Paula Bennett v. Shawn Brewer

RECOMMENDED FOR FULL-TEXT PUBLICATION Pursuant to Sixth Circuit I.O.P. 32.1(b) File Name: 19a0260p.06

UNITED STATES COURT OF APPEALS FOR THE SIXTH CIRCUIT

PAULA BENNETT, ┐ Petitioner-Appellant, │ │ > No. 17-1574 v. │ │ │ SHAWN BREWER, Warden, │ Respondent-Appellee. │ ┘

Appeal from the United States District Court for the Eastern District of Michigan at Ann Arbor. No. 5:12-cv-12054—John Corbett O’Meara, District Judge.

Argued: December 4, 2018

Decided and Filed: October 8, 2019

Before: DAUGHTREY, GIBBONS, and GRIFFIN, Circuit Judges.

_________________

COUNSEL

ARGUED: Sinéad Redmond, UNIVERSITY OF MICHIGAN LAW SCHOOL, Ann Arbor, Michigan, for Appellant. Linus Banghart-Linn, OFFICE OF THE MICHIGAN ATTORNEY GENERAL, Lansing, Michigan, for Appellee. ON BRIEF: Sinéad Redmond, Melissa M. Salinas, UNIVERSITY OF MICHIGAN LAW SCHOOL, Ann Arbor, Michigan, for Appellant. Linus Banghart-Linn, Andrea M. Christensen-Brown, OFFICE OF THE MICHIGAN ATTORNEY GENERAL, Lansing, Michigan, for Appellee. Paula R. Bennett, Ypsilanti, Michigan, pro se.

No. 17-1574 Bennett v. Brewer Page 2

_________________

OPINION _________________

MARTHA CRAIG DAUGHTREY, CIRCUIT JUDGE. Petitioner Paula Bennett was convicted of aiding and abetting first-degree murder, in violation of Michigan Compiled Laws § 750.316(1)(a), for the shooting of Stephanie McClure, who was killed by Bennett’s then- boyfriend, Kyron Benson. Bennett is serving life in prison without the possibility of parole. She seeks a writ of habeas corpus pursuant to 28 U.S.C. § 2254, arguing that her trial counsel rendered constitutionally ineffective assistance. Bennett also claims that she suffered from ineffective assistance of her appellate counsel, who failed to challenge trial counsel’s deficient performance on direct appeal. Respondent, Warden Shawn Brewer, opposes Bennett’s petition, primarily on procedural grounds. Relying on Michigan Court Rule 6.508(D), he contends that because Bennett did not initially raise her claim of ineffective assistance of trial counsel on direct appeal, she is barred from doing so now. Further, he argues that Bennett’s challenge to the competence of appellate counsel does not excuse her procedural default because the underlying claims lack merit. The district court ruled that Bennett could not meet the high standard required to establish constitutionally ineffective assistance and denied her a writ. Because our review is constrained by the Anti-Terrorism and Effective Death Penalty Act of 1996 (AEDPA), we affirm the district court's judgment.

FACTUAL AND PROCEDURAL BACKGROUND

In October 2007, 18-year-old Paula Bennett lived with her boyfriend, Kyron Benson, who, by all accounts, physically and emotionally abused Bennett throughout their relationship. For a period of time that fall, the victim in this case, Stephanie McClure, stayed at Bennett and Benson’s apartment. When she moved out, McClure took with her Benson’s video game console and a number of other items belonging to the couple. In the days after Benson became aware of the missing items, he began making threatening statements to Bennett and their friends about his plans to confront McClure, saying that he was going to shoot and kill her. Meanwhile, on October 5, Bennett reported the theft to the police.

No. 17-1574 Bennett v. Brewer Page 3

Early the following morning, after a night spent out with their friends, Bennett and Benson announced that they were going to reclaim their belongings from McClure. The couple’s first stop was a nearby grocery store, where they picked up Benson’s friend, Michael Larvaidan. Larvaidan later testified at trial that, while at the store, Benson repeated his desire to get back his things, reiterated his threats on McClure’s life, and showed Larvaidan a gun he was carrying. Despite the gun and the repeated threats, Larvaidan—like most of Bennett’s other friends—said he did not believe Benson was serious about harming McClure.

