IMAGINARY IMAGES, INCORPORATED, d/b/a Paper Moon; BTF3, L.L.C., d/b/a Paper Moon; Papermoon-Springfield, Incorporated, d/b/a Paper Moon, Plaintiffs-Appellants, v. Pamela O'Berry EVANS, in her official capacity as Chair of the Virginia Alcohol Beverage Control Board; Susan R. Swecker, in her official capacity as Member of the Virginia Alcohol Control Board; Esther H. Vassar, in her official capacity as Member of the Virginia Alcohol Control Board, Defendants-Appellees, 612 F.3d 736

OPINION WILKINSON, Circuit Judge: Plaintiffs are three nightclubs where women give erotic dance performances wearing only g-strings and pasties. The clubs brought First Amendment, vagueness, and overbreadth challenges to Virginia’s alcohol licensing program, which allows the clubs to serve beer and wine but not mixed beverages. Under the standard of intermediate scrutiny applicable to policies aimed at the harmful secondary effects of sexually oriented entertainment, Virginia’s policy passes constitutional muster.

The public interest served by the policy is substantial, the restriction on the clubs mild and the burden on First Amendment values slight. Moreover, legislatures must have some leeway to draw a regulatory middle ground and Virginia’s is a policy of moderation. Judicial invalidation of carefully drawn distinctions risks chasing lawmakers from the paths of compromise and into absolutes. We thus decline to overturn the classifications here, and accordingly affirm the judgment of the district court. I.

The sale and consumption of alcohol within the Commonwealth of Virginia is governed by the comprehensive regulatory scheme established by the Alcoholic Beverage Control (“ABC”) Act, Va.Code §§ 4.1-100, et seq., and by regulations adopted by the ABC Board, the regulatory body created by the Act. See Va.Code §§ 4.1-101, - 103. Under this regime, establishments where performers offer striptease routines may obtain licenses to sell beer, wine, or both. Such facilities are not eligible, however, for mixed beverage licenses, which permit the sale of distilled spirits. See Va.Code §§ 4.1-226(2)(i), -325(12), (13); 3 Va.

Admin. Code § 5-50-140. The current shape of these provisions stems in part from earlier litigation. In Giovani Carandola, Ltd. v. Bason, 303 F.3d 507 (4th Cir.2002) (“Carandola /”), this court struck down as overbroad certain North Carolina limitations on the availability of alcohol at establishments hosting sexually oriented performances. The offending provisions were then amended and the court upheld the revised scheme against overbreadth and vagueness challenges. See Giovani Carandola, Ltd. v. Fox, 470 F.3d 1074 (4th Cir.2006) (“Carandola II”).

At the time, the Virginia ABC statutes and relevant ABC regulation used language similar to that which Carandola I had invalidated, leading to an injunction in 2007 against enforcement of certain portions of the Virginia program. See Norfolk 302, LLC v. Vassar, 524 F.Supp.2d 728, 742 (E.D.Va.2007). The Virginia General Assembly promptly amended the challenged statutes to bring them into compliance and the ABC Board similarly amended its regulation, after which this court issued an order dismissing the ABC Board’s pending appeal and vacating the injunction as moot.

During the period when Virginia’s rules were suspended, mixed beverage licenses were issued to the plaintiffs in this case. Plaintiffs are three Virginia nightclubs belonging to the Papermoon chain — two in Richmond and one in Springfield — where dancers perform wearing only g-strings and pasties. In June 2008, with the revised licensing program about to take effect and their mixed beverage licenses in jeopardy, plaintiffs, whom we shall refer to as Papermoon, sued the ABC Board’s members to block enforcement.

Paper-moon argued that the scheme violated the First Amendment, was unconstitutionally vague, and was facially overbroad. An evidentiary hearing was held a few months later at which the ABC Board offered the testimony of W. Curtis Cole-burn, its chief operating officer. Coleburn testified that he and the Board had reviewed at least forty-two studies and numerous cases dealing with the negative effects on the surrounding community of sexually oriented businesses.

He explained that Virginia’s decision to limit establishments offering sexually oriented entertainment to beer and wine reflected the fact that distilled spirits more readily lead to intoxication because of their higher alcohol content. He also stated that Virginia’s policy had been modified to incorporate the teachings of the Carandola decisions. In response, Papermoon offered various evidence meant to show that its clubs did not produce secondary effects. This consisted chiefly of testimony from its expert, Professor Daniel Linz of the University of California at Santa Barbara.

