dissenting.
Awkal’s primary defense to the capital charge was that he was not guilty by reason of insanity. Any remotely competent attorney knows that an insanity defense relies mainly upon expert testimony regarding the defendant’s mental state. But Awkal’s counsel’s presentation of expert psychological testimony at the guilt phase sounds like the opening of a bad joke: “Three defense experts walk into court. One got his degree from a mail-in university, is not licensed, and cannot testify. One is well-credentialed but testifies against the defendant. And the third is not certified in psychiatry.” Or perhaps the better analogy is an episode of The Three Stooges.1 But, however one wishes to describe it, it is clear what counsel’s performance was not: the kind of representation constitutionally required in a capital case. I find the majority’s opinion to the contrary unpersuasive, so I respectfully dissent.2
A. Clearly Established Federal Law
To succeed on his ineffective assistance of counsel claim, Awkal must show that his “counsel’s representation fell below an objective standard of reasonableness.” Strickland v. Washington, 466 U.S. 668, 688, 104 S.Ct. 2052, 80 L.Ed.2d 674 (1984). Awkal must then demonstrate prejudice by “showing] that there is a reasonable probability that, but for counsel’s unprofessional errors, the result of the proceeding would have been different. A reason*650able probability is a probability sufficient to undermine confidence in the outcome.” Id. at 694, 104 S.Ct. 2052. Moreover, as our review of this case is controlled by AEDPA, Awkal must “show that the [Ohio Supreme Court] applied Strickland to the facts of his case in an objectively unreasonable manner.” Bell v. Cone, 535 U.S. 685, 699, 122 S.Ct. 1843, 152 L.Ed.2d 914 (2002).
B. Ineffective Assistance of Counsel
1. Deficient Performance
Awkal’s sole affirmative defense to the murder charges was that he was not guilty by reason of insanity. During the guilt phase, Awkal’s counsel presented three experts to testify in support of this defense: Dr. Paul Hewitt, Dr. Magdi S. Rizk, and Dr. Eileen S. McGee.
Dr. Hewitt testified first but, “while attempting to qualify Dr. Hewitt to give an opinion on that issue, counsel established that Dr. Hewitt was not competent to engage in forensic psychology.” State v. Awkal, 76 Ohio St.3d 324, 667 N.E.2d 960, 971 (1996). Most glaringly, Dr. Hewitt was not licensed.3 Because the trial court determined that he was not qualified to enter an opinion on the sole guilt-phase issue — whether Awkal was sane at the time that he committed the attack — the court excluded Dr. Hewitt’s testimony and instructed the jury to disregard what testimony he had already provided. Dr. Hewitt was thus never able to opine on Awkal’s sanity.
Next, the defense called Dr. Rizk, a court-appointed forensic psychiatrist who had originally evaluated Awkal to determine whether he was competent to stand trial. Defense counsel had Dr. Rizk testify on direct examination regarding his conclusion about Awkal’s competency to stand trial. (One may recall that the main issue during the guilt phase was Awkal’s sanity at the time of the attack; the jury was not there to decide whether Awkal could be tried.) Then, on cross-examination by the State, Dr. Rizk opined that, even though Awkal was initially incompetent to stand trial, he was nevertheless sane at the time of the attack.
Finally, Dr. McGee testified that Awkal was not sane at the time of the murders. Dr. McGee was a licensed pediatrician who, though well-trained, “was not yet board certified in psychiatry, had no experience in forensic psychiatry, and had been practicing psychiatry for only one year.” Id. at 972. The state court stated that Dr. McGee’s “opinion may have been diminished by her lack of certification and inexperience.” Id. Thus, her testimony was of very limited value when following Dr. Rizk’s devastating sanity opinion.
