State of Arizona v. Joseph Javier Romero

OPINION

¶ 1 Joseph Romero was convicted after a jury trial of second-degree murder and sentenced to a presumptive term of sixteen years. Romero argues the trial court erred when it denied his motion to dismiss the indictment due to pre-indictment delay, denied his motion to preclude testimony from the state’s firearms expert, and granted the state’s motion to preclude testimony from his proffered expert on firearms examination methodology. Romero also argues the trial court erred by entering á criminal restitution order at sentencing. For the reasons that follow, we vacate the criminal restitution order but otherwise affirm Romero’s convictions and sentences.

Factual and Procedural Background

¶2 We view the facts in the light most favorable to sustaining the jury’s verdict and resolve all reasonable inferences against Romero. State v. Haight-Gyuro, 218 Ariz. 356, ¶ 2, 186 P.3d 33, 34 (App.2008). In June 2000, S.M. was killed by two gunshot wounds to his face and back. Among other items, a cellular telephone and six .40-caliber shell casings were discovered near S.M.’s body. Nearly one month later, when Romero was stopped by police officers in an unrelated matter, he possessed a .40-caliber Glock magazine. Officers also found a .40-caliber Glock handgun without its magazine along the path Romero had travelled just prior to his encounter with the police. This firearm later would be linked to the shell casings discovered near S.M.

¶ 3 Seven years after the homicide, a “cold ease” unit examined information from the cell phone found next to S.M.’s body, which led investigators to Romero. Based on this connection, a firearms expert was asked to conduct a ballistics test of Romero’s Glock handgun. The expert fired the handgun and concluded that the indentations it made on the back of each expelled shell casing matched those on the shell casings found near S.M.’s body.

¶ 4 Romero was charged by indictment with first-degree murder. After a jury trial, he was found guilty of the lesser-ineluded offense of second-degree murder and sentenced to sixteen years’ imprisonment.1 This timely appeal followed.

Pre-indictment Delay

¶ 5 Romero argues the trial court erred by denying his motion to dismiss the charge due to pre-indictment delay based on the seven years that had elapsed between the date of S.M.’s death and when the state began investigating the ease again. We review a court’s ruling on a motion to dismiss for an abuse of discretion. State v. Medina, 190 Ariz. 418, 420, 949 P.2d 507, 509 (App. 1997).

*454¶ 6 “To establish that pre-indictment delay has denied a defendant due process, there must be a showing that the prosecution intentionally delayed proceedings to gain a tactical advantage over the defendant or to harass him, and that the defendant has actually been prejudiced by the delay.” State v. Broughton, 156 Ariz. 394, 397, 752 P.2d 483, 486 (1988). Romero does not allege and the record contains no evidence that the state intentionally delayed indicting him to obtain a tactical advantage. Rather, Romero contends the state was negligent in waiting until 2007 to investigate the cellular telephone found next to S.M.’s body. But even assuming the state had been negligent in this regard, it does not demonstrate the delay had been intentional and designed to “gain a tactical advantage” over Romero or “to harass him.” Id. Because Romero has not established this required element, he is not entitled to relief for pre-indictment delay under the test set forth in Broughton. See id.

¶ 7 Romero argues, however, that he is not required to demonstrate the state intentionally delayed the prosecution to gain a tactical advantage. He contends this requirement is the result of our supreme court’s misinterpretation of United States v. Marion, 404 U.S. 307, 92 S.Ct. 455, 30 L.Ed.2d 468 (1971), and United States v. Lovasco, 431 U.S. 783, 97 S.Ct. 2044, 52 L.Ed.2d 752 (1977). Romero appears to ask that we instead apply a balancing test similar to that adopted by some federal circuit courts. See, e.g., Howell v. Barker, 904 F.2d 889, 894-95 (4th Cir.1990); United States v. Moran, 759 F.2d 777, 782 (9th Cir.1985). But we are “bound by decisions of the Arizona Supreme Court and ha[ve] no authority to overturn or refuse to follow its decisions.” State v. Long, 207 Ariz. 140, ¶ 23, 83 P.3d 618, 623 (App.2004). Accordingly, any changes to the test for determining whether a defendant is entitled to dismissal of charges because of pre-indictment delay “would be in the exclusive purview of [the supreme court].” State v. McPherson, 228 Ariz. 557, ¶ 16, 269 P.3d 1181, 1187 (App. 2012).