Linz explained that he had reviewed crime data for the Papermoon locations and found that there was no increase in crime near the clubs after they obtained mixed beverage licenses and that sexually oriented businesses in Richmond generally were not “hot spots” for crime. In December 2008, the district court rejected the bulk of Papermoon’s claims, holding, with exceptions not relevant here, that Virginia’s policy prohibiting distilled spirits at establishments like the Paper-moon clubs was constitutional. See Imaginary Images, Inc. v. Evans, 593 F.Supp.2d 848, 863 (E.D.Va.2008). Papermoon now appeals. II.

Although it is a far cry from political speech, “nude dancing is not without its First Amendment protections.” Schad v. Borough of Mount Ephraim, 452 U.S. 61, 66, 101 S.Ct. 2176, 68 L.Ed.2d 671 (1981). Regulations of sexually oriented entertainment “receive intermediate scrutiny if they are not premised on a desire to suppress the content of such entertainment, but rather to address the harmful secondary effects” it produces — higher crime rates, lower property values, and unwanted interactions between patrons and entertainers such as public sexual conduct, sexual assault, and prostitution.

Carandola I, 303 F.3d at 513. Under this standard, the government must show that its regulation materially advances its substantial interest in reducing negative secondary effects and that reasonable alternative avenues of communication remain available. City of Los Angeles v. Alameda Books, Inc., 535 U.S. 425, 434, 122 S.Ct. 1728, 152 L.Ed.2d 670 (2002) (plurality); Carandola I, 303 F.3d at 515; see also Ward v. Rock Against Racism, 491 U.S. 781, 799, 109 S.Ct. 2746, 105 L.Ed.2d 661 (1989) (government must show its interest “would be achieved less effectively absent the regulation.”).

But while the government must “fairly support” its policy, it need not settle the matter beyond debate or produce an exhaustive evidentiary demonstration. Alameda Books, 535 U.S. at 438, 122 S.Ct. 1728 (plurality); see also id. at 451, 122 S.Ct. 1728 (Kennedy, J., concurring in the judgment) (“[V|ery little evidence is required.”).

Moreover, its policy expertise is entitled to “deference,” and it may demonstrate the efficacy of its method of reducing secondary effects “by appeal to common sense,” rather than “empirical data.” Id. at 439-40, 122 S.Ct. 1728 (plurality); see also id. at 451-52, 122 S.Ct. 1728 (Kennedy, J., concurring in the judgment). It may also rely on the experiences of other jurisdictions and on findings expressed in other cases. See City of Renton v. Playtime Theatres, Inc., 475 U.S. 41, 51-52, 106 S.Ct. 925, 89 L.Ed.2d 29 (1986).

Once the government makes this showing, the matter is at an end unless the plaintiff “produces clear and convincing evidence” to rebut it. Carandola I, 303 F.3d at 516. Papermoon argues that Virginia’s policy is unconstitutional because a ban on mixed beverages at its clubs is pointless when beer and wine are still allowed. It asserts that the ABC Board produced no studies to support such a restriction, while Paper-moon offered social science evidence undermining it. In assessing Papermoon’s challenge, we first examine the nature of the regulation and its burden on expressive interests.

We next consider whether the ABC Board sufficiently demonstrated the necessary relationship between the mixed beverage restriction and its interest in reducing negative secondary effects. Finally, we turn to Papermoon’s rebuttal evidence. A. We begin by noting that Virginia’s policy regarding alcohol at erotic dancing locales is about as tame as one could imagine. Virginia “has not forbidden these performances across the board. It has merely proscribed such performances in establishments that it licenses to sell liquor by the drink.” California v.

LaRue, 409 U.S. 109, 118, 93 S.Ct. 390, 34 L.Ed.2d 342 (1972); see also Carandola I, 303 F.3d at 513 n. 2 & 519. Indeed, Virginia does not even prohibit all alcohol at sexually oriented businesses, only mixed beverages. Wine and beer are as available at the Papermoon clubs as at any other Virginia bar. And as Paper-moon itself notes, beer remained the drink of choice for its patrons even during the period when it sold mixed beverages. Given that the First Amendment has been held to permit banning any alcohol where dancers strip to g-strings and pasties, Virginia’s policy is hardly censorious.

See Daytona Grand, Inc. v. City of Daytona Beach, 490 F.3d 860, 886 (11th Cir.2007); Ben’s Bar, Inc. v. Village of Somerset, 316 F.3d 702, 728 (7th Cir.2003). A mixed beverage license may well be a moneymaker — Papermoon offered uneontradicted evidence that it was — but in order to fail intermediate scrutiny there must be some greater showing than some loss of revenue. See Renton, 475 U.S. at 54, 106 S.Ct. 925. Indeed, the Court recognizes that a law may result in a mild and incidental diminution of speech without running afoul of the First Amendment. See Ward, 491 U.S. at 799-800, 109 S.Ct. 2746.

