Alvarez v. Guerrero: Limiting “Sleeping Lawyer” Habeas Relief Under AEDPA When Co‑Counsel Is Active

I. Introduction

In Alvarez v. Guerrero, No. 18‑70001 (5th Cir. Dec. 23, 2025), the Fifth Circuit affirmed the denial of federal habeas relief to Juan Carlos Alvarez, a Texas death row inmate convicted of multiple gang‑related murders in 1999. The case presents three important clusters of issues:

a high‑stakes Sixth Amendment claim that lead trial counsel repeatedly slept during Alvarez’s capital trial;
a mitigation‑investigation claim under Strickland, Wiggins, and Andrus alleging inadequate development of Alvarez’s background and mental health at sentencing; and
due process and Brady‑type challenges to DNA evidence originating in the troubled Houston/Harris County crime lab.

The majority opinion (Judge Edith Jones, joined by Chief Judge Elrod) tightly construes federal habeas review under the Antiterrorism and Effective Death Penalty Act of 1996 (“AEDPA”), refuses to treat the “sleeping lawyer” allegation as an automatic or presumed‑prejudice error, and rejects all three certified claims. Judge Richman dissents in a lengthy opinion, arguing that the sleeping‑lawyer claim alone warrants habeas relief or, at minimum, a remand for further state court consideration.

This commentary explains the opinion’s holdings, situates them against prior Supreme Court and Fifth Circuit precedent, and analyzes their likely impact on future habeas litigation—especially “sleeping lawyer” and mitigation‑investigation claims in capital cases.

II. Summary of the Fifth Circuit’s Decision

The Fifth Circuit’s core rulings can be summarized as follows:

Sleeping‑lawyer / deprivation‑of‑counsel claim
- The court treats the state courts’ silent handling of the sleeping‑lawyer allegation as an adjudication on the merits under Johnson v. Williams, triggering AEDPA deference.
- There is no “clearly established” Supreme Court precedent holding that a Sixth Amendment violation occurs when one of two trial attorneys is allegedly sleeping or ineffective while the other is actively representing the defendant.
- Accordingly, under AEDPA, the Texas Court of Criminal Appeals (TCCA) did not “unreasonably apply” Strickland or Cronic by denying relief, nor did it make an unreasonable factual determination in discounting two late juror affidavits alleging repeated sleeping.
- The court refuses to stay the case or remand for consideration of a 2019 affidavit from lead counsel Reyes (admitting he may have slept), both because of AEDPA’s evidentiary limits (especially Shinn v. Ramirez) and because the new affidavit could have been obtained long earlier.
Mitigation investigation / Wiggins–Andrus claim
- Even assuming de novo review (ignoring AEDPA deference), the court holds that trial counsel’s mitigation investigation and presentation at sentencing were constitutionally adequate.
- Counsel consulted a mental‑health expert, met repeatedly with family, reviewed the State’s files, planned strategy with Alvarez and his family, and presented a coherent mitigating narrative at sentencing.
- The additional mitigation Alvarez now offers is largely cumulative or speculative; he has not shown either deficient performance or a “reasonable probability” that the result would have been different.
- Andrus v. Texas is distinguished as involving an almost total failure to investigate and present obvious, powerful mitigating evidence; Alvarez’s case is characterized as nothing of the sort.
Houston Crime Lab / DNA‑taint claim
- The TCCA dismissed Alvarez’s later state habeas applications raising crime‑lab and DNA issues as an “abuse of the writ,” which the Fifth Circuit treats as an independent and adequate state procedural bar.
- The DNA‑taint claim is therefore procedurally defaulted in federal court, and Alvarez has not shown cause and prejudice or actual innocence to overcome the default.
- In any event, on the merits, re‑testing by independent experts and by court order confirmed that the blood on Alvarez’s Mossberg shotgun barrel matched victim José Varela. Given other overwhelming evidence of guilt, any flaws in the original Houston Lab work were not material under Brady/Banks v. Dretke.

In short, the Fifth Circuit affirms the district court’s denial of habeas relief and denies all attendant motions (funding, Rhines stay, record expansion).

