Justice Thomas
delivered the opinion of the Court, except as to Part II-A.
Oklahoma has a semiclosed primary system, in which a political party may invite only its own party members and voters registered as Independents to vote in the party’s primary. The Court of Appeals held that this system violates the right to freedom of association of the Libertarian Party of Oklahoma (LPO) and several Oklahomans who are registered members of the Republican and Democratic Parties. We hold that it does not.
I
Oklahoma’s election laws provide that only registered members of a political party may vote in the party’s primary, [585]*585see Okla. Stat. Ann., Tit. 26, § 1-104(A) (West 1997), unless the party opens its primary to registered Independents as well, see § 1-104(B)(1). In May 2000, the LPO notified the secretary of the Oklahoma State Election Board that it wanted to open its upcoming primary to all registered Oklahoma voters, without regard to their party affiliation. See § 1-104(B)(4) (requiring notice when a party opens its primary to Independents). Pursuant to § 1-104, the secretary agreed as to Independent voters, but not as to voters registered with other political parties. The LPO and several Republican and Democratic voters then sued for declaratory and injunctive relief in the United States District Court for the Western District of Oklahoma, alleging that Oklahoma’s semiclosed primary law unconstitutionally burdens their First Amendment right to freedom of political association. App. 20.
After a hearing, the District Court declined to enjoin Oklahoma’s semiclosed primary law for the 2000 primaries. After a 2-day bench trial following the primary election, the District Court found that Oklahoma’s semiclosed primary system did not severely burden respondents’ associational rights. Further, it found that any burden imposed by the system was justified by Oklahoma’s asserted interest in “preserving the political parties as viable and identifiable interest groups, [and] insuring that the results of a primary election . . . accurately reflect the voting of the party members.” Memorandum Opinion, Case No. CIV-00-1071-F (WD Okla., Jan. 24,2003), App. to Pet. for Cert. 55-56 (hereinafter Memorandum Opinion) (internal quotation marks omitted). The District Court therefore upheld the semiclosed primary statute as constitutional. Id., at 72-73.
On appeal, the Court of Appeals for the Tenth Circuit reversed the judgment of the District Court. The Court of Appeals concluded that the State’s semiclosed primary statute imposed a severe burden on respondents’ associational rights, and thus was constitutional only if the statute was [586]*586narrowly tailored to serve a compelling state interest. 363 F. 3d 1048, 1057-1058 (2004). Finding none of Oklahoma’s interests compelling, the Court of Appeals enjoined Oklahoma from using its semiclosed primary law. Id., at 1060-1061. Because the Court of Appeals’ decision not only prohibits Oklahoma from using its primary system but also casts doubt on the semiclosed primary laws of 23 other States,1 we granted certiorari. 542 U. S. 965 (2004).
t — I I — (
The Constitution grants States broad power to prescribe the ‘Times, Places and Manner of holding Elections for Senators and Representatives,’ Art. I, §4, cl. 1, which power is matched by state control over the election process for state offices.” Tashjian v. Republican Party of Conn., 479 U. S. 208, 217 (1986); Timmons v. Twin Cities Area New Party, 520 U. S. 351, 358 (1997) (quoting Tashjian). We have held that the First Amendment, among other things, protects the right of citizens “to band together in promoting among the electorate candidates who espouse their political views.” California Democratic Party v. Jones, 530 U. S. 567, 574 (2000). Regulations that impose severe burdens on associational rights must be narrowly tailored to serve a compelling state interest. Timmons, 520 U. S., at 358. However, [587]*587when regulations impose lesser burdens, “a State’s important regulatory interests will usually be enough to justify reasonable, nondiscriminatory restrictions.” Ibid, (internal quotation marks omitted).
In Tashjian, this Court struck down, as inconsistent with the First Amendment, a closed primary system that prevented a political party from inviting Independent voters to vote in the party’s primary. 479 U. S., at 225. This case presents a question that Tashjian left open: whether a State may prevent a political party from inviting registered voters of other parties to vote in its primary. Id., at 224, n. 13. As Tashjian acknowledged, opening a party’s primary “to all voters, including members of other parties, . . . raise[s] a different combination of considerations.” Ibid. We are persuaded that any burden Oklahoma’s semiclosed primary imposes is minor and justified by legitimate state interests.