¶ 8 Moreover, under either test Romero was required to demonstrate that he actually was prejudiced by the delay, which he has failed to do. See Howell, 904 F.2d at 895; Moran, 759 F.2d at 782. “To make a showing of actual and substantial prejudice, ‘it is not enough to show the mere passage of time nor to offer some suggestion of speculative harm; rather the defendant must present concrete evidence showing material harm.’ ” State v. Dunlap, 187 Ariz. 441, 450, 930 P.2d 518, 527 (App.1996), quoting United States v. Anagnostou, 974 F.2d 939, 942 (7th Cir.1992).

¶ 9 Romero argues his ability to mount a defense was prejudiced by the passage of time because potential witnesses had died, witnesses’ memories had faded, and he was not on notice to preserve evidence showing his whereabouts at the time of the murder. Romero did not identify unavailable witnesses or possible testimony. Similarly, he has not specified what evidence he could have gathered with respect to ownership of the handgun attributed to him that was not already in the law enforcement record. Thus, Romero has not presented concrete evidence that he was actually and substantially prejudiced by the delay. See Broughton, 156 Ariz. at 397, 752 P.2d at 486. Based on the record before us, the trial court did not err by refusing Romero’s request to dismiss the charge.

Rule 702

¶ 10 Romero next raises two arguments related to the admissibility of expert testimony under Rule 702, Ariz. R. Evid. First, he contends the trial court erred by denying his motion to preclude the testimony of the state’s firearms examiner, Frank Powell, on the ground the examination was not the product of reliable principles and methods. Romero also asserts the court erred in precluding his experimental psychologist expert, Ralph Haber, from testifying at trial about scientific criticisms of all firearm identifications. We review a trial court’s decisions on the admissibility of expert testimony for an abuse of discretion. State v. Davolt, 207 Ariz. 191, ¶ 69, 84 P.3d 456, 475 (2004).

¶ 11 Effective January 1, 2012, Arizona adopted the language of Rule 702, Fed. *455R.Evid., which reflects the principles set forth in Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579, 113 S.Ct. 2786, 125 L.Ed.2d 469 (1993). See Ariz. R. Evid. 702 cmt. to 2012 amend.; State v. Perez, 233 Ariz. 38, ¶ 16, 308 P.3d 1189, 1193 (App.2013). Under Rule 702, the trial court is to serve as a “gatekeeper!]” that admits testimony it initially finds reliable, permitting the jury to weigh what the court has already determined to be “reliable, expert testimony.” Ariz. R. Evid. 702 cmt. to 2012 amend.; see also Perez, 233 Ariz. 38, ¶ 16, 308 P.3d at 1193. This “gatekeeper” function applies not only to scientific evidence, but “also to testimony based on ‘technical’ and ‘other specialized’ knowledge.” Kumho Tire Co. v. Carmichael, 526 U.S. 137, 141, 119 S.Ct. 1167, 143 L.Ed.2d 238 (1999). Specifically, Rule 702, Ariz. R. Evid., provides:

A witness who is qualified as an expert by knowledge, skill, experience, training, or education may testify in the form of an opinion or otherwise if:

(a) the expert’s scientific, technical, or other specialized knowledge will help the trier of fact to understand the evidence or to determine a fact in issue;

(b) the testimony is based on sufficient facts or data;

(c) the testimony is the product of reliable principles and methods; and

(d) the expert has reliably applied the principles and methods to the facts of the case.

