Petitioner ‐ Appellant , v. D AN W INKLESKI , Warden,

Respondent ‐ Appellee . ____________________ Appeal from the United States District Court for the Eastern District of Wisconsin. No. 19 ‐ CV ‐ 33 — William E. Duffin , Magistrate Judge . ____________________ A RGUED M AY 12, 2021— D ECIDED S EPTEMBER 21, 2021

____________________ Before F LAUM , H AMILTON , and B RENNAN , Circuit Judges . B RENNAN Circuit Judge David Minnick pleaded no contest in Wisconsin state court to several crimes that resulted from a violent confrontation involving his then wife. He received sentences totaling years of initial con fi nement. Since then, Minnick has brought a series of unsuccessful challenges to his convictions in state and federal courts. The district court denied Minnick’s request for federal ha beas relief under U.S.C. That court decided that Minnick’s trial counsel was not ine ff ective for advising him that a term of not more than ten years of initial con fi nement was likely. The court also ruled that Minnick did not show that any reasonable trial counsel would have advised him of the possibility of withdrawing his no contest pleas before sen tencing. So not o ff ering that argument did not deny Minnick the right to e ff ective postconviction counsel. Although Minnick’s claims could have been analyzed dif ferently—including whether the state court’s decision on his trial counsel’s sentencing advice warranted deference under the Antiterrorism and E ff ective Death Penalty Act (AEDPA), U.S.C. 2254—the correct result was reached. We a ffi rm the denial of habeas relief.

When David Minnick’s wife told him she was leaving him for another man, Minnick retrieved a ri fl e and struck her in the head. She fl ed to her parents’ house across the street, and Minnick followed, fi ring several shots. He tried to break down the door of his in laws’ house, broke windows, and shot in side the house, grazing his father in law. As a result, Minnick was charged in Kenosha County Circuit Court with aggra vated ba tt ery, a tt empted fi rst degree murder, and several counts of fi rst degree reckless endangerment and a tt empted burglary, all while using a dangerous weapon. Minnick initially pleaded not guilty by reason of mental disease or defect, arguing that his actions were rooted in the post ‐ traumatic stress disorder from which he su ff ers. He later withdrew that plea and agreed to plead no contest to the crimes (except for the a tt empted murder charge, which was dismissed and read in) and leave sentencing up to the court. This exposed Minnick to years of initial con fi nement. The presentence investigation report recommended Minnick re ceive between and years of initial con fi nement. At the sentencing hearing, the state asked for years of initial con fi nement, and Minnick’s trial counsel, Laura Walker, asked for years. The trial court sentenced Minnick to years of initial con fi nement followed by years of extended supervi sion. Minnick appealed that sentence, arguing (by his postcon viction counsel Michael Zell) that he should be able to with draw his no contest pleas because he received ine ff ective assistance of counsel. An a tt orney is constitutionally ine ff ec tive if she performs de fi ciently and this performance preju dices her client. See Strickland v. Washington U.S. Minnick argued Walker met this standard because she improperly guaranteed and unreasonably estimated that he would receive a much shorter sentence. The state trial court held a hearing at which Minnick and Walker testi fi ed. 2 The court found Walker credible that she

