dissenting:
The one thing everyone appears to agree on is that Smith is not intellectually disabled.1 When tested in 2005 the experts found that he had an IQ of between 87 and 93, well within the low-average to average range of intellectual ability. Yet despite this fact, the majority reverses because it is certain that Smith was intellectually disabled in 1980 when he murdered Sandy Owen. The majority reaches this conclusion by disregarding the findings of the state courts, denying those courts the deference they are due, and expressing supreme confidence in its own ability to detect past intellectual disability despite substantial conflicting evidence and the fact that Smith is not now intellectually disabled. Accordingly, I dissent. The majority recognizes that although Smith filed his federal habeas petition pri- or to the effective date Antiterrorism and Effective Death Penalty Act of 1996 (AEDPA), the state court factual findings are entitled to a presumption of correctness.2 Nonetheless, the majority concludes that the state court’s factual determination is “not fairly supported by the record.” Op. 1181. The majority is wrong as an objective review of the record discloses ample evidence to support the Arizona courts’ determination that Smith did not sustain his burden of showing that he was intellectually disabled in 1980.
Section II C 2 of Judge Reinhardt’s “opinion” also contains his view — for which there is no concurrence — that the Arizona courts applied an “unconstitutional standard of proof.” This argument is based on *1218an unreasonable reading of the trial court’s decision and a failure to give the state courts’ decisions the deference they are due. Judge Reinhardt’s opposition to the death penalty in Section 2 C of his opinion is neither the ruling of this panel nor of the Ninth Circuit.
I. The Record Adequately Supports the Determination that Smith Failed to Show He Was Intellectually Disabled at the Time of the Murder and His Conviction.
The majority boldly asserts that the finding that Smith has not shown that he was intellectually disabled at the time of the murder and his trial is not fairly supported by evidence. Op. 1181. The majority recognizes, as it must, that there was conflicting evidence, but argues that “once we look behind each expert’s conclusion and consider the evidence on which he relies,” the majority of the evidence supports a finding of intellectual disability. Op. 1181 n.7. This is not a fair reading of the record.
A. The Record
1. The Underlying Crime and Prior Judicial Proceedings
In our 1999 en banc opinion affirming the denial of Smith’s first federal habeas appeal, we described the crime as follows:
Lambright and Smith were traveling across the country with Lambright’s girlfriend, Kathy Foreman. Smith was troubled by the fact that while Lam-bright and Foreman had intercourse in his presence, he did not have anybody along to satisfy him. For his part, Lam-bright thought that he “would like to kill somebody just to see if he could do it.” [State v. Lambright ], 138 Ariz. [63,] 66 [673 P.2d 1 (1983)].... They decided that both desires could be fulfilled, and they set out with Foreman to find a victim. They found Sandy Owen and kidnaped her. Smith raped her on the way to a mountain site where they all got out of the car and Smith raped Owen again as Lambright and Foreman had intercourse. What happened next was that Smith began choking Owen, and Lambright declared that she must be killed. So, “Lambright took Foreman’s knife out of its sheath and began stabbing the victim in the chest and abdomen, twisting the knife around inside of her. Smith held one of the victim’s arms while she was being stabbed, and Foreman held the other arm.” Id. at 67 [673 P.2d 1]____After that, “Smith unsuccessfully tried to break Ms. Owen’s neck by twisting her head. Then Lam-bright, Foreman or both began cutting deeply into the victim’s neck with the knife.... The victim remained alive, and was at least semiconscious, as she attempted to raise herself up on one arm. Lambright picked up a large rock and hurled it at her head. Foreman testified that as he threw the rock he yelled ‘Die, bitch.’ ” Id. The three then drove off in a celebratory mood, playing the piece ‘We Are the Champions” as they went. See id. Once caught, the trio’s song changed. Foreman turned state’s evidence, was given immunity, and testified against her erstwhile lover and his friend. Lambright confessed, but deemed Smith to be the worst of the three. Smith, too, confessed, but he dubbed Foreman and Lambright as the real killers.
In 1982, an Arizona jury convicted Robert Douglas Smith of first-degree murder, kid-naping, and sexual assault. Schriro v. Smith, 546 U.S. 6,126 S.Ct. 7, 163 L.Ed.2d 6 (2005). He was given the death penalty.
Smith’s appeals and post-conviction proceedings proved unavailing. See Stewart v. Smith, 536 U.S. 856, 122 S.Ct. 2578, 153 L.Ed.2d 762 (2002). It was not until after the Supreme Court had denied him relief, that around 2004 Smith alleged for the *1219first time that he was intellectually disabled and thus pursuant to Atkins v. Virginia, 586 U.S. 304, 122 S.Ct. 2242, 153 L.Ed.2d 335 (2002), could not be executed. See Schriro v. Smith, 546 U.S. 6, 126 S.Ct. 7, 163 L.Ed.2d 6 (2005). In December 2005, we entered an order suspending Smith’s federal proceedings and directing his counsel “to pursue state proceedings in Arizona to determine whether the state is prohibited from executing the petitioner in accordance with Atkins.” Smith v. Schriro, No. 96-99025 (Dec. 12, 2005). The Arizona Superior Court, Pima County, held extensive proceedings concerning Smith’s intellectual disability before issuing its opinion on March 27, 2008.
2. The Evidence Concerning Intellectual Disability
In 1964, when Smith was 15, he took the Otis IQ test and received scores of 62 and 71. The state court found that the “Otis test was developed in approximately the 1920’s and was outmoded at the time it was reportedly given to [Smith] in 1964.” More importantly, “[t]here is no evidence concerning the qualifications of the persons administering the tests, whether an appropriate protocol was followed, the specific circumstances of [Smith] at the times of the tests, or any of the other information required to determine the validity of these school record entries.”
Moreover, by this time, Smith’s dysfunctional and abusive childhood had already had an effect on his education. At age 15 he scored in the 2nd to 5th percentiles on the Stanford Achievement Test, placing him seven years below his age level. Moreover, his school transcripts reveal that he received nearly all “Ds” and “Fs” in his academic studies. Indeed, Smith dropped out of school when he was 16.