III. Procedural and Doctrinal Framework

A. AEDPA’s Deferential Standard

The decision is structured around AEDPA, 28 U.S.C. § 2254(d), which sharply limits federal habeas relief from state convictions. Where a claim has been “adjudicated on the merits” in state court, a federal court may not grant relief unless the state decision:

§ 2254(d)(1): is “contrary to” or involves an “unreasonable application” of clearly established federal law as determined by the U.S. Supreme Court; or
§ 2254(d)(2): is based on an unreasonable determination of the facts in light of the state‑court record.

The majority repeatedly emphasizes the high bar: it is not enough that the state court was wrong, or even clearly wrong. Relief is barred unless the error is “beyond any possibility for fairminded disagreement” (citing Harrington v. Richter, Shoop v. Hill, Shinn v. Kayer, and the Fifth Circuit’s own en banc decision in Langley v. Prince).

B. Silent or Partial State Decisions and Johnson v. Williams

A key threshold issue was whether the Texas courts actually adjudicated the sleeping‑lawyer claim “on the merits.” Neither the state habeas trial court nor the TCCA explicitly discussed that claim, although the juror affidavits describing sleeping were in the record.

Relying on Johnson v. Williams (568 U.S. 289 (2013)) and Harrington v. Richter, the majority applies the presumption that when a state court issues a reasoned decision rejecting a habeas application, it is deemed to have adjudicated all federal claims on the merits—even those not expressly discussed—unless that presumption is rebutted. The Fifth Circuit had recently reinforced this presumption in Jimenez v. Guerrero, 133 F.4th 483 (5th Cir. 2025).

Because Alvarez did not carry the “heavy burden” of rebutting this presumption, the court holds that AEDPA deference applies to the sleeping‑lawyer claim.

IV. The Sleeping‑Lawyer Claim

A. Factual Basis of the Claim

The sleeping‑lawyer allegation rests primarily on:

Two juror affidavits, executed nearly ten years after trial, asserting that lead counsel, Frumencio Reyes, repeatedly fell asleep—sometimes for several minutes at a time—during the State’s examination of witnesses, including at least one occasion where the judge had to call his name to wake him, prompting nervous laughter in the courtroom.
Reyes’s own later affidavit (2019), stating that he now believes he likely did fall asleep at times, and admitting he was previously untruthful when he denied that possibility to earlier habeas counsel. He also notes serious health problems and similar episodes in other matters. (The panel, however, refuses to consider this affidavit, discussed below.)
The fact that second‑chair counsel, John Denninger, was present and active throughout trial, conducting a substantial portion of voir dire and examinations, and delivering one of the closings.

The dissent stresses the trial court’s practice that only one lawyer per side may examine a given witness or make objections during that testimony. Thus, when Reyes was asleep during the direct of a witness he was assigned to cross‑examine, Alvarez effectively had no functioning lawyer for that portion. The majority acknowledges the practice but characterizes the record as too thin and late‑developed to show a constitutional deprivation, given Denninger’s overall active role and the silence of contemporaneous records and earlier affidavits.

B. Strickland Framework and AEDPA Overlay

Under Strickland v. Washington, 466 U.S. 668 (1984), a Sixth Amendment ineffective‑assistance claim requires:

Deficient performance: counsel’s representation fell below an objective standard of reasonableness under prevailing professional norms; and
Prejudice: a reasonable probability that, but for counsel’s errors, the outcome would have been different.

Because the claim was adjudicated on the merits in state court, Alvarez could succeed in federal habeas only by showing that the TCCA’s application of Strickland was not just incorrect but unreasonable under § 2254(d)(1).

