OPINION

Opinion By Justice O'NEILL.

Appellant Christopher Jerome Wilson appeals his conviction for felony driving while intoxicated (DWI). In four issues, appellant contends (1) the trial court erred in denying his motion to suppress evidence, (2) the evidence is factually insufficient to support his conviction, (3) the trial court erred in admitting evidence of the Horizontal Gaze Nystagmus (HGN) test, and (4) the trial court erred in denying his motion for mistrial. For the following reasons, we affirm the trial court's judgment. The grand jury indicted appellant for felony DWI alleging he committed the offense of DWI having twice before been convicted of DWI. Appellant filed a pretrial motion to suppress the results of his field sobriety tests asserting the tests were the fruit of an illegal detention. Following a hearing, the trial court denied appellant's motion to suppress. At trial, the State presented evidence that appellant operated a motor vehicle while intoxicated. Appellant stipulated to two prior DWI convictions, and the jury found appellant guilty as charged. The trial court assessed punishment, enhanced by two prior felony convictions, at twenty-five years' confinement. This appeal followed. In his first issue, appellant contends the trial court abused its discretion in overruling his motion to suppress evidence. He asserts his field sobriety tests should have been suppressed because they were taken after a temporary detention became unreasonable when city police waited for over thirty minutes for a state trooper to take over the DWI investigation. The evidence at the hearing on the motion to suppress showed appellant was initially detained by Melissa Police officer Jimmie Stanley. Stanley suspected appellant was intoxicated. However, because he was new to the police force and not qualified to administer field sobriety tests, he called his superior officer, Sergeant Kyle Babcock. When Babcock arrived, he also suspected appellant was intoxicated. He called state troopers to take over the investigation. The city officers called state troopers because the Melissa police force was small and usually had only two officers on duty at any given time. Turning DWI investigations over to state troopers allows the Melissa officers to be available to respond to emergency calls if necessary. It took state trooper Margarito Lopez about twenty or thirty minutes to arrive. While the Melissa officers were waiting for Lopez, they continued the investigation, asking appellant questions, obtaining consent to search his vehicle, and searching the vehicle. When Lopez arrived, he administered field sobriety tests and then placed appellant under arrest for DWI. Appellant does not contest his initial detention. Instead, he asserts his continued detention while Melissa officers waited for the state trooper exceeded the proper scope of a temporary detention. We review a trial court's ruling on a motion to suppress evidence under a bifurcated standard of review. St. George v. State, 237 S.W.3d 720, 725 (Tex. Crim. App. 2007); Ford v. State, 158 S.W.3d 488, 493 (Tex. Crim. App. 2005). We do not engage in our own factual review; rather, the trial court is the sole trier of fact and judge of the credibility of the witnesses and the weight to be given to their testimony. State v. Ross, 32 S.W.3d 853, 855 (Tex. Crim. App. 2000); Guzman v. State, 955 S.W.2d 85, 89 (Tex. Crim. App. 1997). We give trial courts "almost complete deference in determining historical facts." St. George, 237 S.W.3d at 725; see Carmouche v. State, 10 S.W.3d 323, 327 (Tex. Crim. App. 2000). We review the record to determine whether the trial judge's ruling is supported by the record and correct under some theory of law applicable to the case. Armendariz v. State, 123 S.W.3d 401, 404 (Tex. Crim. App. 2003). A police officer may stop and briefly detain persons suspected of criminal activity on less information than required for probable cause for arrest. See Terry v. Ohio, 392 U.S. 1, 20-22 (1968). The length of a temporary detention may make an otherwise legitimate stop unreasonable. See United States v. Sharpe, 470 U.S. 675, 679 (1985). However, there is no rigid, bright-line time limitation for an investigatory detention. Sharp, 470 U.S. 675, 679; Belcher v. State, 244 S.W.3d 531, 539 (Tex. App.