“Daubert offers additional ‘non-exclusive factors for determining whether scientific evidence is admissible,’ including empirical testing, peer review, error rate, the existence of standards and controls, and the degree to which the theory and technique is generally accepted by a relevant scientific community.” Sandretto v. Payson Healthcare Mgmt., Inc., 234 Ariz. 351, ¶ 12, 322 P.3d 168, 173 (App. 2014), quoting Ariz. State Hosp./Ariz. Cmty. Protection & Treatment Ctr. v. Klein, 231 Ariz. 467, ¶27, 296 P.3d 1003, 1009 (App. 2013); see also Daubert, 509 U.S. at 593-94, 113 S.Ct. 2786.

Admission of Toolmark Analysis for Firearm Identification

¶ 12 Romero moved to preclude Powell’s testimony, asserting the field of firearms identification lacked the reliability required by Daubert and Rule 702. Although he did not challenge Powell’s expert qualifications, he argued that the field is not a science because the theory of unique markings from individual firearms cannot be tested under the scientific method. He also attacked the field’s subjective methods, the structure and functioning of its research literature, and how examiner error rates are calculated. Additionally, Romero relied on Dr. Haber to convey these general arguments, as well as to expand upon criticisms from the National Academy of Science review of forensic sciences in 2009.2 The trial court conducted a Rule 702 evidentiary hearing, and reviewed Powell’s testimony from the first trial in which he described his qualifications as well as the methodology he used to match spent casings to a specific firearm. The court found the firearms identification evidence “reliable and admissible under Arizona’s newly adopted Daubert standard.” On appeal, Romero raises the same arguments he did below.

¶ 13 Before Rule 702 changed in 2012, our supreme court determined that firearms identification testimony was admissible under the previous standard set forth in Frye v. United States, 293 F. 1013, 1014 (D.C.Cir. 1923). See State v. Miller, 234 Ariz. 31, ¶¶ 28-31, 316 P.3d 1219, 1229 (2013); State v. Macumber, 112 Ariz. 569, 570-71, 544 P.2d 1084, 1085-86 (1976). Although Arizona courts have yet to determine whether firearms identification is sufficiently reliable for admission under amended Rule 702, we look *456to federal decisions interpreting Federal Rule 702 for guidance. See State v. Green, 200 Ariz. 496, ¶ 10, 29 P.3d 271, 273 (2001) (“When interpreting an evidentiary rule that predominately echoes its federal counterpart, we often look to the latter for guidance.”); Ariz. R. Evid. 702 cmt. to 2012 amend. (“The 2012 amendment of Rule 702 adopts Federal Rule of Evidence 702, as restyled.”).

¶ 14 Several federal district courts have held that firearms identification testimony is sufficiently reliable under Daubert and Federal Rule 702. See, e.g., United States v. Willock, 696 F.Supp.2d 536, 571-72 (D.Md. 2010); United States v. Taylor, 663 F.Supp.2d 1170, 1179-80 (D.N.M.2009); United States v. Monteiro, 407 F.Supp.2d 351, 354-55 (D.Mass.2006). In Monteiro, after a six-day evidentiary hearing, the court held that “the underlying scientific principle behind firearm identification — that firearms transfer unique toolmarks to spent cartridge cases — is valid under Daubert." 407 F.Supp.2d at 355. Similarly, in Willock, the court determined that the standards governing toolmark examination are sufficient to permit a qualified expert’s testimony to assist jurors in determining whether bullets or cartridges have been fired from a particular firearm. 696 F.Supp.2d at 571-72.

¶ 15 At the first trial, Powell testified about his background, training, and experience in firearms identification. He is a member of the Association of Firearm and Tool-mark Examiners that publishes a quarterly journal. He also testified that he is required to complete an annual proficiency exam and that studies indicate an error rate around one percent for proficiency tests given to firearms examiners. Further, he indicated that the methodology he used to analyze the shell casings is accepted by his scientific community as valid, and that a second examiner was required to review his work and agree with his conclusion before it was reported.