And here, it must be said, there is no indication that expression is being curtailed at all. Although one of the three Papermoon clubs initially decided to keep its mixed beverage license and have its dancers wear extra clothes, it evidently thought better of that decision and returned to pasties and g-strings. From aught that appears, Papermoon dancers continue to express themselves after reinstatement of the regulation without diminution of inhibition — the performance went on as before. B.

Not only does Virginia’s policy regulate with the lightest of touches, but the degree to which it trenches upon First Amendment values is minimal at best. The First Amendment’s pride of place in our constitutional order is a reflection of how essential the institution of free speech is to a democratic society. See Eu v. San Francisco County Democratic Cent. Comm., 489 U.S. 214, 223, 109 S.Ct. 1013, 103 L.Ed.2d 271 (1989); New York Times Co. v. Sullivan, 376 U.S. 254, 270, 84 S.Ct. 710, 11 L.Ed.2d 686 (1964).

But that principle also provides a limitation: activities that have little to do with advocacy, deliberation, or the exposition of ideas have correspondingly little to do with the First Amendment. See City of Erie v. Pap’s A.M., 529 U.S. 277, 294, 120 S.Ct. 1382, 146 L.Ed.2d 265 (2000). Here the kind of acts affected by the laws Papermoon assails are removed from the First Amendment’s core concerns. The challenged provisions pertain only to businesses where performers give strip shows or otherwise expose their buttocks or breasts. See Va.Code §§ 4.1-226(2)(i), -325(12), (13); 3 Va. Admin. Code § 5-50-140(B).

And the policy does not even purport to reach all such displays. Sexual entertainment “expressing matters of serious literary, artistic, scientific, or political value” offered in “establishments that are devoted primarily to the arts or theatrical performances” is entirely exempt. Va.Code §§ 4.1-226(2), -325(C); 3 Va. Admin. Code § 5-50-140(C). In other words, the policy primarily, if not exclusively, applies to bars offering performanees partaking more of “sexuality than of communication.” LaRue, 409 U.S. at 118, 93 S.Ct. 390.

Sexual expression and depictions can and do play an important role both in democratic and artistic discourse, and it is thus crucial to our ruling that Virginia has taken care to narrow its regulatory focus here to the particular context of sexually oriented entertainment at bars. As to this, we are simply not at liberty to ignore the Supreme Court’s emphasis upon the relatively greater protections afforded many other forms and outlets for artistic speech. The Court has instructed that nude dancing is “only marginally” of First Amendment value, Barnes v.

Glen Theatre, Inc., 501 U.S. 560, 566, 111 S.Ct. 2456, 115 L.Ed.2d 504 (1991), and “only within the outer ambit of the First Amendment’s protection.” Pap’s A.M., 529 U.S. at 289, 120 S.Ct. 1382. For quite simply, “it is manifest that society’s interest in protecting this type of expression is of a wholly different, and lesser, magnitude than the interest in untrammeled political debate.” Id. at 294, 120 S.Ct. 1382 (quoting Young v. American Mini Theatres, Inc., 427 U.S. 50, 70, 96 S.Ct. 2440, 49 L.Ed.2d 310 (1976)).

In the context of a restriction as mild as the one to which Papermoon has . been subjected, “[t]he impairment of First Amendment values is slight to the point of being risible, since the expressive activity involved in the kind of striptease entertainment provided in a bar has at best a modest social value.” Blue Canary Corp. v. City of Milwaukee, 251 F.3d 1121, 1124 (7th Cir.2001). And even here, it bears reminding, Virginia has not asked Paper-moon to stop the show or even to cease serving all alcoholic beverages in conjunction with it. c. 1.

With these considerations in mind, we assess the ABC Board’s justifications for the challenged policy. Notwithstanding the policy’s minimal effect on expressive interests, Papermoon requests that we strike it down because the ABC Board did not produce empirical studies showing that a ban on only mixed beverages at sexually oriented businesses will reduce secondary effects — “higher crime rates and lower property values,” and “public sexual conduct, sexual assault, and prostitution.” Carandola I, 303 F.3d at 513.

In making this claim, however, Papermoon asks us to subject the ABC Board to a more stringent standard than is compatible with the Supreme Court’s teachings or the appropriate relationship between courts and policymakers. For starters, Papermoon’s argument takes an ironic turn, namely that the Virginia regulation should be struck because it is too mild to be effective.

But in Pap’s A.M., the Supreme Court rejected the idea that the government’s rationale could be impeached because its regulation was not-as effective as a more restrictive alternative — in that case because the government combated the problems of totally nude dancing by requiring only that dancers wear pasties and g-strings.

