Mark Stephen ELLIS, Petitioner-Appellee, v. Rick RAEMISCH, Executive Director, Colorado Department of Corrections; Cynthia Coffman, Attorney General, State of Colorado, Respondents-Appellants, 872 F.3d 1064

ORDER This matter is before the court on the appellee’s Petition for Panel Rehearing and/or Rehearing En Banc. We also have a response from the appellants. Upon consideration, the original panel grants panel rehearing in part and only to the extent of the changes made to page 1071-72 and footnote 2 of the attached revised opinion. The clerk is directed to file the revised decision nunc pro tunc to the original filing date of May 11, 2017. The Petition for Panel Rehearing and/or Rehearing En Banc and the revised opinion were also transmitted to all the judges of the court in regular active service.

As no judge on the original panel or the en banc court called for a poll, the request for rehearing en banc is denied. HOLMES, Circuit Judge. A jury convicted Mark Ellis of five felony offenses and one misdemeanor offense involving child sexual assault on his adopted daughter, V.E. Child sexual assault allegations against Mr. Ellis first arose during his contentious divorce from V.E.’s mother. At trial, defense counsel Rowe Stayton argued that Mr.

Ellis had been falsely accused; specifically, he contended that V.E.’s vengeful mother was coaching her, and that V.E.’s sexual knowledge came only from admitted sexual abuse by her older brother. After he was convicted, Mr. Ellis filed a motion for postconviction relief in Colorado state district court. He alleged that Mr.

Stayton had been constitutionally ineffective for failing to interview and/or call to testify (1) an expert forensic psychologist who could testify about theories of family dynamics and childhood memory, and (2) several lay witnesses who could testify in particular about the Ellises’ family dynamics when the allegations arose. The state district court denied relief. The Colorado Court of Appeals (“CCA”) affirmed. Mr. Ellis never sought review of his ineffective-assistance claim in the Colorado Supreme Court (“CSC”). Mr.

Ellis, now serving an indeterminate life sentence in the Colorado Department of Corrections, filed an application for a writ of habeas corpus pursuant to 28 U.S.C. § 2254 in the United States District Court for the District of Colorado. He alleged ineffective assistance of trial counsel, among other claims. The federal district court determined that Mr. Stayton had been constitutionally ineffective and granted Mr. Ellis conditional habeas relief. First, the court concluded that Mr. Ellis had not failed to exhaust state remedies even though he never sought review of his ineffective-assistance claim in the CSC.

Then, after ruling in Mr. Ellis’s favor on the merits of his ineffective-assistance claim, the district court ordered the Colorado state respondents (“State”) to retry Mr. Ellis within ninety days or be forever barred from pursuing further proceedings on the same charges. The State now appeals from.the federal district court’s grant of habeas relief. The State argues that the district court erred in (1) finding that Mr. Ellis exhausted state remedies; (2) granting federal habeas relief on his ineffective-assistance claim; and (3) barring the State from retrying him, if they do not act to do so within ninety days.

Exercising jurisdiction under 28 U.S.C. § 1291, we reverse the judgment granting conditional habeas relief and remand with instructions to enter judgment denying relief. As we explain below, we conclude that Mr. Ellis adequately exhausted his ineffective-assistance claim, but that the district court erred in granting him conditional habeas relief on that claim. Any question as to the propriety of the district court’s ninety-day retrial condition is effectively moot because we conclude that the district court should not have granted habeas relief in the first place. Therefore, we do not reach this retrial issue.

I A V.E., a foster child, began living with Mark and his then-wife, Kari Ellis, when she was two years old. When she was seven years old, in 1998, Mr. and Ms. Ellis adopted her. The sexual assault of which Mr. Ellis was convicted occurred when V.E. was approximately eight to ten years old, from 1999 to 2001. In 2000, Kari Ellis filed for divorce after learning that her husband was having an affair. During the contentious divorce proceedings that ensued, V.E.’s older brother M.E. told his mother (i.e., Ms. Ellis) that his father had “screwed” V.E. Aplt’s App. Vol. III, set 154. After M.E. told Ms.

Ellis this, she hid a tape-recorder in her purse and asked V.E. whether “she had any secrets to tell [her] about anybody.” Id. at 158. V.E. did not disclose any abuse. Because V.E. was not “telling [Ms. Ellis] anything,” Ms. Ellis asked M.E. to “talk to [V.E.] and tell her it’s okay to be honest with [Ms. Ellis].” Id. at 159. M.E. obliged and talked to V.E. alone. Afterward, Ms. Ellis talked to V.E. again, and this time, V.E. said “something about her dad tying her to the bed,” “put[ting] a buzzer[ ] ... on her neck,” and “put[ting] his hands down her pants a lot.” Id. at 160. Ms.

Ellis immediately reported these statements to the police. More than six months after the police began investigating the possible sexual assault on V.E., lab results revealed semen on one of her blankets. Shortly thereafter, V.E. revealed for the first time that M.E. also had been sexually assaulting her. M.E. pleaded guilty to sexual assault on a child. He later testified, at his father’s trial, that he “first had the idea[ ] [of sexually assaulting V.E.] after she told [him] what [their] father had been doing to her.” Id. at 35 (M.E.’s Trial Test.). At the time of Mr. Ellis’s trial in 2002, Mr.

Stayton had been working as a criminal defense lawyer for nearly twenty years. He specialized in child sexual assault cases and had handled probably “a couple hundred” of them. Aplt.’s App. Vol. VI, at 195, 197 (Stayton’s Test, at Postcon-viction Hr’g). He had also interviewed “dozens, dozens of juries.” Id. In the months leading up to Mr. Ellis’s trial, however, Mr. Stayton “ha[d] a lot of things on [his] plate that required [him] to be out of town.” Id. at 277. First, Mr. Stayton’s mother shot herself, and although she survived, her attempted suicide triggered a family fight over custody.of Mr.

Stayton’s quadriplegic sister. Then, the month before Mr. Ellis’s trial, Mr. Stayton’s wife filed for .divorce. Finally, during the week before and the week after Mr. Ellis’s trial, Mr. Stayton was in trial for other cases. At Mr. Ellis’s trial, Mr. Stayton’s theory of the case was that Ms. Ellis “despise[d]” Mark Ellis and that she “put this hatred over from her into the children.” Aplt.’s App. Vol. II, at 32 (Opening Statement). Mr. Stayton presented this theory primarily through cross-examination of state witnesses.

For example, he elicited from cross-examination of V.E. that she was angry at her father, that she did not like him, and that she felt closer to her mother. In addition, he elicited from cross-examination of V.E.’s eldest sister, Elizabeth Jefferson, that “[t]hese allegations have split the family up,” and that while she-“allied [her]self with [their] mother,” her sister, Jessica Geer, “allied herself with” their father. Aplt’s App. Vol. Ill, at 256 (Ms. Jefferson’s Test.). Furthermore, Mr.

Stayton elicited from cross-examination of M.E. that he was “probably one of the closest children to” his mother, that he had conversations with his mother about his parents’ divorce, and that his mother “was very hurt by the divorce.” Aplt.’s App. Vol. Ill, at 70, 71 (M.E.’s Test.). Mr. Stayton also elicited that M.E. was “angry at [his father] for what he was doing to [their] family,” that he “dislike[d] [his] father a great deal,” and that his sister, Jessica, “being close to her dad is the same as [him] being close to [their] mom.” Id. at 56, 74. And Mr.

Stayton elicited on cross-examination from both Y.E. and M.E. that M.E. had been sexually assaulting V.E. Mr. Stayton then called several witnesses for the defense. One was a forensic scientist who testified that the amount of Mr. Ellis’s semen found on certain blankets and comforters in the Ellises’ house— the only physical evidence in the case— was only “a small percentage of what would come from a human ejaculation,” in “quantities that could be transferred, for example, if ejaculate got onto somebody’s hands ... and you picked up an item.” Aplt’s App. Vol. IV, at 20 (Taylor’s Trial Test.).

Notably, during one bench conference, the trial judge observed that “[t]his is a very, very well fought case on both sides.” Aplh’s App. Vol. III, at 252-53. Nevertheless, the jury convicted Mr. Ellis on all counts. On direct appeal, the CCA affirmed the convictions. See People v. Ellis, 148 P.3d 205 (Colo. App. 2006). The CSC subsequently denied Mr. Ellis’s petition for a writ of certiorari. B In 2007 Mr. Ellis filed a motion for postconviction relief in Colorado state dis-triet court, alleging ineffective assistance of counsel and arguing that newly-discovered evidence warranted a new trial.

Regarding ineffective assistance—the only claim before us now—Mr. Ellis argued that Mr. Stayton was constitutionally ineffective for failing to consult and/or call an expert forensic psychologist to testify about theories of family dynamics and childhood memory; for failing to consult and/or call several lay witnesses who could have supported the defense themes of parental alienation, witness coaching, and collusion; and for committing other trial errors including weak cross-examination, mishandling of prejudicial evidence, and failure to object to improper questioning.