Counsel’s choice and then presentation of experts is baffling. Not only did he call two experts whose credibility was weak or whose testimony was inadmissible at the guilt phase, but counsel’s one certified psychiatrist undermined Awkal’s sole defense by testifying unequivocally on cross-examination that Awkal was sane at the time of the attack. Counsel’s handling of Dr. Rizk, therefore, was not only troubling, but disastrous. Either counsel was completely unaware of how Dr. Rizk would testify regarding Awkal’s sanity, which shows a glaring lack of basic trial skills as Dr. Rizk had already issued a written report finding Awkal to have been sane, or counsel knew that Dr. Rizk would opine that Awkal was sane at the time of the attack but did not seek to address it on direct examination, in *651which case the decision to call him as a witness is senseless.4 What little value there is in having a psychiatrist testify that Awkal had previously been found incompetent to stand trial was obviously overshadowed by having that same psychiatrist opine that Awkal was sane at the time of the attack. The only plausible conclusion is that counsel’s actions with regard to Dr. Rizk at the guilt phase, combined with counsel’s reliance upon two other inherently unreliable “experts,” constituted the deficient performance necessary to establish ineffective assistance of counsel.
The majority, of course, disagrees and puts forth various arguments in support of its conclusion that counsel’s actions were the result of strategy instead of incompetence. So let us examine some of these arguments in more detail.
The majority asserts that this case is more similar to Morales v. Mitchell, 507 F.3d 916 (6th Cir.2007), in which we rejected an ineffective assistance argument, than Combs, in which we found ineffective assistance. I strongly disagree. The facts of Awkal’s case are materially indistinguishable from Combs. Like Awkal, Combs did not contend that he had not committed the attack. Combs, 205 F.3d at 273. Instead, Combs’s defense was that “he was too intoxicated from alcohol and drugs to form the requisite intent to kill the [two] women or to have committed the killings with prior calculation and design.” Id. Combs’s counsel presented testimony that Combs was intoxicated, but he called only one expert witness on this point. That witness was a psychologist who, like Dr. Rizk, testified on cross-examination that, contrary to Combs’s defense, Combs had killed intentionally. Id. We held that Combs received ineffective assistance of counsel because,
[r]egardless of whether Combs’s counsel should have known or instead actually knew [the expert’s] opinion regarding Combs’s intent, however, counsel’s decision to put him on the stand was objectively unreasonable.... [N]ot only did [the expert’s] testimony destroy any hope of a successful intoxication defense, but it also helped the prosecution to establish one of the elements of its case in chief. Quite simply, this testimony was completely devastating to the defense, and counsel’s decision to present it was objectively unreasonable.
Id. at 288. In sum, we found that calling an expert who, on cross-examination, directly contradicted the defendant’s sole defense constituted ineffective assistance of counsel. Dr. Rizk’s testimony followed exactly the same script.
The majority attempts to distinguish Awkal’s case from Combs by pointing to the fact that Awkal’s counsel presented three expert witnesses regarding the insanity issue rather than only one.5 This *652distinction may carry some weight if it were true that the other two experts offered by Awkal’s counsel had added any value. But instead, one of the expert’s testimony was stricken due to his lack of qualification, and the other expert was a relatively novice, uncertified psychiatrist, opining on the mental state of a man put on trial for his life. In essence, the majority reasons that this case is different from Combs because Combs involved only one terrible decision but this case involved three. This is a distinction to be sure, but it cuts in favor of Awkal, not the State. The majority’s reasoning to the contrary leads to the apparent, but vexing, takeaway that, in the Sixth Circuit, numerous instances of incompetence may aggregate into competence, that defense counsel may insulate themselves from ineffective assistance claims simply by making multiple errors.
The majority’s reliance on Morales is also misplaced. In Morales, a defendant charged with murder claimed that he was not guilty by reason of insanity. 507 F.3d at 921, 936. Morales alleged that his voluntary intoxication rendered him unable to “appreciate the consequences of his acts or refrain from committing those acts, or understand the difference between right and wrong” at the time of the crime. Id. at 921 (internal quotation marks omitted). In spite of the fact that intoxication played a role in Morales’s defense, Morales’s counsel called a forensic psychologist to the stand whom Morales’s counsel knew “did not believe that Morales had a viable voluntary-intoxication defense.” Id. at 936.
