State of Arizona v. Dimitri Polanco Romero

• Court: Arizona Supreme Court
• Decided: May 15, 2026
• Docket: CR-24-0237-PR
• Status: Published
• Author: Maria Elena Cruz

Opinion

IN THE

SUPREME COURT OF THE STATE OF ARIZONA

STATE OF ARIZONA, Appellee,

v.

DIMITRI POLANCO ROMERO, Appellant.

No. CR-24-0237-PR Filed May 15, 2026

Appeal from the Superior Court in Pima County The Honorable Catherine M. Woods, Judge No. CR20203538001 AFFIRMED

Opinion of the Court of Appeals, Division Two 258 Ariz. 237 (App. 2024) VACATED

COUNSEL:

Kristin K. Mayes, Arizona Attorney General, Joshua D. Bendor (argued), Solicitor General, Alice M. Jones, Deputy Solicitor General/Section Chief of Criminal Appeals, Michael T. O’Toole, Assistant Attorney General, Phoenix, Attorneys for State of Arizona

David J. Euchner (argued), Pima County Public Defender’s Office, Tucson, Attorney for Dimitri Polanco Romero

Rosemarie Pena-Lynch, Office of Public Defense Services, Gary Kula, Office of the Public Defender, Kristen Reller, Deputy Public Defender, Sherri McGuire Lawson, Office of the Legal Defender, Shannon Burns, Office of the Public Advocate, Steve Koestner, Office of the Legal Advocate, Phoenix, STATE V. ROMERO Opinion of the Court

Attorneys for Amicus Curiae Maricopa County Public Defense Services

Carol Lamoureux, Law Office of Hernandez & Hamilton, PC, Tucson, Attorneys for Amicus Curiae Arizona Attorneys for Criminal Justice

JUSTICE CRUZ authored the Opinion of the Court, in which CHIEF JUSTICE TIMMER, VICE CHIEF JUSTICE LOPEZ and JUSTICES BOLICK, BEENE, MONTGOMERY, and KING joined.

JUSTICE CRUZ, Opinion of the Court:

¶1 This case requires us to determine whether a claim of prosecutorial error depends upon proof of the prosecutor’s state of mind and whether the court of appeals correctly applied the governing standard. We hold that establishing prosecutorial error does not require consideration of the prosecutor’s mental state. Although Dimitri Polanco Romero demonstrated the existence of prosecutorial error in his trial, he has not shown—under a proper application of the framework set forth in State v. Escalante, 245 Ariz. 135 (2018)—that the cumulative error so profoundly distorted the proceedings as to render his trial fundamentally unfair. Accordingly, we affirm his convictions and sentences.

BACKGROUND

¶2 In September 2020, a fight broke out at a nightclub parking lot around two o’clock in the morning. Several shots were fired, and the victim, S.H., suffered and later died from gunshot wounds.

¶3 A nearby Pima County sheriff’s deputy responded after hearing the gunfire. As the deputy approached in his vehicle, he saw a car leave the nightclub parking lot and commit multiple traffic violations; he also saw that the driver was holding a black object. The deputy attempted a traffic stop, but the vehicle fled, leading officers on a four-mile pursuit before it struck a fence. The officers identified Romero as the driver and arrested him at the scene. They recovered a black firearm from the vehicle and found a plastic bag containing cocaine on Romero’s person. ¶4 During their investigation, detectives obtained “grainy”

2 STATE V. ROMERO Opinion of the Court

surveillance footage of the nightclub exit and the parking lot. The footage depicted S.H. and someone later identified as Romero standing in the parking lot. It showed Romero following the victim and firing shots “in [the victim’s] direction multiple times before leaving.” Around this time, other individuals shot their guns into the air, purportedly to disperse bystanders.

¶5 Although detectives did not recover any of the bullets that struck S.H., they determined, based on the video footage, the location within the parking lot where the shooting occurred. Detectives searched that location and recovered four shell casings from the same manufacturer. Law enforcement also test-fired the gun from Romero’s car and compared the resulting casings with those found at the parking lot through the National Integrated Ballistics Information Network (“NIBIN”). A qualified ballistics expert concluded the casings from Romero’s firearm matched those recovered from the parking lot. The State subsequently charged Romero with first degree murder, unlawful discharge of a firearm, fleeing from law enforcement, and possession of a narcotic drug.