In 2011 the state district court held a three-day evidentiary hearing on Mr. Ellis’s postconviction claims. At the end of the hearing, the court found that— other than “not [being] persuaded by ... the evidence that the victim has affirmatively recanted her testimony”—it “accepted] the testimony of the witnesses presented by the defense as true.” Aplt.’s App. Vol. I, at 268 (State Dist. Ct.’s Ruling & Order). Moreover, the State had called Mr. Stayton as a witness at the hearing, and the court also accepted the truth of his testimony. Mr.

Stayton testified, that he “was adequately prepared for trial,” that he “was all over the facts,” and that he believed he had “adequately represented [his] client at trial.” Aplt’s App. Vol. VI, at '287. Mr. Stayton also testified that he had “attempted to establish throughout [Mr. Ellis’s] trial[] that Kari was really mad at Mark, and that what happened is that [V.E.] had read her mom, and she was more closely aligned with her mom. And ergo we now have allegations.” Id. at 226 (alterations in original). Mr.

Stayton continued that child sexual assault cases come down to whether “the jury believes the child,” and he believed that at Mr. Ellis’s trial, V.E. presented well to the jury. Id. at 229. Notably, Colorado criminal defense attorney and trial-advocacy expert Patrick Mulligan testified at the evidentiary hearing that “the most pressing” problem with Mr. Stayton’s assistance was that he failed to consult and call an expert forensic psychologist “who could have offered ... extremely important testimony.” Aplt.’s App. Vol. VI, at 104, 112. Mr.

Mulligan explained that an expert could have helped the defense prepare “to more appropriately and thoroughly cross-examine prosecution witnesses,” and could have “tak[en] the witness stand and testified] to the jury about some of the concepts that were leading to family dynamics.” Id. at 124-25. In addition, expert forensic psychologist Phillip Esplín testified that “this [was] a case that would have been benefited from an expert in the field of forensic psychology.” Id., Vol.

V, at 142, When asked why he did not call an expert forensic psychologist to explain theories of. family dynamics and childhood memory to the jury, Mr. Stayton testified that he was “well enough familiar with psychological principles [relevant to child sexual assault] cases to prepare [his] cross-examination of the witnesses,” “to prepare [his] presentation of evidence,” and “to prepare both [his] opening statements and closing arguments.” Id., Vol. VI, at 231-32. In fact, Mr.

Stayton testified that he was so familiar with both “parental alienation” theories and “theories on primacy and re-cency and best memory” that he “had taught ... lectures around the country with these theories” to “other defense attorneys.” Id. at 227, 232. He also testified that he believed that he “had adequately brought out the case, the parental alienation,” by eliciting in cross-examination that V.E. was only accusing her father because she was aligned with her mother. Id. at 230 (“I gave them that reason [for V.E.’s accusations], that was mom’s, I had done that in [the state’s] case in chief.”).

In other words, Mr. Stayton “just felt from [his] perception that they [i.e., the defense] were making the points that [they] should.” Id. at 229. And he testified that, if he had called such an expert, he believed that the state would have called its own, and that he did not want to expose the defense case to “a lot of attack.” Id. In this regard, Mr. Stayton expressed the view that, if the defense is going to prevail in child sexual assault cases, it must keep the focus on the weaknesses in the prosecution’s case.

See id. (opining that “if you don’t win these cases as the defense lawyer, in the prosecution case if you don’t win them over there, you’re not going to win on your side”). Moreover, Mr. Stayton testified that he felt that it “would be insulting to the jury to try to point out the parental alienation” because he believed that “it was in front of them, and [he] didn’t think you needed an expert to tell them that[;] they either believed the child or they don’t.” Id. at 230. He. explained: “[T]he theory ... was easy, it’s intuitive.

You don’t have to have a master’s degree to understand that one parent can manipulate a child, you don’t need that.” Id. at 233. At the postconviction evidentiary hearing, Mr. Mulligan also testified that Mr. Stayton was ineffective for failing to call the Ellis children’s special advocate in their parents’ divorce proceeding, Dr. Spiegle, and for failing to introduce letters that Dr. Spiegle had submitted to the divorce court “to support the idea that Kari Ellis was trying to manipulate the children and, in fact, alienate [them from] Mr. Ellis.” Apli’s App. Vol. VI, at 108.

Those letters primarily indicate that Ms. Ellis was obstructing Mark Ellis’s access to their children during the divorce proceedings. When asked why he did not call Dr. Spiegle after endorsing him as a witness, Mr. Stayton responded that “honestly [he] felt that we were doing okay, we didn’t need that.” Id. at 286. Testifying that he “made decisions that [he] really fe[lt] [were] trying to help [his] cliént,” Mr. Stayton remarked that his decision not to call Dr. Spiegle was “part [of his] trial strategy.” Id. at 287. In addition, V.E.’s eldest sister, Elizabeth Jefferson, testified at the postconviction hearing.

She testified that she had perjured herself at trial. Specifically, she testified that when asked at trial whether Ms. Ellis “had been rehearsing and coaching [V.E.],” Ms. Jefferson answered no— even though both she and her mother had “been rehearsing and coaching [YE.] from the very first day this allegation came to light.” Aplt’s App. Vol. V, at 223 (Ms. Jefferson’s Test, at Postconviction Hr’g) (alterations in original). Ms.

Jefferson also testified at the postconviction hearing that at the time of trial, she “wanted to do everything [she] possibly could to help get [her father] put away” because she believed he “was a child molester.” Id. at 224. At the time, Ms. Jefferson said that her knowledge of the molestation came entirely from her mother, who told her things that Ms. Jefferson later discovered were false. Id. at 224-25. Another of V.E.’s sisters, Jessica Geer, also testified at the postconviction hearing. Ms. Geer had not been called to testify at trial.

However, she testified at the postcon-viction hearing that her mother had talked badly about her father in front of her and her siblings, and that she might have witnessed her mother and M.E. planting evidence against her father. Specifically, Ms. Geer testified that she saw her mother and M.E. “with Scotch tape ... going like this on the carpet of the truck [i.e., presumably, pressing the adhesive portion of the tape to the carpet] on the passenger side floorboard,” and that she later learned “that there was an allegation of there being some fiber evidence found in blankets.” Id. at 242-43. Additionally, Ms.

Geer testified that when she expressed her belief that her father was innocent, her mother “threw [her] out of the house,” refused to sign documents that would have allowed her to continue attending school, and attempted to obtain a restraining order against her. Id. at 239-41 (Ms. Geer’s Test.). Ms. Geer further testified that her mother is “very cruel, callous,” and recounted instances of her “being abusive.” Id. at 245. Ms. Geer noted that “even prior to these allegations” against her father, she and her mother “ha[d] an estranged relationship.” Id. at 254.

After hearing all of the evidence, the state district court denied Mr. Ellis’s motion for postconviction relief. Regarding Mr. Ellis’s claim of ineffective assistance, the court held that he had “failfed] to overcome the presumption that sound trial strategy was used.” Aplt’s App. Vol. I, at 273. And the court could not “reach the conclusion that the result would have been different but for trial counsel’s deficient performance.” Id. Mr. Ellis appealed the denial of his motion for postconviction relief to the CCA. The CCA affirmed.

It held, as relevant here, that the state district court had correctly concluded that Mr. Stayton was not constitutionally ineffective. It first reasoned that Mr. Stayton’s failure to consult and/or call an expert forensic psychologist was not deficient because Mr. Stayton “was familiar with the psychological concepts that an expert forensic psychologist would have explained,” and “he elicited lay testimony at trial about the complicated family dynamics and children aligning with mother and against defendant,” and because Mr.

Stayton believed that if he called a forensic psychologist expert, the State would call its own, and “such conflicting expert testimony would have damaged defendant’s theory of the case.” Id. at 290-91 (CCA’s Op.). The CCA explained: “[B]ecause trial counsel’s decision not to utilize a forensic psychologist was strategic, it was not constitutionally deficient.” Id. at 291. The CCA also determined that Mr. Stayton’s failure to consult and/or call four additional witnesses—Ms. Jefferson, Ms. Geer, Dr. Spiegle, and V.E.’s psychologist Dr. Long—did not constitute constitutionally deficient performance. Regarding Ms.

Jefferson, the CCA reasoned that “[although defendant contends that trial counsel should have somehow discovered that [Ms. Jefferson] was lying at trial, without more, we decline to hold that trial counsel was deficient for failing to know of a witness’s apparent perjury.” Id. at 292. Regarding Ms. Geer and Dr. Spiegle, the CCA determined that their proposed testimony “about the general conflict in the family” was “brought ... before the jury by eliciting testimony about the conflict and resulting alignment from M.E. and [Ms. Jefferson].” Id. And regarding Dr. Long, the CCA concluded that Mr.

