# Mark Rogers v. James Dzurenda

> Court of Appeals for the Ninth Circuit · February 14, 2022 · 25 F.4th 1171

URL: https://www.frixlaw.com/law-library/cases/6221506

## Case

- **Court:** Court of Appeals for the Ninth Circuit
- **Decided:** February 14, 2022
- **Citations:** 25 F.4th 1171
- **Precedential status:** Published
- **Opinion:** Opinion
- **Nature of suit:** Habeas
- **Cited by:** 17 later opinions in the Frix Law Library

## Citator (automated)

- No negative treatment found by the automated citator. That is not the same as a confirmation that the case is good law; read the citing cases.
- Full citator and citing cases: https://www.frixlaw.com/law-library/cases/6221506

## How later opinions describe it (automated extraction)

- stating that opening statements help “contextualize the evidence the jury will hear and . . . help the jury understand each side’s theory of the case”
- finding deficient performance when counsel’s opening statement gave the jury “inadequate context for how the evidence would relate to the insanity defense”

## Opinion text

FOR PUBLICATION

UNITED STATES COURT OF APPEALS
FOR THE NINTH CIRCUIT

MARK ROGERS, No. 19-17158
Petitioner-Appellee,
D.C. No.
v. 3:02-cv-00342-
GMN-WGC
JAMES DZURENDA; ADAM PAUL
LAXALT; WILLIAM GITTERE,
Warden, OPINION
Respondents-Appellants.

Appeal from the United States District Court
for the District of Nevada
Gloria M. Navarro, District Judge, Presiding

Argued and Submitted April 19, 2021
Seattle, Washington

Filed February 14, 2022

Before: Ronald M. Gould, Andrew D. Hurwitz, and
Mark J. Bennett, Circuit Judges.

Opinion by Judge Gould;
Concurrence by Judge Hurwitz;
Separate Statement by Judge Hurwitz;
Dissent by Judge Bennett
2 ROGERS V. DZURENDA

SUMMARY *

Habeas Corpus

Affirming the district court’s judgment granting Mark
Rogers’s 28 U.S.C. § 2254 habeas corpus petition
challenging his murder convictions, the panel held that:
(1) Rogers satisfied the Strickland v. Washington two-prong
test for ineffective assistance of counsel; and (2) the district
court did not abuse its discretion in conditionally granting
Rogers’s habeas petition and giving the State of Nevada the
option to adjudicate Rogers not guilty by reason of insanity
(“NGRI”) or to retry him.

Because the ineffective assistance claim before the
district court was never adjudicated on the merits by the
Supreme Court of Nevada, the panel reviewed the claim de
novo.

On Strickland’s deficient performance prong, the panel
held that, even applying the presumption of reasonableness,
trial counsel’s investigation, preparation, and execution of
their chosen insanity defense fell below an objective
standard of reasonableness. The panel wrote that (1) trial
counsel’s most significant error was failing to call as a
witness—or consult at all—the expert the trial court had
appointed to assess Rogers’s competency for trial and sanity
at the time of the offenses; (2) this error was compounded by
the inadequate preparation of counsel’s chosen mental health
experts; (3) trial counsel performed deficiently by not
preparing to rebut the State’s mental health expert; and
*
This summary constitutes no part of the opinion of the court. It
has been prepared by court staff for the convenience of the reader.
ROGERS V. DZURENDA 3

(4) trial counsel’s failure to explain the elements of the
NGRI defense to the jury in their opening statement fell
below an objective standard of reasonableness.

On Strickland’s prejudice prong, and focusing on what a
reasonable, impartial juror would find compelling, the panel
concluded there was a reasonable likelihood that Rogers’s
NGRI defense would have succeeded if trial counsel had
performed effectively.

Although trial counsel’s performance was replete with
errors, the panel emphasized that the State’s staffing and
funding of Rogers’s case contributed to those errors.

The panel concluded that the district court—which
conditionally granted the writ with instructions for the State
to either adjudicate Rogers NGRI or retry him—did not
abuse its discretion in fashioning relief, which is narrowly
tailored to address the ineffective assistance of counsel
without awarding Rogers an unwarranted windfall.

Judge Hurwitz concurred in full. He also filed a separate
statement, joined by Judges Gould and Bennett, in which he
emphasized that the difficult issues confronted in this case
might have been avoided had Nevada paid sufficient
attention to the appointment of qualified capital counsel.

Judge Bennett dissented. He wrote that the majority’s
characterization of the trial as a battle of experts obscures
reality. He agreed that Nevada failed Rogers in allowing
such inexperienced counsel to defend him in a capital case,
but wrote that the facts left little room for masterful counsel,
much less merely adequate counsel, to have proven that
Rogers was legally insane when he committed the killings;
4 ROGERS V. DZURENDA

and that even considered together, the alleged errors
identified by the majority did not prejudice the defense.

COUNSEL

Jessica E. Perlick (argued), Deputy Attorney General; Aaron
D. Ford, Attorney General; Office of the Attorney General,
Las Vegas, Nevada; for Respondents-Appellants.

Heather Fraley (argued) and Randolph M. Fiedler, Assistant
Federal Public Defenders; Rene L. Valladares, Federal
Public Defender; Office of the Federal Public Defender, Las
Vegas, Nevada; for Petitioner-Appellee.
ROGERS V. DZURENDA 5

OPINION

GOULD, Circuit Judge:

When newly minted attorney Virginia Shane was
appointed as lead counsel in a capital case involving a triple
murder, she was not set up for success. Indeed, the deck was
stacked against her. Shane’s client, Mark Rogers, stood
accused of murdering three members of the Strode family.
Shane was appointed as Rogers’s attorney a mere four
months after passing the Nevada bar exam. At that time, she
was the only attorney in a satellite office of the Nevada State
Public Defender (“NSPD”). Shane recognized immediately
after her appointment that a “not guilty by reason of
insanity” (“NGRI”) argument was her client’s strongest—
and his only meaningfully supported—defense. Despite
these circumstances, Shane received little or no help
preparing Rogers’s insanity defense until another public
defender—equally inexperienced in presenting an insanity
defense—became co-counsel shortly before trial.

Trial counsel’s representation of Rogers reflected their
lack of experience. Their performance was characterized by
the failure to take basic steps to prepare their chosen mental
health experts for trial and to rebut the State of Nevada’s
foreseeable evidence. They did not call or even consult the
one expert, Dr. Donald Molde, appointed by the court to
address Rogers’s legal sanity at the time of the offense.
These deficiencies made the defense’s insanity case less
supported, less persuasive, and more vulnerable to
predictable and preventable attacks by the prosecution.

The State appeals the district court’s judgment granting
Rogers’s 28 U.S.C. § 2254 habeas corpus petition, which
challenged his murder convictions. We have jurisdiction
under 28 U.S.C. §§ 1291 and 2253. Reviewing Rogers’s
6 ROGERS V. DZURENDA

ineffective assistance of counsel claim de novo, see infra
Section II.A, we affirm. We hold that: (1) Rogers has
satisfied the Strickland v. Washington two-prong test, having
demonstrated both that (a) trial counsel exhibited deficient
performance and (b) that performance prejudiced Rogers;
and (2) the district court did not abuse its discretion in
conditionally granting Rogers’s habeas petition and giving
the State the option to adjudicate Rogers NGRI or to retry
him.

I

A

On December 1, 1980, while hitchhiking near
Winnemucca, Nevada, Rogers was picked up by Robert
Schott. Schott described Rogers as nervous and speaking
erratically. Rogers blurted out statements like, “You may
not believe it[,] but I’m a good American,” and “You may
not believe it[,] but I’m on your side.” After driving roughly
thirty minutes, and while Schott was “on the top of a bridge,”
Rogers bluntly said, “Let me out, now.” Schott let him out.

Around 12:30 p.m. the next day, David Hartshorn picked
up Rogers, who was then hitchhiking about twenty miles
from Imlay, Nevada. Hartshorn also had a strange
conversation with Rogers. Rogers introduced himself as
“Teepee,” and when Hartshorn asked Rogers where he was
going, he said, “Nowhere.” Among other things, Rogers told
Hartshorn that he lived in a pyramid and that “Somebody is
shooting rockets off of Mount Olympus and one of these
days it will hit my pyramid and blow me up.” Rogers also
told Hartshorn, “This is my land. I own it all.” Hartshorn
gave Rogers a can of soda, which was later found at the
crime scene. Rogers also explained that he had slept “[r]ight
here,” referring to a stretch of highway.
ROGERS V. DZURENDA 7

Around 3:30 p.m. that same afternoon, highway
maintenance foreman Earl Smith saw Rogers walking
roughly seventeen miles south of Denio, Nevada, about
130 miles north of Imlay. Smith and then some of his
workers separately gave Rogers a ride. After their
interactions with Rogers, Smith and another highway
maintenance worker described him as nervous, giving odd
answers to questions, and wearing jeans with drawings on
them. Some time that day, December 2, 1980, three
members of the Strode family were murdered. Rogers v.
State, 101 Nev. 457, 461 (1985).

Rogers was next seen three days later. He tried to enter
Canada through the Washington border, wearing a parka-
like coat with what appeared to be a towel wrapped around
his head or neck. Rogers had gotten into an accident in the
Strodes’ truck and abandoned it about ten miles south of the
Canadian border. When questioned by Canadian
immigration authorities, he claimed alternatively that he was
a United States citizen and a Canadian citizen.

While interacting with Canadian authorities, Rogers’s
behavior quickly became angry and erratic. Among other
things, he stated that: (1) he wanted political asylum in
Canada because he was being persecuted by numerous
organizations, including the CIA, motorcycle gangs, the
FBI, and the Mafia; (2) he was the “King of North America”;
and (3) he was the “Emperor of North America.” Rogers
was denied entry to Canada.

After vanishing for a period, Rogers reemerged in
January 1981. He was arrested in Florida for the Strode
murders while standing on the bumper and hanging on to the
luggage rack of a station wagon driving down the highway.
Rogers told an officer he was standing on the bumper
because two men were pursuing him. Rogers then told the
8 ROGERS V. DZURENDA

officer, “God knew me and that we are all a part of mother
nature.” In his pocket was a small pad which contained no
writing but included drawings of pyramids, figure eights,
and other symbols. During questioning, Rogers wrote “I
belong to the government” on a paper. Later, at the jail,
Rogers claimed that he killed the Strode family in self-
defense. Rogers, 101 Nev. at 462.

B

Rogers was charged in Nevada with capital murder.
Shortly after his arrest, and still in January 1981, Virginia
Shane was appointed as his lead counsel. Shane worked for
the NSPD in its Winnemucca office, where for most of the
time she was the only public defender on site. The
Winnemucca office was three hours from the main office in
Carson City. Shane—who had passed the bar just four
months before her appointment—had little legal experience
and had never handled a capital case. She also had a
workload of over eighty cases, including another capital case
and a first-degree murder case. Approximately three months
before trial, and several months after Shane was appointed,
Robert Bork was added to the defense team. Bork had been
a lawyer for only four years at the time of Rogers’s trial.
Like Shane, Bork had no experience presenting an insanity
defense or handling a capital case.

Shortly after her appointment, Shane decided to pursue
an NGRI defense. Shane and the prosecutor requested a
mental health evaluation for Rogers, to be conducted at the
Lake’s Crossing Center for Mentally Disordered Offenders
(“Lake’s Crossing”).

The trial court appointed three psychiatrists to assess
Rogers. The court directed Dr. Donald Molde to evaluate
Rogers to determine whether he was competent to stand trial
ROGERS V. DZURENDA 9

and assist in his defense, and whether he was sane at the time
of the offense. The court appointed two other psychiatrists
solely to address Rogers’s competency to stand trial:
Dr. Louis Richnak, the Lake’s Crossing medical director,
and Dr. Phillip Rich, who also did evaluations at Lake’s
Crossing and taught at the University of Nevada’s medical
school.

At Rogers’s competency hearing, Dr. Richnak testified
that Rogers was schizophrenic and not competent to consult
with his attorneys. Dr. Molde testified that Rogers “had
signs and symptoms of paranoid schizophrenia,” but he
nevertheless thought Rogers was competent to stand trial.
Significantly, however, Dr. Molde opined that Rogers was
likely not competent at the time of the offense or his arrest.
Lake’s Crossing psychologists Martin Gutride and Robert
Hiller both testified that Rogers was competent to stand trial.
The trial court ordered further testing and evaluation.

Several months after the competency hearing, the trial
court found Rogers competent to proceed to trial.

C

In the eight months before trial, at least ten mental health
professionals evaluated Rogers. Although Shane had
decided to present an NGRI defense immediately after her
appointment, trial counsel waited until the month before trial
to discuss the issue of insanity with any mental health
professional. 1

1
Trial counsel asked one psychiatrist, Dr. Ira Pauly, to do a mental
status examination two months before trial on the issue of competency.
But it was not until October 1981, one month before trial, that trial
10 ROGERS V. DZURENDA

Trial counsel selected three psychiatrists to support
Rogers’s insanity defense: Drs. Richnak (who had testified
at Rogers’s competency hearing), Rich, and Ira Pauly.
Although Dr. Molde had previously testified about Rogers’s
competency at the time of the offense, evaluated Rogers five
times before trial, and diagnosed Rogers with schizophrenia,
trial counsel did not consult him in preparation of Rogers’s
NGRI defense nor ask him to testify at trial.

