delivered the Opinion of the Court.
Wilkerson, who is the defendant in a prosecution for attempted murder, petitioned for relief pursuant to C.A.R. 21, from a pre-trial order of the district court excluding testimony offered by her expert witness. The district court barred the witness from giving his opinion, pursuant to CRE 702, that it was more likely than not that Wilkerson shot the victim accidentally. This court issued its rule to show cause why the order should not be vacated, and the People responded. We conclude that the district court did not abuse its discretion in excluding the proffered opinion, and we therefore discharge the rule.
I.
Victoria Wilkerson was charged with attempted first degree murder, first degree assault, menacing, conspiracy and attempt to commit aggravated robbery, and committing a crime of violence, for her involvement in events occurring on March 3, 2003. According to police reports, she shot the victim with a handgun during a physical melee involving no fewer than four people. Wilkerson later told the police that the handgun she was holding “came off safety,” and when she pointed it at the victim, it fired on its own.
Wilkerson endorsed an expert witness in the field of ergonomics, or human factors, to testify about accidental shootings, and the People objected. At a hearing prior to trial, the witness testified that ergonomics is a scientific discipline that analyzes the design of “systems” to make them safer and more useable, and to avoid human error. In this case, the “system” about which he proposed to testify was a person holding a handgun.
The witness also testified about his qualifications to give an expert opinion in this field. He indicated that he had co-authored two articles based on ergonomic handgun design, handgun training, and deficiencies in handgun training. More specifically, the articles discussed issues relating to trigger design, trigger forces, safeties, grip design, and the impact on people using handguns of such things as stress and lack of prior handgun experience. He also testified that he had studied analyses of accidental shootings by police.
Although the witness indicated that he had actually test-fired the guns involved in the other eases in which he had testified, he did not have access to, or any other information about, the handgun used by Wilkerson in this ease.1 Similarly, the witness conceded that *876he had never tested Wilkerson for specific physical characteristics, such as her reaction time, strength, or hand size, even though he considered those factors important, and that his opinion was based instead entirely on what she had related to him about the weapon and the events. On cross-examination, the witness agreed that his opinion was dependent upon the truthfulness of Wilkerson’s statements about her lack of training with handguns, the stressfulness of the situation, and her desire to protect her husband and keep the situation from getting out of hand.
Nevertheless, the witness offered his expert opinion that it was “more likely than not” that the shooting was accidental. In response to questioning, he indicated that by that opinion he meant there was more than a 50% probability that it was accidental. In arguing for the admissibility of this expert testimony, defense counsel characterized it as being an opinion that under the conditions described by the witness, an accidental shooting could happen 51% of the time. Counsel expressly asked the court to permit the witness to testify and give his opinion about what he thought happened in this case, but if the court decided not permit the witness to give his opinion regarding the guilt or innocence of the defendant then, in the alternative, it should permit “a defense that says that the best minds that study this sort of thing and study stress, and study systems and human factors and study handguns, and have written articles about this sort of thing say that, yes, accidental shootings can happen.”
In its order, the district court characterized the proffer upon which it was ruling as an offer of expert opinion “that it is more likely than not that Ms. Wilkerson fired the semi-automatic weapon accidentally based on persons’ physiological response to stress, tunnel vision response to stress and inexperience with weapons.” The court specifically found that the witness was “eminently qualified in the general area of human factors/ergonomics,” and that he had studied and presented papers concerning accidental shootings. With regard to the specific proffer, however, the court was critical of the lack of any extant body of literature or knowledge concerning standards of measurement and error rates. Noting particularly that the witness was relying almost exclusively on the statements of the defendant, and that there was no specific weapon in evidence from which characteristics like the configuration of the safety, or the trigger mechanism and pull pressure could be taken, the court denied the defense proffer.
The district court did not address the defendant’s alternate request in its order, and the defendant has not presented us with any indication that she requested clarification or a broader ruling. In her petition, the defendant sought disapproval of the district court’s ruling excluding the witness’ expert opinion “based upon his training, experience and knowledge of the allegations in this case that the defendant accidentally fired a handgun,” and nothing more.
II.
In People v. Shreck, 22 P.3d 68, 78-79 (Colo.2001), we made clear that CRE 702 represents the appropriate standard for determining the admissibility of scientific evidence in this jurisdiction, and we held that the focus of the trial court’s inquiry should be on the reliability and relevance of such evidence. We held that the trial court’s inquiry should be broad in nature and consider the totality of the circumstances of each specific case. We emphasized that the factors mentioned by the Supreme Court cases or other courts in specific cases may or may not be pertinent to any particular CRE 702 inquiry; and in light of this liberal standard, the trial court should also apply its discretionary authority under CRE 403 to ensure that the probative value of the evidence is not substantially outweighed by unfair prejudice.
