with whom Circuit Judge O’SCANNLAIN joins, dissenting:
In striking down Proposition 100, the majority sets aside the policy judgment of the Arizona legislature and nearly 80 percent of Arizona’s voting electorate, telling the State it really doesn’t have an illegal immigration problem adversely affecting its criminal justice system. But Arizonans thought Proposition 100 was the solution to an ineffective bail system that was letting too many illegal aliens avoid answering for their serious felony charges. They were concerned that these offenders, who often lack community ties, too often skirt justice by fleeing the state or the country before trial. Plaintiffs-Appellants Lopez-Valenzuela and Castro-Armenta are good examples. Between the two of them they were charged with aggravated assault with a deadly weapon, kidnapping, theft by extortion, assisting a criminal syndicate, and transportation of a dangerous drug.
Today’s holding leaves Arizona nowhere to turn. For years before Arizona passed Proposition 100, it tried to assess flight risk based on family ties, employment, and length of residency in the community on an individualized basis. Immigration status was assuredly a critical consideration in this assessment, even well before Proposition 100’s enactment. The majority ignores reality to suggest otherwise. Yet despite these individualized flight-risk assessments, the Maricopa County Attorney explained in 2006 that Arizona still had a “tremendous problem with illegal immigrants coming into the state, committing serious crimes, and then absconding, and not facing trial for their crimes.” Lou Dobbs Tonight (CNN television broadcast Oct. 13, 2006). Faced with this continuing problem, Arizona took a logical next step *799by denying bail to illegal aliens who commit serious felony offenses. Because Proposition 100 is not excessive in relation to Arizona’s compelling regulatory interest in ensuring that illegal aliens who commit serious felony offenses stand trial, I respectfully dissent.
I
A
The majority’s errors begin with its substantive due process framework. It applies what it calls “Salerno’s heightened scrutiny” to Proposition 100. Maj. Op. at 780; see United States v. Salerno, 481 U.S. 739, 107 S.Ct. 2095, 95 L.Ed.2d 697 (1987). Under this test, it says, the law survives only when it is “narrowly tailored to serve a compelling state interest.” Maj. Op. at 781. This is strict scrutiny. See, e.g., Green v. City of Tucson, 340 F.3d 891, 896 (9th Cir.2003) (Fisher, J.) (Under “strict scrutiny,” “the statute will be upheld only if the state can show that the statute is narrowly drawn to serve a compelling state interest.”). However, you will not find strict scrutiny mentioned, let alone applied, anywhere in Salerno.1
The majority mistakenly applies strict scrutiny by misreading Salerno, which addressed the consideration of danger to the community in bail determinations, as holding that the Bail Reform Act implicated a fundamental liberty interest. Salerno held just the opposite. The Court acknowledged that liberty, in its broadest sense, is a fundamental right. Salerno, 481 U.S. at 750, 107 S.Ct. 2095. It explained, however, that the right asserted in Salerno was more limited; it was the right to bail after “the Government proves by clear and convincing evidence that an arrestee presents an identified and articulable threat to an individual or the community.” Id. at 751, 107 S.Ct. 2095. As to that right, the Court said: “we cannot categorically state that pretrial detention ‘offends some principle of justice so rooted in the traditions and conscience of our people as to be ranked as fundamental.’ ” Id. (quoting Snyder v. Massachusetts, 291 U.S. 97, 105, 54 S.Ct. 330, 78 L.Ed. 674 (1934)). Thus, rather than holding that liberty is a fundamental right in all instances, the Court held that the liberty right “under the[ ] circumstances” of the Bail Reform Act was not fundamental. Id.
■ The Court’s narrow construction of the liberty interest at stake in Salerno is consistent with its instruction that “ ‘[substantive due process’ analysis must begin with a careful description of the asserted right, for the doctrine of judicial self-restraint requires us to exercise the utmost care whenever we are asked to break new ground in this field.” Reno v. Flores, 507 U.S. 292, 302, 113 S.Ct. 1439, 123 L.Ed.2d 1 (1993) (citations and internal quotations omitted). Thus, we must “carefully formulat[e]” the liberty interest, Washington v. Glucksberg, 521 U.S. 702, 722, 117 S.Ct. 2258, 138 L.Ed.2d 772 (1997), and define it “narrowly,” Flores v. Meese, 913 F.2d 1315, 1330 (9th Cir.1990) (emphasis supplied). Only then can we determine whether the asserted right is fundamental.