On direct examination, the psychologist provided favorable testimony, outlining Morales’s psychological problems, which included a “personality disorder” that, when “combined with [Morales’s] alcohol dependency and abuse, deprived him of the ability to refrain from or understand the wrongfulness of his actions.” Id. at 938. On cross-examination the psychologist conceded that, legally, “voluntary intoxication alone [was] not a basis for the insanity defense.” Id. The psychologist went on to testify that “voluntary intoxication was not the only contributing factor in the murder,” citing Morales’s other psychological problems. Id.
Faced with Morales’s claim that his counsel was ineffective for calling the psychologist, the Morales Court determined that the psychologist’s statement regarding the legal standard for insanity was plainly “unhelpful,” but concluded that the case was distinguishable from Combs for several reasons. Id. Most significantly for Awkal, the Court determined that, unlike Combs, the psychologist’s testimony in Morales did not completely undermine Morales’s only affirmative and complete defense. Rather, Morales’s insanity defense was based on a combination of his psychological problems and his intoxication, and the psychologist testified to that effect. Id. Thus, “the substance of [the expert’s] testimony” in Morales “supported the defense theory.” Id.
Here, however, the facts are strikingly different. Dr. Rizk’s testimony that Awkal was sane at the time of the murders completely undermined Awkal’s only affirmative and complete defense — that he was insane when he killed his wife and brother-in-law. Therefore, Dr. Rizk’s statement that Awkal was sane was not just “unhelpful”, cf. id., it was fatal.
*653After seeking, unconvincingly, to distinguish Combs, the majority next engages in that time-honored tradition of ineffective assistance of counsel cases — conjuring up possible explanations to make counsel’s actions seem like anything other than buffoonery. The majority is correct that the Supreme Court has indicated that reviewing courts should seek to contrive an explanation for counsel’s actions, even if it is mere speculation. But, at the very least, these imagined explanations should be plausible and not contradicted by the record.
The majority first suggests that perhaps defense counsel called Dr. Rizk so that he could testify as to Awkal’s competence to stand trial and other mental issues in order to “plant a seed of doubt” regarding his conclusion that Awkal was sane. (Ante at 641.) Relatedly, the majority suggests that perhaps counsel called Dr. Rizk in order to “take the sting out” of Dr. Rizk’s damaging sanity opinion. (Id. at 642.) Perhaps counsel called Dr. Rizk to preempt what would surely have been a decision by the prosecution to call Dr. Rizk in its rebuttal case. (Id. at 643.) Or perhaps counsel called Dr. Rizk in an effort to bolster credibility by acknowledging weaknesses. (Id.) All of these hypotheses rest on the theory that counsel called Dr. Rizk in order to address what counsel knew would be damaging testimony, either by preemptively discrediting the testimony or managing its “sting.” These explanations could be sound trial strategies, if it were the case that defense counsel actually addressed the damaging testimony in his examination of the witness. But here, counsel never attempted to elicit from Dr. Rizk that he was of the opinion that Awkal was sane at the time of the crime; that gem came out on the prosecution’s terms during cross-examination. Nor did counsel seek on redirect examination to discredit Dr. Rizk’s sanity opinion with his testimony on direct examination regarding Awkal’s mental issues and incompetence to stand trial.
In short, the record does not support the majority’s post hoc justifications for counsel’s actions and, in fact, directly contradicts them. It cannot plausibly be speculated that counsel was seeking to “plant a seed of doubt” about Dr. Rizk’s sanity opinion unless there is some indication that counsel actually did anything in his examination of Dr. Rizk to discredit that opinion. Here, Awkal’s counsel neither suggested doubt on direct examination nor discredited the sanity opinion on redirect, so counsel clearly was not questioning Dr. Rizk with an eye toward the damaging sanity opinion. Similarly, as to the “take the sting out” conjecture, this could not possibly have been counsel’s strategy when counsel did not actually elicit the damaging testimony.6 Far from taking the sting out by eliciting the damaging testimony on direct examination, counsel compounded the sting by allowing the prosecution to elicit the opinion on cross-examination after defense counsel had relied on the witness and bolstered his credibility with the jury.
