In 1998, Massachusetts state prisoner Richard Aspen was convicted of raping and sexually assaulting his stepdaughter. During the selection of the petit jury, Aspen claimed that the prosecutor exercised her peremptory challenges to exclude potential male jurors in violation of the Equal Protection Clause of the Fourteenth Amendment, as interpreted in Batson v. Kentucky, 476 U.S. 79, 85, 106 S. Ct. 1712, 90 L. Ed. 2d 69 (1986) (holding that the racially discriminatory use of peremptory challenges violates equal protection), and J.E.B. v. Alabama, 511 U.S. 127, 129, 114 S. Ct. 1419, 128 L. Ed. 2d 89 (1994) (extending Batson to the use of peremptory challenges on the basis of gender). The trial judge denied Aspen's challenge, concluding that he had not established a prima facie case of discrimination.

Aspen appealed to the Massachusetts Appeals Court. The court considered Aspen's equal protection claim under Article 12 of the Declaration of Rights of the Massachusetts Constitution, which it understood to mandate the same analysis as the Fourteenth Amendment. Commonwealth v. Aspen, 53 Mass.App.Ct. 259, 758 N.E.2d 163, 166 (2001). The court stated that to establish a prima facie case, Aspen had to show "that there [was] a pattern of excluding members" of the venire on account of gender, and "it [was] likely that individuals [were] being excluded solely" on the basis of gender. Id. The court also explained that, once presented with this evidence, the trial judge had to determine "whether to draw the reasonable inference that peremptory challenges have been exercised so as to exclude individuals on account of" gender. Id. After stating the standard, the court concluded that Aspen had not established a prima facie case because the trial "judge could properly conclude that ... gender was not shown to be the likely reason for the [prosecutor's] initial challenges." Id. at 168.

Aspen's request for further appellate review by the Massachusetts Supreme Judicial Court (SJC) was denied. Having exhausted state court remedies, Aspen filed a petition for a writ of habeas corpus in federal district court, arguing that the Appeals Court had applied an incorrect legal standard in considering his Batson claim. He asserted that the state court had erred by requiring him, in order to make out a prima facie case, to establish that it was "likely" that the prosecutor exercised peremptory challenges on account of gender. The district court, applying the deferential standards of review under the Antiterrorism and Effective Death Penalty Act of 1996 (AEDPA), rejected Aspen's petition but did grant him a certificate of appealability, see 28 U.S.C § 2254(b)(1)(A).

We review the denial of Aspen's habeas petition de novo. See Ellsworth v. Warden, 333 F.3d 1, 3 (1st Cir.2003) (en banc). Where, as here, the state court considered Aspen's claim on the merits, our review is governed by AEDPA. See Lynch v. Ficco, 438 F.3d 35, 44 (1st Cir. 2006). Under AEDPA, we may not disturb the state court's conclusion unless the "adjudication of [Aspen's] claim resulted in a decision that was contrary to, or involved an unreasonable application of, clearly established Federal law, as determined by the Supreme Court of the United States." 28 U.S.C. § 2254(d)(1).

A state court adjudication is "contrary to" Supreme Court precedent if, when made, it results from the application of a rule that contradicts the governing law set forth by the Supreme Court or is inconsistent with a Supreme Court decision in a case involving "materially indistinguishable" facts. Williams v. Taylor, 529 U.S. 362, 405-06, 120 S. Ct. 1495, 146 L. Ed. 2d 389 (2000). A state court decision is an unreasonable application of the governing law if the state court identifies the correct governing legal principle from the Supreme Court's then-current decisions but unreasonably applies that principle to the facts of the prisoner's case. See Horton v. Allen, 370 F.3d 75, 80 (1st Cir.2004).

The clearly established law governing the discriminatory use of peremptory challenges was set forth in Batson. The Court described a three-part test for adjudicating claims that peremptory challenges have been exercised in a discriminatory manner. The moving party bears the initial burden of demonstrating a prima facie case of discrimination. Batson, 476 U.S. at 96, 106 S. Ct. 1712. If this burden is met, the non-moving party must then offer a non-discriminatory reason for striking the potential juror. Id. at 97, 106 S. Ct. 1712. Finally, the trial court must determine if the moving party has met its ultimate burden of persuasion that the peremptory challenge was exercised for a discriminatory reason. Id. at 98, 106 S. Ct. 1712.

