ORDER
A sua sponte call for a vote on rehearing this case en banc was made by an active judge of this court. The call failed to receive a majority of the votes of the non-reeused active judges. Fed. R.App. P. 35. The sua sponte en banc call is rejected.
O’SCANNLAIN, Circuit Judge, joined by BYBEE and BEA, Circuit Judges, dissenting from the denial of rehearing en banc:
This ease started out as an antitrust dispute between drug manufacturers and came to our court in the posture of an appeal from a simple juror selection ruling during trial. Sadly, it has morphed into a constitutional essay about equal protection and sexual orientation sweeping far beyond mere administration of Batson v. Kentucky, 476 U.S. 79, 106 S.Ct. 1712, 90 L.Ed.2d 69 (1986). The opinion’s unprecedented application of heightened scrutiny to a peremptory strike of a juror who was *991perceived to be gay bears significant implications for the same-sex marriage debate and for other laws that may give rise to distinctions based on sexual orientation.
Indeed, today’s opinion is the only appellate decision since United States v. Windsor, — U.S. -, 133 S.Ct. 2675, 186 L.Ed.2d 808 (2013), to hold that lower courts are “required by Windsor to apply heightened scrutiny to classifications based on sexual orientation for purposes of equal protection.” 740 F.3d 471, 484 (9th Cir.2014). Such holding is wrong, egregiously so. Because of the danger that district courts will be misled by the opinion’s sweeping misinterpretation of Windsor, it is most unfortunate that we denied rehearing en banc. I respectfully must dissent from our regrettable decision to do so.
I
The consequences of this opinion reach far beyond the treatment of Batson challenges. Consider the mischief it has already wrought. In the view of many, the application of heightened scrutiny in this case precludes the survival under the federal Constitution of long-standing laws treating marriage as the conjugal union between a man and a woman. See, e.g., Order at 4, Latta v. Otter, No. 14-35420 (9th Cir. May 20, 2014) (Hurwitz, J., con-eurring) (noting the opinion’s heightened scrutiny determination and concluding that “[g]iven that high burden, it is difficult to see how the ... appellants can make a ‘strong showing’ that they will prevail in their defense of [Idaho’s] measure[.]”). As a result of the decision, state officials charged with defending such laws in this court have already abdicated their task, invoking this case. See Mot. for Leave to Withdraw Brief, Sevcik v. Sandoval, No. 12-17668 (9th Cir. Feb. 10, 2014) (Nevada Governor and Attorney General); Resp. to Mot. for Summ. J. at 13-14, 34, Geiger v. Kitzhaber, No. 6:13-cv-01834 (D.Or. Mar. 18, 2014) (Oregon Governor and Attorney General). As a result, this is not just a Batson decision. It is perhaps all but this court’s last word on the question whether the Constitution will require States to recognize same — sex marriages as such — a question the Supreme Court in Windsor consciously decided not to answer.1
Moreover, as the first opinion among our sister circuits to apply heightened scrutiny to an equal protection claim in light of Windsor, it is likely a bellwether— or, perhaps, a premonitory harbinger. Every circuit court but our own — and the Second Circuit in Windsor, in a maneuver the Supreme Court declined to follow — has held to the contrary.2 Once again we are *992on the short end of a 10-2 split among our sister circuits. We have misled district courts within our Circuit by forcing them to join us out on this limb and have offered those around the country an invitation — to many of them, welcome — to do the same.3 See, e.g., Wolf v. Walker, No. 3:14-cv-00064, 2014 WL 2558444, at *24-29, 986 F.Supp.2d 982, 1004-1015 (W.D.Wis. June 6, 2014) (noting the panel’s opinion in concluding that heightened scrutiny applies); Whitewood v. Wolf, No. 1:13-cv-1861, 2014 WL 2058105, at *11 n. 10, 992 F.Supp.2d 410, 426 n. 10 (M.D.Pa. May 20, 2014) (same); Latta v. Otter, No. 1:13-cv-00482, 2014 WL 1909999 at *16-17, — F.Supp.2d -,-(D.Idaho May 13, 2014) (same); Henry v. Himes, No. 1:14-cv-129, 2014 WL 1418395, at *14, — F.Supp.2d -, - (S.D.Ohio Apr. 14, 2014) (same).4 The motivating question behind the opinion — whether Windsor subjects traditional marriage laws and others that may give rise to distinctions based on sexual orientation to heightened scrutiny — is truly one of exceptional importance, one that should not have been decided in the guise of a challenge to a peremptory strike during jury selection in an antitrust suit.
