Leonardo Franqui v. Secretary, Florida Department of Corrections

2 Opinion of the Court 24-11887 (2002). Leonardo Franqui, who is subject to two Florida death sen- tences, contends that this prohibition prevents Florida from exe- cuting him.

The Florida courts considered that claim in consolidated state postconviction proceedings covering both of Franqui's capital cases. After an evidentiary hearing, they concluded that Franqui did not establish that he is intellectually disabled. Franqui then challenged that ruling in 28 U.S.C. § 2254 ha- beas proceedings. The district court denied Franqui's petition. We granted a certificate of appealability on two questions. First, was the Florida Supreme Court's rejection of Franqui's intel- lectual-disability claim contrary to, or an unreasonable application of, Atkins? Second, did that decision rest on an unreasonable deter- mination of the facts presented in state court?

After reviewing the record and the parties' briefs, and with the benefit of oral argument, we answer "no" to both questions. So we affirm the denial of Franqui's habeas petition.

I. Background

We discuss the relevant background in four parts. First, we identify Franqui's two capital judgments and explain how they pro- duced separate federal habeas actions. Second, we trace the state postconviction proceedings that eventually brought the two intel- lectual-disability claims together. Third, we review the evidence and decisions that resulted from those consolidated proceedings.

2

24-11887 Opinion of the Court 3 Finally, we describe the federal proceedings that produced this ap- peal.

A. Franqui's Two Capital Cases

Franqui is subject to two death sentences from two separate Florida cases. The first stems from the 1991 murder of Raul Lopez during an attempted armed robbery in Hialeah, Florida. See Fran- qui v. State, 699 So. 2d 1312, 1315-16 (Fla. 1997) ("Hialeah case"). The second arises out of the 1992 murder of law-enforcement of- ficer Steven Bauer during an unrelated bank robbery in North Mi- ami, Florida. See Franqui v. State, 804 So. 2d 1185, 1190-91 (Fla. 2001) ("North Miami case"). The Florida Supreme Court affirmed Franqui's conviction and death sentence in each case. See Franqui,

699 So. 2d at 1329; Franqui, 804 So. 2d at 1199. Franqui challenged the two judgments in separate federal habeas actions. He challenged the North Miami judgment in Case No. 07-22384-CIV-KMM. The district court denied that petition in 2008. See Franqui v. Florida, No. 07-22384-CIV-KMM, ECF No. 12 (S.D. Fla. July 10, 2008).¹Franqui challenged the Hialeah judgment

3

4 Opinion of the Court 24-11887 in Case No. 11-22858-CIV-GRAHAM. The district court's judg- ment in the Hialeah action is the subject of this appeal. No court consolidated the two federal habeas actions. Each action challenged a different state judgment, proceeded under a dif- ferent federal case number, and resulted in its own federal orders. So this appeal comes from the federal judgment denying Franqui's petition challenging the Hialeah judgment.

With that in mind, we return to the state postconviction pro- ceedings that occurred after Franqui filed his federal habeas cases. After the Supreme Court decided Hall v. Florida, 572 U.S. 701 (2014), Franqui sought renewed state postconviction relief in both capital cases. Franqui v. State, 211 So. 3d 1026, 1030 (Fla. 2017). The Florida Supreme Court consolidated the two state appeals and re- manded both cases for a single evidentiary hearing on whether Franqui is intellectually disabled. Id. at 1027, 1032. That approach followed from the nature of Franqui's claim. The two murders were separate, but whether Franqui is intellectu- ally disabled concerns Franqui, not either offense. As the state post- conviction court had explained during the earlier North Miami pro- ceedings, Franqui "cannot be [intellectually disabled] in one case and not in the other, as that would defy the definition of [intellec- tual disability]." Id. at 1030.

18 (S.D. Fla. July 10, 2008). And we later denied Franqui a certificate of ap- pealability in proceedings arising from that federal case. See Franqui v. Sec'y, Fla. Dep't of Corr., No. 15-12013 (11th Cir. Sept. 30, 2015).

4

24-11887 Opinion of the Court 5 This distinction defines the scope of our review. Procedur- ally, this appeal comes from the federal habeas action challenging the Hialeah judgment. Substantively, the state-court intellectual- disability determination we review arose from consolidated pro- ceedings covering both death sentences. Because we affirm the de- nial of habeas relief, we need not decide what effect a ruling in Franqui's favor would have on his separate North Miami judg- ment.

B. Franqui's State Intellectual-Disability Proceedings

1. The Hialeah Proceedings

Evidence about Franqui's intellectual functioning first arose during the Hialeah trial. A Florida jury convicted Franqui of first- degree murder and related offenses. Franqui, 699 So. 2d at 1316. At the penalty phase, Franqui presented testimony from Dr. Jethro Toomer. Dr. Toomer administered the Revised Beta Examination and the Wechsler Adult Intelligence Scale-Revised ("WAIS-R") and opined that Franqui's IQ fell within what he called the "retarded range."²Franqui, 699 So. 2d at 1325.

The State responded with testimony from Dr. Charles Mut- ter, who evaluated Franqui and reviewed the tests that Dr. Toomer

5

6 Opinion of the Court 24-11887 had administered. Franqui, 699 So. 2d at 1325-26. Dr. Mutter re- jected Dr. Toomer's conclusions and opined that Franqui was not intellectually disabled. Franqui, 699 So. 2d at 1325-26. The jury recommended death by a vote of nine to three, and the trial court imposed a death sentence. Franqui, 699 So. 2d at 1316. The Supreme Court had not yet decided Atkins, which first held in 2002 that the Eighth Amendment prohibits the execution of an intellectually disabled person. See Atkins, 536 U.S. at 321. On direct appeal, Franqui argued that the trial court should have found nonstatutory mitigation based on his "marginal or re- tarded intelligence and brain damage." Franqui, 699 So. 2d at 1325. The Florida Supreme Court rejected that argument. Id. at 1326. It concluded that "competent, substantial evidence" supported the trial court's determination that Franqui had not established either mitigating circumstance. Id.

After the Supreme Court decided Atkins, Franqui supple- mented his initial state postconviction motion in the Hialeah case with a claim that he was intellectually disabled and therefore con- stitutionally ineligible for execution. Franqui v. State, 59 So. 3d 82, 89-90 (Fla. 2011).

The state postconviction court did not initially rule on Fran- qui's intellectual-disability claim. Franqui, 59 So. 3d at 90. So the Florida Supreme Court returned the case to the postconviction court to address the claim. Id. The postconviction court summar- ily denied the claim. Id. Then the Florida Supreme Court again returned the case and directed the postconviction court to hold an

6

24-11887 Opinion of the Court 7 evidentiary hearing. Id.; see also Franqui v. State, 14 So. 3d 238, 239 (Fla. 2009).

