153 S.W.3d 435 (2005)
The STATE of Texas
Matthew Reid MECHLER, Appellee.

No. 0075-04.
Court of Criminal Appeals of Texas.

January 12, 2005.
Brian W. Wice, Houston, for Appellant.

David C. Newell, Asst. District Atty., Richmond, Matthew Paul, State’s Atty., Austin, for State.


KEASLER, J., delivered the opinion of the Court in which KELLER, P.J., and WOMACK, JOHNSON, HERVEY, and HOLCOMB, JJ., joined.

After hearing testimony on Matthew Mechler’s pretrial motion, the judge suppressed his intoxilyzer results, relying on Mata v. State[1] and Texas Rule of Evidence 403. Reviewing the case de novo, the Court of Appeals reversed the trial court by holding that Mata was inapplicable and Rule 403 was misapplied. We agree with the Court of Appeals’s result but find error in its failure to review for an abuse of discretion. Nevertheless, we affirm its judgment.

I. Factual and Procedural History

Deputy Clarence Golden pulled Mechler over upon observing him commit various traffic violations, including running off the road in executing a sweeping turn, rolling through a stop sign, and other erratic movements. Upon approaching Mechler’s vehicle, Deputy Golden detected an odor of alcohol and noticed Mechler’s movements were slow and sluggish in complying with his request for Mechler’s driver’s license and insurance. When asked if he had been drinking, Mechler stated, “A little.” After administering field-sobriety tests, Golden concluded Mechler was intoxicated and placed him under arrest. Nearly an hour and a half later, Mechler gave a breath sample indicating a blood alcohol concentration (BAC) of .165. The State’s information charged Mechler with driving while intoxicated, alleging intoxication under both an impairment definition, which is the loss of his mental and physical faculties, and a per se definition, or having a blood alcohol concentration of at least . 08.

Mechler filed a motion to suppress the intoxilyzer results on two grounds. First, Mechler argued the State was unable to offer retrograde extrapolation testimony because its expert could not testify on the requisite factors outlined by our decision in Mata v. State.[2] Second, he claimed the intoxilyzer results were irrelevant under Rule 401, or alternatively, inadmissible under Rule 403 because their probative value was substantially outweighed by the danger of unfair prejudice.

At the suppression hearing, the trial court heard testimony from Deputy Golden, an officer who administered the breath test, and a technical supervisor employed by the Texas Department of Public Safety. Deputy Golden testified to his observations and conclusions concerning Mechler’s intoxication. His testimony included the reason for the initial stop, the smell of alcohol when he approached Mechler, and Mechler’s inability to perform various field-sobriety tests. Deputy Ressler testified to the administration of the breath test and was examined on his understanding of the intoxilyzer’s mechanics and tolerances. The trial court also heard testimony from Technical Supervisor Barbosa. After admitting she was unable to testify on the necessary Mata factors, the State conceded its inability to present extrapolation testimony.

The trial court ordered the intoxilyzer results suppressed. The trial court’s order stated,

Pursuant to the application of Texas Rule of Evidence 403 and State v. Mata, [Mata v. State][citation omitted] to the facts of this case, the probative value of the breath test results in this case was substantially outweighed by the danger of unfair prejudice due to the State’s inability to present breath test extrapolation evidence.
The court further stated, “All of the witnesses who testified were credible and this Court did not base its decision upon the credibility of the witnesses.” The State appealed.

II. Court of Appeals

The Court of Appeals reviewed the suppression order de novo.[3] In applying Guzman v. State,[4] the court reasoned that a de novo standard of review was proper because the trial court’s decision did not involve an evaluation of credibility and the facts concerning the suppression issues were undisputed.[5] The Court of Appeals found error in suppressing the intoxilyzer results based on Mata because the Mata decision was inapplicable to this case.[6] After applying the Montgomery factors for analyzing a trial court’s Rule 403 ruling, the court further held the trial court erred in suppressing the test results under Rule 403.[7] We granted Mechler’s petition for discretionary review to address whether the Court of Appeals erred in reversing the trial court’s suppression of Mechler’s breath-test results which was based on (1) Texas Rule of Evidence 403 and (2) Mata v. State.

III. Analysis

A. Mata

We begin with Mata. The Court of Appeals held that Mata was inapplicable to this case because Mata addressed only “the admissibility of expert testimony rather than test results.”[8] We agree. Mata addressed the necessary factors needed to establish the reliability of expert testimony on retrograde extrapolation.[9] In Mata, we stated that

we are not addressing whether retrograde extrapolation is necessary in order for the State to prove a defendant guilty in a DWI case. Nor do we address whether test results showing a defendant’s BAC at some time after the alleged offense are admissible at trial in the absence of retrograde extrapolation.[10]
The trial court erred in suppressing the intoxilyzer results based on Mata.

B. Rule 403

1. Standard of Review

Mechler alleges that the Court of Appeals erred in evaluating the trial court’s suppression order de novo. While acknowledging that an order on a motion to suppress is normally reviewed for an abuse of discretion, the Court of Appeals concluded that a de novo standard is appropriate because there were no disputed issues and no need to observe the credibility and demeanor of witness testimony.[11] We disagree.