See Pap’s A.M., 529 U.S. at 300-01, 120 S.Ct. 1382; see also id. at 310, 120 S.Ct. 1382 (Scalia, J., concurring) (“I am highly skeptical, to tell the truth, that the addition of pasties and G-strings will at all reduce the tendency of establishments such as Kandyland to attract crime and prostitution.... ”). It is one thing to challenge the government’s rationale as pretextual or to argue its restriction advantages certain speakers or ideas to the detriment of others. See City of Ladue v. Gilleo, 512 U.S. 43, 50-52, 114 S.Ct. 2038, 129 L.Ed.2d 36 (1994).

But when the government’s policy is not “a covert attack on speech,” invalidating a regulation because it is too permissive does First Amendment freedoms no favors. Alameda Books, 535 U.S. at 447, 122 S.Ct. 1728 (Kennedy, J., concurring in the judgment). Even setting this objection aside, however, the ABC Board’s position that prohibiting mixed beverages at establishments like Papermoon will curtail adverse secondary effects was hardly unsupported.

The Board considered more than forty studies documenting the negative secondary effects associated with establishments like Papermoon, and at any rate, it is well established that “bars and clubs that present nude or topless dancing” have “a long history of spawning deleterious effects.” Carandola I, 303 F.3d at 516 (citation omitted). Nor can there be any controversy over the proposition that intoxication aggravates such secondary effects. “Common sense indicates that any form of nudity coupled with alcohol in a public place begets undesirable behavior.” N.Y. State Liquor Auth. v.

Bellanca, 452 U.S. 714, 718, 101 S.Ct. 2599, 69 L.Ed.2d 357 (1981) (quoting N.Y. State Legis. Annual 150 (1977)). “Liquor and sex are an explosive combination.” Blue Canary, 251 F.3d at 1124. Common sense equally indicates that more intoxication will likely translate into more of the unwanted effects intoxication produces. The remaining link in the chain of reasoning underlying Virginia’s policy, and the one Papermoon devotes its energies to attacking, is the assumption that allowing mixed beverages to be served will likely produce more intoxication.

Paper-moon notes that the ABC Board’s own website instructs that an ounce-and-a-half shot of eighty-proof liquor contains the same amount of alcohol as a twelve-ounce beer or five-ounce glass of wine. How then, asks Papermoon, could mixed beverages lead to more drunkenness? This argument, however, trips on itself. By Papermoon’s own calculations, mixed beverages contain a higher concentration of alcohol in a smaller volume. The fact that there is as much alcohol in a shot of whiskey as there is in a serving of beer more than six times that volume illustrates how much more concentrated distilled spirits are.

A state is thus entitled to conclude, as the Commonwealth has: “Distilled spirits used in mixed beverages have higher alcohol content per volume than beer or wine. As a result, patrons drinking straight shots of liquor or mixed beverages can become intoxicated with less volume consumed, and, therefore, in less time and more easily, than patrons drinking beer or wine.” Appellee’s Br. at 17.

Virginia could certainly conclude that this higher level of intoxication from mixed beverages translates into higher levels of secondary effects in the surrounding area — namely sexual assaults, prostitution, and a generally higher disorderly conduct rate. Virginia could certainly take notice of the fact that people will visit these clubs throughout the hours of the clubs’ operation and that patrons will stay for varying lengths of time. For those at a club a relatively short period of time, a state of intoxication can be reached more quickly with distilled spirits.

For those there a longer time, the degree of intoxication will be much greater with mixed beverages than it would be for a person drinking beer or wine over the same period. Of course, all of these assumptions will be subject to individual variations dependent upon a variety of factors. But legislatures can pass laws dealing with what will normally happen without making exceptions for individual particularities.

In sum, Virginia has a legitimate interest in reducing the chances of a person leaving a strip club intoxicated by eliminating the sale of distilled spirits and it could further legitimately believe that this modest step could reduce the harmful secondary effects surrounding such establishments. 2. The particular risks of distilled spirits are reflected in the fact that Virginia in a variety of ways saddles them with special burdens. Distilled spirits are taxed more heavily than beer and wine. See Va.Code §§ 4.1-234, -236.

Unlike beer and wine, they generally can only be purchased for home consumption in ABC stores. See id. §§ 4.1-119(A), -210; see also id. § 4.1-221(A). And they may only be served in establishments with full restaurant facilities where at least forty-five percent of gross receipts come from the sale of food or other beverages. See id. § 4.1-210(A)(1). Papermoon argues that because these additional safeguards are in place, there is no need for further restricting the availability of mixed beverage licenses at its clubs. But one sensible precaution does not obviate the need for others.

