Charles Martin v. Jeffrey Nines

Opinion

UNPUBLISHED

UNITED STATES COURT OF APPEALS FOR THE FOURTH CIRCUIT

No. 24-6086

CHARLES BRANDON MARTIN, Petitioner - Appellee,

v.

JEFFREY NINES, Acting Warden; ATTORNEY GENERAL OF MARYLAND, Respondents - Appellants.

Appeal from the United States District Court for the District of Maryland, at Baltimore. Julie R. Rubin, District Judge. (1:20-cv-02602-JRR)

Argued: September 24, 2024 Decided: January 16, 2025

Before NIEMEYER, GREGORY, and HEYTENS, Circuit Judges.

Affirmed by unpublished opinion. Judge Gregory wrote the opinion, in which Judge Heytens joined. Judge Niemeyer wrote a dissenting opinion.

ARGUED: Andrew John DiMiceli, OFFICE OF THE ATTORNEY GENERAL OF MARYLAND, Baltimore, Maryland, for Appellants. Nicole Houston Welindt, SKADDEN, ARPS, SLATE, MEAGHER & FLOM LLP, Palo Alto, California, for Appellee. ON BRIEF: Anthony G. Brown, Attorney General, OFFICE OF THE ATTORNEY GENERAL OF MARYLAND, Baltimore, Maryland, for Appellants. Shay Dvoretzky, Parker Rider-Longmaid, Sylvia O. Tsakos, SKADDEN, ARPS, SLATE, MEAGHER & FLOM LLP, Washington, D.C., for Appellee.

Unpublished opinions are not binding precedent in this circuit.

2

GREGORY, Circuit Judge:

This case concerns the habeas petition of Charles Brandon Martin, who is serving a life sentence for a conviction as an accessory before the fact to murder in the first degree. Martin challenges his conviction based on an undisclosed computer forensics report that substantially undermines the testimony of a key State witness. The court below found that the Maryland Court of Special Appeals (hereafter “state appellate court”) engaged in an unreasonable application of clearly established Supreme Court precedent in finding the suppressed report to be immaterial under Brady v. Maryland. The district court then ordered Martin to be released within sixty days unless retried and convicted. The State of Maryland appealed, and this Court stayed the order of release pending appeal on May 20, 2024.

On review, we agree with the district court’s determination that the state appellate court unreasonably applied clearly established Supreme Court precedent. We also find that the conditional release order was not an abuse of the district court’s discretion. Accordingly, we affirm the judgment of the district court granting habeas relief.

I.

On October 27, 2008, Jodi Lynne Torok was found unconscious on her foyer floor after suffering a gunshot wound to the head. J.A. 727. At the time, she was two months pregnant. Id. Paramedics responded to the scene and took Torok to the hospital, where she would eventually survive. J.A. 728. Police recovered a .380-caliber projectile and shell casing, Torok’s cell phone, and a Gatorade bottle from Torok’s home. J.A. 728–30. The police found the casing and projectile near the front door, while the Gatorade bottle

3

was “on the other side of the couch from where [Torok] was found.” J.A. 1331, 1344– 1345. The mouth of the Gatorade bottle was wrapped in layers of duct tape and white medical tape in a rectangular shape, and there was a hole in the bottom of the bottle surrounded by black soot. J.A. 728, 740, 2958, 2961, 2964.

Prosecutors theorized that Martin was responsible for the attempted murder, though not necessarily the shooter himself. See J.A. 728–29. Torok testified at trial that she had been in a relationship with Martin and had recently informed him that she was pregnant.1 J.A. 727. Martin had asked Torok to obtain an abortion, but she declined, informing Martin of her intent “to go to court and take him for child support.” Id. Martin was married and dating two women beyond Torok, Sheri Carter and Maggie McFadden––both of whom are relevant in this case. See J.A. 2258–59. The State argued that, to protect his marriage and other relationships, Martin needed to ensure that Torok’s pregnancy was not carried to term. See id.

The State also presented text messages that allegedly showed Martin was attempting to confirm Torok was at home. Martin asked Torok “What time do you work?”, to which she responded: “I’m off.” J.A. 1394–95. An hour later, Martin messaged “Hello.” J.A. 1396. At 5:11 PM, roughly two hours after the shooting, Martin sent another message: “I got some stuff with the kids to about 7:00, so any time after. How much did you need?” J.A. 1397–98. These messages, in the State’s view, provided clear evidence that Martin was involved in the shooting.

1

Torok lost all recollection of the day of the shooting and roughly the month thereafter, but she was able to testify to prior events. J.A. 728.

4

However, significant evidence countered the State’s theory of motive. For example, Torok testified that she had been in a romantic relationship with another man, Emmanuel Quarterly, who may have been the father. J.A. 1259. Martin had stated in a police interview that he highly doubted he was the father of Torok’s child because she had a boyfriend. J.A. 2910. And even if Martin were the father, his wife testified that she was aware of Martin’s relationships with other women and that he had two additional children outside of their marriage. J.A. 2190, 2210.

As the State’s theory goes, Martin, upon learning of Torok’s refusal to obtain an abortion, solicited his friend, Jerry Burks, to kill Torok and assisted Burks in the murder attempt by constructing himself, or helping Burks to construct, a silencer made from the Gatorade bottle found at the scene. J.A. 775. The State first tried Burks for the attempted murder, and a jury acquitted him on all counts. J.A. 729. After failing to secure a conviction of Burks, the State turned to Martin, charging him with (1) soliciting Burks to murder Torok and (2) as an accessory before the fact to attempted murder in the first degree.

A.

At trial, the State presented evidence from several relevant witnesses to show both that the Gatorade bottle was a silencer, that Martin constructed said silencer, and that Martin intended the silencer to be used to kill Torok. See J.A. 729–31. Several witnesses testified to forensic evidence linking Martin to the Gatorade bottle and shell casing found at the scene. David Exline, a forensic analyst, testified that he examined the medical tape from the mouth of the Gatorade bottle and discovered two hairs on the tape: one from a cat, the other a human. J.A. 1499–1504, 1526. Exline testified that the tape resembled

5

tape found in the residence of Maggie McFadden, the location where Martin allegedly constructed the silencer. J.A. 1499–1504.

The human hair was then analyzed by Dr. Terry Melton, an expert in mitochondrial DNA testing. J.A. 728, 762. Dr. Melton explained that mitochondrial DNA can show that someone is from the same maternal lineage, but it “can never say for sure this hair absolutely for sure came from this person.” J.A. 1815. Dr. Melton testified that she could rule out 99.94% of North Americans as contributors to that DNA sample, but not Martin. J.A. 1833. Dr. Melton did acknowledge on cross-examination that this left roughly 180,000 Americans and 30,000 people in Maryland with the same mitochondrial DNA profile. J.A. 1945, 1947.

The State next called Anne Arundel County senior forensic chemist, Sarah Chenoweth, to testify about nuclear DNA testing performed on DNA found on the mouth of the bottle. This included DNA from “at least three individuals,” including at least one male and one female. J.A. 1981. Chenoweth could not rule out Martin nor Torok as contributors to the DNA samples. J.A. 1997. She did, however, rule out three other individuals submitted for comparison. Id. However, the DNA was not compared to that of alternative suspects suggested by the Defense, including Sheri Carter, Emmanuel Quarterly, Maggie McFadden, or Michael Bradley, the brother of McFadden. J.A. 2003.