did not guarantee Minnick a certain sentence length, and that Minnick knew Walker provided only an estimate. At the hear ing, a friend of Minnick’s also testi fi ed he had spoken with Walker, who asked the friend to convince Minnick to take the plea. Walker responded in her testimony that Minnick knew his sentence was ultimately up to the judge and that she al ways quali fi ed her statements to Minnick about the length of his sentence by emphasizing that her estimate was not a guar antee. The state court ruled against Minnick and declined to let him withdraw his no contest pleas. The Wisconsin Court of Appeals a ffi rmed this decision in That court ruled: “Minnick has shown no more than that counsel predicted an outcome that did not come to pass. Her misjudgment of the likely sentence is not a basis for an inef fective assistance of counsel claim, The Wisconsin Supreme Court and the Supreme Court of the United States denied review. Minnick then fi led a collateral a tt ack in state court under Wis. Stat. alleging that Walker was constitutionally ine ff ective because she failed to advise him that he could withdraw his no contest pleas before sentencing if he pro vided a “fair and just reason.” Because Minnick had not raised this claim in his fi rst appeal, he argued the state court could consider it because his postconviction counsel Zell was con stitutionally ine ff ective for not raising it. When Walker learned that the presentence investigation report recom mended a sentencing range exceeding what she had advised, Minnick argued, she should have informed him that he could have moved to withdraw his no contest pleas. For habeas petitioners who allege they received ine ff ective assistance of postconviction counsel because an issue was not raised, Wisconsin employs a “clearly stronger” standard to evaluate counsel’s performance under Strickland See State v. Romero Georgana N.W.2d (Wis. (citing State v. Starks N.W.2d (Wis. Under that stand ard, “the defendant must show that a particular nonfrivolous issue was clearly stronger than issues that counsel did present.” Starks N.W.2d at (internal quotation marks omi tt ed). The state court held a hearing on Minnick’s motion. Zell testi fi ed that the claim he raised about Walker’s sentence esti mate was stronger than arguing Walker was ine ff ective for not advising Minnick about presentence plea withdrawal. Zell ex plained that witnesses could testify about Walker’s advice to Minnick about the likely sentence. In contrast, given the rec ord, Zell had concerns that a plea withdrawal motion was not well founded. The state court agreed and denied this motion. In Minnick appealed that denial to the Wisconsin Court of Appeals. That court noted how Minnick’s appeal was premised on the same scenario it had rejected three years ear lier that Walker’s misjudgment of a likely sentence was a basis for an ine ff ective assistance of counsel claim. The appeals court concluded that Zell had not performed de fi ciently,

ruling that the plea withdrawal argument was not clearly stronger than the argument Zell o ff ered. Zell was aware of the law underlying a plea withdrawal motion, and more factors favored the claim Zell made that Walker had misled Minnick concerning his possible sentence than supported an ine ff ec tiveness claim that Walker failed to counsel Minnick to withdraw his pleas before sentencing. Minnick also was not prejudiced, the appeals court ruled, because sentencing was at the discretion of the trial judge and the presentence inves tigation report did not alter that. The Wisconsin Supreme Court denied review of Minnick’s collateral a tt ack.

Minnick then fi led this federal habeas corpus petition un ‐ der 28 U.S.C. § 2254. His petition alleged:

1. Walker was ine ff ective for saying that if Min ‐ nick accepted the plea deal, he would re ‐ ceive only ten years of initial con fi nement; 2. Walker was ine ff ective for not advising Min ‐ nick that he could withdraw his plea before sentencing; and 3. Zell was ine ff ective for not making the plea withdrawal argument in Minnick’s fi rst ap ‐ peal. On the fi rst claim, the district court concluded that Minnick did not make this argument in the state court of appeals, so it could be considered only because the state conceded that Minnick met the exhaustion requirement for habeas petitions. This led the district court to conclude that the Wisconsin Court of Appeals had not adjudicated the fi rst claim on the merits, and that as a result, its dec ision was not entitled to def erence under AEDPA. Even under de novo review, though, the district court denied the fi rst claim because Walker had not shown bad faith, her estimate was not inconsistent with other cases, and there was no suggestion that her estimate was a gross misjudgment. The district court reviewed the second claim through the lens of ine ff ective assistance of appellate counsel. But, inscru tably, that court determined the government had waived any argument that Minnick had procedurally defaulted this sec ond claim. So the district court essentially analyzed Minnick’s second and third claims together. On these claims, the district court concluded that AEDPA deference applied because the Wisconsin Court of Appeals adjudicated the ine ff ective assistance of postconviction counsel claim on the merits. Ac cording to the district court, Minnick could not have been prejudiced by postconviction counsel’s failure because Min nick could not show he would have moved to withdraw his pleas, or that he would have succeeded in withdrawing his plea if he had been advised to do so. So the district court denied Minnick’s petition, but it is sued a certi fi cate of appealability on two issues: whether Min nick was denied e ff ective assistance of counsel based on Walker’s sentence estimate, and by Zell’s failure to make the plea withdrawal argument.

We fi rst examine whether Minnick is entitled to habeas relief for his claim that his trial counsel Walker was constitu ‐ tionally ine ff ective when she told Minnick he was “likely” to receive no more than ten years of initial con fi nement. The dis ‐ trict court answered this question in the negative. We review that determination, as we do all the district court’s decisions on Minnick’s petition, de novo. Maier v. Smith , 912 F.3d 1064, 1069 (7th Cir. 2019).