As Coleburn testified, “Our whole system, as well as that of every state in the United States, is designed to discourage people from drinking distilled spirits in favor of the less intoxicating beer and wine.” Virginia has long favored less potent varieties of drink. The original ABC Act established a policy of “discouraging the consumption of hard liquor by making it harder to obtain while encouraging the consumption of light fermented beverages, such as beers and wines by making them easier to obtain.” Bolick v. Roberts, 199 F.Supp.2d 397, (E.D.Va.2002) (internal quotation omitted) (vacated as moot by Bolide v.

Danielson, 330 F.3d 274 (4th Cir.2003)). Indeed, support for such an approach is literally of early vintage. Thomas Jefferson argued that “[n]o nation is drunken where wine is cheap; and none sober, where the dearness of wine substitutes ardent spirits as the common beverage.” Letter from Thomas Jefferson to Jean Guillaume, Baron Hyde de Neuville (Dec. 13, 1818), in A Jefferson Profile as Revealed in His Letters 301 (Saul K. Padover ed., 1956). And in identifying distilled spirits as a matter of special regulatory concern, Virginia is anything but unique.

Like Virginia, most states adopted ABC statutes that “placed more stringent requirements on interests dealing in distilled spirits ... since distilled spirits, of course, contain a significantly higher alcoholic content than beer and wine.” California Beer Wholesalers Ass’n v. Alcoholic Bev. Control Appeals Bd., 5 Cal.3d 402, 96 Cal.Rptr. 297, 487 P.2d 745, 749 (1971).

Taxes are higher on distilled liquor than on beverages with lower concentrations of alcohol — indeed, one state’s highest court held it a violation of the state constitution to tax beverages whose alcoholic content was no greater than that of wine at the much higher rates applicable to distilled spirits, finding it “impossible to ignore this natural progression in alcoholic content by volume.” See Federated Distributors, Inc. v. Johnson, 125 Ill.2d 1, 125 Ill.Dec. 343, 530 N.E.2d 501, 509 (1988).

Similarly, licenses allowing the sale of mixed beverages are often costlier than licenses allowing only drinks with lower alcoholic contents, reflecting the fact “either that the legislature believed that a restaurant selling all liquors would ordinarily do a different kind of business or that it was contemplated that it would cost more to police it.” JPM Inv. Group, Inc. v. Brevard County Bd. of County Comm’rs, 818 So.2d 595, 599 (Fla.Dist.Ct.App.2002) (quoting Salerni v. Scheuy, 140 Conn. 566, 102 A.2d 528, 530 (1954)).

Widespread legislative recognition of the special need to regulate mixed beverages stands as an empirical demonstration of its own. The prevalence and durability of Virginia’s distinction in no way render it wrong, at least not where a state is exercising its “inherent police powers” “to prohibit the sale of alcoholic beverages in inappropriate locations.” 44 Liquormart, Inc. v. Rhode Island, 517 U.S. 484, 515, 116 S.Ct. 1495, 134 L.Ed.2d 711 (1996).

The provisions at issue comprise part of Virginia’s “long-established alcohol control law” and represent but one facet of the comprehensive regulatory approach the Commonwealth has adopted. Carandola I, 303 F.3d. at 514. Those provisions appear alongside a variety of other measures minimizing secondary effects and plainly related to preserving public order, and they “are most naturally viewed as companion provisions, also intended to prevent such societal ills.” Id. at 515; see Va.Code § 4.1-325(20). Courts have no warrant to supplant a state’s policy preferences with our own.

We have no trouble concluding that Virginia’s “inferences appear reasonable.” Alameda Books, 535 U.S. at 452, 122 S.Ct. 1728 (Kennedy, J., concurring in the judgment). That businesses like Papermoon are capable of producing harmful secondary effects, that intoxication exacerbates those effects, that more intoxication means more exacerbation, and that mixed beverages may lead to more intoxication are propositions whose sensible nature would lead to a Supreme Court slap of any hand that invalidated them.

Taken together, they provide “fair[ ] support” for the Commonwealth’s policy, and the ABC Board carried its burden. Id. at 438, 122 S.Ct. 1728 (plurality). D. Virginia has thus demonstrated the necessary relationship between its mixed beverage restriction and its substantial interest in reducing negative secondary effects. We turn therefore to Papermoon’s rebuttal of the ABC Board’s showing.

Evidence rebutting the government’s justification for a secondary effects regulation, however, must do more than challenge the government’s rationale; it must convincingly discredit the foundation upon which the government’s justification rests. See Carandola I, 303 F.3d at 516. Papermoon largely relied on the study produced by its expert, Professor Daniel Linz, and that evidence falls short of the “clear and convincing” standard necessary to sustain its challenge. Id. For a start, Linz’s before-and-after analysis focused only on Papermoon.