A
At the outset, we note that Oklahoma’s semiclosed primary system is unlike other laws this Court has held to infringe associational rights. Oklahoma has not sought through its electoral system to discover the names of the LPO’s members, see NAACP v. Alabama ex rel. Patterson, 357 U. S. 449, 451 (1958); to interfere with the LPO by restricting activities central to its purpose, see NAACP v. Claiborne Hardware Co., 458 U. S. 886, 895 (1982); NAACP v. Button, 371 U. S. 415, 423-426 (1963); to disqualify the LPO from public benefits or privileges, see Keyishian v. Board of Regents of Univ. of State of N. Y., 385 U. S. 589, 595-596 (1967); or to compel the LPO’s association with unwanted members or voters, see Jones, supra, at 577. The LPO is free to canvass the electorate, enroll or exclude potential members, nominate the candidate of its choice, and engage in the same electoral activities as every other political party in Oklahoma. Oklahoma merely prohibits the LPO from leaving the selection of its candidates to people who are members of [588]*588another political party. Nothing in § 1-104 prevents members of other parties from switching their registration to the LPO or to Independent status.2 The question is whether the Constitution requires that voters who are registered in other parties be allowed to vote in the LPO’s primary.
In other words, the Republican and Democratic voters who have brought this action do not want to associate with the LPO, at least not in any formal sense. They wish to remain registered with the Republican, Democratic, or Reform parties, and yet to assist in selecting the Libertarian Party’s candidates for the general election. Their interest is in casting a vote for a Libertarian candidate in a particular primary election,3 rather than in banding together with fellow citizens committed to the LPO’s political goals and ideas. See Jones, supra, at 573-574, n. 5 (“As for the associational ‘interest’ in selecting the candidate of a group to which one does not belong, that falls far short of a constitutional right, if indeed it [589]*589can even fairly be characterized as an interest”). And the LPO is happy to have their votes, if not their membership on the party rolls.
However, a voter who is unwilling to disaffiliate from another party to vote in the LPO’s primary forms little “association” with the LPO — nor the LPO with him. See Tashjian, supra, at 235 (Scalia, J., dissenting). That same voter might wish to participate in numerous party primaries, or cast ballots for several candidates, in any given race. The issue is not “dual associations,” post, at 601 (O’Connor, J., concurring in part and concurring in judgment), but seemingly boundless ones. “If the concept of freedom of association is extended” to a voter’s every desire at the ballot box, “it ceases to be of any analytic use.” Tashjian, supra, at 235 (Scalia, J., dissenting); cf. Democratic Party of United States v. Wisconsin ex rel. La Follette, 450 U. S. 107, 130 (1981) (Powell, J., dissenting) (“[Not] every conflict between state law and party rules concerning participation in the nomination process creates a burden on associational rights”).
But even if Oklahoma’s semiclosed primary system burdens an associational right, the burden is less severe than others this Court has upheld as constitutional. For instance, in Timmons, we considered a Minnesota election law prohibiting multiparty, or “fusion,” candidacies in which a candidate appears on the ballot as the nominee of more than one party. 520 U. S., at 353-354. Minnesota’s law prevented the New Party, a minor party under state law, from putting forward the same candidate as a major party. The New Party challenged the law as unconstitutionally burdening its associational rights. Id., at 354-355. This Court concluded that the burdens imposed by Minnesota’s law— “though not trivial — [were] not severe.” Id., at 363.
The burdens were not severe because the New Party and its members remained free to govern themselves internally and to communicate with the public as they wished. Ibid. [590]*590Minnesota had neither regulated the New Party’s internal decisionmaking process, nor compelled it to associate with voters of any political persuasion, see Jones, 530 U. S., at 577. The New Party and its members simply could not nominate as their candidate any of “those few individuals who both have already agreed to be another party’s candidate and also, if forced to choose, themselves prefer that other party.” Timmons, supra, at 363.
The same reasons underpinning our decision in Timmons show that Oklahoma’s semiclosed primary system burdens the LPO only minimally. As in Timmons, Oklahoma’s law does not regulate the LPO’s internal processes, its authority to exclude unwanted members, or its capacity to communicate with the public. And just as in Timmons, in which Minnesota conditioned the party’s ability to nominate the candidate of its choice on the candidate’s willingness to disaffiliate from another political party, Oklahoma conditions the party’s ability to welcome a voter into its primary on the voter’s willingness to dissociate from his current party of choice. If anything, it is “[t]he moment of choosing the party’s nominee” that matters far more, Jones, 530 U. S., at 575, for that is “ ‘the crucial juncture at which the appeal to common principles may be translated into concerted action, and hence to political power in the community,’ ” ibid, (quoting Tashjian, 479 U. S., at 216). If a party may be prevented from associating with the candidate of its choice — its desired “ ‘standard bearer,’ ” Timmons, supra, at 359; Jones, supra, at 575 — because that candidate refuses to disaffiliate from another political party, a party may also be prevented from associating with a voter who refuses to do the same.