At the time of that hearing, Florida used a three-part test for intellectual disability. Franqui, 59 So. 3d at 91-92. Franqui had to show (1) significantly subaverage intellectual functioning, (2) con- current deficits in adaptive behavior, and (3) the manifestation of the condition before he turned eighteen. Id. at 92. If Franqui failed to establish any one of those requirements, his claim failed. Id. Florida law then treated an IQ score of 70 as a strict cutoff for the first requirement. Id. at 92-93. A score above 70 meant that a defendant had not established significantly subaverage intellec- tual functioning. Id. And if the defendant failed that first require- ment, the court did not consider adaptive functioning or onset be- fore age eighteen. Id.

At the evidentiary hearing, the parties entered reports from Dr. Trudy Block-Garfield and Dr. Enrique Suarez. Id. at 90-91. Dr. Block-Garfield had evaluated Franqui in 2003. Id. at 90. She ad- ministered the Stanford-Binet Intelligence Scale, which produced a full-scale IQ score of 76, and the Wechsler Adult Intelligence Scale, Third Edition, which produced a full-scale IQ score of 75. Id. Dr. Block-Garfield also considered Dr. Toomer's 1993 administration of an earlier version of the Wechsler test, on which Franqui had scored 83. Id. at 90-91. She estimated that Franqui's true IQ fell between 71 and 80 and concluded that he was not intellectually dis- abled. Id. at 90.

7

8 Opinion of the Court 24-11887 Dr. Suarez evaluated Franqui in 2009. Id. at 91. Franqui scored 75 on the Wechsler Adult Intelligence Scale, Fourth Edition. Id. But Dr. Suarez also administered five tests designed to deter- mine whether Franqui was giving his best effort. Id. The results suggested that Franqui was deliberately performing poorly. Id. Dr. Suarez therefore believed that the IQ scores understated Franqui's actual abilities, and he determined that Franqui was not intellectu- ally disabled. Id.

Based on these reports, the postconviction court concluded that Franqui had not satisfied Florida's test for intellectual disabil- ity. Id. at 91. In particular, none of Franqui's scores on the ap- proved intelligence tests was 70 or below. Id. at 92. The Florida Supreme Court affirmed the denial of relief. Id. at 92-94. It also rejected Franqui's argument that Florida's strict IQ cutoff con- flicted with Atkins. Id. at 94.

2. The North Miami Proceedings

Franqui later raised the same constitutional claim in a suc- cessive postconviction motion in the North Miami case. Franqui v. State, 211 So. 3d 1026, 1029 (Fla. 2017). He relied in part on Dr. Toomer's testimony from the Hialeah trial that one test had pro- duced an IQ score below 60. Id. The State responded with the reports from Drs. Block-Garfield and Suarez that the courts had considered in the Hialeah proceedings. Id.

The postconviction court denied the North Miami claim. Id. It first determined that Franqui had filed the claim too late. Id. It also concluded that the Hialeah proceedings had already resolved

8

24-11887 Opinion of the Court 9 whether Franqui was intellectually disabled. Id. As the court ex- plained, Franqui "cannot be [intellectually disabled] in one case and not in the other, as that would defy the definition of [intellectual disability]." Id.

The Florida Supreme Court affirmed. Id. at 1029-30. It ex- plained that the test producing Dr. Toomer's score below 60 was not one of the tests Florida accepted for determining intellectual disability. Id. Franqui's scores on the accepted tests were all above

70. Id. at 1030. And Franqui had not alleged that his claimed intel- lectual disability manifested before he turned eighteen. Id. at 1030.

3. Hall and the Consolidated Remand

In 2014, the Supreme Court decided Hall, which addressed Florida's strict IQ cutoff and its risk of excluding individuals who may be intellectually disabled. 572 U.S. at 704. As the Court ex- plained, an IQ test does not produce a perfectly precise measure- ment. Id. at 713. Because every test has a standard error of meas- urement, an IQ score represents a range of possible scores rather than one fixed number. Id.

For that reason, Hall held that Florida could not automati- cally reject an intellectual-disability claim merely because the de- fendant obtained an IQ score above 70. Id. at 723. When a defend- ant's IQ scores are above 70 but only to a degree within the test's acknowledged margin of error, the defendant had to be allowed to present other evidence, including evidence of adaptive deficits. Id. at 723-24.

9

10 Opinion of the Court 24-11887 Based on Hall, Franqui filed successive postconviction mo- tions in both capital cases. Franqui, 211 So. 3d at 1030. He argued that Florida's former strict IQ cutoff had prevented the state courts from properly considering all the evidence bearing on his intellec- tual disability. Id. The postconviction court summarily denied both motions. Id.

In the Hialeah case, the postconviction court reasoned that the earlier proceedings had considered all three requirements for intellectual disability, not only Franqui's IQ scores. Id. It also relied on Dr. Block-Garfield's conclusion that Franqui had supported his family and did not display adaptive deficits. Id. In the North Miami case, the court concluded that Hall did not create a new right to bring an intellectual-disability claim and that Franqui's motion was untimely. Id. at 1030-31.

Franqui appealed both rulings, and the Florida Supreme Court consolidated the appeals. Id. at 1027. The court first held that Hall applied retroactively to Franqui's cases. Id. at 1031. It then concluded that the earlier Hialeah hearing may have been lim- ited by Florida's former strict IQ cutoff. Id. at 1031-32. Because none of Franqui's scores on the approved tests were 70 or below, the court reasoned, Franqui and his attorneys may have viewed ad- ditional evidence about adaptive functioning and childhood onset as futile. Id. at 1032.

For these reasons, the Florida Supreme Court concluded that Franqui had not necessarily received the type of holistic eval- uation that Hall required. Id. It reversed the summary denials in

10

24-11887 Opinion of the Court 11 both capital cases and remanded them for a single evidentiary hear- ing on whether Franqui is intellectually disabled. Id. Franqui's counsel agreed that one hearing could resolve the claim in both cases. Id. at 1032 n.6.

C. The Consolidated Evidentiary Hearing

As we have explained, the Florida Supreme Court remanded both cases for a single evidentiary hearing on whether Franqui is intellectually disabled. Under Florida law, Franqui had to establish three requirements: (1) significantly subaverage intellectual func- tioning, (2) concurrent deficits in adaptive behavior, and (3) the manifestation of the condition before he turned eighteen. § 921.137(1), Fla. Stat. (2017); Franqui v. State, 301 So. 3d 152, 154 (Fla. 2020) ("Franqui III"). If Franqui failed to prove any one of those requirements, he could not establish intellectual disability. See Sal- azar v. State, 188 So. 3d 799, 812 (Fla. 2016).

The first requirement focuses on intellectual functioning. Florida defined significantly subaverage intellectual functioning as performance at least two standard deviations below the mean score on a standardized intelligence test. Fla. R. Crim. P. 3.203(b); Fran- qui III, 301 So. 3d at 154. Because standardized IQ tests generally use a mean score of 100 and a standard deviation of 15, a score ap- proximately two standard deviations below the mean is 70. See Hall, 572 U.S. at 711.

But, as Hall explained, an IQ score is not perfectly precise. Id. at 713. Every score has a standard error of measurement, or

"SEM," that accounts for the test's inherent margin of error. Id.

11

12 Opinion of the Court 24-11887 The SEM means that an obtained score represents a range within which the person's true score probably falls. Id. Thus, a court can- not treat a single score of 71, for example, as proving that a person's intellectual functioning necessarily exceeds the intellectual-disabil- ity range. See id. at 713, 723.