Between the time he left school at the age of 16, and his arrest some 15 years later in 1980, Smith had many jobs, a number of unstable relationships, and frequent changes of residence. The state court found that those who knew Smith stated that “he was one of a group of young men which included his co-defendant, Joe Lambert, Sidney LeBalanc and Charles McCarver, who lived together, worked together, and traveled together at various times during the 1970’s.” The state court found that the evidence from the 1970’s showed that Smith was a full participant in his adult life. It found that Smith:
worked as a diesel mechanic, garage mechanic, car repossessor, truck driver, cable installer and apartment maintenance provider, among other things. While Defendant held a large number of jobs, he was consistently employed. Defendant’s last employer indicated, in a form attached to Defendant’s pre-sen-tence report, that Defendant worked for approximately four months as a mechanic employed to maintain and repair equipment, that he received a raise after three months, that his work was rated “satisfactory” for job performance, work skills and attendance, and “excellent” for cooperation with employer/supervisor and other employees, and that he would be considered for reemployment.
Smith’s friend, LeBlanc, who sometimes worked for the same employer as Smith, stated that Smith had difficulties with paperwork and written tests, as opposed to hands-on or mechanical tasks. Similarly, Robert Lebrecque, a former maintenance man with the Arizona Department of Corrections, who worked with Smith for about eight years after he began serving his prison term, commented that although Smith “was a little slow at the beginning ... [i]f I showed him how to do something, I only had to show him once.” Lebrecque recalled that Smith could read, “but *1220seemed to have a hard time understanding what the written words meant.”3
Smith was married five times in approximately 15 years. The state court, however, noted that while Smith “made poor choices in partners and had great difficultly maintaining relationships with women, this fact can be explained as arising from his loveless childhood just as well as it can be viewed as an indicator of the limitations of mental retardation.” Smith’s last marriage, which was entered into between the time of the offense and his arrest, “showed promise of being quite different from the others despite two incidents of violence in the relationship.” His fifth wife had known Smith since childhood. During this marriage Smith worked as a truck driver and performed dry-wall and other work in the apartment complex. His wife did not work outside the home and Smith supported her arid her three children, for whom he was a loved, active father figure. After he was arrested, the pre-sentence report characterized Smith as having a “borderline mentality.” The state court, however, gave this description little weight because the “experience of this probation officer is unknown, and there was no indication that he had training as a psychologist or other mental health professional that would provide the expertise required for any diagnostic observations.”4
In addition, two mental health professionals, Dr. Martin Levy and Dr. John LaWall, performed Rule 11 evaluations of Smith in 1982 for purposes of his criminal trial proceedings. Both found Smith to be competent. Dr. Levy noted that Smith was neatly dressed and displayed logical, coherent thought.5 Dr. LaWall also noted that Smith “was neat and cooperative, and his mood somewhat depressed, with somewhat blunted affect.” Dr. LaWall’s report indicated that Smith was oriented to time, place, and person, with intact memory, and *1221there was “no evidence of any disturbance of the form or content of his thinking whatsoever.” Dr. LaWall indicated that Smith “probably functions in the average range of intelligence,” but probably has a personality disorder with both passive-aggressive and antisocial features.
3. The Post-2005 Evaluations
After our 2005 order, proceedings were commenced in the Arizona Superior Court, Pima County, for the sole purpose of complying with our order that the state court determine whether Arizona was prohibited from executing Smith in accordance with Atkins. The Superior Court noted that the burden was on Smith “to prove the claim of mental retardation by clear and convincing evidence” and that both parties acknowledged that the court was “bound to follow the decision of the Arizona Supreme Court in State v. Grell (Grell II), 212 Ariz. 516, 521, 135 P.3d 696, 701 (2006) in this regard.”
Smith was subjected to testing and evaluations by experts retained by both Smith and Arizona. Testing in August 2005 by Dr. Sergio Martinez, Arizona’s expert, resulted in a finding that Smith “had an IQ score of 93 on the WAIS-III and a score of 89 on a second test, the Slosson Intelligence Test-Revised, within the low-average to average range of intellectual ability.” Testing by Dr. Thomas Thompson, Smith’s expert, utilizing a different appropriate testing instrument, the Reynolds Intellectual Assessment Scale with subtests, resulted in a finding that Defendant, at the time of the testing, had a score of 93. Thus, both experts agree that as of 2005 Smith was not intellectually disabled.
Dr. Thompson, however, was of the opinion that there was a “high probability” that Smith “was mentally retarded at the time the crime was committed in 1980, but that his functioning has improved as a result of his stable, structured prison life and appropriate medication.” Dr. Thompson relied heavily on the Otis test scores from 1964, the Stanford test scores, early grades and recollections by relatives and others concerning Smith’s childhood and early adolescence. Dr. Thompson appeared “to view mental retardation as a fluid condition responsive to any number of changes in a patient’s environment, nutrition, and physical, mental and emotional health.” He considered Smith’s low test scores, low grades, lack of social skills and other deficits as valid indicators of mental retardation.
When asked about evidence of Smith’s intellectual disability in 1980, Dr. Thompson referred to the 1964 IQ tests and noted that the presentence evaluation indicated that he functioned in a borderline range. However, when informed of the two Rule 11 evaluations by Drs. Levy and LaWall, he acknowledged that he would have expected retardation to have been noted in their reports.
Dr. Thompson described Smith’s life after he left school as “characterized by instability in employment, personal relationships and residence, and showed signs of impulsivity and deficits of adaptive functioning, all characteristic of mental retardation.” Nonetheless, he acknowledged that Smith “seemed to have some qualitative independent living skills.”6
In contrast, Dr. Martinez, after giving Smith IQ tests and meeting with him, testified that there was a high degree of *1222probability that he was not retarded at the time of the offense. He agreed that increases in intellectual functioning could occur within a highly enriched learning environment, but that a 30-point increase in IQ was unlikely and he did not view prison as an enriching environment.
4. The Arizona Superior Court’s Decision
On March 27, 2008, the Arizona Superior Court issued a 19-page ruling that Smith had failed to show that he was intellectually disabled at the time of his trial and that Arizona was therefore not precluded by Atkins from executing him.
The court first held that the parties agreed that the burden to prove intellectual disability was on Smith, pursuant to Grell II, 212 Ariz. at 521, 135 P.3d 696. However, in view of the procedural differences between Grell II and Smith’s case, the court “considered the evidence under the preponderance of evidence standard applicable to Rule 32 proceedings.” The court held that Smith had failed to show that he was entitled to relief under either the clear and convincing evidence standard or the preponderance of evidence standard.