We have long held that a trial court is entitled to broad discretion in ruling on a Rule 403 objection.[12] In Montgomery v. State,[13] we noted two significant rationales for this broad discretion. First, Rule 403’s language implies that a determination under this rule is inherently discretionary with the trial court.[14] Rule 403 provides, in pertinent part, that relevant evidence “may be excluded if its probative value is substantially outweighed by the danger of unfair prejudice.”[15] The inclusion of the word “may” displays the drafter’s intent to vest the trial courts with substantial discretion.[16] Second, the trial court is in a superior position to evaluate the impact of the evidence.[17] Unlike the trial court, an appellate court “`cannot weigh on appeal… the intonation and demeanor of the witnesses preceding the testimony in issue … nor can we determine the emotional reaction of the jury to other pieces of evidence….'”[18]

In Guzman, we held that an appellate court may review de novo mixed questions of law and fact not involving an evaluation of credibility and demeanor.[19] But we acknowledged that an abuse-of-discretion review is still the proper standard on a trial court’s evidentiary ruling.[20] We explained that an appellate court’s review of a trial court’s evidentiary rulings generally does not involve mixed questions of law and fact.[21]

It is true that the trial court stated, “All witnesses who testified were credible and this Court did not base its decision upon the credibility of the witnesses.” But a statement in a trial judge’s findings of fact and conclusions of law regarding the role witness credibility played in its decision cannot determine an appellate court’s standard of review. The court’s inclusion of this sentence in its findings does not grant it the ability to control how its rulings will be reviewed.

Finally, the fact that this Rule 403 contention was before the court in the form of a pretrial motion to suppress does not change the standard of review on appeal. When an objection on 403 grounds is raised at trial, we review the judge’s ruling for an abuse of discretion.[22] In this case, the temporal or procedural posture of Mechler’s attempts to suppress the test results does not warrant different treatment or a deviation from the established abuse-of-discretion review. We find the principles set out in Montgomery and its progeny are advanced in this situation by applying an abuse-of-discretion standard.

The test for whether the trial court abused its discretion is whether the action was arbitrary or unreasonable.[23] An appellate court should not reverse a trial judge whose ruling was within the zone of reasonable disagreement.[24] We turn now to the trial court’s remaining ground for suppressing the test results to determine if the court abused its discretion.

2. Application

A proper Rule 403 analysis includes, but is not limited to, four factors: (1) the probative value of the evidence; (2) the potential to impress the jury in some irrational yet indelible way; (3) the time needed to develop the evidence; and (4) the proponent’s need for the evidence.[25] The fact that this Rule 403 determination occurred pretrial in the form of a motion to suppress does not alter either an appellate or trial court’s use of the factors outlined in Montgomery. But in such instances a trial court often will not have enough information before it to adequately apply these factors and assess whether the contested evidence’s probative value is substantially outweighed by its prejudicial effects. In this case, the trial court heard sufficient testimony at the suppression hearing necessary to properly evaluate the motion on Rule 403 grounds.

Probative Value

The first factor looks to the evidence’s probativeness or how compellingly the evidence serves to make a fact of consequence more or less probable.[26] The Court of Appeals held that the intoxilyzer results were probative of intoxication under both the per se and impairment definitions of intoxication.[27] We agree.

We recently held that intoxilyzer results are probative without retrograde extrapolation testimony.[28] Mechler’s intoxilyzer results indicate that Mechler had consumed alcohol. As a result, they tend to make it more probable that he was intoxicated at the time of driving under both the per se and impairment definitions of intoxication. Mechler concedes that this factor weighs in favor of admissibility.

Irrational Impression

The second factor asks whether the evidence has the potential to impress the jury in some irrational but indelible way. The Court of Appeals found the admission of Mechler’s intoxilyzer results would not tend to impress the jury in an irrational way if the trial court instructed the jury to find guilt only if it found that Mechler was intoxicated at the time of driving.[29] We agree with the Court of Appeals’s ultimate decision on this factor, but we reach our conclusion on different grounds.

Rule 403 does not exclude all prejudicial evidence. It focuses only on the danger of “unfair” prejudice.[30] “Unfair prejudice” refers only to relevant evidence’s tendency to tempt the jury into finding guilt on grounds apart from proof of the offense charged.[31] Here, the intoxilyzer results are undoubtedly prejudicial to Mechler, but they are not unfairly prejudicial because this evidence relates directly to the charged offense.[32] This contested evidence does not have a great potential to impress the jury in an irrational way.[33]

Time to Develop

The third factor looks to the time the proponent will need to develop the evidence, during which the jury will be distracted from consideration of the indicted offense. The Court of Appeals found that developing the evidence would not require an undue amount of time. The court reasoned that the intoxilyzer results are evidence of intoxication under both intoxication definitions and the proof under each definition is not mutually exclusive.[34] The court stated

[c]learly, a test showing that blood had a .10 alcohol concentration is probative evidence of a loss of faculties. Conversely, evidence of his failure to pass field sobriety tests immediately after driving his vehicle tends to make it more probable that the failed blood test or breath test taken an hour later accurately reflect the driver’s condition at the time of the offense.[35]
The Court of Appeals also supported its holding by stating the State will not need an undue amount of time devoted to presenting the evidence because the State will not need to explain the reliability of retrograde extrapolation.[36] The injection of retrograde extrapolation and the State’s need to present the extrapolation into this factor analysis is irrelevant to the disposition of this issue. The disputed evidence, and the subject of the motion to suppress, is the intoxilyzer results and not the retrograde extrapolation.