¶6 At trial, the prosecutor elicited testimony from a detective identifying Romero and S.H. in surveillance footage. The detective also testified to information he learned from other officers, including the circumstances of Romero’s apprehension following the vehicular pursuit. Although neither detective was qualified as a ballistics expert, the prosecutor asked two detectives whether Romero’s firearm had been “NIBIN compared” to the shell casings recovered from the parking lot. Both detectives testified that there was a “correlation.”

¶7 Throughout the trial the prosecutor frequently asked leading questions during direct examination, particularly when questioning the lead detective about the surveillance footage and Romero’s identification. Romero did not object. In response to Romero’s challenges to the adequacy of the investigation, the prosecutor also questioned defense investigators about their failure to pursue additional investigative leads. Romero did not object in this regard either. The jury found him guilty on all four counts.

3 STATE V. ROMERO Opinion of the Court

¶8 Romero appealed and, in a divided opinion, the court of appeals affirmed. State v. Romero, 258 Ariz. 237 (App. 2024). In his appeal, Romero primarily argued that he should receive a new trial due to multiple acts of prosecutorial error stemming from leading questions and the use of non-expert witnesses. Because Romero had not objected during trial, the court reviewed his claims for fundamental error. Id. at 243 ¶ 9 & n.4. As the court noted, Romero conceded that no single instance of alleged prosecutorial error warranted reversal. Romero instead argued that four categories of conduct cumulatively deprived him of a fair trial: (1) eliciting improper opinion testimony; (2) pervasive leading questions; (3) inadmissible hearsay; and (4) improper burden shifting through cross-examination and rebuttal argument. Id. at 246 ¶¶ 21, 24. The court concluded that only the leading questions and the use of non-expert testimony constituted prosecutorial error. Id. at 250 ¶ 42, 253 ¶ 57. It determined that the cumulative effect of those errors did not require a new trial because they did not “so profoundly distort the trial that injustice is obvious without the need to further consider prejudice.” Id. at 255 ¶ 65 (quoting Escalante, 245 Ariz. at 141 ¶ 20).

¶9 One judge concurred in the result but disagreed that the leading questions constituted prosecutorial error in the absence of an objection or corrective action by the trial court. Id. at 256 ¶ 70 (O’Neil, J., concurring in part). The concurring judge also acknowledged that although In re Martinez, 248 Ariz. 458 (2020), “arguably eliminated intent from consideration in claims of prosecutorial error,” he did not understand it to do so. Id. at 257 ¶ 74.

¶10 The dissent disagreed with the majority on two points. First, it concluded that eliciting certain lay-opinion testimony identifying Romero as the shooter, based solely on the surveillance footage equally available to the jury, constituted prosecutorial error. Id. at 258 ¶ 78 (Sklar, J., dissenting).

¶11 Second, the dissent concluded that Romero met his burden to show that cumulative prosecutorial error deprived him of a fair trial. Id. ¶¶ 79–80. In the dissent’s view, Romero was not required to demonstrate a “profound distortion” of the trial but instead needed to show that the cumulative errors “so infected the trial with unfairness as to make the resulting conviction a denial of due process” and that it “could have affected” the verdict. Id. ¶ 79. 4 STATE V. ROMERO Opinion of the Court

¶12 Romero petitioned this Court for review. We granted review to resolve two recurring issues of statewide importance: (1) whether a defendant asserting prosecutorial error must establish a prosecutor’s culpable mental state, and (2) the proper application of the Escalante fundamental error test when evaluating cumulative prosecutorial error. We have jurisdiction under article 6, section 5(3) of the Arizona Constitution.

DISCUSSION

I. Prosecutor’s Mental State in Determining Prosecutorial Error

¶13 We first address whether a claim of prosecutorial error requires consideration of the prosecutor’s culpable mental state. It does not. Both parties agree on this point. We address the issue nonetheless to correct the court of appeals’ misapprehension on the matter.

¶14 We begin by addressing the terms “prosecutorial error” and “prosecutorial misconduct” as used in Arizona case law. Just recently, this Court clarified that “[t]he term ‘prosecutorial misconduct’ broadly encompasses any conduct that infringes a defendant’s constitutional rights. It sweeps in prosecutorial conduct ranging from inadvertent error or innocent mistake to intentional misconduct.” Martinez, 248 Ariz. at 469 ¶ 45.