Despite choosing Drs. Richnak and Rich to testify for
the defense, trial counsel never met with them before trial or
prepared them for their testimony by, among other things,
going through the DSM-III 2 with them or discussing the
report of the State’s expert. Trial counsel also did not give
Drs. Richnak and Rich the police reports detailing Rogers’s
extraordinary behavior before and after the offense. Further,
trial counsel did not even discuss Rogers’s mental state at
the time of the offense with the two experts before they took
the stand. The only expert with whom trial counsel
discussed the issue of insanity before trial was Dr. Pauly.
But trial counsel did not meet with Dr. Pauly in person,
discuss with him the other expert reports that contradicted
his conclusion that Rogers was schizophrenic, or give him
daily progress notes from Lake’s Crossing that detailed
Rogers’s behavior while he was being evaluated.

counsel asked Dr. Pauly to do a second evaluation, after which he
diagnosed Rogers with paranoid schizophrenia.

The DSM-III is the third edition of the Diagnostic and Statistical
2

Manual of Mental Disorders, published by the American Psychiatric
Association in 1980.
ROGERS V. DZURENDA 11

D

At Rogers’s trial, the prosecution explained the elements
that the State needed to prove for the jury to convict. Shane,
by contrast, did not explain Rogers’s NGRI defense in her
opening statement.

The defense presented three expert witnesses. Drs. Rich
and Richnak opined that Rogers was schizophrenic, but they
also maintained that it was not possible for them to
determine his mental state at the time of the offense because
of Rogers’s unwillingness or inability to discuss the crime.
Dr. Pauly testified that Rogers was insane at the time of the
offense, and that he had arrived at this conclusion based on
his brief evaluation of Rogers and, “most importantly,”
bystander reports. He also noted that Rogers had displayed
violent and antisocial behaviors in the past.

On rebuttal, the State called Lake’s Crossing
psychologist Martin Gutride. Based on his daily interactions
with Rogers and psychological testing, Dr. Gutride testified
that he had diagnosed Rogers with antisocial personality
disorder (“ASPD”).

The prosecution’s closing argument emphasized
testimony from the defense experts that Rogers was violent
and deceptive. The State argued that Rogers was a theatrical
actor whose bizarre behavior was evidence he knew he was
guilty and was trying to avoid criminal liability. The defense
split their closing argument, with Bork principally arguing
that Rogers did not commit the offense and Shane arguing
that Rogers did not know the nature or quality of his acts at
the time of the offense.
12 ROGERS V. DZURENDA

The jury rejected the NGRI defense and found Rogers
guilty of three counts of first-degree murder. Rogers was
sentenced to death.

E

Rogers’s conviction and sentence were affirmed by the
state appellate court on direct appeal. Rogers then sought
post-conviction relief in state court. Mary E. Boetsch, who
was appointed as post-conviction counsel, spent about eight
hours drafting a five-page petition containing seven claims
for relief. The petition argued that trial counsel were
ineffective “in presenting an insanity defense in that defense
counsel presented the testimony of Dr. Rich[,] who stated
that it would be difficult for him to determine the mental
status of Petitioner at the time of the offenses as he was not
present there and had no facts or history[,] just prior to the
testimony of Dr. Pauly, who testified that he could make
such a determination.” In preparing Rogers’s petition,
appointed post-conviction counsel did not (1) investigate or
meaningfully interview trial counsel; (2) consult experts
regarding Rogers’s mental state at the time of the offense;
(3) interview trial witnesses; (4) request Lake’s Crossing
records; (5) interview Rogers’s family or friends; or
(6) review the complete trial record.

After a hearing, the post-conviction state court found
Rogers competent and proceeded with a hearing on the
petition for post-conviction relief. The court then rejected
the claims for relief and concluded that “any alleged errors
of trial counsel were either trial strategy choices made by the
trial attorneys or errors which were not so serious that
counsel was not functioning as counsel guarantee[d] by the
Sixth Amendment.” The Supreme Court of Nevada affirmed
the denial of post-conviction relief.
ROGERS V. DZURENDA 13

Rogers filed his first federal habeas petition in 1987.
“Twice, his federal petitions contained both exhausted and
unexhausted claims, and twice his federal petitions were
stayed, and ultimately dismissed without prejudice so that
Rogers could return to state court, file new state petitions for
post-conviction relief, and present the unexhausted claims in
state court.” Rogers v. McDaniel, 793 F.3d 1036, 1040 (9th
Cir. 2015). Both successive state petitions were dismissed.
Id.

Rogers filed his third petition, the operative petition in
this appeal, in June 2002. Id. Claim Five contended that
trial counsel were ineffective for failing to investigate or
adequately present the chosen insanity defense. Rogers
asserted, among other things, that trial counsel did not
properly investigate his background to provide that
information to his experts, and trial counsel did not impeach
Dr. Gutride’s testimony with readily available evidence.

In 2011, the district court rejected Claim Five on the
merits. On appeal, we remanded for the district court to
reconsider Claim Five in light of Martinez v. Ryan, 566 U.S.
1 (2012), and Dickens v. Ryan, 740 F.3d 1302 (9th Cir. 2014)
(en banc). See Rogers, 793 F.3d at 1045. On remand, the
district court conducted an evidentiary hearing on Claim
Five and heard testimony from both of Rogers’s trial
attorneys, defense expert Dr. Pauly, Dr. Molde, and a
Strickland expert, Martin Wiener.

At the evidentiary hearing, trial counsel testified to their
lack of experience presenting an insanity defense and the
mistakes they made in preparing their witnesses for trial.
Trial counsel admitted to not taking basic steps to prepare
their witnesses for trial and conceded that they had no
strategic justification for those omissions. While trial
counsel acknowledged that it was important to corroborate
14 ROGERS V. DZURENDA

Rogers’s mental illness in childhood, they admitted that they
did not investigate Rogers’s childhood or provide
information about his childhood to their experts. Trial
counsel testified that they did not give their experts essential
documents that trial counsel already had in their possession,
or which were readily available. Trial counsel admitted to
not preparing to cross-examine Dr. Gutride with evidence
such as Lake’s Crossing records and the DSM-III, and that
they had no strategic reason for not doing so. Finally, trial
counsel admitted to never speaking with Dr. Molde, the only
court-appointed expert to assess Rogers’s mental state at the
time of the offense, and who had stated after his examination
that he believed Rogers was insane when committing the
crime.

Drs. Pauly and Molde’s testimony confirmed trial
counsel’s account of the preparation for Rogers’s trial.
Dr. Pauly confirmed that he never met with trial counsel
face-to-face and never received certain documents.
Dr. Molde confirmed that he had never spoken with trial
counsel. Dr. Pauly explained how the Lake’s Crossing daily
progress notes, which he did not receive before trial, would
have been helpful to him. He also emphasized that evidence
of Rogers’s behavior as a child could have been explained as
prodromal symptoms of schizophrenia. Dr. Molde opined
that Dr. Pauly’s testimony at Rogers’s trial was strong, but
he also described the testimony he would have provided at
trial if he had been asked to testify, including his conclusion
that Rogers was insane at the time of the offense.

Rogers’s Strickland expert, Martin Wiener, concluded
that trial counsel’s performance in presenting Rogers’s
insanity defense “absolutely” did not satisfy the standard of
practice in Nevada in 1981. As he explained, “it was just a
series of compound mistake[s] . . . that really diminished the
ROGERS V. DZURENDA 15

strength of the insanity defense at trial.” Wiener discussed
the importance of preparing experts for their testimony by
providing them all relevant records. He then testified that
the standard of practice required counsel to prepare to rebut
the State’s experts and evidence. For each of the above,
Wiener explained how trial counsel performed deficiently by
failing to meet the prevailing standard.

The district court first found that Claim Five had been
“fundamentally altered” from the ineffective assistance of
counsel claim Rogers had raised in the state post-conviction
proceeding. As a result, Rogers’s Claim Five was
procedurally defaulted under Martinez v. Ryan. The district
court next found that Rogers overcame the procedural
default because his first state post-conviction counsel was
ineffective for failing to raise Claim Five. Accordingly, the
district court reviewed the merits of Claim Five de novo and
determined that Rogers was entitled to relief.

This timely appeal followed.

II

We first address whether Rogers’s trial counsel was
constitutionally ineffective under Strickland v. Washington,
466 U.S. 668 (1984). We conclude that trial counsel’s
execution of their chosen insanity defense fell below
objective standards of reasonableness and was therefore
deficient performance. Id. at 687. We also conclude that
trial counsel’s deficient performance prejudiced Rogers. Id.

A. Standard of Review

“We review de novo the district court’s decision to grant
or deny a petition for writ of habeas corpus.” Rhoades v.
Henry, 638 F.3d 1027, 1034 (9th Cir. 2011). We also review
16 ROGERS V. DZURENDA

ineffective assistance of counsel claims, which present
mixed questions of law and fact, de novo. Id. But we review
“[f]actual findings and credibility determinations made by
the district court in the context of granting or denying the
petition . . . for clear error.” Larsen v. Soto, 742 F.3d 1083,
1091–92 (9th Cir. 2013) (quoting Lambert v. Blodgett,
393 F.3d 943, 964 (9th Cir. 2004)).

Because Rogers filed his operative habeas petition in
2002, his appeal would ordinarily be governed by the
deferential standard of review mandated the Antiterrorism
and Effective Death Penalty Act of 1996 (“AEDPA”),
28 U.S.C. § 2244. But that standard of review only applies
“with respect to any claim that was adjudicated on the merits
in State court.” 28 U.S.C. § 2254(d). Here, the district court
determined that Claim Five, as alleged in the operative
petition, included new factual allegations such that the claim
previously considered by the state court was “fundamentally
altered.” See Dickens, 740 F.3d at 1319. If a claim has been
fundamentally altered, it was not adjudicated on the merits
in state court and is procedurally defaulted. Id. at 1318. As
a result, the district court could review Rogers’s Claim Five
only if he overcame the procedural default. The district
court then found that Rogers overcame the procedural
default because his counsel at the state collateral review
proceeding was ineffective. See Martinez, 566 U.S. at 14.
After finding that Rogers overcame the procedural default,
the district court reviewed the merits of Rogers’s claim de
novo, “[b]ecause this claim, as now presented, was not
adjudicated on its merits in state court.”

On appeal, the State does not challenge the district
court’s decision not to apply AEDPA deference to the state
court’s adjudication of Rogers’s ineffective assistance of
ROGERS V. DZURENDA 17

counsel claim. 3 Because the ineffective assistance of
counsel claim before the district court was never adjudicated
on the merits by the Supreme Court of Nevada, we review
Rogers’s Claim Five de novo. Rodney v. Filson, 916 F.3d
1254, 1258 (9th Cir. 2019) (“[A]ny federally reviewable
claims that were not adjudicated on the merits in state court
are reviewed de novo.”).

B. Ineffective Assistance of Counsel Legal Standard

To succeed on a claim of ineffective assistance of
counsel (“IAC”) under the Sixth Amendment, a petitioner
must prove: (1) that his counsel’s performance fell below an
objective standard of reasonableness (the deficient
performance prong); and (2) that there is a reasonable
probability of a more favorable outcome if counsel
performed effectively (the prejudice prong). Strickland,
466 U.S. at 687–88, 694.

On the deficient performance prong, the petitioner must
show that “counsel made errors so serious that [they] [were]
not functioning as the ‘counsel’ guaranteed the defendant by
the Sixth Amendment.” Id. at 687. Counsel’s errors are
assessed by referring to “prevailing professional norms.” Id.
at 688. There is no checklist of required events for
satisfactory attorney performance, and counsel is “strongly
presumed” to have performed effectively. Id. at 688, 690.
Because “[e]ven the best criminal defense attorneys would
not defend a particular client in the same way,” id. at 689,
leeway must be allowed for tactical decisions at trial. Courts

3
Although a habeas court may exercise its “discretion to raise
procedural default sua sponte if doing so furthers” the interests of
comity, federalism, and judicial efficiency, Boyd v. Thompson, 147 F.3d
1124, 1127 (9th Cir. 1998), we decline to exercise that discretion here.
18 ROGERS V. DZURENDA

must also adopt counsel’s perspective at the time of the
challenged conduct to avoid the “distorting effects of
hindsight.” Id.

Even considering the deference owed to strategic
decisions by counsel, see id., courts have held that counsel
can perform deficiently by, among other things, failing to
adequately investigate the defendant’s background, see
Williams v. Taylor, 529 U.S. 362, 390–91 (2000); to prepare
witnesses, including mental health experts, see Bean v.
Calderon, 163 F.3d 1073, 1080–81 (9th Cir. 1998); and to
prepare to rebut opposing counsel’s witnesses with known
or readily available information, see Harris v. Blodgett,
853 F. Supp. 1239, 1265–66 (W.D. Wash. 1994), aff’d sub
nom. Harris v. Wood, 64 F.3d 1432, 1436, 1438 (9th Cir.
1995).

On the prejudice prong, “[a] reasonable probability is a
probability sufficient to undermine confidence in the
outcome.” Strickland, 466 U.S. at 694. This is not a “more
likely than not” standard, but the likelihood of a different
result must be more than “just conceivable.” Harrington v.
Richter, 562 U.S. 86, 111–12 (2011). The prejudice prong
“focuses on the question whether counsel’s deficient
performance renders the result of the trial unreliable or the
proceeding fundamentally unfair.” Lockhart v. Fretwell,
506 U.S. 364, 372 (1993). The prejudice inquiry does not
focus solely on “mere outcome determination”; attention
must also be given “to whether the result of the proceeding
was fundamentally unfair or unreliable.” Id. at 369.

C. Deficient Performance Prong

On the deficient performance prong, we hold that trial
counsel’s investigation, preparation, and execution of their
chosen insanity defense fell below an objective standard of
ROGERS V. DZURENDA 19

reasonableness. See Strickland, 466 U.S. at 687–88. Trial
counsel made several distinct errors that constitute deficient
performance in preparing and presenting the insanity
defense. We address each in turn.

