Thomas Overton v. Secretary, Florida Department of Corrections

16-10654 Opinion of the Court 3 Thomas Mitchell Overton, who is incarcerated on death row in Florida, appeals the district court's denial of his 28 U.S.C. § 2254 petition for a writ of habeas corpus. This appeal concerns whether his petition was timely filed. It also concerns his claims that trial counsel was ineffective in failing to prepare for and partic- ipate adequately in a preliminary hearing and that the State vio- lated Brady v. Maryland¹by failing to disclose an investigator's his- tory of sloppy evidence collection practices. After a thorough re- view of the record and with the benefit of oral argument, we con- clude that Overton's petition was timely, but that it was properly denied in the alternative on the merits. We therefore affirm the judgment of the district court.²

I. BACKGROUND

In this section, we discuss the crimes of which Overton was convicted and the investigation of these crimes before turning to examine the relevant pretrial proceedings, the trial, and the post- conviction proceedings.

A. The Murders and Law Enforcement's Investigation

The crime scene that confronted investigators was re- counted by the Florida Supreme Court as follows (edited here for brevity):

3

4 Opinion of the Court 16-10654 On August 22, 1991, Susan Michelle MacIvor, age 29, and her husband, Michael MacIvor, age 30, were found murdered in their home in Tavernier Key. Su- san was eight months pregnant at the time with the couple's first child.

. . . .

Once law enforcement officers arrived, a thorough examination of the house was undertaken. In the liv- ing room, where Michael's body was found, investiga- tors noted that his entire head had been taped with masking tape, with the exception of his nose which was partially exposed. . . . The investigators surmised that a struggle had taken place because personal pa- pers were scattered on the floor near a desk, and the couch and coffee table had been moved. . . . Continuing the search toward the master bedroom, a piece of clothesline rope was found just outside the bedroom doorway. Susan's completely naked body was found on top of a white comforter. Her ankles were tied together with a belt, several layers of mask- ing tape and clothesline rope. Her wrists were also bound together with a belt. Two belts secured her bound wrists to her ankles. Around her neck was a garrote formed by using a necktie and a black sash, which was wrapped around her neck several times. Her hair was tangled in the knot. Noticing that a dresser drawer containing belts and neckties had been pulled open, officers believed that the items used to bind and strangle Susan came from inside the home.

. . . Also under the comforter was her night shirt; the

4

16-10654 Opinion of the Court 5 buttons had been torn off with such force that the but- ton shanks had been separated from the buttons themselves. Near the night shirt were her panties which had been cut along each side in the hip area with a sharp instrument.

. . . .

. . . The investigators used a luma light to uncover what presumptively appeared to be seminal stains on Susan's pubic area, her buttocks, and the inside of her thighs. The serologist later testified that he collected what appeared to be semen from Susan's body with swab applicators. Three presumptive seminal stains also appeared on the fitted sheet . . . . Ultimately, the officers took the comforter, fitted sheet, and mattress pad into evidence.

Overton v. State (Overton I), 801 So. 2d 877, 881-83 (Fla. 2001). Investigators could not immediately identify a suspect. A se- rologist, Dr. Donald Pope, "examined the bedding and made cut- tings in accordance with the markings he had made at the scene."

Id. at 883. "One of the stains from the fitted sheet and another stain from the mattress pad tested positive for sperm." Id. The cuttings were not sent to the Florida Department of Law Enforcement ("FDLE") immediately after Dr. Pope detected the presence of sperm cells because, at that time, FDLE had only "recently begun the process of DNA testing and [its] protocol did not allow for test- ing in cases where there was not a suspect." Id. at 883 n.2.

5

6 Opinion of the Court 16-10654

"[A] large-scale investigation" began, and "[o]ver the years following the murders, law enforcement agencies investigated sev- eral potential suspects," including Overton, "a known 'cat burglar,' whom police suspected" in another unsolved murder. Id. at 883-

84. In 1993, the cuttings from the bedding were sent to the FDLE lab. "Through a process known as restriction fragment length pol- ymorphism ('RFLP'), [forensic serologist Dr. James] Pollock was able to develop a DNA profile from two of the cuttings." Id. at 884. However, "[n]o match was made at that time." Id. In late 1996, five years after the murders, Overton was ar- rested during a burglary in progress. Id. While in custody on that arrest, Overton cut himself with a razor blade and the towel that he used to stop the bleeding was turned over to investigators. Id.

"Based on preliminary testing conducted on the blood from the towels, police obtained a court order to withdraw [Overton's] blood for testing." Id. Then, "Pollock was able to compare the pro- file extracted from the stains in the bedding to a profile developed after extracting DNA from Overton's blood. After comparing both profiles at six different loci, there was an exact match at each locus."

Id. (footnote omitted). In layman's terms, "the probability of find- ing an unrelated individual having the same profile was, conserva- tively, in excess of one in six billion Caucasians, African[-]Ameri- cans and Hispanics." Id.

The cuttings were later tested a second time:

In 1998, the cuttings from the bedding were submit- ted to yet another lab, the Bode Technology Group

6

16-10654 Opinion of the Court 7 ("Bode"). Dr. Robert Bever, the director at the Bode lab, testified as to the tests which were conducted on the bedding and the resulting conclusions. The Bode lab conducted a different DNA test, known as short tandem repeat testing ("STR"), from that performed by the FDLE. Overton's DNA and that extracted from a stain at the scene matched at all twelve loci. These results were confirmed by a second analyst and a computer comparison analysis. Asked to describe the significance of the Bode lab findings, Dr. Bever testified that the likelihood of finding another individ- ual whose DNA profile would match at twelve loci was 1 in 4 trillion Caucasians, 1 in 26 quadrillion Afri- can[-]Americans and 1 in 15 trillion Hispanics. Id. at 884-85.

Overton was charged with two counts of first degree mur- der as well as killing an unborn child, burglary, and sexual battery. He pleaded not guilty and proceeded to trial.

B. Relevant Pretrial Proceedings

In the weeks leading up to trial, Overton's counsel at- tempted to get the DNA evidence excluded. Trial counsel filed multiple motions for exclusion of the evidence and moved for a Frye hearing, described below, to require the State to establish the DNA evidence's admissibility.

In these motions, Overton's counsel advanced three theo- ries why the DNA evidence was inadmissible. First, they sought to exclude only the STR testing Bode performed, arguing that the

7

8 Opinion of the Court 16-10654 State had failed to provide discovery the defense needed to chal- lenge the testing. In the alternative, counsel asked that the court continue the trial and compel additional discovery on the STR test- ing. The court rejected the challenge to Bode's STR testing and re- fused to continue the trial.

