OPINION
Defendant Pablo Arenas-Ortiz was convicted of illegally re-entering the United States after having been deported in violation of 8 U.S.C. § 1326. He appeals the district court’s denial of his motion to compel discovery to support his claim that the United States Attorney engaged in a pattern of selective prosecution of Hispanic males under 8 U.S.C. § 1326, in violation of the equal protection guarantees of the Fifth Amendment. Because Arenas-Ortiz has failed to present evidence that similarly situated individuals could have been prosecuted, but were not, we affirm the judgment of the district court.
I. LEGAL FRAMEWORK
We must exercise a high degree of deference to the decision of prosecuting authorities to bring charges, because the Constitution assigns that decision to the executive branch of government. See United States v. Armstrong, 517 U.S. 456, 464, 116 S.Ct. 1480, 134 L.Ed.2d 687 (1996). One important restriction on prosecutorial discretion, however, is that “the decision whether to prosecute may not be based on ‘an unjustifiable standard such as race, religion, or other arbitrary classification.’ ” Id. at 464, 116 S.Ct. 1480 (quoting Oyler v. Boles, 368 U.S. 448, 456, 82 S.Ct. 501, 7 L.Ed.2d 446 (1962)). To establish such a violation of equal protection, “[t]he claimant must demonstrate that the federal prosecutorial policy ‘had a discriminatory effect and that it was motivated by a discriminatory purpose.’ ” Id. at 465, 116 S.Ct. 1480 (quoting Wayte v. United States, 470 U.S. 598, 608, 105 S.Ct. 1524, 84 L.Ed.2d 547 (1985)).
To meet the first requirement, of discriminatory effect, Arenas-Ortiz “must show that similarly situated individuals of a different [ethnic origin] were not prosecuted.” Id.; see also United States v. Bass, 536 U.S. 862, 863, 122 S.Ct. 2389, 153 L.Ed.2d 769 (2002). Specifically, he must show that non-Hispanic-males were not prosecuted even though they: (1) were aliens, (2) had been removed or deported from the United States, and (3) had reentered without the consent of the Attorney General. See 8 U.S.C. § 1326. This standard for demonstrating a violation of equal protection is “a demanding one.” *1069 Armstrong, 517 U.S. at 463, 116 S.Ct. 1480. The showing necessary to obtain discovery is somewhat less: the defendant must produce “some evidence that similarly situated defendants of other races could have been prosecuted, but were not.” Id. at 469, 116 S.Ct. 1480. Even this standard, however, is a “rigorous” one designed to minimize interference with the prosecutorial function. Id. at 468, 116 S.Ct. 1480.
The district court held that Arenas-Ortiz had failed to meet the standard for discovery. We review that determination for abuse of discretion. See United States v. Candia-Veleta, 104 F.3d 243, 246 (9th Cir.1996). For reasons we now set forth, we find no abuse of discretion.
II. EVIDENCE OF SELECTIVE PROSECUTION
Arenas-Ortiz submitted several pieces of statistical evidence in an attempt to demonstrate a racial disparity between eligible § 1326 defendants and actual § 1326 defendants. Arenas-Ortiz first sought to derive the percentage of actual defendants who are Hispanic males. Analysis of the § 1326 caseload for the Federal Public Defender’s office for the Northern District for the period from 1985 to 2001 revealed that of the 1,556 defendants represented by the Public Defender’s office, 1,470, or 94.5%, were Hispanic males. Arenas-Ortiz then attempted to show that the percentage of eligible § 1326 defendants who are Hispanic males is significantly lower than 94.5%. In support of his claim, Arenas-Ortiz submitted a declaration from a statistician, Michael J. Sullivan, that relied principally on two separate figures to establish evidence of selective prosecution. First, Sullivan took census data indicating that 66.6% of aliens in California are Hispanic. From that fact, he extrapolated that 66.6% of the alien prison population in California is Hispanic, and consequently that 66.6% of eligible § 1326 defendants are Hispanic. Sullivan concluded that this difference between the percentage of eligible defendants (66.6%) and the percentage of actual defendants (94.5%) constituted statistically significant evidence of selective prosecution.