1. Deficient performance

The majority acknowledges that actual, significant sleeping by a capital defendant’s lead lawyer during critical testimony would ordinarily be highly problematic under Strickland. But it stresses:

The only evidence of sleeping in the state record are two “vague” juror affidavits, executed almost a decade post‑trial, not tested in an evidentiary hearing.
Reyes submitted three affidavits in state post‑conviction proceedings about his representation; none mentioned any physical incapacity or sleeping.
There is no indication in the trial transcript that the judge or anyone else observed and addressed sleeping (for example, no admonitions, no bench conferences about counsel’s physical state).
Second‑chair counsel Denninger’s role was substantial: he handled many voir dires, questioned multiple witnesses, argued to the jury, and made numerous objections.

Against this backdrop, the majority holds that the TCCA could reasonably conclude that Alvarez had not carried his burden to show that the defense as a whole—considering both lawyers—was constitutionally ineffective. In the court’s view, attacking only one lawyer’s conduct, while the other is actively engaged, does not establish that the defense team’s performance as a whole fell below constitutional minimums.

2. Prejudice

On prejudice, Alvarez relied almost entirely on the jurors’ lay assessments that Reyes “did not put up a real defense” and that his sleeping contributed to a sense that counsel was uninvested. The majority finds this insufficient, noting:

No specific missed objections, missed lines of cross‑examination, or alternative strategies were identified.
The evidence of guilt was “overwhelming”: multiple eyewitnesses, gang insiders testifying against Alvarez, ballistic matches to his weapons, and his own incriminating statements to police.
Given such evidence, the state court could reasonably find no “reasonable probability” of a different result even if portions of testimony had been more aggressively challenged.

The majority quotes cases emphasizing that, historically, courts have denied Strickland claims where petitioners could not articulate what questions should have been asked or how cross‑examination would have changed the outcome.

C. Cronic vs. Strickland: When Is Prejudice Presumed?

Alvarez also argued that Reyes’s sleeping amounted to a constructive denial of counsel under United States v. Cronic, 466 U.S. 648 (1984), and thus should trigger a presumption of prejudice without any need to prove outcome‑determinative harm.

Cronic, as refined by Bell v. Cone, identifies three narrow situations where prejudice is presumed:

Complete denial of counsel at a critical stage of the proceeding (e.g., no lawyer present during arraignment where rights are lost).
Total failure to subject the prosecution’s case to meaningful adversarial testing.
Situations where counsel is forced to operate under circumstances making effective assistance virtually impossible (e.g., capital trial with unprepared appointed counsel days before trial in highly complex case, though even this has been narrowed).

The majority’s key moves are:

It relies heavily on Wright v. Van Patten, 552 U.S. 120 (2008), where counsel participated in a plea hearing by speakerphone. The Supreme Court held there was no clearly established law that such partial “absence” triggered Cronic, and thus AEDPA barred relief.
The court notes that the Supreme Court has never applied Cronic in a setting where the defendant was represented by two lawyers and only one was allegedly “absent” or ineffective.
Under AEDPA, the Fifth Circuit may not extrapolate from general Cronic language to create a new, specific rule: that the “absence” of one co‑counsel, while another is present and functioning, constitutes a per se Sixth Amendment violation.

The majority acknowledges that the Fifth Circuit (en banc) has previously applied Cronic to a “sleeping lawyer” scenario in Burdine v. Johnson, 262 F.3d 336 (5th Cir. 2001), as have several other circuits (e.g., Javor, Tippins, Ragin). But it distinguishes those cases on a crucial fact: in each, the defendant had only one trial lawyer, who slept through substantial portions of the proceedings. By contrast, Alvarez had two lawyers, one of whom was active, and Burdine and its siblings are not Supreme Court decisions and thus cannot supply “clearly established federal law” for AEDPA purposes.

The majority also flags (in a footnote) that to treat “partial denial of the defense team” as equivalent to “denial of counsel” would likely run afoul of Teague v. Lane, 489 U.S. 288 (1989), by announcing a new rule and applying it retroactively on habeas—something AEDPA does not permit.

D. Unreasonable Factual Determination? (§ 2254(d)(2))

Under § 2254(d)(2), Alvarez had to show that the state court’s implicit factual determination (that no constitutionally significant sleeping occurred, or that the defense overall was adequate) was unreasonable in light of the state record, and to rebut the presumption of correctness with “clear and convincing” evidence (28 U.S.C. § 2254(e)(1)).