Second, Overton's counsel took a broader approach, argu- ing for the exclusion of all inculpatory DNA evidence derived from the bedsheet cuttings. They contended that the State "taint[ed]"

the evidence when it impermissibly contacted the defense's expert and had the expert run additional tests on the evidence without the consent of Overton's counsel. Doc. 13-25:15.³Overton's counsel argued that this interference deprived the defense team of the use of the evidence from the cuttings because their intended expert was compromised. Thus, they asked the court to exclude any DNA ev- idence inculpating Overton to "level [the] playing field." Doc. 13- 65 at 35. The court denied the motion, finding no basis for relief because Overton's counsel had not established a confidential ex- pert relationship with the analyst.

Third, Overton's counsel argued that the DNA evidence was inadmissible because it was not scientifically reliable. They re- quested a pretrial hearing pursuant to Frye v. United States, 293 F. 1013 (D.C. Cir. 1923). Frye, "[a]s adopted by Florida's courts, . . . requires that the proponent of expert evidence establish by a pre- ponderance of the evidence the general acceptance of the

8

16-10654 Opinion of the Court 9 underlying scientific principles and methodology." Taylor v. Sec'y, Fla. Dep't of Corr., 64 F.4th 1264, 1270 (11th Cir. 2023) (alterations adopted) (internal quotation marks omitted). The court granted the request for a Frye hearing.

Overton's counsel had argued that the hearing was neces- sary to determine the reliability and therefore admissibility of the State's DNA evidence. But when the hearing began as scheduled, four days before jury selection, defense counsel told the court that they were not prepared and could not meaningfully challenge the information the State provided as to the STR testing performed by Bode. Defense counsel requested to continue the Frye hearing. The district court denied the request and proceeded with the hearing. At the Frye hearing, the State presented expert testimony to show that both RFLP and STR testing were generally accepted methods of DNA analysis and that the testing had been properly conducted in Overton's case. Three experts testified for the State:

Pollock; Bever; and Dr. Martin Tracey, a genetics professor. Over- ton's counsel did not meaningfully question the State's experts and did not present any witnesses of their own.

As to the RFLP testing, the State posited that the technique was admissible under Florida caselaw recognizing it as generally accepted if performed correctly. The State argued that STR testing was not "new" or "unrelated" to RFLP testing; it was an "improve- ment on the basic science already followed and approved by the Florida courts," Doc. 13-70 at 28-29, a "faster, quicker" version of RFLP testing that allowed the testing of smaller amounts of DNA,

9

10 Opinion of the Court 16-10654 Doc. 13-67 at 14. Because Florida courts deemed RFLP testing ad- missible under Frye, the State argued, STR testing should be admis- sible as well.

To support this argument, the State's experts testified to the similarities between the already accepted RFLP testing and STR testing. Bever and Tracey testified that the scientific principles un- derpinning both methodologies were the same or "essentially iden- tical." Doc. 13-70 at 12. Tracey observed that "most of the public laboratories" in the United States were beginning to use STR test- ing, and it was already the "method of choice" in Europe. Id. at 23,

25. The experts agreed that STR testing was generally accepted in the scientific community as an "extremely reliable" method of DNA analysis. Doc. 13-68 at 38.

Bever and Pollock also testified to the DNA testing they had performed on the bedding cuttings taken from the crime scene, the statistical methods they used to interpret their results, and the qual- ity controls in place in their laboratories. Tracey verified the other scientists' results, testifying that the match between Overton's DNA and the cuttings, calculated by reliable methods, was, in his opinion, a certainty.

The court admitted the RFLP and STR DNA testing results, ruling that both modes of analysis were generally accepted in the scientific community.

After the Frye hearing, Overton's counsel continued to con- test the admissibility of the DNA evidence. Maintaining their the- ory that the State had failed to provide adequate discovery on

10

16-10654 Opinion of the Court 11 Bode's STR testing, defense counsel twice renewed their motion to exclude the STR testing before the trial began. Both motions were denied.

C. The Trial and Sentencing

At trial, the State presented the DNA evidence to the jury. The State also presented the testimony of two jailhouse inform- ants. See Overton I, 801 So. 2d at 885. The informants testified that while in jail after his arrest Overton had confessed to the murders and disclosed details of the crimes that only the perpetrator would know. Id. at 885-86.

"The primary thrust of the defense in the case was centered upon a theme that law enforcement officers, Detective Visco in particular, had planted Overton's semen in the bedding, which was essential to the prosecution." Id. at 887. "The defense theorized that [] Visco obtained [Overton's] sperm from [his] one-time girl- friend, Lorna Swaybe, transported the sample in a condom, and placed it on the bedding." Id. Allegedly, Visco had motive to frame Overton because Overton had filed an internal affairs complaint against him. See id. at 887 n.7. Visco denied this. See id. at 887 n.8.

"In an attempt to substantiate this fabrication of evidence theory,"

the defense tested the bedding samples for a compound contained in spermicidal condoms called Nonoxynol-9. Id. at 887. The bed- ding tested positive for Nonoxynol-9. But the analyst who con- ducted the testing admitted at trial that the compound also was

"commonly used in household detergents" and that the amount

11

12 Opinion of the Court 16-10654 found on the bedding was much smaller than what a spermicidal condom contained. Id. at 887-88.

To further attack the DNA evidence's validity, the defense argued that the chain of custody was broken, and the evidence therefore was compromised. To support this argument, Overton's counsel cross-examined Pope, the Monroe County Sheriff's Office ("MCSO") serologist present at the crime scene, at length about his evidence handling practices and gaps in the chain of custody. As the Florida Supreme Court summarized, the defense's cross-examina- tion identified the following issues with Pope's collection and stor- age of crime scene DNA:

(1) envelopes that were used to store DNA evidence were misdated; (2) there were no property receipts to account for the swabs that were used to obtain fluids from Susan's body at the scene; (3) the swabs were transported to [Pope's] home, which was not a certi- fied storage facility; (4) these swabs were placed in [Pope's] home refrigerator; (5) the first property re- ceipt for the envelopes of clippings, which provided a match to Overton's DNA, was dated June 10, 1994 [al- most three years after the crime scene was processed];

(6) the bedding (quilt, mattress pad, comforter, and bed sheet) on which semen stains were found were placed in paper bags and transported to [Pope's] home to be air dried; (7) the bedding was transported to the Key West property evidence storage room on August 26, 1991 [four days after the crime scene was processed]; and (8) [Pope] transported the mattress

12

16-10654 Opinion of the Court 13 pad in a paper bag by car to Orlando to have a psychic conduct an inspection.

Overton v. State (Overton II), 976 So. 2d 536, 551 (Fla. 2007). Overton's counsel also vigorously cross-examined Detective Robert Petrick, the MCSO crime scene investigator assigned to the MacIvor murders. As the Florida Supreme Court summarized, de- fense counsel's cross-examination identified the following issues with Petrick's collection of the evidence from which DNA was eventually extracted and tested:

(1) the paper bags in which [Petrick] collected evi- dence did not resemble the particular paper bag [in- troduced at trial] that allegedly had his signature on it; (2) this alleged signature on the paper bag, which read "Detective R. Petrick," was not [Petrick's] signa- ture; and (3) the property receipts with regard to the [bedding] clippings in envelopes had writing on them that was not [Petrick's] writing.