1

Trial counsel’s most significant error was failing to call
as a witness—or consult at all—Dr. Molde, the expert the
trial court had appointed to assess Rogers’s competency for
trial and sanity at the time of the offenses.

Shane settled on an NGRI defense “[i]mmediately” upon
her appointment in January 1981. When Shane first visited
Rogers at the jail, she recognized his bizarre behavior as
consistent with schizophrenia. Dr. Molde was appointed to
evaluate Rogers to determine not only if he was competent
to assist in his defense, but also to determine whether, on the
date of the crime, Rogers “was in possession of mental
capacities sufficient to distinguish between right and wrong
and to comprehend the nature and quality of the acts with
which he has been charged.” At the competency hearing,
Dr. Molde testified that Rogers “had signs and symptoms of
paranoid schizophrenia,” but he believed Rogers was
competent at that time to proceed to trial. Dr. Molde also
testified, however, that Rogers was probably not competent
at the time of the offense or his arrest. Despite being aware
of Dr. Molde’s opinion on Rogers’s competence at the time
of the offense, trial counsel did not consult him in preparing
Rogers’s NGRI defense, nor did they call Dr. Molde to
testify as a trial witness. At the evidentiary hearing before
the district court, Shane testified that she did not talk to
Dr. Molde about his testimony at any time after the
competency hearing.
20 ROGERS V. DZURENDA

Trial counsel’s decision not to consult or use Dr. Molde
as a witness fell far below an objective standard of
reasonableness. See Strickland, 466 U.S. at 687–88. The
NGRI defense required the jury to apply the M’Naghten rule
and determine whether Rogers “knew the nature and quality
of [his] acts, had the capacity to determine right from wrong
or knew whether [he] was doing wrong when [he] committed
the crime.” Clark v. State, 588 P.2d 1027, 1029 (Nev. 1979).
Under the M’Naghten rule, the “ultimate issue” was not
whether Rogers was schizophrenic, but whether that
schizophrenia rendered him incapable of determining right
from wrong at the time of the offense.

Because the jury had to determine legal sanity, trial
counsel had no tactical reason not to ascertain Dr. Molde’s
opinion to see if it was consistent with his prior opinion that
Rogers was not competent at the time of the offense. The
State contends, on the other hand, that it was reasonable for
trial counsel not to call Dr. Molde as a witness because trial
counsel called Dr. Pauly, who opined that Rogers was
legally insane at the time of the crime. Even if Dr. Molde
would have been a better expert to select than Dr. Pauly, the
State contends, “counsel[’s] decisions regarding which
witnesses to present is a strategic decision entitled to almost
absolute deference.” But deference is afforded only for
“strategic choices made after thorough investigation of law
and facts relevant to plausible options.” Hernandez v.
Chappell, 923 F.3d 544, 550 (9th Cir. 2019) (quoting
Strickland, 466 U.S. at 690); see also Lambright v. Schriro,
490 F.3d 1103, 1120 (9th Cir. 2007) (per curiam) (“Only
after a thorough investigation can a less than complete
presentation of . . . evidence ever be deemed reasonable, and
only to the extent that a reasonable strategy supports such a
presentation.”). Here, trial counsel made no strategic choice.
Shane never spoke to Dr. Molde and admitted that she
ROGERS V. DZURENDA 21

simply “dropped the ball. I mean, he was there. He was
already paid by the court. I should have brought him in
because his opinion was consistent.”

If counsel had consulted with Dr. Molde and then
decided to present only Dr. Pauly on the insanity issue, such
a decision might warrant traditional deference. But we do
not give deference to counsel’s choice of one expert over
another when they did not investigate the choice and had no
informed view on whether the testifying witness would be
stronger than the non-testifying witness. See Correll v.
Ryan, 539 F.3d 938, 949 (9th Cir. 2008) (“An uninformed
strategy is not a reasoned strategy. It is, in fact, no strategy
at all.”); see also Porter v. McCollum, 558 U.S. 30, 39–40
(2009) (per curiam) (counsel ineffective for failing to “take
the first step of interviewing witnesses” and “ignor[ing]
pertinent avenues for investigation of which he should have
been aware”).

Furthermore, the failure to call Dr. Molde as a witness
was deficient performance because Dr. Molde’s testimony
would have lacked deficiencies that existed with regard to
other expert witnesses that trial counsel did decide to call. 4
First, Dr. Molde was a neutral, court-appointed expert,

4
The State contends that it was reasonable for trial counsel not to
call Dr. Molde because his testimony would have been “cumulative” to
Dr. Pauly’s. This argument conflates the Strickland deficient
performance prong with the prejudice prong. We view counsel’s
performance at the time of counsel’s conduct, and at the time counsel
failed to consult Dr. Molde, they could not have known whether
Dr. Molde’s testimony would be cumulative and therefore could not
have made a strategic choice not to call him for that reason. Whether
Dr. Molde’s testimony would have created “a reasonable probability” of
different result at trial, on the other hand, is a question about Strickland
prejudice. We address the prejudice prong in Section II.D.
22 ROGERS V. DZURENDA

whereas Dr. Pauly was hired by defense counsel. Second,
the testimony Dr. Molde would have given at trial was
consistent with his earlier testimony that Rogers was not
competent at the time of the offense. Third, unlike
Dr. Pauly, Dr. Molde had reviewed all the relevant records
and had seen Rogers on more occasions than any of the
mental health professionals to testify other than Dr. Gutride.
Indeed—and in a very negative contrast—at trial, Dr. Pauly
admitted that he came to his conclusion on insanity after
spending only three-and-a-half hours with Rogers and
without reviewing the Lake’s Crossing daily progress notes.
During closing arguments, the prosecutor emphasized
Dr. Pauly’s shortcomings as a witness:

Again, all I can tell you, ladies and
gentlemen, is that we had ten various
psychiatrists who indicated that they had
examined this defendant, trained psychiatric
people, over a period of eight to ten months,
appointed by the court, and we find one man
that will come into the courtroom and give an
opinion, who will give an opinion at all, with
regards to whether at that time the defendant
knew right from wrong.

Do you know why that man is here?

Well, he had an interest in being here.

He was hired especially by the defense to
come into this courtroom.
ROGERS V. DZURENDA 23

He was not court appointed as he indicated
. . . .5

Dr. Molde’s testimony, on the other hand, would not have
been vulnerable to such “hired gun” attacks, and with his
testimony to bolster Dr. Pauly’s, the prosecution could not
have faulted Rogers for presenting only one expert on the
ultimate issue. Because trial counsel’s decisions about
Dr. Molde cannot be explained as trial strategy, we hold that
trial counsel performed deficiently in failing to consult
Dr. Molde and to call him to testify.

2

Trial counsel’s error in not consulting and calling
Dr. Molde was compounded by the inadequate preparation
of their chosen mental health experts: Drs. Pauly, Richnak,
and Rich. We distinguish this error from trial counsel’s
threshold decision to use these experts to support the NGRI
defense at all. The State misconstrues one of Rogers’s
arguments, stating that “the main thrust of the [district]
court’s conclusion is based on [trial counsel’s] presentation
of Rogers’[s] insanity defense through” these three experts,
two of whom did not opine on the ultimate issue of insanity.
But even if we assume that counsel had good reasons to call
these experts and that their proposed testimony was likely to
be helpful in some ways to Rogers, it does not excuse the
lack of preparation once those experts were chosen.

5
The dissent argues “[t]hat Dr. Pauly acknowledged being hired by
defense counsel is a non-sequitur.” Dissent at 79. Clearly, however, the
prosecution believed that it was relevant that Dr. Pauly was hired by the
defense.
24 ROGERS V. DZURENDA

We conclude that trial counsel’s inadequate preparation
of their mental health experts constitutes deficient
performance. To begin, trial counsel left Dr. Pauly too little
time to prepare an opinion. Although Shane settled on the
NGRI defense immediately after her appointment, she and
Bork did not begin preparing Dr. Pauly for his testimony—
testimony on what Shane believed was Rogers’s sole
supported defense—until the month before trial. Trial
counsel has given no reason why they waited so long to
contact Dr. Pauly, who was also the only expert they
intended to ask about the ultimate issue of insanity.

Even after finally contacting Dr. Pauly, trial counsel did
not provide him with documents that were essential to his
testimony, and the documents they did provide were not
adequately discussed with him. We have repeatedly found
trial counsel ineffective for failing to adequately prepare
experts or provide them with sufficient “informational
foundations.” See, e.g., Smith v. Stewart, 189 F.3d 1004,
1012 (9th Cir. 1999) (holding that “[a] lawyer who should
have known but does not inform his expert witnesses about
essential information going to the heart of the defendant’s
case” was constitutionally ineffective).

Critically, Dr. Pauly never received the Lake’s Crossing
progress notes that detailed Rogers’s daily conduct in the
months leading up to trial, which contained “quite a bit” of
information that supported his diagnosis of schizophrenia,
even though trial counsel had access to them. See Caro v.
Calderon, 165 F.3d 1223, 1228 (9th Cir. 1999) (trial counsel
ineffective where “counsel failed to provide those who did
examine [the defendant] with the information that he had”).
Shane admitted that not providing the daily progress notes to
Dr. Pauly “was definitely a mistake.”
ROGERS V. DZURENDA 25

Furthermore, trial counsel’s failure to investigate
Rogers’s background to fully prepare their experts was
objectively unreasonable. Counsel in a capital case has an
“obligation to conduct a thorough investigation of the
defendant’s background.” Williams, 529 U.S. at 396; see
also Porter, 558 U.S. at 39 (“It is unquestioned that under
the prevailing professional norms at the time of Porter’s trial,
counsel had an ‘obligation to conduct a thorough
investigation of the defendant’s background.’” (quoting
Williams, 529 U.S. at 396)); Bean, 163 F.3d at 1080 (trial
counsel ineffective for failing to adequately investigate the
case to provide informational foundations for mental health
experts’ conclusions). This obligation is especially pertinent
in preparing mental health experts, because any gaps in the
experts’ knowledge about the defendant will severely
undercut the efficacy of the defense. See Brown v. Sternes,
304 F.3d 677, 696 (7th Cir. 2002) (“[I]t is the imprecise and
imperfect nature of the science known as psychiatry that
makes a review of the past available psychiatric records an
essential part of an evaluation of a defendant’s competency
. . . .”).

Here, trial counsel did not investigate Rogers’s
childhood before trial despite wanting to do so and knowing
such information would be important. The NSPD had an in-
house staff investigator, but trial counsel did not use him for
Rogers’s case. Instead, trial counsel used an outside
investigator who was unable to investigate Rogers’s
childhood in Ohio due to limited funding. 6 The failure to
investigate was significant because “[t]he entire theory of the

6
After exhausting their investigative funds, counsel moved for
additional funds for the penalty phase, but did not move for additional
investigative funds for the guilt phase, i.e., funds to support their NGRI
defense.
26 ROGERS V. DZURENDA

prosecution was that [Rogers] was an actor” and thus a
malingerer trying to escape criminal consequences.
Establishing a historical foundation for Rogers’s
schizophrenia diagnosis was also essential because the
State’s expert was prepared to testify that Rogers was faking
his symptoms to gain an advantage in the legal proceedings.

Trial counsel has provided no tactical reason for not
developing the proper historical foundation. Making clear
there was no supportive tactical reason, Shane
acknowledged the need to corroborate Rogers’s mental
illness in childhood, stating that she “wanted to find out what
had been going on with him at the time he was in Ohio at the
onset of his schizophrenia.” Trial counsel also knew that
“criteria from the DSM Three”—on which the mental health
experts who evaluated Rogers relied—required her “to show
certain things starting with [Rogers] at a certain age.”

For similar reasons, trial counsel performed below the
required standard of competence by not ensuring that
Drs. Rich and Richnak received and reviewed all relevant
documents. Trial counsel did not ensure that either expert
was provided the police reports detailing Rogers’s bizarre
behavior, and they did not discuss with them the Lake’s
Crossing daily progress notes or Dr. Gutride’s reports.
Shane did not recall a reason for not providing the police
reports to Dr. Rich, especially because she gave them to
Dr. Pauly. In fact, trial counsel could not recall “do[ing]
anything” to prepare these experts for their testimony, and
no entries in the case activity log indicate that they did so.
Without this necessary information, Drs. Rich and Richnak
were unable to provide a full picture of Rogers’s purported
ROGERS V. DZURENDA 27

schizophrenia. 7 See Wallace v. Stewart, 184 F.3d 1112,
1118 (9th Cir. 1999) (finding counsel ineffective for not
fulfilling “duty to seek out [evidence of the defendant’s
background] and bring it to the attention of the experts”); see
also Sternes, 304 F.3d at 696–97 (“We think it is obvious
that a psychiatrist’s diagnosis, especially when dealing with
a chronic schizophrenic, must necessarily rely heavily on the
patient’s past psychiatric history, family history, criminal
activity, and medical records.” (emphasis omitted)).

