De Novo Federal Review When Florida Applies Carratelli’s “Actual Bias” Prejudice Standard to Appellate-IAC Claims

Introduction

Case: Gary Michael Hilton v. Secretary, Florida Department of Corrections
Court: United States Court of Appeals for the Eleventh Circuit (unpublished)
Date: April 1, 2026

Gary Michael Hilton, a Florida death-row prisoner convicted of first-degree murder, kidnapping, and grand theft, sought federal habeas relief under 28 U.S.C. § 2254 after the Florida Supreme Court rejected multiple ineffective-assistance theories. The Eleventh Circuit granted a certificate of appealability on four issues: (1) alleged “disarray” of the defense team; (2) an allegedly split guilt/penalty strategy; (3) appellate counsel’s failure to appeal the denial of a change of venue; and (4) appellate counsel’s failure to appeal denials of for-cause challenges during jury selection.

The central legal significance of the decision is not a new Strickland rule on the merits, but the court’s application of a crucial AEDPA framework: when the Florida Supreme Court applies Carratelli v. State’s heightened “actual bias” prejudice test to appellate-ineffectiveness claims, that is an unreasonable application of clearly established federal law, requiring de novo federal review—while still demanding a showing that the omitted appellate issue had a reasonable probability of success.

Summary of the Opinion

  • Trial-counsel “disarray” claim: The Eleventh Circuit held the Florida Supreme Court reasonably denied relief for lack of Strickland prejudice, because Hilton’s state postconviction pleading contained largely generalized complaints and did not show a reasonable probability of a different outcome.
  • Split guilt/penalty strategy claim: The court again affirmed on the prejudice prong, emphasizing overwhelming evidence and lack of a plausible probability that additional mitigation or a different approach would have changed either phase’s result.
  • Change-of-venue appellate-IAC claim: The Florida Supreme Court used Carratelli’s “actual bias” framing; under Guardado v. Secretary, Florida Department of Corrections, that was an unreasonable application of Strickland, so the Eleventh Circuit reviewed de novo. Even so, Hilton failed to show prejudice because a venue claim would likely not have succeeded on direct appeal given the voir dire record and the high deference owed to trial-court findings of juror impartiality.
  • For-cause challenge appellate-IAC claim: For similar reasons, the court reviewed de novo (again due to the state court’s use of Carratelli) but found no prejudice because the prospective appellate issue was unlikely to succeed: jurors need not be wholly ignorant, and the record supported impartiality after voir dire rehabilitation.

Disposition: The Eleventh Circuit affirmed the denial of Hilton’s federal habeas petition and upheld his conviction and death sentence.

Analysis

Precedents Cited

1) AEDPA Deference and “Clearly Established” Law

  • Pye v. Warden, Ga. Diagnostic Prison and Davis v. Ayala: Used to frame AEDPA as “highly deferential” when the state court adjudicated the claim on the merits, limiting federal relief to unreasonable applications of Supreme Court holdings or unreasonable fact determinations.
  • Woods v. Donald and White v. Woodall: Reinforced that “clearly established Federal law” under § 2254(d)(1) consists only of Supreme Court holdings, not dicta—important because petitioners often rely on broader constitutional language that is not clearly established as a holding.
  • Williams v. Taylor and Harrington v. Richter: Provided the core “unreasonable application” framework. The opinion relies on Richter’s “beyond any possibility for fairminded disagreement” standard to reject Hilton’s trial-counsel claims under AEDPA deference.
  • Calhoun v. Warden, Baldwin State Prison and Adkins v. Warden: Cited for the proposition that when AEDPA deference does not apply, federal courts review de novo.

2) The Governing Ineffective-Assistance Framework

  • Strickland v. Washington: The controlling standard for all four claims—requiring both deficient performance and prejudice. The court repeatedly resolves claims on prejudice alone, consistent with Strickland’s instruction that courts need not address both prongs if one fails.
  • Thornell v. Jones: Cited to clarify the prejudice inquiry as “reasonable probability,” not “more likely than not,” and to emphasize that the difference between these standards matters only rarely.