On this fi rst claim, the district court concluded that the deferential standard of AEDPA does not apply. A threshold question is whether that decision is correct. AEDPA deference applies only to claims in habeas petitions that were adjudi ‐ cated on the merits in state court. Adorno v. Melvin , 876 F.3d 917, 921 (7th Cir. 2017). We presume a state court adjudicated a claim on the merits unless the state court relied wholly on state law grounds for its decision or expressly declined to con ‐ sider the claim. Win fi eld v. Dorethy , 871 F.3d 555, 560 (7th Cir. 2017). The district court construed Minnick’s claim in the state court that he was “guaranteed” a certain sentence as di ff erent from his fi rst claim here that Walker’s advice as to the likely sentence constituted ine ff ective assistance. If Minnick’s claim was not presented to the state court—as the district court be lieved—it could not have been adjudicated on the merits, as required for AEDPA deference. Win fi eld F.3d at Gen erally, a claim not raised in the state appellate court would be an unexhausted claim that a district court could not consider on collateral review. See U.S.C. But according to the district court, because the state conceded that there was no exhaustion issue, the district court had to consider this claim so it conducted de novo review. We do not agree with the district court that the Wisconsin Court of Appeals did not adjudicate Minnick’s first claim on the merits. The state appeals court not only concluded that Walker did not make a guarantee to Minnick about the length of his sentence, but it expressly ruled that Walker’s misjudg ment of the likely sentence was not ineffective assistance of counsel. For the purpose of habeas review, there is no func tional distinction between Minnick complaining about a “guarantee” and about “advice.” Walker offered her opinion about what sentence Minnick would receive. However her opinion is characterized—a “guarantee,” a “likely outcome,” an “estimate,” or a “prediction”—it was her counsel to Min nick, which the Wisconsin Court of Appeals ruled was not in effective. The state appeals court did not rely wholly on state law grounds for its decision or expressly decline to consider the claim, so the presumption of adjudication on the merits con trols. We conclude that AEDPA deference applies to the Wis consin Court of Appeals’ resolution on Minnick’s first claim.

Resolving this threshold question leads us to recall what AEDPA deference means. Minnick’s appeal comes to us as a collateral a tt ack on a state court judgment. Under AEDPA’s strict standard of review, federal courts defer to state court decisions. By its text, AEDPA precludes a federal court from granting a state prisoner’s habeas petition unless the state court’s merits adjudication “resulted in a decision that was contrary to or involved an unreasonable application of clearly established Federal law, as determined by the Supreme Court of the United States.” U.S.C. 2254(d)(1) (emphases added). “Section 2254(d) re fl ects the view that habeas corpus is a ‘guard against extreme malfunctions in the state criminal justice systems,’ not a substitute for ordinary error correction through appeal.” Harrington v. Richter U.S. (quoting Jackson v. Virginia U.S. n.5 (Stevens, J., concurring in judgment)). This standard is “di ffi‐ cult to meet and highly deferential.” Makiel v. Butler F.3d (7th Cir. (internal quotation marks omi tt ed). Re lief is precluded “so long as fairminded jurists could disagree on the correctness of the state court’s decision.” Harrington U.S. at (internal quotation marks omi tt ed). AEDPA’s strictness is grounded in comity. “AEDPA rec ognizes a foundational principle of our federal system: State courts are adequate forums for the vindication of federal rights.” Burt v. Titlow U.S. “AEDPA’s re quirements re fl ect a ‘presumption that state courts know and follow the law.’” Woods v. Donald U.S. (quoting Woodford v. Viscio tt i U.S. (per cu riam)). “When reviewing state criminal convictions on collat eral review, federal judges are required to a ff ord state courts due respect by overturning their decisions only when there could be no reasonable dispute that they were wrong.” Woods U.S. at This is particularly true when state courts ad judicate ine ff ective assistance of counsel claims. Titlow U.S. at For an ine ff ective assistance of counsel claim such as this, “[t]he federal courts as a whole engage in ‘doubly deferential’ review” under AEDPA. Wilborn v. Jones F.3d (7th Cir. (quoting Knowles v. Mirzayance U.S. Deference is layered upon deference in these cases be cause federal courts must give “both the state court and the defense a tt orney the bene fi t of the doubt.” Titlow U.S. at Even without AEDPA, ine ff ective assistance of counsel claims remain di ffi cult to prove, as “counsel is strongly pre sumed to have rendered adequate assistance and made all signi fi cant decisions in the exercise of reasonable professional judgment.” Strickland , 466 U.S. at 690. “That hill is even steeper” for claims governed by AEDPA. Myers v. Neal , 975 F.3d 611, 620 (7th Cir. 2020). A claim for constitutionally ine ff ective assistance of coun sel during the plea process is governed by the Strickland standard. See La fl er v. Cooper U.S. To be de fi cient, counsel’s performance must be unreasonable such that “counsel was not functioning as the counsel guar anteed the defendant by the Sixth Amendment.” Strickland U.S. at (internal quotation marks omi tt ed). In the plea bargaining context, the prejudice prong is satis fi ed if “there is a reasonable probability that, but for counsel’s errors, the [petitioner] would not have pleaded guilty and would have insisted on going to trial.” La fl er U.S. at (internal quo tation marks omi tt ed). There is “a strong presumption that counsel’s conduct falls within the wide range of reasonable professional assistance.” Strickland U.S. at