The trial court expressed serious concerns with Dr. Thompson’s perspective. It noted that his view of “mental retardation as a fluid condition responsive to any number of changes in a patient’s environment, nutrition, and physical, mental and emotional health,” was not necessarily consistent “with the definition of mental retardation provided by Arizona law, and the procedures by which mental retardation is to be determined under A.R.S. § 13-703.02.”7 The court further noted that Dr. Thompson placed considerable weight on the 1964 IQ tests and the pre-sentence report’s indication of Smith’s “borderline functioning.” It also observed that Dr. Thompson gave little weight to Smith’s ability to live on his own for 15 years between the time he left school and the murder. The court concluded that Dr. Thompson’s “analysis does not permit a finding, with any degree of accuracy, of Defendant’s level of ‘general intellectual functioning’ either before the age of 18, or in the period 1980-82.” Thus, his evaluation “does not support the conclusion that during the pertinent time period, Defendant was mentally retarded.”
The Superior Court recognized that Smith’s “dysfunctional family and troubled early life undoubtedly affected his circumstances in an adverse way,” and that he likely “has suffered from clinically cognizable conditions probably including a personality disorder.” However, the circumstances “do not point to mental retardation with any degree of certainty.” Based on all the evidence the Superior Court found that:
Defendant, has failed to meet his burden of showing that he was mentally retarded at the time of the offense and trial in this case. There was insufficient evidence from which this Court could find that Defendant exhibited “significantly subaverage general intellectual functioning” during the period of the offense and his trial. While unorthodox and unstable, Defendant’s pre-arrest life did not show “significant impairment in adaptive *1223behavior” existing concurrently with the deficit in general intellectual functioning. In the absence of adequate information concerning the early Otis IQ tests, and in view of the alternative explanations for his early school and social deficits, Defendant failed to show the onset of mental retardation before the age of 18. The Court therefore FINDS that the State of Arizona is not precluded, on Atkins grounds from executing Defendant.
5. The Arizona Court of Appeals’ Opinion
Smith appealed to the Arizona Court of Appeals, which unanimously affirmed the Superior Court. Smith v. Kearney, No. 2CA-SA-2008-0019, 2008 WL 2721155 (Ariz.App. Jul. 11, 2008). It noted that the trial court “had considered the evidence under the applicable clear-and-convincing standard as well as under the lesser burden of a preponderance of the evidence standard that applies to post-conviction proceedings.” The appellate court reviewed the evidence and the Superior Court’s decision noting that the trial court had found: (a) Dr. Thompson’s opinion was based on an approach to defining mental retardation that was inconsistent with the requirements of Arizona law; (b) lay witness Martha Hight’s opinion that Smith was mentally retarded was inconsistent with the testimony of witnesses who had lived with Smith in the 1970’s; and (c) Smith’s own written statements were “lengthy, neatly written, logical, detailed, structured and coherent.” The appellate court concluded that the Superior Court had carefully considered all the evidence and “exercised its discretion in resolving conflicts in the evidence, in assessing the reliability of the test results, and credibility of the witnesses, and in weighing evidence.” The court concluded that it had no basis for interfering with the Superior Court’s discretionary judgments or for reweighing the evidence.8
6. The District Court’s Order Denying Habeas Relief
Following the conclusion of his state court proceedings, Smith renewed his proceedings in the United States District Court for the District of Arizona. In a 21-page order issued on December 3, 2012, the court found that Smith’s Atkins related claims were without merit.
The court recognized that because Smith filed his initial federal habeas petition pri- or to AEDPA’s effective- date, pre-AEDPA standards applied. Accordingly, the court reviewed de novo mixed questions of law and fact as well as pure questions of law. See Robinson v. Schriro, 595 F.3d 1086, 1099 (9th Cir.2010). However, the court held, citing Robinson, that the state court factual findings were entitled to a presumption of correctness, subject to eight exceptions enumerated in the previous version of 28 U.S.C. § 2254(d).9 The district *1224court, citing Marshall v. Lonberger, 459 U.S. 422, 432, 103 S.Ct. 843, 74 L.Ed.2d 646 (1983), held that before it could reject a state court’s factual determination it would have to conclude “that the state court’s findings lacked even fair support in the' record.” It further determined that whether Smith “is mentally retarded is a question of fact.”
The district court noted that Smith did not dispute that the state court’s finding as to intellectual disability was entitled to a presumption of correctness.10 The district court concluded that there was ample evidence in the record to support the state court’s conclusion that Smith had failed to establish either “subaverage general intellectual functioning” or “significant impairment in adaptive behavior” before the age of 18. In particular, the district court agreed that the 1964 Otis tests were unreliable and entitled to little weight. It noted that the record was “devoid of any evidence concerning the testing, including the raw test data, identification of the administrators and their qualifications, or the protocols followed.” Moreover, both Drs. Thompson and Martinez testified that the Otis test was outdated when administered to Smith in 1964 and would not be used today to determine an individual’s IQ.
The district court further commented:
[A]s noted by the state court, the evidence presented at the hearing indicated it is just as likely that Petitioner’s poor school performance and unstable lifestyle was the result of a severely dysfunctional upbringing and personality disorder as it was mental retardation. Petitioner led a transient lifestyle, frequently changing employment and residences, but many of his jobs (such as being a mechanic, a cable installer, and a truck driver) required at least a minimal degree of intellectual functioning. Although witnesses agreed that Petitioner is not “book smart,” he learns quickly when shown how to do something. In addition, the record supports a finding that Petitioner began living an independent life after dropping out of school at age 16 and was not dependent on others to function in his daily life.
The district court concluded that Smith had not overcome the presumption of correctness that attached to the state court’s finding that he was not intellectually disabled at the time of the offense.
*1225B. Analysis
The majority does not question the adequacy of the state court’s proceedings. Instead, invoking the eighth exception enumerated under the 1994 version of § 2254(d), it concludes that the state court’s findings are “not fairly supported by the record.” Op. 1181. This is simply wrong. An objective review of the conflicting evidence reveals that there is substantial un-refuted evidence supporting the state court’s determination that Smith was not intellectually disabled in 1980. The majority’s preference for Dr. Thompson’s perspective does not justify its setting aside all the evidence — including portions of Dr. Thompson’s testimony — that supports the determinations by the state court and the district court that Smith has not demonstrated that he was intellectually disabled in 1980.
First, the majority cannot deny that as of 2005, Smith was not intellectually disabled. Even Dr. Thompson’s test indicated that Smith had an IQ of 93. Thus, this case is relatively unique in that the courts are required to determine whether a person who now is clearly not intellectually disabled, was intellectually disabled some 25 years earlier when he committed a murder and was tried. The fact that Smith was not intellectually disabled in 2005, gives rise to at least a presumption that Smith was not intellectually disabled in 1980.