3) Procedural Default and Expansion of Trial-IAC Theories

  • Martinez v. Ryan: Addressed Hilton’s attempt to add new trial-counsel deficiency allegations in federal habeas. The panel explains that Martinez can excuse default only for a “substantial” underlying trial-IAC claim; generalized complaints (e.g., “not doing a good job”) without a meritorious, outcome-relevant showing do not satisfy the standard.

4) Florida’s Carratelli Standard and Federal Correction via Guardado

  • Carratelli v. State: The Florida Supreme Court’s postconviction standard requiring a showing akin to “actual bias” for prejudice in certain appellate-ineffectiveness contexts. The Eleventh Circuit treats this as materially more demanding than Strickland’s “reasonable probability.”
  • Guardado v. Secretary, Florida Department of Corrections: The critical Eleventh Circuit precedent holding that Carratelli is an unreasonable application of clearly established federal law to the extent it heightens prejudice beyond Strickland. Relying on Guardado, the panel applies de novo review to Hilton’s two appellate-IAC claims.

5) Appellate Counsel Ineffectiveness: “Reasonable Probability of Success on Appeal”

  • Philmore v. McNeil (quoting Heath v. Jones): Supplies the prejudice test for omitted appellate issues—whether the neglected claim had a “reasonable probability of success on appeal.”
  • Smith v. Robbins and Jones v. Barnes: Support the principle that appellate lawyers need not raise every nonfrivolous or “colorable” claim; effective advocacy includes winnowing weaker issues.

6) Venue, Pretrial Publicity, and Juror Impartiality

  • Patton v. Yount: Sets the key inquiry—whether jurors had such fixed opinions that they could not judge guilt impartially, not whether they remembered the case.
  • Skilling v. United States (citing Mu'Min v. Virginia): Emphasizes “pinnacle” deference to trial courts on juror impartiality and the “manifest error” threshold for reversal—central to the panel’s conclusion that a venue argument was unlikely to succeed on direct appeal.
  • Murphy v. Florida and Irvin v. Dowd: Reinforce that jurors need not be totally ignorant; the question is whether they can set aside impressions and decide based on evidence.

7) Penalty-Phase Preparation (Raised in Concurrence)

  • Rowland v. Chappell: Cited by Judge Abudu for the proposition that failure to timely prepare for the penalty phase may constitute deficient performance in capital litigation—used to underscore the seriousness of the defense team’s dysfunction even where no prejudice is shown.
  • Kuren v. Luzerne Cnty. and Betschart v. Oregon: Invoked to contextualize structural underfunding and systemic impediments to effective counsel—supporting the concurrence’s concern that focusing solely on prejudice can mask unacceptable lawyering.
  • Davis v. Comm'r, Ala. Dep't of Corr.: Cited by the concurring judge for a capital-case-specific view of Sixth Amendment concerns.

Legal Reasoning

1) Two Lanes of Review: AEDPA Deference vs. De Novo

The opinion demonstrates a disciplined separation between (a) claims governed by AEDPA deference because the state court adjudicated them on the merits under a proper Strickland framework, and (b) claims where the state court applied a prejudice standard inconsistent with clearly established Supreme Court law, thereby forfeiting AEDPA deference on that point and triggering de novo review.

For Hilton’s trial-counsel “disarray” and “split strategy” claims, the Eleventh Circuit treated the Florida Supreme Court’s prejudice determinations as reasonable applications of Strickland, thus ending the inquiry under § 2254(d) absent an extreme, clearly unjustifiable error under Richter.

2) The Guardado Correction: Rejecting Carratelli’s Heightened Prejudice Test

On the appellate-IAC claims, the panel explicitly identifies the Florida Supreme Court’s reliance on language derived from Carratelli v. State, describing it as demanding “actual bias.” Under Guardado v. Secretary, Florida Department of Corrections, that heightened test is incompatible with Strickland because it effectively requires more than a “reasonable probability” that the result would have differed.

The remedy is methodological: the federal court does not grant relief automatically; it instead removes AEDPA deference and performs its own de novo Strickland analysis using the appropriate prejudice standard.