The more significant issue here is whether the jury would be distracted from consideration of the charged offense during the time needed to present the intoxilyzer results. Like the second factor’s analysis, the charged offense charges Mechler with driving while intoxicated. Because the intoxilyzer results relate directly to the charged offense, a jury could not be distracted away from the charged offense regardless of the required time to present the results.[37]

Need for the Evidence

The fourth factor focuses on the proponent’s need for the evidence. This last factor encompasses the issues of whether the proponent has other evidence establishing this fact and whether this fact is related to a disputed issue.[38] Based on the suppression hearing testimony, the Court of Appeals found this factor weighed in favor of exclusion because the State had other probative evidence to establish Mechler’s intoxication.[39] We agree.

The trial court heard testimony from the arresting officer sufficient to dispel the argument for any “great need” of the test results. Officer Golden testified that Mechler rolled through a stop sign and was driving erratically. He further testified that he smelled alcohol on Mechler’s breath and when he approached Mechler’s car. Mechler admitted to drinking “[a] little” alcohol and failed field-sobriety tests. In light of this other evidence probative of Mechler’s intoxication, the State does not have a great need for the test results.

In evaluating all of the Montgomery factors, we hold the sum of the factors weigh in favor of admissibility. The trial court abused its discretion in suppressing Mechler’s test results on Rule 403 grounds.

IV. Conclusion

The Court of Appeals erred in applying a de novo standard of review instead of reviewing for an abuse of discretion. We find that the trial court erred in suppressing the test results under Mata because it is inapplicable to the facts here. After applying the Montgomery factors, we find the factors weigh in favor of admissibility and the trial court erred in suppressing the intoxilyzer results. Therefore, we affirm the judgment of the Court of Appeals.

MEYERS, J., filed a concurring opinion.

COCHRAN, J., filed a concurring opinion in which MEYERS, PRICE, and JOHNSON, JJ. joined.

MEYERS, J., filed a concurring opinion.

I agree with the majority that the court of appeals erred in failing to review this case for an abuse of discretion. I also agree with the holding that the trial court erred in suppressing the intoxilyzer results, but I reach this conclusion for a different reason than the majority.

The majority states that the court of appeals erred in conducting a de novo review, however, the majority makes the same mistake by conducting a Rule 403 analysis. As Judge Cochran states in her concurring opinion, there are factors that develop at trial that must be considered in a Rule 403 analysis. Therefore, a pretrial hearing is not appropriate in this situation. Because the majority is essentially conducting a de novo pretrial Rule 403 analysis, I cannot join the opinion and concur only in the result.

COCHRAN, J., filed a concurring opinion, in which MEYERS, PRICE, and JOHNSON, JJ., joined.

I concur in the majority’s resolution and agree with much of its reasoning concerning the admissibility of intoxilyzer results in the absence of retrograde extrapolation testimony. I write separately to emphasize two points: 1) the difficulty of making individualized Rule 403 rulings in a pretrial setting; and 2) the need to make individualized Rule 403 rulings on proffered intoxilyzer test results that account for the degree to which the result exceeds the legal limit of 0.08% as well as the time elapsed from driving until the test is taken. I do not understand the majority opinion to imply that intoxilyzer test results are always admissible in a DWI trial under Rule 403, only that, given the specific evidence in this case, the trial court abused its discretion in excluding this particular test result under Rule 403 at the pretrial stage.

A. Rule 403 rulings are context-driven and depend upon the specific evidence before the trial court at the time of the ruling.

First, as a general rule, most of Rule 403’s work of balancing probative value against the risk of unfair prejudice or confusion of issues is done during trial, not pretrial.[1] As the Third Circuit has stated in discussing pretrial rulings concerning scientific evidence, it is rare that Rule 403 is an appropriate basis for the pretrial exclusion of evidence because the trial judge cannot ascertain potential relevance or the impact of countervailing factors without “`a virtual surrogate for a trial record.'”[2]

Second, Rule 403 rulings are largely idiosyncratic. Such rulings usually depend upon the precise evidentiary context of a particularized trial setting, taking into consideration the ebb and flow of trial testimony, the unique circumstances and facts, and the specific contested issues.[3] Rule 403 rulings, because of their finely tuned balance of probative value versus unfair prejudicial effect, do not travel well from case to case. Indeed, legal precedents are of so little value in dictating the proper weight of the balancing process that the American Law Institute specifically forbade the use of Rule 403 rulings as legal precedent when it drafted the Model Code of Evidence.[4] As aptly put by the Fifth Circuit, the specific result of the trial court’s conscientious balance of unique facts and circumstances under Rule 403 “is not subject to scrutiny by an appellate Bureau of Weights and Standards that balances the factors gram for gram.”[5]

Thus, the authority of the trial judge to make individualized, discretionary rulings under Rule 403 during trial is extensive, but it is not boundless. All Rule 403 rulings are subject to three general considerations:[6]

1) the trial judge should exercise his power to exclude evidence under Rule 403 sparingly;[7]
2) the trial judge’s discretion under Rule 403 is not an invitation to rule reflexively or without careful reasoning;[8]
3) the trial judge may not exclude evidence merely because he disbelieves the testimony.[9]
As long as the trial court’s individualized, context-driven Rule 403 ruling is “within the zone of reasonable disagreement, the appellate court will not intercede.”[10]

The problem in this case is that the trial court’s pretrial Rule 403 determination does not appear to be individualized or based upon the specific evidence and context that the trial court had before it. In balancing the probative value of an intoxilyzer test result against its potential for creating unfair prejudice or confusion of the issues, the two most significant items in that balance are: 1) the actual test result itself and how much it exceeded the legal limit of 0.08% BAC; and 2) the time interval between the defendant’s driving and the taking of the test. Yet there is nothing in the trial court’s ruling, reflection, or analysis in this case that accounts for these two crucial considerations. The trial judge explicitly stated that his ruling was not based upon the credibility of any witness, and he did not point to the time lapse between Mr. Mechler’s driving and the intoxilyzer test or the extent to which this BAC test result of 0.165% was in excess of the legal limit as significant factors in his analysis.