Stay-ton’s performance was not deficient because (1) he “interviewed Dr. Long before trial,” (2), “it [was] unclear from the record whether Dr. Long was still treating V.E. at the time of the sexual abuse,” and (3) “[t]he tactical decision not to call an expert witness is within the discretion of trial counsel.” Id. at 293. The CCA also rejected “Additional Alleged Deficiencies” that Mr. Ellis had raised regarding Mr. Stayton’s performance. Namely, it found meritless Mr. Ellis’s arguments “that trial counsel’s unspecified ‘mishandling of Rule 404(

b) evidence, his ineffective cross-examination of prosecution witnesses, and his failures to object to improper questioning by the prosecutor’ constituted deficient performance.” Id. Importantly, because the CCA concluded that Mr. Ellis had “failed to establish deficient performance,” it never reached the issue of whether Mr. Stayton’s performance was prejudicial. Id. at 294. Mr. Ellis did not petition the CSC for certiorari review of the denial of his motion for postconviction relief.

Instead, in 2014, he filed a petition for a writ of habe-as corpus in the United States District Court for the District of Colorado, alleging, inter alia, ineffective assistance of trial counsel. Specifically, Mr. Ellis alleged that Mr. Stayton’s failures to consult and/or call an expert forensic psychologist, Dr. Long, Dr. Spiegle, Ms. Jefferson, and Ms. Geer were deficient; Mr. Stayton’s weak cross-examinations, mishandling of prejudicial evidence, and failure to object to improper prosecution questioning were deficient; and Mr. Stayton’s deficient performance was prejudicial.

Contrary to both state postconviction court rulings, the federal district court determined that Mr. Stayton’s representation was “glaringly deficient.” Aplt’s App. Vol. I, at 181. The court decided that Mr.

Stay-ton: (1) delegated too many pre-trial responsibilities to an investigator and delayed in testing DNA and fiber evidence; (2) should have consulted and called an expert forensic psychologist; and (3) should have consulted and/or called “multiple members of the Ellis family and psychological professionals who ... had firsthand insights into the severe dysfunction in the Ellis family and Kari Ellis’s behavior.” Id. at 182. The federal district court then concluded that these deficiencies were prejudicial.

Because the CCA never reached the prejudice prong, the federal district court reviewed de novo the state district court’s decision and determined that it had applied Strickland unreasonably. The court explained: “Because the evidence in the record did not overwhelmingly support Ellis’s guilt, ... had the jury heard an effective defense presentation of the physical evidence in the case and the testimony of the witnesses presented by the defense in the [postconviction] hearing, at least one of the jurors would have had a reasonable doubt as to Ellis’s guilt.” Id. at 186. Accordingly, the court granted Mr.

Ellis ha-beas relief, but conditioned it on the State’s decision not to retry him within ninety days. In response to this ruling, the State filed a motion for additional findings, a motion to alter order and judgment, and a motion to clarify.

Regarding the motion for additional findings, the district court issued an order stating that, “Having now considered all of the relevant state court record, this Court finds that there is additional support for the ruling that Rowe Stayton’s deficient performance undermined eonfi-dence in the outcome of the Ellis trial thereby meeting the prejudice prong of Strickland.” Id. at 246 (Order on Mot. for Additional Findings) (emphasis added).

Regarding the motion to clarify, the court issued an order clarifying that “the State of Colorado must re-try Mark Stephen Ellis on the charges upon which he was convicted within 90 days from the entry of the amended judgment, failing in which he shall be released from custody on those convictions, which have been vacated by this Court, and upon which no further proceedings shall be pursued.” Id. at 244 (Order Clarifying J.). The district court stayed enforcement of its amended judgment through any appeal. The State timely appealed. II On appeal, the State raises three claims.

It alleges that the district court: (1) erred in concluding that Mr. Ellis exhausted state -remedies on the ineffective-assistance claim; (2) erred in granting conditional habeas relief on the ineffective-assistance claim; and (3) abused its discretion in barring retrial after ninety days. We conclude that Mr. Ellis did'exhaust his remedies, but that the district court erred in granting him habeas relief on the ineffective-assistance claim. These rulings render the third issue—involving the propriety of the district court’s ninety-day retrial condition—effectively moot, so we do not reach it.

A First, the State argues that Mr. Ellis failed to exhaust state remedies because he never petitioned the CSC for discretionary review of his ineffective-assistance claim. We review de novo a district court’s

interpretation of a federal statute. Hain v. Mullin, 436 F.3d 1168, 1176 (10th Cir. 2006) (en banc). 1 The statutory basis for Mr. Ellis’s petition, 28 U.S.C. § 2254, codifies certain provisions of the Antiterrorism and Effective Death Penalty Act of 1996 (AEDPA). See Pub. L. No. 104-132, § 104, 110 Stat. 1228; see, e.g., Paxton v. Ward, 199 F.3d 1197, 1204 (10th Cir. 1999); Nobles v. Johnson, 127 F.3d 409, 412-13 (5th Cir. 1997). Under AEDPA, “a state prisoner generally must exhaust available state-court remedies before a federal court can consider a' habeas corpus petition.” Bland v.

Sirmons, 459 F.3d 999, 1011 (10th Cir. 2006). More specifically, AEDPA prohibits federal courts from granting habeas relief to state prisoners. who have not exhausted available state remedies. In this regard, § 2254(b)(1) states, “An application for a writ of habeas corpus ... shall not be granted unless it appears that[ ] ... the applicant has exhausted the remedies available in the courts of the State....” 28 U.S.C. § 2254(b)(1).

Section 2254(

c) elaborates that “[a]n applicant shall not be deemed to have exhausted the remedies available in the courts of the State[ ] ... if he has the right under the law of the State to raise, by any available procedure, the question presented.” Id. § 2254(c) (emphasis added). In O’Sullivan v. Boerckel, 526 U.S. 838, 119 S.Ct. 1728, 144 L.Ed.2d 1 (1999), the Supreme Court interpreted AEDPA’s exhaustion requirement to mean that “a state prisoner must present his claims to a state supreme court in a petition for discretionary review in order to satisfy [AEDPA’s] exhaustion requirement.” Id. at 839-40, 119 S.Ct. 1728.

Critically, however, the Court stated that its holding does not require “federal courts to ignore a state law or rule providing that a given [state appellate review] procedure is not available.” Id. at 847-48, 119 S.Ct. 1728. In 2006 the CSC promulgated Colorado Appellate Rule 51.1, which states: In all appeals from criminal convictions or post-conviction relief ..., a litigant shall not be required to petition for rehearing and certiorari following an adverse decision of the Court of Appeals in order to be deemed to have exhausted all available state remedies....

Rather, when a claim has been presented to the Court of Appeals or Supreme Court, and relief has been denied, the litigant shall be deemed to have exhausted all available state remedies. Colo. App. R. 51.1(a) (emphasis added). This court has not yet decided on the effect of Rule 51.1 in the federal habeas context, when viewed through the prism of O’Sullivan’s exhaustion holding. 2 The district court concluded that Mr. Ellis exhausted all available state remedies because he raised his ineffective-assistance claim “before the CCA and was denied relief,” which is all that Rule 51.1 requires. Aplt’s App. Vol.

I, at 170. Appealing that determination, the State argues that O’Sullivan made clear that a prisoner must seek review in the state -supreme court in order to exhaust state remedies. The State admits that in O’Sullivan, the Court “acknowledged] the existence of rules like 51.1,” and that “the majority did not ultimately rule.on their effect.” Id. at 16. However, the State argues that to give effect to rules like 51.1 would be “inconsistent with the reasoning of O’Sullivan.” Id. at 23. In response, Mr.

Ellis argues that the majority opinion in O’Sullivan indicated that state rules like 51.1 “could render certiorari review not ‘available’ for purposes of federal habeas doctrine.” Aplee.’s Br. at 18. He relies on the fact that “the Illinois system addressed in O’Sullivan did not have such a rule, and the majority opinion did not invalidate any such rule.” Id. And he points out that one concurring and three dissenting justices in O’Sullivan “all expressed their view that the majority left the question open—an

interpretation the majority did not contradict.” Id. at 21. Thus, Mr. Ellis encourages us to hold that Rule 51.1 “sets forth the circumstances under which a litigant ‘shall be deemed to have exhausted all available state remedies.’ ” Id. at 22 (quoting Colo. App. R. 51.1(a)). 3 Whether Rule 51.1 permits state prisoners to exhaust all available state remedies without seeking discretionary relief from the CSC is a question of first impression in this court.

As explicated further infra, all of our sister circuits who have considered analogous exhaustion questions under rules like Rule 51.1 have concluded that they do permit state prisoners to effect exhaustion of available state remedies. We reach this same result with respect to Rule 51.1. As a result, we hold that the district court correctly concluded that Mr. Ellis exhausted all available state remedies on his ineffective-assistance claim.

In O’Sullivan, the Supreme Court held that a state prisoner had failed to exhaust state remedies for claims that he did not include in his petition for leave to appeal to. the Illinois Supreme Court, a court of discretionary review, because “a state prisoner must present his claims to a state supreme court in a petition for discretionary review in order to satisfy [AEDPA’s] exhaustion requirement.” 526 U.S. at 839-40, 119 S.Ct. 1728.