But officials “need not show that each individual adult establishment actually generates the undesired secondary effects.” Independence News, Inc. v. City of Charlotte, 568 F.3d 148, 156 (4th Cir.2009). Since the Virginia policy could be sustained if sexually oriented businesses “as a category” produce secondary effects when mixed beverages are served, Linz’s study hardly undermined the government’s case. Richland Bookmart, Inc. v. Knox County, 555 F.3d 512, 532 (6th Cir.2009).

Moreover, the study was based on only nine months of data, yet as Linz candidly acknowledged, a study of crime rates should be based on at least three years of information. And more generally, there was reason to be skeptical about how well his conclusions about Papermoon matched the governing legal standard since Linz has sought to debunk altogether the idea that sexually oriented businesses generate secondary effects as a “legal myth.” The Commonwealth also contests the data sets he used.

The Richmond data he obtained referred to “founded” incidents of crime but he acknowledged that the term is “not defined further by the police department” and that he had “found no other definition.” He also admitted that he did not know whether the addresses included with the crime data always referred to the place where a crime was committed, and in any event he did not account for crime that may be linked to Papermoon but that actually occurred outside the narrow zone of geographic proximity he had designated.

His Springfield data, meanwhile, did not include many relevant crimes, including disorderly conduct, drunkenness, driving under the influence, homicide, interference with police, prostitution, threatening bodily harm, various weapons offenses, and so on. So while the Linz study and others may well be of interest to legislatures or those formulating policy, it does not provide the kind of “clear and convincing” evidence needed to rebut the government’s showing and invalidate the regulation. E.

We need not dwell further on the particulars of the ABC Board’s showing or the problems with Papermoon’s rebuttal, however, for there is a simpler principle to be respected. The notion that the decisions of democratically accountable bodies must be set aside because of an absence of some unspecified quantum of social science support or the presence of a conflicting study commissioned by a litigant is one we must approach with skepticism. “As a general matter, courts should not be in the business of second-guessing fact-bound empirical assessments” made by lawmakers.

Alameda Books, 535 U.S. at 451, 122 S.Ct. 1728 (Kennedy, J., concurring in the judgment). A local policymaking body “knows the streets ... better than we do. It is entitled to rely on that knowledge; and if its inferences appear reasonable, we should not say there is no basis for its conclusion.” Id. at 452, 122 S.Ct. 1728 (citation omitted). Papermoon insists, however, that there is no evidence in the record showing that people drinking liquor at strip clubs cause more problems than people drinking beer or wine.

We agree with Papermoon that no empirical study has been presented that correlates criminal activity to the particular alcoholic beverage consumed, but we disagree that empirical support is needed for the perfectly sensible legislative proposition that someone drinking liquor at a strip club will get more intoxicated than someone drinking beer or wine over the same amount of time and hence be more likely to cross permissible lines. Óf course there will be many occasions when legislators will wish to consult empirical woi*k.

But much in life is not easily reduced to data sets, and there are limits on how much lawmakers’ judgment can be subjected to the argumentative rounds and elusive requirements of statistical validation. Policymakers “must be allowed a reasonable opportunity to experiment with solutions to admittedly serious problems.” Renton, 475 U.S. at 52, 106 S.Ct. 925 (quotation marks omitted). Legislative bodies have the advantage both of commonsense practicality and constituent accountability.

And “appeal to common sense,” Alameda Books, 535 U.S. at 439, 122 S.Ct. 1728 (plurality), and “common experience,” id. at 452, 122 S.Ct. 1728 (Kennedy, J., concurring in the judgment), are what the Supreme Court has approved. We risk violence, then, to democratic principles and to prudent and innovative governance when we make the validity of legislation turn on wars of competing studies. There remains a place in the legislative process for the exercise of simple reason and sound judgment, just as there remains a place in the judicial process for the exercise of some restraint.

These considerations are particularly salient where, as here, lawmakers have sought a middle ground that balances competing demands. Courts often are not equipped to craft such compromises and must take special care not to hamstring those who are. Where compromise embodies invidious distinctions, special scrutiny is demanded. But the distinction between beer and wine on the one hand and distilled spirits on the other is anything but invidious, and to strike down such classifications risks pushing lawmakers away from compromise and toward more polar postures.

It bears repeating that more severe policies, under which alcohol is completely forbidden at establishments like Paper-moon, have been repeatedly upheld in the face of constitutional challenge. See Daytona Grand, 490 F.3d at 886 (“[A]ny artistic or communicative elements present in such conduct are not of a kind whose content or effectiveness is dependent upon being conveyed where alcoholic beverages are served.”) (quoting Grand Faloon Tavern, Inc. v.