Arnold Esposito, an expert in firearms and ammunition, J.A. 1587, then testified to the shell casing collected from the scene. He concluded that the gun used in the shooting could have come from any one of sixteen manufacturers, including Jennings. J.A. 1596–

6

97. Reviewing a firearms tracing report, Esposito testified that Martin had purchased two .380 caliber Jennings firearms in 2003. J.A. 1599–1600.

The only source of testimony about Gatorade bottles being used as silencers came from Detective Richard Alban, who was called to testify about discovering the bottle at the scene. While not rendering any opinion as to whether the bottle was or was not a silencer, Detective Alban testified that the bottle reminded him of a makeshift silencer he’d seen in a Steven Seagal movie. J.A. 1336. He also testified to having watched YouTube videos in which plastic bottles were used to produce silencers. Id. Detective Alban acknowledged on cross that Gatorade bottles could be used to smoke marijuana. J.A. 1345. No other witness––expert or otherwise––testified to Detective Alban’s speculation that a Gatorade bottle could be used as a silencer. While another detective at the scene testified to not smelling marijuana in the bottle, J.A. 1625, no testing was ever done to determine whether marijuana or gunshot residue was present inside. J.A. 1783.

B.

After presenting evidence that Martin’s DNA was present on the Gatorade bottle and Detective Alban’s view that the bottle was a silencer, the State called two key witnesses to link Martin to the creation of that Gatorade bottle silencer.

The State first called Michael Bradley, who testified that he may have seen Martin and Burks constructing the Gatorade-bottle silencer. For background, Bradley and his sister, Maggie McFadden, lived together with their other siblings, Frank and Dennis, as well as McFadden’s daughter. J.A. 2011. Bradley knew Martin because Martin had been dating McFadden and frequently stayed over at their home. J.A. 2008–11. Bradley had

7

seen Martin with a handgun in McFadden’s room in the year prior to the shooting. J.A. 2015–16.

Bradley testified that, on the day of the shooting, Frank, Burks, Martin, and himself were in the house smoking marijuana. J.A. 741–42. They “smoked five or six blunts that day,” and Bradley was admittedly intoxicated. Id.; J.A. 2021–22. Bradley testified that he saw (1) Frank take white medical tape to the kitchen, (2) Martin and Frank go upstairs to McFadden’s room, and (3) Frank come down to retrieve a Gatorade bottle from the kitchen and return upstairs. J.A. 2023–24. Martin and Burks then left the house. J.A. 2025. Bradley initially testified that when they returned, he did not see Martin and Frank have any conversation. J.A. 2038. But after further prodding from prosecutors, he changed his tune, recalling seeing Martin hand Frank a brown paper bag and instruct Frank to “get rid of this.” J.A. 2039. Bradley never testified to seeing Martin with white medical tape or the Gatorade bottle.

On cross examination, Bradley’s testimony was substantially impeached beyond just his intoxication. Bradley admitted that he received immunity from prosecution for Torok’s shooting in exchange for his testimony. J.A. 2041. Further, Bradley received a benefit to his pending obstruction of justice charges in New Jersey, though it is unclear if Bradley realized that he received any benefit. J.A. 2047, 2053, 2297. Bradley admitted that the New Jersey charges stemmed from lying to the police. J.A. 2046. Bradley also admitted that he only testified because McFadden asked him to because “she was involved” in the attempted murder. J.A. 2043.

8

The second key witness, and perhaps the most important, was Sheri Carter. Carter had been in a relationship with Martin for three years. J.A. 2061–62. She testified that Martin kept a computer at her house, and that in September or October of 2008, she saw Martin looking up gun silencers on that computer. J.A. 731. This computer was “unique” in that it came from Martin’s employer, the College of Southern Maryland (“CSM”), and had many restrictive settings. J.A. 2065–66. This included the inability to modify system files or download new software without an administrator password. J.A. 2066. Carter testified that Martin “got rid of” the computer because they “had looked up so many crazy things on the internet that in case [her] apartment got searched he didn’t want it found there.” J.A. 731, 2066. She also testified that she had seen Martin with a handgun in September and October 2008. J.A. 729–730. Carter’s testimony was the only testimony at trial linking Martin to silencers, and her testimony was unimpeached.

C.

The Defense presented multiple alternative theories of the case. To begin, Martin argued that the bottle was not a silencer but, rather, a marijuana smoking device. Thus, the presence of Martin’s DNA on the bottle would not mean that he had any role in constructing the silencer, if it even was one. See J.A. 2299–2300. This theory was corroborated by the fact that Torok’s DNA was found on the Gatorade bottle despite Torok’s testimony that she never drank Gatorade. J.A. 1262. The state’s silencer theory failed to explain the presence of Torok’s DNA, but the bottle’s use as a smoking device would explain such presence.

9

Martin also pointed to several others with motive to harm Torok due to Torok’s pregnancy, including Quarterly, who was potentially the father; McFadden and Sheri Carter, as Martin’s disgruntled romantic partners; and Michael and Frank Bradley, who, again, were McFadden’s brothers.

Chief among these alternate suspects was Maggie McFadden, whose erratic and threatening behavior toward other women in relationships with Martin made her a particularly compelling suspect. For example, Sheri Carter testified about volatile interactions she had with McFadden after McFadden learned of Carter’s relationship with Martin. J.A. 2073. Carter testified that, in 2009, McFadden called her and asked to meet in person. J.A. 2072. During that meeting, McFadden told Carter that “she liked to beat people up” and “beat [Martin] up on a regular basis.” J.A. 2074. McFadden expressed that “if people got in her way she, you know, knew how to take care of it.” Id. She told Carter that she “had a gun on her and that she’d brought it for protection because she didn’t know what [Carter] was like.” J.A. 2073. She even bragged that “she’d had someone shot at one point.” J.A. 2074. After this meeting, McFadden continued to threaten Carter on social media and made disparaging comments about her in emails to Carter’s employer. J.A. 2087–88. Additional evidence also pointed to McFadden’s involvement in the shooting, as McFadden’s friend, Steve Burnette, had lied to the police about his whereabouts during the shooting and “was never able to explain what he was doing” that day. J.A. 2292.

The Defense’s theory that McFadden was responsible for the shooting was consistent with other evidence in the record that may otherwise have pointed toward Martin. For example, the State had presented evidence that Martin was an avid Gatorade drinker to link

10

Martin to the alleged silencer, J.A. 729; however, Defense counsel highlighted that any bottle in McFadden’s home would likely have had his DNA on it. Rather than demonstrating Martin’s guilt, the presence of Martin’s DNA on the bottle was entirely consistent with anyone with access to McFadden’s home having committed the shooting.

The State’s rebuttal to the McFadden theory was that it was far-fetched because McFadden’s volatile and aggressive behavior would “draw attention” to her as a suspect. J.A. 2269. In their view, it was simply “not logical” that someone who would “threaten people and e-mail people and send nasty pictures to people,” and who “had someone shot” in the past, could have been responsible for the attempt on Torok’s life. Id.; J.A. 2074.

Finally, the Defense also criticized investigators’ sloppy handling of the case. J.A.

2292–93. For example, the State had repeatedly “lost” pieces of potentially exculpatory evidence. This included street camera footage that would have showed whether Martin had driven toward Torok’s house that day and a recording of a police interview in which Bradley endorsed Martin’s stated alibi, in direct conflict with Bradley’s trial testimony. See J.A. 1634, 2295. Detectives also repeatedly failed to conduct DNA testing on any of the alternative suspects put forward by Martin, instead “focusing on [Martin] from the very first day of the investigation.” J.A. 2292.

D.