From all appearances, the trial court seems to have created an implicit blanket prohibition: Under Rule 403 no intoxilyzer test results shall be admissible without scientifically reliable retrograde extrapolation testimony. That Rule 403 ruling is too sweeping in scope to be decided in a pretrial motion.[11] That ruling is not a discretionary, individualized, context-driven ruling under Rule 403. That ruling is instead the enactment of a newly minted rule of evidence, one that is not contained within the Texas Rules of Evidence. A trial court abuses its discretion and acts arbitrarily and without reference to the guiding principles of law if it creates a universal rule of exclusion under Rule 403 applicable to all cases without regard to the specific facts or context. For this reason alone I would conclude that, based upon the present record, the trial court abused its discretion in excluding the intoxilyzer test result evidence.

B. Balancing the probative value of intoxilyzer test results against the danger of unfair prejudice or confusion of issues under Rule 403 in the absence of retrograde extrapolation testimony.

Because the issue of admitting or excluding intoxilyzer test results under Rule 403 is one that arises in Texas trial courts on a daily basis, we should provide guidance to trial courts concerning the process of balancing probative value against unfair prejudicial effect when retrograde extrapolation testimony is not available.

The primary indicator of whether a person is intoxicated is his blood alcohol concentration (BAC) level. The BAC describes the concentration of alcohol in a person’s blood expressed as weight per unit of volume. At 0.10% BAC, a person has a concentration of 100 mg of alcohol per 100 ml of blood. The BAC may be determined by a blood test, urine test, or, most frequently in DWI prosecutions, by an intoxilyzer test which analyzes a driver’s exhaled breath.

Scientific studies have shown that alcohol may affect one’s driving ability at BAC levels as low as 0.02%.[12] The probability of causing an automobile accident begin to increase significantly at a BAC level of 0.05% and climbs rapidly after about 0.08%.[13] As of 2004, every state in the nation except Minnesota had adopted a per se DWI or DUI statute specifying that a BAC of 0.08% or greater is sufficient proof, by itself, of legal intoxication.[14] The push for a national standard of 0.08% was “based on laboratory and on-road research which demonstrated that the majority of drivers, regardless of experience, are significantly impaired at 0.08 percent BAC [.]”[15] This national standard was based upon the recognition that the “average” driver has lost the normal use of his mental or physical faculties at this level. Laws and statutory presumptions are written for the “average” case although a specific driver, because of individual variables such as weight, gender, physical condition, metabolic rate, time of day, mental state, presence or absence of food in the digestive system, medications, and tolerance for alcohol, might not be impaired at this precise level. But the Texas per se intoxication law, like that in the other 49 states, was written with the hypothetical “average” driver in mind, regardless of individual variables.

Of course, a BAC result of 0.08% on a breath test taken some time after a person has stopped driving does not necessarily correspond to the BAC result that person would have tested at had he been tested at the time he was driving.[16] A number of state legislatures, recognizing that it is not possible for law enforcement officers to test a driver’s BAC while he is driving,[17] have drafted their per se DWI laws such that the law expressly relates the BAC test back to the time of driving, either by a rebuttable presumption or by the definition of the offense.[18] For example, the Alaska DWI statute makes it a crime to operate a motor vehicle when, “as determined by a chemical test taken within four hours after the alleged operation or driving,” a person’s blood contains 0.08% or more alcohol.[19] Similarly, Arizona law provides that it is illegal to have a BAC of 0.08% or more within two hours of driving.[20]

Regardless of their use of statutory presumptions or definitions of DWI incorporating a time lapse between driving and the taking of a breath test, the vast majority of states do not require the prosecution to relate breath test BAC levels back to the time of driving as a predicate to their admission.[21] Courts have rejected the necessity for such “relation-back” or “retrograde extrapolation”[22] evidence for a variety of reasons:

* To create such a requirement would place an impossible burden on the prosecution because only the defendant knows the pertinent variables;[23]
* The DWI statute does not require such relation-back evidence;[24]
* The test result, standing alone, creates a prima facie showing of intoxication at the time of driving which the defendant may rebut;[25]
* The test results alone constitute sufficient evidence of intoxication to obtain a DWI conviction;[26] and
* Other evidence of intoxication corroborates the test results.[27]
Of course, these statutes allow, and courts permit, the defendant to offer expert retrograde extrapolation testimony to show that his individual BAC was, in fact, below the legal limit at the time he was driving.[28]

The “average” alcohol absorption and elimination rate is 0.15% per hour, thus the higher the intoxilyzer test result, the more likely that the driver was intoxicated not only at the time of the test, but also at the time of driving. Expert retrograde extrapolation testimony can exclude the possibility that a driver’s BAC level was still rising in the absorption stage and had not yet reached the legal limit at the moment he stopped driving. But as one court has stated in rejecting an argument that the State must offer extrapolation evidence, “the law was not intended to encourage a perilous race to reach one’s destination, whether it be home or the next bar, before the blood alcohol concentration reaches the prohibited level.”[29] While it may be within the realm of possibility that a specific driver “chug-a-lugged” enough alcohol immediately before or after getting behind the wheel of a car to race to his next destination before his BAC level reached the legal limit of 0.08%, this is a sufficiently remote scenario for state legislatures to have discounted it in drafting their DWI laws which do not require extrapolation testimony.

