OPINION
{1} In State v. Gardner, 1998-NMCA-160, 126 N.M. 125, 967 P.2d 465, we held that compliance with .the number of minutes of continuous observation stated in the Department of Health Regulation was a prerequisite to admissibility of breathalyzer test results. In this case, the major issue we determine is whether the evidence supported the trial court’s ruling that an eight-minute wait before administering the second test was all that was necessary following a twenty-minute wait before the first test when the officer testified that Defendant did not do any of the things that trigger a second twenty-minute wait. We hold that substantial evidence supported the trial court’s ruling.
{2} Defendant appeals her conviction of aggravated driving while intoxicated (DWI). She raises three issues: (1) the court erred in allowing the breathalyzer test results into evidence, (2) the court erred in allowing testimony about horizontal gaze nystagmus (HGN) tests, and (3) the aggravated portion of the DWI charge should be dismissed because the breathalyzer test was taken more than an hour and a half after Defendant was driving and registered .16, the exact reading required to raise DWI to the aggravated level. The State concedes that, under the circumstances of this case, in which the officer testified that there was no way of knowing what Defendant’s blood alcohol content was at the time of driving, a rational juror could not find beyond a reasonable doubt the facts necessary to raise the DWI to an aggravated level. We agree and therefore reverse and remand for entry of judgment for simple DWI unless one of Defendant’s other issues results in reversal. See State v. Cavanaugh, 116 N.M. 826, 829-30, 867 P.2d 1208, 1211-12 (Ct.App.1993) (holding that blood alcohol content almost 30% over the limit, together -with testimony of intoxication at the time of driving, was sufficient to convict, contrasting a case in which the blood alcohol content was just barely over the limit, which was insufficient to convict). We hold that Defendant’s other issues do not require reversal.
FACTS
{3} Defendant was involved in an accident in which another vehicle struck hers as she was making a left-hand turn. Because the officer dispatched to the scene was initially unable to tell who was the driver (Defendant or her passenger) and because Defendant smelled like alcohol, the officer gave her an HGN test. This test gauges an individual’s ability to track a moving object with his or her eyes. Generally, a bouncing of the eye during the HGN test indicates a level of intoxication. During voir dire of the officer, he explained that if the test is performed properly, failing the HGN test predicts with 77% accuracy that the subject has a blood alcohol content of .10 or greater. Yet, when confronted with a photocopy of the training manual that the officer used when learning how to give HGN tests, he admitted that he used improper procedure on virtually every aspect of the test. Specifically, (1) he was looking for smooth tracking of the eyes after, rather than before, the test;. (2) he checked for all three of the required clues during the same pass of the object before the subject’s eyes, and he checked for these clues in two total passes, rather than checking for each of the three clues during two separate passes, for a total of six passes; (3) when he checked for maximum deviation, he held the object for two or three seconds, rather than the required four; and (4) he never spent the required four seconds getting to the 45-de-gree" point. The trial court ruled that any improper procedure went to the weight, not the admissibility, of the HGN evidence. The officer testified that Defendant failed the *714test, and his testimony in front of the jury was similar to his testimony on voir dire.
{4} Based on the failure of the HGN test, Defendant was arrested and taken to the station for a breathalyzer test. After the required twenty-minute wait, Defendant blew into the breathalyzer machine, and it registered .16. A second test was rejected due to a mouth alcohol warning; there was no readout from this test. The officer then reinput Defendant’s information into the machine and after eight minutes took another two readings. Both of these registered .16 also. The officer testified that the mouth alcohol reading could be saliva or a burp. He also testified that he did not see Defendant either regurgitate or burp, although he admitted he could have missed a small burp. He finally testified that because the machine accepted the second and third sample without rejecting them for mouth alcohol, he believed that the residual mouth alcohol was gone. The court ultimately admitted the breathalyzer test results because the officer did not see Defendant regurgitate or burp.
DISCUSSION
1. Waiting Period
{5} Department of Health Regulation 12.1.1, on which Defendant relies, states:
Two breath samples shall be collected and/or analyzed by certified Operators or Key Operators only, and shall be end expiratory in composition. Breath shall be collected only after the subject has been under continuous observation for at least 20 minutes prior to collection of the first breath sample. If during this time the subject regurgitates or introduces any foreign substance suspected of containing alcohol into his mouth or nose, another 20 minutes observation period must be initiated. The two breath samples shall be taken not more than 15 minutes apart. If the difference in the results of the two samples exceeds 0.02 grams per 210 liters (BrAC), a third sample of breath or blood shall be collected and analyzed. If the subject declines or is physically incapable of consent for the second and/or third samples, it shall be permissible to collect and/or analyze fewer samples.