With Larvaidan in the car, Bennett told Benson that McClure lived in a local trailer park, and Benson drove them there. Bennett then pointed out McClure, who was standing outside her trailer and Benson got out of the car. Bennett moved into the driver’s seat, and drove around the area for a few minutes. Eventually, Bennett and Larvaidan heard shots ring out. Bennett immediately began crying. When she saw Benson running from McClure’s trailer, Bennett drove to meet him, and, after collecting Benson, the three left the scene. On their return to their apartment, Bennett reported to her friends that Benson shot McClure. Police arrested Bennett and Benson a few days later.

Bennett was charged with first-degree murder on a theory of aiding and abetting because she had directed Benson to McClure’s trailer park, pointed her out, and then driven the getaway car. She and Benson—who was similarly charged, as the principal defendant—were tried jointly, before separate juries. At trial, Larvaidan offered eye-witness testimony that Benson shot McClure. Other witnesses testified regarding Benson’s threats on McClure’s life, although all but one of these witnesses also testified that they did not consider Benson’s threats credible or believe that he posed an actual danger to McClure. In the face of this overwhelming evidence, Benson’s attorney adopted a feeble theory of defense: that the police and prosecutor single- mindedly focused on Benson as the culprit and failed to investigate two other potential suspects. Bizarrely, Bennett’s counsel followed suit, arguing that the jury could not convict Bennett unless it found that Benson had in fact killed McClure.

However, proving Benson’s guilt alone was not sufficient to convict Bennett. The second element of the crime required the state to show that, at the time that she offered the alleged aid, Bennett intended McClure’s death or knew of Benson’s intent to kill McClure. To

No. 17-1574 Bennett v. Brewer Page 4

establish Bennett’s awareness of Benson’s intention, the prosecution relied on Benson’s repeated threats. Bennett’s counsel did little to rebut this inference, despite testimony from multiple witnesses that Benson’s threats did not seem serious. Nor did Bennett’s attorney highlight the various aspects of Bennett’s behavior that suggested she lacked the requisite mens rea— specifically, that she alerted the police about the missing items less than 24 hours before Benson killed McClure, that Bennett unnecessarily involved Larvaidan in the supposed plot, and that she displayed extreme emotional distress after hearing the gun go off. Instead, Bennett’s attorney focused his efforts on cooperating with Benson’s counsel, even going so far as to cross-examine some witnesses whose testimony bore little or no relevance to Bennett’s case. Indeed, Bennett’s counsel concentrated so intently on Benson’s innocence that he did not proffer a single witness in support of her defense. Predictably, the juries convicted both defendants, and Bennett was sentenced to life imprisonment without the possibility of parole—the mandatory sentence for first-degree murder, regardless of the amount of aid or encouragement offered by an accessory to a crime. See Mich. Comp. Laws §§ 750.316(1) and 767.39.

Bennett appealed her conviction to the Michigan Court of Appeals, raising claims of prosecutorial misconduct and insufficient evidence in support of her conviction. The appeals court found that although there was “some evidence” presented at trial suggesting that Bennett did not intend to aid Benson in killing McClure, “there was considerable evidence from which the jury could have inferred that Bennett knew of Benson’s intent.” People v. Bennett, 802 N.W.2d 627, 634 (Mich. Ct. App. 2010). Further, the court found that the record established only that Bennett was “unenthusiastic” about Benson’s plan and did not negate her awareness of Benson’s objective. Id. The court also rejected Bennett’s prosecutorial-misconduct claim and therefore affirmed her conviction. Id. at 635-37.

In an impassioned dissent, Judge Douglas Shapiro disagreed with the majority’s determination of Bennett’s insufficient-evidence claim and, among other things, argued sua sponte that Bennett’s trial counsel was constitutionally ineffective. Specifically, Judge Shapiro argued that the overwhelming evidence against Benson “mandated the defense that Bennett did not believe that Benson would murder the victim and that she did not want any harm to come to the victim.” Id. at 645 (Shapiro, J., dissenting). Bennett and Benson’s defenses, he pointed out,