As this Court aptly noted in Mata v. State,[30] retrograde extrapolation testimony is scientifically unreliable unless the expert witness knows a vast amount of personal information about the driver including:

the presence and type of food in the stomach, the person’s gender, the person’s weight, the person’s age, the person’s mental state, the drinking pattern, the type of beverage consumed, the amount consumed, and the time period of alcohol consumption.[31]
This is all information that is within the defendant’s personal knowledge, but, under the Fifth Amendment, it is not within the State’s power to compel the defendant to disclose this information.[32] Although Mata held that retrograde extrapolation evidence that fails to account for individual variables is inadmissible, in Stewart v. State,[33] we held that a BAC test result, taken eighty minutes after the defendant was arrested, still had some probative value without retrograde extrapolation evidence, and thus was relevant under Rule 401.[34] All relevant evidence, even evidence with a low probative value, is admissible under Rule 402 unless excluded under some other statute or rule, such as Rule 403.

In discussing Rule 403, the majority sets out the four major, but non-exclusive, factors that Texas courts use for balancing probative value and prejudicial effect. The single most important factor in the context of admitting breath test results in the absence of extrapolation testimony is the inherent probative value of the test result. In turn, the relative probative value of the test depends primarily upon two variables: 1) the degree to which the test result exceeds the legal limit of 0.08%; and 2) the amount of time elapsed between driving and the taking of the test. The higher the test result and the shorter the time between driving and testing, the more likely the logical inference is that the defendant had a BAC level at or above 0.08% at the time of driving. The stronger the inference of a BAC of 0.08% or greater at the time of driving, the less significant is the necessity for expert retrograde extrapolation testimony. Conversely, the weaker the inference of guilt, the more vital is the necessity for evidence that relates an accused’s BAC test result back to the time of driving. When a test is obtained long after the arrest and the result is at or below the legal limit, the logical inference that the person had a 0.08% BAC at the time of driving may be so tenuous that a trial judge appropriately exercises his discretion by excluding that specific test result under Rule 403 absent expert testimony that extrapolates the test result back to the time of driving.

In this case, however, the trial judge did not conduct such an individualized assessment, and he failed to offer any rationale as to why the danger of unfair prejudice of evidence of an intoxilyzer test result of 0.165% — twice the legal limit — taken a mere hour and a half after Mr. Mechler stopped driving “substantially” outweighed its probative value. There might be something truly extraordinary about the specific circumstances in this case that would support such a discretionary ruling, but if so, the pretrial record does not reflect those specific facts.

Under these circumstances, therefore, I agree with the majority’s conclusion that the trial judge abused his discretion in excluding the evidence of Mr. Mechler’s intoxilyzer test results.

[1] 46 S.W.3d 902 (Tex.Crim.App.2001).

[2] Id. at 902.

[3] State v. Mechler, 123 S.W.3d 449, 451 (Tex.App.-Houston [14th Dist.] 2003).

[4] Guzman v. State, 955 S.W.2d 85 (Tex.Crim.App.1997).

[5] Mechler, 123 S.W.3d at 451.

[6] Id. at 451-53.

[7] Id. at 454-58.

[8] Id. at 453.

[9] Mata, 46 S.W.3d at 910.

[10] Id.

[11] Mechler, 123 S.W.3d at 451.

[12] See Manning v. State, 114 S.W.3d 922, 926 (Tex.Crim.App.2003).

[13] Montgomery v. State, 810 S.W.2d 372, 378-79 (Tex.Crim.App.1990).

[14] Id.

[15] Tex.R. Evid 403.

[16] Manning, 114 S.W.3d at 926.

[17] Montgomery, 810 S.W.2d at 378-79.

[18] Id. at 379 (citing United States v. Robinson, 560 F.2d 507, 512-16 (2d Cir.1977)).

[19] Guzman v. State, 955 S.W.2d 85, 89 (Tex.Crim.App.1997).

[20] Id.

[21] Id.

[22] See, e.g., Erazo v. State, 144 S.W.3d 487, 489 (Tex.Crim.App.2004); Manning, 114 S.W.3d at 926; Wheeler v. State, 67 S.W.3d 879, 888 (Tex.Crim.App.2002).

[23] Manning, 114 S.W.3d at 926.

[24] Id.; Montgomery, 810 S.W.2d at 380.

[25] Erazo, 144 S.W.3d at 489; Montgomery, 810 S.W.2d at 389-90 (op. on reh’g).

[26] Manning, 114 S.W.3d at 928.

[27] Mechler, 123 S.W.3d at 455.

[28] Stewart v. State, 129 S.W.3d 93, 97 (Tex.Crim.App.2004).

[29] Mechler, 123 S.W.3d at 456.

[30] Manning, 114 S.W.3d at 927; see Tex.R. Evid. 403.

[31] Manning, 114 S.W.3d at 928.

[32] See id.

[33] Id.

[34] Mechler, 123 S.W.3d at 456.

[35] Id. (citations omitted).

[36] Id.

[37] See Manning, 114 S.W.3d at 928.