7 NMAC 33.2.12.12.2.1 (emphasis added). Defendant argues that this regulation was violated and, for the reasons outlined in Gardner, holding that the twenty-minute waiting period could not be satisfied by a fifteen-minute observation, the trial court erred in admitting the breathalyzer test results into evidence. We disagree for several reasons.
{6} First, it is undisputed that the first twenty-minute waiting period was satisfied, and there was a valid .16 reading. Although the factual basis for the last sentence of the regulation is not present in this case, the last sentence indicates that, under some circumstances, it is permissible to rely on only one sample.
{7} Second, the sentence Defendant contends was violated is the highlighted one. By its terms (“during this time”), it appears to apply only to the first twenty-minute period. As just noted, the first twenty-minute period was satisfied.
{8} Third, the highlighted sentence applies only when “the subject regurgitates or introduces any foreign substance” into the mouth or nose. In this case, there was substantial evidence upon which the trial court could have found that the subject neither regurgitated nor introduced anything foreign into her mouth or nose. The officer testified that he did not see it happen, and the trial court’s express rationale for admitting the evidence was the officer’s testimony to that effect. When a court makes findings preliminary to admissibility, this Court will not reverse those findings if they are supported by substantial evidence. See Boone v. State, 105 N.M. 223, 227, 731 P.2d 366, 370 (1986). When reviewing for substantial evidence, this Court reviews tbe evidence in the light most favorable to the ruling below. See id. Under that standard of review, and considering the officer’s testimony, we cannot say that the trial court erred in admitting the breathalyzer test results.
{9} Fourth, the officer also testified that once the machine accepted the sample and registered a reading, any residual mouth alcohol was gone. Part of the State’s argument below was that the machine would abort if mouth alcohol were present; having *715received a reading for. the second test, the State argued that second sample was necessarily proper. The officer’s testimony and the State’s argument are similar to the testimony and argument considered in a case from a sister jurisdiction. See Williams v. State, 884 P.2d 167, 173-74 (Alaska Ct.App.1994). The Williams court considered the. admissibility of breathalyzer test results where the officer performed a second test after the first test was aborted due to mouth alcohol without waiting the requisite time period. See id. at 174. The court agreed with the officer’s testimony that if the subject’s mouth alcohol had not disappeared by the time of the next attempt, the breathalyzer machine would have once again aborted the test. See id. The Williams court held that the test results were admissible but that the defendant could argue to the jury that the test results deserved little weight in light of the government’s level of compliance with the testing procedures. See id.
{10} We would be reluctant to rely on the testimony in the Williams case to support the State’s argument if that argument were made without the benefit of any testimony whatsoever in this case. Here, however, the officer testified in a manner consistent with the Williams testimony. Given our primary holding, that the trial court was entitled to accept the officer’s testimony that Defendant did not regurgitate or introduce anything foreign into her mouth or nose, the officer’s testimony that the reading after the aborted test was proper appears to further support the trial court’s ruling. Thus, we hold that there was no error in admitting the breathalyzer test results. Because there was no violation of the regulation, any arguable shortcoming in the testing procedure goes to the weight of the evidence, not its admissibility. See State v. Watkins, 104 N.M. 561, 563-64, 724 P.2d 769, 771-72 (Ct.App.1986).
2. HGN Testimony
{11} Defendant contends that the trial court erred in admitting the HGN test results because the officer administered the test poorly, and the test results were scientific evidence that should not be admitted without the proper evidentiary foundation. While we share some of Defendant’s concerns, under the unique circumstances of this case, we do not find any reversible error.
{12} It appears to us that HGN evidence can be used in one of two ways. First, it can be used like other field sobriety tests, e.g., that a person who fails the touch-your-nose test, the stand-on-one-foot test, or the heel-to-toe walking test is particularly intoxicated. Although some cases hold that HGN evidence used in this matter is similar to lay testimony of intoxication, e.g ., the person smelled of alcohol; the person was unsteady; the person had bloodshot, watery eyes; the person had slurred speech; etc., see State v. Cissne, 72 Wash.App. 677, 865 P.2d 564, 567-68 (1994), we would be inclined to agree with the holding in Cissne that such use is not lay opinion under Rule 11-701 NMRA 1998. Rather, it would be more like expert testimony under that portion of Rule 11-702 that deals with testimony based on specialized knowledge an expert gains by experience. Second, HGN evidence can be used as scientific, expert testimony of a particular level of intoxication. See Cissne, 865 P.2d at 567-68 (noting that cases admitting HGN testimony as lay testimony do not admit it as evidence of a specific degree of intoxication).