A petitioner’s habeas claim is considered against the last reasoned state court decision on the merits. Wilson v. Sellers S. Ct. Makiel F.3d at For this fi rst claim, that is the decision of the Wisconsin Court of Ap peals on direct appeal. That court rested its analysis on the de fi cient performance prong of Strickland so we con fi ne our analysis to that prong. Minnick argues that by applying a categorical rule, the reasoning of the Wisconsin Court of Appeals was contrary to Strickland According to Minnick, this rule was that an a tt or ney’s sentence miscalculations could never result in constitu tionally de fi cient performance. True, the Supreme Court has instructed that Strickland requires a “circumstance speci fi c reasonableness inquiry,” Roe v. Flores Ortega U.S. But the reasoning of the Wisconsin Court of Appeals was very similar to the standard this court has applied in anal ogous cases, and which we do not read as a categorical rule. This court has stated that “a mistaken prediction is not enough in itself to show de fi cient performance” because “the sentencing consequences of guilty pleas (or, for that ma tt er, guilty verdicts) are extraordinarily di ffi cult to predict.” United States v. Barnes F.3d (7th Cir. see also Bridgeman v. United States F.3d (7th Cir. (“[C]ounsel’s alleged miscalculation, standing alone, could never su ffi ce to demonstrate de fi cient performance unless the inaccurate advice resulted from the a tt orney’s failure to un dertake a good faith analysis of all of the relevant facts and applicable legal principles.”). Still, a miscalculation might constitute de fi cient performance if it is a gross miscalculation and there is evidence of the a tt orney’s bad faith. Id. see also United States v. Martinez F.3d (7th Cir. In its opinion, the Wisconsin appeals court did not include the phrases “in itself” or “standing alone.” Even so, its decision can reasonably be interpreted as describing a sim ilar standard. Nothing in Minnick’s case required the state ap peals court to explain that a miscalculation might be grounds for de fi cient performance if there was evidence of bad faith. Further, Minnick did not present any evidence—besides the incorrectly estimated sentence—to suggest that Walker per formed de fi ciently. The Wisconsin Court of Appeals determined that Minnick knew at sentencing that the state court could impose the max imum sentences, as well as that the term of initial con fi nement which Walker opined Minnick would receive was not a guar antee. Minnick cites Walker’s newness to legal practice (two years of experience when she represented Minnick) and the di ff erence between the initial con fi nement term she estimated and the range in the presentence investigation report years versus years) and contends Walker acted unreasona bly in her prediction. But Walker had worked on hundreds of criminal cases before Minnick’s, including shooting cases, and the federal district court provided examples of cases in which defendants facing similar charges to Minnick’s re ceived sentences that were within the range that Walker pre dicted. And the state court did not conclude or suggest facts that showed bad faith on Walker’s part in making her predic tion. Even more, Walker gave reasons for her prediction in the state court. She believed that because Minnick had no crimi nal record, was years of age, and was a military veteran that he would receive a sentence far less severe than the max imum. Although Supreme Court decisions provide the relevant lodestar for cases under AEDPA review, if a state court’s rea soning largely follows circuit precedent, that is persuasive ev idence the state court did not improperly apply the Supreme Court caselaw. See Thill v. Richardson F.3d (7th Cir. (“This ambiguity is not enough to demonstrate that the court applied a standard contrary to clearly established federal law.”) (internal quotation marks omi tt ed). Because the Wisconsin Court of Appeals’ resolution of this case was con sistent with this court’s precedent, its decision was not con trary to, or an unreasonable application of Strickland What is more, that a sentence miscalculation is not enough by itself to demonstrate de fi cient performance is consistent with Strick land ’s presumption of deference to a tt orneys. The state appeals court did not unreasonably apply Strickland when it concluded that Walker’s misjudgment of Minnick’s likely sentence length, without more, could not es tablish de fi cient performance. Under AEDPA, that decision is due deference. So relief is not available to Minnick on his fi rst claim.