-Fort Worth 2007, no pet.). Instead, common sense and ordinary human experience must govern. Id. at 685. The reasonableness of the duration of a detention depends on whether the police diligently pursued a means of investigation that was likely to confirm or dispel their suspicions quickly, during which time it was necessary to detain the defendant. Id. at 686. In determining the reasonableness of the duration of a detention, the trial and appellate courts may consider legitimate law enforcement purposes served by any delay in the officer's investigation. Id. at 685. In this issue, appellant maintains the Melissa police officers' detention of him, while they waited for a state trooper to administer field sobriety tests, was unreasonable. Appellant's argument is twofold. First, he suggests it was not a legitimate law enforcement purpose to detain him while the Melissa officers waited for a state trooper. Second, he contends the sheer length of the delay was unreasonable. In Belcher, the defendant was stopped for a traffic offense. Belcher v. State, 244 S.W.3d 531, 539 (Tex. App.-Fort Worth 2007, no pet.). The police officer that stopped the defendant, Officer Willenbrock, suspected the defendant was intoxicated. Although Willenbrock was authorized to conduct DWI investigations, he called another officer, Officer Martin, who was more qualified to take over the investigation. Willenbrock waited over thirty minutes for Martin to arrive. While Willenbrock performed some minimal investigation, the reason for the delay was waiting for Martin. The Fort Worth Court of Appeals concluded that a legitimate law enforcement purpose was served by the delay because Martin was more efficient in conducting DWI investigations and could complete it more quickly. Id. at 541. We likewise conclude that it was a legitimate law enforcement purpose for the Melissa officers to call the state trooper so that the City of Melissa would have officers available to respond to emergency calls. We also cannot agree with appellant that the length of the time the officers waited for Lopez to was unreasonable. The evidence before the trial court regarding the length of this delay was not entirely clear. Stanley testified it took Babcock about five minutes to arrive and then Lopez an additional fifteen or twenty minutes. He however acknowledged it could have taken Lopez as long as thirty or thirty-five minutes to arrive. During this time period, contrary to appellant's assertions, some investigation did occur. Specifically, both Stanley and Babcock questioned appellant, and Babcock asked appellant for consent to search the vehicle and searched the vehicle. We cannot conclude based on the totality of the circumstances that the length of the delay was unreasonable. See Belcher, 244 S.W.3d at 534. We resolve the first issue against appellant. In his second issue, appellant contends the evidence is factually insufficient to support his conviction. In reviewing the factual sufficiency of the evidence, we view all of the evidence in a neutral light to determine whether the jury's verdict of guilt was rationally justified. See Lancon v. State, 253 S.W.3d 699, 705 (Tex. Crim. App. 2008); Roberts v. State, 220 S.W.3d 521, 524 (Tex. Crim. App. 2007). We determine whether (1) the proof of guilt is so obviously weak as to undermine confidence in the fact finder's verdict, or (2) the proof of guilt, even if sufficient standing alone, is greatly outweighed by contrary proof. See Johnson v. State, 23 S.W.3d 1, 11 (Tex. Crim. App. 2000); Clewis v. State, 922 S.W.2d 126, 129 (Tex. Crim. App. 1996). Unless the record clearly reveals a different result is appropriate, we must defer to the fact finder's determination concerning what weight to give contradictory testimony. Lancon, 253 S.W.3d at 705. At trial, Stanley testified he began following appellant in response to a call about a reckless driver. Appellant swerved from the left hand lane, quickly to the right hand lane, and then stopped on the shoulder. Stanley approached appellant and then asked for his license and insurance. According to Stanley, appellant had difficulty responding to him. Appellant seemed to fumble through his wallet looking for his license even though it was right in front. Appellant's eyes were blood shot and he seemed disoriented. Stanley testified that when appellant got out of the vehicle, he had visible trouble balancing. Stanley also noticed that appellant smelled of alcohol. Appellant denied, at that time, that he had anything to drink. Meanwhile, Officer Kyle Babcock arrived at the scene. Babcock asked appellant some questions and then asked for consent to search. Appellant agreed and