concurring in the judgment:
I agree with the result reached by the majority and its analysis of the exhaustion issues. I write separately to consider the claim that Judge Weinstein, in taking the rare step of issuing a certificate of appeal-ability, invited this Court to consider: “petitioner’s claim that he was denied the effective assistance of trial counsel because [counsel] failed to have the petitioner assessed by a psychologist or psychiatrist and failed to investigate or prosecute an extreme emotional disturbance defense.” I think the majority wrongly implies (without explicitly saying as much) that Murden failed to satisfy both prongs of Strickland’s inquiry. See Strickland, 466 U.S. at 687, 104 S.Ct. 2052. In my view, as discussed below, the record demonstrates that Murden proved in his first § 440 motion filed in state court and on habeas review in the district court that trial counsel’s performance was not objectively reasonable. I concur in the mandate, however, because I agree with the majority that Murden failed to show prejudice. Only for that reason would I hold that the ineffective assistance of counsel claim cannot succeed.
*200The majority’s discussion of the objective reasonableness of trial counsel’s performance is limited to the conclusion that it was entirely reasonable for counsel to pursue self-defense as a primary defense. Maj. Op. at 3830. But this conclusion assumes that any deficiency in pursuing the EED defense is therefore excused. I do not believe that Strickland condones deficient performance in investigating one defense if another also happens to be available. See, e.g., Michael v. Crosby, 430 F.3d 1310, 1320 (11th Cir.2005) (“Counsel’s duty to investigate requires that counsel conduct a substantial investigation into any of his client’s plausible lines of defense.”) (citation and internal quotation marks omitted); see also Eze v. Senkowski, 321 F.3d 110, 135-36 (2d Cir.2003) (ineffective assistance may be demonstrated where counsel performs competently in some respects but not others). In addition, in focusing only on the self-defense issue, the majority does not consider the range of Murden’s allegations properly before us: that trial counsel did not consult with Mur-den regarding an EED defense, but should have done so; that counsel did not interview witnesses for that defense, but should have done so; and that counsel did not explore Murden’s past attempted suicide, but should have done so.
Murden’s allegations are supported by the record. Pre-trial transcripts show that counsel requested time to review the file on the January 9, 1991 calendar call, and to “sit with the defendant on one more occasion to prepare it.” 1/9/91 Hearing Tr. at 6. Counsel’s notes indicate that he met with Murden for one hour each on January 9 and on January 23, the day of the next calendar call. At jury selection five days later (after a three-day vacation), counsel was unable to name the witnesses he intended to call until he consulted with his client. Although counsel testified at the December 2004 hearing before the district court that he did not remember whether he had met with Murden at any other point before trial, he provided no evidence to establish that he had done any investigation into the EED defense.
Nothing in the record contradicts Mur-den’s assertions that counsel did not consult with him about the EED defense or otherwise pursue it. Even setting aside what has been asserted about counsel’s pre-trial preparation, however, his actions at trial proclaim that he did not prepare or investigate the EED defense. Counsel neither mentioned the EED defense in his opening statement, nor on cross-examination did he ever inquire of witnesses about Murden’s emotional state. He did not call any witnesses to bolster the defense. He introduced no evidence of Murden’s psychiatric history or prior suicide attempt. Unfortunately, it is little wonder that his request for an EED charge at the conclusion of his case was denied for lack of evidence supporting it.
“[CJounsel has a duty to make reasonable investigations or to make a reasonable decision that makes particular investigations unnecessary.” Strickland, 466 U.S. at 691, 104 S.Ct. 2052. It is readily apparent that counsel here gave no thought to laying the foundation for the EED defense until the end of the trial. Had counsel adequately prepared for the case, he would have thought to raise the defense from the beginning. Had he done some investigation, it would have revealed that Murden had a psychiatric history and that he had threatened suicide and to hurt his previous girlfriend when she threw him out of their apartment. We also know that at least one psychologist believes Murden was suffering from a mood and personality disorder and could not manage his emotions. It is logical to conclude that timely investigation would have revealed similar expert testimony for use at the trial.
*201I can conceive of no strategic reason for an attorney who wishes to assert an EED defense to decide not to: (1) consult with his client about the defense; (2) interview witnesses with the defense in mind; and (3) explore a documented past suicide attempt made by his client. See id. at 690, 104 S.Ct. 2052 (a court “must judge the reasonableness of counsel’s challenged conduct on the facts of the particular case, viewed as of the time of counsel’s conduct”). Moreover, if, despite what the record clearly indicates to me, counsel did in fact do some investigation into this defense, I do not understand what strategy would compel counsel to avoid presenting the defense in opening statements or introducing any evidence or testimony supporting the defense. See Eze, 321 F.3d at 136 (noting that, when strategic considerations do not account for a challenged decision by counsel, the deficient-performance prong of Strickland will likely be met); Hart v. Gomez, 174 F.3d 1067, 1071 (9th Cir.1999) (finding it “inconceivable” that strategy played a part in counsel’s decision not to introduce readily-available evidence that would have corroborated a particular line of defense). Accordingly, to the extent the state court and district court found trial counsel’s performance effective despite this lack of preparation, and to the extent the majority implicitly adopts these findings, I respectfully disagree.