No. 17-1574 Bennett v. Brewer Page 5

were inherently antagonistic to one another. Id. As evidence of the obviousness of this dynamic, Judge Shapiro highlighted Benson’s defense counsel’s motion to sever the two cases, noting that Bennett’s counsel made no argument at the hearing related to that motion. Id. Further, he argued that defense counsel’s strategy of focusing on Benson’s innocence made Bennett’s outcome reliant on Benson’s and undercut Bennett’s argument that she was unaware of Benson’s intention. Id. at 646. Finally, the dissent questioned Bennett’s trial counsel’s failure to impeach Breanna Kandler, the one witness who thought Benson’s threat had any validity, and his failure to request an instruction on accessory after the fact. Id. at 646–47. Judge Shapiro concluded that, at minimum, Bennett should have received an evidentiary hearing pursuant to People v. Ginther, 212 N.W.2d 922 (Mich. 1973). Id. at 648.

After the Michigan Supreme Court denied review of Bennett’s appeal, People v. Bennett, 796 N.W.2d 75 (Mich. 2011), she filed a pro se petition in federal district court seeking a writ of habeas corpus based on her claims of insufficient evidence, inadequate jury instructions, and prosecutorial misconduct, as well as the ineffective representation of trial counsel highlighted by Judge Shapiro and, finally, a claim of ineffective assistance of appellate counsel. The district court held the petition in abeyance, pending Bennett’s exhaustion of her ineffective-assistance claims. Bennett v. Warren, No. 5:12-CV-12054, 2012 WL 6737768 (E.D. Mich. Dec. 28, 2012).

Accordingly, Bennett returned to state court, still pro se. The state trial court denied her appeal for post-conviction relief pursuant to Michigan Court Rule 6.508(D)(3), which bars relief for claims that “could have been raised on appeal from the conviction.” People v. Bennett, No. 07-024733-02-FC, slip op. at *2 (Mich. Cir. Ct. Crim. Div. Cty. of Wayne Dec. 10, 2013). The court also found that the claim as to appellate counsel’s incompetence could not serve as a “good cause” to overcome the procedural bar, pursuant to Michigan Court Rule 6.508(D)(2). Id. But this ruling was based on an error—the trial judge mistakenly believed that Bennett’s appellate-counsel claim had been raised and denied in a prior adjudication. It had not.1

1On collateral review, the state trial court determined that a prior court adjudication barred Bennett from relying on her appellate counsel’s errors as a shield from procedural default. A careful look at the record reveals the trial court’s error. Bennett first raised her appellate-counsel claims in her initial federal petition for habeas corpus, filed in March 2012, and then again in front of the state trial court in February 2013. With the exception of the federal district court, which temporarily deferred review, no court had the opportunity to adjudicate the claim before

No. 17-1574 Bennett v. Brewer Page 6

Regardless of this error and despite the procedural ground for denying relief to Bennett, the trial court engaged in a substantive discussion of Bennett’s ineffective-assistance claims and found that she failed to meet the prejudice prong of the test announced in Strickland v. Washington, 466 U.S. 668, 694–95 (1984). Id. at *3-5. In the court’s view, it was not “reasonably probable that the results of the proceeding would have been different had it not been for counsel’s error.” Id. at *3. The Michigan Court of Appeals and the Michigan Supreme Court both affirmed the trial court’s ruling. See People v. Bennett, 866 N.W.2d 432 (Mich. 2015).

Having exhausted her post-conviction remedies in state court, Bennett returned to federal court with an amended petition. She abandoned her prosecutorial-misconduct claim and her insufficient-evidence claim but reasserted her jury-instruction claim as well as her claims regarding the ineffectiveness of her trial and appellate counsel. In addition to habeas relief, Bennett sought an evidentiary hearing. The district court denied both requests, finding no error in the trial court’s jury instructions and determining that none of her ineffective-assistance claims met Strickland’s high standard. Bennett v. Warren, No. 5:12-01-12054, 2017 WL 1344775 at *4-13 (E.D. Mich. Apr. 12, 2017). The district court did, however, recognize that other reasonable jurists might disagree and granted Bennett a certificate of appealability. Id. at *13.