Id.

Overton's counsel argued at trial that Pope's and Petrick's testimony, along with that of other detectives and evidence custo- dians, suggested "probable tampering," Doc. 13-129 at 27, and that the chain of custody between the collection of DNA evidence and the first DNA testing done on that evidence—Pollock's RFLP test- ing on the bedsheet cuttings—was broken. They therefore ob- jected to the introduction of the cuttings of the victims' bedding, attempting to exclude all testimony about the DNA testing done on the cuttings. This attempt was unsuccessful. After the State

13

14 Opinion of the Court 16-10654 pointed to testimony by Pope, Pollock, and other evidence custo- dians supporting that the chain of custody remained intact, the court overruled the objection and admitted the cuttings into evi- dence, allowing the State's experts to testify about the DNA evi- dence inculpating Overton.

The jury found Overton guilty on all charges. Overton I,

801 So. 2d at 888. After a sentencing hearing, the jury recom- mended a death sentence for Susan's murder by a vote of 9 to 3 and for Michael's murder by a vote of 8 to 4. Id. at 888-89. The trial court imposed a death sentence. Id. at 889. The Florida Supreme Court affirmed Overton's convictions and sentences on direct ap- peal, see id. at 881, and the Supreme Court of the United States de- nied certiorari, see Overton v. Florida, 535 U.S. 1062 (2002).

D. Postconviction Proceedings

After his unsuccessful direct appeal, Overton filed a state postconviction motion under Florida Rule of Criminal Procedure

3.851. The timing and procession of Overton's postconviction liti- gation is relevant to the timeliness of his federal habeas petition. Rather than recounting that timing here, we do it below in our timeliness analysis. See infra Part III.A.

As relevant to this appeal, Overton raised an ineffective as- sistance of trial counsel claim and a Brady claim. Overton con- tended that his trial counsel was ineffective in failing to prepare for and challenge the State's evidence at the Frye hearing. And he con- tended that the State violated Brady when it failed to disclose that Pope had engaged in sloppy evidence collection practices in other

14

16-10654 Opinion of the Court 15 cases, most notably Allen v. State, 854 So. 2d 1255 (Fla. 2003). In Al- len, the FDLE rejected DNA evidence that Pope collected from the crime scene because he submitted the evidence in an incorrectly labeled envelope, contravening standard procedures. The postconviction trial court held an evidentiary hearing on the ineffective assistance of counsel claim. Overton's postcon- viction counsel called his trial attorneys, Manuel Garcia and Jason Smith, to testify, attempting to show that trial counsel was ineffec- tive, in part, for failing to participate in the Frye hearing. Both attorneys testified that they made a tactical decision not to present or question witnesses at the Frye hearing. Smith, as the lead defense attorney, made the ultimate decision not to partic- ipate. Smith testified that even successful participation in the Frye hearing would have served a "limited purpose" in Overton's de- fense, for three reasons. Doc. 13-276 at 2.

First, trial counsel lacked the necessary discovery to chal- lenge the State's DNA evidence. The defense retained a DNA ex- pert, Dr. Gary Litman, who, according to Smith, advised that he needed more information on Bode's procedures to be able to aid the defense in challenging the State's STR DNA testing. The ave- nues of attack against the State's STR testing were limited. Smith testified that Litman believed the science behind STR testing was sound. So the only way to challenge the evidence was through the methodology that Bode used. But, according to Smith, the defense was unable to obtain the discovery necessary to mount this chal- lenge before the Frye hearing. The court had denied the

15

16 Opinion of the Court 16-10654 continuances that trial counsel requested to facilitate further dis- covery. At the "last[] minute" before the hearing and trial, the Court permitted the defense team to spend a week at Bode's labor- atory to examine his methods, but Smith testified that because this would not have been a good use of counsel's time with trial ap- proaching, he declined the court's offer. Doc. 13-292:22. Second, trial counsel believed that inculpatory DNA evi- dence would have been admitted regardless of the defense's partic- ipation in the hearing. According to Smith, Litman advised that the RFLP DNA testing done by the FDLE would have been admitted regardless because RFLP had previously passed scientific muster under Frye. And the RFLP testing, like the STR testing, would have linked Overton to the crime scene.

Third, Smith agreed on cross-examination that challenging the DNA evidence on science and methodology did not comport with the defense's main theory, which was that Overton's DNA had been planted. Under this theory, it was unnecessary to dispute that the STR and RFLP tests accurately matched Overton's known DNA to the crime scene samples.

For these reasons, Smith believed that "the best thing [he] could do" for Overton's defense was to preserve the discovery issue for appeal. Doc. 13-276 at 3. To preserve the issue for appeal, not participating in the hearing was a better strategy than participating and doing an "inadequate job." Doc. 13-275 at 25-26. Declining the visit to Bode also was a strategic decision. Smith stated that he "felt [he] could make an issue o[n] appeal" out of the fact that the

16

16-10654 Opinion of the Court 17 defense's only option to obtain necessary discovery was a time-con- suming trip on the eve of trial. Doc. 13-292 at 23. Smith also testi- fied that he made a conscious choice not to raise chain-of-custody issues at the Frye hearing, believing they were better addressed at trial, during which the State would have to prove a clear chain of custody to admit the DNA evidence.

After the evidentiary hearing, the trial court denied relief. The Florida Supreme Court affirmed. Overton II, 976 So. 2d at 575. The Court concluded that Overton's ineffective assistance of counsel claim failed because he had shown neither deficient per- formance nor prejudice. Id. at 549-53. First, the Court concluded that "the limited participation of counsel during the Frye hearing did not constitute deficient performance because it was a strategic decision made by counsel." Id. at 550. The Court recounted that counsel "was of the view that they would not participate due to the lack of discovery with regard to the procedures and protocols that the Bode Lab used in testing," as well as the court's denial of their requested "continuance to provide more time to prepare" for the hearing. Id. "The fact that counsel may not have been prepared to fully participate during the Frye hearing does not establish they were not equipped to make a strategic decision with regard to whether they should participate to a greater extent." Id. And the decision was indeed strategic: counsel "testified that the defense made a strategic decision not to participate further to properly pre- serve the issue of the lack of discovery with regard to the Bode Lab, which could then be attacked on direct appeal." Id. "Consistent

17

18 Opinion of the Court 16-10654 with this strategy, appellate counsel argued the discovery issue on direct appeal, but this Court found the argument to be without merit." Id.

The Florida Supreme Court further explained that "[i]n making the strategic decision, Overton's trial counsel understood that even if they were able to prevent the STR DNA testing by the Bode Lab from being admitted into evidence, the RFLP DNA test- ing by the FDLE Lab would still be admitted and would similarly link Overton to the crime." Id. at 550-51. Before the Frye hearing, the trial court had "acknowledged that case law established that RFLP DNA testing results would be admitted here and the Frye hearing was unnecessary on that DNA matter." Id. at 551. Plus, de- fense expert Litman had advised trial counsel that the RFLP DNA evidence "should be admitted in this case." Id. So, trial counsel "re- quested the Frye hearing to challenge only the newer STR technol- ogy." Id.