Wicker, 670 F.2d 943, 948 (11th Cir.1982)); Ben’s Bar, 316 F.3d at 726 (“The First Amendment does not entitle Ben’s Bar, its dancers, or its patrons, to have alcohol available during a ‘presentation’ of nude or semi-nude dancing.”). Were we to invalidate a policy restricting only distilled spirits, the Commonwealth’s response might well be to ban alcohol at sexually oriented businesses outright. The First Amendment does not demand that we distort the political process in such a fashion. None of this is to say that Virginia’s policy is unassailable or even right.

But the primary means to challenge legislative misconceptions is through the channels of representative government: hearings, speeches, conversations, debates, the whole clamorous drama of democracy that leads to the enactment of the given law. In the First Amendment context, those affected by restrictions designed to combat secondary effects may of course demonstrate that the justification for a particular restriction rests on “shoddy data or reasoning.” Alameda Books, 535 U.S. at 438, 122 S.Ct. 1728 (plurality).

But to invoke the power of the judiciary to set the policy aside, such evidence must be sufficiently convincing to “prove[] unsound” the government’s justification for its policy. See id. at 453, 122 S.Ct. 1728 (Kennedy, J., concurring in the judgment). Here the evidence is not. III. Papermoon also challenges portions of the Virginia ABC statutes as unconstitutionally vague and unconstitutionally over-broad. As with its substantive First Amendment claim, we find these objections to be without merit. A. We begin with Papermoon’s vagueness challenge.

In assessing a vagueness challenge, a court must ask whether the government’s policy is “set out in terms that the ordinary person exercising ordinary common sense can sufficiently understand and comply with.” Carandola II, 470 F.3d at 1079 (citation omitted). While laws that regulate expression are subjected to “stricter standards,” Smith v. California, 361 U.S. 147, 151, 80 S.Ct. 215, 4 L.Ed.2d 205 (1959), “perfect clarity and precise guidance have never been required even of regulations that restrict expressive activity.” Ward, 491 U.S. at 794, 109 S.Ct. 2746.

Two different sections of the ABC Act prohibit mixed beverages at establishments like Papermoon. The first requires that the ABC Board suspend the mixed beverage license of any establishment hosting “what is commonly called stripteasing, topless entertaining, and the like, or which has employees who are not clad both above and below the waist, or who uncommonly expose the body.” See Va. Code § 4.1-226(2)(i); see also id. § 4.1-223(3)(i).

The second provides that a mixed beverage licensee may not allow “any striptease act on the licensed premises” or “persons connected with the licensed business to appear nude or partially nude.” See id. §§ 4.1-325(12), (13). Papermoon argues that the terms “stripteasing” and “striptease,” and the phrases “clad both above and below the waist” and “partially nude” are unconstitutionally vague because it is unclear how much clothing has to be worn to satisfy their requirements. We find this argument unpersuasive.

As the district court noted, “striptease” is defined straightforwardly as “a burlesque act in which a performer removes clothing piece by piece.” Merriam-Webster’s Collegiate Dictionary 1166 (10th ed.1999); see also Barnes, 501 U.S. at 581, 111 S.Ct. 2456 (Souter, J., concurring in the judgment) (describing a striptease as “a dancer’s acts in going from clothed to nude ... integrated into the dance and its expressive function.”). The term is clearly one of common usage and given the erotic fashion in which clothes are removed, “a ‘striptease’ performance, we think, speaks for itself.” City of New Orleans v.

Kiefer, 246 La. 305, 164 So.2d 336, 339 (1964). Nor do we think that in this context the term “partially nude” is vague. Nudity, as a matter of everyday speech, refers to the absence of clothing, exposing those parts of the body commonly denominated “private.” Partial nudity would thus refer to the partial exposure of the private parts. Not surprisingly, that is precisely what the ABC regulation governing mixed beverage licenses provides, forbidding “less than a fully-opaque covering of the genitals, pubic hair or buttocks, or any portion of the breast below the top of the areola.” 3 Va. Admin.

Code § 5-50-140(B). This language tracks the definition of “nudity” elsewhere in Virginia law. See Va.Code § 18.2-390(2); see also, e.g., Iowa Code § 709.21(2)(a); Md.Code Ann., Crim. Law § 11-203(a)(6); Mass. Gen. Laws ch. 272, § 105(a); Mo.Rev.Stat. § 565.250(1); 18 Pa. Cons.Stat. § 7507.1(e); Utah Code Ann. § 76-5a-2(6); W. Va.Code § 61-8-28(a)(1); Wis. Stat. § 942.08(l)(a).