Even when trial counsel provided their experts with
useful documents, they failed to adequately discuss those
documents with the experts or discuss how the expert would
respond to expected lines of inquiry on cross-examination.
For example, trial counsel provided Dr. Pauly with the
reports of other mental health professionals who had
evaluated Rogers before trial and police reports detailing
Rogers’s behavior before and after the offense, but they
never actually met with Dr. Pauly face-to-face or discussed
those records with him. Trial counsel also never discussed
with Dr. Pauly the reports from Lake’s Crossing
psychologist Martin Gutride, whose opinions contradicted

7
The district court characterized Drs. Rich and Richnak’s testimony
as “mostly pointless, and worse . . . damaging to Rogers’[s] NGRI
defense.” We need not decide whether counsel’s decision to present two
experts who did not intend to opine on sanity was deficient because it
was deficient to present them without consulting and calling Dr. Molde
and providing them necessary and available documents. We agree with
the State that a successful insanity defense requires counsel to educate
the jury on the nature of mental illness, but the fact that these two
witnesses may have helped the jury to understand why they believed
Rogers was schizophrenic does not cure counsel’s objectively
unreasonable decisions to ignore Dr. Molde and fail to provide the
requisite informational foundations for Drs. Rich and Richnak’s
testimony.
28 ROGERS V. DZURENDA

Dr. Pauly’s conclusions, even though it would be an obvious
line of expected prosecutorial attack on cross-examination.

Trial counsel also did not discuss the ultimate issue of
insanity with either Dr. Rich or Dr. Richnak because they
did not intend to ask the experts to opine on that issue. 8 Trial
counsel offers no strategic justification for failing to at least
discuss the issue with these experts before putting them on
the stand. Without a discussion, trial counsel could not have
known what the experts would say about the insanity issue
on cross-examination. See Combs v. Coyle, 205 F.3d 269,
288 (6th Cir. 2000) (finding trial counsel ineffective in
presenting an expert in support of a diminished capacity
defense but failing to ask that expert for an opinion on the
ultimate issue before trial).

As Rogers’s Strickland expert testified, the prevailing
norms at the time of Rogers’s trial required counsel to give
their competency experts “everything you possibly could”
because “you’d want them to be prepared to . . . render a
reliable opinion.” We conclude that trial counsel performed
deficiently by not adequately preparing their experts, thereby

8
The State appears to construe Rogers’s argument as contending
that Drs. Rich and Richnak were not useful witnesses because they could
not opine on the ultimate issue of insanity. We acknowledge that some
aspects of Drs. Rich and Richnak’s testimony were useful to the defense,
but emphasize that counsel performs deficiently when counsel fail to
even discuss an issue with an expert that counsel knows will be the focal
point of the cross-examination of those witnesses and the critical issue at
trial. Indeed, the trial judge interrupted Dr. Richnak’s direct examination
to ask counsel when they would be asking for his opinion on whether
Rogers was insane at the time of the offense. When informed that would
not be the case, the State then objected to Dr. Richnak’s testimony as
having no bearing on the issues at trial.
ROGERS V. DZURENDA 29

undercutting the persuasiveness of the NGRI defense as
demonstrated through expert testimony.

3

Trial counsel also performed deficiently by not preparing
to rebut the State’s mental health expert, Dr. Martin Gutride.
Dr. Gutride was a psychologist at Lake’s Crossing. At trial,
called as the prosecution’s rebuttal witness, he opined that
Rogers was not schizophrenic or otherwise severely
mentally ill; instead, Dr. Gutride diagnosed Rogers with
ASPD. Dr. Gutride also testified to his concern that Rogers
was malingering, i.e., faking symptoms of schizophrenia to
escape criminal consequences. These conclusions directly
contradicted those of Dr. Pauly, who testified that Rogers
was schizophrenic, and that his mental illness rendered him
legally insane at the time of the offense. Thus, it was
especially important for trial counsel to at least attempt to
undermine Dr. Gutride’s testimony because he examined
Rogers numerous times over several months; this longevity
is in stark contrast to the only expert the defense called to
opine on the ultimate issue, Dr. Pauly, who admitted at trial
that he only saw Rogers for a total of three-and-a-half hours.

We have held, as have our sister circuits, that trial
counsel can be ineffective for failing to challenge a
prosecution expert. See Wood, 64 F.3d at 1436, 1438
(affirming district court’s conclusion that trial counsel was
ineffective for failing to challenge qualifications and
conclusions of prosecution ballistics expert); see also Rivas
v. Fischer, 780 F.3d 529, 550 (2d Cir. 2015) (finding trial
counsel ineffective in part for failing to adequately impeach
a vulnerable State expert at guilt phase); Couch v. Booker,
632 F.3d 241, 247 (6th Cir. 2011) (finding trial counsel
ineffective in part for failing to present expert testimony
refuting the State’s theory on cause of death).
30 ROGERS V. DZURENDA

Here, as the Strickland expert and the district court
correctly pointed out, any reasonable defense attorney would
have anticipated that the State would call Dr. Gutride based
on his reports and his testimony at the competency hearing.
Trial counsel should have known that they needed to prepare
their witnesses to explain the basis for their disagreement
with Dr. Gutride because trial counsel possessed the reports
of the multiple psychiatrists who had disagreed with
Dr. Gutride’s conclusions—and his conclusions were
extremely favorable to the prosecution. Shane admitted she
knew at the time that Dr. Gutride’s report was “very bad for
our defense,” and that she knew if the prosecution had
harmful evidence, they would typically present it.

Even though rebuttal evidence was readily available,
trial counsel did not adequately prepare their experts to rebut
Dr. Gutride’s testimony. First, trial counsel did not discuss
Dr. Gutride’s reports with Dr. Pauly. With adequate
discussion, Dr. Pauly would have been better prepared to
explain the basis not only for his disagreement with
Dr. Gutride’s diagnosis, but also to explain that Dr. Gutride
did not have a sufficient basis for his conclusions.
Specifically, pursuant to the DSM-III, Dr. Gutride lacked
sufficient information to diagnose ASPD because an ASPD
diagnosis requires records from the period before age 15.
Also, it is medically insufficient to base a diagnosis of ASPD
on the patient’s self-reporting, although that is exactly what
Dr. Gutride did. Trial counsel used the DSM-III to prepare
for trial, so the information regarding an appropriate ASPD
diagnosis was available to them.

Second, trial counsel did not prepare to impeach
Dr. Gutride with the fact that it was accepted at the time of
Rogers’s trial that a diagnosis of schizophrenia preempts, or
precludes, a diagnosis of ASPD. This information was
ROGERS V. DZURENDA 31

readily available in the ASPD section of the DSM-III.
Acknowledging that she possessed this information, Shane
testified that she had no reason for failing to prepare her
experts to testify that a schizophrenia diagnosis preempts an
ASPD diagnosis. As Dr. Molde later testified at the
evidentiary hearing before the district court, ASPD by
definition requires a normal mental status examination. The
preemption line of questioning was important because
Dr. Gutride diagnosed Rogers with ASPD, but his reports
described symptoms consistent with schizophrenia, and
therefore symptoms that were inconsistent with the normal
mental status examination that ASPD requires. Similarly,
trial counsel did not effectively confront Dr. Gutride with
evidence that supported a schizophrenia diagnosis and
undermined a finding of malingering. For example,
although trial counsel asked Dr. Gutride about a report in
which he indicated Rogers’s affect was “bland,” they did not
confront Dr. Gutride with the evidence that numerous
psychiatrists found abnormalities of perception and thought,
flatness of affect, and looseness of thought—all
characteristics of an abnormal mental status examination. It
was objectively unreasonable for trial counsel to not point
out that Dr. Gutride himself observed symptoms of a mental
illness that would preempt his own diagnosis.

The State’s contention that trial counsel “preemptively
undermined the credibility and findings of Dr. Gutride” is
unpersuasive. A prepared cross-examination is one of the
most effective tools in a search for the truth. See John Henry
Wigmore, 5 Evidence in Trials at Common Law, § 1367,
at 32 (J. Chadbourn ed. 1974) (“[C]ross-examination is . . .
beyond any doubt the greatest legal engine ever invented for
the discovery of truth.”). For this reason, trial counsel
cannot discharge their duty to anticipate and rebut the State’s
expert merely by presenting their own expert with differing
32 ROGERS V. DZURENDA

views. This is especially true if the reason trial counsel was
ill-prepared to effectively cross-examine the State’s witness
is because of inadequate investigation and preparation. See
Jones v. Ryan, 1 F.4th 1179, 1192 (9th Cir. 2021) (finding
trial counsel ineffective where the error was “based not on
strategy, but on lack of preparation, which left counsel
unaware of the importance of [the] evidence”); Couch,
632 F.3d at 247 (finding trial counsel ineffective for failing
to take “simple steps” to prepare to rebut the State’s
causation evidence). Accordingly, we conclude that trial
counsel performed deficiently by failing to adequately
challenge Dr. Gutride.

4

Trial counsel’s failure to explain the elements of the
NGRI defense to the jury in their opening statement likewise
fell below an objective standard of reasonableness.

The prosecution’s opening statement informed the jury
of the elements of the crimes charged, explaining that “[a]s
you progress and we progress in this trial and as the
witnesses testify it is important that you have some
understanding as to what elements you may have to look
for.” After explaining the elements of first-degree murder,
the prosecutor told the jury that these “elements” are like a
“recipe or a chemical formula,” and at the end of the trial the
jury “can go through and check them off.” Trial counsel for
Rogers, on the other hand, did not tell the jury the elements
of their insanity defense. The prosecution informed the jury
that Rogers had entered a plea of NGRI and had the burden
of proving that defense, but Rogers’s counsel did not define
legal insanity or tell the jury what elements the defense had
the burden of proving. Thus, as Rogers points out, the jury
was given context for how the evidence that would be
presented related to the murder charge, but it was given
ROGERS V. DZURENDA 33

inadequate context for how the evidence would relate to the
insanity defense.

Opening statements are essential to contextualize the
evidence the jury will hear and to help the jury understand
each side’s theory of the case. For this reason, trial counsel
admitted that their failure to explain the insanity defense at
the outset was “a big mistake.” Rogers’s Strickland expert,
Wiener, explained at the evidentiary hearing below why this
omission was objectively unreasonable. He testified that due
to the State’s strong circumstantial evidence, “the possibility
of avoiding a guilty verdict all fell on the insanity defense.”
But because Shane never discussed the definition of insanity
that would be presented to the jury at the end of the trial, the
jury “must have wondered why are we listening to all this
stuff about is it paranoid schizophrenia or is it antisocial
behavior or did he malinger or didn’t he, did he fake, didn’t
he? Why are we listening to psychiatrists?” A reasonably
competent defense attorney would have known that omitting
the proper context to the most important question before the
jury made the NGRI defense less likely to succeed.

5

Shane, as lead counsel for Rogers, knew that an NGRI
verdict was the best, and only, defense strategy. Rogers and
the State agree on appeal that there was strong evidence
connecting Rogers to the murder scene. Given that it would
be difficult for Rogers’s defense counsel to seriously argue
that he did not commit the murders, properly preparing and
executing the NGRI defense was trial counsel’s most
important task. Trial counsel’s failure to prepare their
experts, provide them with critical documents, discuss the
insanity issue with them before putting them on the stand,
investigate Rogers’s background to provide information to
support its experts’ diagnoses, explain the elements of the
34 ROGERS V. DZURENDA

insanity defense to the jury, and use readily available
evidence to rebut the State’s expert, all prevented Rogers
from subjecting the State’s case to “reliable adversarial
testing.” Strickland, 466 U.S. at 688. Even granting
Rogers’s trial counsel the proper presumption of
reasonableness under Strickland, we hold that Rogers’s trial
counsel performed deficiently.

D. Prejudice Prong

We next hold that Rogers has shown there was a
reasonable probability that he would have received a more
favorable outcome had trial counsel performed competently.
See id. at 694.

1

The State’s primary contention on appeal is that NGRI
verdicts are “exceedingly rare” in Nevada. The State bases
its argument that there can be no prejudice around a factually
similar case in which the defendant unsuccessfully asserted
an insanity defense despite using Dr. Molde as an expert.
We decline the State’s invitation to subjectively determine
what a particular jury in rural Nevada is likely to do. Instead,
we ask whether a reasonable juror impartially applying the
totality of the evidence to the legal standards would be
reasonably likely to find those standards met. In Strickland,
the Supreme Court stated that “[t]he assessment of prejudice
should proceed on the assumption that the decisionmaker is
reasonably, conscientiously, and impartially applying the
standards that govern the decision. It should not depend on
the idiosyncra[s]ies of the particular decisionmaker, such as
unusual propensities toward harshness or leniency.” Id. at
695. “[E]vidence about, for example, a particular judge’s
sentencing practices, should not be considered in the
prejudice determination.” Id.
ROGERS V. DZURENDA 35

In effect, the State asks us to look at Nevada juries’
histories regarding a particular defense. This request is
contrary to Strickland’s guidance, and more importantly, it
ignores the fact that the evidence Rogers was insane at the
time of his crime was especially compelling, and thus could
have been more persuasive than the evidence in a
comparable case. Cf. Chappell, 923 F.3d at 554 (finding no
prejudice where counsel did not present diminished capacity
evidence because the strength of defendant’s intent
“contrast[ed] sharply with the relatively weak” evidence on
diminished capacity, which would have negated the requisite
intent). The State points to the factual similarities in the
comparator case, but the fact that the case was also a rural
triple murder and used the same expert tells us little about
the strength of the evidence that the comparator defendant
did not know right from wrong at the time of that defendant’s
offense.