The majority finds:

The two juror affidavits are late, untested, and somewhat generalized; they emphasize not just sleeping but the defense’s perceived overall weakness.
No other juror, court staff, or professionals supplied corroborating affidavits in state court. (Co‑counsel Denninger never submitted an affidavit before his death in 2011.)
Trial transcripts contain no hint that the judge or any party noted sleeping; indeed, a pretrial colloquy includes the judge praising Reyes as a lawyer no one “takes advantage of.”
Direct appeal counsel raised 39 issues; initial state habeas counsel—experienced capital attorneys—did not assert a sleeping‑lawyer claim, despite having the juror affidavits.

Given these gaps, the majority holds that the TCCA’s implicit rejection of the juror affidavits’ more dramatic inferences was not an unreasonable factual determination.

E. Andrew v. White and the Dissent’s Broader “Clearly Established Law” Theory

The dissent, invoking Andrew v. White, 604 U.S. 86 (2025), argues that general Supreme Court principles—such as that a defendant’s deprivation of counsel at a critical stage violates the Sixth Amendment (see Holloway, Geders, Powell, Glasser)—are sufficiently specific to be “clearly established law” for AEDPA, even if no case presents identical facts. The dissent analogizes Andrew’s use of a general due‑process principle (unduly prejudicial evidence can render trial fundamentally unfair) to Alvarez’s claim: general Sixth Amendment principles should dictate that sleeping through critical testimony is a denial of counsel.

The majority rejects that extension, asserting that:

Andrew involved a general principle already applied in closely analogous contexts (unduly prejudicial evidence), whereas there is no Supreme Court decision even roughly on point regarding one of two trial lawyers sleeping.
Merely “incanting” broad language about absence of counsel at critical stages is insufficient under AEDPA; the law must be clearly established in a way “directly related to the evidentiary issue presented.”

In short, the majority reads Andrew narrowly and refuses to treat it as license to generalize Sixth Amendment doctrine beyond existing Supreme Court holdings.

F. The 2019 Reyes Affidavit and Rhines Stay

Alvarez moved to supplement the record in the Fifth Circuit with Reyes’s 2019 affidavit and to obtain a stay and abeyance under Rhines v. Weber, 544 U.S. 269 (2005), to return to state court.

The majority denies this on two main grounds:

AEDPA’s evidentiary limits (Cullen v. Pinholster, Shinn v. Ramirez)
- Under Pinholster, federal courts reviewing claims adjudicated on the merits must confine themselves to the state‑court record for purposes of § 2254(d).
- Shinn v. Ramirez, 596 U.S. 366 (2022), further holds that federal habeas courts generally may not consider new evidence to overcome procedural defaults where the prisoner is at fault, absent narrow statutory exceptions under § 2254(e)(2).
- Alvarez does not argue that any statutory exception applies; thus, the panel cannot consider Reyes’s new affidavit on the merits.
No good cause / undue delay under Rhines
- Rhines allows a stay only where (i) the petitioner shows good cause for not exhausting earlier, (ii) the unexhausted claims are potentially meritorious, and (iii) the request is not abusive or dilatory.
- The majority finds no good cause: nothing prevented Reyes from acknowledging his lapses in earlier affidavits, and Alvarez had years—and multiple habeas rounds—during which he could have sought such a statement.
- Moreover, because the court concludes the underlying sleeping‑lawyer claim is not meritorious under AEDPA due to active co‑counsel, the new affidavit would not make the claim “potentially meritorious.”

The dissent would go the other way: because Reyes now admits prior dishonesty and does not dispute the juror accounts, the new affidavit significantly strengthens the claim and, under pre‑AEDPA exhaustion case law (Joyner, Graham) as carried forward in Anderson and Ward, warrants remitting the claim to state court. The dissent sees good cause in Reyes’s prior refusal to be candid and emphasizes that Alvarez filed the new affidavit within days of obtaining it.