DISCUSSION

Standard of Review

In habeas proceedings, we typically apply a mixed standard of review, considering a district court’s legal conclusions and mixed questions of law and fact de novo and reviewing its factual findings for clear error. King v. Westbrooks, 847 F.3d 788, 795 (6th Cir. 2017). However, when a district court bases its factual determinations on trial transcripts and court records, as it did here, we review its conclusions de novo. Dando v. Yukins, 461 F.3d 791, 796 (6th Cir. 2006).

Our review of Bennett’s claims is limited by AEDPA. King, 847 F.3d at 795. If a state court has adjudicated a claim on the merits, we will not grant a writ of habeas corpus unless the

the trial court considered it. Thus, no earlier decision precludes Bennett from raising that claim here to overcome the procedural bar imposed by Rule 6.508(D)(3). See McFarland v. Yukins, 356 F.3d 688, 699 (6th Cir. 2004).

No. 17-1574 Bennett v. Brewer Page 7

state court’s decision was “contrary to, or involved an unreasonable application of, clearly established Federal law, as determined by the Supreme Court of the United States” or “was based on an unreasonable determination of the facts in light of the evidence presented in the State court proceeding.” 28 U.S.C. § 2254(d)(1)-(2). Thus, unlike in a de novo review of the merits of a petitioner’s ineffective-assistance claims, “[t]he pivotal question is whether the state court’s application of the Strickland standard was unreasonable.” Harrington v. Richter, 562 U.S. 86, 101 (2011). The Supreme Court has admonished that “an unreasonable application of federal law is different from an incorrect application of federal law.” Id. (quoting Williams v. Taylor, 529 U.S. 362, 410 (2000)). This standard is a heavy burden for a petitioner to satisfy. “[E]ven a strong case for relief does not mean the state court’s contrary conclusion was unreasonable.” Id. at 102. Indeed, “so long as fairminded jurists could disagree on the correctness of the state court’s decision,” the law precludes this court from granting habeas relief. Id. at 101 (internal quotation marks and citation omitted).

The warden argues, however, that we should not reach the merits of Bennett’s claims. Relying on Michigan Court Rule 6.508(D)(3)(a) and (b), the warden urges us to dismiss her claims as procedurally defaulted because she did not raise them in her direct appeal. Bennett responds that she has good cause for her failure to raise the claims initially: her appellate counsel’s ineffective assistance. The way forward in evaluating these competing contentions proves to be somewhat circular.

We have held that “[i]neffective assistance of counsel can supply the cause that, together with prejudice, would excuse a procedural default.” McFarland v. Yukins, 356 F.3d 688, 699 (6th Cir. 2004). To surmount that obstacle, however, a petitioner must meet the standard laid out in Strickland by showing both that her appellate counsel’s failure to raise her ineffective- assistance-of-trial-counsel claims was objectively unreasonable and that the failure prejudiced her on appeal. See id. Because failure to raise a meritless claim is not ineffective assistance, Bennett must establish a reasonable probability that her trial-counsel claims would have prevailed on direct appeal. Mapes v. Coyle, 171 F.3d 408, 413–14 (6th Cir. 1999); see also McFarland, 356 F.3d at 699. If no reasonable probability exists that raising Bennett’s trial- counsel claims would have altered the outcome of her direct appeal, her appellate counsel’s

No. 17-1574 Bennett v. Brewer Page 8

decision to omit the claims was neither deficient nor prejudicial and cannot in any instance provide good cause as required by Michigan Court Rule 6.508(D)(3)(a). Cf. McFarland, 356 F.3d at 699–700.

However, this maze of initial assessments leads only to our principal task in this case: considering whether Bennett’s ineffective-assistance claims survive AEDPA’s exacting standard. That is, we must determine whether, when adjudicating Bennett’s ineffective-assistance claims, the state court unreasonably applied Strickland. See Richter, 562 U.S. at 101. Because Bennett cannot succeed in securing habeas relief without meeting this stringent standard, even if she were able to overcome procedural default, we move forward to consider the merits.2 See Lambrix v. Singletary, 520 U.S. 518, 525 (1997) (permitting appellate courts to proceed directly to the merits where judicial economy counsels in favor); Aria v. Hudson, 589 F.3d 315, 316 (6th Cir. 2009) (applying Lambrix).