[38] Id.

[39] Mechler, 123 S.W.3d at 456-57.

[1] See Spain v. Gallegos, 26 F.3d 439, 453 (3d Cir.1994) (noting “cautious approach to Rule 403 exclusions at the pretrial stage”); Petruzzi’s IGA Supermarkets, Inc. v. Darling-Delaware Co., 998 F.2d 1224, 1240 (3d Cir.1993) (same).

[2] In re Paoli RR Yard PCB Litig., 35 F.3d 717, 747 (3d Cir.1994) (quoting In re Paoli R.R. Yard PCB Litig., 916 F.2d 829, 855 (3d Cir.1990)).

[3] 1 Christopher B. Mueller and Laird C. Kirkpatrick, Federal Evidence § 93 at 477 (2d ed.1994); see Li v. Canarozzi, 142 F.3d 83, 88 (2d Cir.1998) (trial judge has broad discretion in performing balancing analysis under Rule 403 because he “sees the witnesses, the parties, the jurors, and the attorneys, and is thus in a superior position to evaluate the likely impact of the evidence”); United States v. Esdaille, 769 F.2d 104, 108 (2d Cir.1985) (trial judge is “in a superior position to evaluate the likely impact of the evidence, since he sees the witnesses, the parties, the jurors, and the attorneys, and their mannerisms and reactions”).

[4] See A.L.I. MODEL CODE OF EVIDENCE RULE 303 (Discretion of Judge to Exclude Admissible Evidence), Comment (1942):

The application of this rule should depend so completely upon the circumstances of the particular case and be so entirely in the discretion of the trial judge that a decision in one case should not be used as precedent in another.

See also 22 CHARLES A. WRIGHT & KENNETH W. GRAHAM, JR., FEDERAL PRACTICE AND PROCEDURE § 5214, at 265 (1977) (past decisions useless in weighing process under Rule 403).

[5] Caldarera v. Eastern Airlines, Inc., 705 F.2d 778, 782 (5th Cir.1983).

[6] See generally, 1 MUELLER & KIRKPATRICK, § 93, at 479.

[7] Montgomery v. State, 810 S.W.2d 372, 388 (Tex.Crim.App.1990) (op. on reh’g) (presumption under Rule 403 is that probative value outweighs prejudicial effect “unless in the posture of the particular case the trial court determines otherwise”); see Conner v. State, 67 S.W.3d 192, 202 (Tex.Crim.App.2001) (“Rule 403 requires exclusion of evidence only when there exists a clear disparity between the degree of prejudice of the offered evidence and its probative value”); Joiner v. State, 825 S.W.2d 701, 708 (Tex.Crim.App.1992) (Rule 403 envisions exclusion of evidence only when there is a “clear disparity between the degree of prejudice of the offered evidence and its probative value”); McFarland v. State, 845 S.W.2d 824, 837 (Tex.Crim.App.1992); see generally 1 JACK WEINSTEIN & MARGARET BERGER, WEINSTEIN’S EVIDENCE ¶ 403[01], at 403-10 (1998) (“If there is any doubt about the existence of unfair prejudice, confusion of the issues, misleading, undue delay or waste of time, it is generally better practice to admit the evidence taking necessary precautions by way of contemporaneous instructions to the jury followed by additional admonitions in the charge”).

[8] Montgomery, 810 S.W.2d at 389 (stating that Rule 403 imposes a duty on trial judge to engage in a balancing process and noting that he “would do well” to inquire of the opponent what his view of the prejudice is and ask the proponent to articulate his need for the evidence). Although a trial judge is not required to articulate his Rule 403 analysis on the record, it is most helpful to reviewing courts. “Rulings that provide more than bare conclusions and indicate the nature of the danger or the underlying considerations are more satisfying and more likely to survive review.” 1 MUELLER AND KIRKPATRICK, § 93, at 481; see, e.g., United States v. Himelwright, 42 F.3d 777, 781 (3d Cir.1994) (when trial court “fails to explain” its grounds for 403 ruling and the reasons for doing so “are not otherwise apparent from the record,” reviewing court “need not defer” to trial court’s ruling and may “examine the record and perform the required balancing” itself); United States v. Dwyer, 539 F.2d 924, 928 (2d Cir.1976) (noting that “where the reasons for a discretionary ruling are not apparent to counsel, they will probably not be apparent to an appellate court”).

[9] Montgomery, 810 S.W.2d at 390 n. 3 (re-emphasizing that credibility of the evidence is not a proper inquiry for the trial judge under Rule 403).

[10] Id. at 391.

[11] See National Union Fire Ins. Co. of Pittsburgh, Pa. v. L.E. Myers Co. Group, 937 F.Supp. 276, 287 (S.D.N.Y.1996) (reserving a ruling on pretrial motion in limine until trial when admission of particular evidence is in an appropriate factual context).

[12] Margaret C. Jasper, DWI, DUI AND THE LAW 13 (Oceana 2004).

[13] Id.

[14] Id. at 115-16, Appendix 18 (noting that all 50 states have enacted a BAC level of 0.08% per se intoxicated DWI statute, but that Minnesota’s statute would not become effective until August 1, 2005); see also Kimberly S. Keller, Sobering Up Daubert: Recent Issues Arising in Alcohol-Related Expert Testimony, 46 S.TEX.L.REV. 111, 111-12 & n. 2 (Fall 2004) (noting that President Clinton, in 1998, called for a national 0.08 BAC per se intoxication limit to reduce the number of alcohol-related deaths on the nation’s highways); see generally Alcohol Policy Information System, A Project of the National Institute on Alcohol Abuse and Alcoholism, accessible at http://alcoholpolicy.niaaa.nih.gov. (The statutory DWI provisions of all 50 states are listed and linked at this site).