G. Practical Impact on “Sleeping Lawyer” Claims

Alvarez v. Guerrero cements a significant limitation in the Fifth Circuit:

Where a capital defendant was represented by two trial lawyers, and only one is alleged to have slept or been otherwise intermittently absent, there is no clearly established Supreme Court authority that such partial absence automatically triggers Cronic’s presumption of prejudice.
As a result, such claims will almost invariably be analyzed under Strickland + AEDPA, requiring proof of concrete prejudice and deference to state‑court factfindings.
Late‑developed affidavits (from jurors or counsel) are unlikely to overcome § 2254(d) and (e), particularly where trial transcripts and earlier affidavits are silent and co‑counsel was active.

For capital and non‑capital defendants in the Fifth Circuit, this opinion sharply narrows the circumstances under which a “sleeping lawyer” claim can yield federal habeas relief, especially when there was co‑counsel.

V. The Wiggins / Andrus Mitigation‑Investigation Claim

A. Legal Framework: Strickland, Wiggins, and Andrus

In the capital sentencing context, Strickland’s two‑prong test is applied with specific focus on the duty to investigate mitigating evidence. In Wiggins v. Smith, 539 U.S. 510 (2003), the Court held that counsel can be constitutionally deficient for failing to undertake a reasonable mitigation investigation, especially where red flags (e.g., severe childhood abuse) are evident. Andrus v. Texas, 590 U.S. 806 (2020), reaffirmed and extended Wiggins, condemning a capital defense in which counsel presented minimal, unprepared mitigation testimony and failed to uncover a “tidal wave” of readily available evidence of extreme childhood deprivation, trauma, and mental illness.

Crucially, even where performance is deficient, a petitioner must also show prejudice, i.e., a substantial likelihood that, had the mitigation case been properly developed and presented, at least one juror would have opted for a life sentence rather than death.

B. Alvarez’s Allegations

Alvarez claimed his trial team failed to:

investigate and present evidence that he suffered from organic brain damage and PTSD;
fully develop his history of poverty, trauma, violence, and the murder of loved ones (his father and a close friend);
explain his gang involvement in a more contextualized, sympathetic way; and
construct a cohesive, expert‑supported narrative tying these factors together as mitigating circumstances.

C. The Actual Mitigation Case Presented at Trial

The majority relies heavily on what trial counsel did do. At sentencing, the defense presented six witnesses, including:

Four family members describing:
- the family’s life in Mexico and the trauma of the father’s murder;
- subsequent moves to Monterrey and Houston;
- Alvarez’s work ethic, support for family, lack of drug use, generally quiet demeanor; and
- the emotional impact of a close friend’s killing shortly before the charged murders.
A corrections officer stating Alvarez was quiet and not problematic in custody.
A psychologist (Dr. Laval) who:
- interviewed Alvarez multiple times;
- administered psychological testing;
- found depression and poor anger management but no major mental illness;
- offered the opinion that the murders were a reaction to the friend’s killing; and
- concluded Alvarez would not be a future danger if incarcerated.

Trial counsel also averred in state habeas affidavits that:

they made “many trips” to the DA’s office to review open files;
had “numerous meetings” with family to understand Alvarez’s mental and emotional state;
worked with Alvarez and his family to plan sentencing strategy; and
relied heavily on Dr. Laval’s expertise, whose evaluations at the time did not suggest brain damage or PTSD.

D. Majority’s Analysis: No Deficient Performance, No Prejudice

The majority’s core points are:

Mental‑health investigation was reasonable: Counsel retained and used a qualified psychologist who conducted interviews and testing. Nothing in those evaluations suggested psychosis, major mental illness, or neurocognitive impairment. Only years later did a different psychologist speculate about possible brain damage/PTSD. Counsel were entitled to rely on the contemporaneous expert assessment.
Mitigation themes were developed and presented: The sentencing presentation did, in fact, tell a story about:
- childhood instability and trauma;
- the father’s violent death and its lasting impact;
- the friend’s killing as emotional trigger;
- gang membership as a maladaptive response to bereavement and identity needs; and
- Alvarez’s manageable risk in a prison environment.
New evidence is largely cumulative: Much of what Alvarez now adduces overlaps with, or modestly elaborates, themes already presented. There is no “tidal wave” of qualitatively different mitigation comparable to Andrus.
No reasonable probability of a different sentence: The State’s punishment evidence showed an escalating pattern of gratuitous gang violence, including prior aggravated assaults, drive‑by shootings, and disregard for bystanders. Even assuming some additional mental‑health or background detail could have been presented, the majority sees no substantial likelihood that at least one juror would have spared Alvarez death in light of this record.