Ineffective Assistance of Trial Counsel

Bennett alleges three errors on the part of her trial counsel. First, she argues that trial counsel was constitutionally ineffective because his trial strategy was “predicated upon Benson’s unlikely innocence” and did not address Bennett’s questionable intent or culpability. Second, Bennett contends that her counsel erred by failing to impeach one of the prosecution’s key witnesses. Third, Bennett argues that counsel should have investigated and presented evidence regarding the intimate-partner abuse she suffered at Benson’s hands.

To establish ineffective assistance in violation of the Sixth Amendment, Strickland requires a petitioner to show that her counsel’s performance was deficient, as measured against “an objective standard of reasonableness.” 466 U.S. at 688. Additionally, a petitioner must establish “a reasonable probability that, but for counsel’s unprofessional errors, the result of the proceeding would have been different. A reasonable probability is a probability sufficient to

2We note also that Bennett never received an evidentiary hearing, even though evidentiary hearings are usual practice in Michigan courts, under People v. Ginther, 212 N.W.2d 922 (Mich. 1973). Without additional evidence regarding her trial and appellate counsels’ reasoning in fashioning each of their cases, our review of her ineffective-assistance-of-counsel claims de novo for purposes of overcoming procedural default would be significantly hindered. We believe this factor, too, counsels in favor of proceeding immediately to AEDPA review.

No. 17-1574 Bennett v. Brewer Page 9

undermine confidence in the outcome.” Id. at 694. Thus, to prevail on any of her claims, Bennett must show that the error she alleges was objectively unreasonable and that, absent that error, she had a reasonable probability of acquittal.

Bennett’s first allegation of error is best summed up by Judge Shapiro of the Michigan Court of Appeals, writing in dissent in Bennett’s direct appeal:

There was overwhelming evidence that Benson committed this cold-blooded murder. Further, there was no evidence to contradict the testimony of Larvaidan that Bennett provided Benson with the location of the victim’s trailer. Both the practicality of trial tactics and the actual events related to by multiple witnesses mandated the defense that Bennett did not believe that Benson would murder the victim and that she did not want any harm to come to the victim. In terms of trial strategy, the entire defense effort would have to focus on separating Bennett’s actions from Benson’s in the eyes of the jury. Even Benson’s counsel expected that this would be the core of Bennett’s defense, because before the onset of trial, he moved to sever his client’s case from Bennett’s . . . . At the hearing on Benson’s motion, the prosecutor objected to separate trials, but indicated that she would not object to separate juries. Benson’s counsel was satisfied with this approach because it ensured that his client’s jury would not hear Bennett’s statements that Benson shot the victim . . . . Given that Bennett’s jury, unlike Benson’s jury, had heard multiple witnesses testify that Bennett told them that Benson killed McClure, it is difficult to fathom why her counsel tried to argue to the jury that Bennett should be acquitted because Benson did not commit the murder. This was even more peculiar given Larvaidan’s eyewitness testimony clearly demonstrating that Benson committed the murder. Nevertheless, in his closing argument, Bennett’s counsel spent more time arguing that there was reasonable doubt regarding the guilt of Benson than he did arguing that despite Benson’s evident guilt, Bennett should not be convicted of aiding and abetting. This bizarre focus on Benson’s possible innocence was not merely a useless argument; it was wholly counterproductive because, by “defending” Benson, Bennett’s attorney linked her fate to his.

Bennett, 802 N.W.2d at 645-46 (Shapiro, J., dissenting).

Likewise, Judge Shapiro outlined Bennett’s second claim of ineffective assistance, which is closely entwined with the first. In short, Bennett asserts that her counsel performed deficiently in choosing not to impeach Bennett’s friend Breanna Kandler. As Judge Shapiro explained: “Kandler was the only witness who testified at trial that she believed Benson’s threats. However, Kandler admitted at the preliminary examination that in her statement to police, she

No. 17-1574 Bennett v. Brewer Page 10

wrote that she ‘didn’t really think he was gonna do it,’. . . [and that] she thought Benson was ‘just blowing off steam.’” Id. at 646–47. Bennett’s attorney never attempted to raise this prior inconsistent statement or challenge Kandler’s assertion that she found Benson’s threats credible. In Bennett’s view, Kandler was a critical witness for the prosecution and, in light of the dearth of evidence demonstrating that Bennett knew of Benson’s intent, counsel’s decision not to attack her credibility was not reasonable. Bennett argues that this error—alone and especially when considered cumulatively with her counsel’s prioritization of Benson’s defense—prejudiced her.