Oklahoma’s semiclosed primary system imposes an even slighter burden on voters than on the LPO. Disaffiliation is not difficult: In general, “anyone can ‘join’ a political party merely by asking for the appropriate ballot at the appropriate time or (at most) by registering within a state-defined reasonable period of time before an election.” Jones, supra, [591]*591at 596 (Stevens, J., dissenting). In Oklahoma, registered members of the Republican, Democratic, and Reform Parties who wish to vote in the LPO primary simply need to file a form with the county election board secretary to change their registration. See Okla. Stat. Ann., Tit. 26, §4-119 (West Supp. 2005). Voters are not “locked in” to an unwanted party affiliation, see Kusper v. Pontikes, 414 U. S. 51, 60-61 (1973), because with only nominal effort they are free to vote in the LPO primary. For this reason, too, the registration requirement does not unduly hinder the LPO from associating with members of other parties. To attract members of other parties, the LPO need only persuade voters to make the minimal effort necessary to switch parties.
B
Respondents argue that this case is no different from Tashjian. According to respondents, the burden imposed by Oklahoma’s semiclosed primary system is no less severe than the burden at issue in Tashjian, and hence we must apply strict scrutiny as we did in Tashjian. We disagree. At issue in Tashjian was a Connecticut election statute that required voters to register with a political party before participating in its primary. 479 U. S., at 210-211. The State’s Republican Party, having adopted a rule that allowed Independent voters to participate in its primary, contended that Connecticut’s closed primary infringed its right to associate with Independent voters. Ibid. Applying strict scrutiny, this Court found that the interests Connecticut advanced to justify its ban were not compelling, and thus that the State could not constitutionally prevent the Republican Party from inviting into its primary willing Independent voters. Id., at 217-225.
Respondents’ reliance on Tashjian is unavailing. As an initial matter, Tashjian applied strict scrutiny with little discussion of the magnitude of the burdens imposed by Connecticut’s closed primary on parties’ and voters’ associational [592]*592rights. Post, at 605 (O’Connor, J., concurring in part and concurring in judgment). But not every electoral law that burdens associational rights is subject to strict scrutiny. See, e. g., Nader v. Schaffer, 417 F. Supp. 837, 849 (Conn.) (“There must be more than a minimal infringement on the rights to vote and of association . . . before strict judicial review is warranted”), aff’d, 429 U. S. 989 (1976). Instead, as our cases since Tashjian have clarified, strict scrutiny is appropriate only if the burden is severe. Jones, supra, at 582; Timmons, 520 U. S., at 358. In Tashjian itself, Independent voters could join the Connecticut Republican Party as late as the day before the primary. 479 U. S., at 219. As explained above, supra, at 590-591, requiring voters to register with a party prior to participating in the party’s primary minimally burdens voters’ associational rights.
Nevertheless, Tashjian is distinguishable. Oklahoma’s semiclosed primary imposes an even less substantial burden than did the Connecticut closed primary at issue in Tashjian. In Tashjian, this Court identified two ways in which Connecticut’s closed primary limited citizens’ freedom of political association. The first and most important was that it required Independent voters to affiliate publicly with a party to vote in its primary. 479 U. S., at 216, n. 7. That is not true in this case. At issue here are voters who have already affiliated publicly with one of Oklahoma’s political parties. These voters need not register as Libertarians to vote in the LPO’s primary; they need only declare themselves Independents, which would leave them free to participate in any party primary that is open to registered Independents. See Okla. Stat. Ann., Tit. 26, § 1-104(B)(1) (West 1997).
The second and less important burden imposed by Connecticut’s closed primary system was that political parties could not “broaden opportunities for joining ... by their own act, without any intervening action by potential voters.” Tashjian, 479 U. S., at 216, n. 7. Voters also had to act by registering themselves in a particular party. Ibid. That is [593]*593equally true of Oklahoma’s semiclosed primary system: Voters must register as Libertarians or Independents to participate in the LPO’s primary. However, Tashjian did not characterize this burden alone as severe, and with good reason. Many electoral regulations, including voter registration generally, require that voters take some action to participate in the primary process. See, e. g., Rosario v. Rockefeller, 410 U. S. 752, 760-762 (1973) (upholding requirement that voters change party registration 11 months in advance of the primary election). Election laws invariably “affec[t] — at least to some degree — the individual’s right to vote and his right to associate with others for political ends.” Anderson v. Celebrezze, 460 U. S. 780, 788 (1983).