"Moreover," the Court continued, "despite the decision to not participate further during the Frye hearing, other attempts were made by Overton's counsel to exclude these DNA testing results."

Id. Counsel requested twice that the evidence be excluded. Id. And they continued to assert the chain-of-custody issues: "[a]n alleged broken chain of custody was significant to the defense to support the defense theory that law enforcement had the opportunity to plant Overton's DNA," and so "it was reasonable for Overton's counsel to believe that an alleged broken chain of custody did not need to be addressed during the Frye hearing, but rather, should be

18

16-10654 Opinion of the Court 19 addressed during trial." Id. And that is precisely what trial counsel did, "thoroughly cross-examin[ing] Dr. Pope and Detective Petrick, both of whom worked for law enforcement agencies and gathered evidence from the crime scene, on the alleged broken chain of custody." Id.

The Court also concluded "that the decision by Overton's counsel to not address a potential degradation of the DNA evi- dence during the Frye hearing on the basis of an alleged broken chain of custody was reasonable." Id. This was because (1) the fact that Overton's DNA was on the evidence "would be consistent with his theory that his DNA had been planted there," and (2) Lit- man "had dismissed the dangers of degradation and false positives from an alleged broken chain of custody." Id. at 552. Alternatively, the Florida Supreme Court concluded that Overton suffered no prejudice from trial counsel's limited partici- pation at the Frye hearing. Id. "First, the chain of custody was in- tact." Id. "Second, even if the chain of custody was broken, there was not sufficient evidence to establish a probability of tampering, which would support exclusion of the evidence" under Florida law. Id. "Third, [the Florida Supreme Court] conclude[d] that the STR DNA testing completed at the Bode Lab [met] the requirements of the Frye test." Id. at 553. "[T]his indicates that if Overton's counsel had attempted to challenge the STR DNA testing here, it [was] highly unlikely that the evidence would have been excluded." Id. In addition, although counsel did not challenge the RFLP testing, there was no prejudice "because the RFLP results were clearly

19

20 Opinion of the Court 16-10654 admissible and the results from this testing also matched Overton."

Id. at 553 n.14.

As to the Brady claim, the Florida Supreme Court first opined that "the alleged evidence with regard to Pope's perfor- mance in Allen is of minimal value," citing the facts that "Overton has not identified whether this alleged similar 'sloppy' work oc- curred before or after Pope's DNA work in the instant case" and that the "evidence reflects only that which occurred in another case, rather than providing evidence of that which occurred in"

Overton's case. Id. at 563 (citation omitted). Second, the Court ex- plained that trial counsel had presented "significant" challenges to Pope during trial:

Pope was impeached with evidence of his conduct in the instant case. Along with other forms of impeach- ment, Overton's counsel elicited evidence from Pope that he transported pieces of evidence to his home and placed evidence in his household refrigerator, which is not certified as a storage facility or lab. This evidence did impeach Pope, and the alleged evidence of similar "sloppy work" in another case would be cu- mulative.

Id. Thus, the Court concluded that Overton failed to show preju- dice, rendering his Brady claim meritless Id. Overton then filed a federal habeas petition. The district court concluded that his federal petition was untimely under

28 U.S.C. § 2244. Overton v. Jones (Overton III), 155 F. Supp. 3d 1253, 1267-70 (S.D. Fla. 2016). The court alternatively concluded that

20

16-10654 Opinion of the Court 21 Overton's claims failed on their merits. Id. at 1270, 1310. The court determined that the Florida Supreme Court reasonably rejected Overton's ineffective assistance of counsel and Brady claims. Id. at 1277-85, 1288-91. The district court declined to issue Overton a certificate of appealability, id. at 1310, but this Court issued him a certificate on the claims we discuss in this opinion.

II. STANDARDS OF REVIEW

We review de novo a district court's determination that a fed- eral habeas petition is time-barred. Hall v. Sec'y, Dep't of Corr.,

921 F.3d 983, 986 (11th Cir. 2019).

"When reviewing a district court's grant or denial of habeas relief, we review questions of law and mixed questions of law and fact de novo, and findings of fact for clear error." Reaves v. Sec'y, Fla. Dep't of Corr., 717 F.3d 886, 899 (11th Cir. 2013) (internal quotation marks omitted). An ineffective assistance of counsel claim "pre- sents a mixed question of law and fact that we review de novo." Pope

v. Sec'y, Fla. Dep't of Corr., 752 F.3d 1254, 1261 (11th Cir. 2014). The Antiterrorism and Effective Death Penalty Act of 1996 ("AEDPA") governs our review of federal habeas petitions decided after the Act's passage. "AEDPA prescribes a highly deferential framework for evaluating issues previously decided in state court." Sears v. Warden GDCP, 73 F.4th 1269, 1279 (11th Cir. 2023). AEDPA bars federal courts from granting habeas relief to a peti- tioner on a claim that was "adjudicated on the merits in [s]tate court" unless the decision (1) "was contrary to, or involved an un- reasonable application of, clearly established Federal law, as

21

22 Opinion of the Court 16-10654 determined by the Supreme Court of the United States[;]" or

(2) "was based on an unreasonable determination of the facts in light of the evidence presented in the [s]tate court proceeding." 28 U.S.C. § 2254(d).

A state-court decision is "contrary to" clearly established law if the court "applie[d] a rule that contradicts the governing law" set forth by the United States Supreme Court or confronted facts that were "materially indistinguishable" from Supreme Court prece- dent but arrived at a different result. Williams v. Taylor, 529 U.S. 362, 405-06 (2000). To meet the "unreasonable application" stand- ard, a petitioner "must show far more than that the state court's decision was merely wrong or even clear error." Shinn v. Kayer,

592 U.S. 111, 118 (2020) (internal quotation marks omitted). In- stead, the decision must be "so obviously wrong that its error lies beyond any possibility for fairminded disagreement." Id. (internal quotation marks omitted). AEDPA's standard is "difficult to meet"

and "demands that state-court decisions be given the benefit of the doubt." Raulerson v. Warden, 928 F.3d 987, 996 (11th Cir. 2019) (in- ternal quotation marks omitted).

A federal habeas court must defer to a state court's determi- nation of the facts unless the state court's decision "was based on an unreasonable determination of the facts in light of the evidence presented in the [s]tate court proceeding." 28 U.S.C. § 2254(d)(2). We must afford state courts "substantial deference" un- der § 2254(d)(2) and "may not characterize . . . state-court factual determinations as unreasonable merely because we would have

22

16-10654 Opinion of the Court 23 reached a different conclusion in the first instance." Brumfield v. Cain, 576 U.S. 305, 313-14 (2015) (alteration adopted) (internal quotation marks omitted). We presume a state court's factual de- terminations are correct absent clear and convincing evidence to the contrary. See Pye v. Warden, Ga. Diagnostic Prison, 50 F.4th 1025, 1035 (11th Cir. 2022) (en banc).