This unfolding of events is devastating as a matter of trial dynamics. In essence, counsel examined Dr. Rizk on direct, without any preemptive impeachment, and then tendered the witness to the prosecution, leaving the jury with the impression that Awkal’s own counsel endorsed Dr. Rizk’s entire testimony. Thus, when the prosecution elicited on cross-examination that Dr. Rizk believed Awkal to have been *654sane at the time of the attack, the jury likely assumed either that defense counsel also endorsed this testimony or had hoped to suppress it. Counsel reinforced this conclusion when he ignored the sanity opinion on redirect examination. This is not sound trial strategy, and the majority’s attempts to justify counsel’s actions simply do not hold water.
The rest of the majority’s speculation about possible reasons for counsel’s actions is similarly unpersuasive. Perhaps counsel wanted Dr. Rizk’s favorable testimony regarding Awkal’s initial incompetence to stand trial. (Ante at 642.) But whether Awkal was at one point incompetent to stand trial (as he had obviously been subsequently found competent) was at most only marginally probative as to whether he was sane at the time of the attack. It cannot be sound trial strategy to seek weak and likely irrelevant favorable testimony when you know that doing so is sure to bring in highly damaging and relevant testimony. Under Ohio law effective at the time of Awkal’s trial, “[a] person is ‘not guilty by reason of insanity’ ... only if he proves ... that at the time of the commission of the offense, he did not know, as a result of severe mental disease or defect, the wrongfulness of his acts.” Ohio Rev. Code § 2901.01(N) (emphasis added). Thus, the only portion of Dr. Rizk’s testimony relevant at the guilt phase was his opinion regarding Awkal’s sanity at the time of the attack. Other testimony regarding Awkal’s competence to stand trial or matters related to his family background had, at best, no impact on the guilt phase and seems unlikely to have constituted any kind of trial strategy.
On a related note, perhaps counsel sought to keep Dr. Rizk’s favorable testimony but downplay the bad by bracketing his testimony with that of Dr. Hewitt and Dr. McGee. (Ante at 643.) However, this also is not plausible explanation based on the record. As stated above, counsel did not purposefully elicit any damaging testimony from Dr. Rizk, so he could not have been meaning to hide the damaging testimony between the favorable testimony of Drs. Hewitt and McGee. Moreover, Dr. Rizk’s damaging sanity opinion was unquestionably a heavy piece of evidence, so it would take some rather strong brackets to contain it. Counsel could not have expected Dr. Hewitt to provide the necessary front-end support in the jury’s mind, and, indeed, the court struck Dr. Hewitt’s testimony in its entirety. Counsel also could not have expected the testimony of Dr. McGee — a former pediatrician and uncertified psychiatrist with no experience in forensic psychiatry — to provide the back-end support needed to remedy Dr. Rizk’s devastating testimony.
Also related, perhaps counsel called Dr. Rizk in Awkal’s defense to remove one arrow from the prosecution’s quiver so that the prosecutor would have only one expert instead of two on rebuttal. (Id.) Like many of the majority’s other hypotheses, however, this speculation also incorrectly presumes that counsel called Dr. Rizk for the purpose of eliciting his damaging sanity opinion. As shown above, the record contradicts that line of reasoning. If counsel had called Dr. Rizk with an eye towards addressing his sanity opinion, one would expect that he would have actually elicited it.
In sum, the majority valiantly but vainly attempts to invent an explanation for counsel’s actions that could be characterized as strategic, as none of the proffered explanations are plausible when one looks at what defense counsel actually did. The majority attempts to respond to my reasoning, but at no point does it offer an explanation for what I see as the most damning fact for its position: counsel never addressed *655Dr. Rizk’s sanity opinion. At some point, Ockham’s Razor must apply — the simplest answer is usually the correct one. Here, in light of the majority’s numerous failed attempts to justify counsel’s actions, the simplest answer is that counsel did not have the foggiest idea what he was doing when he decided to put Drs. Hewitt, Rizk, and McGee on the stand. There was no grand plan.