Before the case was submitted to the jury on May 4, 2010, the State dismissed all counts except attempted first degree murder and solicitation, dropping the lesser included offenses to attempted murder. J.A. 317. At closing arguments, the State emphasized the importance of Carter’s testimony in linking Martin to the use of the Gatorade bottle as a

11

silencer. The Prosecution noted Carter’s credibility, explaining that “the things that [Carter] says are very powerful and she is someone who is very sane and very normal but caught up in a bad situation.” J.A. 2330. They asked:

So is anyone surprised that Sheri Carter saw the Defendant researching silencers on the internet? Natural place to go. Is anyone surprised that the Defendant got rid of that computer after the police talked to him? No, it fits precisely with the evidence.

J.A. 2262. The Prosecution used this testimony as evidence of Martin’s “planning” of the attempted murder, arguing the research of silencers was evidence of his “intent to kill.” J.A. 2276. And to demonstrate the importance of this testimony, the Prosecution summed up its case as follows: “If you decide that [Martin] made that silencer and that [the] silencer was intended to be used upon the victim then he is guilty.” J.A. 2265–66.

Carter’s testimony also yielded a jury instruction regarding the concealment of evidence. J.A. 731. The jury was instructed as follows:

You have heard evidence that the Defendant removed a computer from the house of Sheri Carter. Concealment of evidence is not enough by itself to establish guilt but may be considered as evidence of guilt. Concealment of evidence may be motivated by a variety of factors, some of which are fully consistent with innocence. You must first decide whether the Defendant concealed any evidence in this case. If you find that the Defendant concealed evidence in this case then you must decide whether that conduct shows a consciousness of guilt.

J.A. 2913. Crucially, this instruction allowed the jury to infer guilt from Martin’s alleged decision to get “rid of” of the laptop due to his search history. J.A. 731. The only basis for this instruction was Carter’s testimony, which even the State agreed was “highly damaging.” J.A. 2423.

12

The jury found Martin guilty of attempted first degree murder but acquitted him on the solicitation charge. J.A. 731. Martin was sentenced to life imprisonment. J.A. 2533. Martin directly appealed to the Maryland intermediate appellate court, and the verdict was affirmed. J.A. 732. His petitions for certiorari to Maryland’s highest court and the Supreme Court of the United States were denied. Id.

E.

After his conviction, Martin’s mother sent a public records request to the police, which yielded a never-produced forensic computer report analyzing five computers belonging to Martin. J.A. 732, 2879–86. Martin filed a postconviction claim in state court under Brady v. Maryland based on the content of this suppressed report. J.A. 735.

The report indicated that Detective Regan requested forensic analysis of five computers seized from Martin’s residence, one of which was the laptop from the College of Southern Maryland (“CSM”) with its many restrictive settings. J.A. 2879–86. The computers were searched for various terms including “handgun,” “Gatorade,” “silencer,” “contract murder,” “murder for hire,” “homemade silencer,” and “hitman.” J.A. 732–33. For the CSM laptop and three others, these search terms failed to return any results. Id.; see also J.A. 2879–86. No data could be obtained from the fifth computer. J.A. 2886.

At a postconviction hearing, Martin’s trial counsel testified that the report would have undermined Carter’s testimony because it showed that the CSM laptop was never used to research silencers, as Carter claimed. J.A. 733. Carter had testified in detail to the CSM laptop being the one that Martin specifically kept at her apartment, and this report directly contradicted her devastating testimony. J.A. 484–85. Further, the postconviction

13

state court held that, given that the allegedly concealed laptop was in police custody, there is a substantial probability that the concealment of evidence jury instruction would not have been given. J.A. 489, 734.

At the same hearing, the State called Daniel Giambra of the Anne Arundel Police Department to testify to a second forensic analysis of the computer. See J.A. 2650–51. He testified that his analysis was limited, as the hard drive failed ten minutes into his analysis. J.A. 2661, 2666. However, based on those ten minutes with the computer, he formed the opinion that the computer had not been accessed since 2005, years prior to Torok’s shooting. Id. This opinion was based only on the system files which, as both Martin and Carter testified, could only be modified by an administrator. Id.; J.A. 2065–66.

On October 5, 2018, the postconviction state circuit court ordered a new trial, finding that a Brady violation had occurred. J.A. 734. That court determined that law enforcement’s failure to disclose the report was “likely willful.” J.A. 486. The report was both exculpatory and impeaching because it undermined Sheri Carter’s testimony and showed that Martin did not destroy evidence. J.A. 484–86. Because Carter’s testimony was the only link between Martin and homemade silencers, and the only basis for the concealment of evidence jury instruction, the Court found the suppression of this evidence to be material. J.A. 734. The post-conviction court also noted that “[Maggie] McFadden had just as much of a motive, if not more, to get rid of [] Torok” than Martin did, particularly because Martin “already had . . . several other children out of wedlock” of which his wife was aware. J.A. 2440–41.

14

The state appellate court reversed, finding that the report was not material. J.A. 726.

The court discussed the substantial evidence it considered supportive of the verdict, including the DNA evidence on the Gatorade bottle, the similarity of the medical tape found on the bottle and in McFadden’s home, the testimony of Michael Bradley, the firearms analysis, the evidence of Martin’s motive, and text messages from Martin to the victim which the State argued were sent to confirm the victim was home at the time of the shooting. J.A. 739–43. While the state appellate court acknowledged that there would have been no basis for the concealment of evidence jury instruction given the report, the court nonetheless concluded that there was no reasonable probability that trial would have yielded a different result had the report not been hidden. Id.

Maryland’s highest court and the Supreme Court of the United States again denied certiorari. J.A. 2893–94.

F.

Martin filed a petition for federal habeas relief under 28 U.S.C. § 2254. J.A. 2894.

The petition was originally filed pro se and modified twice after attaining two sets of counsel. J.A. 2887. Martin asserted five claims 2 before the district court, including a claim that his conviction violated Brady v. Maryland because of the suppressed forensic laptop report. J.A. 2894.

2

The other claims asserted that (1) “trial counsel was ineffective because he failed to object to compound voir dire questions;” (2) “trial counsel was ineffective because he failed to object to the state’s burden-shifting closing argument;” (3) “the trial court erred when it failed to require the state to provide a bill of particulars;” and (4) “the state denied Martin due process when it changed its theory of the case mid-trial.” J.A. 2894.

15

The district court granted Martin’s petition with respect to the Brady claim but denied all other claims. J.A. 2888. That court found that the state appellate court had engaged in a Brady analysis “contrary to” clearly established federal law and “imposed an unreasonable application of facts to the law” when it found that the suppressed computer report was immaterial. J.A. 2908; see also J.A. 2911. The court then issued a conditional order of release unless Martin was retried within sixty days and stayed its order for thirty days pending an appeal. J.A. 2921. The court revised and clarified this order on January 30, 2024. J.A. 2941.

The State filed a notice of appeal and sought an emergency stay pending appeal.

The stay was granted on May 20, 2024, ECF No. 29, and argument in this case was heard on September 24, 2024.

II.

We review de novo the district court’s grant of habeas relief under 28 U.S.C.

§ 2254(d), based on the state record. Tucker v. Ozmint, 350 F.3d 433, 438 (4th Cir. 2003). When a state court has adjudicated a petitioner’s claim on the merits, we apply the standard of review set out in the Antiterrorism and Effective Death Penalty Act of 1996 (“AEDPA”), which provides that federal courts may grant relief if an individual is held “in custody in violation of the Constitution.” 28 U.S.C. § 2254. This standard is “highly deferential,” Lindh v. Murphy, 521 U.S. 320, 333 n.7 (1997), and “intentionally difficult to meet.” Woods v. Donald, 575 U.S. 312, 316 (2015) (quotation marks omitted).