{13} In this case, the officer said that, in his experience, people who fail the test the way he gives it are intoxicated. He also explained that asserted deficiencies in the way he gave the test would be meaningless. He explained that he saw the jerkiness in less than the required four seconds; waiting more seconds would not matter. The officer also said that a failure of the test is 77% accurate in telling that a person had a blood alcohol level of .10 or greater when the test is performed properly. The former testimony is more like expert testimony based on experience; the latter, establishing a particular blood alcohol content, is more like scientific, expert testimony.
{14} In this case, based on the officer’s testimony, the objection to it, its impeachment, and the way the evidence was used by the attorneys, the HGN evidence was more in the nature of expert evidence that was not *716based on scientific principles. We consider it noteworthy that Defendant did not object below that there was no scientific basis for the officer’s testimony; Defendant never mentioned either State v. Alberico, 116 N.M. 156, 861 P.2d 192 (1993), or Rule 11-702. Thus, to the extent that Defendant contends that the “trial court should have done a full Alberico analysis,” we hold that the trial court was never asked to do so. See State v. Casteneda, 97 N.M. 670, 674, 642 P.2d 1129, 1133 (Ct.App.1982).
{15} Defendant’s objection below went to the foundation for the HGN testimony; Defendant contended that the foundational requirement was that the test must be performed properly. When used as nonseientific, expert testimony, we believe our Supreme Court would rule that deficiencies in conducting the HGN test such as were shown below would go to the weight, not the admissibility, of the evidence, as the trial court ruled. As expert testimony of a specific degree of intoxication, however, we would be inclined to believe that the trial court may have erred in admitting the testimony, but we hold that any error was harmless under the facts of this case.
{16} The prosecutor never argued to the jury that the HGN evidence established any particular level of intoxication. In fact, she recognized that the jury may look entirely askance at the HGN evidence: “this is a little complicated and it sounds a little scientific and a little bit like *what are you talking about?’ But the officer’s looking for things ... and based on what he saw, she’s under the influence.” The way the prosecutor used the HGN evidence was like other observations that people make of intoxicated people. The prosecutor’s emphasis, as one might expect, was on the very scientific and precise breath test, on which Defendant scored .16.
{17} With the officer admitting that so much of the way he conducted the HGN test was not in accordance with his training, defense counsel was able to credibly argue that the HGN test was “garbage.” Because the focus of the evidence showing Defendant was actually intoxicated was the breath test, because the HGN evidence as explained by the officer could be viewed as not showing more than what is ordinarily shown with field sobriety tests, because that is how the prosecutor used the evidence, and because Defendant never specifically voiced an objection to the trial court based on scientific evidence, we cannot say that this is a case in which we should undertake to definitively decide whether HGN evidence is scientific evidence subject to strict standards of admissibility. We note that Defendant informs us that the Supreme Court has at least one case pending before it in which it may decide issues concerning HGN evidence. See State v. Torres, S.Ct. No. 23,334. We also note that the Supreme Court has ruled in an analogous case that infirmities in DNA testing go to the weight, not the admissibility, of such evidence. See State v. Anderson, 118 N.M. 284, 299, 881 P.2d 29, 44 (1994). Thus, we would be hard pressed to say that the trial court’s similar ruling here, under the circumstances in which it was made, constituted reversible error.
{18} In addition, we cannot say that HGN evidence played such an important part in the simple DWI conviction, which we uphold, that reversible error was present. We are supported in this ruling by the Williams case, cited above, 884 P.2d at 172-73. The appellate court in that case did not rule on a challenge to HGN testimony, finding any reference to it harmless. In that case, the judge denied a motion to exclude the HGN evidence; the officer testified about how the defendant took the test, but did not say whether the defendant passed or failed the test (apparently, the prosecutor’s oversight); and the prosecutor argued, without objection based on a lack of evidence, in both opening and closing that the defendant failed the HGN test. The comments in opening and closing were in long lists of factors showing that the defendant was intoxicated.
{19} Our case is similar enough to warrant the same result. For the reasons given above, we are confident that the conviction was not based in any way on expert, scientific evidence that was admitted without a necessary foundation going to admissibility. If the HGN evidence contributed to the conviction at all, it was as testimony in which any deficiencies in the testing would go to *717weight under Anderson. . Considering that the strongest testimony of a specific level of intoxication was the breathalyzer results of .16 and considering that Defendant admitted that she had been drinking, we cannot say that the HGN evidence amounted to reversible error. See Clark v. State, 112 N.M. 485, 487, 816 P.2d 1107, 1109 (199Í) (error may be deemed harmless only when there is no reasonable possibility that it contributed to the conviction).
CONCLUSION
{20} Defendant’s conviction for aggravated DWI is vacated, and the case is remanded with instructions to enter a judgment reducing that conviction to simple DlVI. In other respects, Defendant’s convictions are affirmed.
{21} IT IS SO ORDERED.
DONNELLY, J., concurs.
HARTZ, C. J., concurring in part and dissenting in part.