On each claimed basis for relief, we review "the last state- court adjudication on the merits." Greene v. Fisher, 565 U.S. 34, 40 (2011). Here, that decision is the decision of the Florida Supreme Court in Overton v. State, 976 So. 2d 536 (Fla. 2007).

III. DISCUSSION

We divide our discussion into three parts. First, we address the timeliness of Overton's federal habeas petition. Second, we ad- dress Overton's claim that his trial counsel was ineffective for fail- ing to prepare adequately for and present a challenge to the State's DNA evidence at Frye hearing. Third, we address Overton's claim that the State violated Brady by failing to disclose Pope's sloppy ev- idence handling in other cases.

A. Overton's Federal Habeas Petition Was Timely Filed.

The State contends, and the district court agreed, that Over- ton's federal habeas petition was untimely because he filed it after AEDPA's one-year limitations period expired. Overton's state court judgment of conviction became final in May 2002, when the United States Supreme Court denied his petition for a writ of certiorari after the Florida Supreme Court denied relief in his direct appeal. Yet his federal habeas petition was not filed until October 2013.

23

24 Opinion of the Court 16-10654 Although the pendency of a properly filed state postconviction ap- plication tolls the federal limitations period, the State argues that Overton's facially inadequate Rule 3.851 motions, dismissed seria- tim by the state habeas court, did not meet this standard. There- fore, the State argues, they did not toll the period. For the reasons we explain below, we disagree.

We begin by reviewing the chronology of Overton's state postconviction proceedings.

Overton's counsel filed a state postconviction motion under Florida Rule of Criminal Procedure 3.851 on April 30, 2003, less than one year after the Supreme Court of the United States denied his petition for a writ of certiorari on May 13, 2002. The State moved to strike the motion, arguing that the motion lacked the re- quired factual basis for its claims. The postconviction trial court, finding Overton's motion "legally insufficient" under Rule 3.851(e), granted the State's motion to strike. Doc. 13-194 at 13. The court ordered Overton to "file an amended motion on or before July 11, 2003." Id. at 17.

On July 10, 2003, Overton filed an amended Rule 3.851 mo- tion. The trial court struck this motion, too—this time because it

"was not signed by [Overton] under oath" as Rule 3.851 requires. Doc. 13-198 at 20. The court noted that Overton had refused to sign the motion because of unresolved issues with his counsel over whether to assert a claim of ineffective assistance of trial counsel based on their failure to offer any mitigating evidence at the penalty phase. These issues, the court noted, "were apparently being

24

16-10654 Opinion of the Court 25 resolved with [c]ounsel." Id. The court ordered Overton to file an amended, conforming Rule 3.851 motion by October 31, 2003. On October 30, 2003, Overton filed a second amended Rule

3.851 motion in response to the court's order. This one conformed to the procedural requirements of Rule 3.851. The court held a hearing on and, ultimately, denied this motion. Overton's state postconviction proceedings continued until October 31, 2013, when the Florida Supreme Court denied rehearing on his remain- ing claims. See Overton v. State, 129 So. 3d 1069 (Fla. 2013) (un- published).⁴He filed his federal habeas petition eight days later, on November 8, 2013.

AEDPA permits a state prisoner to file a petition for a fed- eral writ of habeas corpus once he exhausts all available state court remedies. 28 U.S.C. § 2254(b)(1)(A). It imposes a one-year limita- tions period, which begins to run on "the date on which the [state court] judgment became final by the conclusion of direct review."

Id. § 2244(d)(1)(A). But "[t]he time during which a properly filed application for State post-conviction or other collateral review with respect to the pertinent judgment or claim is pending shall not be counted toward any period of limitation" under AEDPA. Id. § 2244(d)(2). "So the application must be 'properly filed' and

25

26 Opinion of the Court 16-10654 'pending.'" Bates v. Sec'y, Dep't of Corr., 964 F.3d 1326, 1328 (11th Cir.

2020).⁵

Whether the running of AEDPA's limitations period was tolled between the filing of Overton's original, deficient Rule 3.851 motion in April 2003 and his compliant motion in October 2003 thus depends on whether his motion was properly filed and pend- ing during this period. Our precedent answers this question. "[A] compliant [Florida postconviction] motion relates back to the date of filing of a noncompliant motion, such that the compliant mo- tion was 'properly filed' and 'pending' as of that date for purposes of tolling the limitations period in section 2244 of Title 28." Id.; see Hall, 921 F.3d at 990 ("[F]or the purposes of tolling under 28 U.S.C.

26

16-10654 Opinion of the Court 27 § 2244(d)(2), a petitioner's [Florida postconviction] motion is 'pend- ing' until it is denied with prejudice.").

And so, however "intuitive" it may be "that when a motion is stricken from the record, the motion is no longer pending. . . [,] this approach upends the procedure Florida courts have developed for processing facially deficient postconviction motions." Hall,

921 F.3d at 989 (alterations adopted) (internal quotation marks omitted). This procedure, laid out by the Florida Supreme Court in Spera v. State, 971 So. 2d 754, 761 (Fla. 2007), establishes that "a trial court abuses its discretion when it fails to provide a . . . petitioner at least one opportunity to amend his facially insufficient" Rule

3.851 motion. Hall, 921 F.3d at 989. In this case, the district court erred in concluding that Over- ton's federal habeas petition was untimely under AEDPA. The judgment in his case became final on May 13, 2002, when the Su- preme Court of the United States denied his petition for a writ of certiorari from the denial of his direct appeal. See Overton v. Florida,

535 U.S. 1062 (2002). That started the clock on AEDPA's one-year statute of limitations, giving Overton until May 13, 2003 to file a federal habeas petition or a state postconviction motion that would toll the federal limitations period. 28 U.S.C. § 2244(d). He filed a first Rule 3.851 motion on April 30, 2003, 13 days before the federal statute of limitations expired. Although that motion did not com- ply with the rule—and the following one did not, either—the state postconviction court twice gave Overton the opportunity to file a compliant motion. He timely filed one on October 30, 2003. Under

27

28 Opinion of the Court 16-10654 Bates, this compliant motion related back to the April 30 motion, such that it was both "properly filed" and "pending" from April 30, 2003 until October 31, 2013. See Bates, 964 F.3d at 1328. At that point, Overton still had 13 days remaining to file a federal habeas petition. He filed eight days later, on November 8, 2013. That peti- tion was timely filed under 28 U.S.C. § 2244(d).⁶We thus reject the State's argument that Overton's federal habeas petition was filed too late.

We now turn to the merits of the substantive claims for which we granted Overton a certificate of appealability.