16

Under AEDPA, a federal court may not grant a writ of habeas corpus unless the state court’s adjudication was either (1) “contrary to, or involved an unreasonable application of, clearly established federal law, as determined by the Supreme Court of the United States”; or (2) “based on an unreasonable determination of the facts in light of the evidence presented in the State court proceeding.” 28 U.S.C. § 2254(d). “‘[C]learly established Federal law’ under § 2254(d)(1) is the governing legal principle or principles set forth by the Supreme Court at the time the state court renders its decision.” Lockyer v. Andrade, 538 U.S. 63, 71–72 (2003).

“[A] state court’s decision is ‘contrary to’ clearly established federal law ‘if the state court applies a rule different from the governing law set forth in [the Supreme Court’s] cases.’” Appleby v. Warden, N. Reg’l Jail & Corr. Facility, 595 F.3d 532, 535 (4th Cir. 2010) (quoting Bell v. Cone, 535 U.S 685, 694 (2002)). A state court’s decision involves an “unreasonable application” of clearly established federal law if it “correctly identifies the governing legal principle from [Supreme Court] decisions but unreasonably applies it to the facts of the particular case.” Id. at 535–36 (quoting Bell, 535 U.S. at 694). A petitioner must show that the state courts’ application of federal law was “so lacking in justification that there was an error well understood and comprehended in existing law beyond any possibility for fair minded disagreement.” White v Woodall, 572 U.S. 415, 419–20 (2014) (quoting Harrington v. Richter, 562 U.S. 86, 103 (2011)). These well- established standards impose a “high bar for habeas relief.” Virginia v. LeBlanc, 582 U.S. 91, 96 (2017).

17

III.

“Clearly established federal law, determined in Brady v. Maryland, . . . and its progeny, provides that a state violates a defendant’s due process rights when it fails to disclose to the defendant prior to trial, ‘evidence favorable to an accused . . . where the evidence is material either to guilt or to punishment, irrespective of the good faith or bad faith of the prosecution.’” Spicer v. Roxbury Corr. Inst., 194 F.3d 547, 555 (4th Cir. 1999) (quoting Brady v. Maryland, 373 U.S. 83, 87 (1963)). Prosecutors must “disclose [favorable] evidence . . . even though there has been no request [for it] by the accused,” including evidence known only to police. Strickler v. Greene, 527 U.S. 263, 280–81 (1999); see also Kyles v. Whitley, 514 U.S. 419, 438 (1995). This is because, in the American system, the prosecutor “is the representative not of an ordinary party to a controversy, but of a sovereignty . . . whose interest . . . in a criminal prosecution is not that it shall win a case, but that justice shall be done.” U.S. v. Bagley, 473 U.S. 667, 675 n.6 (1985) (quoting Berger v. United States, 295 U.S. 78, 88 (1935)).

While prosecutors have a “broad duty of disclosure [,] . . . not every violation of that duty necessarily establishes that the outcome was unjust.” Strickler, 527 U.S. at 281. A Brady violation, thus, has three critical elements. First, the evidence must be “favorable to the accused, either because it is exculpatory, or because it is impeaching.” Id. at 281–82. Second, the evidence “must have been suppressed by the State, either willfully or inadvertently.” Id. at 282. Lastly, the withheld evidence must be “material” in that, were it not withheld, there is “a reasonable probability” that the jury would have reached “a different result.” Kyles, 514 U.S. at 434.

18

As for materiality, “[t]he question is not whether the defendant would more likely than not have received a different 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.” Id. Furthermore, this inquiry “is not a sufficiency of the evidence test[,]” and therefore “[a] defendant need not demonstrate that after discounting the inculpatory evidence in light of the undisclosed evidence, there would have been enough left to convict.” Id. at 434–35. A court commits legal error if it rejects a Brady claim because “the remaining evidence is sufficient to support the jury’s conclusions.” See Strickler, 527 U.S. at 290.

IV.

The parties agree, both here and in the courts below, that the first two Brady elements are satisfied. See J.A. 2909; Resp. Br 7. 3 As for the first element, the CSM report is both impeaching and exculpatory, both in how it undermines the testimony of Carter and in how it shows that Martin did not conceal evidence, where the laptop was in possession of police. For the second element, the evidence was undisputedly suppressed by the State. Thus, the dispute centers on materiality: whether the report “undermines confidence in the outcome of the trial.” Kyles, 514 U.S. at 434.

3

Despite conceding the first two elements in state court and continuing to concede that the report was impeachment evidence, the State now attempts to argue that the report is somehow unhelpful to Martin’s defense because he could have theoretically been researching silencers on a different computer. See Reply Br. 14. But the State abandoned that argument before the state appellate court and the state appellate court accepted that concession. See JA 733 (noting that the State had “abandoned on appeal” any argument “that the laptop mentioned in the Computer Analysis was not the same laptop discussed in Ms. Carter’s testimony”). We thus conclude that the State has forfeited any argument that Martin could have been researching silencers on a different computer.

19

A.

As discussed above, the Prosecution’s basic theory of the case was that Martin solicited Burks to murder Torok after learning she would not obtain an abortion, and he helped Burks in this attempted murder by constructing or helping construct a silencer made from a Gatorade bottle. Evidence at trial suggested that the Gatorade bottle was a silencer and linked Martin to the Gatorade bottle. But the only evidence connecting Martin to his potential construction of the Gatorade bottle for use as a silencer was Carter’s testimony, which was severely undermined by the forensic computer report. The question before us is whether any reasonable jurist could have concluded that the suppression of this report was immaterial.

The state appellate court began by correctly summarizing the relevant Supreme Court precedent on materiality. J.A. 738–39. But after stating the proper rule, the Court instead applies a different rule entirely, concluding that even without Carter’s testimony, there was still “strong evidence of [Martin’s] guilt.” J.A. 740–43. Despite repeating its conclusory statements that Martin “has not met his burden of showing that . . . there is a reasonable probability that the result of his trial would have been different,” J.A. 743, the court’s actual analysis goes to the sufficiency of the evidence, an approach that Kyles squarely rejected.

In Kyles, the Supreme Court explained that “[t]he possibility of an acquittal on a criminal charge does not imply an insufficient evidentiary basis to convict.” Kyles, 514 U.S. at 435. Rather, the operative question is whether there is a reasonable probability that one juror would have changed their mind based on the undisclosed evidence. Id. The

20

majority of the Court critiqued the dissent for applying the wrong standard, which appeared to “assume that [the petitioner] must lose because there would still have been adequate evidence to convict even if the favorable evidence had been disclosed.” Id. at 435 n.8. The Court pointed to specific examples of the dissent’s flawed analysis, including the dissent’s statements that the “possibility that [an individual] planted evidence ‘is perfectly consistent’ with [petitioner’s] guilt”; “[t]he jury could well have believed [portions of the defense theory] and yet have condemned petitioner because it could not believe that all four of the eyewitnesses were similarly mistaken”; “the Brady evidence would have left two prosecution witnesses ‘totally untouched’”; and that “Brady evidence ‘can be logically separated from the incriminating evidence that would have remained unaffected.’” Id. As the majority explained, the dissent stated the correct rule for materiality as being a reasonable probability of a different result. However, in its listing of evidence that supported the verdict without analyzing the impact of the undisclosed evidence on the entire case, the dissent applied the rule as a sufficiency of the evidence standard. This, the majority emphasized, was legal error. Id.