The Court reasoned that “the exhaustion doctrine is designed to give the state courts a full and fair opportunity to resolve federal constitutional claims before those claims are 'presented to the federal courts.” Id. at 845, 119 S.Ct. 1728. And, the Court said, “[b]y requiring state prisoners to give the Illinois Supreme Court the opportunity to resolve constitutional errors in the first instance, the rule we announce today serves the comity interests that drive the exhaustion doctrine.” Id. at 846, 119 S.Ct. 1728.

When O’Sullivan was decided, Illinois did not have a state rule akin to Colorado’s Rule 51.1—namely, a rule stating that a petitioner need not petition for discretionary state supreme court review in order to exhaust all available state remedies. The OSullivan Court acknowledged that some state courts may “not wish to have the opportunity to review constitutional claims before those claims are presented to a federal habeas court.” Id.

The Court stated that “[u]nder these circumstances,” the petitioner “may be correct that the increased, unwelcome burden on state supreme courts disserves the comity interests underlying 'the exhaustion doctrine.” Id. at 847, 119 S.Ct. 1728. The Court continued that because § 2254(c) “directs federal courts to consider whether a habeas petitioner has ‘the right under the law of the State to raise, by any available procedure, the question presented,’ ” the exhaustion doctrine actually “turns on an inquiry into what procedures are ‘available’ under state law.” Id.

As a result, the Court “note[d] that nothing in [its] decision today requires the exhaustion of any specific state remedy when a State has provided that that remedy is unavailable.” Id. (emphasis added). “In sum,” the Court concluded, “there is nothing in the exhaustion doctrine requiring federal courts to ignore a state law or rule providing that a given procedure is not available”; it held “only that the creation of a discretionary review system does not, without more, make review in the Illinois Supreme Court unavailable.” Id. at 847-48, 119 S.Ct. 1728. Three justices wrote separately.

Though one authoring justice concurred and the other two dissented, all three emphasized the limited nature of the majority’s holding—namely, that the majority did not speak to the circumstances under which states could declare resort to certain discretionary review procedures unnecessary for purposes of exhaustion, thereby rendering those procedures effectively unavailable. All three authoring justices at the very least intimated that such circumstances may well exist, and two seemed to take a more definitive, affirmative stance on the matter.

First, writing in concurrence, Justice Souter stated: “I understand the Court to have left open the question (not directly implicated by this case) whether we should construe the exhaustion doctrine to force a State, in effect, to rule on discretionary review applications when the State has made it plain that it does not wish to require such applications before its petitioners may seek federal habeas relief.” Id. at 849, 119 S.Ct. 1728 (Souter, J., concurring).

Thus, Justice Souter clarified that the majority opinion “should not be read to suggest ... that however plainly a State may speak[,] its highest court must be subjected to constant applications for a form of discretionary review that the State wishes to reserve for truly extraordinary cases, or else be forced to eliminate that kind of discretionary review.” Id. at 849-50, 119 S.Ct. 1728 (emphasis added).

In conclusion, he reiterated: I understand that we leave open the possibility that a state prisoner is likewise free to skip a procedure even when a state court has occasionally employed it to provide relief, so long as the State has identified the procedure as outside the standard review process and has plainly said that it need not be sought for the purpose of exhaustion. It is not obvious that either comity or 'precedent requires otherwise. Id. at 850, 119 S.Ct. 1728 (emphasis added). Second, Justice Stevens dissented in an opinion joined by Justice Ginsburg and Justice Breyer.

He lamented that the majority’s holding would “impose unnecessary burdens on habeas petitioners; it w[ould] delay the completion of litigation that is already more protracted than it should be; and, most ironically, it w[ould] undermine federalism by thwarting the interests of those state supreme courts that administer discretionary dockets.” Id. at 859, 119 S.Ct. 1728 (Stevens, J., dissenting).

Justice Stevens explained: “If, as the Court has repeatedly held, the purpose of our waiver doctrine is to cultivate comity by respecting state procedural rules, then ... we should not create procedural obstacles when state prisoners follow those rules.” Id. at 859-60, 119 S.Ct. 1728.

However, notably, he added: Thankfully, the Court leaves open the possibility that state supreme courts with discretionary dockets may avoid a deluge of undesirable claims by making a plain statement—as Arizona and South Carolina have done—that they do not wish the opportunity to review such claims before they pass into the federal system. I agree with Justice SOUTER that a proper conception of comity obviously requires deference to such a policy.

Id. at 861-62, 119 S.Ct. 1728 (citations omitted); see also id. at 862, 119 S.Ct. 1728 (“The key point is that federal courts should not find procedural default when a prisoner has relied on a state supreme court’s explicit statement that criminal defendants need not present to it every claim that they might wish to assert as a ground for relief in federal habeas proceedings.”). Third, Justice Breyer dissented in an opinion joined by Justice Ginsburg and Justice Stevens.

Justice Breyer wrote that whether a state prisoner (who failed to seek discretionary review in a state supreme court) can seek federal habeas relief depends upon the State’s own preference. If the State does not want the prisoner to seek discretionary state review (or if it does not care), why should that failure matter to federal ha-beas law? Id. at 862, 119 S.Ct. 1728 (Breyer, J., dissenting).

However, Justice Breyer “nonetheless s[aw] cause for optimism,” because “Justice SOUTER’s concurring opinion suggests that a federal habeas court should respect a State’s desire that prisoners not file petitions for discretionary review, where the State has expressed the desire clearly.” Id. at 864, 119 S.Ct. 1728.

Justice Breyer concluded that “today’s holding creates a kind of presumption that a habeas petitioner must raise a given claim in a petition for discretionary review in state court prior to raising that claim on federal habeas, but the State could rebut the presumption through state law clearly expressing a desire to the contrary. South Carolina has expressed that contrary preference. Other States may do the same.” Id. at 864, 119 S.Ct. 1728 (citation omitted). Indeed, Justice Breyer “wr[o]te to emphasize the fact that the majority has left the matter open.” Id.

Consistent with O’Sullivan’s emphasis on comity, some of our sister circuits— specifically, four—-when considering analogous exhaustion questions under state rules similar to Rule 51.1, have concluded that those rules permit prisoners to exhaust available state remedies. First, in Swoopes v.

Sublett, 196 F.3d 1008 (9th Cir. 1999) (per curiam), the Ninth Circuit held that because “the Arizona Supreme Court has announced that ... review need not be sought before the Arizona Supreme Court in order to exhaust state remedies,” “post- . conviction review before the Arizona Supreme Court is a remedy that is ‘unavailable’ within the meaning of O’Sullivan.” 196 F.3d at 1010. The court reasoned: Although review before the Arizona- Supreme Court is discretionary, it is “available” under O’Sullivan; thus, at least facially, Arizona prisoners are not relieved of their duty to file an appeal with that court.

However, the question is whether Arizona has identified discretionary Supreme Court review ‘as outside the standard review process and has plainly said that it need not be sought for the purpose of exhaustion.’ It plainly has. Id. (citation omitted).

In short, the Ninth Circuit concluded that because “Arizona has declared that its ‘complete round’ [of the State’s appellate review process] does not include discretionary review before the Arizona Supreme Court,” “post-conviction review before the Arizona Supreme Court is a remedy that is ‘unavailable’ within the meaning of O’Sullivan.” Id. at 1010 (stating that it “must credit Arizona’s choice”). The court explained: “The import of O’Sullivan is that exhaustion is not required when a state declares which remedies are ‘available’ for exhaustion. Arizona has done so.” Id. at 1011. Second, in Randolph v.

Kemna, 276 F.3d 401, 404-05 (8th Cir. 2002), the Eighth Circuit gave effect to a Missouri Supreme Court rule stating that its discretionary review “is an extraordinary remedy that is not part of the standard review process for purposes of federal habeas corpus review.” Id. at 404 (quoting Mo. Sup. Ct. R. 83.04).

The court reasoned that “O’Sullivan held that where a state articulates that a certain avenue is not part of its appellate review process, it is not necessary that prisoners pursue that avenue,” and the court concluded that Missouri’s state rule “constitute[d] an unequivocal statement about where Missouri’s ‘one complete round of the state’s established appellate review process’ stops and ma[d]e[ ] clear that Missouri does not consider a petitioner who bypasses its supreme court in favor of federal habeas review to have denied the State its rightful ‘opportunity to resolve federal constitutional claims.’ ” Id. (quoting O’Sullivan, 526 U.S. at 845, 119 S.Ct. 1728).

Because the court “c[ould] ask for no more clear statement than that,” it gave effect to Missouri’s state rule. Id. at 405. Third, in Adams v. Holland, 330 F.3d 398, 401-03 (6th Cir. 2003), the Sixth Circuit upheld a Tennessee Supreme Court rule declaring that “a litigant shall not be required to petition for rehearing or to file an application for permission to appeal to the Supreme Court of Tennessee following an adverse decision of the Court of Criminal Appeals in order to be deemed to have exhausted all available state remedies.” Id. at 401 (quoting Tenn. Sup. Ct. R. 39).

Observing that “[t]he O’Sullivan Court[] ... explicitly excepted from its holding cases in which the state has explicitly disavowed state supreme court review as an ‘available state remedy,”’ id., the Sixth Circuit reasoned that the state rule “clearly removed Tennessee Supreme Court review as an antecedent for habeas purposes,” id. at 402.