B. Overton's Ineffective Assistance of Counsel Claim Does Not Withstand AEDPA Deference.

The Sixth Amendment entitles criminal defendants to the

"effective assistance of counsel"—that is, representation that does not fall "below an objective standard of reasonableness" relative to

"prevailing professional norms." Strickland v. Washington, 466 U.S. 668, 686-88 (1984) (internal quotation marks omitted). "To prevail on a claim of ineffective assistance of counsel, [Overton] must prove that his counsel's performance was objectively deficient and that this deficient performance prejudiced him." King v. Warden, Ga. Diagnostic Prison, 69 F.4th 856, 873 (11th Cir. 2023) (citing Strickland, 466 U.S. at 687). The Florida Supreme Court denied Overton's ineffective assistance of counsel claim on the merits, "so

28

16-10654 Opinion of the Court 29 we must defer to the state court's decision . . . unless it was not only erroneous, but objectively unreasonable." Id. (internal quota- tion marks omitted).

As to the first Strickland prong—whether trial counsel per- formed deficiently—the Florida Supreme Court concluded that trial counsel made a reasonable strategic decision not to participate in the Frye hearing based on several known factors, including that the RFLP DNA evidence was admissible notwithstanding any chal- lenge to the admissibility of the STR DNA evidence (a fact that both defense expert Litman and the trial court told trial counsel), the trial court's denial of continuances and the lack of discovery from the STR DNA lab could provide an issue for appeal, the de- fense was pushing a planted-evidence theory accepting that Over- ton's DNA was on the crime scene bedding, and trial counsel could attempt to undermine the STR DNA evidence at trial by cross-ex- amining Pope about the chain-of-custody problems. Overton chal- lenges this conclusion and argues, primarily, that trial counsel un- reasonably failed to challenge the DNA evidence at the Frye hearing based on Pope's evidence collection and storage practices—that is, the alleged break or breaks in the chain of custody of the evidence. But we need not address this conclusion, because even if Overton could establish that the Florida Supreme Court's determi- nation that his trial counsel rendered adequate performance was unreasonable, he cannot make the same showing as to the other prong of his ineffective assistance of counsel claim—that he suf- fered prejudice as a result. See Carey v. Dep't of Corr., 57 F.4th 985,

29

30 Opinion of the Court 16-10654 989 (11th Cir. 2023) ("Because a petitioner must prove both defi- cient performance and prejudice, a court need not address one ele- ment if it determines that the petitioner has failed to prove the other.").

To establish that his trial counsel's deficient performance caused him prejudice, Overton "must show that there is a reason- able probability that, but for counsel's unprofessional errors, the result of the proceeding would have been different." Strickland,

466 U.S. at 694. "A reasonable probability is a probability sufficient to undermine confidence in the outcome." Id. And, under AEDPA,

"the question is not whether a federal court believes the state court's determination [that there was no prejudice] under the Strickland standard was incorrect but whether that determination was unreasonable—a substantially higher threshold." Nejad v. Att'y Gen., State of Ga., 830 F.3d 1280, 1290 (11th Cir. 2016) (alteration adopted) (quoting Knowles v. Mirzayance, 556 U.S. 111, 123 (2009)). The Florida Supreme Court concluded that Overton was not prejudiced by trial counsel's failure to participate in the Frye hearing for several reasons, including that the chain of custody re- mained unbroken, there was insufficient evidence of tampering with the evidence, the STR DNA testing met the Frye require- ments, and the RFLP DNA test results were admissible and matched Overton's DNA. Overton challenges the Florida Supreme Court's prejudice determination with arguments about how trial counsel's purportedly deficient performance in failing to challenge the admissibility of the STR and RFLP evidence at the Frye hearing

30

16-10654 Opinion of the Court 31 prejudiced him. If the DNA evidence had been properly excluded, Overton says, the State would have been left with only the testi- mony of jailhouse informants in their case against him, leading to a "strong likelihood" the jury would have acquitted him or voted against death. Reply Br. 33. His arguments fall short of establishing that "no fairminded jurist could agree with [the Florida Supreme Court's] determination" that he suffered no prejudice from trial counsel's failure to participate in the Frye hearing. King, 69 F.4th at 867 (internal quotation marks omitted). Therefore, the district court did not err in denying federal habeas relief on Overton's in- effective assistance of counsel claim.

As a preliminary matter, although Overton argues that he was prejudiced because the STR and RFLP DNA results were un- reliable and likely would have been excluded had trial counsel chal- lenged their admission at the Frye hearing, his only evidence to sug- gest that the DNA testing should have been excluded is of Pope's sloppy evidence handling practices. Overton presented no evi- dence in the state postconviction proceedings that the DNA evi- dence had been tampered with, no evidence that the RFLP or STR DNA evidence was inadmissible for any other reason under Frye, and no evidence that the specimen from which the DNA had been sourced had meaningfully degraded.

Instead, Overton points to flaws in the collection and han- dling of the DNA evidence that he argues fell below "the standards of scientific reliability required by Frye." Reply Br. 23. These flaws, in Overton's view, are sufficient to demonstrate prejudice because

31

32 Opinion of the Court 16-10654 they would have rendered the DNA evidence inadmissible if ade- quately presented to the trial court. To bolster his argument, he points to a Florida appellate court decision recognizing that DNA evidence "might be excluded in a specific case due to defects" in the collection of evidence. Brim v. State, 779 So. 2d 427, 438 (Fla. Dist. Ct. App. 2000) (emphasis added).

Even accepting that Florida courts have in some cases ex- cluded DNA evidence because of tampering during the evidence's collection, see Murray v. State, 838 So. 2d 1073, 1082-83 (Fla. 2002), the Florida Supreme Court decided that the evidence was admissi- ble after considering the flaws in its collection and handling. Begin- ning by addressing the chain of custody issue, the Florida Supreme Court concluded the DNA evidence's chain of custody was intact, but even if it was broken, "there was not sufficient evidence to es- tablish a probability of tampering, which would support exclusion of the evidence." Overton II, 976 So. 2d at 552. And the Court ex- plained that it "has not held that a broken chain of custody alone is enough by itself to establish probable tampering." Id. Put differ- ently, the Florida Supreme Court concluded that Overton pointed to no evidence of tampering sufficient to establish that the defects in the chain of custody made the DNA evidence excludable. It de- termined that "the record does not support the contention that Overton's counsel could have established a probability of tamper- ing, which[,]" the Court explained, "would have arguably led to an exclusion of both the STR DNA testing and the RFLP DNA testing results, had evidence been introduced during the Frye hearing with regard to the alleged broken chain of custody." Id. at 553. Thus,

32

16-10654 Opinion of the Court 33 even though DNA evidence may be inadmissible under Florida law due to defects in the evidence's chain of custody where there is ev- idence of tampering, the Florida Supreme Court determined this is not one of those cases.