The state appellate court makes the same error as the dissent in Kyles, resulting in an unreasonable application of clearly established Supreme Court precedent. See id. (“This rule is clear.”). As appellate courts have summarized when applying this rule, we must “both add to the weight of the evidence on the defense side all of the undisclosed exculpatory evidence and subtract from the weight of the evidence on the Prosecution’s side the force and effect of all the undisclosed impeachment evidence.” Juniper, 876 F.3d at 568 (citations omitted). In contrast to a sufficiency of the evidence analysis, Kyles

21

requires courts to “exhaustively examine[] the suppressed evidence as well as the evidence introduced at trial . . . then carefully assess[] what purposes the suppressed evidence might have served and how that evidence might have affected the jury’s consideration of the evidence that was introduced.” Boss, 263 F.3d at 745–46. The state appellate court did no such thing.

As the district court noted, rather than focusing on how the removal of Carter’s testimony might have impacted the character of the entire case, the state appellate court focused on

the evidence it considered supportive of the verdict, including the mitochondrial and nuclear DNA on the Gatorade bottle, the similar characteristics of the medical tape removed from the Gatorade bottle, the tape found during the search of Maggie McFadden’s home, the testimony of Michael Bradley, the ballistic evidence showing Martin owned a gun that could have shot the recovered shell from the crime scene, the State’s theory that Martin had a motive to kill the victim, and text messages from Martin to the victim the morning of the shooting the State argued were sent for the purpose of confirming the victim was home at the time of the shooting.

J.A. 2910. The state appellate court focused on the “substantial” evidence “connecting” Martin to the attempted murder and focused on how a jury “could infer” a nefarious nature to Martin’s text messages. J.A. 740–43.

However, in contravention of what the law requires, the state appellate court never engaged with how the lack of evidence connecting Martin to silencers would severely weaken the Prosecution’s case. This should have included an analysis of whether impeaching Carter’s testimony would weaken the then-corroborative testimony of Michael Bradley––an admittedly intoxicated witness with substantial bias concerns and a pending obstruction of justice charge for lying to the police––and how such impeachment might

22

instill greater belief in the Defense’s theory that the Gatorade bottle was a smoking device, which at the time had been substantially undermined by Carter’s testimony that Martin was researching homemade silencers. Nor did the court grapple with how the Prosecution relied on and emphasized Carter’s statements and reliability in closing arguments. The state appellate court also entirely discounted the impact of the jury instruction regarding Martin’s concealment of evidence, despite agreeing that it may not have been given but for Carter’s unimpeached testimony. J.A. 743 n.14. And, as the district court noted, the state appellate court disregarded or misconstrued key facts supporting its finding there was substantial evidence of Martin’s guilt, including Torok’s admission that her unborn child may have been fathered by Quarterly, who she was seeing at the time, and Martin’s police interview expressing similar doubt. J.A. 2910 n.5. By failing to consider how Carter’s statements impacted the State’s case as a whole, the state appellate court did not properly analyze “whether the favorable evidence could reasonably be taken to put the whole case in such a different light as to undermine confidence in the verdict.” Kyles, 514 U.S. at 434.

On appeal, the State incorrectly claims that Kyles is the wrong framework for our analysis. Rather, it argues that the state appellate court’s opinion properly tracks the Supreme Court’s analysis in Strickler v. Greene. There, the Court declined to find that evidence that impeached the testimony of an eyewitness was material where two other eyewitnesses also placed the defendant at the crime scene, and the offense of conviction “did not depend on proof” offered in that testimony. Strickler, 527 U.S. at 292–96. While the Court did list the overwhelming amount of evidence supporting the conviction, it did so in the context of demonstrating the strength of the prosecution’s case, even had the

23

witness’s testimony been totally discredited. Id. Strickler never purported to alter the materiality analysis. To the contrary, the Supreme Court reiterated that it would be legal error to find undisclosed evidence immaterial on the basis of “the remaining evidence [being] sufficient to support the jury’s conclusions.” Id. at 290. Thus, the State’s argument is nothing more than an attempt to rewrite the “clear” rule established in Kyles, as was applied in Strickler, through selective quotes taken out of their broader context. Furthermore, Strickler’s facts also limit its applicability to the case now before us. As the Supreme Court explained, the remaining evidence in Strickler would have been untouched by impeachment of that witness’s testimony, in stark contrast to the case now before us, where the Prosecution’s other evidence would have been severely weakened by any impeachment of Carter. Further, the witness’s testimony in Strickler was not the basis of a jury instruction, nor was it “relied upon by the prosecution at all during its closing argument.” Id. at 295. These facts make Strickler a poor comparator and the State’s argument unavailing.

Clearly established Supreme Court precedent requires a nuanced analysis of the impact of the suppressed evidence on both sides of the case. While the state appellate court stated the correct rule from Kyles v. Whitley, its application was unreasonable, as explained by the majority opinion in Kyles itself.

B.

In conducting our analysis of whether the state court unreasonably applied clearly established law, AEDPA requires us to consider whether the state court’s application was an “error . . . beyond any possibility for fair minded disagreement.” White, 572 U.S. at

24

419–20. To do so, “a habeas court must determine what arguments or theories supported, or [] could have supported, the state court’s decision; and then it must ask whether it is possible fairminded jurists could disagree that those arguments or theories are inconsistent with the holding in a prior decision of [the Supreme] Court.” Harrington, 562 U.S. at 102. Considering all possible arguments and theories when applying Kyles, which we have established is the governing law, we now conclude that all fairminded jurists would agree that the undisclosed forensic computer report was material.

To begin, the forensic computer report was particularly valuable impeachment evidence because it provided the only means of undermining Carter’s damaging testimony. As this Court has held, impeachment evidence is more likely to be material under Brady when it undermines evidence of an essential element of the offense, especially when it provides the only significant basis for impeachment of that evidence. See United States v. Bartko, 728 F.3d 327, 339 (4th Cir. 2013); United States v. Parker, 790 F.3d 550, 558 (4th Cir. 2015). While Carter’s testimony was not the sole evidence for an essential element of Martin’s conviction as an accessory before the fact to first degree murder, it was by far the strongest. Without Carter’s testimony, the link between Martin and the construction of the alleged silencer is exceedingly weak. The presence of Martin’s DNA on the Gatorade bottle is consistent with entirely innocuous explanations, such as simply drinking or smoking from it. And Bradley’s testimony, the only other evidence linking Martin to the construction of the silencer, was already significantly undermined by his own admitted intoxication, the substantial benefit he received in exchange for his testimony, his history of lying to the police, and his decision to testify only at McFadden’s urging. Thus, Carter’s

25

testimony was crucial both for its own merit and in its bolstering of Bradley’s testimony. Under clearly established precedent, any reasonable jurist would find this to weigh strongly in favor of the suppression of the report being material.

For further evidence of the significance of Carter’s testimony in the State’s overall case against Martin, the Court need look no further than the Prosecution’s closing argument. As the Supreme Court has held, suppressed evidence is material where it impeaches a witness whose testimony was “stress[ed]” by the prosecution and “uncorroborated by any other witness.” Banks, 540 U.S. at 700; see also Dennis v. Sec’y, Penn. Dep’t of Corr., 834 F.3d 263, 298–99 (3d Cir. 2016) (en banc) (holding that a state court unreasonably applied Brady when it found that suppressed evidence that would have impeached testimony “emphasized” by the prosecution in closing was not material). Furthermore, the statements of prosecutors are especially probative, as “[t]he likely damage is best understood by taking the word of the prosecutor” in what they “contended during closing arguments.” Kyles, 514 U.S. at 444.