¶ 16 We find the reasoning in Monteiro and Willock persuasive and likewise conclude that the methodology governing firearms identification is sufficiently reliable, under Daubert and Arizona Rule 702, to permit a qualified expert to provide in-court technical testimony.3 See Daubert, 509 U.S. at 593-94, 113 S.Ct. 2786. First, Romero failed to develop an argument that changes in firearms identification methods call into question Arizona case law admitting such testimony under Frye. Nor does he identify a reason Arizona’s adoption of the Daubert standard would justify a different result. See Miller, 234 Ariz. 31, ¶¶ 28-31, 316 P.3d at 1229; cf. Favela, 323 P.3d 716, ¶¶ 6, 9, 323 P.3d at 718, 719. Accordingly, the trial court did not abuse its discretion in denying Romero’s motion to preclude Powell’s testimony. See Willock, 696 F.Supp.2d at 571-72; Monteiro, 407 F.Supp.2d at 354-55; cf. Favela, 323 P.3d 716, ¶¶6, 9, 323 P.3d at 718, 719.

¶ 17 Romero further argues that even if the trial court properly allowed Powell to testify, the court erred by “failing to limit his testimony regarding the certainty of his conclusions.” He appears to rely on Monteiro in support of this argument. 407 F.Supp.2d at 355. Although the court in Monteiro held that the underlying scientific principle behind firearms identification is valid under Dau-bert, it determined that “the subjective nature of the matching analysis,” meant “a firearms examiner must be qualified through training, experience, and/or proficiency testing to provide expert testimony.” Id. The court further concluded that a firearms expert may give an opinion of a match “to a reasonable degree of certainty in the ballistics field,” but may not testify that there is a match “to an exact statistical certainty.” Id.

¶ 18 But Monteiro is distinguishable. Here, unlike the examiners in Monteiro, who testified essentially that they could be 100 percent sure of a match, Powell testified that there was a match to “a reasonable degree of scientific certainty.” See 407 F.Supp.2d at 372. Moreover, in Ruiz-Troche v. Pepsi-Cola of Puerto Rico Bottling Company, upon *457which the Monteiro court relied in support of its holding, the court of appeals approved allowing an accident reconstruction expert to testify to a reasonable degree of scientific certainty. Ruiz-Troche v. Pepsi-Cola of Puerto Rico Bottling Co., 161 F.3d 77, 82 (1st Cir.1998); see also Monteiro, 407 F.Supp.2d at 372. Accordingly, the trial court did not err by permitting Powell to testify to a reasonable degree of scientific certainty.

Preclusion of Expert Testimony Criticizing Firearms Identification

¶ 19 Romero next argues the trial court erred in precluding his psychology expert from testifying at trial about criticisms of firearms identification. The court found Haber not qualified to challenge or rebut the testimony, foundation, or opinions of Powell. The court also found that Romero sought to introduce Haber’s testimony to conduct what amounted to a second Daubert hearing before the jury.4 Accordingly, the court granted the state’s motion to preclude Haber’s proposed testimony.

¶20 Unlike most Rule 702 issues that courts have faced in the last two decades, the question of whether an expert is qualified to express a particular opinion is largely unaffected by Daubert, its progeny, or the changes to Rule 702. Almost a century ago it was black letter law that a person offering an expert opinion must have the requisite qualifications on the particular matter.5 1 Wigmore on Evidence § 560 (2d ed.1923); see also Gaston v. Hunter, 121 Ariz. 33, 51, 588 P.2d 326, 344 (App.1978), citing Myers v. Cessna Aircraft Corp., 275 Or. 501, 553 P.2d 355, 370 (1976). It also was recognized that expertise is specific, and experience in one area does not confer expertise in a related area. Myers, 553 P.2d at 370-71. Stated differently: no expert is competent to express an opinion on every subject. Wigmore, supra, § 555.

¶ 21 As the proponent of expert testimony, Romero had the burden of demonstrating Haber’s qualifications on the particular issues. Sandretto, 234 Ariz. 351, ¶ 15, 322 P.3d at 174. The trial court has broad discretion in admitting or excluding expert testimony, and we will not reverse its ruling “unless there is a clear abuse of discretion.” Gemstar Ltd. v. Ernst & Young, 185 Ariz. 493, 505, 917 P.2d 222, 234 (1996); see also Salem v. U.S. Lines Co., 370 U.S. 31, 35, 82 S.Ct. 1119, 8 L.Ed.2d 313 (1962) (trial judge has broad discretion in exclusion of expert testimony; ruling will be sustained unless manifestly erroneous). Moreover, the trial court determines “whether the expertise of the witness is applicable to the subject.” Englehart v. Jeep Corp., 122 Ariz. 256, 258, 594 P.2d 510, 512 (1979).