[15] Jasper at 13.

[16] See Mata v. State, 46 S.W.3d 902, 909-10 (Tex.Crim.App.2001) (noting that “if a driver is tested while in the absorption phase, his BAC at the time of the test will be higher than his BAC while driving. If tested while in the elimination phase, his BAC at the time of the test could be lower than while driving, depending on whether he had reached his peak before or after he was stopped. Obviously, the greater the length of time between the driving and the test, the greater the potential variation between the two BACs”).

[17] Indeed, Texas regulations, like those in virtually all states, prohibit law enforcement officers from conducting a DWI breath test until after the completion of a fifteen-minute observation period. 37 T.A.C. § 19.3(c)(1).

[18] See generally, D.E. Evins, Construction and Application of Statutes Creating Presumption or Other Inference of Intoxication from Specified Percentages of Alcohol in System, 16 A.L.R.3d 748 (1967 & 2004 updates); V. Woerner, Validity of Legislation Creating Presumption of Intoxication or the Like from Presence of Specified Percentage of Alcohol in Blood, 46 A.L.R.2d 1176 (1956 & 2000 updates).

[19] Alaska Stat. § 28.35.030(a)(2); see Conrad v. State, 54 P.3d 313, 313-15 (Alaska Ct.App.2002) (holding that DWI statute defines offense in terms of BAC at time of driving, not on test result itself, but concluding that “[i]f a chemical test is administered to the defendant within the statutorily prescribed four hours, the test result will create a presumption that the defendant’s blood alcohol level was at least as high at the time the defendant operated or controlled the vehicle”; defendant may offer evidence to rebut that presumption); see also Fla. Stat. § 316.1934(2)(c) (statutory presumption that one whose breath test BAC registers at 0.08% or higher was impaired at the time of driving); Disbro v. State, 791 N.E.2d 774, 778 (Ind.App.2003) (upholding use of statutory presumption that driver whose BAC tested 0.08% or more within three hours of driving was intoxicated at time of driving despite defendant’s rebuttal evidence); Commonwealth v. Murray, 2000 Pa.Super. 84, 749 A.2d 513, 520-21 (2000) (upholding statutory “common sense permissive evidentiary inference [that a driver was at or above 0.08% BAC] based upon the BAC, if the testing occurred within three hours of driving”); Welch v. City of Pratt, Kan., 214 F.3d 1219, 1223-25 (10th Cir.2000) (upholding Kansas statutory permissive presumption of driver’s intoxication if “while driving, [he] had an alcohol concentration in his breath of.08 or more as measured within two hours of the time of operating the vehicle”).

[20] Ariz.Rev.Stat. 28-692(A)(2); See State v. Martin, 174 Ariz. 118, 847 P.2d 619, 623 (App.1992); see also Ga.Code Ann § 40-6-391(a)(5) (“[a] person shall not drive or be in actual physical control of any moving vehicle while … [t]he person’s alcohol concentration is 0.08 grams or more at any time within three hours after such driving or being in actual physical control from alcohol consumed before such driving or being in actual physical control ended”); N.C. Gen.Stat. § 20-138.1(a)(2) (“[a] person commits the offense of impaired driving if he drives any vehicle upon any highway, any street, or any public vehicular area within this State … [a]fter having consumed sufficient alcohol that he has, at any relevant time after the driving, an alcohol concentration of 0.08 or more”); State v. Tischio, 107 N.J. 504, 506, 527 A.2d 388, 389 (1987) (holding “that a defendant may be convicted under N.J.S.A. 39:4-50(a) when a breathalyzer test that is administered within a reasonable time after the defendant was actually driving his vehicle reveals a blood-alcohol level of at least 0.10%. We rule that it is the blood-alcohol level at the time of the breathalyzer test that constitutes the essential evidence of the offense”).

[21] See generally, Jim Frasier, Admissibility and Sufficiency of Extrapolation Evidence in DUI Prosecutions, 119 A.L.R.5th 379 (2004).

[22] Retrograde extrapolation is the scientific process of “working backward” from the BAC test to an estimate of the person’s actual BAC at the time of driving. See LAWRENCE TAYLOR, DRUNK DRIVING DEFENSE § 5.2.1 (5th ed.2000).

[23] See, e.g., State v. Taylor, 132 N.H. 314, 319-20, 566 A.2d 172, 175-76 (1989) (rejecting defendant’s claim that prosecution was required to provided retrograde extrapolation when breath test was given over an hour after he stopped driving; noting that extrapolation testimony “requires evidence that the State will rarely be able to acquire because of the defendant’s constitutional right to remain silent. Extrapolation requires evidence as to when, and in what amounts, the defendant consumed the alcohol prior to driving. Without this information, which is wholly within the defendant’s knowledge in the vast majority of cases, extrapolation of blood alcohol content back to the time of driving becomes an impossible task. The legislature could not have intended to place such an impossible burden on the State”); State v. Kubik, 235 Neb. 612, 619-24, 456 N.W.2d 487, 493-96 (1990) (quoting and following Taylor); People v. Campbell, 236 Mich.App. 490, 501, 601 N.W.2d 114, 119 (1999) (quoting and relying upon Taylor, also rejecting need for extrapolation testimony because “[t]he difficulty of establishing evidence of a nexus between the offense and the test results would have the effect of encouraging defendants to delay taking a blood alcohol test for as long as possible while providing no information regarding their actions near the time of the offense. A defendant’s own actions should not be allowed to effect the suppression of evidence against him”); State v. Greenwood, 115 S.W.3d 527, 532 (Tenn.Crim.App.2003) (quoting and following Kubik and stating that “[t]he legislature obviously knew a blood alcohol test could not be conducted while the defendant was driving. We do not believe it intended to place upon the state the impossible burden of extrapolation in order to prove DUI”).