Finally, Bennett contends that her trial counsel should have used evidence that she suffered from Battered Woman’s Syndrome to shed light on her state of mind and to explain her behavior on the night of McClure’s death. Bennett’s argument here is novel. She concedes that Michigan law does not provide duress as a defense against homicide but argues that Michigan allows the admission of evidence of abuse if it reveals pertinent information about a defendant’s behavior when her actions might otherwise be “incomprehensible to average people.” See People v. Christel, 537 N.W.2d 194, 201–02 (Mich. 1995); see also People v. Kowalski, 821 N.W.2d 14, 26–27 (Mich. 2012) (applying Christel).

Evaluating these claims on collateral review, the state trial court “agree[d] [with Bennett] that trial counsel’s ‘performance fell below objective standards of reasonableness.’” Bennett, No. 07-024733-02-FC, slip op. at *5. Nevertheless, the court found that Bennett failed to meet Strickland’s prejudice prong because, absent counsel’s deficiencies, it remained reasonably probable that Bennett would have faced the same outcome. Id. The court found that sufficient evidence existed to allow the jury to infer that, despite her potentially “clouded . . . judgement,” Bennett understood the seriousness of Benson’s threats, knew he had a gun in his possession, and still “decided to enable, and thus encourage Benson to murder the decedent.” Id.

The court’s conclusion was justified by the record. True, Bennett’s trial counsel failed to present a strong defense on her behalf and, instead, relied on Benson’s weak protestation of innocence. A reasonably competent attorney likely would have understood that challenging Benson’s guilt in light of the substantial evidence against him would prove to be unavailing. Contesting Bennett’s knowledge of Benson’s intent might well have been a wiser strategic decision. Similarly, a more thorough impeachment of Kandler might have put Bennett on a

No. 17-1574 Bennett v. Brewer Page 11

better footing with the jury. As the state trial court found, these errors point to significant deficiencies on the part of Bennett’s trial counsel.

But deficient performance alone does not satisfy Strickland. 466 U.S. at 691–92. In considering whether the trial court properly determined that counsel’s errors did not prejudice Bennett, we “must consider the totality of the evidence before the judge or jury.” Id. at 695; see also Jackson v. Bradshaw, 681 F.3d 753, 760–61 (6th Cir. 2012). In doing so, we are struck by two facts in particular. First, Bennett did not exit the car with Benson when they arrived at McClure’s home. Rather, she stayed in the car and moved it to a more convenient location in order to pick up Benson after he confronted McClure. Second, after Bennett heard shots coming from McClure’s home, she drove Benson away from the scene. Both of these facts, alone and certainly when taken together, undermine Bennett’s assertion that she did not know of Benson’s plan. A jury could reasonably surmise that, if Bennett believed Benson to have innocent intentions, she would have confronted McClure with him, and she would not have arranged their exit strategy. Crucially, the errors Bennett identifies in counsel’s performance have no impact on these facts—they would have been presented to the jury regardless of his impeachment of Kandler or reliance on a more convincing defense strategy. Bennett has not provided evidence or argument suggesting otherwise.

Likewise, we cannot find that counsel’s decision not to elicit evidence of the abuse Bennett experienced prejudiced her. Her contention that it might have explained her state of mind could be true. However, it is just as possible that the jury would have taken Benson’s demonstrated violent tendencies as evidence that Bennett had reason to believe his threats. Moreover, because it is well understood that, under Michigan law, duress is not a viable defense to first degree murder, an attempt by counsel to admit evidence of Battered Woman’s Syndrome could have been entirely rejected. “In assessing prejudice under Strickland, the question is not whether a court can be certain counsel’s performance had no effect on the outcome or whether it is possible a reasonable doubt might have been established if counsel acted differently.” Richter, 562 U.S. at 111 (emphases added). The standard is higher. Strickland requires that “[t]he likelihood of a different result . . . be substantial, not just conceivable.” Id. Although it is

No. 17-1574 Bennett v. Brewer Page 12

possible that presenting evidence of Battered Woman’s Syndrome could have led to Bennett’s acquittal, Bennett has not established that the possibility is a likely one.