Second, the evidence supporting the majority’s determination is inherently problematic. Smith’s 1964 IQ tests, his poor performance on the Stanford achievement tests, and his poor grades could be signs of intellectual disability. But the test administered in 1964 was out-moded and there is nothing in the record as to how the test was administered. Also, by the time that the test was administered, Smith had failed in school, and there is nothing to suggest that he made any effort to perform well on the test.11 Moreover, there is substantial evidence that Smith had difficulty with the written word, which indicates that his written test results were likely to underrate his intelligence.
Other than the test scores and his academic performance, the evidence of Smith’s alleged intellectual disability is primarily the testimony of lay witness Martha Hight who compared Smith to her sister who had been diagnosed as intellectually disabled. However, the trial court found that her statement “was inconsistent with the testimony of others who lived with [Smith] at or near the same period of time in the 1970’s.”
There was evidence, as the majority notes, of Smith’s horrendous childhood. That his stepfather frequently belittled him and beat him. That his mother frequently ignored her children and was promiscuous in front of them. See Op. 1187-88. This upbringing, the majority notes, led Dr. Thompson to opine that Smith “became intellectually disabled with frontal lobe abnormalities.” Op. 1187.
But it is Dr. Thompson’s view of intellectual disability “as a fluid condition responsive to any number of changes in a patient’s environment, nutrition and physical, mental and emotional health” that renders his diagnosis problematic. If a person’s mental ability is fluid, if it can change in response to changes in the person’s environment, nutrition, and physical, mental and emotional health, then by definition, even assuming that Smith was intellectually disabled in 1964, when he was 16 years *1226old, he was not necessarily disabled in 1980, when he committed the murder.
Critically, by 1980, Smith had lived independently for 15 years, had been married a number of times, and had held numerous jobs. This inherently raises questions as to whether, assuming that Smith was intellectually disabled in 1964, the subsequent improvement of his mental ability was due to his living alone away from his oppressive family for 15 years, as Dr. Martinez suggests, or to his being in prison from 1980 to 2005, as Dr. Thompson suggests. Smith’s independent life from 1964 to 1980 is strong circumstantial evidence that by 1980 he was not intellectually disabled, even if he had been intellectually disabled in 1964.
In addition, Dr. Thompson’s perception of intellectual disability is in tension with Arizona’s definition. Arizona’s statute assumes that a person’s intellectual ability is relatively stable. See Ariz.Rev.Stat. Ann. § 13-703.02(K)(3). In fact, the statute requires that the onset of intellectual disability be before an individual is 18 years of age. Dr. Thompson’s approach begs the question of whether Smith was intellectually disabled as defined by Arizona when he was 16, or only suffered from “frontal lobe abnormalities” that affected his performance but cleared up once he was in a less toxic environment. It is not clear that the alleged “frontal lobe abnormalities” resulted in “significant subaverage intellectual functioning” or caused “significant impairment of adaptive behavior” as required by the Arizona statute. See A.R.S. § 13-703.02(E).
In any event, the critical issue here is not whether Smith was intellectually disabled in 1964, but in 1980. Dr. Martinez’s perspective that Smith was never disabled is certainly supported by the 2005 IQ tests. Dr. Thompson’s suggestion that Smith was disabled in 1980 depends first on a determination that Smith was disabled in 1964 and second on the acceptance that prison rather than 15 years of living alone, explains Smith’s present IQ level. However, as noted, Smith’s ability to live on his own for 15 years from 1964 to 1980 is strong evidence that even if Smith had developed “frontal lobe abnormalities” as a result of his horrendous childhood, they had dissipated by 1980. Moreover, this conclusion is supported by the fact that the two doctors who examined Smith for competency to stand trial for the murder in 1980 failed to detect any signs of intellectual disability. Even Dr. Thompson admitted that he would have expected the doctors to note some sign of intellectual disability.
In sum, there is substantial — if not overwhelming — evidence to support the state court’s determination that Smith had failed to demonstrate that he was intellectually disabled in 1980. Indeed, the majority does not really try to refute this evidence. Instead, it explains at length why it prefers Dr. Thompson’s perspective to that of Dr. Martinez. But that is not the proper inquiry. The question is whether the state court’s factual determination is “fairly supported by the record.” Marshall, 459 U.S. at 432, 103 S.Ct. 843. Perhaps if all the conflicting evidence could be explained away, the majority’s approach might be acceptable. But the evidence remains obstinate in support of the state court’s determinations: Smith is not now intellectually disabled, he lived independently for 15 years before he committed the murder, and the doctors who examined him for competency in 1980 failed to detect any signs of intellectual disability. Moreover, although Dr. Thompson offers an explanation for how Smith’s mental ability could change over time, his own theory cannot pinpoint when Smith overcame his alleged initial intellectual disability. The factual record fully supports the state court’s determinations that Smith failed to carry his *1227burden, and that, accordingly, he was not intellectually disabled in 1980.
In addition, the Supreme Court’s recent opinion in Hall v. Florida, — U.S.—, 134 S.Ct. 1986, 188 L.Ed.2d 1007 (2014), supports the denial of relief. The Supreme Court was critical of Florida’s over-reliance on the measurement of an IQ test.12 It concluded that “[i]ntellectual disability is a condition, not a number.” 134 S.Ct. at 2001. It held that courts “must recognize, as does the medical community, that the IQ test is imprecise.”13 Id. The Supreme Court concluded:
Florida’s rule is in direct opposition to the views of those who design, administer, and interpret the IQ test. By failing to take into account the standard error of measurement, Florida’s law not only contradicts the test’s own design but also bars an essential part of a sentencing court’s inquiry into adaptive functioning. Freddie Lee Hall may or may not be intellectually disabled, but the law requires that he have the opportunity to present evidence of his intellectual disability, including deficits in adaptive functioning over his lifetime.
Id.
Here, Smith had precisely this opportunity. He had a full “opportunity to present evidence of his intellectual disability, including deficits in adaptive functioning over his lifetime.” Id. However, the state courts and the district court concluded that the evidence did not show that he was intellectually disabled in 1980. It is the majority and Dr. Thompson who cling to the 1964 test results. But other evidence such as Smith’s 2005 IQ test results, his living independently on his own for 15 years before the murder, and the failure of the doctors who examined Smith for mental competence in 1980 to detect any sign of intellectual disability, strongly support the state court’s determination.14
Accordingly, because the district court’s denial of Smith’s petition should be affirmed, I dissent from the majority’s opinion.