Even though a general area of scientific knowledge is determined to be reliable, if the results of a scientific test or *877comparison are not self-evident, the test itself lacks relevance unless there is also reliable expert interpretation of its results. People v. Lee, 18 P.3d 192, 197 (Colo.2001); see also Fishback v. People, 851 P.2d 884, 893 (Colo.1993). In Fishback, we held that the method used to calculate the probability of a random match between DNA recovered from a crime scene and that of the defendant was also subject to the general acceptance test, just as the underlying scientific methods used for DNA profiling. Id. at 893. Although Shreck later replaced the Frye2 general acceptance test with the CRE 702 analysis, Shreck, 22 P.3d at 77, it nevertheless remains true that “a declared [DNA] match, unaccompanied by its statistical significance, is essentially meaningless,” Fishback, 851 P.2d at 893 n. 18. Even where the underlying scientific evidence would be sufficiently reliable under a CRE 702 analysis, that in no way demonstrates the reliability of the method of reaching statistical or numerical conclusions related to the underlying evidence. A CRE 702 analysis is independently necessary to show that statistical or numerical results are also relevant and reliable. See Lee, 18 P.3d at 197 (citing Fishback, 851 P.2d at 893 & n. 18); see also Shreck, 22 P.3d at 76-78.
III.
Understood as narrowly as it was by the district court, the defendant’s proffered expert opinion was virtually without any support in the record whatsoever, much less support sufficient to demonstrate the reliability of a numeric calculation of the probability that the shooting in this case was accidental. Although the witness may have intended nothing more than an expression of personal confidence (based on his experience studying handgun usage) that this particular shooting was probably an accident, by expressing his expert opinion as a numerical conclusion, he could only be understood as offering a scientifically quantifiable probability statement about the likelihood of a particular occurrence.
An expert opinion that a shooting was probably accidental necessarily implies that it also was probably not purposeful, or intentional. To the extent that it could be understood as an opinion directly about the defendant’s state of mind, it was clearly beyond the scope of ergonomics, and the defendant’s witness did not purport to have any expertise concerning the defendant’s psychological condition or will. To the extent that it could be understood as an opinion based on circumstantial evidence of the shooting, or even a hypothetical opinion based on circumstances and conditions similar to those described by the defendant, such a numeric probability statement was nevertheless completely without empirical or methodological justification in the record.
Given the specific terms in which the opinion was offered in this case, and especially the danger that the expert would be understood simply as vouching for the defendant’s account of events, it would be difficult to find an abuse of discretion in rejecting such an opinion as more confusing, or even misleading, than helpful. See CRE 403. In any event, however, the defense offered no foundation for a quantification of the likelihood that a shooting occurred, or would occur, accidentally, even in the absence of any intent by the shooter.
Until cross-examination, the witness made no attempt to quantify the probability that the shooting was accidental and apparently had no mathematical or scientific technique in mind. While he testified about ergonomic research into shootings, conducted in apparent attempts to discover the extent to which accidental shootings occur and to isolate contributing factors and hopefully eliminate them and avoid accidental shootings altogether, the witness never referenced a scientific methodology for quantifying, or a standard for measuring, the likelihood that a particular shooting was accidental, much less error rates for such testing. To the extent that he suggested the existence of measurement techniques at all, he referred to measurements like reaction time, strength, hand size, safety configuration, and trigger pull pressure, which he admittedly failed to make in this ease, instead relying on less measurable factors like stressfulness and lack of experi-*878enee. In perhaps his only reference to a quantifiable factor, the witness presumed that the weapon used was a “.22 or .23 caliber semiautomatic,” with a relatively light trigger pull, despite all indications from the police reports that a .38 caliber shell casing was found at the scene of the shooting.
Given the preciseness of the district court’s order, rejecting an expert conclusion expressed in terms of quantifiable, scientific test results, the court’s emphasis on the expert’s failure to test a specific weapon and the absence of peer reviewed literature concerning standards for measurement and error rates did not evidence a misunderstanding of the flexible test we described in Shreck. Rather than excluding expert opinion merely because it was incapable of quantification, the district court was faced with an opinion expressed in numeric terms, and it appears to have considered precisely those factors most applicable to the admissibility of such a numeric test result.
Although the district court did not address defense counsel’s alternate request to permit testimony concerning factors relevant to the question of accidental shootings generally, it specifically found the witness to be eminently qualified in the general area of human factors/ergonomics and did not exclude as unreliable all testimony concerning application of that field to accidental shootings. In light of the district court’s failure to address more than the specific opinion that it was more likely than not that the shooting in this case was accidental, we offer no opinion concerning the reliability or relevance of any other proffered testimony of the defense expert.
IV.
In light of the defendant’s specific proffer and the district court’s narrow holding, we find that the district court did not abuse its discretion in excluding the expert’s proffered opinion that it was more likely than not that the defendant shot the victim accidentally. We therefore discharge our rule.
Justice BENDER dissents, and Justice MARTINEZ joins in the dissent.. In this original proceeding, we have not been provided with any information about the circum*876stances of the shooting or arrest of the defendant, nor any explanation why the defense expert did not have access to the weapon or even know the make and model of weapon that was actually used.
. Frye v. United States, 293 F. 1013 (D.C.Cir. 1923).