In doing so, the court rejected an argument that the rule did not technically make Tennessee Supreme Court review “unavailable, ... since litigants have not been explicitly prohibited from appealing to the state supreme court”; the court explained that this argument “fails to grasp the meaning of the word ‘available’ as it is used in O’Sullivan, and instead dwells upon a hypertechnical

interpretation of that term.” Id. The court continued, moreover, that this “misinterpretation is revealed by the O’Sullivan Court’s examples of what might constitute making state supreme court review unavailable: namely, rules passed by the Supreme Courts of South Carolina and Arizona,” id., and it noted that its own “view of the O’Sullivan opinion is bolstered by the concurring opinion of Justice Souter from that case.” Id. at 403. Thus, the Sixth Circuit held that Tennessee’s state rule “rendered Tennessee Supreme Court review ‘unavailable’ in the context of habeas relief.” Id. And, fourth, in Lambert v.

Blackwell, 387 F.3d 210 (3d Cir. 2004), the Third Circuit held that a Pennsylvania Supreme Court order (i.e., Order No. 218) declaring that “a litigant shall not be required to petition for rehearing or allowance of appeal following an adverse decision by the Superior Court in order to be deemed to have exhausted all available state remedies” rendered “review from the Pennsylvania Supreme Court ‘unavailable’ for purposes of exhausting state court remedies under § 2254(c).” 387 F.3d at 233 (quoting Order No. 218).

The court reasoned that the O’Sullivan Court “took pains[ ] ... to state that ‘there is nothing in the exhaustion doctrine requiring federal courts to ignore a state law or rule providing that a given procedure is not available.’” Id. at 232 (quoting O’Sullivan, 526 U.S. at 847-48, 119 S.Ct. 1728). And the court concluded that Pennsylvania’s state rule “is the something ‘more’ that makes the Pennsylvania Supreme Court’s discretionary review system ‘unavailable.’ ” Id. at 233 (citation omitted).

In analyzing the federal-habeas exhaustion implications of Rule 51.1, we are persuaded by the holdings of the four circuits and believe that they correctly give effect to the concerns expressed in O’Sullivan. Concern for comity primarily motivated the outcome of that case: the Supreme Court stated that it wanted to give state courts the opportunity to resolve state cases in the first instance before federal courts could intrude.

See O’Sullivan, 526 U.S. at 844, 119 S.Ct. 1728 (“Comity thus dictates that ... the state courts should have the first opportunity to review [a] claim and provide any necessary relief.”); id. at 845, 119 S.Ct. 1728 (“[T]he exhaustion doctrine is designed to give the state-courts a full and fair opportunity to resolve federal constitutional claims before those claims are presented to the federal courts.”); id. (“Comity, in these circumstances, dictates that Boerckel use the State’s established appellate review procedures before he presents his claims to a federal court.”); id. at 846, 119 S.Ct. 1728 (“[T]he rule we announce today serves the comity interests that drive the exhaustion doctrine.”).

Further, when the Court seemingly left open the question we address today, it clearly suggested that failing to give effect for exhaustion purposes to state rules like Colorado’s Rule 51.1 might not serve the interests of comity that underlay its holding.

See id. at 847, 119 S.Ct. 1728 (stating that the petitioner “may be correct that the increased, unwelcome burden on state supreme courts disserves the comity interests underlying the exhaustion doctrine”); see also id. (stating that because § 2254(c) “directs federal courts to consider whether a habeas petitioner has ‘the right under the law of the State to raise, by any avail able procedure, the question presented,’ ” the exhaustion doctrine actually “turns on an inquiry into what procedures are ‘available’ under state law”); id. (“[T]here is nothing in the exhaustion doctrine requiring federal courts to ignore a state law or rule providing that a given procedure is not available.”).

In light of the concern for comity underlying O’Sullivan, and the persuasive holdings of our four sister circuits, we conclude that Rule 51.1 renders CSC review “unavailable” for purposes of AED-PA exhaustion. Thus, we hold that pursuant to Rule 51.1, Mr. Ellis exhausted state remedies for his ineffective-assistance claim, and we proceed to consider the merits of that claim. B Having determined that Mr. Ellis properly exhausted state remedies, we turn to the question of whether he received constitutionally ineffective assistance of counsel. Mr. Ellis and the State dispute whether Mr.

Stayton was constitutionally ineffective for failing to consult and/or call the following witnesses: (1) an expert forensic psychologist;

(2) Dr. Long, a clinical psychologist who had counseled V.E.;

(3) Dr. Spiegle, the Ellis children’s court-appointed special advocate in their parents’ divorce proceedings;

(4) Ms. Jefferson, V.E.’s eldest sister; and

(5) Ms. Geer, another of V.E.’s sisters. Mr. Ellis argues that these alleged deficiencies were both individually and cumulatively prejudicial; the State rejects this conclusion. In addition, the State argues that the federal district court erred “by sua sponte raising and aggregating prejudice from ineffective-assistance claims Petitioner never raised” in the CCA—specifically, by ruling that “counsel’s failure to secure additional forensic testing” and his “agreement to waive the recording of closing arguments” constituted deficient performance. Aplt’s Opening Br. at 45. Mr.

Ellis responds that even though the district court’s “ruling is somewhat expansive and encompasses additional grounds not raised by Mr. Ellis,” his ineffective-assistance claim “as a whole” is not procedurally defaulted, and he notes that we “may affirm the judgment on any basis evident in the record.” Aplee.’s Br. at 25. We begin by analyzing each of the claims relating to particular witnesses and conclude that the federal district court erred in concluding that the CCA’s ineffective-assistance determinations were unreasonable and thus legally untenable. Specifically, regarding Mr.

Stayton’s decisions not to interview and/or call an expert forensic psychologist, Dr. Spiegle, Ms. Jefferson, and Ms. Geer, we conclude that the CCA was not unreasonable in determining that these decisions were strategic and not constitutionally deficient. Additionally, regarding Mr. Stayton’s decision not to consult and/or call Dr. Long—assuming ar-guendo that the CCA unreasonably deemed this decision an instance of constitutionally deficient performance, but not definitively determining what standard of review should apply to any prejudice analysis of Mr.

Stayton’s decision—we conclude that even under the standard of review most favorable to Mr. Ellis, that is, de novo review, he cannot show prejudice stemming from Mr. Stayton’s failure to consult and/or call Dr. Long. Thus, in sum, we conclude that Mr. Ellis is not entitled to habeas relief on the basis of Mr. Stay-ton’s failure to consult and/or call each of these witnesses. And, because there is only one assumed error regarding Dr. Long, we do not conduct a cumulative-error analysis.

We lastly turn to the State’s argument that the federal district court erred by factoring into its grant of habeas relief ineffective-assistance claims that Mr. Ellis never raised before the CCA. Mr. Ellis does not dispute that he never presented these ineffective-assistance claims to the CCA but he contends that the attorney errors on which the claims are based can individually and cumulatively provide alternative grounds for affirming the district court’s ineffective-assistance ruling. We disagree. We conclude that the district court erroneously considered Mr.

Ellis’s additional ineffective-assistance claims— which were unexhausted and procedurally defaulted—in granting habeas relief. And it would be improper for us to consider them now. We reverse the district court’s judgment. 1 “Under AEDPA, the standard of review applicable to a particular claim depends on how that claim was resolved by the state courts.” Byrd v. Workman, 645 F.3d 1159, 1165 (10th Cir. 2011). “[I]f there has been no state-court adjudication on the merits of the petitioner’s claim,” “we review the district court’s legal conclusions de novo and its factual findings for clear error.” Id. at 1166-67.

In contrast, “[w]here the state court has adjudicated a claim on the merits, we may only grant relief if the state court’s decision ‘was contrary to, or involved an unreasonable application of, clearly established Federal law, as determined by the Supreme Court of the United States,’ or ‘was based on an unreasonable determination of the facts in light of the evidence presented in the State court proceeding.’ ” Id. (citation omitted) (quoting 28 U.S.C. § 2254(d)(1), (d)(2)).

For claims adjudicated on the merits, therefore, “AEDPA imposes a ‘highly deferential standard for evaluating state-court rulings’—one that ‘demands that state-court decisions be given the benefit of the doubt,’ ” and that “prohibits us from substituting our own judgment for that of the state court.” Id. at 1166 (citations omitted). “If this standard is difficult to meet, that is because it was meant to be.” Harrington v. Richter, 562 U.S. 86, 102, 131 S.Ct. 770, 178 L.Ed.2d 624 (2011).

AEDPA provides that “[a]s a condition for obtaining habeas corpus from a federal court, a state prisoner must show that the state court’s ruling on the claim being presented in federal court was so lacking in justification that there was an error well understood and comprehended in existing law beyond any possibility for fairminded disagreement.” Id. at 103, 131 S.Ct. 770. Under this standard, “[e]ven a strong case for relief does not make the state court’s contrary conclusion unreasonable.” Id. at 102, 131 S.Ct. 770.