*1228II. The Arizona Courts Applied the Appropriate Standard of Proof.
A. The Superior Court’s Use of the Words “with any degree of certainty” Does Not in Any Way Suggest that it Applied an Inappropriate Standard of Proof.
Despite the lack of any concurrence, Judge Reinhardt includes in his opinion an argument that the Arizona courts applied an unconstitutional standard of proof. Section II C.2, pages 42-52. Accordingly, I offer the following rebuttal to his inaccurate accusations.
1. The State Court and District Court Decisions
As noted, the Arizona Superior Court held extensive hearings and admitted considerable evidence as to whether Smith was intellectually disabled at the time of his trial. It agreed with the parties that the burden to prove intellectual disability was on Smith, pursuant to Grell II, 212 Ariz. 516, 135 P.3d 696. However, in view of the procedural differences between Grell II and Smith’s case, the court “considered the evidence under the preponderance of evidence standard applicable to Rule 32 proceedings.” The court held that its decision was the same under this lower standard.
After carefully considering all the evidence, the Superior Court concluded:
Although Defendant’s dysfunctional family and troubled early life undoubtedly affected his circumstances in an adverse way, and while it is likely Defendant has suffered from clinically cognizable conditions probably including a personality disorder, the circumstances described at the hearing do not point to mental retardation with any degree of certainty. The Court has carefully considered all of the testimony presented at the hearing, and has reviewed and considered all of the exhibits received in evidence at that proceeding. Based on all of the evidence, the Court FINDS that Defendant has failed to meet his burden of showing that he was mentally retarded at the time of the offense and trial in this case. There was insufficient evidence from which this Court could find that Defendant exhibited “significant subaverage general intellectual functioning” during the period of the offense and his trial. While unorthodox and unstable, Defendant’s pre-arrest life did not show “significant impairment in adaptive behavior” existing concurrently with the deficit in general intellectual functioning. In the absence of adequate information concerning the early Otis IQ tests, and in view of the alternative explanations for his early school and social deficits, Defendant failed to show the onset of mental retardation before the age of 18. The Court therefore FINDS that the State of Arizona is not precluded, on Adkins grounds, from executing Defendant.
The Arizona Court of Appeals affirmed the Superior Court. It noted that the trial court “had considered the evidence under the applicable clear-and-convincing evidence standard as well as under the lesser burden of a preponderance of the evidence that applies to post conviction proceedings ... and concluded that under either standard Smith had failed to establish he was mentally retarded at the time of the offense and at trial.”
Similarly, the district court denied Smith’s petition. It noted that although Smith did not “identify deficiencies in the state court’s ruling,” he contended that his proffered evidence “overwhelmingly established the subaverage intellectual functioning and adaptive skills prongs of Arizona’s mental retardation test as of the time of the offense in 1980.” The district court rejected this contention. It noted that *1229Smith did not dispute that the 1964 Otis tests were unreliable, and commented that “the evidence presented at the hearing indicated it is just as likely that Petitioner’s poor school performance and unstable lifestyle was the result of a severely dysfunctional upbringing and personality disorder as it was mental retardation.” The district court concluded that there was “ample evidence in the record to support the state courts’ conclusion that Petitioner failed to establish either ‘subaverage general intellectual functioning’ or ‘significant impairment in adaptive behavior’ before the age of 18.” Accordingly, he had “not overcome the presumption of correctness attached to the state court’s finding that he was not mentally retarded.” Moreover, in denying a certificate of appealability the district court found “that reasonable jurists could not debate its resolution of Petitioner’s Ai/fcms-related claims,” and that “[t]he question of whether the state court erred in finding that Petitioner was not mentally retarded under Arizona law is not debatable among jurists of reason.”
2. Discussion
In light of the unanimous perspective of the Arizona trial and appellate courts and the district court, how does Judge Reinhardt conclude that the Superior Court applied an unconstitutional standard of proof? His concurrence does so by first disbelieving the state courts’ statements that the Superior Court applied the preponderance of the evidence standard. Second, the concurrence ignores the court’s factual findings and misconstrues the Superior Court’s statement that “the circumstances described at the hearing do not point to mental retardation with any degree of certainty.” Nee Op. 1197. Thus, the concurrence takes five words from the trial court’s decision out of context and then gives them an improper definition. By attacking this incorrect -definition, the concurrence, in essence, argues that the death penalty cannot be constitutionally applied.15
Both the Superior Court and the Arizona Court of Appeals stated that the Superior Court applied the lesser preponderance of the evidence standard. The concurrence dismisses their considered opinions in a footnote arguing that the body of the Superior Court’s opinion “did not fulfill that promise, however, but, rather, the court concluded after reviewing all the evidence that it did not meet a ‘certainty’ standard.” Op. 1196-97 n.25. This is wrong on a number of levels.
The concurrence takes “with any degree of certainty” out of context, endows it with a incorrect meaning and then argues that the stilted meaning it has conjured up is unconstitutional. But, considering the factual evidence in this case, an objective jurist must admit to the lack of some precision in an evaluation of Smith’s intellectual ability in 1980. On the one hand, there are the Otis IQ tests from 1964 and Dr. Thompson’s testimony that Smith’s intellectual disability was not a constant. On the other hand, both Dr. Thompson and Dr. Martinez agreed that by 2005 Smith was not intellectually disabled, and there is evidence that for some 15 years after dropping out of school, and before committing the murder, Smith lived independently and was not dependent on anyone. Thus, given that Smith had the burden to prove an intellectual disability, the Superior Court reasonably concluded that he had failed to do so, even by a preponderance of the evidence. In other words, the evidence as presented by Smith did not *1230“point to mental retardation with any degree of certainty.” Both the Arizona Court of Appeals and the District Court agreed.
But the concurrence eschews the trial court’s intent and suggests that “the ‘any degree of certainty’ standard ... is more akin to the ‘reasonable doubt’ standard than the clear and convincing standard mandated by Arizona’s Atkins statute, which requires only that the issue under consideration be ‘highly probable.’ ”16 Op. 1197. In support of this assertion, the concurrence cites a 27-year old Arizona case that involved a jury instruction. State v. King, 158 Ariz. 419, 763 P.2d 239 (1988). This case, which affirmed placing the burden on the defendant to prove insanity, disapproved defining “clear and convincing evidence” as evidence that “is certain, plain to the understanding, unambiguous, and convincing in the sense that it is so reasonable and persuasive as to cause you to believe it.” Id. at 241. The court held that “the better instruction would inform a jury that clear and convincing evidence is evidence that makes the existence of the issue propounded highly probable.” Id. at 244.