The hearing also involved a different concept known as the Flynn Effect. The Flynn Effect describes the tendency of IQ scores to rise over time after a test is standardized on a particular popula- tion. See Raulerson v. Warden, 928 F.3d 987, 1008 (11th Cir. 2019). As a result, administering an older version of an IQ test may pro- duce a score that overstates the test-taker's intellectual functioning. Id. The SEM and the Flynn Effect therefore address different con- cerns. The SEM reflects the imprecision in an individual test score. The Flynn Effect concerns the possible inflation of scores caused by using aging testing norms.

The second requirement focuses on adaptive behavior. § 921.137(1), Fla. Stat. (2017). Adaptive behavior concerns how ef- fectively a person "meets the standards of personal independence and social responsibility expected" for someone of the same "age, cultural group, and community." Id. Clinicians generally consider functioning in three domains: conceptual skills, social skills, and practical skills.

Finally, Franqui had to establish that his significantly subav- erage intellectual functioning and adaptive deficits manifested be-

12

24-11887 Opinion of the Court 13 fore he turned eighteen. § 921.137(1), Fla. Stat. (2017). This re- quirement reflects the developmental nature of intellectual disabil- ity. See Atkins, 536 U.S. at 308 n.3.

With that framework in mind, the postconviction court heard testimony from three experts. Franqui presented Dr. Gor- don Taub and Dr. Jethro Toomer. The State presented Dr. En- rique Suarez. Their testimony focused principally on Franqui's in- tellectual functioning, his adaptive behavior, and whether the avail- able evidence showed that any limitations began before he turned eighteen.

1. Dr. Gordon Taub

In 2015, one of Franqui's attorneys asked Dr. Taub, a psy- chologist with a doctorate in school psychology, to review Dr. Block-Garfield's 2003 report. So Dr. Taub reviewed her admin- istration of the Stanford-Binet Intelligence Scale, Fourth Edition, and the Wechsler Adult Intelligence Scale, Third Edition. Those tests produced full-scale IQ scores of 76 and 75, respectively. Fran- qui, 59 So. 3d at 90-91.

Dr. Taub testified that the Flynn Effect applies across intelli- gence tests and likely inflated both scores. The Stanford-Binet test had been standardized in 1985, eighteen years before Dr. Block- Garfield administered it to Franqui. After adjusting for the Flynn Effect, Dr. Taub reduced Franqui's score on that test from 76 to 71. The Wechsler test had been standardized in 1995, eight years before Franqui took it. Dr. Taub adjusted Franqui's score on that test from 75 to 73. His calculations assumed that Dr. Block-

13

14 Opinion of the Court 24-11887 Garfield had not already accounted for the Flynn Effect because her report did not mention an adjustment.

Dr. Taub then addressed the SEM. When asked to identify the SEM for the Wechsler test, he said that it was between four and five points. Using five points as an estimate, he calculated ranges around Franqui's scores after first adjusting them for the Flynn Ef- fect. Dr. Taub explained that experts sometimes apply both adjust- ments in death-penalty cases because the consequences are so sig- nificant.

The postconviction court also questioned Dr. Taub about whether Franqui's background might have affected the testing. Franqui's native language is Spanish, but Drs. Toomer and Suarez interviewed and tested him in English. Dr. Taub noted that Fran- qui had arrived in the United States from Cuba more than twenty years before the relevant testing and that the available information described him as fluent in English. So Dr. Taub did not believe that testing Franqui in English had materially affected his scores. Dr. Taub also considered Franqui's limited formal educa- tion. He explained that intelligence and academic achievement do not have a one-to-one relationship. According to Dr. Taub, intelli- gence generally accounts for only about one-half, and at most sixty percent, of a person's academic performance. Thus, in his view, Franqui's limited schooling did not by itself establish the level of his intellectual functioning.

2. Dr. Jethro Toomer

14

24-11887 Opinion of the Court 15 Franqui retained Dr. Toomer, a clinical and forensic psy- chologist, in March 1992. Dr. Toomer examined Franqui on three separate occasions. To collect background information, Dr. Toomer also interviewed Franqui and his uncle, Mario Franqui Suarez.

Dr. Toomer administered the Revised Beta Examination, which produced the low score that formed the basis for Dr. Toomer's original opinion that Franqui fell within the intellectual- disability range. See Franqui, 211 So. 3d at 1030. But Florida did not recognize that examination as an approved test for determining in- tellectual disability. Id. On the other hand, the approved Wechsler test produced a full-scale IQ score of 83. Id. at 1029 n.4. At the consolidated hearing, Dr. Toomer did not dispute Dr. Taub's opinion that the Flynn Effect should reduce the Wechsler score from 83 to 78. Even with that adjustment, though, the score remained above 70. So Dr. Toomer also relied on Franqui's per- sonal history and functioning to support his opinion. Dr. Toomer testified that Franqui experienced academic dif- ficulties while attending school in Cuba. Those difficulties contin- ued after Franqui moved to Miami when he was nine or ten years old. According to Dr. Toomer, Franqui also experienced problems involving his education, interpersonal relationships, and connec- tions with family members. Franqui ultimately left school before completing the seventh grade.

15

16 Opinion of the Court 24-11887 Dr. Toomer relied in part on information supplied by Fran- qui's uncle. Franqui's uncle described Franqui's academic difficul- ties in both Cuba and the United States. He also reported that Fran- qui became increasingly isolated and behaved more erratically after the death of Franqui's younger sibling. According to the uncle's account, Franqui's father then abandoned Franqui and coped with his own grief through substance abuse.

Based on that history, Dr. Toomer concluded that Franqui's limitations began before he turned eighteen. That testimony ad- dressed the third requirement for intellectual disability, which asks whether the claimed intellectual and adaptive limitations mani- fested during childhood. Dr. Toomer also viewed Franqui's poor academic performance and difficulties in school as evidence of def- icits in the conceptual domain of adaptive functioning. The State questioned whether Franqui's activities showed that he could function independently. The prosecutor asked Dr. Toomer whether any evidence showed that Franqui had needed assistance with daily activities. Dr. Toomer responded that the need for assistance did not resolve the adaptive-functioning in- quiry. In his view, the relevant question was how effectively Fran- qui performed those activities.

The prosecutor also pointed to Franqui's ability to steal cars and hold several jobs. Dr. Toomer did not view those activities as proof that Franqui functioned effectively. He explained that Fran- qui's jobs generally involved repetitive skills and that Franqui did

16

24-11887 Opinion of the Court 17 not keep any of his jobs for a significant period. Dr. Toomer there- fore believed that Franqui's employment history remained con- sistent with adaptive limitations. Dr. Toomer also testified that he had reviewed the prior testimony of Dr. Brad Fisher, who had ex- amined Franqui and assessed his intelligence as average. Finally, Dr. Toomer addressed why Franqui left school. He testified that Franqui's school records reflected trauma at home and erratic functioning. Dr. Toomer attributed Franqui's decision to leave school to those difficulties rather than to a simple lack of interest in education.