Here, the Prosecution elevated Carter’s testimony in its closing arguments, emphasizing her reliability and noting that “the things that [Carter] says are very powerful and she is someone who is very sane and very normal but caught up in a bad situation.” J.A. 2330. The prosecutor also asked whether anyone was “surprised that Sheri Carter saw the Defendant researching silencers on the internet.” J.A. 2262. The Prosecution used this testimony as evidence of Martin’s “planning” of the attempted murder, arguing the research of silencers was evidence of his “intent to kill.” J.A. 2276. The Prosecution summed up its case as follows: “If you decide that [Martin] made [the] silencer and that the silencer

26

was intended to be used upon the victim then he is guilty.” J.A. 2265–66. Carter’s testimony played a central role in the Prosecution’s closing argument and entire case. Any reasonable jurist would be compelled to find that evidence that successfully impeaches Carter’s testimony, at the time touted by the Prosecution as airtight, would be material.

Further, no fairminded jurist could accept the state appellate court’s discounting of the concealment of evidence jury instruction simply due to the otherwise “strong evidence of [Martin’s] guilt.” J.A. 743. That instruction is impossible to square with the suppressed impeachment evidence that contradicts its factual basis, as the CSM report showed that Martin neither researched homemade silencers nor concealed evidence. As the district court correctly noted, the instruction “permitted the jury to infer consciousness of guilt as to all elements of the crime from Carter’s testimony regarding destruction of the computer,” J.A. 2913, which, again, the forensic report reveals never happened. The removal of this instruction alone may likely have caused one juror to change their mind, as such a fundamental change to the jury’s instructions “undermines confidence in the outcome of the trial.” Kyles, 514 U.S. at 434.

We also consider the broader impact that impeaching Carter’s testimony would have had on the entirety of each party’s case. Such impeachment would have painted the Prosecution’s case in an entirely different light. Cf. Kyles, 514 U.S. at 441–54; see also Boss, 263 F.3d at 745–46. As discussed above, Carter corroborated the testimony of an intoxicated, admitted liar who received a substantial benefit for his testimony, and her testimony also turned Detective Alban’s “mere theory” seen in a movie and on YouTube into reality. See J.A. 2912–14. As a result, the forensic computer report, through

27

undermining Carter’s testimony, would have weakened the testimony of other State witnesses.

Additionally, impeaching Carter’s testimony also would have likely pushed the jury toward the Defense. This Court has found that evidence exposing “that a major prosecution witness was testifying falsely” is “likely” to have made the jury “more sympathetic to [the defendant’s] entire defense.” Monroe v. Angelone, 323 F.3d 286, 315 (4th Cir. 2003). Carter’s impeachment would have also strengthened the Defense’s alternative explanations of the evidence: that the Gatorade bottle was a smoking device, not a silencer, and that Maggie McFadden was responsible for the attempted murder. J.A. 2004, 2072–74, 2087, 2299–2300. Carter’s unimpeached testimony cut both arguments at their knees.

First, if Defense counsel had the opportunity at trial to use the forensic computer report to show that Martin had not been researching homemade silencers, it is far more plausible that a juror would believe that the Gatorade bottle served an alternative purpose, thus explaining away Martin’s DNA being present on the bottle. The bottle was found on the opposite side of the couch from where Torok was shot. J.A. 1344–45, 2955. The soot inside the bottle was never tested for gunshot or marijuana residue. J.A. 1783. And the bottle had saliva from three individuals, likely including Torok. J.A. 1997. The State never had an explanation for why Torok’s saliva would be found inside the Gatorade bottle given that she never drank Gatorade. J.A. 1262, 1285. Martin did have an explanation––that it was because Torok and Martin had both used the bottle to smoke––but this argument was

28

significantly undercut by Carter’s testimony about silencers. 4 A juror may have rightly questioned Detective Alban’s uncorroborated silencer theory had Carter’s testimony been impeached with the forensic computer report.

Second, with Carter’s testimony called into question, it becomes far more likely that at least one juror would have had reasonable doubt as to whether the true culprit was Maggie McFadden. McFadden, who was also dating Martin at the time, demonstrated extremely erratic and violent behavior. See J.A. 2073. As the post-conviction circuit court noted, “McFadden had just as much of a motive, if not more, to get rid of Ms. Torok” than Martin, particularly because Martin “already had . . . several other children out of wedlock”

4

The State argues that the district court erred in rejecting the state appellate court’s factual finding that the Gatorade bottle was a silencer. Pursuant to 28 U.S.C. § 2254(e)(1), “the federal court must assume the state court’s factual determinations are correct unless there is clear and convincing evidence to the contrary.” Kelley v. Bohrer, 93 F.4th 749, 755 (4th Cir. 2024).

The state appellate court wrote that “[t]he Gatorade bottle, which the evidence indicated was used as a silencer for the gun used to shoot the victim, was wrapped in duct tape, with white medical tape underneath” during a discussion of the entirety of the evidence that remained in the absence of Carter’s testimony. J.A. 740. The State argues that this is a factual finding, and therefore the district court was required to accept this interpretation of the facts as binding. This is not so.

Habeas courts adopt the “best interpretation” of ambiguous legal or factual findings by state courts. Kelley, 93 F.4th at 756. Nothing in the court’s discussion indicates that the state appellate court—which was sitting in an appellate posture—was making a factual finding. Further, the materiality analysis turns on what the jury may have concluded if the CSM report had not been suppressed. It would be, arguably, legally improper for the state appellate court to have made factual findings on the evidence while engaging in an inquiry into whether a juror may have come to a different conclusion as to Martin’s guilt. Because federal courts are required to give state courts “the benefit of the doubt” and avoid finding legal errors or contradictions when unnecessary, id. at 757, we construe the state court’s statement as an evidentiary description and not a misapplication of law.

29

of which his wife was aware and unbothered. J.A. 2440–41. The tape on the bottle was like that from McFadden’s house, J.A. 741, and the gun like that seen in McFadden’s room, J.A. 2015–16. The Defense’s theory that McFadden attempted or solicited Torok’s murder after learning Torok was pregnant with her lover’s child is substantially stronger in the absence of a link between Martin and silencers. On these facts, any reasonable jurist would find the suppressed report to be material.

* * *

Carter’s testimony fundamentally altered the character of this case. It formed the entire basis for a devastating jury instruction, bolstered the State’s weaker witnesses, and undermined the Defense’s theories. Applying clearly established Supreme Court precedent to this record, no reasonable jurist could conclude that the suppression of the forensic computer report was immaterial. We therefore affirm the district court’s grant of Martin’s habeas petition.

V.

Turning now to the conditional order of release, the State challenges the district court’s provision of only sixty days for retrial or release. District courts have “broad discretion in conditioning a judgment granting habeas relief.” Hilton v. Braunskill, 481 U.S. 770, 775 (1987). Courts of appeals “review the district court’s formulation of an appropriate habeas corpus remedy for abuse of discretion.” Douglas v. Workman, 560 F.3d 1156, 1176 (10th Cir. 2009); see also Wolfe v. Clarke, 718 F.3d 277, 284–85 (4th Cir. 2013). The “award of an unconditional writ does not, in and of itself, preclude the

30

authorities from rearresting and retrying a successful habeas petitioner” after the deadline for release expires. Id. at 288–89.

The parties provide several cases to illustrate the spectrum of relief in similar habeas cases. The State cites to nearly a dozen cases 5 where courts provided their preferred form of relief: sixty days to decide to retry, or 120 days to complete retrial. See Opening Br. at 64– 65. These are countered by Martin’s similarly extensive number of cases, 6 in which courts provided the same form of relief selected by the district court here. See Resp. Br. at 61.