In Smith’s proceedings, no court used the definition of clear and convincing evidence disapproved in King. Indeed, our precedent requires that we presume that the state judges know and follow the law. See Lopez v. Schriro, 491 F.3d 1029, 1043 (9th Cir.2007). We have further held that we should not lightly disregard the trial court’s determinations. Id. In Parker v. Dugger, 498 U.S. 308, 315, 111 S.Ct. 731, 112 L.Ed.2d 812 (1991), the Supreme Court held “[w]e must assume that the trial judge considered all this [mitigation] evidence before passing sentence. For one thing, he said he did.” Thus, we have no basis for finding that the Superior Court did not apply the preponderance of the evidence standard that it said it did (and which the Arizona Court of Appeals affirmed) or that the Superior Court somehow applied “certainty” in a way that has been improper in Arizona since King was decided in 1988.
Furthermore, even if there were some ambiguity in the Superior Court’s decision — which there is not — we would still have to construe any ambiguity in the language in the state court’s favor. See Woodford v. Visciotti 537 U.S. 19, 24, 123 S.Ct. 357, 154 L.Ed.2d 279 (2002) (“This readiness to attribute error is inconsistent with the presumption that state courts know and follow the law”).17 Indeed, the Supreme Court has indicated that we should not “demand a formulary statement” by state courts. Early v. Packer, 537 U.S. 3, 9, 123 S.Ct. 362, 154 L.Ed.2d 263 (2002) (per curiam) (an AEDPA case, but citing a pre-AEDPA decision, Lowenfield v. Phelps, 484 U.S. 231, 108 S.Ct. 546, 98 L.Ed.2d 568 (1988), in support of its admonition). Thus, the concurrence’s approach goes against both Supreme Court and Ninth Circuit case law when it not only construes the trial court’s clear language as ambiguous, but then interprets the ambiguity as reflecting an unconstitutional standard.
The concurrence then proceeds to argue that Atkins proceedings are different. “Consequently, where a state court analyz*1231ing an Atkins claim fails to follow binding state law, its decision does not simply violate state law, but also violates the Eighth Amendment right provided by Atkins and the violation is therefore cognizable by a federal habeas court.” Op. 1197. Again relying on its stilted definition of “certainty,” the concurrence asserts:
Here, the “certainty” standard applied by the state trial court was plainly contrary to the clear and convincing standard required by Arizona’s statute and adopted by its supreme court. See Ariz. Rev.Stat. Ann. § 13-703.02(G); Grell II, 135 P.3d at 701 (“The statute places on ‘the defendant ... the burden of proving mental retardation by clear and convincing evidence’ in the pretrial hearing.” (quoting § 13-703.02(G)) (alteration in original)). Accordingly, the standard of proof applied by the state trial court was not simply contrary to state law but was also unconstitutional under Atkins, see Williams, 792 F.3d 606, 612; Black, 664 F.3d at 97, and, accordingly, the state court’s findings are not due any deference. See Lafferty, 949 F.2d at 1551 n. 4; Walker, 167 F.3d at 1345.
But this is not right. First, as noted, the state court applied the less demanding preponderance of the evidence test. Second, there is nothing in either the trial court’s decision or the state appellate court’s memorandum disposition that suggests that either court defined “certainty” in a way that violated King, 763 P.2d 239. Third, Grell II, which the majority cites, affirms that Smith had the burden of proving mental retardation by clear and convincing evidence. Fourth, a review of the record in this case fully supports the Superior Court’s factual determination that Smith had failed either by clear and convincing evidence or a preponderance of the evidence to show that he was intellectually disabled at the time of the crime and his trial. Cherry-picking words from the trial court’s decision and then giving them an incorrect meaning does not undermine the clear logic of the decision as affirmed by the state appellate court.
Perhaps aware that its first argument is less than persuasive, the concurrence offers a second argument: that “the standard of proof applied by the state trial court is unconstitutional.” Op. 1198. Again, based largely on its incorrect definition of “certainty,” the concurrence asserts that “a ‘certainty’ standard of proof transgresses the limits of the state’s authority to craft appropriate procedures to enforce Atkins and, in doing so, encroaches on the substantive constitutional right.” Op. 1198.
This argument appears, in essence, to be an argument against the constitutionality of the death penalty. The majority claims that it does not “need to determine what standard of proof the federal Constitution requires,” but “only whether the Arizona court applied a standard it forbids.” Op. 1198. It posits that “[w]hen the natural operation of a state’s procedures for rendering factual determinations transgresses a substantive constitutional right, those procedures are unconstitutional.” Op. 1198. The concurrence argues that “[i]t is elementary that the ‘natural operation’ of applying a heightened standard of proof can determine the outcome of litigation, and thus the availability of a constitutional right.”18 Op. 1198.
The concurrence next objects to the death penalty based on the “inherent imprecision of psychiatric determinations of mental illness.” Op. 1199. Citing Addington v. Texas, 441 U.S. 418, 430-32, 99 S.Ct. 1804, 60 L.Ed.2d 323 (1979), it comments: *1232“[a]s the Supreme Court explained in rejecting the argument that the Constitution requires use of a reasonable doubt standard in the context of civil commitment proceedings, the unique nature of psychiatric diagnosis renders factual determinations uniquely unsusceptible to certainty.” Op. 1199. The concurrence posits that when, as in this case, the determination of an individual’s intellectual ability at the time of the crime is not made until years after the crime, certainty is “even less attainable and a certainty standard is even less constitutionally acceptable in such cases.”19 Op. 1200.
The concurrence concludes with the assertion that capital punishment requires a “heightened degree of certainty.” Op 1200. Accordingly, “where, as in Atkins, the Eighth Amendment renders a class of individuals categorically ineligible for execution, the procedures used to determine whether a defendant falls into that class may not allocate nearly all of the risk of an erroneous determination to the defendant.” Op. 1201. It reasons that by requiring Smith to demonstrate with a “degree of certainty” that he was intellectually disabled, the Arizona court allocated “nearly the entire risk of an erroneous determination to Smith.” Op. 1201. Thus, according to the concurrence, because “the factual determination in question concerned an issue for which certainty may be unattainable ... and a penalty for which a greater degree of reliability is required ... [,] the constitutional violation [is] even more clear.” Op. 1201.