Still, Overton argues that this conclusion was unreasonable because at the Frye hearing stage, "trial counsel was not required to establish a probability of tampering." Appellant's Br. 86. Instead, he contends that under Frye, "even if the general scientific principles and techniques [were] accepted by experts in the field, the same experts could testify that the work done in a particular case was so flawed" to make the evidence inadmissible. Id. (alteration adopted) (quoting Murray, 838 So. 2d at 1078)). And in Overton's view, if trial counsel had challenged the STR DNA evidence at the Frye hearing on this basis, it likely would have been excluded. The problem for Overton, however, is that even if counsel had successfully challenged the STR DNA evidence at the Frye hearing, the State still would have been able to offer at trial the RFLP DNA evidence connecting Overton to the scene of the crime. Overton does not now, nor has he ever, explained why the RFLP DNA in this case was inadmissible aside from the evidence collec- tion problem. And as to his argument that the evidence collection methods rendered the RFLP DNA evidence inadmissible as well, he must overcome AEDPA deference as to the Florida Supreme Court's determination that the RFLP DNA evidence here was

"clearly admissible" because it is generally accepted, testimony at the evidentiary hearing "illustrate[d] the proper procedures and

33

34 Opinion of the Court 16-10654 protocols that existed at the FDLE Lab with regard to the RFLP testing," and Pollock testified that the FDLE Lab's "quality assur- ance program . . . ensured that evidence was stored properly." Over- ton II, 976 So. 2d at 553 n.14. And thus he was not prejudiced by trial counsel's failure to challenge the RFLP DNA evidence because it was admissible and the RFLP DNA testing results matched his

DNA.

To show that the Florida Supreme Court unreasonably de- termined that the RFLP DNA evidence was admissible, Overton notes that RFLP DNA evidence "is not per se admissible." Appel- lant's Br. 75. He again points to Brim to say that the trial court

"might" have excluded the evidence due to defects in evidence col- lection. 779 So. 2d at 438 (emphasis added). Further, he advances that if the STR DNA evidence had properly been excluded, the RFLP "in-house DNA result alone lends significantly more cre- dence to trial counsel's argument that the DNA was at the very least contaminated, if not planted." Appellant's Br. 84-85 (internal quotation marks omitted).

We are troubled by Pope's handling of the DNA evidence in this case. And we agree that Overton need not "definitively prove that DNA test results would have been excluded" to warrant ha- beas relief. Reply Br. 25. But establishing prejudice under Strickland requires Overton to show a reasonable probability of a different result. His arguments show, at best, that the RFLP DNA testing results might have been excludable. They fall far short of showing that the Florida Supreme Court's conclusion that the evidence was

34

16-10654 Opinion of the Court 35 admissible was so wrong as to be unreasonable. After considering the record evidence that RFLP DNA testing is generally accepted, the FDLE lab followed proper procedures when it conducted RFLP testing of the DNA evidence, and the RFLP DNA test results matched Overton's DNA, and applying the deference AEDPA re- quires, we cannot say that "no fairminded jurist could agree" with the Florida Supreme Court's no-prejudice determination. King,

69 F.4th at 867 (internal quotation marks omitted). The district court did not err in rejecting Overton's ineffec- tive assistance of counsel claim.

C. Overton's Brady Claim Does Not Withstand AEDPA Def- erence.

Overton contends that the Florida Supreme Court unrea- sonably concluded that Pope's mishandling of evidence in another case, Allen, 854 So. 2d at 1255, was not material, such that the State's failure to disclose evidence of it did not prejudice his de- fense. In his view, "even if the suppressed information was 'just' impeachment evidence," it was an unreasonable application of Brady to conclude that evidence of Pope's evidence mishandling would have been cumulative of the evidence presented at trial. Ap- pellant's Br. 95. He argues that because the evidence would have demonstrated a pattern of mishandling, would have impeached the State's star witness, and was of a different nature than the impeach- ment evidence in this case, it was material. Likewise, he contends that the aggregate effect of the suppressed evidence made the evi- dence material because his trial counsel, armed with the suppressed

35

36 Opinion of the Court 16-10654 evidence, would have been able to draw concessions from several of the state's witnesses at trial. In addition, Overton asserts that even if the suppressed evidence was cumulative, the Florida Su- preme Court unreasonably applied Brady in concluding that it was immaterial. Finally, he contends that the Florida Supreme Court unreasonably applied Brady in assessing materiality by failing to ac- count for the fact that, had defense counsel obtained evidence of Pope's sloppy evidence handling practices in other cases, they not only would have impeached him with that evidence but also would have used it to have the DNA evidence Pope collected in this case excluded.

"As recognized in Brady and its progeny, principles of due process dictate that, in a criminal proceeding, the prosecution must disclose evidence favorable to the defendant." Rimmer v. Sec'y, Fla. Dep't of Corr., 876 F.3d 1039, 1053 (11th Cir. 2017). The Supreme Court has identified "three components of a true Brady violation:

[(1)] [t]he evidence at issue must be favorable to the accused, either because it is exculpatory, or because it is impeaching; [(2)] that ev- idence must have been suppressed by the State, either willfully or inadvertently; and [(3)] prejudice must have ensued." Strickler v. Greene, 527 U.S. 263, 281-82 (1999). Because it is determinative, we need only discuss the third element here.

"To establish prejudice, the defendant must show that the suppressed evidence was material." Rimmer, 876 F.3d at 1054.

"[F]avorable evidence is material, and constitutional error results from its suppression by the government, if there is a reasonable

36

16-10654 Opinion of the Court 37 probability that, had the evidence been disclosed to the defense, the result of the proceeding would have been different." Kyles v. Whit- ley, 514 U.S. 419, 433 (1995) (internal quotation marks omit- ted). "In determining whether disclosure of the suppressed evi- dence might have produced a different result, we must consider the 'totality of the circumstances.'" Rimmer, 876 F.3d at 1054 (quoting United States v. Bagley, 473 U.S. 667, 683 (1985)). "We must examine the trial record, evaluate the withheld evidence in the context of the entire record, and determine in light of that examination whether there is a reasonable probability that, had the evidence been disclosed, the result of the proceeding would have been dif- ferent." Turner v. United States, 582 U.S. 313, 324-25 (2017) (altera- tion adopted) (internal quotation marks and citation omitted). Noting that the undisclosed evidence of Pope's sloppy evi- dence collection practices "occurred in another case" and not this one, the Florida Supreme Court concluded that the evidence had

"minimal value." Overton II, 976 So. 2d at 563. Further, the Court explained that at trial Overton's counsel cross-examined Pope ex- tensively on his flawed handling of the DNA evidence tying Over- ton to the murder scene, which included:

(1) envelopes that were used to store DNA evidence were misdated; (2) there were no property receipts to account for the swabs that were used to obtain fluids from Susan's body at the scene; (3) the swabs were transported to [Pope's] home, which was not a certi- fied storage facility; (4) these swabs were placed in [Pope's] home refrigerator; (5) the first property