2

Focusing on what a reasonable, impartial juror would
find compelling, we conclude there was a reasonable
likelihood that Rogers’s NGRI defense would have
succeeded if trial counsel performed effectively. In
analyzing Strickland prejudice, we “compare the evidence
that actually was presented to the jury with the evidence that
might have been presented had counsel acted differently.”
Id. at 551 (quoting Clark v. Arnold, 769 F.3d 711, 728 (9th
Cir. 2014)). Here, trial counsel’s deficiencies in failing to
prepare their witnesses and failing to impeach or rebut the
State’s evidence pervaded the entire trial. See Weaver v.
Massachusetts, 137 S. Ct. 1899, 1913 (2017) (suggesting
that ineffective assistance is more likely to be shown where
deficient performance “pervade[d] the whole trial”).
Rogers’s case for prejudice is bolstered by the fact that we
36 ROGERS V. DZURENDA

have identified multiple errors. “When an attorney has made
a series of errors that prevents the proper presentation of a
defense, it is appropriate to consider the cumulative impact
of the errors in assessing prejudice.” Turner v. Duncan,
158 F.3d 449, 457 (9th Cir. 1998); see also Wood, 64 F.3d
at 14s39 (“By finding cumulative prejudice, we obviate the
need to analyze the individual prejudicial effect of each
deficiency.”).

Due to trial counsel’s ineffectiveness during opening
statements, the jury was not told the legal standard for
insanity or what precisely Rogers had to prove. As a result,
the jury had little context for understanding the significance
of the State’s evidence of Rogers’s bizarre behavior before
and after the crime, or the defense evidence of mental illness.

The jury also heard from only a single defense witness,
Dr. Pauly, who was willing and able to testify to whether
Rogers was legally insane at the time of the offense. 9 The
jury heard Dr. Pauly’s opinion only after two mental health
experts testified to being unable to render an opinion, despite
being defense witnesses. 10 Exacerbating the fact that

9
The dissent argues that “[t]he key missing evidence, as opposed to
anything missing from the defense presentation, were facts from which
a jury could conclude that Rogers didn’t know that he was killing the
Strodes or that killing them was wrong.” Dissent at 95–96. This is belied
by the record. Drs. Pauly and Molde directly opined that Rogers was
insane at the time of the crime.
10
Even if it were not constitutionally deficient for trial counsel to
present two experts to opine that Rogers is schizophrenic without also
asking those experts their opinion on legal sanity, it was prejudicial to
inadequately prepare those experts. If Drs. Rich and Richnak had been
adequately prepared, they would have had the same materials as
Dr. Pauly and would have therefore had the tools to form an opinion on
the ultimate issue. As Rogers’s Strickland expert testified: “the only
ROGERS V. DZURENDA 37

Dr. Pauly was the only defense expert to testify to the
ultimate issue, his testimony was subject to attack on several
fronts. The jury heard Dr. Pauly admit that he only met with
Rogers twice (for under four hours total), he was hired by
defense counsel for his opinion, and he failed to review daily
progress notes in rendering his opinion. See Jones, 1 F.4th
at 1196–1201 (finding prejudice where trial counsel
presented a mental health expert who spent four hours with
the defendant instead of a mental health expert who spent an
estimated 130 hours working on the defense case).
Dr. Gutride’s cross-examination even highlighted the fact
that reading his reports, which summarized the daily
progress notes, would not provide the same comprehensive
information as the notes themselves.

reason [Drs. Rich and Richnak] couldn’t express an opinion [on insanity]
was [that] their only source of a possible opinion would have been from
what Mr. Rogers told them. But there was so much other material
available that didn’t require Mr. Rogers to say anything about the event.”
If the two experts had been adequately prepared and still did not have an
opinion on insanity or believed that Rogers was legally sane at the time
of the offense, trial counsel could have presented fewer or different
experts.

Alternatively, Drs. Rich and Richnak could have been prepared to
testify in a way that did not contradict Dr. Pauly’s conclusion that he had
enough information to opine on insanity. Instead, both experts were
forced to admit that they were unable to form an opinion because Rogers
refused to discuss the crime, rather than explain that they had not been
asked to draw such a conclusion. Moreover, Dr. Gutride, a state expert,
testified that he too could not opine on insanity because he only saw
Rogers the year after the crime was committed and it would have been a
“tremendous leap” for him to draw a conclusion without “enough data
available to [him] at the time of the crime.” Thus, the jury was left to
wonder why only one of the testifying mental health experts—and,
conveniently, the hired defense expert—had enough data to draw a
conclusion on Rogers’s sanity.
38 ROGERS V. DZURENDA

The jury also heard Rogers’s mother describe his violent
and impulsive behavior growing up, but because counsel did
not investigate Rogers’s childhood or prepare their experts
with that information, the defense experts were not prepared
to place those behaviors in the proper context. If trial
counsel had investigated Rogers’s background, Dr. Pauly
would have explained to the jury that Rogers’s behavior in
childhood was “quite typical of schizophrenia,” as it
represented “the prodromal stage of a disorder that evolves
later that’s called schizophrenia.” This testimony would
have bolstered Dr. Pauly’s opinion that Rogers was
schizophrenic and undermined Dr. Gutride’s testimony that
Rogers was malingering. See United States v. Laureys,
866 F.3d 432, 440 (D.C. Cir. 2017) (holding that counsel’s
failure to present expert testimony that would have
“significantly bolstered” defense theory prejudiced the
defendant).

Proper investigation also would have uncovered
evidence of Rogers’s mother drinking while pregnant and
Rogers’s childhood abuse, and Dr. Pauly would have
testified that both are contributing factors to the development
of schizophrenia. Courts have consistently held that
evidence of a defendant’s background is relevant and
impactful to the defendant’s mental state while committing
a crime. See Boyde v. California, 494 U.S. 370, 382 (1990)
(“[E]vidence about the defendant’s background and
character is relevant because of the belief, long held by this
society, that defendants who commit criminal acts that are
attributable to a disadvantaged background, or to emotional
and mental problems, may be less culpable than defendants
who have no such excuse.” (quoting Penry v. Lynaugh,
492 U.S. 302, 319 (1989))); Jones, 1 F.4th at 1200 (“We are
persuaded that testimony explaining Jones’s [childhood and
upbringing] would have significantly impacted the overall
ROGERS V. DZURENDA 39

presentation of Jones’s culpability with respect to his mental
state[.]”).

Instead, the jury was able to pair Rogers’s mother’s
testimony with Dr. Pauly’s testimony that Rogers exhibited
antisocial behaviors in adolescence and during the time he
lived with his ex-roommate. On direct examination, trial
counsel asked Dr. Pauly about the prosecutor’s reports that
covered Rogers’s background in Ohio without preparing him
to talk about how those reports fit into his schizophrenia
diagnosis. Based on his review of the reports, Dr. Pauly
testified that in Rogers’s adolescence, he displayed
antisocial behaviors, including incidents of theft and
stabbing, and engaged in “heavy use” of drugs. Although
trial counsel had framed the question to ask about the early
onset of Rogers’s schizophrenia, Dr. Pauly equivocally
stated, “the best I could say from [the evidence provided by
the district attorney’s reports regarding Rogers’s
adolescence] would be that his adolescence was marked by
a good deal of antisocial behavior, fighting, and . . . a good
deal of drug abuse.” Dr. Pauly then stated that he reviewed
evidence from Rogers’s ex-roommate that Rogers’s
behavior had been “erratic,” characterized by “outbursts of
anger,” and Rogers had even thrown an ashtray into a
television set. Not only did Dr. Pauly not show the
connection between the adolescent antisocial behaviors and
his schizophrenia diagnosis (the “prodromal stage”
connection), he also did not testify that his schizophrenia
diagnosis preempts an ASPD diagnosis. If Dr. Pauly had
testified that his diagnosis of schizophrenia preempted any
diagnosis of ASPD, the damage to the defense from
Dr. Pauly’s references to antisocial behavior would have
been mitigated.
40 ROGERS V. DZURENDA

Finally, the jury heard from Dr. Gutride on rebuttal that
Rogers may have been faking his psychotic symptoms and
suffered from ASPD. Because Dr. Gutride was the only
rebuttal witness, he was the last witness to testify. Due to
trial counsel’s ineffectiveness, the jury did not hear defense
counsel impeach Dr. Gutride’s conclusions or his credibility
with readily available evidence. See supra, Section II.C.3;
see also Pointer v. Texas, 380 U.S. 400, 404 (1965)
(“[P]robably no one, certainly no one experienced in the trial
of lawsuits, would deny the value of cross-examination in
exposing falsehood and bringing out the truth in the trial of
a criminal case.”). Without an effective cross-examination,
and because no expert had testified that an abnormal mental
status examination preempts ASPD, the jury was left to
choose between two seemingly plausible theories of
Rogers’s mental health.

3

On the other hand, if counsel had performed effectively,
the jury would have heard the standard for legal insanity
during opening arguments and would have been primed to
consider the testimony of the mental health experts in the
context of the defense’s NGRI theory. Perhaps most
importantly, if trial counsel had contacted Dr. Molde and
given him all the relevant documents, he would have
confirmed Dr. Pauly’s testimony that Rogers “did not know
right from wrong or know the nature and quality of his act.”
Dr. Molde would have explained how the bystanders’
descriptions of Rogers’s behavior and his numerous
examinations allowed him to reach the conclusion that
Rogers was insane at the time of the offense. He would have
testified that Rogers had a “rock-solid” case for NGRI.
Unlike Dr. Pauly, as a court-appointed expert, Dr. Molde
would have been immune from the State’s “hired gun”
ROGERS V. DZURENDA 41

attacks. He would have also been largely immune from
attacks related to the amount of time spent with Rogers
because Dr. Molde conducted more examinations of Rogers
than any other psychiatrist or psychologist except for
Dr. Gutride.

Properly prepared, Dr. Molde’s testimony would have
severely undercut the State’s witnesses. 11 He would have
testified that a diagnosis of ASPD “absolutely requires a
normal mental status examination. There cannot be any
abnormalities of perception or thought or there can’t be any
hallucinations or delusions, any flatness of affect, any
looseness of thought, any of that stuff, because antisocial
personality disorders by definition don’t have any of that.”
Dr. Molde would have testified that “the record is replete
with oddities of perception and thinking and delusional
activity that [Rogers] had.” He would have noted that
“things like loose thought associations, concrete proverb
interpretations, flat affect, a stare, suspicious behavior,

11
Again, even if it was not deficient to call Drs. Rich and Richnak
for the defense, presenting those experts instead of Dr. Molde was
deficient and prejudicial. In contrast to the testimony Dr. Molde would
have given, Drs. Rich and Richnak’s testimony did little to further the
insanity defense for two reasons. First, the opinions of these two experts
lacked credibility because trial counsel did not provide them with the
relevant historical records necessary to support their diagnosis, which the
prosecutor highlighted during cross-examination. See Sternes, 304 F.3d
at 696–97 (emphasizing that a psychiatrist’s diagnosis of chronic
schizophrenic must necessarily rely on defendant’s background and
records). Second, Drs. Rich and Richnak’s opinions that Rogers was
schizophrenic did not reach the ultimate question for the defense.
Because they were unable to testify to the ultimate issue, and said as
much on the stand, their opinions also could not prevent the prosecution
from arguing that only one out of many mental health experts who
examined Rogers were willing to say he was legally insane at the time
of the murders.
42 ROGERS V. DZURENDA

thought-blocking, the grandiosity, odd comments, poverty of
thought, posturing, doodling, food being poisoned . . . all
that stuff is highly consistent with mental illness,
schizophrenia in particular and not due to anything else.”

This testimony would have been important because
Dr. Gutride’s reports described symptoms on Dr. Molde’s
list, including flat affect and paranoid ideations. In other
words, if Dr. Molde’s testimony about preemption had been
presented and accepted, the jury would have had to believe
that the multiple psychiatrists and psychologists who
examined Rogers and found abnormalities of perception or
thought, hallucinations or delusions, flatness of affect, or
looseness of thought—including Dr. Gutride—were
incorrect for Dr. Gutride’s ASPD diagnosis to be medically
valid under the DSM-III. Instead, Dr. Gutride’s statement
that paranoid ideation is “consistent with [paranoid
schizophrenia] but also consistent with many other things”
went unchallenged.

We reject the State’s contention that Dr. Molde’s
testimony would have been merely “cumulative” to
Dr. Pauly’s testimony. Unlike Dr. Pauly, Dr. Molde would
have testified that Dr. Gutride’s ASPD diagnosis was invalid
because he lacked sufficient historical information and
observed traits of abnormal perception and thought that
would preempt an ASPD diagnosis. Dr. Molde’s proposed
testimony would have been more persuasive than
Dr. Pauly’s because Dr. Molde was a neutral court-
appointed expert who saw Rogers many more times than
Dr. Pauly. Dr. Molde’s testimony would have prevented the
State from arguing that only “one man” was willing to testify
that Rogers was insane at the time of the crime. And, in a
battle of the experts, presenting two mental health experts to
ROGERS V. DZURENDA 43

bolster the defense theory would have been more compelling
than one. See Laureys, 866 F.3d at 440. 12

If trial counsel had adequately prepared their experts,
Dr. Pauly also would have been a more effective witness.
Specifically, if trial counsel had discussed the DSM-III
definition of ASPD with Dr. Pauly, he would have agreed
with Dr. Molde that a schizophrenia diagnosis preempts an
ASPD diagnosis. If trial counsel had investigated and
prepared Dr. Pauly with Rogers’s adolescent behavior, he
would have explained that behavior as emblematic of the
prodromal stage of schizophrenia and not ASPD. He would
have stated that Rogers’s mother drinking during pregnancy
and Rogers’s childhood trauma and abuse were contributing
factors to the development of schizophrenia. See Jones,
1 F.4th at 1200–01 (holding that defendant was prejudiced
where trial counsel did not call an expert who could have
provided testimony explaining the contributing factors to the
defendant’s cognitive dysfunction, including prenatal
substance abuse). If trial counsel had provided Dr. Pauly the

12
The dissent repeatedly asserts that additional testimony by
Drs. Rich, Richnak, and Molde would have been merely cumulative,
citing United States v. Schaflander, 743 F.2d 714, 718 (9th Cir. 1984).
See Dissent at 70, 85, 86. In Schaflander, this Court held that the
testimony of twenty-eight uncalled witnesses would have been
cumulative of the fifteen defense witnesses called at trial. Id. at 718. The
facts of this case are far different. If Rogers had even one additional
medical expert testify about the central point of his defense—whether he
was insane at the time of the crime—he would have rebutted the
prosecution’s attack that the only person willing to testify that he was
insane at the time of the crime was a hired gun. Our conclusion is
strengthened by the fact that we are engaged in a de novo review. See
supra Section II.A; see also Dissent at 62 n.11 (agreeing that review of
Rogers’s IAC claim is de novo). Under de novo review, there was a
reasonable probability that additional testimony by Drs. Rich, Richnak,
and Molde would have led to a different outcome.
44 ROGERS V. DZURENDA

Lake’s Crossing daily progress notes, those notes would
have supported his schizophrenia diagnosis. For example,
the notes provided evidence that Rogers cried when he could
not get a blue “point sheet”—a document related to the
institution’s work program—instead of a pink one, and that
he refused food only to later be found looking into the
garbage. The notes also indicated that on the first day
Rogers was at Lake’s Crossing, his “movements were
mechanical and robot-like.” If Dr. Pauly had been prepared
with the progress notes, he would not have had to admit on
cross-examination that he did not review the notes, which he
conceded could have been “of some value,” and his
credibility would not have been so undermined.