Batson also described the moving party's burden at the prima facie stage. To establish a prima facie case, the moving party must "raise an inference that the prosecutor used [peremptory challenges] to exclude the veniremen from the petit jury" because of their membership in a protected class. Id. at 96, 106 S. Ct. 1712. The judge may consider all "relevant circumstances" in making this determination. Id.

Most significantly for present purposes, Batson established that the moving party's burden in meeting the prima facie requirement is not substantial. The Batson opinion relied on the Court's earlier Title VII jurisprudence to define "the operation of [the] prima facie burden of proof rules." 476 U.S. at 94 n. 18, 106 S. Ct. 1712. Among the Title VII cases cited in Batson was Texas Department of Community Affairs v. Burdine, 450 U.S. 248, 253, 101 S. Ct. 1089, 67 L. Ed. 2d 207 (1981), in which the Court held that the plaintiff's burden in establishing a prima facie case of discrimination was "not onerous." Our post-Batson precedents reflect the understanding that the burden is not substantial. See United States v. Escobarde Jesus, 187 F.3d 148, 164 (1st Cir.1999); United States v. Bergodere, 40 F.3d 512, 516 (1st Cir.1994).1 Even more importantly, the Supreme Court has recently reiterated that the Batson prima facie standard is not onerous. See Johnson v. California, 545 U.S. 162, 170, 125 S. Ct. 2410, 162 L. Ed. 2d 129 (2005).2

In Johnson, the Supreme Court considered whether it was consistent with Batson to require the moving party to show, at the prima facie stage, that it was "more likely than not" that a peremptory challenge had been exercised on a prohibited ground. Id. at 164, 125 S. Ct. 2410. The Court held that such a "standard [was] at odds with the prima facie inquiry mandated by Batson." Id. at 173, 125 S. Ct. 2410. In reaching this conclusion, the Court relied exclusively on Batson as precedent. It explained that the Batson first step was not intended "to be so onerous that a defendant would have to persuade the judge . . . that the challenge was more likely than not the product of purposeful discrimination. Instead, a defendant satisfies the requirements of Batson's first step by producing evidence sufficient to permit the trial judge to draw an inference that discrimination has occurred." Id. at 170, 125 S. Ct. 2410.

In holding that Aspen had not carried his burden because he had not shown that it "was likely" that the prosecutor had exercised peremptory challenges on the basis of gender, the Appeals Court judged Aspen's prima facie burden by a more rigid standard than that established by Batson. Batson clearly established that Aspen was only required to make a "likelihood" showing at the final stage of the burden-shifting framework. 476 U.S. at 94 n. 18, 106 S. Ct. 1712; see also Purkett v. Elem, 514 U.S. 765, 768, 115 S. Ct. 1769, 131 L. Ed. 2d 834 (1995) (per curiam) ("It is not until the third step [of Batson] that the persuasiveness of the justification becomes relevant — the step in which the trial court determines whether the opponent of the strike has carried his burden of proving purposeful discrimination.") (emphasis in original).

The Commonwealth argues that the Appeals Court applied the appropriate standard because the opinion stated that when "[c]onfronted with a claim that a peremptory challenge is being used to exclude members of a discrete group, the judge must determine whether to draw the reasonable inference that peremptory challenges have been exercised on account of their group affiliation." Aspen, 758 N.E.2d at 166. The Commonwealth correctly states that an "inference" of discrimination is the Batson prima facie case standard. But it is apparent that the Appeals Court equated an "inference" of discrimination with a showing that gender was the "likely" reason that the prosecutor exercised her peremptory challenges.3 As explained above, Batson establishes that the "inference of discrimination" standard is not so demanding.4 See Wade v. Terhune, 202 F.3d 1190, 1196 (9th Cir.2000) (concluding that state appellate court misapplied Batson where it understood the "inference" standard to require the moving party to show a "strong likelihood" that peremptory challenges were being exercised on an improper basis).