3) Why Hilton Still Loses Under De Novo Review

The de novo prejudice inquiry for omitted appellate issues asks whether the unraised claim had a reasonable probability of success on direct appeal. Applying that standard, the panel reasons:

  • Change of venue: Given Skilling v. United States’s “manifest error” bar and extensive voir dire rehabilitation reflected in the record, a direct-appeal challenge to venue was unlikely to succeed.
  • For-cause challenges: Under Murphy v. Florida and Irvin v. Dowd, juror familiarity with a case does not equal bias. With the record showing seated jurors had not made up their minds and the trial court’s impartiality determination not meaningfully undermined, the omitted appellate issue again lacked a reasonable probability of success.

4) The Concurrence’s Institutional Warning (Without Changing the Result)

Judge Abudu’s concurrence agrees Hilton cannot show prejudice, but argues courts should sometimes still identify deficient performance—especially in capital cases—because consistently skipping the performance prong may reduce accountability and obscure substandard defense practices. The concurrence relies on Strickland’s permissive sequencing (courts may resolve on prejudice alone) to argue courts also may reach and condemn deficient lawyering when the record warrants it.

Impact

  • Operational impact in Florida habeas litigation: The decision reinforces that when Florida courts apply Carratelli’s “actual bias” prejudice requirement to appellate-ineffectiveness claims, federal courts in the Eleventh Circuit will treat that as an unreasonable application of clearly established federal law and conduct de novo review (per Guardado).
  • No automatic relief: The opinion illustrates the practical limitation of that doctrinal win for petitioners: de novo review still requires a persuasive showing that the omitted appellate claim was likely to succeed. In jury-bias/venue contexts, the “manifest error” deference to trial judges recognized in Skilling makes prejudice difficult to establish.
  • Appellate strategy validation: By invoking Jones v. Barnes and Smith v. Robbins, the decision strengthens the principle that appellate counsel’s winnowing is constitutionally permissible—even in capital cases—so long as omitted issues were unlikely to succeed.
  • Normative influence (concurrence): While not binding, the concurrence may be cited to urge more explicit findings on deficient performance in capital ineffectiveness litigation, potentially shaping how future panels write opinions even when denying relief.

Complex Concepts Simplified

  • AEDPA deference (28 U.S.C. § 2254(d)): Federal habeas courts generally cannot overturn a state court’s merits decision unless it was not just wrong, but unreasonably wrong under Supreme Court holdings, or based on an unreasonable view of the facts.
  • “Clearly established Federal law”: Only Supreme Court holdings count; lower-court cases and Supreme Court dicta cannot create “clearly established” rules for AEDPA purposes.
  • Strickland’s two prongs: A defendant must show (1) deficient attorney performance and (2) prejudice—i.e., a reasonable probability the outcome would have been different.
  • De novo review: The federal court decides the issue fresh, without deferring to the state court’s legal conclusion—here, triggered because the state court used the wrong prejudice standard (Carratelli as rejected by Guardado).
  • Change of venue due to publicity: The question is not whether jurors heard about the case, but whether they could still be fair and impartial after questioning.
  • For-cause vs. peremptory challenges: A for-cause challenge argues a juror must be removed due to bias or disqualification; a peremptory challenge allows removal without stating a reason (though limited in number). A key practical concern is being forced to use peremptories to cure for-cause denials.
  • Juror “rehabilitation”: Follow-up questioning by the judge or lawyers to determine whether a juror who expressed concern or prior knowledge can set it aside and decide based solely on evidence.

Conclusion

The Eleventh Circuit’s decision underscores a targeted but important rule for federal habeas review in Florida cases: when the Florida Supreme Court evaluates appellate-ineffectiveness claims under Carratelli v. State’s heightened “actual bias” prejudice concept, that approach conflicts with Strickland v. Washington and triggers de novo federal review under Guardado v. Secretary, Florida Department of Corrections.

Yet the case also illustrates the high practical barriers to relief even under the correct standard—especially where the omitted appellate issues concern venue and juror impartiality, areas where trial-court determinations receive exceptional deference under Skilling v. United States.

Finally, Judge Abudu’s concurrence highlights a broader institutional concern: even when prejudice is absent, courts should sometimes explicitly identify deficient lawyering—particularly in capital cases—to reinforce professional norms and the Sixth Amendment’s promise of genuinely effective representation.