Babcock searched the vehicle. After the officers found an open, partially drunk, bottle of beer, appellant then admitted having drunk that beer. On cross-examination, appellant attempted to show various discrepancies between Stanley's testimony, the police report, and the videotape of the stop. For example, Stanley testified that he recalled appellant fumbling for his license, but he did not note that fact in his police report. Stanley's police report stated that he asked appellant for his driver's license several times, but he can only be heard on the video asking for the license one time. Stanley also admitted his police report said appellant had "visible trouble standing" and "almost fell over," but the video at most showed appellant was swaying. Officer Margarito Lopez also testified for the State. Lopez testified he administered three standardized sobriety tests to appellant — the HGN test, the walk-and-turn test, and the one-leg stand. Appellant failed all three tests. Regarding the HGN test, appellant exhibited all six clues of intoxication. Appellant exhibited six of seven clues on the walk-and-turn test and two of four clues on the one-leg stand. Lopez also told appellant to say his ABCs, but appellant stopped before he was finished. Lopez believed appellant was intoxicated and placed him under arrest for DWI. On cross-examination, appellant attempted to establish Lopez improperly administered the field sobriety tests. Appellant called one witness, Mark Frost. Frost is the pastor at appellant's church as well as a Dallas County Probation officer. As a probation officer, Frost works primarily with alcoholics and drug addicts. Frost has known appellant for about fifteen months. Frost testified that when he first met appellant, he thought he was "high," but later discovered that was just how appellant normally acted. In this issue, appellant asserts the jury's finding that he was intoxicated was against the great weight and preponderance of the evidence as to be clearly wrong and unjust. We disagree. The State presented evidence that appellant smelled of alcohol and had blood shot eyes. The videotape of the stop shows appellant swaying somewhat during the detention. Officer Lopez testified appellant failed all three field sobriety tests. Further, appellant was drinking in his vehicle and refused to take an intoxilyzer test. See Tex. Transp. Code Ann. § 724.061 (Vernon 1999) (refusal to take an intoxilyzer test can be considered as evidence of intoxication); Griffith v. State, 55 S.W.3d 598, 601 (Tex. Crim. App. 2001) (same). To show the evidence was factually insufficient he relies primarily on some discrepancies in Stanley's testimony. Appellant generally ignores Lopez's testimony, discounting it in an conclusory manner based on his assertion the field sobriety tests were not properly administered. The only evidence he discusses to support this claim is evidence that Lopez followed an improper procedure in administering one part of the HGN test. We have reviewed all the evidence in a neutral light. Having done so, we cannot conclude the jury's finding of guilt was so clearly against the overwhelming weight of the evidence as to be clearly wrong and manifestly unjust. We resolve the second issue against appellant. In his third issue, appellant contends the trial court erred in refusing to suppress results of the HGN test asserting it was not properly administered. Testimony concerning HGN testing is scientific evidence. See Emerson v. State, 880 S.W.2d 759, 763 (Tex. Crim. App. 1994). To constitute scientific knowledge that will assist the trier of fact, the proposed testimony must be relevant and reliable. Id. To be considered reliable, evidence based on a scientific theory must satisfy three criteria: (1) the underlying scientific theory must be valid, (2) the technique applying the theory must be valid, and (3) the technique must have been applied properly on the occasion in question. Id. With respect to the HGN test, the court of criminal appeals has taken judicial notice of (1) the validity of the scientific theory and (2) the validity of the technique applying the theory. Id. at 768-69. Thus, the issue is whether the officer properly applied the theory in this case. See id. When administering an HGN test, an officer must follow the standardized procedures outlined in the DWI detection manual published by National Highway Traffic Safety Administration (NHTSA). Id. at 768. Slight variations in the administration of the HGN test do not necessarily render the evidence inadmissible or unreliable. See Compton v. State, 120 S.W.3d 375, 378 (Tex. App.