dissenting:
I respectfully dissent. Even if the district court were to determine on remand that all of Soto-Lopez’s allegations are true, those allegations are not sufficient to satisfy either the deficient performance or the prejudice prongs of Strickland v. Washington, 466 U.S. 668, 104 S.Ct. 2052, 80 L.Ed.2d 674 (1984). That conclusion would have been correct before the Supreme Court’s recent decision in Lafler v. Cooper, — U.S. -, 132 S.Ct. 1376, 182 L.Ed.2d 398 (2012), and it remains correct today.
I. Performance
Our decision in Turner v. Calderon, 281 F.3d 851 (9th Cir.2002), which the majority properly relies on in its disposition, holds that, in the plea context, “[c]ounsel cannot be required to accurately predict what the jury or court might find, but he can be required to give the defendant the tools he needs to make an intelligent decision.” Turner, 281 F.3d at 881; see also Lafler, 132 S.Ct. at 1391 (“[A]n erroneous strategic prediction about the outcome of a trial is not necessarily deficient performance.”). In Turner, the defendant alleged that his lawyer was ineffective because, among other things, the lawyer told him that the worst sentence he faced was 15 years to life, and that his case was not a “death penalty” case. As a result, the defendant turned down a second-degree murder plea offer and went to trial, where he was convicted of first-degree murder and robbery and later sentenced to death. Id. at 879-81. We held that the defendant “was informed that he was subject to the death penalty, and of the plea offer,” in contrast to cases where an attorney failed to advise his client of a plea offer or misled his client about the law. “That counsel and [the defendant] chose to proceed to trial based on counsel’s defense strategy and presumably sincere prediction that the jury would not award a sentence of death, does not demonstrate that Turner was not fully advised of his options.” Id. at 881.
In this case, Soto-Lopez had the information he needed to make an informed decision. Soto-Lopez’s federal defender told him of the government’s fast-track plea offer, and told Soto-Lopez that if he rejected the fast-track plea offer, his Sentencing Guidelines range would be much higher. Soto-Lopez nonetheless made a strategic decision to reject the binding 48-month sentence and roll the dice on a shorter sentence. The district court already found that Soto-Lopez’s decision was not the result of “any affirmative misrepresentations of law or fact” by De Oli-vas, or by interference with Soto-Lopez’s “previous understanding of the potential consequences of failing to obtain a new *149plea deal.”1 Critically, Soto-Lopez declined the district court’s offer to withdraw his guilty plea even after he learned of De Olivas’s suspension from practice.
The majority dismisses as irrelevant the advice Soto-Lopez received from his federal defender, reasoning that the court should look only to what De Olivas did or did not do. Mem. Disp. at 145-46 n.l. In the majority’s view, this narrow focus is warranted because De Olivas told Soto-Lopez that his federal defender had given him bad advice. Id. However, we cannot evaluate a lawyer’s performance in a vacuum, but instead must look to the totality of circumstances. See Strickland, 466 U.S. at 688, 104 S.Ct. 2052 (“[T]he performance inquiry must be whether counsel’s assistance was reasonable considering all the circumstances.”). A defendant who is as familiar with the courts as Soto-Lopez,2 and who receives conflicting advice from two lawyers, is highly unlikely to take at face value the advice of only one of them. Whatever De Olivas said, Soto-Lopez knew that he ran some risk of receiving a higher sentence if he rejected the fast-track plea deal.
To be sure, the record contains little evidence that De Olivas had a strong basis for promising Soto-Lopez a 24- to 30-month sentence. But that fact shows only that De Olivas was an imperfect lawyer, not that he rendered constitutionally deficient assistance. See Turner, 281 F.3d at 881 (defendant has no “right to receive an accurate prediction of the outcome of his ease”); Lafler, 132 S.Ct. at 1391 (“[A]n erroneous strategic prediction about the outcome of a trial is not necessarily deficient performance.”).
The majority tries to fill this gap by pointing to De Olivas’s suspension from practice in the Southern District of California for unprofessional conduct. Mem. Disp. at 146. While the suspension certainly raises doubts about De Olivas’s general competence, it does not show that he was deficient in this case. As the district court said, there is no “direct connection between Mr. De Olivas’ advice and the Standing Committee’s petition. The petition mentions that the Standing Committee became aware of ‘acts of alleged incompetence and malpractice,’ but states that the Standing Committee chose not to include such matters in the petition, instead limiting the petition to unprofessional conduct.”
The majority also asserts that “there is no evidence that De Olivas had any experience” securing 24- to 30-month plea offers or sentences for other defendants. Mem. Disp. at 146. However, the fact that De Olivas subsequently obtained a 30-month sentence, following a fast-track plea deal, for a similarly-situated defendant shows at the very least that such a sentence was legally possible.
Finally, we must be mindful of our obligation to give the lawyer the benefit of the doubt. See Strickland, 466 U.S. at 689, 104 S.Ct. 2052 (“A court must indulge a strong presumption that counsel’s conduct *150falls within the wide range of reasonable professional assistance.”); Premo v. Moore, — U.S. -, 131 S.Ct. 733, 741, 178 L.Ed.2d 649 (2011) (explaining that “strict adherence to the Strickland standard [is] all the more essential when reviewing the choices an attorney made at the plea bargain stage.... In determining how searching and exacting them review must be, habeas courts must respect their limited role in determining whether there was manifest deficiency in light of information then available to counsel.”).
II. Prejudice
The majority concludes that Soto-Lopez “has alleged sufficient facts to show prejudice” because, if Soto-Lopez had never met De Olivas, he “would not have rejected the government’s plea,” he and the government would have finalized the fast-track deal, and the district court would not have “had cause to exercise its discretion to reject” the deal. Mem. Disp. at 146^47.
The record does not support these conclusions. Although Soto-Lopez asserted in his 28 U.S.C. § 2255 motion that he “would have accepted the Government’s 48-month plea deal” if “De Olivas had not lied” to him, he declined the opportunity to withdraw his open guilty plea after he learned of De Olivas’s suspension from practice. In addition, nothing in the district court’s decision suggests that it necessarily would have accepted the fast-track plea deal for Soto-Lopez, who previously had been convicted of an aggravated felony.
III. Conclusion
Even if we take all of Soto-Lopez’s allegations to be true, there is insufficient evidence that he received ineffective assistance of counsel. I therefore would affirm the district court.
. Compare Nunes v. Mueller, 350 F.3d 1045, 1049, 1054 (9th Cir.2003) (finding ineffective assistance where counsel misinformed his client that government’s plea offer was twice as long as what the government was actually offering); United States v. Blaylock, 20 F.3d 1458, 1465-66 (9th Cir.1994) (finding deficient performance where counsel did not inform defendant of plea offer); United States v. Day, 969 F.2d 39, 42-44 (3d Cir.1992) (finding deficient performance where counsel affirmatively misrepresented the maximum sentence and did not tell the defendant that he would be classified as a career offender at sentencing).
. Soto-Lopez previously had been removed from the United States after being convicted of an aggravated drug trafficking felony.