That the Appeals Court applied a standard that was contrary to clearly established federal law does not automatically entitle the petitioner to relief, however. A writ of habeas corpus will issue only upon a showing that the petitioner "is in custody in violation of the Constitution or laws or treaties of the United States." 28 U.S.C. § 2254. This means that a petitioner must show that his underlying detention is unlawful and not just that the state court employed faulty reasoning in his case. See Bronshtein v. Horn, 404 F.3d 700, 724 (3d Cir.2005). As we have explained, habeas relief is available only if the petitioner demonstrates that "Supreme Court precedent requires an outcome contrary to that reached by the relevant state court." O'Brien v. Dubois, 145 F.3d 16, 24-25 (1st Cir.1998) (emphasis added), abrogated on other grounds, McCambridge v. Hall, 303 F.3d 24 (1st Cir.2002) (en banc). Accordingly, Aspen may not obtain habeas relief if, applying the correct Batson standard, he would still not prevail on his claim.

We consider de novo whether Aspen is entitled to relief under the correct Batson standard. See Bronshtein, 404 F.3d at 724 (considering de novo whether habeas petitioner established prima facie case under Batson where state appellate court had incorrectly applied Batson standard); Paulino v. Castro, 371 F.3d 1083, 1090 (9th Cir.2004) (similar). In so doing, we limit our review to facts gleaned from the state court record concerning jury selection at Aspen's trial. The trial court conducted a three-day voir dire. After potential jurors had been stricken for cause, the venire consisted of twenty-eight men and sixteen women, and the first panel of prospective jurors included twelve men and four women, with no distinction as to which four would be alternates.

Each side was allowed sixteen peremptory challenges. In the prosecutor's first round of peremptory challenges, she struck six men. These six men were replaced by four men and two women. The prosecutor struck one of the male replacements who was, in turn, replaced by another man. She then stated that she was content with the jury. At this point, the panel consisted of ten men and six women.

The trial court then permitted Aspen to exercise his peremptory challenges. He struck five women and two men. These seven potential jurors were replaced by three men and four women. Aspen then struck three more women. Two men and one woman were seated as replacements, and Aspen challenged the woman. A man was selected to replace the last stricken juror, and Aspen then stated that he was satisfied. At this point, the jury consisted of fourteen men and two women.

The prosecutor then struck five men. Two men and three women were seated as replacements, and the prosecutor challenged one of the men. A woman replaced the stricken man and the prosecutor stated that she was satisfied. The potential jury was then ten men and six women.

Aspen exercised his second round of peremptory challenges by eliminating three of the women. They were replaced by one woman and two men. Aspen challenged one of the men, who was replaced by another man. Aspen announced that he was satisfied, and the prosecutor agreed, even though she had three challenges remaining. As did the original panel, the final jury, with alternates, consisted of twelve men and four women.5

In total, the prosecutor had exercised thirteen peremptory challenges, all of which were against prospective male jurors. Aspen had exercised fifteen challenges, twelve of which were against prospective female jurors. He contends that the prosecutor's use of all of her challenges against men is sufficient evidence to establish a prima facie case under Batson and J.E.B.

We have cautioned that a party "who advances a Batson argument ordinarily should come forward with facts, not just numbers alone." Bergodere, 40 F.3d at 516.6 Here, Aspen has relied entirely on the number of prosecution strikes against men. It is at least questionable whether this evidence is adequate to surpass the prima facie hurdle. See United States v. Esparsen, 930 F.2d 1461, 1467 (10th Cir. 1991) ("By itself, the number of challenges used against members of a particular [group] is not sufficient to establish . . . a prima facie case."); Sangineto-Miranda, 859 F.2d at 1521 (stating that evidence "standing alone" that the government exercised all of its peremptory challenges against black members of the venire "does not raise the necessary inference of purposeful discrimination" to establish a prima facie case). But even assuming that Aspen could establish a prima facie case, it is clear that he cannot ultimately establish a Batson violation.