Although the State insists that Dr. Pauly explained his
disagreement with Dr. Gutride, Dr. Pauly’s testimony would
have been significantly stronger with proper preparation. If
trial counsel had discussed Dr. Gutride’s reports with
Dr. Pauly, Dr. Pauly would have testified that despite
diagnosing malingering and ASPD, Dr. Gutride’s reports
actually described symptoms consistent with schizophrenia.
Dr. Pauly would have testified that the malingering
diagnosis did not follow from the symptoms described in
Dr. Gutride’s March 17 report, in which Dr. Gutride
described flat affect and paranoid ideations. He would have
testified that Dr. Gutride’s diagnosis was undermined by
other professionals’ reports about Rogers’s behavior.
Dr. Pauly would have testified that the State’s witnesses
described behavior before the offense that was consistent
with schizophrenia, such as paranoid and persecutory
delusions. And he would have explained that Rogers had no
reason to feign mental illness at the time those behaviors
were witnessed, thereby undermining Dr. Gutride’s
conclusion of malingering. Dr. Pauly would also have
explained that a proper ASPD diagnosis requires records
ROGERS V. DZURENDA 45

from the period before age 15, and that it is medically
insufficient to base a diagnosis of ASPD on the patient’s
self-reporting, as Dr. Gutride did. Dr. Pauly’s assertions
about what data is necessary to diagnose ASPD would have
been supported by the DSM-III, which Dr. Gutride admitted
to using to guide his diagnosis.

Had trial counsel adequately prepared to rebut
Dr. Gutride’s testimony, they could have impeached him
with the fact that under the DSM-III he “lacked sufficient
evidence to make th[e] diagnosis” of ASPD. Trial counsel
would also have pointed out internal inconsistencies in
Dr. Gutride’s reports, his lack of experience in diagnosing
malingering, the daily progress notes that undermined his
conclusions, and the evidence the State itself presented
regarding Rogers’s behavior before the offense that
undermined a malingering finding. Trial counsel would
have challenged Dr. Gutride’s assertion that schizophrenia
and ASPD are not necessarily inconsistent with the
information in the DSM-III about preemption.

In the end, the dissent’s prejudice analysis (conducted de
novo, with no deference to the state proceedings) rests on the
assumption that presenting the unequivocal opinion of a
state-employed psychiatrist not hired by the defense, who
examined Rogers close to the time of the events in question
and concluded that Rogers was insane, could not have led a
single reasonable juror to arrive at a different verdict, despite
the fact that the only expert evidence offered by the State on
insanity was the opinion of a state psychologist who
examined the defendant months later, and the State made
much of the fact in summation that the only defense expert
who testified was a hired gun. That analysis erroneously
conflates an individual judge’s subjective post-hoc
speculation about what verdict the jury would actually reach
46 ROGERS V. DZURENDA

with the required Strickland analysis, which requires only a
“reasonable probability” of a different result.

We reiterate that for Rogers to meet his burden on
Strickland’s prejudice prong, he need not show that absent
trial counsel’s errors the NGRI defense would have
definitely succeeded. Rather, “[a] reasonable probability is
a probability sufficient to undermine confidence in the
outcome.” Strickland, 466 U.S. at 694. The prejudice prong
thus “focuses on the question whether counsel’s deficient
performance renders the result of the trial unreliable or the
proceeding fundamentally unfair.” Lockhart, 506 U.S.
at 372. Here, in a case turning on a battle of the experts, trial
counsel’s failure to reasonably choose which experts to call,
investigate Rogers’s history, prepare their experts, and rebut
the State’s expert undermines our confidence in the outcome
of the trial. Applying de novo review, we conclude that
Rogers has shown there was a reasonable probability of a
more favorable outcome if trial counsel had performed
effectively.

E. Institutional Obstacles

Although trial counsel’s performance was replete with
errors, we emphasize that the State’s staffing and funding of
Rogers’s case contributed to those errors. Virginia Shane
became lead trial counsel for Rogers’s capital case in
January 1981, just four months after she passed the Nevada
bar exam. Shane worked at the Winnemucca office of the
NSPD, a satellite of the main NSPD office in Carson City.
There was only one other attorney in Shane’s office from
October to December 1980, before preparation began for
Rogers’s trial, and Shane had little contact with attorneys in
other NSPD offices.
ROGERS V. DZURENDA 47

Shane had been practicing for less than a year when
Rogers’s trial was held in 1981. Despite this, she received
little to no guidance in preparing for Rogers’s trial—a trial
in which he faced the death penalty. Although Robert Bork
was subsequently staffed on the case, the record shows that
Bork did not begin assisting Shane until August 1981,
shortly before trial. Shane decided on and began preparing
the NGRI defense herself despite never having presented an
NGRI defense before Rogers’s trial. She was tasked with
choosing and preparing mental health experts, but she
admitted that she did not know the difference between the
various categories of mental health experts from which to
choose.

Shane’s office—and the NSPD’s office generally—was
overburdened and understaffed. Although the American Bar
Association recommends an attorney’s caseload not exceed
35 to 50 cases, her caseload “was always upwards of 80.” 13
Shane testified that she did not use NSPD’s in-house
investigator in Rogers’s case, and that it was likely because
he was the only public defender investigator for the whole

13
One leading criminal procedure scholar has noted that “[s]tudies
repeatedly reveal that indigent defense delivery systems are woefully
underfunded and accordingly staffed by defense attorneys who often lack
adequate training or supervision and whose caseloads are impossible for
any attorney to manage effectively.” Eve Brensike Primus,
Disaggregating Ineffective Assistance of Counsel Doctrine: Four Forms
of Constitutional Ineffectiveness, 72 STAN. L. REV. 1581, 1612 (2020).
Professor Primus contends that in cases involving systemic institutional
failure, ineffectiveness may be so “structural and pervasive” as to
constructively deny a defendant their Sixth Amendment right to effective
assistance of counsel. Id. at 1594 (citing Powell v. Alabama, 287 U.S.
45, 53–56, 68–71 (1932)). The Supreme Court has similarly expressed
concern with the structural problems in state indigent defense delivery
systems. Id. at 1619 (citing Luis v. United States, 136 S. Ct. 1083, 1087,
1095 (2016) (plurality op.)).
48 ROGERS V. DZURENDA

state outside of Clark and Washoe Counties. Then, when
Shane was forced to turn to an outside investigator, the court
granted only $3,000 in funding, consistent with the limit
prescribed by state law. Shane knew that she needed further
investigation into Rogers’s background in Ohio, but the
court denied her request for additional funds for the penalty
phase.

Given these systemic obstacles, it is clear that we should
not resolve this case merely by pointing out that counsel
have fallen on their own sword in admitting inadequate
performance, but rather should note a caution to the state
government that should have adequately trained and staffed
the case. Although we rest our holding on Strickland, we are
troubled by the adequacy of the representation in Rogers’s
case due at least in part to factors outside of trial counsel’s
personal control. If, as we believe, the right-to-counsel
guarantee is meant to ensure that our adversarial system
reliably tests the prosecution’s case before subjecting a
defendant to incarceration—and here, death—states and
municipalities providing defense in serious criminal cases
should ensure that advocates have the means to do so.

III

We finally address the State’s contention that “the
district court abused its discretion in fashioning relief for
Rogers” because “the federal court has the power of release,
but it has no other power.” After finding Rogers’s trial
counsel ineffective, the district court conditionally granted
the writ with instructions for the State to either, “(1) within
90 days from the date of this order, vacate [Rogers’s]
judgment of conviction and adjudge him [NGRI], and adjust
his custody accordingly, consistent with Nevada law, or
(2) within 90 days from the date of this order, file a notice of
the State’s intent to grant [Rogers] a new trial, and, within
ROGERS V. DZURENDA 49

180 days from the date of this order, commence jury
selection in the new trial.” We conclude that the district
court did not abuse its discretion in fashioning relief for
Rogers. 14

A reviewing “court has broad discretion in conditioning
a judgment granting habeas relief.” Hilton v. Braunskill,
481 U.S. 770, 775 (1987). Indeed, “[f]ederal courts are
authorized, under 28 U.S.C. § 2243, to dispose of habeas
corpus matters ‘as law and justice require.’” Id. “Sixth
Amendment remedies should be ‘tailored to the injury
suffered from the constitutional violation and should not
unnecessarily infringe on competing interests.’” Lafler v.
Cooper, 566 U.S. 156, 170 (2012) (quoting United States v.
Morrison, 449 U.S. 361, 364 (1981)). For instance, in
Murtishaw v. Woodford, we held that an erroneous jury
instruction constituted constitutional error that could not be
deemed harmless. 255 F.3d 926, 974 (9th Cir. 2001). As for
remedy, we reversed the district court’s denial of the habeas
petition with regard to the defendant’s death sentence and
remanded “to the district court with instructions to grant the
writ to the extent that the death penalty sentence is vacated
and a sentence of life imprisonment without parole is
substituted unless the State resentences him within a
reasonable time.” Id.

Similarly, here, the district court gave the State the
option to adjudge Rogers NGRI, thereby preserving state
resources, or retry him. Thus, contrary to the State’s
contention, the district court did not “compel” the State to

14
Despite arguing that federal courts have only one power in
granting a habeas petition—release—the State does not want the option
to release Rogers. The State also does not challenge the aspect of the
district court’s conditional writ that provided an opportunity for retrial.
50 ROGERS V. DZURENDA

adjudicate Rogers NGRI—if the State does not want to
adjudicate him NGRI, it is free to retry him. The relief
ordered is narrowly tailored to address the ineffective
assistance of counsel without awarding Rogers an
unwarranted windfall. Giving due consideration to the
district court’s broad discretion in granting conditional
habeas relief, we hold that the district court did not abuse its
discretion. 15

15
The dissent in its footnote 16 unleashes a range of contentions
primarily by its own assertions. For example, the dissent says that it
doesn’t “see how defense counsel discussing legal requirements for
insanity, along with highlighting some of the evidence . . . , would have
even benefitted Rogers,” with the dissent’s theory being that discussing
legal requirements for the NGRI defense “might have been off-putting
in the opening or might have predisposed the jury against Rogers.” See
Dissent at 66 n.16. But that theory negates the value of opening
argument on any case where a defense of NGRI is permissible and the
elements of a defense should be specified to give the best chance that a
jury will understand the context for the evidence to be presented at trial.

Strangely, the dissent also contends that, because it is clear that
“Rogers brutally murdered three innocent people,” the “evidence of both
his guilt and sanity was overwhelming.” Dissent at 100. But the fact
that Rogers killed the victims, even if clearly established, does not
establish that evidence of “his sanity was overwhelming.” Wholly apart
from the expert opinions, the jury heard that Rogers told Hartshorn that
he lived in a pyramid and that “Somebody is shooting rockets off of
Mount Olympus and one of these days it will hit my pyramid and blow
me up”; told Canadian authorities that he was “Emperor of North
America”; and, when arrested in Florida, had been standing on the
bumper and hanging on to the luggage rack of a station wagon driving
down the highway. See supra pp. 6–8.

The dissent doesn’t adequately explain why it concludes that the
evidence of sanity was “overwhelming.” To be sure, Rogers killed his
victims in calculated and cold-blooded ways and persistently attempted
to avoid capture, including fleeing to the Canadian border. But these
ROGERS V. DZURENDA 51

IV

For the foregoing reasons, we affirm the district court’s
grant of Rogers’s habeas petition based on his ineffective
assistance of counsel claim. The deficiency of counsel
prejudiced Rogers because a better result was reasonably
probable had there been a competent presentation by the
defense, and the error was not harmless because there was a
substantial and injurious effect or influence of the IAC in
determining the jury verdict. See O’Neal v McAninch,
513 U.S. 432, 436 (1995).

AFFIRMED.

HURWITZ, Circuit Judge, concurring:

I join Judge Gould’s opinion in full and have also filed a
separate statement in which Judges Gould and Bennett
concur.

