concurring.
I fully concur with the lead opinion’s conclusion that Lyons’s counsel was constitutionally ineffective because of the lawyer’s failure to inform Lyons that the state could appeal the decision of the trial judge to sentence him as a juvenile. My reason for writing separately is to explain, in what I regard as an extremely close case, why I find the well-written dissenting opinion less persuasive than the lead opinion.
Like the lead opinion, I believe that the Michigan Court of Appeals’s determination that Lyons’s counsel performed competently was an unreasonable application of Strickland v. Washington, 466 U.S. 668, 104 S.Ct. 2052, 80 L.Ed.2d 674 (1984). The dissent, in contrast, finds counsel’s performance to be constitutionally competent, a conclusion with which I disagree for several reasons.
First, in reaching the conclusion that Lyons’s counsel provided competent representation, the dissent does not address the failure of counsel to inform Lyons that the state could appeal the trial judge’s sentencing decision. The dissent instead focuses solely on whether counsel’s strategy was reasonable in light óf his knowledge of the trial judge’s sentencing tendencies, thereby overlooking the very omission that in my opinion rendered the performance of Lyons’s counsel deficient.
Second, because Lyons’s counsel never considered the possibility of the state appealing the trial judge’s sentencing decision, his advising Lyons to plead guilty to first-degree murder cannot be considered a reasonable strategic decision to which deference should be accorded. Strickland, 466 U.S. at 690-91, 104 S.Ct. 2052 (explaining that “strategic choices made after thorough investigation of law and facts relevant to plausible options are virtually unchallengeable; and strategic choices made after less than complete investigation are reasonable precisely to the extent that reasonable professional judgments support the limitations on investigation”). As the lead opinion notes, a reasonable attorney would have considered the possibility of an appeal before advising Lyons to plead guilty under the circumstances of the present case.
Finally, I believe that the dissent’s consideration of the strength of the state’s case against Lyons in assessing whether counsel provided competent representation, including the suggestion that his only available defense was a hope for jury nullification, focuses on the wrong issue. Lyons, in my opinion, was entitled to be advised of the possibility of an appeal before deciding whether to plead guilty, because without that knowledge his decision was not based upon all of the relevant facts. Although the Supreme Court has held that the Sixth Amendment does not require a correct assessment of the risks and benefits of pleading guilty as opposed to proceeding to trial, it recognizes that counsel must at least be aware of such risks, especially where the lack of awareness directly impacts the reasoning behind whatever advice is provided. See McMann v. Richardson, 397 U.S. 759, 770, 90 S.Ct. 1441, 25 L.Ed.2d 763 (1970) (“Questions like these [relating to the risks of trial] cannot be answered with certitude; yet a decision to plead guilty must necessarily rest upon counsel’s answers, uncertain as they may be.”); United States v. Hanley, 906 F.2d 1116, 1121 (6th Cir.1990) (recognizing that “a reversal for ineffective assistance would be in order” if the defendant had pled guilty solely in reliance upon his counsel’s erroneous advice “that he would be eligible for parole in one-third the time he received for sentence”).
I also agree with the lead opinion that if Lyons’s counsel had informed Lyons that the state could appeal the trial judge’s *602sentencing decision, a reasonable probability exists that Lyons would have proceeded to trial rather than plead guilty. In considering Strickland’s “prejudice” prong, we are not bound by AEDPA’s deferential standard of review, because the Michigan Court of Appeals never reached the issue of prejudice when it determined that counsel’s performance was not deficient. See Gonzales v. McKune, 247 F.3d 1066, 1075-76 (10th Cir.2001) (explaining that because the state court considered only two of three requirements necessary to establish a violation of Brady v. Maryland, 373 U.S. 83, 83 S.Ct. 1194, 10 L.Ed.2d 215 (1963), a de novo standard of review applied to the third element). I therefore believe that the dissent’s application of AEDPA’s “objectively unreasonable” test to the prejudice component of Strickland is not appropriate. Williams v. Taylor, 529 U.S. 362, 409, 120 S.Ct. 1495, 146 L.Ed.2d 389 (2000) (clarifying the meaning of an “unreasonable application” of law under 28 U.S.C. § 2254(d)(1)).
The dissent concludes that Lyons’s effort to withdraw his guilty plea constituted an attempt to get “a second bite at the apple,” and that Lyons entered his guilty plea voluntarily and with full knowledge of the possible sentence that he faced. I respectfully disagree. Although Lyons did not seek to withdraw his guilty plea during the state appellate process, the record reflects that he was not advised by his attorney that the state had even appealed. His first knowledge of the state’s appeal was approximately two years later, when a relative read about the Michigan Court of Appeals’s decision ordering him to be sentenced as an adult. Thus, the dissent’s inference that Lyons was “gaming the system” is unwarranted under these circumstances.
With respect to whether Lyons entered his guilty plea knowingly and intelligently, I believe that the dissent underestimates the significance of a possible appeal in any calculation that Lyons or his counsel made. Lyons might have believed that he had, say, an 80% chance of being sentenced as a juvenile by the particular trial judge before whom he pled guilty. If he had been advised that the state could appeal, however, he might have evaluated the odds of the trial judge’s determination being sustained on appeal as only 50-50. The risk of the trial judge’s sentencing decision being reversed on appeal would, in this example, reduce the ultimate likelihood of Lyons being sentenced as a juvenile from 80% to only 40%.
We are unable, of course, to determine the exact odds that Lyons believed he faced prior to pleading guilty, but the above example illustrates the significance of failing to consider the possibility of the state appealing. The fact that Lyons was aware of the maximum penalty he faced does not, therefore, negate the significance of the information that neither he nor his counsel considered. Perhaps Lyons would have pled guilty even if he had known that the trial judge’s sentencing decision was not the only hurdle that he had to overcome in order to be sentenced as a juvenile. But he was entitled to be advised of all the risks that he faced before choosing to plead guilty and foregoing his right to a jury trial.
Because I am not confident that Lyons would have pled guilty if his counsel had informed him that his sentence could be appealed, I concur with the lead opinion in concluding that the decision of the Michigan Court of Appeals was an unreasonable application of Strickland. 28 U.S.C. § 2245(d)(1) (providing that a writ of habe-as corpus is available for a person in state custody if the state court’s adjudication of the petitioner’s claim “resulted in a decision that ... involved an unreasonable ap*603plication of ... clearly established Federal law, as determined by the Supreme Court of the United States”); Hill v. Lockhart, 474 U.S. 52, 59, 106 S.Ct. 366, 88 L.Ed.2d 203 (1985) (“[I]n order to satisfy the ‘prejudice’ requirement, the defendant must show that there is a reasonable probability that, but for counsel’s errors, he would not have pleaded guilty and would have insisted on going to trial.”); Strickland, 466 U.S. at 694, 104 S.Ct. 2052 (“A reasonable probability is a probability sufficient to undermine confidence in the outcome.”).