-Texarkana 2003, pet. ref'd); see also Plouff v. State, 192 S.W.3d 213, 221 (Tex. App.-Houston [14th Dist.] 2006, no pet.). In this case, appellant asserts Lopez improperly administered the test because he used an improper procedure in estimating a forty-five degree angle. The forty-five degree angle is relevant in determining two of the six clues of intoxication for that test. When Lopez was trained to administer the HGN test, the NHTSA guidelines taught officers to estimate forty-five degrees by holding the stimulus in front of the nose twelve to fifteen inches and move it to the shoulder. The current NHTSA manual still instructs the officer to start twelve to fifteen inches in front of the suspect's nose, but now instructs the officer to move the stimulus out twelve to fifteen inches. According to Lopez, twelve to fifteen inches would be about to the shoulder. Appellant has not shown any practical difference between the guidelines and how the test was administered in this case. While appellant may have shown Lopez did not use the current procedure for estimating forty-five degrees, he has not shown Lopez did not nevertheless accurately estimate forty-five degrees. We conclude that under the facts presented here, the trial court did not abuse its discretion in refusing to suppress the results of the HGN test. We resolve the third issue against appellant. In his fourth issue, appellant contends the trial court erred in overruling his motion for mistrial after Lopez testified that failure of the HGN test equated to a blood-alcohol concentration (BAC) of .08 or higher. The HGN test is sufficiently reliable to be admissible as evidence of intoxication, but not as evidence of a precise BAC. See Emerson, 880 S.W.2d at 769. Prior to trial, the trial court granted appellant's motion in limine with respect to any attempts by the prosecutor to correlate the results on the HGN test to a BAC. However, on direct examination, in a nonresponsive answer to the prosecutor's question, Lopez testified that once a suspect reaches four clues on the HGN test, it is probable his BAC is above a ".10 or .08." Appellant timely objected and the trial court sustained the objection and instructed the jury to disregard the officer's testimony insofar as it attempted to correlate results on the HGN test to a specific BAC. The trial court denied appellant's motion for mistrial. When reviewing a trial court's ruling on a motion for mistrial, an appellate court must uphold the trial court's ruling if it was in the zone of reasonable disagreement. Archie v. State, 221 S.W.3d 695, 699-700 (Tex. Crim. App. 2007) (citing Wead v. State, 129 S.W.3d 126, 129 (Tex. Crim. App. 2004)). Only in extreme circumstances where the prejudice is incurable, will a mistrial be required. Gallo v. State, 239 S.W.3d 757, 775 (Tex. Crim. App. 2007); Archie, 221 S.W.3d at 69. A mistrial is appropriate only for "a narrow class of highly prejudicial and incurable errors" and may be used to end trial proceedings if the error is "so prejudicial that expenditure of further time and expense would be wasteful and futile." Hawkins v. State, 135 S.W.3d 72, 77 (Tex. Crim. App. 2004). When determining whether to grant a mistrial, the trial court should balance three factors: (1) the severity of the misconduct; (2) the trial court's measures to cure the misconduct (i.e. the efficacy of any cautionary instruction by the judge); and (3) the certainty of conviction absent the misconduct. Archie, 221 S.W.3d at 700; Hawkins. 135 S.W.3d at 77. Ordinarily, a prompt instruction to disregard will cure error associated with an improper question or answer. Ovalle v. State, 13 S.W.3d 774, 783 (Tex. Crim. App. 2000). Here, Lopez testified about the BAC correlation in an unresponsive answer to the prosecutor's question. When the evidence was offered, the trial court sustained appellant's objection and promptly instructed the jury to disregard. More significantly, in this case, the jury was not charged that it could find appellant was intoxicated based on a BAC. Thus, it is unlikely the jury used the evidence in an improper manner. Under these facts, we cannot conclude the error was either highly prejudicial or incurable. The trial court did not abuse its discretion in denying appellant's motion for mistrial. We resolve the fourth issue against appellant. We affirm the trial court's judgment.