¶ 22 Romero’s proof of Haber’s qualifications was limited to general background statements in advance of his testimony about firearms identification. Romero did not proffer a curriculum vitae, bibliography of published articles, or other record of Haber’s experiences and training. Haber’s graduate education and professional background are predominantly in the field of experimental psychology. His professional experience included psychology-related work in academia as well as consulting in the area of eyewitness testimony. He eventually branched out to fingerprint analysis after he underwent fingerprint examiner training. Haber’s firearms identification experience consisted solely of his reviewing the relevant literature and writing a “chapter in the California Bar Association’s publication on evidence in the criminal courts on firearms and handgun identification.” This was the first time he *458had been retained as a proposed expert in firearms identification. Haber admitted he had “no idea what an examiner does when he carries out an examination.”

¶ 23 Romero challenges the description of Haber as only a psychologist. He posits him as an expert in the scientific field of experimental design. Haber’s self-description was not so broad. For instance, he taught for six years “as an assistant professor in psychology and primarily in experimental psychology and statistics and experimental design.” At subsequent academic positions as an experimental psychologist, he also taught experimental design. He explained that he has been a peer and grant reviewer “on a variety of experimental topics where I review them, analyze them both in terms of the appropriate experimental designs, the way the experiment was carried out, the conclusions reached, the interpretations and the statistical methods that were used.” One such item involved handgun identification, but Haber provided no details about the grant application he reviewed. He also has done review work for several national academies and twenty different journals, although apparently none involving journals read by firearms and toolmark analysts.

¶ 24 Accepting for the purpose of addressing Romero’s argument that Haber has expertise in experimental design, we address whether that background qualifies him to testify as an expert in firearms identification, where he has “studied this literature for three or four years,” but has no practical experience. First, we note that experimental design is not a separate field of study, but generally describes various empirical models to study measurable phenomena. It is a critical component of the scientific method. Erica Beecher-Monas, The Heuristics of Intellectual Due Process: A Primer for Triers of Science, 75 N.Y.U. L.Rev. 1563, 1578 (2000) (“Science Primer”) (science consists of assumptions about the way the world works, coupled with canons of experimental design and theoretical exemplars to address problems and explanations). Experimental design is employed in virtually any area susceptible to statistical analysis, such as the social, biological, and physical sciences. See generally, David H. Kaye & David A Freedman, Reference Guide on Statistics, in Reference Manual on Scientific Evidence 90-97 (2d ed.2000). It is not a one-size-fits-all approach. The application of experimental design principles “differ[s] widely from field to field.” Science Primer, 75 N.Y.U. L.Rev. at 1629. A classic text on experimental design cautions that the researcher cannot casually transfer design principles across fields. D.R. Cox, Planning of Experiments at vi (1958) (“[T]he practical importance of different parts of [experimental design] varies greatly between different applied fields.”). The issue is whether Haber could apply his knowledge of experimental design to firearms identification.

¶ 25 Assuming that Haber described all of his relevant experience, training, and knowledge, the omissions in his ability to apply theoretical design knowledge to firearms identification are numerous. Before this case, Haber never conducted a toolmark analysis, never attempted to identify different firearms, and never conducted research on firearms identification. He has no experience in any physical sciences on which tool-mark analysis rests, such as ballistics, metallurgy, or physics. Despite his general study of the firearms identification literature, Haber could not describe the methods or protocols of a toolmark analyst. Had Haber demonstrated relevant experience or knowledge in one or more of these areas, the issue of his qualifications would have been moot or at least a much closer question. See, e.g., Kum-ho Tire Co., 526 U.S. at 153, 119 S.Ct. 1167 (tire expert qualified based on master’s degree in mechanical engineering, manufacturing experience, and tire failure analysis); Lo-gerquist v. McVey, 196 Ariz. 470, ¶¶ 15, 32, 1 P.3d 113, 117, 124 (2000) (psychiatrist qualified to testify about amnesia for traumatic experiences based on education and clinical experience); Lohmeier v. Hammer, 214 Ariz. 57, ¶¶ 3, 29, 148 P.3d 101, 104, 108-09 (App. 2006) (biomechanical engineer qualified to testify about forces involved in vehicle collision based on education, industry experience, and research).