In sum, Dr. Toomer addressed all three requirements for in- tellectual disability. He accepted a Flynn-Effect-adjusted IQ score of 78 as evidence of Franqui's intellectual functioning. He relied on Franqui's academic, employment, and interpersonal history as evidence of adaptive limitations. And he relied on Franqui's child- hood history to conclude that those limitations manifested before Franqui turned eighteen.

3. Dr. Enrique Suarez

The State presented testimony from Dr. Suarez, a licensed psychologist who had evaluated Franqui on August 31 and Septem- ber 4, 2009. Dr. Suarez regularly served as an expert in capital cases and had received appointments in more than 1,400 criminal cases between 1998 and 2017.

To prepare his opinions, Dr. Suarez reviewed Franqui's background materials and conducted telephone interviews with correctional officers who had observed Franqui in prison. The

17

18 Opinion of the Court 24-11887 background materials included Dr. Toomer's report, which rec- orded information that Dr. Toomer had obtained from Franqui and Franqui's uncle. Dr. Suarez therefore considered the family history reported during Dr. Toomer's earlier evaluation, but the record does not indicate that Dr. Suarez personally interviewed Franqui's uncle.

Dr. Suarez first discussed Franqui's life before his arrest. Franqui reported that he met Vivian Gonzalez when he was seven- teen. Soon afterward, they moved into Franqui's uncle's home. Franqui worked for his uncle, who deducted rent from Franqui's paycheck. Franqui said that he worked to support his family finan- cially. About a year later, Franqui and Gonzalez moved into their own home, where they remained until Franqui's arrest. Dr. Suarez also reviewed evidence of Franqui's functioning in prison. Between 1997 and 2015, prison officials received several grievances filed in Franqui's name. Dr. Suarez believed that the grievances appeared to share the same handwriting, although he could not determine with certainty that Franqui had written them. Dr. Suarez also noted that Franqui selected his own items from a prison canteen offering more than 170 products, and Franqui cor- responded with a pen pal, either independently or with assistance. Dr. Suarez next addressed the intelligence testing. He ex- plained that Dr. Block-Garfield's report did not identify which ver- sion of the Wechsler test she had administered. Dr. Suarez con- tacted her in 2009 and learned that she had administered the Wechsler Adult Intelligence Scale, Third Edition. He also agreed

18

24-11887 Opinion of the Court 19 with Dr. Toomer's decision to administer the standard English-lan- guage Wechsler test rather than a version standardized in Puerto Rico. In Dr. Suarez's view, the Puerto Rican version was not suffi- ciently reliable.

Dr. Suarez disagreed with Dr. Taub about the Flynn Effect. He testified that the Flynn Effect was not universally accepted and that experts disagreed about how it should be applied to an individ- ual test-taker. He identified two principal concerns. First, research- ers had derived the effect from aggregated data across different countries, but the size of the observed effect was not uniform. Sec- ond, no professional consensus established how to translate those population-level results into a numerical adjustment for one per- son.

Dr. Suarez further testified that, even if the Flynn Effect were applied, the adjusted scores would not establish that Franqui was intellectually disabled. And in Dr. Suarez's view, it was statis- tically improper to adjust an IQ score for the Flynn Effect and then apply the same SEM used for the unadjusted score. He explained that the Flynn adjustment introduced an additional variable into the calculation. So according to Dr. Suarez, the margin of error surrounding the adjusted score could not simply be assumed to be the same as the margin surrounding the original score. Dr. Suarez also believed that several features of Franqui's background could have caused the tests to understate, rather than overstate, his intellectual abilities. Those features included Fran-

19

20 Opinion of the Court 24-11887 qui's limited formal education, his first language, and his incom- plete immersion in the culture in which the tests had been stand- ardized. Dr. Suarez therefore believed that Franqui's actual intel- lectual functioning might exceed his obtained scores. That possibility was especially important to Dr. Suarez be- cause of the validity testing. Dr. Suarez administered several tests designed to measure whether Franqui was making his best effort. Franqui "basically failed all of them." According to Dr. Suarez, the pattern of responses suggested that Franqui sometimes knew the correct answer but intentionally selected an incorrect one. Dr. Sua- rez testified that such a pattern was highly unusual for an intellec- tually disabled person. He therefore believed that Franqui had ma- lingered and that his IQ scores understated his actual abilities. Dr. Suarez also evaluated Franqui's adaptive functioning. In assessing Franqui's conceptual skills, he considered Franqui's com- munication, functional academics, and self-direction. Franqui told Dr. Suarez that he had left school after repeated absences because he wanted to spend time with other young people. Based on his interviews with Franqui, Dr. Suarez observed no adaptive deficits in these areas.

Other evidence reinforced that view. Dr. Suarez believed that Franqui had written grievances for other inmates. He used the Flesch-Kincaid Index to evaluate two letters attributed to Franqui and determined that one was written at a twelfth-grade level and the other at an eleventh-grade level. Franqui also recalled each of the nine vehicles he had owned, how he acquired each one, and

20

24-11887 Opinion of the Court 21 when he traded it. Dr. Suarez viewed that detailed recall as incon- sistent with a significant memory deficit.

Finally, Dr. Suarez noted that previous experts had not con- cluded that Franqui was intellectually disabled. Based on the test- ing, the evidence of malingering, and Franqui's demonstrated func- tioning before and during incarceration, Dr. Suarez likewise con- cluded that Franqui did not satisfy the requirements for intellectual disability.

4. The State Postconviction Court's Decision

On May 28, 2018, the state postconviction court denied Franqui's intellectual-disability claim. The court evaluated each of the three requirements and concluded that Franqui had failed to establish any of them by clear and convincing evidence. For the first requirement, the court considered Franqui's scores on the approved Wechsler and Stanford-Binet intelligence tests. According to the court, even after accounting for the SEM, the scores discussed by Franqui's experts produced a range from 71 to 92. The court did not consider the score from the Revised Beta Examination because Dr. Toomer had conceded that it was not a recommended test for determining intellectual disability. Franqui III, 301 So. 3d at 154-55.

The court also declined to reduce Franqui's scores based on the Flynn Effect. It credited Dr. Suarez's testimony that experts did not universally accept the Flynn Effect and did not agree on how to apply it to an individual test-taker. Besides that, the court relied

21

22 Opinion of the Court 24-11887 on Quince v. State, 241 So. 3d 58, 62 (Fla. 2018), which upheld a de- cision not to adjust an IQ score for the Flynn Effect when substan- tial record evidence supported that decision. The court therefore concluded that Franqui had not shown that his intellectual func- tioning fell two or more standard deviations below the mean score of 100.

For the second requirement, the court considered Franqui's functioning in the conceptual, social, and practical domains. See Franqui III, 301 So. 3d at 155. In the conceptual domain, the court recognized that Franqui's home life had affected his school perfor- mance. But it determined that Franqui's repeated absences, rather than intellectual disability, caused his academic difficulties. The court also considered Franqui's employment history and the skills his various jobs required. The court concluded that those experi- ences did not support a deficit in the practical domain. After con- sidering all three domains, the court determined that Franqui had not established significant adaptive deficits.