The majority ignores that instruction and concludes that the Bail Reform Act *800and Proposition 100 impinge on some generalized fundamental liberty right. In fact, the majority makes no effort to even define the right at stake. But it isn’t simply “liberty” as the majority would have it. Rather, Proposition 100 implicates an arrestee’s alleged right to bail where the proof is evident or the presumption great that the arrestee committed a serious felony offense,2 and there is probable cause to believe the arrestee has entered or remained in the United States illegally. That is, the right is not merely to be free from detention, but to be free from detention “under these circumstances.” Salerno, 481 U.S. at 751, 107 S.Ct. 2095.
In my view, this asserted right is not “so rooted in the traditions and conscience of our people as to be ranked fundamental.” Id.; see, e.g., Demore v. Kim, 538 U.S. 510, 522, 123 S.Ct. 1708,155 L.Ed.2d 724 (2003) (stating, in approving the detention of aliens awaiting deportation, that “this Court has firmly and repeatedly endorsed the proposition that Congress may make rules as to aliens that would be unacceptable if applied to citizens”); Carlson v. Landon, 342 U.S. 524, 545, 72 S.Ct. 525, 96 L.Ed. 547 (1952) (“The bail clause was lifted with slight changes from the English Bill of Rights Act. In England that clause has never been thought to accord a right to bail in all cases[.]”); United States v. Edwards, 430 A.2d 1321, 1327 (D.C.App. 1981) (“[A] fundamental right to bail was not universal among the colonies or among the early states;” “the language of several state constitutions explicitly limiting the power of the judiciary to set excessive bail negates any suggestion that the excessive bail clause was intended to restrict the definition of bailable offenses by the legislature.”), cert, denied, 455 U.S. 1022, 102 S.Ct. 1721, 72 L.Ed.2d 141 (1982). Because the right is not fundamental, strict scrutiny does not apply.3
B
Whatever substantive due process standard Salerno provides — and I concede that *801there is some ambiguity — Proposition 100 meets it. Nobody disputes that “Arizona has a compelling interest in ensuring that persons accused of serious crimes, including undocumented immigrants, are available for trial.” Maj. Op. at 782. And Proposition 100 is “careful[ly] delineat[ed]” and “carefully limited” — it is even “narrowly focuse[d].” Salerno, 481 U.S. at 750-51, 755, 107 S.Ct. 2095. Several fundamental errors in the majority’s opinion lead it to conclude otherwise.
First, by ignoring all evidence to the contrary, the majority concludes that “unmanageable flight risk posed by undocumented immigrants ... has not been shown to exist.” Maj. Op. at 791. Tell that to the Maricopa County Attorney who, from his vantage on the front line just a month before the Proposition 100 vote, thought flight risk among illegal aliens in Arizona was “a tremendous problem.” Lou Dobbs Tonight (CNN television broadcast Oct. 13, 2006) (emphasis supplied).4 The majority insists that the record does not substantiate this claim, but the claim — by the Maricopa County Attorney no less — is part of the record.5 It is one thing for my colleagues to declare that Proposition 100 is an excessive measure to address Arizona’s flight problem; reasonable minds can disagree. It is quite another for an Article III court to tell Arizona, based on this record and considering the majority vote of the Arizona legislature and electorate in favor of Proposition 100, that its perceived problem is not really a problem. Maj. Op. at 791. Even the Arizona courts have concluded that “Proposition 100 reflects that [the Arizona] electorate and Legislature perceived pretrial detention as a potential solution to a pressing societal problem.” Hernandez v. Lynch, 216 Ariz. 469, 167 P.3d 1264, 1274 (Ariz.Ct.App.2007) (internal quotations and citations omitted). They are the ones who have to live with it.
Second, the majority takes issue with Proposition 100 not being a widely shared legislative judgment because few states categorically deny bail for illegal aliens. Maj. Op. at 787-88. While factually true, perspective here is critical. As an initial matter, novelty alone does not render a law unconstitutional. See Ewing v. California, 538 U.S. 11, 24, 123 S.Ct. 1179, 155 L.Ed.2d 108 (2003) (holding that defendant’s sentence under California’s novel “three strikes” law was not unconstitutional). In Ewing, the Supreme Court explained that “[t]hough three- strikes laws may be relatively new, our tradition of deferring to state legislatures in making and implementing such important policy decisions is longstanding.” Id. My colleagues give no deference to the policy judgment of the Arizona legislature or the democratic views of the electorate.