Separate statement of HURWITZ, Circuit Judge, joined by
GOULD and BENNETT, Circuit Judges:

I write separately to emphasize a point Judge Gould
makes in his majority opinion—that the difficult issues we

actions of Rogers are consistent with insanity. The methodical actions
of a killer in a delusional state who does not know right from wrong do
not prevent satisfying the standard M’Naghten definition of insanity.
The critical issue is whether Rogers knew the difference between right
and wrong at the time of the murders, not whether he acted in an entirely
bizarre or plainly methodical manner.
52 ROGERS V. DZURENDA

confront today might have been avoided had Nevada paid
sufficient attention to the appointment of qualified capital
counsel.

I do not doubt that appointed defense counsel tried their
best and were fully competent to try noncapital cases. But
capital defense is not simply routine criminal defense with
higher stakes. Rather, as the Supreme Court aptly put it,
“death is different.” Ford v. Wainwright, 477 U.S. 399, 411
(1986) (plurality op.). It is indisputably the “most
challenging aspect of criminal defense.” Larry Hammond &
Robin M. Maher, The ABA Guidelines: The Arizona
Experience, 47 HOFSTRA L. REV. 137, 138 (2018). The
American Bar Association long ago recognized this,
enacting detailed Guidelines for the Appointment and
Performance of Defense Counsel in Death Penalty Cases.
And the Supreme Court, while not treating the Guidelines as
mandatory, has emphasized their usefulness in evaluating
the adequacy of capital defense counsel’s performance. See
Bobby v. Van Hook, 558 U.S. 4, 8 (2009); Wiggins v. Smith,
539 U.S. 510, 524 (2003).

Rogers nonetheless went on trial for his life defended by
two relatively inexperienced and overworked lawyers. Their
appointments would not have passed muster under the ABA
Guidelines. Neither had tried a capital case, presented an
insanity defense, or received any specialized training in
capital defense. See Am. Bar Ass’n, ABA Guidelines for the
Appointment and Performance of Defense Counsel in Death
Penalty Cases, 31 HOFSTRA L. REV. 913, 961–62, 976–77
(2003) (Guidelines 5.1, 8.1). And both retained oppressive
workloads during this assignment, with each responsible for
up to eighty other cases in the time leading up to Rogers’
trial. Id. at 965 (Guideline 6.1). Had Nevada paid sufficient
attention to counsel’s qualifications and workloads, most of
ROGERS V. DZURENDA 53

what counsel acknowledge as errors in handling this case
would have been avoided.

Since this case was tried, Nevada has taken steps to
address the adequacy of capital representation. The Nevada
Supreme Court now presumptively requires that lead
counsel in a capital case have “(1) acted as lead defense
counsel in five felony trials, including one murder trial, tried
to completion (i.e., to a verdict or a hung jury); (2) acted as
defense co-counsel in one death penalty trial tried to
completion; and (3) been licensed to practice law at least
three years.” See Nev. Sup. Ct. Rule 250(2)(b). The Nevada
Supreme Court has also adopted detailed, if non-binding,
standards for counsel in capital cases that appear to be
consistent with the ABA Guidelines. See Supreme Court of
the State of Nevada, ADKT No. 411, Order (Oct. 16, 2008).

But it remains as true today as it was at the time Rogers
was tried that if a State chooses to impose capital
punishment, it has the obligation to provide experienced
defense counsel with the ability, resources, and time to
handle life and death matters. “The process of sorting out
who is most deserving of society’s ultimate punishment does
not work when the most fundamental component of the
adversary system, competent representation by counsel, is
missing.” Stephen B. Bright, Counsel for the Poor: The
Death Sentence Not for the Worst Crime but for the Worst
Lawyer, 103 YALE L.J. 1835, 1837 (1994).

Far too often our court sees defective performance in
capital cases from trial counsel confronting the death penalty
for the first time. See, e.g., Elmore v. Sinclair, 799 F.3d 1238
(9th Cir. 2015); Hamilton v. Ayers, 583 F.3d 1100 (9th Cir.
2009). This is a highly specialized area of the law, and if a
State believes it is important to have a death penalty, it must
provide counsel competent to provide an effective defense,
54 ROGERS V. DZURENDA

both as to guilt and penalty. We are here today not only
because of counsel’s representation, but also because
Nevada failed to honor its obligation to provide competent
capital counsel. This case serves as a continuing caution of
the consequences of such a failure, both for parties and
victims of crime.

BENNETT, Circuit Judge, dissenting:

As the majority explains, our law requires a petitioner
seeking a writ of habeas corpus based on ineffective
assistance of counsel to show two things: (1) that counsel’s
representation was deficient, meaning it “fell below an
objective standard of reasonableness”; and (2) that such
deficiency was “prejudicial to the defense,” meaning it
created “a reasonable probability that, but for counsel’s
unprofessional errors, the result of the proceeding would
have been different.” Strickland v. Washington, 466 U.S.
668, 688, 692, 694 (1984).

The majority characterizes Mark Rogers’s 1980 Nevada
triple murder trial as a battle of the experts. In its view, the
search for “cumulative prejudice” centers on the testimony
of one expert who defense counsel failed to call as a witness
and other experts who defense counsel should have better
prepared. E.g., Harris ex rel. Ramseyer v. Wood, 64 F.3d
1432, 1439 (9th Cir. 1995). That narrative, however,
obscures reality. I agree that Nevada failed Rogers in
allowing such inexperienced counsel to defend him in a
capital case. But the facts left little room for masterful
counsel, much less merely adequate counsel, to have proven
that Rogers was legally insane when he committed the
killings. Even considered together, the alleged errors
identified by the majority did not prejudice Rogers’s
ROGERS V. DZURENDA 55

defense. I therefore respectfully dissent from the majority’s
affirmance of the district court’s issuance of the writ. 1

I. Factual Background

On December 2, 1980, Emery and Mary Strode and their
disabled daughter Meriam Strode Treadwell were murdered
in their home near Majuba Mountain, Nevada. Rogers v.
State, 705 P.2d 664, 667, 672 (Nev. 1985) (per curiam).
Emery, seventy-one, was shot three times and stabbed twice
with a knife left in his chest. Id. at 667. Mary, seventy-two,
was stabbed in the back and shot in the chest. 2 Id. Meriam,
forty-one, was shot in the back, her wrists bound with an
electric cord. 3 Id. Frank Strode, Emery and Mary’s son and
1
Although I dissent from the majority’s affirmance of the grant of
habeas relief (and would grant Rogers no relief), I agree with the
majority about the form of the relief it allows: Nevada may either adjudge
Rogers not guilty by reason of insanity or retry him. Majority Op. at 49.
I join the majority in rejecting Nevada’s argument that the district court
lacked the power to order that form of relief.
2
The prosecutor told the jury in closing, based on the forensic
evidence and testimony: “This knife was driven by the defendant to its
very hilt breaking through two ribs of Mrs. Strode with tremendous force
and at the same time that knife had to be drawn back out” with such
“great force” that her blood reached the ceiling. “But, as she was close
to death the pathologist testified she was shot approximately no more
than a foot away.”
3
The prosecutor argued: “Meriam was executed. She wasn’t merely
shot. Meriam, this feeble woman, . . . had a gun put up against her back[,]
either touching or so very close it might as well have been touching as
she was shot through the back. Simply said she was executed. There are
no other words for it.” Meriam’s sister-in-law, Linda Strode, noted
Meriam’s various ailments: She suffered from diabetes since age eight,
had only one eye with limited vision, had her pituitary removed, had
brittle bones from her use of cortisone, almost always used a cane and
also had a wheelchair and walker, and slept in a hospital bed.
56 ROGERS V. DZURENDA

Meriam’s brother, and his wife, Linda, discovered the bodies
the next day.

The evidence pointed only to Mark Rogers, who was
twenty-three, the murderer. Rogers was an aspiring actor
living in Los Angeles. A few weeks before Thanksgiving in
1980, Rogers visited friends in nearby San Bernardino.
Before Thanksgiving, while in San Bernardino, Rogers
called his roommate, Doug Morrison, to ask about a friend
of Morrison’s who lived in Denver, Colorado. Rogers
planned to pursue his performing (acting, singing, and
dancing) career there. Rogers called Morrison again at
8:00 a.m. on December 1, the day before the murders, this
time from a payphone in Wells, Nevada. Rogers said that he
wanted to return to Los Angeles and pursue his career there.
He sounded normal during the call. 4 That afternoon, Rogers
caught a ride bringing him from Winnemucca to Imlay,
Nevada, about thirty-five miles closer to the murder scene.
See id.

Rogers continued hitchhiking the next day, when the
murders occurred. That day, “between approximately 12:15
and 12:45 p.m.,” a geologist working nearby offered Rogers
a ride, gave him a can of Seven-Up to drink, and dropped
him off at the Strode residence. Id. Later, authorities lifted
Rogers’s fingerprints “from various items in the Strode
residence, including a Seven-Up can and a glass jar found in
the bedroom under the blanket with the victims’ bodies.” Id.
at 668. Among the other items with Rogers’s prints were a
plate of beans, a cup of coffee, a sugar jar, a creamer jar, and
a dirty coffee pot on the stove. The Strode property included

4
Morrison did not describe any erratic behavior by Rogers during
the time they were roommates. He only noted that Rogers had a few
angry outbursts, discussed below, which he did not view as alarming.
ROGERS V. DZURENDA 57

a sign announcing that it was protected by Winchester,
presumably referring to the rifle company. Other signs in
the home reflected the Strodes’ belief in their right to keep
and bear arms in self-defense. And there was at least one
shotgun stored inside.

The prosecution offered a theory of the murder sequence
based on various facts no longer disputed in this habeas
appeal. 5 When Rogers arrived at the Strode residence, there
was no smoke coming from the chimney of the residence,
contrary to the family’s usual practice on cold days. The
prosecution argued that this showed that the Strodes were
not home. Rogers then, according to the prosecution,
entered the Strodes’ home and fed himself, preparing a plate
of beans and coffee. The Strodes returned home while
Rogers was still there. At some point soon after, Rogers
stabbed Mary in the back before stabbing Emery in the chest.
Next, as the prosecution told it, Rogers shot Emery twice.
As Mary neared death, Rogers shot her from no more than a
foot away. Rogers then shot Emery in the back. He also
shot Meriam at point blank range through the back. Rogers
then cut the cord off an electric drill, tied it around Meriam’s
arm, dragged her body, and dumped it on top of the bodies
of her mother and father. Rogers then placed bed covers
over the bodies. The final entries in Emery and Meriam’s
daily diaries were dated the morning they were murdered,
and a pocket watch found in Emery’s shirt pocket and

5
At trial, defense counsel advanced a merits defense as well as an
insanity defense. As Rogers claimed to have no memory of the events,
defense counsel were likely unable to obtain his permission to abandon
a merits defense. The record is silent on this issue, and Rogers has never
advanced this as a habeas claim.
58 ROGERS V. DZURENDA

damaged by a bullet stopped at 1:00 p.m. Id. at 667. No
guns were found. 6

Evidence also revealed Rogers’s flight after he killed the
Strodes. The plate of beans was left partially eaten and the
cup of coffee partially drunk. “Between 12:30 and 2 p.m.
that same day,” a man driving a blue truck fired shots at a
mechanic driving on a road near the murder scene. 7 Id. That
afternoon, “between 3:30 and 4 p.m.,” a highway
maintenance worker gave Rogers a ride after seeing him on
the side of the road having run out of gasoline. Id. And
Rogers was spotted speeding in the Strodes’ blue truck after
that. 8 Id. Along with the murders, Rogers was convicted of
stealing the Strodes’ blue truck.

Rogers’s flight from the murders also included an
international border crossing. Rogers was seen next at the
Canadian border three days later. That day, about ten to
twelve miles from the border in Washington state,
authorities recovered the Strodes’ truck after it had been in
an accident during a blizzard. Rogers then tried to cross into
Canada on foot. He explained to border security that he was
headed to a nearby town and then to Quebec, where he hoped
to work as an entertainer and a singer. But he was denied
entry because he lacked the proper employment visa. At one
point, Rogers expressed a desire to seek political asylum in

6
The prosecutor argued: “If the defendant didn’t know that killing
these people was wrong, why would he take the guns?”

The prosecutor argued: “Why was he shot at? It was because the
7

defendant had just left the home of the Strodes and he was afraid
someone had discovered his crimes.”
8
The prosecutor argued: “[H]e peel[ed] out of there in a way that
indicate[d] that he was in an awful hurry to be leaving.”
ROGERS V. DZURENDA 59

Canada, though he claimed to be a Canadian citizen at
another point. He also told border security that he was the
emperor of North America and that the FBI and the mafia
were after him. Within a month, Rogers made it to Montreal.
While in Montreal, Rogers called his roommate, Morrison.
This call, as well, was normal. 9

About a month after the murders, on January 4, 1981,
Rogers was arrested in Florida for soliciting a ride on an
interstate. He had been standing on the bumper of a vehicle,
holding onto the luggage rack. He told the trooper who
arrested him that he was standing on the bumper because two
men who had shot at him were after him. At the county jail,
law enforcement recognized that Rogers was wanted in
Nevada. They told Rogers that they would not question him
about the charges but would “advise him about the warrants
lodged against him” and “talk to him about a possible waiver
of extradition.” With no question to prompt him, Rogers
then “made a statement that he had done it in self-defense.” 10

Rogers was charged with three counts of first-degree
murder, attempted murder (for shooting at the mechanic),
and grand larceny (for stealing the Strodes’ truck). Id.
at 668. He went to trial claiming not to remember the events
9
The prosecutor argued: “Isn’t it strange how he always can recall
the telephone number of Mr. Morrison and every time he talks to
Mr. Morrison everything is coherent and fine.”
10
The prosecutor argued: “He said, ‘I did it in self-defense.’ . . .
[A]fter being informed of the charges against him that he was wanted in
Nevada for the murder of three people . . . he said, ‘I did it in self
defense.’ . . . [T]his is not one person but two different people that he
told this to, one clear across the country, and he made the same statement,
‘I did it in self-defense.” . . . There is no self-defense in any of the bodies
that were found. Those bodies weren’t killed in self-defense. But . . .
the defendant recognized and knew that he had committed the acts.”
60 ROGERS V. DZURENDA

of those months. After deliberating for less than seven
hours, the jury returned guilty verdicts on all charges. After
deliberating another four-and-a-half hours, the jury found
two aggravating circumstances: First, that the murders were
committed by a person who was previously convicted of a
felony involving the use or threat of violence to the person
of another. And second, that the murders involved torture,
depravity of mind, or mutilation of the victims. The jury also
found insufficient mitigating circumstances to outweigh the
aggravating circumstances. And so it imposed a death
sentence for each of the three first-degree murders.