Judge Reinhardt is certainly entitled to his opinion, but it is not the opinion of the panel or of the Ninth Circuit. The concurrence is clearly contrary to the position of the Arizona Supreme Court. State v. Grell (Grell III), 212 Ariz. 516, 135 P.3d 696, 702 (2006) (en banc) (“We find no constitutional bar to imposing the burden of proving mental retardation on the defendant.”). The concurrence cites no Ninth Circuit case to support its perspective. And it is not supported by any decision by the United States Supreme Court, which denied certiorari in Grell II (Grell v. Arizona, 550 U.S. 937, 127 S.Ct. 2246, 167 L.Ed.2d 1095 (2007)), and accepted certain provisions of Arizona’s proceedings in Hall, 134 S.Ct. at 1996-97.
In sum, the assertion that the Arizona courts applied an unconstitutional standard of proof fails first because it misconstrues the Arizona Superior Court’s decision, and ignores the ruling of the Arizona Court of Appeals, contrary to both Supreme Court and Ninth Circuit precedent. See Woodford, 537 U.S. at 24,123 S.Ct. 357; Parker, 498 U.S. at 315, 111 S.Ct. 731; Lopez, 491 F.3d at 1043 (stating “there are ways to construe the state court’s ruling that would not make it ‘clearly untenable,’ and we are therefore bound by the state courts’ inter*1233pretation and application of its own procedural rules.”). The second arrow in the concurrence’s quiver — that there is not sufficient certainty to impose the death penalty on Smith — similarly lacks support from either the Supreme Court or the Ninth Circuit, as Arizona’s placement of the burden on the defendant to prove his intellectual disability has not been disturbed in the eight years that have passed since the Arizona Supreme Court decided Grell II in 2007.
III. Conclusion
The district court’s denial of the Smith’s petition should be affirmed because an objective review of the extensive record reveals that there is substantial evidence, if not overwhelming evidence, that Smith failed to meet his burden of showing that he was intellectually disabled in 1980-82, when he murdered Sandy Owen and was tried, convicted and sentenced. At a minimum, this conclusion is compelled by the undeniable facts that: (a) Smith in 2005 had an IQ between 87 and 98; (b) Smith lived independently and supported himself for 15 years after he dropped out of school and before the murder; and (c) the doctors who examined Smith in 1980 to determine his competency to be tried found no signs of intellectual disability. Furthermore, Dr. Thompson’s approach of considering intellectual ability to be fluid, while allowing for Smith’s alleged intellectual disability to dissipate, offers no assurance as to when it did so. The state courts took Smith’s claim of intellectual disability seriously, and gave his assertions and the evidence full consideration. An objective review of this record will not support a finding — and certainly not a finding by this court on review of a state habeas petition — that Smith met his burden of showing that he was intellectually disabled in 1980. I would affirm the district court’s denial of Smith’s petition20 and I therefore dissent.
. As does the majority opinion, I use the term "intellectually disabled” rather than "mentally retarded” except where the term is used in quoted material.
. Under pre-AEDPA law:
We review the district court's decision to grant habeas relief de novo. We review de novo questions of law and mixed questions of law and fact, whether decided by the district court or the state courts. The district court’s factual findings are reviewed for clear error. We therefore accept its findings "absent a definite and firm conviction that a mistake has been committed.” State court factual findings are entitled to a presumption of correctness, subject to eight exceptions enumerated in the previous version of 28 U.S.C. § 2254(d).
Sivak v. Hardison, 658 F.3d 898, 905-06 (9th Cir.2011) (internal citations omitted).
. Lebrecque further noted that because of Smith's "lack of basic grade-school academic skills, and his short stature, Lebrecque initially thought his maturity level was that of a 12 to 14 year old.”
. The state court further explained:
The pre-sentence report contains detailed descriptions of the offense, including a recitation of information obtained from written statements of Defendant to the court and his statements to law enforcement officers. Information concerning Defendant's social, marital, educational, religious, and employment history, was also apparently obtained primarily from Defendant. The pre-sen-tence report does not mention any difficulty in obtaining this information from Defendant. The section entitled "Physical and Mental Health,” while noting Defendant's depression, "veiy poor self concept,” sexual issues, drug abuse and early psychiatric treatment, did not mention mental retardation. The references to retardation were two. In the "education” section, the pre-sentence report writer stated: "Appended records indicate that intelligence tests administered during his eighth grade year revealed an IQ of 71, indicating that he is borderline but educable.” The “Evaluation Summary” includes the following sentence: "His borderline mentality probably makes him an easy person to manipulate and somewhat of a follower in social situations.” Defendant’s so-called "borderline mentality” is not mentioned as a mitigating circumstance — in fact, the pre-sentence report noted that, "In view of the defendant’s known history and the circumstances of the instant offense, the Court may feel that there are no applicable mitigating circumstances.” It is at least as likely that the “borderline” language in the Evaluation Summary section simply reflected the pre-sentence report writer’s knowledge of the 1964 IQ test referenced in the school records, as that the pre-sentence report writer based the comment on an analysis of Defendant’s history and characteristics grounded in the appropriate expertise. There simply is no way to know.
.Smith apparently complained of memory problems, but there was no evaluation for organic brain syndrome or seizure disorder.
. In reference to Smith’s prior 1970 hospitalization and diagnosis for personality disorder with psychotic features, Dr. Thompson "acknowledged that this disorder included features described as 'inadequate and immature,' and that this condition included anxiety and depression which could display the impulsivity causing the job changes and relationship issues characterizing Defendant's early adulthood.”
. The state court commented:
The State ... contends that the family dysfunction and abuse, faulty nutrition, depression and anxiety rather than mental retardation contributed to the low test scores, low grades and other signs. In other words, the defense view is that Defendant's early difficulties cause his retardation, and that he got better in prison. The prosecution’s perspective is that Defendant’s dysfunctional background and other mental health problems rather than mental retardation caused the factors pointed to by Dr. Thompson as diagnostic of retardation.
. The Arizona Supreme Court summarily denied Smith’s petition for review.