Moreover, wé cannot judge this case ■with blinders on. Arizona faces unique challenges as one of four states bordering Mexico. Indeed, “Arizona bears many of the consequences of unlawful immigra*802tion.... Unauthorized aliens who remain in the State comprise, by one estimate, almost six percent of the population.” Arizona v. United States, — U.S. -, 132 S.Ct. 2492, 2500, 183 L.Ed.2d 351 (2012). “[I]n the State’s most populous county, these aliens are reported to be responsible for a disproportionate share of serious crime.” Id. I find it hardly surprising that other states have not enacted their own Proposition 100 laws. Kansas, for example, may have its share of illegal immigrants, but certainly not to the extent of Arizona. More importantly, only four states — Arizona among them — provide serious felony offenders with such a quick and convenient route into Mexico.6 Thus, while Proposition 100 may be relatively unique among the fifty states, that’s nothing more than a reflection of Arizona’s unique flight risk problem and geography.
Third, the majority mistakenly demands “findings, studies, statistics or other evidence ... showing that undocumented immigrants as a group pose either an unmanageable flight risk or a significantly greater flight risk than lawful residents.” Maj. Op. at 783. Arizona is not required to support its legislative judgment with empirical studies. This is a slippery slope on which the majority is quick to tread, and one which threatens the delicate balance between the judiciary and the people we serve.
While we have imposed empirical fact-finding requirements in a few contexts,7 those are the exceptions to the rule. See, e.g., Gonzales v. Raich, 545 U.S. 1, 21, 125 S.Ct. 2195, 162 L.Ed.2d 1 (2005) (“[W]e have never required Congress to make particularized findings in order to legislate, absent a special concern such as the protection of free speech.” (citations omitted)). Indeed, the Supreme Court has explained that “[wjhile Congressional findings are certainly helpful ... the absence of particularized findings does not call into question Congress’ authority to legislate.” Id. We have never held that a state legislature — let alone a state’s voting electorate, as here — must prove its flight-risk concerns with empirical data. Nor have we been instructed to require supporting data in connection with pretrial detention schemes — not in Salerno and not in Demore,8 The majority’s empirical data re*803quirement, imposed today by judicial fiat, is particularly inappropriate and dangerous in this case, where Arizona voters passed Proposition 100 by a whopping 78 percent. See Nixon, 528 U.S. at 394, 120 S.Ct. 897 (observing, even under the heightened requirements of First Amendment challenges, that a 74-percent statewide vote passage rate “certainly attested to the perception” that the challenged law was necessary to remedy the state’s identified concern).
Finally, the majority errs by effectively precluding the use of irrebuttable presumptions in the bail context. Proposition 100 “plainly is not carefully limited,” says the majority, “because it employs an overbroad, irrebuttable presumption rather than an individualized hearing” to assess flight risk. Maj. Op. at 784. That conclusion relies on the premise that individualized hearings would solve Arizona’s troubles. In Demore, the Supreme Court upheld a categorical' bail prohibition in part because individualized hearings had proven unsuccessful. 538 U.S. at 528, 123 S.Ct. 1708. Here, like in Demore, Arizona has already tried determining flight risk on an individualized basis, considering factors such as family ties, employment, and length of residency in the community. If you believe the record — and Plaintiffs-Appellants have given us no reason not to — the pressing problem of illegal aliens absconding before trial survived these individualized hearings. Thus, Proposition 100’s lack of individualized flight risk determinations cannot render its irrebuttable presumption excessive. See Demore, 538 U.S. at 528, 123 S.Ct. 1708.
Applying well-established substantive due process principles to this record reveals that Proposition 100 — while distasteful to some — survives substantive due process review.
II
The majority also errs by finding Proposition 100 impermissibly punitive, although it purports to leave untouched the district court’s factual conclusion that Proposition 100 was not intended to impose punitive restrictions. Salerno provides the two-part test to determine whether the pretrial detention scheme at issue constitutes “impermissible punishment or permissible regulation.” Salerno, 481 U.S. at 747, 107 S.Ct. 2095. First, we look to legislative intent to determine whether the legislature “expressly intended to impose punitive restrictions.” Id. Second, we ask “whether an alternative purpose to which the restriction may rationally be connected is assignable for it, and whether it appears excessive in relation to the alternative purpose assigned to it.” Id (alterations and internal quotation marks omitted). In other words, in the absence of an express punitive intent, “the punitive/regulatory distinction turns on” whether the pretrial detention scheme is “excessive in relation to the regulatory goal [the legislature] sought to achieve.” Id.