¶ 26 Romero and our specially concurring colleague draw a different conclusion about *459Haber’s qualifications, principally relying on his general experience in a forensic consulting firm and experimental background. That consulting is primarily in the area of eyewitness identification and fingerprint analysis. The first area is not surprising because eyewitness identification experts frequently have psychology backgrounds due to the interplay between perception and memory. See, e.g., United, States v. Moore, 786 F.2d 1308, 1312 (5th Cir.1986) (noting conclusions of psychological studies serve to “‘explode common myths about an individual’s capacity for perception’”), quoting United States v. Smith, 736 F.2d 1103, 1105 (6th Cir.1984); State v. Chapple, 135 Ariz. 281, 291, 660 P.2d 1208, 1218 (1983) (expert on eyewitness identification a professor specializing in area of experimental and clinical psychology dealing with perception, memory retention and recall). To qualify as a fingerprint expert, Haber undertook professional training, which was the “equivalent to what a fingerprint examiner would take to be employed in a crime laboratory.” Haber offered no such specialized training or experience with firearms identification. Equally important, it is not the role of this court to re-weigh the evidence proffered to qualify a person as an expert. Cauble v. Osselaer, 150 Ariz. 256, 258, 722 P.2d 983, 985 (App.1986) (abuse of discretion standard requires appellate court to uphold trial court’s determination unless unsupported by evidence or absolutely contrary to uneontradicted and unconflicting evidence).

¶27 Romero alternatively offered to limit Haber’s testimony to a general critique of the field, specifically avoiding anything Powell did. But this position is implicitly grounded on the assumption that a person with experimental design knowledge applicable to one field can apply the same principles to an entirely different field. Science does not support such an assumption and neither does the law.

¶ 28 For instance, in Myers, 553 P.2d at 370, the expert was proffered to opine about the probable cause of an airplane crash based on his experience in “technical, engineering aspects of accident investigations.” He was a member of the Society of Ah’ Safety Investigators and had flown in the Air Force. Id. The trial and appellate courts found specific absences more significant than his admittedly pertinent experience in limited areas. The expert “had no formal training as an accident investigator, had never attended a seminar on that subject, was not an aeronautical engineer, was not accredited as an instrument flight pilot, did not have a current pilot’s license, and had never flown a light aircraft similar to the one involved in this crash.” Id.

¶29 Similarly, in United States v. Paul, 175 F.3d 906, 912 (11th Cir.1999), the proponent sought to use an evidence law professor who had co-written an article critical of forensic document examiners to rebut the opinion of an expert in that field. Despite his obvious expertise in evidence and having reviewed the literature about document examiners, his lack of knowledge about handwriting analysis precluded his opinions about the examination conducted or the field itself. Id. at 911-12. Simply stated, even a person with expertise in one area must demonstrate sufficient knowledge or experience in the pertinent area to qualify as an expert in the particular case regarding a specific opinion.