The district court handled Minnick’s second and third claims in tandem. Because the Wisconsin Court of Appeals viewed Minnick’s second claim against Walker through the prism of his third claim for ine ff ective assistance of postcon viction counsel, the district court reasoned that it must do so as well. Yet, incongruously, the district court also rejected the argument that Minnick’s second claim against Walker was procedurally defaulted, concluding that the government had waived that argument by failing to mention it in its opening brief. That reasoning is contradictory. If Minnick’s second claim was not procedurally defaulted, the question is whether the Wisconsin Court Appeals adjudicated that claim on the merits. If, however, this second claim was procedurally de faulted, then only Minnick’s third claim for ine ff ective assis tance of postconviction counsel would remain. What cannot be true is that Minnick both did not procedurally default his second claim and that we view it only in the context of his third claim. We address this contradiction by examining only Min nick’s third claim for ine ff ective assistance of postconviction counsel. Although this is what the district court did as well, we do so for di ff erent reasons. The third claim is included in the certi fi cate of appealability, our charge for this appeal. In addition, although the district court concluded that the gov ernment waived the procedural default argument, that court essentially treated Minnick’s second claim—that Walker was ine ff ective for not advising Minnick that he could withdraw his plea before sentencing—as procedurally defaulted by not viewing the claim on its own terms and by granting the cer ti fi cate of appealability on the third claim. The parties also ap pear to treat the second claim as procedurally defaulted, and they focus their arguments on the third claim. So we turn to the merits of that third claim.