II
In concluding that heightened scrutiny applies to distinctions based on sexual orientation, the panel abandoned our circuit precedents, arrogating to itself, regrettably, the power of an en banc court. The panel’s reliance on Witt v. Dep’t of Air Force, 527 F.3d 806 (9th Cir.2008), reveals the driving force behind this opinion, the panel’s eagerness to reach the heightened scrutiny question. Our existing precedents had already settled that rational basis review, not heightened scrutiny, applies to this case — the panel had only to follow them. See Philips v. Perry, 106 F.3d 1420, 1425 (9th Cir.1997); High Tech Gays v. Defense Indus. Security Clearance Office, 895 F.2d 563, 574 (9th Cir.1990). Even when, in a misreading of Lawrence v. Texas, 539 U.S. 558, 123 S.Ct. 2472, 156 L.Ed.2d 508 (2003), we applied heightened scrutiny to distinctions based on sexual orientation in the substantive due process context, we declined to do so in the equal protection context because Lawrence is not an equal protection case. See Witt, 527 F.3d at 821. But today a three-judge panel, dissatisfied with the existing state of the law, casts off our precedents prescribing rational basis review of the juror selection claim in this case.
No three-judge panel has the power to overrule existing Ninth Circuit precedent. In an extremely narrow exception, our court requires invocation of “clearly irreconcilable” higher authority for one panel to overrule the prior decision of another. See Miller v. Gammie, 335 F.3d 889, 893 (9th Cir.2003) (en banc).5 Although the panel *993fails to explain why Windsor is clearly irreconcilable with our precedents, it stunningly determines that its sweeping reading of that case justifies one three-judge panel in overruling two binding prece-dential decisions. If our court wished to adopt the “new perspective” divined in Windsor, 133 S.Ct. at 2689, as its own, we should have reheard this case en banc to permit legitimate reconsideration of Philips and High Tech Gays. Instead, we have allowed one three-judge panel to cast aside our circuit precedents on its own in a display of judicial will that reflects little of the judgment we are charged to exercise.
Ill
A
And nothing in Windsor compels the application of heightened scrutiny to this juror selection challenge. Far less can Windsor be considered “clearly irreconcilable” with our rational basis precedents in a way that would justify such disregard for them. The Windsor dissenters considered the opinion to be “rootless and shifting,” noting crucially that it “does not even mention what had been the central question in this litigation: whether, under the Equal Protection Clause, laws restricting marriage to a man and a woman are reviewed for more than mere rationality,” 133 S.Ct. at 2705-06 (Scalia, J., dissenting). Even the majority in Windsor declined to adopt the reasoning of the Second Circuit, which had expressly applied heightened scrutiny to the equal protection claim in the case. See Windsor v. United States, 699 F.3d 169, 181 (2d Cir.2012).
The essential aspects of Windsor in fact cut against our own panel’s view. After the Court declined there to identify the applicable standard of review, it significantly limited its holding in a way the panel simply ignored. The Court explained that “[t]he class to which DOMA directs its restrictions and restraints are those persons who are joined in same-sex marriages made lawful by the State,” and that DOMA’s “purpose and effect [is] to disparage and to injure those whom the State, by its marriage laws, sought to protect in personhood and dignity.” 133 S.Ct. at 2695-96. The Windsor Court expressly identified the classification relevant to its inquiry, but the panel’s opinion simply invented a new classification, concluding that heightened scrutiny applies any time “state action discriminates on the basis of sexual orientation.” 740 F.3d at 483. And the panel prefers entirely to disregard Windsors closing instruction: “This opinion and its holding are confined to those lawful marriages” that States like New York had chosen to recognize. 133 S.Ct. at 2695-96. As the Chief Justice observed, “[Windsor’s ] analysis leads no further.” Id. (Roberts, C.J., dissenting). An opinion so limited compels “not only our usual obedience, but also our self-conscious restraint.” Witt, 548 F.3d at 1275 (O’Scannlain, J., dissenting from denial of rehearing en banc).