The State argues it is “self-evident that 60 days to retry a decade-old complex attempted murder case such as this is unreasonable,” Reply Br. 26. But as conceded at oral argument, the State has provided no case in which a district court’s sixty-day conditional

5

See, e.g., Herrera v. Collins, 506 U.S. 390, 403 (1993); Luna v. Cambra, 311 F.3d 928, 929 (9th Cir. 2002) (remanding “with instructions to issue the writ of habeas corpus, unless California elects, within 90 days of the issuance of the mandate, to retry [petitioner],” adding that “[a]ny such retrial shall commence within a reasonable time thereafter”); Norde v. Keane, 294 F.3d 401, 403 (2d Cir. 2002) (remanding “with instructions to issue the writ unless within sixty days the State elects to retry” the petitioner); Wilson v. Sec’y, Penn. Dept. of Corr., 782 F.3d 110, 117 (3d Cir. 2015) (180 days); D’Ambrosio v. Bagley, 656 F.3d 379, 385 (6th Cir. 2011) (180 days); Woods v. Dugger, 923 F.2d 1454, 1456 (11th Cir. 1991) (180 days to resentence petitioner); accord Wansing v. Hargett, 115 Fed. Appx. 27, 31 (10th Cir. 2004) (“Allowing 120 days for a retrial is a common practice.”).

6

See, e.g., Arroyo v. Jones, 685 F.2d 35, 41 (2d Cir. 1982) (affirming district court “decision requiring the State to release or retry [petitioner] for attempted murder within sixty days”); Browder v. Dir., Dep’t of Corr. of Ill., 434 U.S. 257, 260 (1978) (affirming 60-day conditional release order); Foxworth v. Maloney, 515 F.3d 1, 2 (1st Cir. 2008) (affirming 60-day conditional release order); Whitehead v. Wainwright, 609 F.2d 223, 224 (5th Cir. 1980) (affirming 60-day conditional release order but remanding due to unrelated legal error); Engel v. Buchan, 710 F.3d 698, 701 (7th Cir. 2013) (discussing 60-day conditional release order); see also Jones v. Cunningham, 313 F.2d 347, 353 (4th Cir. 1963) (reversing denial of habeas petition and directing district court to enter a 60-day conditional release order to be further stayed if the State chose to appeal to the Supreme Court).

31

release order was reversed as an abuse of discretion. The State has failed to show that sixty days is so unreasonable as to constitute an abuse of the district court’s broad discretion, especially in light of the many cases cited by Martin granting this exact form of relief.

Further, the State never moved for an extension in the district court, and this Court looks unfavorably upon the failure to do so. See Wolfe, 718 F.3d at 287. If it continues to maintain that sixty days is insufficient, the State will have the opportunity to seek an extension from the district court on remand.

Accordingly, we find that the conditional release order was not an abuse of the district court’s discretion.

VI.

For the foregoing reasons, the judgment is

AFFIRMED.

32

NIEMEYER, Circuit Judge, dissenting:

A jury in Anne Arundel County, Maryland, convicted Charles Martin of the attempted murder of his pregnant girlfriend, allegedly because she refused to have an abortion. The girlfriend, Jodi Torok, told Martin that she would have the baby and sue him for child support. Under the State’s theory, a confederate of Martin used Martin’s .380 caliber handgun, which had been fitted with a silencer fabricated by Martin from a Gatorade bottle, and shot Torok in the head at Martin’s instruction. The Gatorade bottle had been attached to the barrel of the gun through which the gun would fire, thus muffling its sound. Torok narrowly survived with her life.

The critical evidence at trial linking Martin to the assault was the Gatorade bottle, which was found in Torok’s apartment a few feet away from her as she lay unconscious on the floor. Its mouth was wrapped with tape pressed into the rectangular shape of a handgun’s barrel, and the hole in the bottom of the bottle projected outward, as would be created by an exiting bullet. The bottle contained DNA matching Martin’s and, according to eyewitness testimony, Martin participated in fabricating the silencer. The State asked the jury to find that Martin “was involved in the shooting of Jodi Torok because he helped make that silencer.” And the jury so found.

After Martin was convicted, the state court sentenced him to life imprisonment.

Martin later filed a petition in state court for postconviction relief, arguing, among other things, that the State committed a violation of Brady v. Maryland, 373 U.S. 83 (1963), by failing to disclose to him a police forensic computer analysis report, which indicated that the police had in their custody a laptop computer registered to “CSM” (College of

33

Southern Maryland), where Martin was formerly employed. The report indicated that the police searched the laptop for incriminating evidence but found nothing of “investigative value.” Martin claimed that this was important because the lack of relevant evidence on the computer contradicted the testimony of government witness Sheri Carter, who stated that she had seen Martin “looking up gun silencers” on a computer that he had gotten from a former employer and that Martin later took the computer from her apartment and told her that “he got rid of it.” The forensic report indicated that when an analyst searched for the term “silencer” and other key terms on the CSM laptop in police possession, “no data of investigative value” was found. But the report also showed that the laptop subject to the report had “last shut down” on May 18, 2005, more than three years before Torok was shot. Nonetheless, the State conceded that the report should have been disclosed to Martin and had not been. It argued, however, that while the report may have provided potential impeachment evidence, the evidence was not material because even if Carter’s testimony had been totally discredited, the other evidence of Martin’s culpability was overwhelming. The state trial court, nonetheless, agreed with Martin and granted his petition for postconviction relief.

On the State’s appeal, the Appellate Court of Maryland reviewed the postconviction court’s ruling in detail and concluded that the undisclosed evidence was not material, thus reinstating the judgment of conviction. State v. Martin, No. 3207, 2019 WL 4567473 (Md. Ct. Spec. App. Sept. 20, 2019). The court concluded that there was no reasonable probability that the result of the trial would have been different had the report been disclosed. First, it noted that the evidence indicated that the Gatorade bottle found near

34

Torok had been used as a silencer, rather than as a smoking device, as Martin had theorized. Id. at *8. Most significantly, the State introduced the bottle itself into evidence, allowing the jury to see how the taped circular mouth of the bottle had been molded into the rectangular shape of a gun barrel and that the sides of the small hole in the bottom of the bottle projected outward, indicating that the hole had been formed by something passing through the bottle. And there was also the evidence that the bottle was recovered in Torok’s apartment a few feet from where Torok was found lying, even though Torok’s roommate testified (1) that neither she nor Torok drank Gatorade or kept it in the house; (2) that she and Torok had recently cleaned the apartment and would never have left trash lying around; and (3) that the bottle had not been there when she left for work that morning.

The Appellate Court of Maryland also recognized that “the silencer/Gatorade bottle” had been “wrapped in duct tape, with white medical tape underneath.” Martin, 2019 WL 4567473, at *8. And on the white medical tape, investigators recovered a human hair fragment with a mitochondrial DNA profile that matched Martin’s profile. Thus, as the Appellate Court explained, Martin “was in the 0.06 percent of North Americans who could have left [the] hair” recovered from in between the two layers of tape that had been wrapped around the bottle’s mouth. Id. The court also pointed to other evidence linking Martin to the bottle that was provided by Michael Bradley’s testimony. Bradley stated that approximately two hours before the shooting, he saw his brother Frank go into the kitchen of their sister Maggie McFadden’s house while holding some white medical tape; that he then saw Frank and Martin go upstairs to McFadden’s bedroom; and that he then saw Frank run downstairs for a Gatorade bottle before returning to the bedroom, where Martin

35

remained. See id. at *9. In addition, the Appellate Court noted that the white medical tape on the bottle had the exact same characteristics as a roll of tape recovered from McFadden’s house, such that it could have come from that particular roll. Id. In addition, there was Bradley’s testimony that Martin and Jerry Burks left the house shortly before 2:00 p.m. and that when they returned at some point later that afternoon, Martin handed Frank Bradley a brown paper bag and told him to “get rid of” it. Id. That testimony, in turn, gave extra meaning to the State’s evidence that, at the time of the shooting, Martin owned a .380 caliber semi-automatic handgun, which was made by one of the manufacturers that could have made the .380 handgun that was used to shoot Torok, but that Martin’s handgun was never found. Id.