In considering Batson claims, courts examine both numeric and non-numeric forms of evidence. Relevant numeric evidence includes the percentage of strikes directed against members of a particular group, e.g., Paulino, 371 F.3d at 1091, the percentage of a particular group removed from the venire by the challenged strikes, e.g., Turner v. Marshall, 63 F.3d 807, 813 (9th Cir.1995), abrogated on other grounds, Tolbert v. Page, 182 F.3d 677, 684 (9th Cir.1999), and a comparison of the percentage of a group's representation in the venire to its representation on the jury, e.g., United States v. Sangineto-Miranda, 859 F.2d 1501, 1521-22 (6th Cir. 1988). Relevant non-numeric evidence includes the striking party's questions and statements during the voir dire, e.g., Brewer, 119 F.3d at 1004, whether the striking party had unused peremptory challenges through which he or she could have eliminated more members of the allegedly targeted group, e.g., United States v. Allison, 908 F.2d 1531, 1538 (11th Cir.1990), apparent non-discriminatory reasons for striking potential jurors based on their voir dire answers, e.g., United States v. Stephens, 421 F.3d 503, 515-16 (7th Cir.2005), and whether similarly situated jurors from outside the allegedly targeted group were permitted to serve, Boyd v. Newland, 467 F.3d 1139, 1148-50 (9th Cir.2006).

Looking first to the numeric evidence, the prosecutor's strikes were all aimed at men. But these challenges removed only forty-six percent of the males from the venire and even after the prosecutor's strikes (not all of which were used), men remained the dominant group. This fact at least tends to mitigate an inference of discrimination. See United States v. Brisk, 171 F.3d 514, 523 (7th Cir.1999) (holding no Batson violation established, in part, because even though two-thirds of strikes were directed against women, the strikes did not remove a significant percentage of the women in the venire).

More importantly, the strikes had little effect on the composition of the actual jury seated. Indeed, the jury agreed to by the prosecutor was seventy-five percent male, even though the original venire, after the strikes for cause, was only sixty-three percent male. See Escobar-de Jesus, 187 F.3d at 165 (no Batson violation based on strikes of African-American jurors where six African-Americans served on the jury); Sangineto-Miranda, 859 F.2d at 1522 (treating the fact that the "percentage of minority members [on] the ultimate jury is the same or greater" than in the venire as tending to "negate the inference of discrimination"). Additionally, the final jury agreed to by the prosecutor had the exact gender composition as the original panel faced by the prosecutor.

Aspen's emphasis on the raw number of strikes made against men loses force when this species of numeric evidence is considered in context. The context here was that Aspen struck most of the women members of the venire and that women constituted a minority of the venire from the outset. In fact, Aspen struck seventy-five percent of the prospective female jurors. The result was that the panels of potential jurors faced by the prosecutor were predominantly male. Indeed, most of the time, the potential juror panel from which the prosecutor made her strikes was between sixty-nine percent and eighty-eight percent male. Therefore, the raw number of strikes against men is not particularly telling in these circumstances because Aspen's use of peremptory challenges kept the number of men on the panel disproportionately high.7 See United States v. Ochoa-Vasquez, 428 F.3d 1015, 1047 n. 47 (11th Cir.2005) (stating that strikes made by the party raising a Batson claim are relevant to the extent that they explain how the composition of the final jury was formed).

The non-numeric evidence further supports the conclusion that a discriminatory motive did not infect the prosecutor's use of peremptory challenges. There are no voir dire statements or questions by the prosecutor that suggest a discriminatory motive. Moreover, seven of the jurors challenged by the prosecutor provided voir dire answers that could reasonably have been understood to mean that the jurors may have been inclined to acquit the defendant even if he committed the conduct alleged. In particular, certain male jurors questioned whether it was legally possible for a person to rape a member of his own family, or whether a woman could, as a factual matter, be raped without having provoked the attack.8 Finally, the prosecutor could have challenged three more men from the male-dominated panel finally selected but did not do so. In the circumstances at hand, this is an important indicum. See Bronshtein, 404 F.3d at 724 (finding relevant to Batson analysis that the prosecutor "passed up" the opportunity to strike other prospective jurors who were members of the allegedly targeted group).

In sum, the Appeals Court misapplied Batson and J.E.B. in analyzing Aspen's equal protection claim. But, even if the correct standard had been applied, Aspen would not have benefitted because the totality of the circumstances demonstrates that the prosecutor did not issue peremptory challenges on the basis of gender. Accordingly, Aspen's habeas petition was correctly denied.

Affirmed.