Reviewing the record as a whole, including legislative history and election-related materials, it is clear that Arizona legislators and voters passed Proposition 100 as a regulatory measure to ensure illegal aliens who commit serious felony offenses stand trial. But don’t just take my word for it, see Lopez-Valenzuela v. Cnty. of Maricopa, 719 F.3d 1054,1060-61 (9th Cir. 2013) (reviewing the record de novo); District Judge Susan Bolton and the Arizona Court of Appeals conducted their own in*804dependent review of the record and reached the same conclusion, Lopez-VaImzuela v. Maricopa Cnty., No. 08-00660, at 10 (D.Ariz. March 29, 2011) (“Having reviewed the voluminous evidence submitted in this case, the Court finds that the record as a whole does not support a finding that Proposition 100 was motivated by an improper punitive purpose.”); Hernandez, 167 P.3d at 1273 (“[W]e hold that the purpose behind Proposition 100 was not to punish illegal aliens, but to prevent them from fleeing before trial.”). The majority wisely declines to disturb these findings. Maj. Op. at 790 (“[W]e see no reason to revisit those conclusions on appeal.”). Yet my colleagues spend several pages attempting to impeach them.
Turning to the second step, we all agree that “Arizona has a compelling interest in ensuring that persons accused of serious crimes, including undocumented immigrants, are available for trial.” Maj. Op. at 782. But the majority concludes that Proposition 100 is excessive in relation to its purpose. Maj. Op. at 791. It isn’t.
Proposition 100, like the Bail Reform Act, “carefully limits the circumstances under which detention may be sought.” Salerno, 481 U.S. at 747,107 S.Ct. 2095. Two threshold requirements must be met before an individual can be denied bail pursuant to Proposition 100. First, the court must find, at the arrestee’s initial appearance, that there is probable cause to believe that the arrestee has entered or remained in the United States illegally. Ariz. R.Crim. P. 7.2(b). But more than that, the court must also find that the proof is evident or the presumption great that the individual committed a “serious felony offense.” Id.: Ariz. Const, art. II § 22(4).
An arrestee is also able to challenge this initial determination by “mov[ing] for reexamination of the conditions of release.” Ariz. R.Crim. P. 7.4(b). Upon such a motion — and whether or not the arrestee alleges new facts — a hearing must be held on the record “as soon as practicable but not later than seven days after filing of the motion.” Id. These procedural and substantive protections make Proposition 100 far from a “scattershot attempt to incapacitate those who are merely suspected” of being in the country illegally or of committing a serious felony offense. Salerno, 481 U.S. at 750, 107 S.Ct. 2095 (emphasis supplied).
For years, Arizona tried making individualized flight-risk determinations for aliens alleged to have committed serious felony offenses. That system didn’t solve Arizona’s particularly acute flight-risk problem. Its voters then overwhelmingly approved Proposition 100 as a measured response in light of the state’s prior efforts. How can that be excessive?
Ill
The majority does not reach Plaintiffs-Appellants’ remaining claims because it doesn’t have to. However, I remain convinced that the district court properly awarded Defendants-Appellees summary judgment on those claims as well. I stand by the reasons set forth in that opinion. See Lopez-Valenzuela, 719 F.3d at 1064-73. For all these reasons, I respectfully dissent.
. If you find it peculiar that the Supreme Court would apply strict scrutiny without telling us, you have good reason. It certainly knows how to say that it is applying strict scrutiny when it does so. See, e.g., Johnson v. California, 543 U'.S. 499, 507, 125 S.Ct. 1141, 160 L.Ed.2d 949 (2005) (explaining that it is applying "strict scrutiny”); Bush v. Vera, 517 U.S. 952, 976, 116 S.Ct. 1941, 135 L.Ed.2d 248 (1996) (same); Miller v. Johnson, 515 U.S. 900, 920, 115 S.Ct. 2475, 132 L.Ed.2d 762 (1995) (same); Bernal v. Fainter, 467 U.S. 216, 227, 104 S.Ct. 2312, 81 L.Ed.2d 175 (1984) (same).
. "Serious felony offense” is statutorily defined as "any class 1, 2, 3 or 4 felony or [aggravated driving under the influence].” Ariz.Rev.Stat. Ann. § 13-3961(A)(5)(b).