The district court did not abuse its discretion in ruling that this evidence was fatally flawed. The first problem with the statistical evidence is Sullivan’s assumption that 66.6% of the alien prison population is Hispanic because 66.6% of the statewide alien population is Hispanic. The Supreme Court has cautioned that such an assumption is faulty because members of a particular racial group often do not commit crimes at a rate proportionate to their representation in the overall population. See Armstrong, 517 U.S. at 469-70, 116 S.Ct. 1480. For example, in Armstrong, the Supreme Court observed that evidence that 90% of individuals sentenced for crack cocaine trafficking in 1994 were black is not sufficient evidence of selective prosecution, even though the black population in the United States is much less than 90%. See id. Similarly, here, sociological and other factors may cause the percentage of Hispanic aliens in prison to be either higher or lower than the overall percentage of Hispanic aliens. See United States v. Turner, 104 F.3d 1180, 1184-85 (9th Cir.1997) (observing that a number of different factors influence the frequency that members of certain groups may commit crimes relative to other groups). Thus, it is unreasonable to infer that 66.6% of aliens in California prisons are Hispanic simply because 66.6% of aliens in California are Hispanic.
Even if Sullivan’s assumption that 66.6% of the alien prison population is Hispanic were reasonable, that figure nonetheless fails to provide “some evidence” concerning individuals similarly situated to Are*1070nas-Ortiz. The data cited by Sullivan does not reveal the percentage of incarcerated Hispanic male aliens who have been previously deported or the percentage of incarcerated Hispanic male aliens who have illegally re-entered the United States.1 Thus, it does not follow from the fact that 66.6% of criminal aliens are Hispanic males that 66.6% of criminal aliens who have violated § 1326 are Hispanic males.
There are equivalent problems with the next analysis presented by Arenas-Ortiz’s expert. Sullivan took Immigration and Naturalization Service (INS) deportation data indicating that for the year 1999, 89% of deported individuals came from predominately Hispanic countries. Building on this data, he assumed that if 89% of deported individuals are Hispanic, then 89% of § 1326 violators also should be Hispanic instead of 94.5%. According to Sullivan, this disparity between eligible and actual defendants also was statistically significant evidence of selective prosecution.
This analysis is faulty. Simplified, Sullivan’s proposition is that selective prosecution may be inferred because 94.5% of § 1326 defendants are Hispanic males, while only 89% of persons deported from the United States are Hispanic. This statistic is probative of the treatment of similarly situated individuals only if one assumes that deported individuals of different ethnic origins return to the United States at relatively equal rates, and therefore that Hispanics comprise 89% of individuals illegally re-entering the United States. As the district court aptly noted, however, common sense suggests that it would be substantially more difficult for an alien removed to China to return to the United States than for an alien removed to Mexico to do so. Thus, it is entirely possible that the percentage of Hispanic male aliens illegally re-entering the United States approaches 94.5% even if Hispanics comprise only 89% of all deportees.2 Certainly the data offered do not tend to negate that possibility. In the absence of evidence regarding the relative frequency that members of different races illegally re-enter the United States, the district court did not abuse its discretion in concluding that Arenas-Ortiz was not entitled to discovery.3
III. “INSUPERABLE TASK”
Arenas-Ortiz next contends the district court’s ruling was an abuse of discretion because it would have been an “insuperable task” to produce the evidence required by the court to justify discovery. Arenas-Ortiz relies on a single fine in Armstrong in which the Supreme Court, in holding that the defendant’s evidence of selective prosecution was insufficient to warrant discovery, noted that “it should not have been an insuperable task to prove that persons of other races were being treated differently than respondents.” Armstrong, 517 U.S. at 470, 116 S.Ct. 1480. From this statement, Arenas-Ortiz posits that a district court must order discovery when it would be an “insuperable *1071task” to produce some evidence of differential treatment of members of other races similarly situated.
Arenas-Ortiz extends the language of Armstrong beyond its intended meaning. The Supreme Court merely observed that, in the case before it, the defendant should have been able to produce the information concerning similarly situated individuals that the Court found lacking. It did not articulate a standard that discovery is required in every case in which the defendant has no feasible way of augmenting an inadequate evidentiary showing. It is in the nature of a standard that there will be times when that standard cannot be met. Merely demonstrating that better evidence cannot be obtained without discovery does not suddenly render otherwise insufficient evidence sufficient.4
IY. CONCLUSION
The district court did not abuse its discretion in ruling that Arenas-Ortiz had not met the selective prosecution standard for discovery set forth by the Supreme Court in Armstrong. The judgment of the district court is
AFFIRMED.