Minnick’s claim of ine ff ective assistance of postconviction counsel is also subject to the double deference of Strickland and AEDPA. See Shaw v. Wilson F.3d (7th Cir. Strickland standard governs ine ff ective assistance of ap pellate counsel). Initially, Minnick argues that the Wisconsin Court of Appeals applied the wrong standard when it limited its de fi ciency inquiry to whether the claim Zell failed to press was clearly stronger than the argument he raised. Per Min nick, that standard applies only to postconviction counsel’s intentional decisions, not oversight, which Minnick submits was the case here based on Zell’s testimony. Nevertheless, the Supreme Court and this court employ that same standard. See Davila v. Davis, S. Ct. (“Declining to raise a claim on appeal, therefore, is not de fi cient performance unless that claim was plainly stronger than those actually presented to the appellate court.”); Ramirez v. Tegels F.3d (7th Cir. (noting that if an a tt orney “abandoned a nonfrivolous claim that was both ‘obvious’ and ‘clearly stronger’ than the claim[s] that [s]he actually presented, [her] performance was de fi cient, unless [her] choice had a strategic justi fi cation”); Makiel F.3d at These authorities do not admit of the distinction Minnick advances between intent and negligence in postconviction counsel’s decision making. See id. (at evidentiary hearing appellate counsel “testi fi ed that she could not remember spe ci fi cally considering and rejecting these issues.”) (internal quotation marks omi tt ed). As the district court recognized, the clearly stronger standard does not impermissibly add to the Strickland test, but instead applies the de fi ciency prong in the appellate context where counsel is encouraged to winnow the issues selected for appeal. By applying the clearly stronger standard, the Wisconsin Court of Appeals did not apply a test that was contrary to clearly established law. Because the state appeals court properly applied this standard, we limit our analysis to the de fi cient performance prong of the Strickland evaluation. “[P]roving that an unraised claim is clearly stronger than a claim that was raised is generally di ffi cult ‘because the com parative strength of two claims is usually debatable.’” Makiel F.3d at The Wisconsin Court of Appeals concluded that the unraised plea withdrawal claim was not clearly stronger than the challenge to Walker’s advice on Minnick’s likely sentence, giving two reasons. The fi rst was its evalua tion of the underlying claim on Walker’s sentencing advice. The second was Zell’s understanding that Minnick would have di ffi culty showing he had a “fair and just reason” for withdrawing his plea before sentencing. See State v. Jenkins N.W.2d (Wis. (applying “fair and just reason” standard to plea withdrawals before sentencing). Under Zell’s reasoning, Walker did not perform de fi ciently in failing to ad vise Minnick to fi le such a motion. To the contrary, Minnick would simply be expressing a desire to go to trial, which Wis consin precedent says does not qualify as such a fair and just reason. See State v. Garcia N.W.2d (Wis. (“A fair and just reason is some adequate reason for defendant ʹ s change of heart other than the desire to have a trial.”) (in ternal quotation marks omi tt ed). Minnick does not directly confront either of these reasons. Rather, he contends his case is akin to those in which counsel fail to consult their criminal defense client about the possibil ity of taking an appeal. But in those decisions, the strength of the underlying claim is not the main consideration; instead “counsel has a constitutionally imposed duty to consult with the defendant about an appeal when there is reason to think either that a rational defendant would want to appeal (for example, because there are nonfrivolous grounds for appeal), or that this particular defendant reasonably demonstrated to counsel that he was interested in appealing.” Flores Ortega U.S. at We are not persuaded by Minnick’s analogy to an a tt orney failing to enter a notice of appeal. There, the proceedings end if the defendant does not take an appeal; nothing is left for the trial court to do. Here, the only change in circumstances Min nick suggests should have triggered Walker’s duty to advise him to withdraw his pleas was the distribution of the presen tence investigation report. But such a report does not bind the sentencing court. Everyone knew the latitude at sentencing a ff orded to the prosecution—and, more importantly, to the court—and Minnick con fi rmed this when he entered his no contest pleas. Moreover, in instances too numerous to count, a presentence investigation report recommends a sentence longer than the term the sentencing court imposes. Under Minnick’s argument, if a presentence investigation report or the possibility of a lengthy sentence is su ffi cient basis to with draw a plea, that would invite substantial gamesmanship. Minnick also argues that the unraised plea withdrawal claim was clearly stronger than Zell’s challenge to Walker’s advice on Minnick’s likely sentence because a presentencing plea withdrawal motion is reviewed under a more deferential standard than if made after sentencing. But just because the standards di ff er based on the timing of the motion does not suggest which ine ff ective assistance of counsel claim is stronger. What ma tt ers is whether Walker, when the motion had to be made, acted reasonably in not advising Minnick to make it. As noted above, there are reasons to conclude Walker acted within the scope of competent counsel in not advising Minnick to so move. It is not clear that any presentence plea withdrawal motion that Minnick made would have succeeded. At most, it is de batable whether such a motion is clearly stronger than the sentencing advice argument Zell advanced. This case is not like Shaw in which an essentially frivolous su ffi ciency of the evidence claim was raised, rather than counsel pressing a po tentially meritorious statutory claim. Shaw F.3d at The state appeals court could reasonably conclude that Walker would not think there was a fair and just reason for Minnick to withdraw his no contest pleas at that stage of the proceedings. No Wisconsin authority expressly holds that this circumstance constitutes a fair and just reason to with draw a plea. That rend ers weaker the unmade claim. Because it is debatable whether the unraised claim was clearly stronger, the state appeals court’s conclusion cannot be un reasonable. Whether he would have succeeded in withdrawing his pleas is beside the point, Minnick maintains, because he was denied legal process. But at sentencing, the state court was not bound by any previous recommendations, and Minnick could argue that his sentence should be lower than the state or presentence investigation report recommended. Minnick ad mits he was fully aware that the state court had full sentencing discretion and that the ultimate disposition was up to the court. These facts all suggest that the claim a tt orney Zell raised on the sentencing advice was arguably stronger than an unraised plea withdrawal request. The Wisconsin Court of Appeals did not unreasonably ap ply the Strickland de fi cient performance prong when it con cluded that a plea withdrawal claim was not clearly stronger than the argument that his postconviction counsel advanced. That decision is due the further deference of AEDPA. For these reasons, Minnick was not incorrectly denied relief on this claim.

* * * Minnick’s postconviction claims fall short, even more so given the deferential AEDPA standard. Accordingly, we A F FIRM the judgment of the district court denying his request for habeas relief.