The court underscores that Strickland and Wiggins require deference to reasonable strategic judgments and caution against “Monday morning quarterbacking” merely because more could always have been done. It labels Alvarez’s criticisms as the type of “just not doing enough” argument the Fifth Circuit has historically treated warily (Carty v. Thaler).

E. Distinguishing Andrus v. Texas

Andrus is a recent and important Supreme Court capital case, so the panel takes care to distinguish it:

In Andrus, trial counsel presented almost no mitigation—essentially only brief testimony from the parents—during a capital sentencing, despite abundant, easily discoverable evidence of horrific childhood abuse, severe mental illness, and systemic neglect.
The Supreme Court described Andrus’s post‑conviction case as a “tidal wave” of powerful, qualitatively new mitigation that was never investigated or presented, and the state habeas trial court had already found counsel ineffective.
By contrast, Alvarez’s counsel:
- investigated, consulted a mental‑health expert, met with family, and presented a multi‑witness mitigation case;
- knew and leveraged the core trauma narratives (father’s murder; friend’s killing; migration turmoil); and
- did not ignore clear signs of psychosis or serious mental illness (because, on this record, such signs were not evident at the time).

Accordingly, the court concludes that Alvarez’s case is far closer to routine, contested Strickland/Wiggins cases than to the stark breakdown in representation condemned in Andrus.

F. Impact on Mitigation Practice and Future Wiggins Claims

The opinion reinforces several practical lessons for capital defense practitioners and habeas litigants in the Fifth Circuit:

Retaining and substantively using a qualified mental‑health expert—and documenting that process—strongly insulates counsel from later claims that they ignored mental‑health “red flags.”
Counsel who meet extensively with family, review State files, and present at least a moderately detailed, coherent mitigation narrative will be difficult to second‑guess under Strickland, especially on habeas.
To prevail on a Wiggins claim post‑Andrus, a petitioner likely must show something close to:
- a near‑total failure to investigate obvious, readily available mitigation; and
- a body of new evidence that qualitatively changes the jury’s understanding of the defendant (not merely cumulative elaboration).

VI. The Houston Crime Lab / DNA‑Taint Claims

A. Background

In the early 2000s, serious problems at the Houston Police Department and Harris County crime laboratories—particularly in DNA and serology—led to a wave of challenges to convictions reliant on those labs. Alvarez claimed:

the blood‑stain DNA evidence linking his Mossberg shotgun to victim José Varela was “tainted” or “sloppy”; and
the State committed Brady violations by not disclosing lab misconduct or problems with the testing.

B. Procedural Default: Abuse‑of‑the‑Writ Dismissals

Alvarez pursued these claims in a second and later in a fourth state habeas application. The TCCA dismissed both as an abuse of the writ under Texas Code of Criminal Procedure art. 11.071 § 5, which bars successive state habeas petitions absent narrow exceptions.

The Fifth Circuit treats this as an independent and adequate state procedural ground. Under longstanding doctrine, such a state procedural ruling generally precludes federal habeas review unless the petitioner can:

show “cause” and “prejudice” to excuse the default (e.g., ineffective post‑conviction counsel, external impediments); or
demonstrate a fundamental miscarriage of justice, typically by establishing actual innocence.

The panel agrees with the district court that Alvarez has not made such a showing. His DNA‑taint claims are therefore procedurally defaulted and barred from federal review.