As for the substantive standards applicable here, to prevail on a claim of ineffective assistance, “[fjirst, the defendant must show that counsel’s performance was deficient.... Second, the defendant must show that the deficient performance prejudiced the defense.” Strickland v.

Washington, 466 U.S. 668, 687-88, 104 S.Ct. 2052, 80 L.Ed.2d 674 (1984); see Byrd, 645 F.3d at 1167 (noting that a defendant “must show both that his counsel’s performance ‘fell below an objective standard of reasonableness’ and that ‘the deficient performance prejudiced the defense’ ” (quoting Strickland, 466 U.S. at 687-88, 104 S.Ct. 2052)).

Under the first Strickland prong, performance, “the standard for judging counsel’s representation is a most deferential one” under which “[t]he question is whether an attorney’s representation amounted to incompetence under ‘prevailing professional norms,’ not whether it deviated from best practices or most common custom.” Harrington, 562 U.S. at 105, 131 S.Ct. 770 (quoting Strickland, 466 U.S. at 690, 104 S.Ct. 2052).

The Supreme Court has recognized that there are “countless ways to provide effective assistance in any given case,” “the best criminal defense attorneys would not defend a particular client in the same way,” and “[r]are are the situations in which the ‘wide latitude counsel must have in making tactical decisions’ will be limited to any one technique or approach.” Id. at 106, 131 S.Ct. 770 (quoting Strickland, 466 U.S. at 689, 104 S.Ct. 2052).

As a result, the Court has held that “[c]ounsel [i]s entitled to formulate a strategy that [i]s reasonable at the time and to balance limited resources in accord with effective trial tactics and strategies.” Id. at 107, 131 S.Ct. 770. When a petitioner raises a claim of ineffective assistance in a habeas petition, our “highly deferential” assessment of counsel’s performance under Strickland’s first prong, [Danny] Hooks v. Workman, 606 F.3d 715, 723 (10th Cir. 2010), is “doubly” so, Knowles v.

Mirzayance, 556 U.S. 111, 123, 129 S.Ct. 1411, 173 L.Ed.2d 251 (2009), because we deferen tially evaluate counsel’s performance through the prism of AEDPA’s deference to the state court’s assessment of counsel’s performance. See Byrd, 645 F.3d at 1168 (noting that “[w]e defer to the state court’s determination that counsel’s performance was not deficient and, further, defer to the attorney’s decision in how to best represent a client” (alteration in original) (quoting Crawley v. Dinwiddie, 584 F.3d 916, 922 (10th Cir. 2009))).

Put another way, applying the first layer of deference, we review only “whether the state court’s ap plication of the Strickland standard was unreasonable.” Harrington, 562 U.S. at 101, 131 S.Ct. 770 (emphasis added). Applying the second layer, we “apply a ‘strong presumption’ that counsel’s representation was within the ‘wide range’ of reasonable professional assistance.” Id. at 104, 131 S.Ct. 770 (quoting Strickland, 466 U.S. at 689, 104 S.Ct. 2052).

As the Court cautioned in Harrington, “Federal habeas courts must guard against the danger of equating unreasonableness under Strickland with unreasonableness under § 2254(d). When § 2254(

d) applies, the question is not whether counsel’s actions were reasonable. The question is whether there is any reasonable argument that counsel satisfied Strickland’s deferential standard.” Id. (emphasis added); accord [Victor] Hooks v. Workman, 689 F.3d 1148, 1187 (10th Cir. 2012). As for the prejudice prong (i.e., the second prong), the Strickland Court opined: “The defendant must show that there is a reasonable probability that, but for counsel’s unprofessional errors, the result of the proceeding would have been different.

A reasonable probability is a probability sufficient to undermine confidence in the outcome.” 466 U.S. at 694, 104 S.Ct. 2052; accord Harrington, 562 U.S. at 107, 131 S.Ct. 770; [Victor] Hooks, 689 F.3d at 1187. “It is not enough for the defendant to show that the errors had some conceivable effect on the outcome of the proceeding.... [N]ot every error that conceivably could have influenced the outcome undermines the reliability of the result of the proceeding.” Strickland, 466 U.S. at 693, 104 S.Ct. 2052 (citation omitted); accord [Victor] Hooks, 689 F.3d at 1187; see also Byrd, 645 F.3d at 1168 (noting that “mere speculation is not sufficient to satisfy this burden” under Strickland’s prejudice prong).

Expanding on this proposition, the Court in Harrington observed: In assessing prejudice under Strickland, the question is not whether a court can be certain counsel’s performance had no effect on the outcome or whether it is possible a reasonable doubt might have been established if counsel acted differently. Instead, Strickland asks whether it is “reasonably likely” the result would have been different.

This does not require a showing that counsel’s actions “more likely than not altered the outcome,” but the difference between Strickland’s prejudice standard and a more-probable-than-not standard is slight and matters “only in the rarest ease.” The likelihood of a different result must be substantial, not just conceivable. 562 U.S. at 111-12, 131 S.Ct. 770 (emphasis added) (citations omitted) (quoting Strickland, 466 U.S. at 693, 696, 104 S.Ct. 2052). In order to succeed on an ineffective-assistance claim, a petitioner must satisfy both prongs of the Strickland test.

Strickland, 466 U.S. at 687, 104 S.Ct. 2052; accord [Victor] Hooks, 689 F.3d at 1186; Byrd, 645 F.3d at 1167. But, notably, “[c]ourts are free to address these two prongs in any order.” Byrd, 645 F.3d at 1168; see Strickland, 466 U.S. at 668, 104 S.Ct. 2052 (“Although we have discussed the performance component of an ineffectiveness claim prior to the prejudice component, there is no reason for a court deciding an ineffective assistance claim to approach the inquiry in the same order or even to address both components of the inquiry if the defendant makes an insufficient showing on one.”). We address each of Mr.

Ellis’s ineffective-assistance claims below. With the exception of the claim involving Mr. Stay-ton’s failure to consult and/or call Dr. Long, which we resolve under Strickland’s prejudice prong, we rule on the claims by examining whether the district court erred in concluding that the CCA unreasonably determined that counsel’s performance was not constitutionally deficient (i.e., Strickland’s performance prong). 2 First, we address Mr. Stayton’s decisions not to consult and/or call four witnesses. Under AEDPA’s doubly deferential standard of review—in which we defer to both the CCA’s determination that Mr.

Stayton’s performance was not constitutionally deficient and Mr. Stayton’s decisions regarding how to represent his client—we conclude that the federal district court erred in determining that the CCA unreasonably concluded that Mr. Stayton was constitutionally ineffective. We address each of the witnesses below. a First, we conclude that the CCA’s application of Strickland’s performance prong to Mr. Stayton’s decision not to consult an expert forensic psychologist was not unreasonable. The CCA concluded that Mr.

Stayton’s decision fell within the range of reasonable performance because he believed that, if he called an expert forensic psychologist to testify, the State would call its own, and “such conflicting expert testimony would have damaged defendant’s theory of the case.” Aplt.’s App. Vol. I, at 290-91. Instead, he chose to elicit the relevant psychological theories—with which he was very familiar—through cross-examination of state witnesses. The record provides support for the CCA’s conclusion. As he recounted at the postconviction evidentiary hearing, Mr.

Stayton decided at trial that calling an expert would open the defense to “a lot of attack”—attacks that, based on his experience, Mr. Stayton feared would undercut his chances of prevailing. Aplt.’s App. Vol. VI, at 229. He believed that he could convey the relevant psychological themes to the jury through cross-examination of state witnesses, instead. And he was well equipped to do so. Mr. Stayton, a specialist in child sexual assault cases, had taught lectures involving the psychological .theories involved in Mr. Ellis’s ease to defense attorneys around the country. Specifically, at the hearing, Mr.

Stayton testified that he was familiar enough with theories of parental alienation and childhood memory “to prepare [his] cross-examination of the witnesses,” “to prepare [his] presentation of evidence,” and “to prepare both [his] opening statements and closing arguments” in this case. Aplt.’s App. Vol. VI, at 231-32. Accordingly, “throughout the trial,” Mr. Stayton worked to lay the groundwork for his theory of parental alienation and, more specifically, to establish that V.E. was falsely accusing Mr.

Ellis because “Kari was really mad at Mark, and ... [V.E.] had read her mom, and she was more closely aligned with her mom.” Id. at 226 (alteration in original). Mr. Stayton elicited from cross-examination of V.E. that she was angry at her father, that she did not like him, and that she felt closer to her mother. He elicited from Ms. Jefferson that she had “allied [her]self with [their] mother” and that Ms. Geer had “allied herself with” their father. Aplt.’s App. Vol. III, at 256. He was able to get M.E. to confirm that Ms.

Geer “being close to her dad is the same as [him] being close to [their] mom,” id. at 74, that he “dislike[d] [his] father a great deal,” and that he was “angry at [his father] for what he was doing to [their] family,” id. at 56. Mr. Stayton believed—not unreasonably—that he was effectively conveying the relevant psychological theories to the jury; he testified that he “just felt from [his] perception that they were making the points that we should.” Aplt’s App. Vol. VI, at 229. Not only did Mr.