After reviewing the collective import of all of this evidence, the Maryland Appellate Court reasonably concluded that Martin had not established that there was “a reasonable probability that the result of his trial would have been different” had the forensic computer report been disclosed to him before trial. Martin, 2019 WL 4567473, at *9.

Despite the Appellate Court’s comprehensive decision analyzing the evidence in detail, the district court below disagreed with the state court’s conclusion and conducted its own factual analysis. Indeed, rather than extending to the state court’s decision the proper deference required by 28 U.S.C. § 2254, the district court misconstrued the state decision, reweighed the evidence in a manner that was contrary to the record and the state court’s reasonable assessment of it, and found that the state court had failed to conduct a proper Brady analysis, even though it acknowledged that the state court correctly set forth

36

the Brady principles. And even as the district court cited the § 2254 restrictions on its review of state proceedings, it nonetheless proceeded to violate those very restrictions.

The majority now affirms, pursuing the same erroneous course committed by the district court — conducting a de novo review of the evidence and making findings — rather than assessing whether the state court decision resulted from an “extreme malfunction” of the state judicial system such that all “fairminded jurists” would agree “that the state court’s decision conflicted” with Supreme Court precedents. Harrington v. Richter, 562 U.S. 86, 102 (2011) (cleaned up). While the majority opinion reasonably recognizes that the undisclosed forensic computer report could have been used to impeach Carter’s testimony — a fact that was also acknowledged by the state court — the majority goes on to make its own finding, contrary to the evidence, that Carter’s testimony “was by far the strongest” evidence inculpating Martin, disregarding the bottle. Supra at 24. It also states, again making its own finding, that “Martin’s DNA on the Gatorade bottle [was] consistent with entirely innocuous explanations, such as simply drinking or smoking from it,” which are inconsistent with the evidence. Id. Resting on its own limited analysis of a few facts, the majority affirms the district court’s grant of a new trial. And to do so, it had to overlook the many facts presented to the jury that the Maryland Appellate Court recognized as directly inculpating Martin, even without Carter’s testimony. The majority’s opinion simply does not recognize that the bottle was the best evidence linking Martin to the attempted murder and that the bottle had clearly been used as a silencer. One does not drink from or smoke with a bottle containing layers of tape. Moreover, the tape was multi- layered and shaped like the barrel of a .380 caliber handgun. The bottle also had a hole in

37

its bottom created by something like a bullet fired through the bottle. And the bottle had Martin’s DNA in between the layers of tape, directly indicating his involvement in the fabrication of the bottle as a silencer.

In short, the majority’s opinion fails to heed the limitations imposed on it by 28 U.S.C. § 2254.

Section 2254 provides a most narrow review by federal courts of state court criminal proceedings, well honoring the distinct sovereignty of the States by authorizing interference only when the state process clearly tramples well-established federal constitutional rights. It provides that a federal court “shall not” grant a writ of habeas corpus “unless” the earlier State-court decision (1) “was contrary to, or involved an unreasonable application of, clearly established Federal law, as determined by the Supreme Court of the United States” or (2) “was based on an unreasonable determination of the facts.” 28 U.S.C. § 2254(d). And the Supreme Court has made emphatically clear that this standard is to be understood as “difficult to meet.” Harrington, 562 U.S. at 102. For our system of dual sovereignty to retain its vitality, as the Supreme Court instructs, state courts must remain “the principal forum” in which “constitutional challenges to state convictions” are asserted and adjudicated. Id. at 103. Consequently, as the Harrington Court recognized, federal habeas review of state criminal convictions comes at a cost, as it “frustrates both the States’ sovereign power to punish offenders and their good-faith attempts to honor constitutional rights.” Id. (quoting Calderon v. Thompson, 523 U.S. 538, 555–56 (1998)). Indeed, as the Supreme Court has characterized it, a federal court’s grant of habeas relief to a person convicted in State court “intrudes on state sovereignty to a

38

degree matched by few exercises of federal judicial authority.” Id. (cleaned up). Thus, the Court has gone to great lengths in defining the limited circumstances for granting habeas pursuant to § 2254.

It has explained that federal habeas relief under § 2254 is extended to state prisoners only in those extraordinary cases where there has been an “extreme malfunction in the state criminal justice system.” Harrington, 562 U.S. at 102 (cleaned up). To this end, “§ 2254(d) stops short of imposing a complete bar on federal-court relitigation of claims already rejected in state proceedings,” but in lieu of a total ban, it limits federal courts’ authority to issue the writ to cases where all “fairminded jurists” would agree “that the state court’s decision conflicts with [the Supreme] Court’s precedents.” Id. Thus, “[a]s a condition for obtaining habeas corpus from a federal court, 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.” Id. at 103 (emphasis added); see also, e.g., Mays v. Hines, 592 U.S. 385, 391 (2021) (per curiam) (“The term ‘unreasonable’ [in § 2254(d)] refers not to ‘ordinary error’ or even to circumstances where the petitioner offers ‘a strong case for relief,’ but rather to ‘extreme malfunctions in the state criminal justice system.’ In other words, a federal court may intrude on a State’s ‘sovereign power to punish offenders’ only when a decision ‘was so lacking in justification beyond any possibility for fairminded disagreement’” (cleaned up)).

The majority opinion has simply not honored these restrictions, as it has conducted a reanalysis of some of the facts, ignored others, and never deferred to those reasonably

39

considered by the state court. While the Appellate Court of Maryland acknowledged that the undisclosed forensic report should have been disclosed to enable Martin to attempt to impeach Carter’s testimony, it concluded that the Brady violation was not material — that there was not a reasonable probability that the outcome would have been different even if Carter’s testimony were entirely excluded. This was a reasoned conclusion.

Moreover, for further context, it is far from clear that the jury would have taken the undisclosed report to prove that Carter was lying. Instead, it could have concluded that Martin had two different computers from the College of Southern Maryland, an older one that was in his home and a newer one that he had previously kept at Carter’s apartment. Or it could have concluded that the laptop that the police recovered was the one Carter referenced in her testimony, but that the fact that the last dates shown were from 2005 indicated that its more recent data had been lost, including its data from the period shortly before the shooting. Either way, had the report been disclosed, it might well have had no impact on the jury’s assessment of Carter’s credibility.

In short, the Appellate Court of Maryland reasonably concluded that Martin failed to establish that there was “a reasonable probability that the result of his trial would have been different” had the forensic computer report been disclosed. Martin, 2019 WL 4567473, at *9. And even if that issue were debatable, the state court’s holding still was not “so lacking in justification that [it] was an error . . . beyond any possibility for fairminded disagreement.” Harrington, 562 U.S. at 103. Yet, in a decision defying the established standards and bereft of judicial humility, the majority concludes that “no

40

reasonable jurist could conclude that the suppression of the forensic computer report was immaterial.” Supra at 29.

I do not agree with this assessment, and therefore I respectfully dissent.