C. Alternative Merits Holding

The court adds an alternative merits analysis:

Subsequent DNA re‑testing—both by experts retained by Alvarez’s own counsel and by court order—confirmed that the blood on the shotgun barrel matched victim Varela.
The record strongly indicates that any deficiencies in the original Houston Lab work did not change the central forensic fact: the gun seized from Alvarez’s apartment bore the victim’s blood.
Even if there were some residual question, the DNA was not material given the overwhelming independent evidence connecting Alvarez to the Woodfair murder:
- witness testimony that she saw Alvarez loading the shotgun and rifle, leaving with them, and returning with blood on his shirt;
- the victim’s sister’s eyewitness identification of Alvarez as the shooter, hearing him shout “Southwest Cholos” as he fired; and
- Alvarez’s own admissions about driving to the scene with his wife, armed, knowing a killing was planned.

Under Banks v. Dretke, 540 U.S. 668 (2004), and Brady, suppressed evidence is “material” only if there is a reasonable probability of a different result had it been disclosed. The panel sees no plausible path to a different outcome had the contested DNA been excluded.

D. Implications for Crime‑Lab‑Based Habeas Claims

Alvarez illustrates the difficulty of obtaining habeas relief based on generalized crime‑lab scandals:

Successive state habeas restrictions (like Texas’s art. 11.071 § 5) will often render such claims procedurally barred unless raised in initial state petitions or tied to new statutory exceptions.
Even when lab misconduct is shown, materiality remains a high hurdle where independent evidence of guilt is strong and subsequent retesting confirms key forensic links.

VII. Complex Concepts Simplified

A. AEDPA & “Clearly Established Federal Law”

Under AEDPA (§ 2254(d)(1)), federal habeas relief is limited to instances where the state court’s decision contradicts or unreasonably applies Supreme Court holdings—not lower federal decisions, and not general principles drawn from dicta. The law must be “clearly established” at the time of the state conviction becoming final.

This is why many of Alvarez’s arguments, which rely on Fifth Circuit cases like Burdine, or on broad readings of older Supreme Court language, run into difficulty: those authorities cannot be used to say the TCCA unreasonably applied clearly established Supreme Court law.

B. Strickland vs. Cronic

Strickland: most ineffective‑assistance claims; you must prove (1) deficient performance and (2) prejudice (a reasonable probability of a different result).
Cronic: very narrow category where prejudice is presumed, such as:
- no lawyer at all at a critical stage;
- counsel literally does almost nothing to challenge the prosecution’s case; or
- extreme circumstances where effective assistance is practically impossible.

The central fight in Alvarez is whether a lead lawyer’s sleeping, in the presence of active co‑counsel, falls into Cronic (no need to prove prejudice) or Strickland (prejudice required). The majority says Strickland + AEDPA governs because the Supreme Court has never applied Cronic in this partial‑absence, co‑counsel scenario.

C. Law of Parties (Texas)

Alvarez was tried under Texas’s law of parties, which allows conviction not just for the shooter, but also for those who intentionally aid, direct, encourage, or assist the crime. Thus, even if Alvarez had not fired a shot, his role in planning, supplying weapons, and driving could support capital murder liability. This background matters because it affects prejudice analysis: even aggressive efforts to show another gang member pulled the trigger may not have saved him from a capital conviction.

D. Procedural Default & Abuse of the Writ

Procedural default: If a state court rejects a federal claim based on a state‑law procedural ground (e.g., “you filed too late,” “you used the wrong vehicle,” “you abused the writ”), federal habeas courts normally respect that bar.
Abuse of the writ: Texas’s rule that a prisoner must raise all known habeas claims in the first application; later applications are barred unless narrow exceptions apply (e.g., newly available facts or law).

To overcome a default, petitioners must show “cause and prejudice” or actual innocence—both difficult standards in practice.

E. Rhines Stay‑and‑Abeyance

Rhines v. Weber allows a federal court to stay a “mixed” habeas petition (containing exhausted and unexhausted claims) and hold it in abeyance while the prisoner returns to state court—but only if:

there is good cause for not having exhausted earlier;
the unexhausted claim is potentially meritorious; and
the request is not part of a pattern of intentional delay.