Stayton assert that the relevant theories were already “in front of’ the jury, id. at 230, but, based on his experience, he opined that the theories were “easy” and “intuitive,” id. at 233. He testified that “[y]ou don’t have to have a master’s degree to understand that one parent can manipulate a child,” id., that he “didn’t think you needed an expert to tell [the jury] that[;] they either believed the child or they don’t,” and that he believed that it “would be insulting to the jury to try to point out the parental alienation” with an expert, id. at 230. In Mr.

Stayton’s experienced opinion, therefore, there were multiple reasons why he did not need to call an expert forensic psychologist to convey the relevant psychological theories to the jury. In our view, based on this record, it was not unreasonable for the CCA to conclude that Mr. Stayton’s decision to not consult or call as a witness a forensic psychologist was strategic and reasonable. This is sufficient to defeat a Strickland claim.

See Harrington, 562 U.S. at 105, 131 S.Ct. 770; see also Strickland, 466 U.S. at 681, 104 S.Ct. 2052 (“Because advocacy is an art and not a science, and because the adversary system requires deference to counsel’s informed decisions, strategic choices must be respected in these circumstances if they are based on professional judgment.”); cf. Harrington, 562 U.S. at 108, 131 S.Ct. 770 (stating that, even if certain expert testimony could have supported a defense, it is “reasonable to conclude that a competent attorney might elect not to use it”).

Thus, deferring to both “the state court’s determination that counsel’s performance was not deficient and, further, ... to the attorney’s decision in how to best represent a client,” Byrd, 645 F.3d at 1168 (quoting Crawley, 584 F.3d at 922), we conclude that the CCA reasonably determined that Mr. Stayton’s decision not to consult and/or call an expert forensic psychologist was not constitutionally deficient. b Second, we conclude that the CCA’s application of Strickland’s performance prong to Mr. Stayton’s decisions not to call—even after endorsing him as an expert—Dr.

Spiegle, and not to present to the jury the letters that Dr. Spiegle wrote about Ms. Ellis’s conduct in the divorce case, was not unreasonable. Mr. Ellis claims that Dr. Spiegle’s testimony and letters to the divorce court “could have provided important evidence of Kari Ellis’s manipulative conduct and how she placed her own needs for revenge ahead of the best interests of her children.” Aplee.’s Br. at 49. Dr. Spiegle’s letters to the divorce court show that Ms. Ellis was obstructing Mr.

Ellis’s access to their children during the divorce proceedings by ignoring his requests to give them birthday and Christmas gifts. They also show that she (unsuccessfully) moved to dismiss Dr. Spiegle as the children’s special advocate in the divorce proceedings. However, the GCA found that Mr. Stayton presented the substance of this evidence to the jury through the cross-examination of both M.E. and Ms. Jefferson. Mr. Stayton testified, and the CCA agreed, that he had deliberately decided, as part of his trial strategy, not to call Dr. Spiegle because he believed that the defense did not need Dr.

Spiegle’s testimony. It would have been reasonable for the CCA to infer from the fact that Mr. Stay-ton was prepared to call Dr. Spiegle and endorsed him as a defense witness—but ultimately decided not to call him because Mr. Stayton believed that he had already presented the evidence through cross-examination of state witnesses—that Mr. Stayton’s decision was an “exercise of rea7 sonable professional judgment” regarding how best to represent his client. Byrd, 645 F.3d at 1168 (citation omitted).

Furthermore, we have repeatedly held that whether to call a' particular witness is within the sound discretion of trial counsel. See Boyle v. McKune, 544 F.3d 1132, 1139 (10th Cir. 2008) (“[T]he decision of which witnesses to call is quintessentially a matter of strategy for the trial attorney.”); DeLozier v. Sirmons, 531 F.3d 1306, 1324 (10th Cir. 2008) (“Generally, the decision whether to call a witness rests within the sound discretion of trial counsel.” (quoting Jackson v. Shanks, 143 F.3d 1313, 1320 (10th Cir. 1998))); United States v.

Miller, 643 F.2d 713, 714 (10th Cir. 1981) (“Whether to call a particular witness is a tactical decision and, thus, a ‘matter of discretion’ for trial counsel.” (citation omitted)); United States v. Dingle, 546 F.2d 1378, 1385 (10th Cir. 1976) (“It is a matter of discretion on the part of trial counsel to exercise judgment in the determination of witnesses to be called and the testimony to be elicited.”). In sum, we conclude that it was not unreasonable for the CCA to determine that Mr. Stayton was not constitutionally deficient for declining to call Dr.

Spiegle as a witness and to present to the jury the letters that Dr. Spiegel wrote about Ms. Ellis’s conduct during the divorce. c Third, we conclude that the CCA’s application of Strickland’s performance prong to Mr. Stayton’s decision not to interview Ms. Jefferson fell within the range of reasonableness. The CCA stated, “Although defendant contends that trial counsel should have somehow discovered that [Ms. Jefferson] was lying at trial, without more, [it could not] hold that trial counsel was deficient for failing to know of a witness’s apparent perjury.” Aplt.’s App. Vol. I, at 292.

In addition, the CCA found that Mr. Stayton “elicited testimony from the detective who interviewed V.E. suggesting that [the] mother had coached or influenced V.E. during the investigation, as well as testimony from M.E. and [Ms. Jefferson] that they had aligned with mother against defendant.” Id. at 297. It was reasonable for the CCA to decide that Mr. Stayton was not constitutionally ineffective for failing to interview Ms. Jefferson before trial. Ms. Jefferson was a state witness.

At the time of trial, she was receiving all of her information about her father from her mother, and she “wanted to do everything [she] possibly could to help get [her father] put away.” Aplt.’s App. Vol. V, at 224. In fact, Ms. Jefferson was so strongly aligned with her mother— and hostile to her father—that she “rehears[ed] and coach[ed] [V.E.] from the very first day this allegation came to light,” and then lied about it on the stand. Id. at 223 (alterations in original). Mr. Stayton knew that the Ellis family was extremely divided, and that Ms. Jefferson was aligned with her mother. And Mr.

Stayton could well expect that she would testify for the state. It was within the range of reasonableness for him to spend his time on other preparations for trial instead of on trying to interview an uncooperative witness.

See Harrington, 562 U.S. at 107, 131 S.Ct. 770 (“Counsel was entitled to formulate a strategy that was reasonable at the time and to balance limited resources in accord with effective trial tactics and strategies.”); Strickland, 466 U.S. at 691, 104 S.Ct. 2052 (concluding that, when counsel has “reason to believe that pursuing certain investigations would be fruitless or even harmful, counsel’s failure to pursue those investigations may not later be challenged as unreasonable”); Morva v.

Zook, 821 F.3d 517, 528-29 (4th Cir. 2016) (finding no deficient performance under Strickland when attorney failed to interview several witnesses prior to death-penalty phase of defendant’s trial); Hanson v. Sherrod, 797 F.3d 810, 828 (10th Cir. 2015) (finding no deficient performance under Strickland when attorney declined to interview a potentially hostile witness before trial or to call that witness to testify); Hoffman v.

Cain, 752 F.3d 430, 446 (5th Cir. 2014) (finding no, deficient performance under Strickland in attorneys’ failure to interview witnesses before trial, and noting that the attorneys “made a reasonable decision not to investigate [the potential witnesses] further” and instead “chose a strategy of discrediting the State’s theory of the case”); see also Byrd, 645 F.3d at 1168 (stating that to be deficient, performance “must have been completely unreasonable, not merely wrong” (emphasis added) (citation omitted)). For these reasons, we conclude that the CCA’s determination regarding Mr.

Stay-ton’s decision not to interview Ms. Jefferson was not unreasonable. d Fourth, we conclude that the CCA’s application of Strickland’s performance prong to Mr. Stayton’s decision not to call Ms. Geer was not unreasonable. The CCA found that Ms. Geer would have testified “about the general conflict in the family, including how the fighting between defendant and mother caused the children to align with one parent or the other.” Aplt.’s App. Vol. I, at 292. However, the CCA found that Mr.

Stayton “brought this evidence before the jury by eliciting testimony about the conflict and resulting alignment from M.E. and [Ms. Jefferson].” Id. Specifically, Mr. Stayton deliberately chose to elicit the evidence about which Ms. Geer could have testified—evidence of Ms. Ellis’s allegedly vengeful attitude and her children’s alignment with her— through cross-examination of state witnesses. He testified that throughout trial, he tried to establish “that Kari was really mad at Mark, and that what happened is that [V.E.] had read her mom, and she was more closely aligned with her mom.” Aplt.’s App. Vol.

VI, at 226. Mr. Stayton had significant discretion to make this decision. See Boyle, 544 F.3d at 1139 (“[T]he decision of which witnesses to call is quintessentially a matter of strategy for the trial attorney.”); accord Dingle, 546 F.2d at 1385. And because Mr. Stayton’s decision reflected an intentional, strategic choice, within a trial lawyer’s customary discretion, we cannot conclude that the CCA was unreasonable in determining that Mr. Strayton’s decision not to call Ms. Geer was not constitutionally deficient performance. Put another way, the CCA did not unreasonably apply Strickland in concluding that Mr.