¶15 In Martinez, we recognized the American Bar Association’s (“ABA”) distinction between prosecutorial error and ethical misconduct in the context of “prosecutorial misconduct claims.” Id. at 470 ¶ 47; ABA Recommendation 100B (2010). “Misconduct” may imply an ethical rule violation, whereas “error” does not. Martinez, 248 Ariz. at 470 ¶ 47. We explained:

The term “prosecutorial misconduct” has become a term of art in criminal law that is sometimes used to describe conduct by the government that violates a defendant’s rights whether or not that conduct was or should have been known by the prosecutor to be improper and whether or not the prosecutor intended to violate the Constitution or any other legal or ethical requirement.

5 STATE V. ROMERO Opinion of the Court

Id. at 469–70 ¶ 46 (quoting ABA Recommendation 100B (2010)).

¶16 We further cautioned that “any finding of error or misconduct may entitle a defendant to relief, but courts should not conflate that inquiry with the collateral issue of a prosecutor’s ethical culpability.” Id. at ¶ 47. Thus, whether a defendant’s rights were violated does not depend on whether the prosecutor acted intentionally. See also State v. Bible, 175 Ariz. 549, 601 (1993) (“The focus is on the fairness of the trial, not the culpability of the prosecutor.”); Smith v. Phillips, 455 U.S. 209, 219 (1982) (“Past decisions of this Court demonstrate that the touchstone of due process analysis in cases of alleged prosecutorial misconduct is the fairness of the trial, not the culpability of the prosecutor.”).

¶17 Although Martinez arose in a disciplinary context, we confirmed in State v. Murray, 250 Ariz. 543, 548 ¶ 12 (2021), that the same principle governs in criminal cases. There, we observed that “[t]he characterization of the prosecutorial misconduct makes no difference to our ultimate decision in th[e] case, but it makes a world of difference to the prosecutor by whom the error was committed.” Id. Read together, Martinez and Murray make clear that a prosecutor’s culpable mental state is not dispositive for determining whether a defendant’s substantive rights have been violated. The relevant inquiry is whether error occurred and whether that error affected the fairness of the proceeding—not whether the prosecutor acted with a particular mental state.

¶18 We acknowledge that in Pool v. Superior Court we considered a prosecutor’s intent. 139 Ariz. 98, 105–09 (1984). But that consideration was expressly and narrowly limited to “the resolution of the question of when jeopardy attaches,” specifically whether retrial would be barred where a prosecutor engaged in intentional, improper conduct. Id. at 108. Pool addressed double jeopardy principles and the propriety of retrial—not whether a conviction should be reversed based on prosecutorial error. Id. at 105. Accordingly, to the extent cases have drawn more broadly from Pool, we clarify that its intent analysis does not apply to claims seeking reversal for prosecutorial error.

¶19 We further recognize that in several cases we have discussed whether a prosecutor acted intentionally. See, e.g., State v. Montoya, 258 Ariz. 128, 151 ¶ 54 (2024) (“We consider whether ‘persistent and pervasive’ [error] occurred and whether the ‘cumulative effect of the 6 STATE V. ROMERO Opinion of the Court

[errors] shows that the prosecutor intentionally engaged in improper conduct and did so with indifference, if not a specific intent, to prejudice the defendant.’” (alterations in original) (quoting State v. Gallardo, 225 Ariz. 560, 570 ¶ 46 (2010))); see also State v. Robinson, 253 Ariz. 121, 143 ¶ 68 (2022); State v. Morris, 215 Ariz. 324, 339 ¶ 67 (2007); State v. Hughes, 193 Ariz. 72, 80 ¶ 31 (1998). In those cases, however, we did not impose an intent requirement as a prerequisite to establishing prosecutorial error. Rather, intent was noted in connection with the surrounding circumstances of the conduct. We did not hold that a defendant must prove a culpable mental state as an antecedent to establishing prosecutorial error.