In Alvarez, the court finds none of these satisfied with respect to Reyes’s late admission; the dissent sees good cause and potential merit.

VIII. Broader Significance and Impact

A. Narrowing the Path for “Sleeping Lawyer” Habeas Relief

Substantively, Alvarez is one of the most restrictive federal appellate opinions to date on “sleeping lawyer” claims under AEDPA in the presence of co‑counsel. Combined with McFarland v. Lumpkin and Hall v. Thaler, it signals that in the Fifth Circuit:

only single‑counsel sleeping‑lawyer cases resemble Burdine and may still potentially benefit from Cronic’s presumption of prejudice (subject to AEDPA);
where co‑counsel is present and functioning, even repeated sleeping by lead counsel will likely be evaluated under Strickland, and AEDPA will make federal relief extremely rare.

B. Constraining the Use of General Principles as “Clearly Established Law”

The opinion also marks a restrictive reading of Andrew v. White. While Andrew endorsed the idea that some general legal principles can be “clearly established” for AEDPA purposes, Alvarez insists that the general principle must be closely tethered to similar factual settings. This will likely make it harder for petitioners to invoke broad doctrinal statements (e.g., about denial of counsel at critical stages) as the basis for relief in novel or hybrid fact patterns.

C. Reinforcing High Bars for Mitigation‑Investigation Claims

On sentencing‑phase ineffective assistance, Alvarez underscores:

the Fifth Circuit’s skepticism of “they could have done more” arguments in mitigation where counsel has engaged an expert and presented a non‑trivial case;
the centrality of qualitative difference in new mitigation—courts will look for evidence that fundamentally recasts the defendant’s life story, not just more detail on familiar themes.

D. Message for Crime‑Lab‑Based Forensic Challenges

The opinion signals that generalized scandal about a lab’s overall quality is not enough for habeas relief where:

case‑specific retesting confirms key results; and
independent, non‑forensic evidence robustly supports guilt.

Procedural default doctrines will often prevent such claims from even reaching merits review if they are first raised in late, successive state petitions.

E. Fairness Concerns and Public Confidence (as Highlighted by the Dissent)

The dissent powerfully underscores the fairness and legitimacy concerns raised when a capital defendant’s lead lawyer repeatedly sleeps through testimony—particularly in front of the jury, which sees the system tolerating apparent indifference to the defendant’s fate. Although the majority holds AEDPA forecloses relief here, the dissent’s reasoning may influence future advocacy in state courts (which are not bound by AEDPA deference) and in direct appeals or initial state habeas proceedings.

IX. Conclusion

Alvarez v. Guerrero is a consequential Fifth Circuit capital habeas decision that:

strictly applies AEDPA deference to a serious Sixth Amendment complaint involving a sleeping lead lawyer, holding that active co‑counsel and the absence of on‑record corroboration preclude relief under either Strickland or Cronic as “clearly established” law;
reaffirms a demanding standard for Wiggins/Andrus mitigation claims, emphasizing the need for qualitatively new, powerful mitigation to establish both deficient performance and prejudice; and
treats crime‑lab‑related DNA challenges as both procedurally barred and substantively non‑material where retesting confirms key results and other evidence of guilt is overwhelming.

The decision’s most significant doctrinal contribution lies in its treatment of partial or co‑counsel “absence” under AEDPA: it refuses to expand Cronic’s presumption of prejudice to cases where one of multiple trial lawyers intermittently sleeps, absent Supreme Court authority squarely on point. Going forward, defendants in the Fifth Circuit will face an especially steep climb in “sleeping lawyer” habeas cases where any co‑counsel was present and participating, and in mitigation claims where some reasonable investigation and presentation occurred.

Whether this tight approach to AEDPA deference and narrow reading of “clearly established law” best vindicate the Sixth Amendment’s guarantee of counsel—especially in capital trials—remains a live normative debate, vividly captured in the clash between the majority and Judge Richman’s dissent. But as a matter of binding precedent in the Fifth Circuit, Alvarez substantially constrains federal habeas relief on these fronts.