Stayton’s decision not to call Ms. Geer was within the “wide range of reasonable professional assistance.” Strickland, 466 U.S. at 689, 104 S.Ct. 2052; accord Byrd, 645 F.3d at 1168. 3 Relying on Strickland’s second prong, we conclude that Mr. Ellis has failed to show that he was prejudiced by Mr. Stayton’s decision not to consult and/or call Dr. Long after endorsing him as a defense expert. The CCA resolved this issue under Strickland’s performance prong, but the parties agree that the CCA’s analysis was based on an erroneous factual finding.

Specifically, contrary to what the CCA found, there is no evidence in the record that Mr. Stayton “interviewed Dr. Long before trial.” Aplt’s Opening Br. at 72. As a result of this uncontested error, the parties agree that we should review the performance prong de novo, but they disagree about how we should review the prejudice prong. Mr. Ellis argues that because the CCA never reached the prejudice prong, we should review it de novo.

The State responds that we should review the prejudice prong with AEDPA deference because: (1) a state court—specifically, the state district court—adjudicated prejudice on the merits, even though this was not the court of last resort (here, the CCA); and (2) the CCA’s factual error affected only the performance prong that it adjudicated, not the prejudice prong that it did not reach. Given the parties’ agreement on the matter, we assume for purposes of our analysis that the CCA in fact erred at Strickland’s performance prong, but we elect to elide a de novo analysis of this prong of Strickland.

See, e.g., Strickland, 466 U.S. at 668, 104 S.Ct. 2052. Instead, we advance to the question of prejudice. There, we need not resolve the standard-of-review question because, even applying the standard of review most, favorable to Mr. Ellis—de novo review—Mr. Ellis has not shown that Mr. Stayton’s failure to consult and/or call Dr. Long caused prejudice; this holding dooms Mr. Ellis’s ineffective-assistance claim. As we explicated supra, to satisfy Strickland’s prejudice prong, Mr.

Ellis must establish “a reasonable probability that, but for counsel’s unprofessional errors, the result of the proceeding would have been different.” Id. at 694, 104 S.Ct. 2052; accord Byrd, 645 F.3d at 1168; United States v. Challoner, 583 F.3d 745, 749 (10th Cir. 2009). Applying de novo review, we conclude that there is no reasonable probability that Dr. Long’s testimony would have changed the outcome of trial. Mr. Ellis argues that Dr.

Long could have testified that “as a licensed psychologist in Colorado, he has a duty to make an official report if a child patient discloses having been sexually abused,” and that he “never reported any allegation that Mr. Ellis had sexually abused anyone.” Aplee.’s Br. at 47. In short, Mr. Ellis argues .that Dr. Long could have testified that Y.E. never disclosed to him that her father was sexually assaulting her; Mr. Ellis reasons that this testimony would have been strong evidence to the jury that Mr. Ellis did not in fact sexually assault V.E.

However, as the State reasonably suggests, “the fact that an adopted nine-year-old girl did not volunteer to her male adult counselor that her father was molesting her is hardly an impressive fact that, had the jury known it, would likely have produced an acquittal.” Aplt/s Opening Br. at 72. Indeed, V.E. “never disclosed M.E.’s abuse in counseling, and ... M.E. molested [V.E.] during the same time frame.” Aplt.’s Reply Br. at 14-15. Thus, it is unlikely that the jury would have found to be significantly exculpatory V.E.’s non-disclosure of Mr. Ellis’s alleged sexual assault. For this reason alone, Mr.

Stayton’s failure to consult and/or call Dr. Long does not undermine our confidence in the outcome of the trial. In addition, the CCA found that “it is unclear from the record whether Dr. Long was still treating V.E. at the time of the sexual abuse.” Aplt.’s App., Vol. I, at 293. This assessment of the facts is presumed to be correct, and Mr. Ellis has not rebutted its correctness “by clear and convincing evidence.” 28 U.S.C. § 2254(e)(1); see also Aplt.’s App. Vol. Ill, at 246-47 (evincing Ms. Ellis’s testimony that V.E. saw Dr.

Long only when she first moved in with the Ellises, when she was three or four, for a couple of years related to the termination of her biological mother’s parental rights, and then “when the allegations came up,” but that she did not see him for “several years” in between). Accordingly, we accept it as being accurate. Without proof that Dr. Long was even treating V.E. at the time of the assault, we have yet another reason for concluding that Mr. Ellis has failed to establish prejudice related to Mr. Stayton’s decision (assumed to be erroneous) for failing to consult and/or call as an expert witness Dr. Long.

In sum, even applying de novo review, we conclude that Mr. Ellis has failed to show any reasonable probability that, but for Mr. Stayton’s failure to consult and/or call Dr. Long, the result of the proceeding would have been different—viz., Mr.

Ellis has failed to establish that he was prejudiced by this. 4 “In the federal habeas context, the only otherwise harmless errors that can be aggregated are federal constitutional errors, and such errors will suffice to permit relief under cumulative error doctrine only when the constitutional errors committed in the state court trial so fatally infected the trial that they violated the trial’s fundamental fairness.” Matthews v. Workman, 577 F.3d 1175, 1195 n.10 (10th Cir. 2009) (quoting Young v. Sirmons, 551 F.3d 942, 972 (10th Cir. 2008)); accord Littlejohn v. Trammell, 704 F.3d 817, 868 (10th Cir. 2013).

This doctrine applies in the context of Strickland, where counsel’s individual actions have been found to be constitutionally deficient, but nonprejudical. See Cargle v. Mullin, 317 F.3d 1196, 1207 (10th Cir. 2003) (“Strickland errors require us to assess whether there is a reasonable probability that counsel’s deficient performance affected the trial outcome .... [S]uch claims should be included in the cumulative-error calculus if they have been individually denied for insufficient prejudice.” (footnotes omitted)).

However, because we have discerned through the lens of AEDPA only one possible instance of deficient performance (i.e., error)—that is, the assumed Strickland error stemming from counsel’s failure to consult or call as an expert witness Dr. Long—we need not address Mr. Ellis’s argument for cumulative prejudice; there must be more than one error to conduct cumulative-error analysis. See United States v. Willis, 826 F.3d 1265, 1280 (10th Cir. 2016) (“[A]t most, we have assumed one error for purposes of our analysis: that Agent Willis was permitted to vouch for K.M.’s credibility.

Under these circumstances, we reject Mr. Willis’s argument because there are not multiple errors to cumulate.”). 5 Lastly, we address the State’s claim that the federal district court “erred at the outset by sua sponte raising and aggregating prejudice from ineffective-assistance claims Petitioner never raised.” Aplt.’s Opening Br. at 45. The State argues that the court inappropriately considered:

(1) Mr. Stayton’s failure to secure additional forensic testing; and- (2) his agreement to waive the recording of closing arguments. Specifically, the State explains that because Mr. Ellis “never asserted these acts as independent Strickland claims in state or federal court,” they are proeedurally defaulted. Id. Mr. Ellis does not dispute that he never before raised these ineffective-assistance claims, but he contends that the errors on which they are based can individually and cumulatively provide alternative grounds for affirmance.

See Aplee.’s Br. at 25 (“Respondents argue that because the district court’s IAC ruling is somewhat expansive and encompasses additional grounds not raised by Mr. Ellis, the claim as a whole is somehow proeedurally defaulted.... [But] the reviewing court may affirm the judgment on any basis evident in the record” (emphasis added)). Notably, Mr. Ellis does not assert any legal grounds that might conceivably excuse this procedural failing. We conclude that because Mr.

Ellis failed to raise these ineffective-assistance claims before the CCA, the claims are now—and were at the time the district court considered them in granting habeas relief—procedurally defaulted. Consequently, the district court erred in taking these claims into account in granting habeas relief, and we cannot consider them here. As noted, a federal court may grant habeas relief only with respect to federal claims that state prisoners have appropriately exhausted by adequately presenting the substance of the claims to the appropriate state court for review. See, e.g., Hawkins v.

Mullin, 291 F.3d 658, 669 (10th Cir. 2002) (“In order to exhaust his state remedies, a federal habeas petitioner must have first fairly presented the substance of his federal habeas claim to state courts.”); 2 Randy Hertz & James S. Lieb-man, Federal Habeas Corpus Practice & Procedure § 23.3[a], at 1280 (7th ed. 2016) (“As a general rule, a petitioner satisfies the exhaustion requirement if she presents the federal claim to the appropriate state courts in the manner required by state law....”).

Notably, it is not enough that a prisoner has generally presented a Strickland claim of ineffective assistance to the appropriate state court; the prisoner must have presented—and thereby exhausted— the substance of the exact ground of ineffectiveness (i.e., deficient attorney conduct) upon which the prisoner later seeks habeas relief in federal court. See, e.g., Hawkins, 291 F.3d at 669 (“The fact that Hawkins asserted some ineffective-assistance claims in state cour

[…]