¶20 Today we make that point explicit. A claim of prosecutorial error does not require proof that the prosecutor acted intentionally, recklessly, negligently, or with any other culpable mental state. A prosecutor’s intent is not a prerequisite to establishing prosecutorial error or to obtaining relief under Escalante. But see Morris, 215 Ariz. at 335 ¶ 47 (“[A]n incident may nonetheless contribute to a finding of persistent and pervasive misconduct if the cumulative effect of the incidents shows that the prosecutor intentionally engaged in improper conduct and did so with indifference, if not a specific intent, to prejudice the defendant.” (quoting State v. Roque, 213 Ariz. 193, 228 ¶ 155 (2006))).

¶21 By contrast, the operative inquiry in determining prosecutorial error is whether the prosecutor engaged in improper, mistaken conduct and, if so, any prejudicial effect. To the extent any prior language could be read to require proof of intent to establish prosecutorial error, we disapprove of such a reading.

¶22 Prosecutorial error occurs when a prosecutor engages in proscribed conduct, including trial error; violates rules of professional responsibility, evidence, or procedure; violates court orders or rulings; or violates constitutional or statutory provisions. But “[w]e do not reverse cases for mere technical errors when it appears substantial justice has been done.” State v. Jordan, 80 Ariz. 193, 198 (1956).

¶23 Applying that principle here, we begin by acknowledging that the prosecutor’s repeated use of leading questions on direct examination constituted error. Arizona Rule of Evidence 611(c) provides that “[l]eading questions should not be used on direct examination except as necessary to develop the witness’s testimony.” This rule reflects the 7 STATE V. ROMERO Opinion of the Court

principle that “[w]hat is desired is that the trier of fact hear what the witness perceived, not the acquiescence of the witness in counsel’s interpretation of what the witness perceived.” Shirley J. McAuliffe, Law of Evidence (Arizona Practice Series, vol.1, 4th ed. 2024). We have likewise stated that the “general rule is that questions that put the answer into the mouth of one’s witness in chief should not be asked.” State v. McKinney, 185 Ariz. 567, 575 (1996) (quoting Ball v. State, 43 Ariz. 556, 558 (1934)). The prosecutor’s repeated use of such questioning was error, and that conclusion is not dependent upon the prosecutor’s mental state.

¶24 The prosecutor’s use of non-expert testimony was also error. Arizona Rule of Evidence 702(a) requires a witness to be “qualified as an expert by knowledge, skill, experience, training or education” before offering testimony based on “scientific, technical, or other specialized knowledge.” Here, the prosecutor elicited NIBIN and ballistics-matching testimony from a witness who was not qualified as an expert in that field and therefore was not permitted to offer such technical or scientific opinions. This was error. As above, this prosecutorial error arises from a violation of an evidentiary rule, not from any assessment of the prosecutor’s mental state.

II. Whether the Court of Appeals Misapplied the Escalante Test

¶25 The next question before us is whether the court of appeals misapplied the Escalante framework in concluding that Romero failed to establish cumulative prosecutorial error warranting reversal. Although we clarify certain aspects of the governing standard and ultimately vacate its opinion, we agree with its disposition.

¶26 When a defendant fails to object at trial, he must establish fundamental error to obtain appellate relief. Escalante, 245 Ariz. at 138 ¶ 1. The defendant must first demonstrate that trial error occurred. Id. at 142 ¶ 21. He must then establish the error was fundamental by showing that “(1) the error went to the foundation of the case, (2) the error took from the defendant a right essential to his defense, or (3) the error was so egregious that he could not possibly have received a fair trial.” Id. Under prongs one and two, the defendant “must make a separate showing of prejudice.” Id. No separate showing of prejudice is required under prong three because satisfying that prong necessarily establishes both fundamental error and prejudice. Id. At every step, the defendant bears the burden of 8 STATE V. ROMERO Opinion of the Court

persuasion. Id.

¶27 In State v. Vargas, we directed courts and litigants to apply the Escalante framework for claims of cumulative prosecutorial error. 249 Ariz. 186, 189 ¶ 12 (2020). Vargas did not modify Escalante; instead, it illustrated how to apply the third prong in the context of cumulative prosecutorial error. Accordingly, the governing standard remains the Escalante test—not an “Escalante/Vargas” test. As Vargas explained, “[c]onsistent with the third prong of Escalante, a defendant claiming cumulative error based on prosecutorial misconduct need not separately assert prejudice since a successful claim necessarily establishes the unfairness of a trial.” Id. at 190 ¶ 13. We reiterated that a successful claim is one in which the misconduct “so infected the trial with unfairness as to make the resulting conviction a denial of due process.” Id. (quoting Roque, 213 Ariz. at 228 ¶ 152).