II. Discussion

The evidence was overwhelming that Rogers murdered
the Strodes. His counsel, an overworked public defender
who had been a lawyer for only four months and had never
handled a capital case, immediately decided to pursue a not
guilty by reason of insanity (“NGRI”) defense. Another
public defender joined the team about three months before
trial. He had never handled a capital case and had never put
on an NGRI defense.

A group of mental health professionals participated in
the case. Rogers’s defense team hired Dr. Ira Pauly, a board-
certified psychiatrist, and the chair of the Department of
Psychiatry at the University of Nevada School of Medicine,
to assist in the defense. The prosecution called Dr. Martin
Gutride, a psychologist at a State facility, Lake’s Crossing
Center for Mentally Disordered Offenders (“Lake’s
Crossing”), as its expert witness in rebutting the NGRI
affirmative defense. The trial court appointed three
psychiatrists to assess Rogers. Dr. Phillip Rich, a private
practitioner and faculty member of the University of Nevada
School of Medicine who did evaluations at Lake’s Crossing,
was appointed to assess Rogers’s competency to stand trial.
ROGERS V. DZURENDA 61

Dr. Louis Richnak, the medical director at Lake’s Crossing,
was also appointed to assess competency. And Dr. Donald
Molde, a private practitioner who had worked as a
psychiatrist for the Nevada Department of Prisons, was
appointed to assess competency and to assist in Rogers’s
defense.

The majority identifies five primary deficiencies in trial
counsel’s representation that it claims prejudiced Rogers’s
NGRI defense: (1) counsel’s failure to explain the elements
of the NGRI defense to the jury in their opening statement;
(2) counsel’s inadequate preparation of Drs. Rich and
Richnak to testify on the ultimate issue of Rogers’s sanity
when he killed the Strodes; (3) counsel’s inadequate
preparation of Dr. Pauly by failing to provide him the daily
progress notes from Lake’s Crossing or to investigate
Rogers’s background; (4) counsel’s failure to call or consult
Dr. Molde as a potential expert witness; and (5) counsel’s
failure to rebut the State’s mental health expert, Dr. Gutride.
Majority Op. at 35–46. In assessing the cumulative effect of
these alleged errors, the majority concludes that “[h]ere, in a
case turning on a battle of the experts, trial counsel’s failure
to reasonably choose which experts to call, investigate
Rogers’s history, prepare their experts, and rebut the State’s
expert undermines our confidence in the outcome of the
trial.” Id. at 46.

I disagree. Only the failure to call Dr. Molde as an expert
witness was deficient under Strickland. “Even under de
novo review, the standard for judging counsel’s
representation is a most deferential one” under which courts
ask “whether an attorney’s representation amounted to
incompetence under ‘prevailing professional norms,’ not
whether it deviated from best practices or most common
custom.” Harrington v. Richter, 562 U.S. 86, 105 (2011)
62 ROGERS V. DZURENDA

(quoting Strickland, 466 U.S. at 690). The rest of counsel’s
alleged errors do not show such incompetence. 11

Yet even under the majority’s assessment that counsel’s
performance was more broadly deficient, the majority
materially overstates the impact on Rogers’s case. To assess
prejudice, courts ask “whether it is ‘reasonably likely’ the
result would have been different.” Id. at 111 (quoting
Strickland, 466 U.S. at 696). The question is not whether
the court is “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.”
Id. Instead, in all but “the rarest case,” the prejudice showing
is functionally indistinguishable from “a showing that
counsel’s actions ‘more likely than not altered the

11
I join the majority in reviewing Rogers’s ineffective assistance of
counsel claim here de novo. See Majority Op. at 15–17. The Supreme
Court of Nevada concluded in 1987 that “Rogers [had] failed to show
that his counsels’ performance was unreasonable or that it prejudiced
him.” Even so, the district court found that new factual allegations
“fundamentally altered” Rogers’s claim from the one he presented in
state court, meaning that the Supreme Court of Nevada never adjudicated
it on the merits and so it is procedurally defaulted. See Dickens v. Ryan,
740 F.3d 1302, 1319 (9th Cir. 2014) (en banc). The district court also
found that Rogers overcame that procedural default because his state
post-conviction counsel was ineffective. See Martinez v. Ryan, 566 U.S.
1, 14 (2012). It therefore reviewed the merits of the claim de novo. See
Atwood v. Ryan, 870 F.3d 1033, 1060 n.22 (9th Cir. 2017). Nevada does
not challenge the district court’s decision not to apply Antiterrorism and
Effective Death Penalty Act of 1996 (“AEDPA”), 28 U.S.C. § 2254(d),
deference to the state court’s adjudication of Rogers’s ineffective
assistance of counsel claim. See Oral Argument at 1:53–2:02 (“Q: I think
we all agree that because of the history of this case, our review is not
constrained by AEDPA deference. Do you agree? A: That’s correct.”).
And we have no duty to consider procedural default, and the attendant
question of AEDPA deference, sua sponte. See, e.g., Vang v. Nevada,
329 F.3d 1069, 1073 (9th Cir. 2003).
ROGERS V. DZURENDA 63

outcome.’” Id. at 111–12 (quoting Strickland, 466 U.S.
at 693, 697). As a result, “[t]he likelihood of a different
result must be substantial, not just conceivable.” Id. at 112.
In making this assessment, courts must “compare the
evidence that actually was presented to the jury with the
evidence that might have been presented had counsel acted
differently.” Hernandez v. Chappell, 923 F.3d 544, 551 (9th
Cir. 2019) (citation omitted).

Rogers was not prejudiced by counsel’s performance.
Although defense counsel could have done a better job, even
a better presentation of the NGRI defense would have been
very unlikely to succeed. “Under Nevada law, insanity is an
affirmative defense which the presumedly sane defendant
must prove by a preponderance of the evidence.” Ford v.
State, 717 P.2d 27, 33 (Nev. 1986). The inquiry looks at a
defendant’s sanity when the offense is committed. Id. As
for the content of the defense, Nevada applies the
M’Naghten rule, under which a defendant can rebut the
presumption of sanity only by showing that he did not
understand “the nature and quality of [his] acts,” had no
“capacity to determine right from wrong,” or did not know
“whether [he] was doing wrong when [he] committed the
crime.” 12 Clark v. State, 588 P.2d 1027, 1029 (Nev. 1979).
Under this test, “[l]egal insanity is not circumscribed in
meaning or purpose by medical criteria concerning human
psychosis,” and so “[t]he presumption of sanity is not
rebutted merely by a history of prior institutional

12
The jury was instructed: “[I]f you find that the defendant was not
capable of knowing or understanding the nature and quality of his act,
you will find that he was legally insane or, if you find that he was
incapable of knowing or understanding that his acts were wrong, you will
find that he was legally insane.”
64 ROGERS V. DZURENDA

commitments or diagnoses of mental deficiency or
derangement.” Ford, 717 P.2d at 33.

Even a perfectly presented insanity defense, with
additional and better prepared experts, 13 would have been
unlikely to rebut the presumption that Rogers was sane. The
sequence of Rogers’s actions preceding, during, and after the
killings confirmed that he understood he was killing the
Strodes and that he understood that killing the Strodes was
wrong. Yet in granting habeas relief, the district court hardly
engaged with the evidence the prosecution presented or the
strength of its case. 14

The jury rejected the NGRI defense and convicted
Rogers as charged: three counts of first-degree murder in
addition to attempted murder and grand larceny. In
following the court’s instructions, the jury necessarily found,
beyond a reasonable doubt, that each of the three killings
“was preceded and accompanied by a clear, deliberate intent
on the part of the defendant to kill which was the result of

At trial, defense counsel called Drs. Rich and Richnak to testify
13

that Rogers suffered from schizophrenia and Dr. Pauly to testify that
Rogers was schizophrenic and legally insane at the time of the offense.
14
The majority misunderstands the significance of the prosecution’s
evidence and the strength of its case. The State, from start to finish, has
argued both that there was no ineffective assistance, and that there was
no prejudice because of the significance of the prosecution’s evidence
and the strength of its case. And, of course, the district court must
compare “the evidence that actually was presented to the jury with the
evidence that might have been presented had counsel acted differently”
in light of the strength of the prosecution’s case. Hernandez, 923 F.3d
at 551 (citation omitted); see also id. at 554 (“In sum, the jury heard
overwhelming evidence that Hernandez had the specific intent to rape
both Bristol and Ryan, and that he murdered both women willfully,
deliberately, and with premeditation.”).
ROGERS V. DZURENDA 65

deliberation and premeditation, so that it must have been
formed upon preexisting reflection and not under a sudden
heat of passion or other condition precluding the idea of
deliberation.” It also found that Rogers “weigh[ed] and
consider[ed] the question of killing and the reasons for and
against such a choice and, having in mind the consequences,
he decide[d] to and [did] kill.” And the jury found beyond a
reasonable doubt that Rogers acted with the “unjustifiable or
unlawful motive or purpose to injure another, which
proceeds from a heart fatally bent on mischief or with
reckless disregard of consequences and social duty.” 15

As addressed below, the five errors identified by the
majority do not together constitute “cumulative prejudice.”
Wood, 64 F.3d at 1439. The factual sequence of the murders
and the prosecution’s evidence and closing arguments
together reveal the strength of the case for legal sanity.
Although anything is theoretically “possible,” Richter,
562 U.S. at 111, it is not reasonably likely that Rogers would
have obtained a different verdict with competent (or even
perfect) counsel. After all, “even the most competent of
counsel could not have altered the facts.” Hurel Guerrero v.
United States, 186 F.3d 275, 282 (2d Cir. 1999) (citation
omitted).

A. Opening Statement

The majority begins its prejudice analysis with defense
counsel’s opening statement. And it makes much of defense
counsel’s failure to define legal insanity and list the elements
15
For attempted murder, the jury necessarily found, beyond a
reasonable doubt, that Rogers “[t]ried and failed to murder Ray Horn,”
“[w]ith premeditation” and “malice aforethought.” And for grand
larceny, the jury found that Rogers “willfully” and “[w]ith the specific
intent to permanently deprive the owner,” stole a motor vehicle.
66 ROGERS V. DZURENDA

of the NGRI defense at that time. But the majority neglects
to mention what defense counsel did say. Defense counsel’s
opening highlighted Rogers’s bizarre statements and
behavior and told the jury that the evidence would show that
Rogers “was not behaving like a person who knew that he
had murdered three people and was running away from that
type of a crime”—the essence of the insanity defense.
Defense counsel also told the jury: “The most important
things that will come out of the District Attorney’s proof to
you is not whether Mark is a person who did these things but
the most important thing for you to remember is Mark’s
mental state at the time this happened through lay witnesses
and through the psychiatrists who later examined Mark.” 16

16
I don’t see how defense counsel discussing the legal requirements
for insanity, along with highlighting some of the evidence (as counsel
did), would have even benefited Rogers. The discussion of legal
requirements might have been off-putting in the opening or might have
predisposed the jury against Rogers as to the insanity defense before the
jury even heard the trial testimony. “By telling a simple story in an
opening, the lawyer will establish a theme, capture the jurors’ interest,
and help the jurors focus on the important parts of the testimony that is
coming.” Hon. Richard B. Klein, Trial Communication Skills § 21:3 (2d
ed., Nov. 2021 Update). Putting the legal elements of proving insanity
in such a story is unlikely to be something that would hold the jury’s
attention or even be remembered. And it is not intuitive that it would
help the defense. Defense counsel told the jury: “The officers from
Canada will tell you that Mark pulled out a pen . . . and said, ‘This proves
I’m the king of North America.’” The jury would remember that. The
same can’t be said of something like: “Now the judge will instruct you
at the conclusion of the case that you must find Mark not guilty by reason
of insanity if the defense proves to you by a preponderance of the
evidence—preponderance means more likely than not—that Mark was
legally insane. Now Mark is presumed legally sane, and the judge will
tell you what that means. Legal insanity is a diseased or deranged
condition of the mind which makes a person incapable of knowing or
ROGERS V. DZURENDA 67

The majority maintains that defense counsel’s failure to
explain the NGRI defense meant that “the jury was given
context for how the evidence that would be presented related
to the murder charge, but it was given inadequate context for
how the evidence would relate to the insanity defense.”
Majority Op. at 32–33. “As a result,” under the majority’s
assessment, “the jury had little context for understanding the
significance of the State’s evidence of Rogers’s bizarre
behavior before and after the crime, or the defense evi

[Text truncated at 120,000 characters. The full text is on the page linked above.]

---

Source: Frix Law Library, https://www.frixlaw.com/law-library/cases/6221506. Public record. Not legal advice.