*994B
The panel grasps at Witt — a substantive due process case — as the best straw possible to justify its departure from our equal protection precedents.6 But even the analysis expressly prescribed by Witt cannot support today’s conclusion that heightened scrutiny applies to distinctions based on sexual orientation in the equal protection context. Witt divined from three entrails that Lawrence prescribed heightened scrutiny in the substantive due process context: first, Lawrence declined to examine hypothetical justifications for the law; second, Lawrence required a “legitimate” justification for the law; and third Lawrence cited substantive due process cases applying heightened scrutiny. 527 F.3d at 817. But Windsor reflects none of the viscera Witt considered to be indicia of heightened scrutiny.
Indeed, the Witt factors reveal only rational basis review at work in Windsor. To employ rational basis review in the equal protection context did not require Windsor to consider hypothetical justifications for Section 8 of DOMA. See Romer v. Evans, 517 U.S. 620, 685, 116 S.Ct. 1620, 134 L.Ed.2d 855 (1996). In declaring that Section to be motivated by no “legitimate” purpose, Windsor only applies rational basis review in the same way that Romer reviewed Colorado’s Amendment 2 for rational basis. See id And, unlike Lawrence, Windsor relied on rational basis cases: Romer, a rational basis case by the panel’s own admission, 740 F.3d at 483; U.S. Dept. of Agriculture v. Moreno, 413 U.S. 528, 93 S.Ct. 2821, 37 L.Ed.2d 782 (1973), a rational basis case according to Lawrence, see 539 U.S. at 580, 123 S.Ct. 2472; and Lawrence itself, which the panel admits “declined to address equal protection,” 740 F.3d at 480.
In a final flourish of legerdemain, the panel pleads that Windsor cites Lawrence and therefore must be applying something other than rational basis review because Lawrence “is a heightened scrutiny case.” Id. at 483. But Lawrence is not a “heightened scrutiny” ease, but rather a substantive due process case, and for that reason cannot govern the equal protection analysis here. Even Witt acknowledged as much, see 527 F.3d at 821—but this panel is not so modest. Its opinion offers no justification for such an extraordinarily expansive reading of Windsor in light of these contrary indications. Indeed, there can be none.
IV
Recall that this appeal started out as a Batson case about striking one juror allegedly based on perceived sexual orientation. Without even acknowledging the consequences of its decision, the panel has produced an opinion with far-reaching — and mischievous — consequences, for the same-sex marriage debate and for the many other laws that may give rise to distinctions based on sexual orientation, without waiting for appropriate guidance from the Supreme Court.7 And in doing so, it plain*995ly misread Windsor, abandoned our own equal protection precedents, and disregarded our procedures for departing from settled constitutional doctrine.
I respectfully dissent from our regrettable failure to rehear this case en banc. While this case may end here — neither party is likely to seek certiorari given that neither party urged en banc reconsideration of the applicable standard of review— reliance on the panel’s analysis as an example of anything more than an exercise of raw judicial will would be most unwise.
. States, of course, remain entirely free to legislate changes to the definition of marriage to include same-sex unions, and eleven states have done so, including two within our Circuit. See, e.g., Haw.Rev.Stat. §§ 572-1-572-1.8 (2013); Wash. Rev.Code § 26.04.020 (2012). Interestingly, Oregon advocates of same-sex marriage dropped a planned ballot measure in light of the Oregon district court decision in Geiger holding Oregon's law unconstitutional. See Jeff Mapes, Gay Marriage: Supreme Court declines to halt same-sex marriages in Oregon, The Oregonian, June 4, 2014, http://www.oregonlive.com/mapes/ index, ssi/2014/06/gay_marriage_supreme_ court_dec.html (last visited June 11, 2014).