¶28 To the extent the court of appeals suggested that a claim of prosecutorial error must be advanced under prong three, that understanding is incorrect. A defendant may pursue relief for a claim of prosecutorial error under any of the three Escalante prongs. Here, however, Romero expressly invoked prong three. Thus, he was required to establish that the cumulative effect of the errors “was so egregious that he could not possibly have received a fair trial.” Id. at 189–90 ¶ 12 (quoting Escalante, 245 Ariz. at 142 ¶ 21).

¶29 We note that the court of appeals erred in stating that it would “[v]iew the facts in the light most favorable to sustaining the jury’s verdicts.” Romero, 258 Ariz. at 255 ¶ 66. Such deference is generally applied only when reviewing a sufficiency of the evidence claim. See State v. Strong, 258 Ariz. 184, 211 ¶ 118 (2024) (sufficiency of the evidence); State v. Atwood, 171 Ariz. 576, 596 (1992) (sufficiency of the evidence); Bible, 175 Ariz. at 595 (motion for judgment of acquittal); State v. Silvas, 91 Ariz. 386, 387, 392 (1962) (motion for new trial); State v. Bentlage, 192 Ariz. 117, 118 ¶ 2 (App. 1998) (suppression rulings). However, fundamental error review of prosecutorial error does not incorporate that same deferential standard. Rather, it presents an objective inquiry that considers the totality of the circumstances. Escalante, 245 Ariz. at 142 ¶ 21. An appellate court must “analyze the entire record to determine whether ‘reasons clearly demonstrat[e] that the case falls within our definition of fundamental error.’” Id. (quoting State v. Gendron, 168 Ariz. 153, 155 9 STATE V. ROMERO Opinion of the Court

(1991)). Accordingly, we evaluate Romero’s claim under this objective framework.

¶30 Applying that objective review to the entire record, we conclude that Romero has established the existence of prosecutorial error but has not demonstrated that the cumulative effect of that error so profoundly distorted the trial that he could not possibly have received a fair trial.

¶31 A defendant asserting cumulative prosecutorial error need not show that each individual error independently constituted fundamental error. Vargas, 249 Ariz. at 190 ¶ 14. Rather, he must establish that, taken together, the errors denied him a fair trial. Id. ¶ 15.

¶32 As noted above, Romero identified multiple instances of alleged prosecutorial error, and the court of appeals concluded that two categories constituted error: (1) the introduction of NIBIN testimony through unqualified detectives, and (2) the repeated use of leading questions that included elements necessary to establish Romero’s guilt. Romero, 258 Ariz. at 250 ¶ 42, 252 ¶ 52. We agree.

¶33 We pause to raise a concern about compliance with our evidentiary rules. The error here occurred in the wake of Cisco, in which the use of a detective’s unreliable and unqualified testimony as to a NIBIN match was deemed error. State v. Cisco, No. 2 CA-CR 2019-0065, 2020 WL 1130744, at *3–*5 ¶¶ 10, 17 (Ariz. App. Mar. 9, 2020) (mem. decision). Cisco was prosecuted by the same deputy county attorney who tried Romero in this case. Our evidentiary rules are not advisory. When a court has already held that certain testimony violates Rule 702, repetition of that conduct undermines confidence in the fairness of the proceedings and the State’s adherence to its constitutional obligations. Although the error here does not warrant reversal under prong three of Escalante, we reiterate that compliance with our evidentiary rules is mandatory.

¶34 Returning to the court of appeals’ decision, the court next addressed whether the cumulative error was fundamental error. Romero, 258 Ariz. at 254–55 ¶ 64. Quoting Escalante, it recognized that cumulative error must “so profoundly distort the trial that injustice is obvious without the need to further consider prejudice.” 245 Ariz. at 141 ¶ 20. Emphasizing the demanding nature of this standard, the majority reasoned 10 STATE V. ROMERO Opinion of the Court

that the cumulative effect must “so infect the trial with unfairness as to make the resulting conviction a denial of due process.” Romero, 258 Ariz. at 255–56 ¶ 68 (quoting Robinson, 253 Ariz. at 143 ¶ 64). That accurately states the governing standard as to prong three under Escalante.