. The district court listed the exceptions as:
(1) that the merits of the factual dispute were not resolved in the State court hearing;
(2) that the factfinding procedure employed by the State court was not adequate to afford a full and fair hearing;
(3) that the material facts were not adequately developed at the State court hearing;
(4) that the State court lacked jurisdiction of the subject matter or over the person of the applicant in the State court proceeding;
(5) that the applicant was an indigent and the State court, in deprivation of his constitutional right, failed to appoint counsel to represent him in the State court proceeding;
(6) that the applicant did not receive a full, fair and adequate hearing in the State court proceeding; or
*1224(7) that the applicant was otherwise denied due process of law in the State court proceeding;
(8) or unless ... the Federal court on consideration of [the relevant] part of the record as a whole concludes that such factual determination is not fairly supported by the record.
28 U.S.C. § 2254(d) (1994).
. Smith nonetheless sought de novo review for at least one of four grounds: "(1) an inadequate factfinding procedure by the state court; (2) the failure to adequately develop material facts at the state court Atkins hearing; (3) the failure to provide Petitioner a full, fair and adequate hearing; and (4) a violation of Petitioner's due process rights.” The district court rejected Smith's claim to a right to funding for a Positron Emission Tomography (PET) scan and his assertion that he was entitled to jury determination on mental retardation. The majority opinion does not discuss either of these claims. Rejecting Smith’s other grounds, the district court noted that the "state court provided a lengthy period of time for Petitioner to prepare for the Atkins hearing and authorized Petitioner's expert of choice, his initial diagnostic testing requests, investigative resources, and numerous depositions of lay witnesses.” It noted that Smith "identifies nothing to support a finding that the state court's factfinding procedures were inadequate, that material facts were left undeveloped, that the state court failed to provide a full and fair hearing, or that his due process rights were violated. Accordingly, the district court found that Smith had not overcome the presumption of correctness on any of these grounds.
. Dr. Martinez testified as to the importance of an awareness of an individual’s behavior during an assessment. He noted that in a group-administered test you don't have the ability to "directly assess how the individual is doing, whether they’re paying attention or not.”
. The Court explained:
Florida’s rule disregards established medical practice in two interrelated ways. It takes an IQ score as final and conclusive evidence of a defendant’s intellectual capacity, when experts in the field would consider other evidence. It also relies on a purportedly scientific measurement of the defendant’s abilities, his IQ score, while refusing to recognize that the score is, on its own terms, imprecise.
134 S.Ct. at 1995.
. The Court continued:
This is not to say that an IQ test score is unhelpful. It is of considerable significance, as the medical community recognizes. But in using these scores to assess a defendant's eligibility for the death penalty, a State must afford these test scores the same studied skepticism that those who design and use the tests do, and understand that an IQ test score represents a range rather than a fixed number.
134 S.Ct. at 2001.
.Finally, it should be noted that the Supreme Court in Hall considered the Arizona’s statute and suggested that it passed constitutional muster.
Arizona’s statute appears to set a broad statutory cutoff at 70, Ariz.Rev.Stat. Ann. § 13-753 (F) (West 2013), but another provision instructs courts to “take into account the margin of error for a test administered.” Id. at § 14-753(K)(5). How courts are meant to interpret the statute in a situation like Hall’s is not altogether clear. The principal Arizona case on the matter, State v. Roque, 213 Ariz. 193, 141 P.3d 368, (2006), states that "the statute accounts for margin of error by requiring multiple tests,” and that "if the defendant achieves a full-scale score of 70 or below on any one of the tests, then the court proceeds to a hearing.” Id. at 403.
134 S.Ct. at 1996.
. Although Judge Reinhardt specifically addresses the constitutionality of the Arizona death penalty statute in his separate concurrence, its spirit clearly informs his concurrence set forth in the majority opinion.
. Of course, this argument is only relevant if one ignores the trial court’s determination, affirmed by the state court of appeals, that Smith had failed to demonstrate an intellectual disability by the preponderance of the evidence.
. Although Woodford concerned review under AEDPA, the Supreme court indicated, citing Parker, 498 U.S. at 314-16, 111 S.Ct. 731, and other cases, that the presumption that state courts know and follow the law was established before the enactment of AEDPA and thus applies to pre-AEDPA cases. Wood-ford, 537 U.S. at 24, 123 S.Ct. 357.
. The concurrence's authority for these assertions is a 1911 Supreme Court opinion, Bailey v. Alabama, 219 U.S. 219, 31 S.Ct. 145, 55 L.Ed. 191 (1911), which held that an Alabama statute that created a presumption of intent to injure violated the 13th Amendment.
. In footnote 26, the concurrence asserts:
It is of no consequence to the analysis that Addington and Atkins involve different burdens of proof than the case at bar, because the focus here is on the effect of the standard of proof. Under Addington, a state desiring the civil commitment of an individual must demonstrate that he suffers from mental illness, whereas under Atkins an individual seeking to avoid execution by the state must demonstrate intellectual disability. In both situations, the determination heavily relies upon psychiatric opinion, and thus in both situations a standard of proof requiring "any degree of certainty” as defined by Arizona law will often render it impossible for a party to carry its burden. See Addington, 441 U.S. at 432, 99 S.Ct. 1804.
Again, this argument is based on the concurrence’s misinterpretation of the trial court's decision. Deciding whether a person has carried his burden of showing an intellectual disability may well be difficult, but here the state court carefully did so. Indeed, it is telling that the majority attacks the decision by giving the words "any degree of certainty” a meaning that they do not have.
. I would also affirm the district court’s holding that Smith has failed to demonstrate cause to overcome his procedural default of his ineffective assistance of counsel claim. His invocation of Martinez v. Ryan, -U.S. -, 132 S.Ct. 1309, 182 L.Ed.2d 272 (2012) fails because he has not demonstrated ineffective assistance of counsel in his first state post-conviction proceeding, and even if he did, this would not excuse Smith's new counsel from raising ineffective assistance of trial counsel in Smith's second state post-conviction proceeding. Moreover, Smith has not shown a reasonable probability that he received ineffective assistance of counsel at sentencing. See Strickland v. Washington, 466 U.S. 668, 104 S.Ct. 2052, 80 L.Ed.2d 674 (1984). Smith's other arguments are unavailing. He fails to demonstrate the existence of a conflict of interest pervading the Pima County Public Defender’s Office; the state habeas court’s denial of a psychological exam does not provide cause for his procedural default; and to the extent that his claim was not covered by Stewart v. Smith, 536 U.S. 856, 860-61, 122 S.Ct. 2578, 153 L.Ed.2d 762 (2002), the state court’s application of its post-conviction waiver rule is adequate and independent.