. See Massachusetts v. Dep’t of Health and Human Servs., 682 F.3d 1, 9-10 (1st Cir.2012) (applying rational basis review); Price-Cornelison v. Brooks, 524 F.3d 1103, 1113 n. 9 (10th Cir.2008) (same); Scarbrough v. Morgan Cnty. Bd. of Educ., 470 F.3d 250, 261 (6th Cir.2006) (same); Citizens for Equal Protection v. Bruning, 455 F.3d 859, 866 (8th Cir.2006) (same); Johnson v. Johnson, 385 F.3d 503, 532 (5th Cir.2004) (same); Lofton v. Sec’y of Dep't of Children and Family Servs., 358 F.3d 804, 818 (11th Cir.2004) (en banc) (same); Nabozny v. Podlesny, 92 F.3d 446, 458 (7th Cir.1996) (same); Thomasson v. Perry, 80 F.3d 915, 928 (4th Cir.1996) (same); Steffan v. Perry, 41 F.3d 677, 684-85 (D.C.Cir.1994) (same); Woodward v. United States, 871 F.2d 1068, 1076 (Fed.Cir.1989) (same).
. Of course some such courts have followed a different line of reasoning than the panel’s after noting, to their credit, that the opinion was not final. See, e.g., Geiger v. Kitzhaber, No. 6:13-cv-01834, 994 F.Supp.2d 1128, 1140-42, 2014 WL 2054264, at *8-9 (D.Or. May 19, 2014) (McShane, J.).
. See also De Leon v. Perry, 975 F.Supp.2d 632, 652 (W.D.Tex.2014) (considering that panel’s application of heightened scrutiny increased likelihood of success on the merits at injunction stage); Bostic v. Rainey, 970 F.Supp.2d 456, 482 n. 16 (E.D.Va.2014) (concluding law failed rational basis review but noting inclination to apply heightened scrutiny based on the panel’s opinion); Bourke v. Beshear, No. 3:13-cv-750, 996 F.Supp.2d 542, 547-49, 2014 WL 556729, at *4 (W.D.Ky. Feb. 12, 2014) (noting panel’s opinion in discussing likely future application of heightened scrutiny).
.Our three-judge panels frequently reject invitations to overrule binding circuit precedent. See, e.g., United States v. Albino-Loe, 747 F.3d 1206, 1212-14 (9th Cir.2014); United States v. Dunn, 728 F.3d 1151, 1156-58 (9th Cir.2013); Johnson v. Bay Area Rapid Transit Dist., 724 F.3d 1159, 1171 (9th Cir.2013); United States v. Green, 722 F.3d 1146, 1149-51 (9th Cir.2013). They often do so *993even when the panel believes that binding precedent is clearly wrong. See, e.g., United States v. Chandler, 743 F.3d 648, 661-62 (9th Cir.2014) (Bybee, J., concurring); Al-Ramahi v. Holder, 725 F.3d 1133, 1138 n. 2 (9th Cir.2013); United States v. Hernandez-Estrada, 704 F.3d 1015, 1025-26 (9th Cir.2012) (Kozinski, C.J., concurring), rev’d en banc, 749 F.3d 1154 (9th Cir.2014). What is more, sitting en banc, we have chastised a three-judge panel for presuming to overrule binding circuit precedent when it is not "clearly irreconcilable” with intervening higher authority. See United States v. Contreras, 593 F.3d 1135, 1136 (9th Cir.2010) (en banc). As explained below, Windsor simply is not "clearly irreconcilable” with Philips and High Tech Gays. As a result, the panel in this case had only two legitimate options — to defer to binding circuit precedent or to issue a sua sponte en banc call. It did neither.
. Six of us argued that Witt was wrong when it was decided. See 548 F.3d 1264, 1265 (9th Cir.2008) (O’Scannlain, I, joined by Bea, M. Smith, and N.R. Smith, JJ., dissenting from denial of rehearing en banc); id. at 1276 (Kleinfeld, J. joined by Bea, J., dissenting from denial of rehearing en banc); id. at 1280 (Kozinski, C.J., joined by Bea and M. Smith, JJ., dissenting from denial of rehearing en banc). Nevertheless, it is the law of our Circuit and should be followed according to its terms.
. Even when the Supreme Court has extended Batson, it has done so only when the classification at issue had already been subjected to heightened scrutiny. See J.E.B. v. Alabama, 511 U.S. 127, 135-36, 114 S.Ct. 1419, 128 L.Ed.2d 89 (1994). It has not extended the Batson analysis, as today's decision does, to classifications never previously regarded with constitutional suspicion.