For the third requirement, the court evaluated whether any intellectual and adaptive limitations had manifested before Franqui turned eighteen. It acknowledged the evidence concerning Fran- qui's difficult childhood, family trauma, and school performance. But the court concluded that the evidence did not establish the re- quired developmental onset. Instead, the court found that Franqui had "functioned normally in society" before his incarceration and had "shown no adaptive deficits" while incarcerated.

22

24-11887 Opinion of the Court 23 Because the court found that Franqui had not established any of the three requirements, it denied his intellectual-disability claim.

5. The Florida Supreme Court's Decision

Franqui appealed, and the Florida Supreme Court affirmed in Franqui III. 301 So. 3d at 153. The court reviewed the postcon- viction court's determination for competent, substantial evidence. Id. at 154. Under that standard, the court did not reweigh the evi- dence or second-guess the postconviction court's credibility find- ings. Id.

Franqui first challenged Florida's requirement that he prove intellectual disability by clear and convincing evidence. Id. at 155. But the Florida Supreme Court concluded that Franqui had not raised that challenge in time to preserve it for appellate review. Id. The court then addressed Franqui's argument that the post- conviction court had failed to conduct the holistic review that Hall required. Id. It rejected that argument because the postconviction court had evaluated each of the three requirements for intellectual disability. Id.

On intellectual functioning, the Florida Supreme Court noted that the postconviction court had accounted for the SEM and determined that Franqui's lowest score was 71. Id. The court also pointed out that the postconviction court had excluded the Revised Beta score based on testimony that the Revised Beta was not a rec- ommended intelligence test. Id. The Florida Supreme Court con-

23

24 Opinion of the Court 24-11887 cluded that competent, substantial evidence supported the deter- mination that Franqui had not established significantly subaverage intellectual functioning. Id.

Then the Florida Supreme Court considered adaptive func- tioning. It explained that the postconviction court had evaluated Franqui's functioning in the conceptual, social, and practical do- mains. Id. The court concluded that competent, substantial evi- dence supported the determination that Franqui had not estab- lished significant adaptive deficits. Id.

The Florida Supreme Court likewise upheld the finding on developmental onset. Id. It quoted the postconviction court's con- clusion that Franqui had functioned normally before his incarcera- tion and had displayed no adaptive deficits while incarcerated. Id. The Florida Supreme Court concluded that competent, substantial evidence supported the postconviction court's findings on all three requirements. Id.

Finally, the Florida Supreme Court rejected Franqui's argu- ment that the postconviction court's analysis conflicted with Moore

v. Texas, 581 U.S. 1 (2017). Franqui III, 301 So. 3d at 155. Although the postconviction court concluded that Franqui's scores did not place him within the clinically recognized range for intellectual- functioning deficits, the Florida Supreme Court explained, the postconviction court did not stop its analysis there. Id. It also con- sidered the evidence of adaptive functioning and childhood onset. Id. The Florida Supreme Court therefore concluded that the post- conviction court had considered the full intellectual-disability claim

24

24-11887 Opinion of the Court 25 rather than treating Franqui's IQ scores as automatically disposi- tive. Id.

Having upheld the findings on each requirement, the Flor- ida Supreme Court affirmed the denial of Franqui's intellectual-dis- ability claim. Id.

D. The Federal Proceedings Producing This Appeal

Meanwhile, after the Florida Supreme Court affirmed the denial of Franqui's initial intellectual-disability claim, Franqui filed a federal habeas petition challenging the Hialeah judgment. Peti- tion, Franqui v. Crews, No. 11-22858-CIV-GRAHAM, ECF No. 1 (S.D. Fla. Aug. 8, 2011). Among other grounds for relief, Franqui asserted that his intellectual disability made him constitutionally ineligible for execution. Id. at 83.

The district court denied the petition in 2014. Franqui v. Crews, No. 11-22858-CIV-GRAHAM, ECF No. 17, at 2 (S.D. Fla. Aug. 20, 2014). The district court concluded that Hall did not apply retroactively to Franqui's federal habeas proceedings. Id. at 75-76. But it granted Franqui a certificate of appealability on his intellec- tual-disability claim. Id. at 79.

While that appeal was pending, the Florida Supreme Court ordered the consolidated evidentiary hearing we've described above. Franqui, 211 So. 3d at 1032. Given that development, this Court vacated the district court's judgment and remanded the Hi- aleah federal habeas action for further proceedings after the state litigation concluded. Franqui v. Florida, No. 14-14278-P, ECF No. 59 (11th Cir. Mar. 20, 2017).

25

26 Opinion of the Court 24-11887 After the Florida Supreme Court decided Franqui III, the par- ties returned to the federal district court and filed supplemental briefs addressing that decision. The district court again denied ha- beas relief. Order at 1, Franqui v. Sec'y, Fla. Dep't of Corr., No. 11- 22858-CIV-GRAHAM, ECF No. 49 (S.D. Fla. May 1, 2024). It con- cluded that Hall did not apply retroactively to Franqui's federal ha- beas claim. Id. at 13-18. The district court therefore evaluated the Florida Supreme Court's decision under Atkins and the deferential standards governing federal habeas review. Id. at 18-28. It con- cluded that Franqui had not shown that Franqui III was contrary to, or an unreasonable application of, Atkins or that the decision rested on an unreasonable determination of the facts. Id. at 28. The dis- trict court also denied a certificate of appealability. Id. A judge of this Court later granted Franqui a certificate of appealability on two issues: (1) whether Franqui III's rejection of Franqui's intellectual-disability claim was contrary to, or an unrea- sonable application of, Atkins; and (2) whether Franqui III rested on an unreasonable determination of the facts in light of the evidence presented in state court. Franqui v. Sec'y, Fla. Dep't of Corr., No. 24- 11887-P, ECF No. 11 (11th Cir. Nov. 14, 2024).

The certificate expressly excludes any claim that Franqui III violated Hall or Moore v. Texas, 581 U.S. 1 (2017), because those de- cisions do not apply retroactively on federal collateral review. Franqui, No. 24-11887-P, ECF No. 11 (11th Cir. Nov. 14, 2024). Franqui moved to expand the certificate so that he could challenge

26

24-11887 Opinion of the Court 27 that retroactivity determination. Motion to Expand Certificate of Ap- pealability at 4-6, Franqui, No. 24-11887-P, ECF No. 15 (11th Cir. Feb. 4, 2025). This Court denied that motion. Franqui, No. 24- 11887, ECF No. 17 (11th Cir. Feb. 24, 2025). So the issues before us are narrower than the issues the Flor- ida courts considered during the proceedings following Hall. We must decide only whether Franqui III survives federal habeas re- view under Atkins and 28 U.S.C. § 2254(d).