Instead, it appears that counsel was merely throwing whatever evidence he could find at the jury and hoping something would stick. Confirming this conclusion is the fact that, when the trial judge asked whether counsel had a strategic reason for calling Dr. Rizk, counsel responded, “I don’t have a response at this time,” (J.A. 2064), and went on to posture that he had no duty to explain himself but of he course he had a reason. This, to me, is code for “no.” Counsel’s actions clearly were not the sound trial strategy that Strickland seeks to protect.7 I therefore would find that counsel rendered constitutionally ineffective assistance.
2. Prejudice
In addition to showing that his counsel’s performance was deficient, Awkal has shown that the deficiency prejudiced him. The State presented other evidence that Awkal was legally sane at the time of the murders and that Awkal acted with prior calculation by pointing to, among other things, the fact that Awkal had diapers and baby supplies in his car at the time of the murders, that he was arrested with a full clip of bullets, and that he had taken steps to move to another area. (J.A. at 2444-50.) Awkal countered this evidence by explaining that he had intended to commit suicide, but his mental instability caused him to break with reality and kill his wife and brother without intent or knowledge of his actions. But, as in Combs, counsel’s decision to call Dr. Rizk as an expert “destroyed] any hope of a successful [insanity] defense” and “was completely devastating to the defense.” Combs, 205 F.3d at 288; see also id. at 290 (“The testimony of the sole defense expert that Combs, though intoxicated, nevertheless acted with purpose and intent was obviously damaging to the defense. Furthermore, [the expert’s] testimony provided the State with its most powerful evidence of purpose.”).
The majority claims that no prejudice resulted from Dr. Rizk’s testimony and even offers that, since the prosecution would have called Dr. Rizk anyway, the trial’s outcome would have been the same regardless of who presented him as a witness. However, that the jury would have heard Dr. Rizk’s testimony at some point does not address the reality that Awkal was prejudiced by his counsel’s presentation and handling of the expert testimony on mental state. The damage was not that testimony countering Awkal’s sole defense was presented at some point in the trial, because, as the majority notes, “it is always desirable in planning one’s trial strategy to anticipate disclosures by the prosecution of highly unfavorable and damaging evidence against the accused ... you can be sure the prosecutor will!” (Ante at 642 (citing Bailey & Rothblatt).) The harm was in the fact that Awkal’s own counsel called three expert witnesses, the most credible of which completely destroyed Awkal’s defense, while the other *656two were not at all persuasive. And, as stated above, counsel conducted the examination of Dr. Rizk in a manner that would lead the jury to conclude that Awkal’s team found Dr. Rizk’s testimony reasonable. Thus, when Dr. Rizk testified on cross-examination that Awkal was sane at the time of the attack, a reasonable jury would have assumed that Awkal accepted this conclusion.8 While it is a normal part of the trial for the prosecution to attack a defendant’s defense, it is highly unusual for defendant’s own counsel to undermine his only defense through his presentation of experts.
Furthermore, reactions from the judge and prosecutor when defense counsel called Dr. Rizk only buoy the conclusion that counsel’s decision was highly unusual and harmful. Before Dr. Rizk took the stand, the judge, at the prompting of the prosecutor, asked Awkal’s counsel if he had “strategic reasons for calling Dr. Rizk.” (J.A. at 2064-66.) Defense counsel replied that he had no response at that time, presumably because there was no strategy. The prosecutor openly wondered why counsel made this choice and, in closing arguments highlighted the damaging decision, telling the jury, “Then [defense counsel] called one of our witnesses, Dr. Rizk, to the stand. I am still trying to figure that one out.” (J.A. at 2457.) Once Awkal’s own expert testified that Awkal was sane at the time of the murders, the jury could not conceivably conclude — with the sole backing of an uncertified psychiatrist — that Awkal was actually insane at the time of the crime. Thus, it is reasonable to assume that defense counsel’s decision to call Dr. Rizk sealed Awkal’s fate.
Furthermore, Awkal has shown prejudice even though, or perhaps because, his counsel called an uncertified psychiatrist who testified that Awkal was legally insane at the time of the murders. During the course of a normal trial, prosecution and defense experts often disagree and it is up to the jury to suss out the conflict. But counsel’s actions rendered this trial distinctly abnormal when, without any apparent explanation or purpose, he offered an expert who agreed with the prosecution’s expert on the most critical fact question in the entire case. Calling a second “expert” of marginal persuasiveness could not possibly undo the damage caused by counsel’s handling of Dr. Rizk.