As the competing opinions in the Michigan Court of Appeals make clear, and as the district court acknowledged, Bennett’s case is a close one, but “AEDPA sharply limits the circumstances in which a federal court may issue a writ of habeas corpus.” Johnson v. Williams, 568 U.S. 289, 298 (2013). We may do so in contravention of a state court’s denial of relief only if we find that the state court’s ruling was “so lacking in justification that there was an error well understood and comprehended in existing law beyond any possibility for fairminded disagreement.” Richter, 562 U.S. at 103. However sympathetic we might be in light of Bennett’s flawed defense, it is simply not the case that her claims are obvious beyond the point of disagreement. In reaching its conclusion, the state trial court considered the evidence presented at trial and looked to the arguments put forth in the majority and dissenting opinions from the Michigan Court of Appeals. Most importantly, the court appropriately identified and applied the standard laid out by the Supreme Court in Strickland v. Washington. We take heed that the Supreme Court has counseled that such a “run-of-the-mill” state-court decision—even if it diverges from the decision a federal appellate court might have reached—is not “contrary to . . . clearly established Federal law.” Williams v. Taylor, 529 U.S. 362, 406 (2000) (considering application of 28 U.S.C. § 2254(d)(1)). Additionally, our review of the record does not lead us to the conclusion that the court made an “unreasonable determination of the facts in light of the evidence presented” at trial. 28 U.S.C. § 2254(d)(2). Thus, we cannot say that the trial court’s assessment of prejudice was “so lacking in justification” as to warrant reversal. Richter, 562 U.S. at 103.

Ineffective Assistance of Appellate Counsel as a Stand-Alone Claim

Bennett makes one last claim in support of securing a writ: she argues that we should evaluate her ineffective-assistance-of-appellate-counsel claim as a stand-alone claim and grant her a new appeal regardless of our ruling on her trial-counsel claims. Bennett contends that because the state court did not reach the claim explicitly, we need not apply AEDPA deference in our evaluation, and she asks that we instead consider it de novo, as one of the exceptions allowed for in Johnson. In Johnson, the Supreme Court recognized that “[w]hen a state court

No. 17-1574 Bennett v. Brewer Page 13

rejects a federal claim without expressly addressing that claim, a federal habeas court must presume that the federal claim was adjudicated on the merits.” 568 U.S. at 301. But the Court allowed for that presumption to be rebutted, including those instances in which “a federal claim [was] rejected as a result of sheer inadvertence.” Id. at 302-03. In Bennett’s view, the Michigan trial court’s refusal to consider her ineffective-assistance-of-appellate-counsel claim was one such oversight.

True, the Michigan trial court seemed to believe, erroneously, that Bennett’s appellate- counsel claim had been previously adjudicated and was therefore procedurally barred from re- consideration. If the court had rested its decision solely on this procedural ground, Bennett might be able to succeed with her claim. See, e.g., Barton v. Warden, S. Ohio Corr. Facility, 786 F.3d 450, 461 (2015). But the court did not end its review at its procedural determination. Instead, it continued to a substantive review and, as detailed above, unequivocally found Bennett’s trial-counsel claims meritless. Having done so, the court had no reason to continue evaluating her appellate-counsel claim. See Johnson, 568 U.S. at 299 (“[T]here are instances in which a state court may simply regard a claim as too insubstantial to merit discussion.”). The trial court found that Bennett could not overcome step one of the merits-adjudication process. We can therefore presume, in accordance with Johnson, that it simply chose not to engage in the remainder of the analysis required for the adjudication of Bennett’s appellate-counsel claims because, as we have noted, counsel cannot be found incompetent for failing to raise meritless claims. See Mapes, 171 F.3d at 408. Bennett has not successfully rebutted that presumption and her claim fails.

The trial court’s assessment of Bennett’s underlying trial-counsel claims was reasonable. Therefore, the court could not have been objectively unreasonable in rejecting her appellate- counsel claim.

CONCLUSION

For the reasons set out above, we AFFIRM the judgment of the district court denying the petitioner a writ of habeas corpus.