II. Standard of Review

We review "de novo the denial of a petition for a writ of habeas corpus." Morrow v. Warden, 886 F.3d 1138, 1146 (11th Cir. 2018) (quotation omitted). But the Antiterrorism and Effective Death Penalty Act of 1996 ("AEDPA") governs our review of fed- eral habeas petitions and prescribes a highly deferential framework for evaluating issues state courts have previously decided. Pye v. Warden, Ga. Diagnostic Prison, 50 F.4th 1025, 1034 (11th Cir. 2022) (en banc); Sealey v. Warden, Ga. Diagnostic Prison, 954 F.3d 1338, 1354 (11th Cir. 2020).

Under AEDPA, we may not grant habeas relief on claims that were "adjudicated on the merits in [s]tate court" unless the state court's decision (1) "was contrary to, or involved an unrea- sonable application of, clearly established Federal law, as deter- mined 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." Sealey, 954 F.3d at 1354 (quoting 28 U.S.C. § 2254(d)). These standards mean

27

28 Opinion of the Court 24-11887 that we must give state-court decisions "the benefit of the doubt."

Evans v. Sec'y, Dep't of Corr., 703 F.3d 1316, 1325 (11th Cir. 2013) (citation and internal quotation marks omitted). A state-court decision is not "contrary to" federal law under § 2254(d)(1) "unless it contradicts the United States Supreme Court on a settled question of law or holds differently than did that Court on a set of materially indistinguishable facts." Id. (citation and in- ternal quotation marks omitted); see also Williams v. Taylor, 529 U.S.

362, 405 (2000).

And a state-court decision is not an "unreasonable applica- tion" of federal law under § 2254(d)(1) "unless the state court iden- tifies the correct governing legal principle as articulated by the United States Supreme Court, but unreasonably applies that prin- ciple to the facts of the petitioner's case, unreasonably extends the principle to a new context where it should not apply, or unreason- ably refuses to extend it to a new context where it should apply."

Evans, 703 F.3d at 1325 (citation and internal quotation marks omit- ted); see also Williams, 529 U.S. at 407 (a state court decision is an

"unreasonable application" of clearly established law if the state court identifies the correct governing legal rule from the Supreme Court's holdings but unreasonably applies it to the facts of the par- ticular defendant's case). The question is not whether the state court was wrong, but whether its decision was so lacking in justifi- cation that there is no possibility for fairminded disagreement. Har- rington v. Richter, 562 U.S. 86, 102-03 (2011).

28

24-11887 Opinion of the Court 29 Section 2254(d)(2) requires a separate inquiry. A state fac- tual determination is not unreasonable merely because the federal court would have weighed the evidence differently. See Wood v. Allen, 558 U.S. 290, 301 (2010). The petitioner must show that the state court's factual determination was objectively unreasonable given the evidence presented. Miller-El v. Cockrell, 537 U.S. 322, 340 (2003). And we presume state-court factual findings are correct un- less they are rebutted by clear and convincing evidence. 28 U.S.C. § 2254(e)(1).

"Our review is limited to the issue specified in the certificate of appealability." Castillo v. United States, 816 F.3d 1300, 1306 (11th Cir. 2016). The COA here confines our § 2254(d)(1) review to At- kins, not Hall or Moore, because our precedent holds that Hall and Moore do not apply retroactively on collateral review. See In re Henry, 757 F.3d 1151, 1159-61 (11th Cir. 2014); Smith v. Comm'r, Ala. Dep't of Corr., 924 F.3d 1330, 1338-40 (11th Cir. 2019).

III. Discussion

AEDPA's restrictions on our review mean the questions we must consider are narrower than the ultimate question of whether Franqui is intellectually disabled. Rather, we ask only whether the Florida Supreme Court's rejection of Franqui's claim of intellectual disability was contrary to or an unreasonable application of Atkins, and whether its decision rested on unreasonable factual determina- tions. Franqui's arguments fail in both respects.

A. The Florida Supreme Court's decision was not con- trary to, or an unreasonable application of, Atkins.

29

30 Opinion of the Court 24-11887 In Atkins v. Virginia, 536 U.S. 304 (2002), the Supreme Court held that the Eighth Amendment prohibits the execution of intel- lectually disabled individuals. But Atkins expressly acknowledged

"serious disagreement" over how intellectual disability should be defined, so it "le[ft] to the State[s] the task of developing appropri- ate ways to enforce the constitutional restriction." Id. at 317 (quot- ing Ford v. Wainwright, 477 U.S. 399, 416-17 (1986)). In other words, Atkins articulated a constitutional floor, not a detailed diag- nostic framework. Id. at 318. That floor requires state standards to be grounded in clinical definitions that account for (1) subaver- age intellectual functioning, (2) significant limitations in adaptive functioning, and (3) onset before age eighteen. Id. Franqui does not identify a rule that the Florida courts ap- plied that contradicts Atkins. Nor does he identify materially indis- tinguishable Supreme Court precedent that reaches the opposite result of the Florida courts. Instead, he argues primarily that the Florida Supreme Court discussed Hall and Moore. But as we've noted, this Court's COA limits the federal habeas inquiry to Atkins. According to Franqui, that creates a "paradox" because the state court applied Hall and Moore, while the COA prevents him from relying on those cases.

We disagree. The Florida courts grounded their analysis in clinical concepts of intellectual functioning and adaptive behavior and considered both IQ evidence and adaptive functioning. Then they made a case-specific determination—precisely the approach

30

24-11887 Opinion of the Court 31 Atkins permits. See Ledford v. Warden, Ga. Diagnostic & Classification Prison, 818 F.3d 600, 633-34 (11th Cir. 2016).

Under AEDPA and the COA, the relevant clearly established federal law for this appeal is Atkins. Hall and Moore do not apply retroactively on collateral review. See Henry, 757 F.3d at 1159-61; Smith, 924 F.3d at 1338-40. Nor does the Supreme Court's later disposition in Hamm v. Smith, No. 24-872, slip op. at 1 (U.S. May 21, 2026) (per curiam), change the inquiry. The Court dismissed the writ of certiorari in Hamm as improvidently granted and did not issue a merits holding on how courts must evaluate multiple IQ scores in Atkins cases. See id. A dismissal as improvidently granted supplies no new rule of clearly established federal law for purposes of § 2254(d)(1).

Even though Hall and Moore do not supply the governing law for Franqui's federal habeas claim, explaining those decisions helps frame his argument. As we have noted, Hall rejected Flor- ida's use of an IQ score of 70 as a rigid cutoff. 572 U.S. at 723. Be- cause an IQ score reflects a range rather than a fixed measurement, Hall required courts to account for the test's standard error of measurement. Id. at 713. When the test's margin of error placed the defendant within the clinically recognized range for intellectual disability, the defendant had to be allowed to present other evi- dence, including evidence of adaptive deficits. Id. at 723. Moore built on Hall in two respects. First, it explained that a State's intellectual-disability determination must remain informed by the medical community's diagnostic framework. 581 U.S. at 13-

31

32 Opinion of the Court 24-11887

15. A court may not substitute lay stereotypes about intellectual disability for that clinical framework. Id. at 14-16. Second, Moore explained that the adaptive-functioning inquiry focuses on adaptive deficits. Id. at 15-16. A court therefore may not allow a defendant's perceived strengths to overwhelm substantial evidence of adaptive limitations. Id. Moore also cautioned against assuming that other conditions or traumatic experiences necessarily disprove intellec- tual disability because those conditions can coexist with intellectual disability or contribute to it. Id. at 14-15.