C. Ohio Supreme Court’s Application of Strickland to These Facts
In addition to showing that his counsel’s performance was deficient and that he was prejudiced by this deficiency, Awkal has “show[n] that the [Ohio Supreme Court] unreasonably applied Strickland to the facts of his case in an objectively unreasonable manner.” Cone, 535 U.S. at 699, 122 S.Ct. 1843. After considering Awkal’s ineffective assistance of counsel claims, the Ohio Supreme Court concluded that “Awkal was not deprived of a fair trial by his trial counsel. Given the evidence supporting his conviction, he was not prejudiced by what occurred at trial.” Awkal, 667 *657N.E.2d at 971. The Ohio Supreme Court explained its holding:
Counsel also called Dr. Rizk to testify during the guilt phase. Dr. Rizk is adequately qualified and has testified in numerous other similar circumstances. However, Dr. Rizk testified that Awkal was sane at the time of the murders. This testimony obviously damaged Awkal’s affirmative defense that he was not sane when he committed the murders. Yet, portions of Dr. Rizk’s testimony assisted the defense, including testimony about religion as a basis for Awkal’s marital problems, his medication levels, and his hallucinations.
Counsel concluded Awkal’s affirmative defense by calling Dr. McGee, a psychiatrist. Dr. McGee was not yet board certified in psychiatry, had no experience in forensic psychiatry, and had been practicing psychiatry for only one year. Dr. McGee testified that Awkal, as evidenced by his hallucinations involving his wife in his cell in Dayton, had broken with reality at the time of the murders. This break with reality impaired his ability to know right from wrong at the time of the murders. Although her opinion may have been diminished by her lack of certification and inexperience, Dr. McGee supported Awkal’s affirmative defense.
Thus, of the three doctors called to testify for the defense, the testimony of one was stricken from the record, one gave an opinion contradicting Awkal’s affirmative defense but also gave other evidence that assisted that defense, and one testified that Awkal was not mentally responsible for his acts. However, Drs. McGee and Hewitt were called to testify and did testify during the penalty phase of the trial, giving pro-defense opinions. Dr. Sarny also gave a pro-defense opinion in the penalty phase.
Awkal’s counsel obviously had some plan in mind. Dr. Hewitt conceivably could have been allowed to testify as an expert witness, and Dr. Rizk did make an earlier finding that Awkal was incompetent to stand trial. In hindsight it appears that Awkal may have been better served to call only Dr. McGee during the guilt phase, and call her and the other defense doctors during the penalty phase, if the trial would reach that stage. However, the end result of tactical trial decisions need not be positive in order for counsel to be considered “effective.” We do not believe the record establishes that Awkal’s attorneys were ineffective at trial.
Awkal, 667 N.E.2d at 971-72 (citation omitted).
This analysis constitutes an unreasonable application of Strickland. The state trial court obviously recognized the extent of the harm caused by defense counsel’s decision to call Dr. Rizk — it concedes that calling Dr. Rizk was not a good idea in hindsight. But the court nevertheless disregards the prejudice by boldly proclaiming that “Awkal’s counsel obviously had some plan in mind.” This strikes me more as wishful thinking rather than legal analysis based on the evidence in the record. The court assumes that counsel had a plan because, otherwise, it would have to answer tough questions and face uncomfortable realities about the state of indigent, and especially capital, representation. Though Strickland is, for better or for worse, extremely solicitous of lawyers, it is not a license to rely on baseless assumptions to escape hard questions.