[24] See, e.g., Kubik, 456 N.W.2d at 493-501 (collecting and discussing cases from numerous jurisdictions holding that DWI “per se intoxication” statutes did not explicitly require extrapolation testimony and courts would not engraft such a requirement for admissibility or sufficiency of evidence purposes); Greenwood, 115 S.W.3d at 532-33 (stating that legislative statute did not require expert extrapolation testimony in a DUI prosecution, and concluding that “a proper blood alcohol test administered at a reasonable time after the defendant has been driving, which reflects a blood alcohol content of .10% or higher, constitutes circumstantial evidence upon which the trier of fact may, but is not required to, convict the defendant of DUI”).

[25] See, e.g., Commonwealth v. Wirth, 936 S.W.2d 78, 83-84 (Ky.1996) (rejecting defendant’s claim that extrapolation testimony necessary and holding that legislature intended that prima facie proof of “per se” DWI arose from properly administered BAC test result of 0.10% or more); Commonwealth v. Yarger, 538 Pa. 329, 334-35, 648 A.2d 529, 531 (1994) (construing DWI statute and concluding that “once the Commonwealth has established that the driver’s blood alcohol content reflects an amount above 0.10%, the Commonwealth has made a prima facie case…. At this point, the defendant is permitted to introduce expert testimony to rebut the Commonwealth’s prima facie evidence. If the defendant decides to rebut the prima facie evidence against him with expert testimony, then the Commonwealth may present its own expert to refute this testimony”). In Haas v. State, 597 So.2d 770 (Fla.1992), the Florida Supreme Court surveyed similar DWI statutes in various jurisdictions before concluding that:

In attempting to combat the scourge of drunk driving, we do not believe the legislature intended to place upon the State the difficult and often impossible burden of extrapolation as a condition precedent to conviction under the [per se DWI] statute. Though our statute is not as specific as that of Minnesota or California, we interpret Florida’s statutory scheme to mean that the test results shall be prima facie evidence that the accused had the same blood-alcohol level at the time of his operation of the vehicle. Properly obtained test results which reflect a blood-alcohol level of 0.10 or more, standing alone, constitute circumstantial evidence upon which the finder of fact may (but is not required to) convict the accused driver of DUI either by impairment or [per se DWI]. However, contrary to the New Jersey position, we do not view the test results as conclusive. The accused is at liberty to seek to demonstrate through cross-examination or the introduction of other evidence that the test results do not accurately reflect his or her blood-alcohol level at the time the vehicle was being operated.

Id. at 774-75.

[26] Haas, 597 So.2d at 774-75; Wirth, 936 S.W.2d at 83-84; Terry v. City of Montgomery, 549 So.2d 566, 567 (Ala.Crim.App.1989) (evidence sufficient to support DWI conviction when driver’s BAC was 0.15% fifty minutes after auto accident; no extrapolation testimony necessary under statute to prove intoxication at time of driving); State v. Wetzel, 7 Haw.App. 532, 538-39, 782 P.2d 891, 895 (1989) (rejecting defendant’s claim that intoxilyzer tests results were inadmissible and insufficient to support conviction without “relation back” testimony; holding that DWI per se statute showed legislative intent that BAC test results of 0.10% or more taken within three hours after time of driving created a permissible inference of guilt, so no further evidence needed for conviction).

[27] Commonwealth v. Kasunic, 423 Pa.Super. 112, 118-19, 620 A.2d 525, 528-29 (1993) (when defendant’s BAC level was twice legal limit fifty minutes after he was found lying on the roadway smelling of alcohol and appearing to officer to be intoxicated, evidence of intoxication sufficient without any extrapolation testimony; although “evidence does not indicate the exact time when appellant operated his vehicle[,] …. in light of the frigid temperature of four degrees below zero Fahrenheit that night, appellant could not have been at the scene laying on the road for a substantial period of time”); Commonwealth v. Montini, 712 A.2d 761, 766 (Pa.Super.1998) (state not required to offer extrapolation testimony in per se DWI prosecution when defendant’s BAC was 0.19% fifty-five minutes after driving and there was evidence he drove erratically and failed to perform field tests adequately).

[28] See generally, Frasier, 119 A.L.R. 5th 379, § 10 (collecting cases from various jurisdictions that hold that defendant may introduce extrapolation testimony to rebut BAC test results).

[29] State v. Tischio, 208 N.J.Super. 343, 348, 506 A.2d 14, 16 (1986).

[30] 46 S.W.3d 902 (Tex.Crim.App.2001).

[31] Id. at 909 (footnotes omitted).

[32] See State v. Taylor, 132 N.H. at 319-20, 566 A.2d at 175-76.

[33] 129 S.W.3d 93 (Tex.Crim.App.2004).

[34] Id. at 96.