¶35 The court of appeals’ dissent and Romero, however, contend that this “profound distortion” standard is too exacting and propose instead that a defendant need only show that the errors “infected the trial with unfairness” and “could have affected” the verdict. Id. at 258 ¶ 79 (Sklar, J., dissenting). But Escalante expressly states that to satisfy prong three, “the error must so profoundly distort the trial that injustice is obvious without the need to further consider prejudice.” 245 Ariz. at 141 ¶ 20. Nothing in Vargas altered that language. We therefore clarify that the “profound distortion” standard applies to all claims presented under prong three, including those based on cumulative prosecutorial error.

¶36 The rigor of prong three reflects Escalante’s recognition that fundamental error under that prong is reserved for those rare cases in which the integrity of the trial itself has been compromised. Absent such a profound distortion of the proceedings, cumulative error—though improper—does not satisfy prong three of Escalante to justify reversal where the verdict remains supported by properly admitted evidence.

¶37 It is also important to distinguish fundamental error review from harmless error review. “Harmless error review places the burden on the state to prove beyond a reasonable doubt that the error did not contribute to or affect the verdict or sentence.” State v. Henderson, 210 Ariz. 561, 567 ¶ 18 (2005). In contrast, under fundamental error review, the defendant bears the burden of establishing that the cumulative effect of the errors rendered the trial fundamentally unfair. He must prove both the existence of error and, under prongs one and two, resulting prejudice, or under prong three, that “the error was so egregious that he could not possibly have received a fair trial.” Escalante, 245 Ariz. at 142 ¶ 21. Because Romero contends fundamental error existed under prong three, the relevant inquiry is whether the cumulative effect of the errors so profoundly distorted the proceedings that Romero could not have received a fair trial. We do not examine whether the prosecutor’s error was harmless under any prong of fundamental error review.

11 STATE V. ROMERO Opinion of the Court

¶38 After reviewing the entire record, excluding the improperly admitted evidence, and accounting for the excessive leading questions, we conclude that the cumulative errors did not so infect the trial with unfairness as to deprive Romero of a fair trial. The properly admitted evidence independently linked Romero to the shooting. A qualified ballistics expert testified that the cartridge casings recovered from the shooting scene were fired from the firearm found in Romero’s vehicle. Surveillance footage placed Romero at the nightclub and depicted an individual resembling him moving to and from the vehicle he was driving immediately before and after the shooting. The vehicle fled the scene and was stopped only after a pursuit, which culminated in Romero’s arrest. Thus, the evidence properly admitted at trial captured continuous video surveillance or law enforcement observation of Romero from before, during, and after the shooting, including his pursuit and arrest.

¶39 In light of this properly admitted evidence, the prosecutor’s leading questions and the erroneous NIBIN testimony did not so profoundly distort the trial that injustice is obvious. See Escalante, 245 Ariz. at 141 ¶ 20.

¶40 Romero further argues that the “profoundly distorted” standard from Escalante risks creating the anomalous result that the burden becomes higher simply because more errors occurred. See id. We disagree. Where, as here, a defendant concedes that no single error deprived him of a fair trial and does not contend that he was prejudiced under prongs one or two of Escalante, it is appropriate to require a showing that the cumulative effect of the errors created “injustice [that] is obvious without the need to further consider prejudice.” Id.; see also Vargas, 249 Ariz. at 190 ¶ 14. That standard ensures that only those cumulative errors that fundamentally undermine the integrity of the trial warrant reversal.

¶41 Having applied the proper objective review to the entire record, we conclude that although the pervasive use of leading questions and the improper admission of NIBIN testimony constitute error, Romero has not demonstrated that the cumulative effect of those errors so profoundly distorted the trial that he could not possibly have received a fair trial. The court of appeals did not err in affirming his convictions and sentences.

12 STATE V. ROMERO Opinion of the Court

CONCLUSION

¶42 For the foregoing reasons, we affirm the court of appeals’ disposition but vacate its opinion. Romero’s convictions and sentences are affirmed.