Franqui relies on those principles to challenge both the in- tellectual-functioning and adaptive-functioning components of the state courts' analysis. As to intellectual functioning, he argues that the state courts failed to give adequate effect to the standard error of measurement and improperly declined to adjust his IQ scores for the Flynn Effect. As to adaptive functioning, he argues that the courts placed too much weight on his perceived strengths, includ- ing his employment, family relationships, prison activities, and par- ticipation in the charged offenses. He also argues that the courts improperly discounted his academic and social limitations by at- tributing them to his absences from school, difficult home life, and other causes.

Even if we could consider Hall and Moore for purposes of evaluating Franqui's arguments, neither decision shows that Fran- qui III was unreasonable under AEDPA. The Florida Supreme Court explained that, although the postconviction court deter-

32

24-11887 Opinion of the Court 33 mined that Franqui's IQ scores did not place him within the clini- cally recognized range for intellectual disability, the postconviction court still considered other evidence bearing on intellectual disabil- ity. Franqui III, 301 So. 3d at 155. The postconviction order sup- ports that description. It reflects that the postconviction court con- sidered the standard error of measurement, evaluated multiple IQ scores, heard competing expert testimony, and then separately considered adaptive functioning and childhood onset. Put simply, the postconviction court didn't treat a score above 70 as dispositive or end its analysis after reviewing Franqui's IQ scores. And alt- hough the postconviction court declined to apply the Flynn Effect, Hall addressed the standard error of measurement; it did not re- quire courts to adjust IQ scores for the Flynn Effect. See Raulerson,

928 F.3d at 1008. Franqui III also reasonably concluded that the postconvic- tion court considered evidence beyond Franqui's IQ scores. 301 So. 3d at 155. The postconviction court's order organized its adap- tive-functioning analysis around the conceptual, social, and practi- cal domains. Within each domain, the postconviction court evalu- ated Franqui's claimed limitations, evidence of his functioning, and the competing expert testimony. Id. The court considered Fran- qui's perceived strengths, but it did not treat any one strength, in- cluding his conduct during the offenses, as dispositive. Id. So the state-court decision did not rest on the rigid or nonclinical analysis that Hall and Moore prohibit.

33

34 Opinion of the Court 24-11887 Franqui next argues that the Florida Supreme Court's failure to cite Atkins in its opinion shows that it applied the wrong law. But AEDPA does not require a state court to cite the governing Supreme Court decision, or even be aware of it, so long as neither the court's reasoning nor its result contradicts Supreme Court prec- edent. Early v. Packer, 537 U.S. 3, 8 (2002). Here, they don't. Franqui next argues that the Florida Supreme Court unrea- sonably applied Atkins by reviewing the postconviction court's findings for competent, substantial evidence rather than inde- pendently reviewing and reweighing the entire record. But Atkins does not prescribe the standard that a state appellate court must use when reviewing a postconviction court's intellectual-disability determination. Nor does Atkins require a state appellate court to conduct de novo factfinding or independently reweigh credibility determinations made after an evidentiary hearing. See Atkins, 536 U.S. at 317. And a state court "need not make detailed findings addressing all the evidence before it." Miller-El, 537 U.S. at 347. Finally, Franqui faults Franqui III for offering too little expla- nation. But for the reasons we've explained, Franqui III's explana- tion for its decision was not unreasonable. The Florida Supreme Court identified the governing intellectual-disability standard, de- scribed the postconviction court's treatment of Franqui's intellec- tual-functioning and adaptive-functioning evidence, and explained why competent, substantial evidence supported the denial of relief. Franqui III, 301 So. 3d at 154-55.

34

24-11887 Opinion of the Court 35 Franqui therefore has not shown that Franqui III was con- trary to, or involved an unreasonable application of, Atkins.

B. Franqui III did not rest on an unreasonable determina- tion of the facts.

As we've noted, in Florida, a person must meet three stand- ards to be intellectually disabled: (1) significantly subaverage intel- lectual functioning; (2) concurrent deficits in adaptive behavior; and (3) manifestation of the condition before age eighteen. § 921.137(1), Fla. Stat. "Subaverage intellectual functioning" means

"performance that is two or more standard deviations from the mean score on a standardized intelligence test." Fla. R. Crim. P. 3.203(b). The state postconviction court found that Franqui failed to establish all three prongs. The Florida Supreme Court affirmed. Franqui III, 301 So. 3d at 154-55.

As we explain below, we conclude that the state courts' de- terminations were not objectively unreasonable given the evidence presented.

1. Intellectual Functioning

Franqui contends that the state courts unreasonably deter- mined that his IQ scores do not show significantly subaverage in- tellectual functioning. He relies mainly on Dr. Taub's testimony that, after applying the Flynn Effect and the SEM, some of Fran- qui's IQ scores could fall near or within the intellectually disabled range.

The state courts did not unreasonably reject that argument.

35

36 Opinion of the Court 24-11887 Franqui emphasizes that Dr. Taub testified that the SEM could place a score of 75 in a range of 70 to 80, while the state court used 71 to 80. But that dispute does not make the state court's de- termination objectively unreasonable. First, Dr. Suarez testified that the SEM range was 71 to 80, and the state court was entitled to credit his testimony. Second, even crediting 70 as the lowest possible endpoint for one score, the state court did not rest its de- cision on a single number. It considered the full body of IQ evi- dence, the disputed Flynn Effect adjustment, the absence of any re- liable score below 70 on a preferred test, and evidence suggesting malingering. On that record, fairminded jurists could agree that Franqui failed to prove significantly subaverage intellectual func- tioning by clear and convincing evidence.

Nor does Hall compel a different conclusion, even assuming we could consider it. Hall held that a court can't treat an IQ score above 70 as an automatic bar to considering other evidence when the score's SEM places the defendant within the clinically recog- nized range for intellectual disability. 572 U.S. at 723. But the state postconviction court here did not refuse to consider other evidence after seeing a score above 70. Rather, it heard and evaluated evi- dence about intellectual functioning, adaptive functioning, onset, and malingering. Only then did it find that Franqui had failed to carry his burden.

Based on this record, that factual determination was not ob- jectively unreasonable. United States v. Stein, 964 F.3d 1313, 1322 (11th Cir. 2020) ("[Trial courts] frequently must choose between

36

24-11887 Opinion of the Court 37 dueling experts, and if that decision is reasonably based on evidence found in the record, the choice is not clear error.").

2. Adaptive Functioning

Franqui also contends that the state courts unreasonably found that he failed to establish significant deficits in adaptive func- tioning. He argues that the postconviction court minimized evi- dence of school difficulties, ignored testimony from his uncle, mis- understood school records, and overstated evidence of his work history, family life, prison functioning, and criminal conduct. We note first that, because the state courts reasonably deter- mined that Franqui failed to prove significantly subaverage intel- lectual functioning, his claim necessarily fails regardless of whether he established adaptive-functioning deficits.