. Given that the Court in Salerno said that the right at issue was not fundamental, id. at 751, 107 S.Ct. 2095, and never applied strict scrutiny, how does the majority divine that strict scrutiny applies to Proposition 100? Apparently from a couple of fractured Supreme Court opinions that hint, but do not hold, that Salerno may have meant something it never said. See Flores, 507 U.S. at 302, 113 S.Ct. 1439 (including Salerno in a string cite about defining "substantive due process”); Foucha v. Louisiana, 504 U.S. 71, 80-83, 112 S.Ct. 1780, 118 L.Ed.2d 437 (1992) (discussing Salerno but not elucidating its standard). I am not convinced.
Flores cited Salerno merely for the proposition that there is "a substantive component” to the Constitution’s due process guarantee. Flores, 507 U.S. at 302-03, 113 S.Ct. 1439. Surely the majority would not suggest that the other cases cited in the same string cite— Collins v. City of Harker Heights, Tex., 503 U.S. 115, 112 S.Ct. 1061, 117 L.Ed.2d 261 (1992), and Bowers v. Hardwick, 478 U.S. 186, 106 S.Ct. 2841, 92 L.Ed.2d 140 (1986) — applied strict scrutiny to any fundamental right. Moreover, the page of Salerno cited simply defines “substantive due process”; it does not mention fundamental rights or strict scrutiny. See 481 U.S. at 746, 107 S.Ct. 2095. Similarly, Foucha cites the same page of Salerno for the same limited proposition, surrounded by citations to cases that merely describe substantive due process. 504 U.S. at 80, 112 S.Ct. 1780 (citing Zinermon v. Burch, 494 U.S. 113, 110 S.Ct. 975, 108 L.Ed.2d 100 (1990); Salerno, 481 U.S. at 746, 107 S.Ct. 2095; and Daniels v. Williams, 474 U.S. 327, 106 S.Ct. 662, 88 L.Ed.2d 662 (1986)). True, Foucha discusses Salerno at greater length, but for the circumstances in which a person can be detained because he poses "a danger to others or to the community,” not because of flight risk. Id. at 81, 112 S.Ct. 1780.
. The prosecutor also testified before the Arizona Senate Judiciary Committee, explaining that there were “numerous examples of serious and violent criminals that [the] Maricopa County Attorney's Office has prosecuted in the past that have escaped justice because they have [] slipped back across the border after they’ve been released.” Senate Judiciary Committee Meeting on H.B. 2389, Mar. 28, 2005, 47th Leg., 1st Regular Sess. (Ariz.2005).
. The majority responds to the Maricopa County Attorney’s statements by attempting to discredit his testimony through extra-record evidence not cited to or relied on by any party. Maj. Op. at 783-84, n. 6. Of course, the majority cannot so easily impeach the four out of five Arizona voters who must live with the problem the majority concludes does not exist.
. The majority’s argument also ignores a host of practical concerns facing border states like Arizona, such as the time and expense in trying to apprehend felony fugitives in Mexico and elsewhere, and the cumbersome and lengthy extradition process required to bring fugitives back to Arizona to face justice.
. The most prevalent example is in- the context of content-based regulations under the First Amendment. See Video Software Dealers Ass'n v. Schwarzenegger, 556 F.3d 950, 962 (9th Cir.2009). But even in the First Amendment context, the Supreme Court has explained that ''[t]he quantum of empirical evidence needed to satisfy heightened judicial scrutiny of legislative judgments will vary up or down with the novelty and plausibility of the justification raised.” Nixon v. Shrink Mo. Gov’t PAC, 528 U.S. 377, 378, 120 S.Ct. 897, 145 L.Ed.2d 886 (2000). I find it neither novel nor implausible that a border state struggles to prevent serious felony offenders who are in the country illegally from fleeing before trial. The majority decrees that Arizonans will just have to live with it.
.In Salerno, the Supreme Court merely mentioned, in a single sentence and without any hint of a requirement, that Congress made findings in enacting the Bail Reform Act. 481 U.S. at 750, 107 S.Ct. 2095. In Demore, the respondent argued that Congress had no evidence that individualized bond hearings would be ineffective, but the Court observed that Congress had evidence suggesting that permitting discretionary release of aliens pending their removal hearings would lead to large numbers of deportable criminal aliens absconding. 538 U.S. at 528, 123 S.Ct. 1708. Demore did not hold that congressional fact-finding was required; but even if we read in such a requirement, Arizona's legislators met that requirement by taking testimony to that *803effect. See Senate Judiciary Committee Meeting on H.B. 2389, Mar. 28, 2005, 47th Leg., lst Regular Sess. (Ariz.2005).