37

38 Opinion of the Court 16-10654 receipt for the envelopes of clippings, which provided a match to Overton's DNA, was dated June 10, 1994 [almost three years after the crime scene was pro- cessed]; (6) the bedding (quilt, mattress pad, com- forter, and bed sheet) on which semen stains were found were placed in paper bags and transported to [Pope's] home to be air dried; (7) the bedding was transported to the Key West property evidence stor- age room on August 26, 1991 [four days after the crime scene was processed]; and (8) [Pope] trans- ported the mattress pad in a paper bag by car to Or- lando to have a psychic conduct an inspection. Id. at 551. Given this "significant" challenge to Pope's reliability at trial, the Court concluded that the "alleged evidence of similar sloppy work in another case would be cumulative" and unlikely to change the result of Overton's trial. Id. at 563. The Florida Supreme Court's conclusion was not an unrea- sonable application of Brady. The United States Supreme Court ex- amined the materiality of undisclosed evidence that could have been used to impeach a state's witness in Turner, 582 U.S. at 313. There, several petitioners brought Brady claims challenging their murder convictions after they learned that the state failed to dis- close exculpatory evidence. Id. at 316. Among other examples of undisclosed evidence, the petitioners pointed to "a prosecutor's un- disclosed note [that] revealed [a state's witness] said she had been high on PCP during a January 9, 1985, meeting with investigators."

Id. at 323. The Court concluded that the undisclosed impeachment evidence "was largely cumulative of impeachment evidence

38

16-10654 Opinion of the Court 39 petitioners already had and used at trial." Id. at 327. The Court ex- plained that "the jury heard multiple times about [the state wit- ness's] frequent PCP use, including" the witness's testimony that she "smoked PCP shortly before" she witnessed the attack in ques- tion. Id. Therefore, "it would not have surprised the jury to learn that [the witness] used PCP on yet another occasion." Id. The Court noted that its decision was not to "suggest that impeachment evidence is immaterial with respect to a witness who has already been impeached with other evidence." Id. But the Court concluded that "in the context of this trial, with respect to these witnesses, the effect of the withheld evidence is insufficient to 'undermine confi- dence' in the jury's verdict.'" Id. at 328 (quoting Smith v. Cain,

565 U.S. 73, 75-76 (2012)). Here, the undisclosed information showing Pope's im- proper handling of DNA evidence in Allen was cumulative of the evidence of his similar DNA mishandling in this case. Just as in Turner, "it would not have surprised the jury to learn" of Pope's mishandling of DNA evidence in another case because of defense counsel's impeachment of Pope with his mishandling of the DNA evidence in this case. 582 U.S. at 327. And the evidence of his mis- handling of DNA in Overton's case was likely more compelling than evidence of Pope's similar conduct in another case. Id. Nor can we agree with Overton that the aggregate effect of the evidence and the evidence's "different nature" made it material. Reply Br. 37. Again, Pope was impeached with evidence of his "sloppy evidence collection practices" in this case. Appellant's Br. 101. We are un- convinced that if additional state's witnesses were confronted on

39

40 Opinion of the Court 16-10654 cross examination with the suppressed information it would have

"put the whole case in such a different light as to undermine confi- dence in the verdict." Kyles, 514 U.S. at 433. Therefore, we cannot say that the Florida Supreme Court unreasonably applied Brady in concluding that the undisclosed evidence was cumulative of Pope's impeachment at trial and immaterial for the purposes of Overton's Brady claim.

Still, Overton argues that even if the withheld evidence was cumulative, it was material nonetheless because it "tends to im- peach or discredit the key evidence or testimony against" him. Re- ply Br. 35. In support of this argument, he contends that this case

"closely mirrors" our decision in United States v. Arnold, 117 F.3d 1308 (11th Cir. 1997), in which we held that the undisclosed im- peachment evidence was material even though the witness had been impeached with other evidence. Reply Br. 36. Arnold is distin- guishable, however. In Arnold, we reversed on direct appeal the dis- trict court's denial of the defendants' motion for a new trial where the government withheld recordings that contradicted the govern- ment's key witness's trial testimony. Arnold, 117 F.3d at 1318. The witness "testified that he had no expectation of a sentence reduc- tion in return for cooperating with the government[,]" but in a re- cording in the government's possession, the witness "discussed his expectation of a five-year reduction of his" sentence. Id. We con- cluded that the defendants' Brady claim had merit even though the defendants impeached the witness at trial with other evidence— evidence that he was in a romantic relationship with one of the government's agents investigating the case. Id. at 1315. And not

40

16-10654 Opinion of the Court 41 only did the withheld recordings show that the witness may have perjured himself, but they also indicated that he may have been

"performing for the prosecution" and had his testimony coached by the government. Id. at 1318 (internal quotation marks omitted). Overton does not assert that the state withheld information indicating that Pope lied in his trial testimony. Nor does he contend that the state unduly influenced Pope's testimony. Unlike in this case, in Arnold it would have "surprised the jury to learn" of the re- cordings contradicting the witness's trial testimony and demon- strating the government's coaching of his testimony, even though the defendants had already impeached him with evidence that he was romantically involved with the government's agent. Turner,

582 U.S. at 327. So Arnold does not suggest that the Florida Su- preme Court unreasonably applied Brady in deciding that the cu- mulative evidence of Pope's evidence handling in Allen was imma- terial.

The other cases Overton offers in support of his argument are distinguishable for similar reasons. See Banks v. Dretke, 540 U.S. 668, 703 (2004) (concluding that a petitioner's Brady claim was mer- itorious where the state withheld the informant status of a witness who was impeached for other reasons, including that he was an informant in different cases); Guzman v. Sec'y, Dep't of Corr.,

663 F.3d 1336, 1351 (11th Cir. 2011) (concluding that a petitioner's Giglio claim was meritorious where the state withheld that its key witness, who previously told police she had no information about the case, was paid $500 for her testimony, even though she was

41

42 Opinion of the Court 16-10654 impeached at trial for agreeing to testify for a lesser charge in her own case and for being reimbursed for expenses during trial, among other reasons).

We are likewise unpersuaded by Overton's argument that the trial court may have ruled Pope's DNA evidence inadmissible at the Frye hearing if it had been aware of Pope's deficient handling of evidence in the Allen case. The Florida Supreme Court con- cluded that the RFLP DNA results "were clearly admissible and the results from this testing also matched Overton." Overton II, 976 So. 2d at 553 n.14. Overton's prejudice argument cannot succeed by challenging as erroneous this state law evidentiary ruling. See Snow- den v. Singletary, 135 F.3d 732, 737 (11th Cir. 1998) (concluding that federal courts are not empowered to correct erroneous evidentiary rulings of state courts unless the error rises to the level of a denial of fundamental fairness to the habeas petitioner in violation of the Due Process Clause).

* * *

"Because the Florida Supreme Court's decision on [Over- ton's] Brady claim did not contain an error so clear that fair-minded people could not disagree about it, we defer to that decision deny- ing [him] relief on his Brady claim." Rimmer, 876 F.3d at 1057.

IV. CONCLUSION

For these reasons, we affirm the district court's denial of Overton's petition for a writ of habeas corpus.

AFFIRMED.

42