¶ 30 Romero indirectly seeks to counter Paul by relying on United States v. Velasquez, 64 F.3d 844, 848 (3d Cir.1995), in which the appellate court concluded the trial court erred in precluding the same law professor from criticizing handwriting standards. The appellate court did not explicitly address the professor’s qualifications. Id. Instead, it “point[ed] to the Professor’s eight years of self-directed research on handwriting analysis and his co-authorship of a law review article on the subject.” Id. at 851. It also noted that the government’s expert was aware of the professor’s scholarship, the professor’s criticisms were similar to critiques that had been subject to peer review, and the professor’s opinions were specific to the methods used by the government’s expert. Id. at 851-52. Additionally, the professor had read “nearly all of the literature on the subject,” and he had been named an American Bar Association Fellow for creating a testing mechanism to certify handwriting analysts and to validate the accuracy of their identifications. Id. at 847 n. 4. On first look, the Eleventh and Third circuits appear to be *460in conflict because they came to contrary conclusions regarding the same law professor arguably offered for the same purpose. The differences, however, illustrate that the proponent in Velasquez made a considerably more detailed record concerning the professor’s actual experience and work in handwriting analysis.6 Whether the circuits still would have disagreed about the professor’s qualification to testify had the pi’oponents made identical proofs of expertise is unknowable, but from the perspective of reported qualifications, there is no conflict in the decisions.

¶31 Romero also relies on State v. Lehr, 201 Ariz. 509, 38 P.3d 1172 (2002), for intertwined propositions that expert testimony generally supporting a defense argument is sufficient under Rule 702 and, in any event, preclusion would violate a defendant’s Sixth Amendment right to present a defense.7 Similarly, our colleague extends the argument with reliance on a more recent expert witness ease, State v. Salazar-Mercado, 234 Ariz. 590, 325 P.3d 996 (2014). We discuss them together. First, in both eases the Arizona Supreme Court noted that the qualifications of the defense experts were not challenged or in doubt. Lehr, 201 Ariz. 509, n. 12, 38 P.3d at 1181 n. 12; Salazar-Mercado, 234 Ariz. 590, ¶ 12, 325 P.3d at 999. Second, while there are few rights “more fundamental than that of an accused to present witnesses in his own defense,” the exercise of the right must comply with evidence rules designed to ensure fairness and reliability. Chambers v. Mississippi 410 U.S. 284, 302, 93 S.Ct. 1038, 35 L.Ed.2d 297 (1973). As the Eleventh Circuit Court of Appeals explained, “a court may constitutionally enforce evidentiary rules to limit the evidence an accused (or for that matter any party) may present in order to ensure that only reliable opinion evidence is admitted at trial.” United States v. Frazier, 387 F.3d 1244, 1272 (11th Cir.2004) (right to put on meaningful defense did not include unfettered, unreviewable opportunity to present expert testimony inadmissible under Rule 702).

¶ 32 Finally, we observe that Romero was not deprived of the right to challenge Powell’s testimony using the same materials Haber referenced. There was a spirited cross-examination of Powell about the 2009 NAS Report and several studies criticizing particular aspects of firearms identification. The jury heard, by quotation and paraphrase, the essence of Haber’s criticisms because much of his analysis was derived from the NAS Report. We conclude the preclusion of Haber’s testimony on the ground he lacked knowledge or practical experience in toolmark analysis did not violate Romero’s right to present a defense.

Criminal Restitution Order

¶33 Romero lastly argues, and the state concedes, that the trial court erred in entering a criminal restitution order (CRO) at sentencing. We agree and find fundamental error associated with the CRO. See State v. Lopez, 231 Ariz. 561, ¶ 2, 298 P.3d 909, 910 (App.2013). In the sentencing minute entry, the trial court ordered “all fines, fees, assessments and/or restitution are reduced to a Criminal Restitution Order, with no interest, penalties or collection fees to accrue while the defendant is in the Department of Corrections.” The trial court’s imposition of the CRO before the expiration of Romero’s sentence “ ‘constitute^] an illegal sentence, *461which is necessarily fundamental, reversible error.’ ” Lopez, 231 Ariz. 561, ¶'2, 298 P.3d at 910, quoting State v. Lewandowski, 220 Ariz. 531, ¶ 15, 207 P.3d 784, 789 (App.2009). This remains true even though the court ordered the imposition of interest be delayed until after Romero’s release. See id. ¶5.

Disposition

¶ 34 For the foregoing reasons, we vacate the CRO but otherwise affirm Romero’s convictions and sentences.