounted to inadequate performance. In addition, as to prejudice, the Pennsylvania Supreme Court opined that the circumstances of the shooting of Officer Miller were so “aggravating” that additional mitigation evidence could not have supported Strickland relief: not “when the aggravating circumstances were so patently grave.”54 However, as an attorney in a first-degree murder case, Marsh had a duty “to discover all reasonably available mitigating evidence.”55 There is no quantity of aggravation that cuts off the right—or the duty—of defense counsel to 51 539 U.S. at 527. 52 529 U.S. at 395. 53 466 U.S. at 691. 54 15 A.3d at 385. 55 ABA Guidelines for the Appointment and Performance of Counsel in Death Penalty Cases 11.4(C) (1989). 18 present mitigating evidence. A defense attorney, representing a defendant in a horrible crime, is not justified in ignoring mitigating evidence just because of the horror of the crime. Lesko’s miserable childhood, the problems, physical and psychological, caused by that childhood, especially the organic brain damage and the related inability to appreciate the criminality of his actions, were all items of reasonably available mitigating evidence. The horror of the crimes committed in no way cut off either Lesko’s right to effective counsel or the defense attorney’s duty to present mitigating evidence. Because of Marsh’s delay in obtaining and reviewing records and in consulting Dr. Levit and because of Marsh’s failure to follow up on the mitigating issue of brain damage, clearly his trial preparation was not reasonable. It was ineffective. Lesko’s right to the effective assistance of counsel, as defined in Strickland v. Washington,56 was violated. As a result, Lesko was prejudiced by the failure to present this evidence to the jury. It took only one juror to vote for life imprisonment. For the above reasons, I believe that Marsh’s representation of Lesko was ineffective, that this ineffective assistance of counsel prejudiced Lesko and that Lesko’s sentence should be vacated, and a new sentencing hearing be awarded. 56 466 U.S. 668 (1984). 19