The meaning of the phrase “clad both above and below the waist” is similarly apparent: Papermoon’s dancers may not dance “topless” or “bottomless.” Again, the regulations make it clear that if mixed beverages are to be served, g-strings, pasties, and other such fig leaves will not do, as Papermoon itself well understands. Quite frankly, Papermoon’s vagueness challenge depends on wishful thinking. It is clear what conduct the ABC mixed beverage policy reaches — and that what it reaches is what Papermoon’s dancers do.

The risk that dancers at clubs like Paper-moon will be “chilled” into donning more clothing than the law requires is slim indeed. B. Finally, we consider Paper-moon’s overbreadth challenge. The over-breadth doctrine allows a party to “challenge a statute on its face because it also threatens others not before the court— those who desire to engage in legally protected expression but who may refrain from doing so rather than risk prosecution or undertake to have the law declared partially invalid.” Bd. of Airport Comm’rs v.

Jews for Jesus, Inc., 482 U.S. 569, 574, 107 S.Ct. 2568, 96 L.Ed.2d 500 (1987) (internal quotation marks omitted). Accordingly, the overbreadth doctrine is “strong medicine” to be applied “sparingly and only as a last resort.” Broadrick v. Oklahoma, 413 U.S. 601, 613, 93 S.Ct. 2908, 37 L.Ed.2d 830 (1973). A court properly holds a law facially invalid on overbreadth grounds only where its overbreadth is “substantial ... judged in relation to the statute’s plainly legitimate sweep.” Id. at 615, 93 S.Ct. 2908.

As discussed, Virginia’s prohibition on mixed beverages at venues like Papermoon is within the statutes’ legitimate sweep. And cultural venues offering “matters of serious literary, artistic, scientific, or political value” are properly exempted. Va.Code §§ 4.1-226(2), - 325(C); 3 Va. Admin. Code § 5-50-140(C). Papermoon stresses that the overwhelming majority of establishments licensed to sell mixed beverages in Virginia are not “adult entertainment establishments” and still would not come within the exception for cultural venues.

But that is beside the point since an ordinary bar is unlikely to have its employees strip or otherwise have their breasts or buttocks exposed (much less for expressive purposes). And one that did allow such displays might plausibly be linked to the secondary effects Virginia has targeted. Perfection is not required to survive an overbreadth challenge — a statute that shields “most protected activity” is permissible. Carandola II, 470 F.3d at 1085. Here, we see few, if any, likely applications of the policy that would be forbidden by the Constitution.

Indeed, the matter should be beyond debate since the exception for cultural venues Virginia adopted uses word-for-word the same language that cured North Carolina’s over-breadth problem in Carandola II. See Carandola II, 470 F.3d at 1083-84. Papermoon attempts to distinguish the case, arguing that the policy Carandola addressed only prohibited outright nudity while Virginia’s additionally prohibits even the near-nudity of g-strings and pasties where mixed beverages are served. We do not see, however, how this additional requirement is likely to inflict further collateral damage on protected expression.

The range of expressive activities dependent upon exposing the buttocks or breasts pretty well coincides with the range of those dependent on exposing the genitals. And if, as Papermoon strenuously urges, Carandola II’s conclusion that the statutes there at issue were not overbroad depended on the mildness of the restriction they imposed, then we must point out once again that a policy merely forbidding mixed beverages where erotic performances take place is likewise anything but draconian. IV. Where governmental action is involved, a constitution exists in part to prune extremes.

Where intermediate scrutiny is concerned, it is not wrong for moderation in the political process to find a constitutional home. The Commonwealth has demonstrated moderation in its efforts to balance the expressive value in erotic dancing with the unwanted encouragement of secondary effects. That courts should not be turned into appellate legislatures should go without saying, but it is particularly true where the political process has not sought to push the constitutional envelope and where lawmakers have responded conscientiously to prior opinions of this and other courts.

For the foregoing reasons, the judgment is AFFIRMED. . Justice Kennedy’s separate opinion in Alameda Books accepted the four-member plurality's holding on the evidentiary standard that governs the secondary effects inquiry. See Alameda Books, 535 U.S. at 449, 122 S.Ct. 1728 (Kennedy, J., concurring in the judgment) ("The plurality ... gives the correct answer” to the question "how much evidence is required?”); see also Carandola I, 303 F.3d at 516. . Papermoon's reliance on Joelner v. Village of Washington Park, 508 F.3d 427 (7th Cir.2007), is accordingly misplaced.

The policy there at issue, “prospectively banning alcohol in strip clubs opened in the future,” "was adopted to stifle competition with current license holders,” not to combat secondary effects, and was subjected to strict scrutiny. Id. at 429, 431, 433.