But in any case, we can't conclude that the state courts un- reasonably determined that Franqui failed to show significant defi- cits in adaptive functioning. To be sure, the record contained evi- dence on both sides. But the mere existence of conflicting evidence does not establish an unreasonable factual determination. See Eg- gers v. Alabama, 876 F.3d 1086, 1102 (11th Cir. 2017). The state post- conviction court held an evidentiary hearing, considered compet- ing expert testimony, and made findings across the conceptual, so- cial, and practical domains. As we discuss below, the Florida Su- preme Court did not unreasonably conclude that the evidence of record supported the state postconviction court's findings.

a. Conceptual Domain

37

38 Opinion of the Court 24-11887 With respect to the conceptual domain (including skills such as language, reading, writing, math reasoning, memory, and the acquisition and use of practical knowledge), Franqui focuses on one sentence in the state postconviction court's order after the eviden- tiary hearing: "No evidence was presented regarding the Defend- ant's preschool experience, except that his mother was absent and he did not know his father." Franqui contends this statement is false. He points to testimony that he participated in "special clas- ses" in Cuba, had trouble in school there, and was "very slow" as a child. Then he argues that the state court did not review the record or did not understand how to evaluate the evidence. But the state postconviction court did not ignore that evi- dence. It acknowledged that Franqui had academic difficulties and that his home life affected his school performance. Even so, it found that excessive absences and behavioral choices principally drove Franqui's academic problems. Yet the postconviction court also acknowledged the evidence of family disruption and trauma. That factual finding was not unreasonable. The State pre- sented evidence that Franqui had performed as an average student before his attendance deteriorated, that he later missed substantial amounts of school, and that he said he skipped school because he wanted to spend time with friends. And Dr. Suarez testified that he observed no adaptive deficits during his interviews with Fran- qui. To the contrary, Dr. Suarez reported, Franqui showed func- tional communication, memory, and practical reasoning abilities.

38

24-11887 Opinion of the Court 39 Evidence that Franqui wrote grievances, corresponded with oth- ers, requested legal materials, and wrote letters at high reading lev- els also supported the postconviction court's conceptual-domain finding.

Franqui's uncle's testimony and school records might sup- port a different inference. But § 2254(d)(2) does not authorize relief whenever the record permits competing interpretations. The question is whether the state court's interpretation was objectively unreasonable. And on this record, it was not.

b. Social Domain

In the social domain, Franqui argues that the state court overstated his ability to form relationships and function socially. Again, the record contained evidence on both sides of the question. But the postconviction court reasonably relied on evidence that Franqui formed friendships, maintained romantic and family rela- tionships, had children, associated with others before his arrest, and interacted socially with other inmates while incarcerated. The court did not need to find that Franqui had no limita- tions. It needed only to determine whether Franqui proved signif- icant adaptive deficits in the social domain. On the record before it, including Dr. Suarez's testimony and the evidence of Franqui's relationships and prison interactions, the court determined that the evidence did not support a finding of significant social adaptive def- icits. That conclusion was not unreasonable, even if Franqui points to evidence that might support a different view. See Holsey v. War- den, Ga. Diagnostic Prison, 694 F.3d 1230, 1257 (11th Cir. 2012) ("[I]f

39

40 Opinion of the Court 24-11887 some fairminded jurists could agree with the state court's decision, although others might disagree, federal habeas relief must be de- nied." (citation and internal quotation marks omitted)).

c. Practical Domain

The same is true of the practical domain. Franqui challenges the state court's reliance on his work history, ability to drive, family support, canteen use, hygiene, prison functioning, and involve- ment in the charged crimes. He argues that the state court over- stated much of this evidence. In this regard, Franqui asserts that his jobs were menial, his uncle deducted rent from his paycheck, some vehicles were gifts or trade-ins, and prison life is structured. But those points do not make the state court's determination unreasonable. The postconviction court could consider evidence that Franqui maintained employment, lived with a partner and children, had a driver's license, handled daily activities, selected canteen items, wrote or obtained assistance with written materials, and functioned independently in prison. After all, Dr. Suarez testi- fied that those facts were inconsistent with significant practical adaptive deficits.

The postconviction court also considered evidence of plan- ning and goal-directed behavior in the charged offenses. True, if we were considering Moore, Moore cautions against focusing on

"perceived adaptive strengths" instead of adaptive deficits. 581 U.S. at 15-18. But that's not what happened here. Rather, the postcon- viction court did not rely on the sophistication of the crimes alone.

40

24-11887 Opinion of the Court 41 It considered the offense conduct together with employment, fam- ily life, prison functioning, written communication, expert testi- mony, and the broader record. That was not objectively unreason- able.

d. Overall Adaptive-Functioning Determination

Franqui's adaptive-functioning argument ultimately asks us to reweigh the evidence. He emphasizes evidence of deficits; the State emphasizes evidence of functioning. The state court reason- ably credited the State's view after an evidentiary hearing. Under AEDPA, we must defer to that reasonable determination. See Rice

v. Collins, 546 U.S. 333, 338-39 (2006). Nor did the postconviction court treat adaptive functioning as an all-or-nothing inquiry or conclude that Franqui had no limi- tations. It separately evaluated Franqui's functioning in the con- ceptual, social, and practical domains. It acknowledged Franqui's claimed limitations but concluded, after weighing the competing evidence, that he had not proved adaptive deficits of the signifi- cance required to establish intellectual disability. The Florida Su- preme Court held that competent, substantial evidence supported that determination. Franqui III, 301 So. 3d at 155. On this record, the state courts' finding that Franqui failed to prove significant adaptive deficits was not based on an unreason- able determination of the facts.

3. Onset Before Age of Eighteen

41

42 Opinion of the Court 24-11887 Finally, Franqui challenges the state courts' finding that Franqui failed to establish that significantly subaverage intellectual functioning and concurrent adaptive deficits manifested before Franqui turned eighteen.

Because the state courts reasonably determined that Franqui failed to prove the first two prongs, his claim necessarily fails re- gardless of the onset prong. A finding of intellectual disability re- quires all three elements. See § 921.137(1), Fla. Stat.; Fla. R. Crim. P. 3.203(b).

But in any event, we can't say that the onset finding was ob- jectively unreasonable. The postconviction court considered evi- dence of childhood school difficulties and family instability but credited evidence that absences, trauma, and behavioral issues, ra- ther than intellectual disability, caused Franqui's problems in these respects. The court also relied on evidence that Franqui was not diagnosed with intellectual disability before he turned eighteen and that his later functioning did not show the kind of lifelong adaptive impairment associated with intellectual disability. Fairminded ju- rists could agree with that conclusion.

In sum, Franqui has not shown that the Florida Supreme Court's decision upholding the postconviction court's decision rested on an unreasonable determination of the facts.

IV. Conclusion

Franqui has not established that the Florida Supreme Court's rejection of his intellectual-disability claim was contrary to, or an unreasonable application of, Atkins. Nor has he shown that

42

24-11887 Opinion of the Court 43 the Florida Supreme Court's decision rested on an unreasonable determination of the facts. We therefore affirm the district court's denial of habeas relief.

AFFIRMED.

43