The court also mischaracterized Dr. Rizk’s testimony as somehow helpful to Awkal. This too overlooks reality. Though Dr. Rizk testified about Awkal’s family and psychiatric history, that testimony was neither here nor there at the *658guilt phase where the only relevant issue was Awkal’s sanity at the time of the attacks. Ohio Rev.Code § 2901.01(N). Thus, Dr. Rizk’s other statements at the guilt phase were, at best, neutral and could not be said to “assist” Awkal’s defense. Dr. Rizk’s relevant testimony destroyed Awkal’s only affirmative defense, a fact that the Ohio Supreme Court ignored when it characterized Dr. Rizk’s testimony as part of a larger strategy. Thus, the court unreasonably applied Strickland to Awkal’s case.
Because the Ohio Supreme Court’s application of Strickland was objectively unreasonable, and because Awkal received ineffective assistance of counsel at the guilt phase of the trial, I would grant habeas relief and order a new trial within ninety days.
I respectfully dissent.
. When I say that counsel's handling of expert testimony on the sanity issue reminds me of The Three Stooges, I do not mean to cast aspersions at the intelligence of the three doctors. I have no reason to believe that they are anything other than diligent, intelligent individuals, though counsel clearly could have found two doctors with more impressive credentials for purposes of supporting this most important issue. Rather, The Three Stooges comes to mind from a director's point of view. The director of The Three Stooges created a product that came across to his audience as bungling. Based on my review of the record, Awkal's counsel engaged in a similar task.
. Awkal brings several habeas claims, alleging errors at both the guilt and the sentencing phases of his trial. The majority addresses and holds against Awkal on all of these claims. Although I do not necessarily agree with the majority’s disposition as to each claim, I focus solely on the guilt-phase ineffective assistance of counsel claim. Because Awkal received constitutionally ineffective assistance of counsel at the guilt phase of his trial, he should receive a new trial. Therefore, I need not address his other claims.
. Though it is unclear how important it was to the trial court’s decision to strike Dr. Hewitt, it was also revealed that he had received his Master’s Degree and Doctorate from Century University in California, a mail-in university.
. The majority suggests that the fact that Awkal’s counsel possessed Dr. Rizk’s report before calling him as a witness demonstrates that the decision to do so was strategic. (Ante at 641.) First, regardless of whether Awkal’s counsel knew or should have known of Dr. Rizk's opinion, the decision to call an expert witness that contradicted Awkal’s only affirmative defense amounts to deficient performance. See Combs v. Coyle, 205 F.3d 269, 288 (6th Cir.2000) (“Regardless of whether Combs's counsel should have known or instead actually knew [the expert's] opinion regarding Combs’s intent, however, counsel's decision to put him on the stand was objectively unreasonable.”). Secondly, counsel's performance was not deficient solely by virtue of having called Dr. Rizk to testify, but rather, having called Dr. Rizk to testify, counsel was deficient in failing to impeach or even address Dr. Rizk's devastating sanity opinion.
. The majority also distinguishes Combs on the basis that it is pre-AEDPA. Though this distinction is maybe material for purposes of Combs’s habeas ruling, it does not affect *652Combs's constitutional finding of deficient performance under Strickland on materially indistinguishable facts.
. I am sure Mssrs. Bailey and Rothblatt, see ante at 642 (citing Bailey and Rothblatt, Successful Techniques for Criminal Trials), would not read this transcript and conclude that counsel was engaging in the “take the sting out” maneuver.
. If counsel's unexplainable actions, when viewed in light of the entire record, do indeed qualify as the strategy that Strickland says is constitutionally sufficient to satisfy the Sixth Amendment’s guaranty of competent counsel, strategy has become “stragedy” and Strickland itself needs to be revisited. The Constitution demands better, and we should demand better.
. The majority claims that this is merely unsupported speculation. I would first note that the very nature of ineffective assistance cases is engaging in post hoc speculation. The majority’s opinion is rife with it. Furthermore, I do not believe that my conclusion is unsupported. Having eliminated, based on the flow of counsel's direct and redirect examination, the prospect that counsel called Dr. Rizk for the purposes of debunking his sanity opinion, the only conclusion remaining is that counsel called Dr. Rizk to endorse his testimony. One cannot endorse some of a witness's testimony but reject other aspects of that testimony unless one actually engages in the exercise of rejecting the unfavorable aspects of the testimony. Here, counsel never took issue with Dr. Rizk’s sanity opinion.