These minor barriers between voter and party do not compel strict scrutiny. See Bullock v. Carter, 405 U. S. 134, 143 (1972). To deem ordinary and widespread burdens like these severe would subject virtually every electoral regulation to strict scrutiny, hamper the ability of States to run efficient and equitable elections, and compel federal courts to rewrite state electoral codes. The Constitution does not require that result, for it is beyond question “that States may, and inevitably must, enact reasonable regulations of parties, elections, and ballots to reduce election- and campaign-related disorder.” Timmons, supra, at 358; Storer v. Brown, 415 U. S. 724, 730 (1974). Oklahoma’s semiclosed primary system does not severely burden the associational rights of the State’s citizenry.
C
When a state electoral provision places no heavy burden on associational rights, “a State’s important regulatory interests will usually be enough to justify reasonable, nondiscriminatory restrictions.” Timmons, supra, at 358 (internal quotation marks omitted); Anderson, supra, at 788. Here, Oklahoma’s semiclosed primary advances a number of regulatory interests that this Court recognizes as important: It [594]*594“preserves] [political] parties as viable and identifiable interest groups,” Nader, 417 F. Supp., at 845; enhances parties’ electioneering and party-building efforts, id., at 848; and guards against party raiding and “sore loser” candidacies by spurned primary contenders, Storer, supra, at 735.
First, as Oklahoma asserts, its semiclosed primary “preserves] the political parties as viable and identifiable interest groups, insuring that the results of a primary election, in a broad sense, accurately reflec[t] the voting of the party members.” Amended and Supplemental Trial Brief of Defendants 10, Record Doc. 63 (quoting without attribution Nader, supra, at 845). The LPO wishes to open its primary to registered Republicans and Democrats, who may well vote in numbers that dwarf the roughly 300 registered LPO voters in Oklahoma. See Memorandum Opinion 31-32 (at least 95% of voters in LPO’s 1996 primary were independents, not Libertarians). If the LPO is permitted to open its primary to all registered voters regardless of party affiliation, the candidate who emerges from the LPO primary may be “unconcerned with, if not... hostile to,” the political preferences of the majority of the LPO’s members. Nader, supra, at 846. It does not matter that the LPO is willing to risk the surrender of its identity in exchange for electoral success. Oklahoma’s interest is independent and concerns the integrity of its primary system. The State wants to “avoid primary election outcomes which would tend to confuse or mislead the general voting population to the extent [it] relies on party labels as representative of certain ideologies.” Brief for Petitioners 12 (quoting without attribution Nader, supra, at 845); Eu v. San Francisco County Democratic Central Comm., 489 U. S. 214, 228 (1989).
Moreover, this Court has found that “ ‘[i]n facilitating the effective operation of [a] democratic government, a state might reasonably classify voters or candidates according to political affiliations.’” Nader, supra, at 845-846 (quoting Ray v. Blair, 343 U. S. 214, 226, n. 14 (1952)). But for that [595]*595classification to mean much, Oklahoma must be allowed to limit voters’ ability to roam among parties’ primaries. The purpose of party registration is to provide “a minimal demonstration by the voter that he has some ‘commitment’ to the party in whose primary he wishes to participate.” Nader, supra, at 847. That commitment is lessened if party members may retain their registration in one party while voting in another party’s primary. Opening the LPO’s primary to all voters not only would render the LPO’s imprimatur an unreliable index of its candidate’s actual political philosophy, but it also “would make registered party affiliations signifi-. cantly less meaningful in the Oklahoma primary election system.” Memorandum Opinion 59. Oklahoma reasonably has concluded that opening the LPO’s primary to all voters regardless of party affiliation would undermine the crucial role of political parties in the primary process. Cf. Jones, 530 U. S., at 574.
Second, Oklahoma’s semiclosed primary system, by retaining the importance of party affiliation, aids in parties’ electioneering and party-building efforts. “It is common experience that direct solicitation of party members — by mail, telephone, or face-to-face contact, and by the candidates themselves or by their active supporters — is part of any primary election campaign.” Nader, supra, at 848. Yet parties’ voter turnout efforts depend in large part on accurate voter registration rolls. See, e. g., Council of Alternative Political Parties v. State Div. of Elections, 344 N. J. Super. 225, 231-232, 781 A. 2d 1041, 1045 (2001) (“It is undisputed that the voter registration lists, with voter affiliation information, .. . provide essential information to the [party state committees] for other campaign and party-building activities, including canvassing and fundraising”).
When voters are no longer required to disaffiliate before participating in other parties? primaries, voter registration rolls cease to be an accurate reflection of voters’ political preferences. And without registration rolls that accurately [596]*596reflect likely or potential primary voters, parties risk expending precious resources to turn out party members who may have decided to cast their votes elsewhere. See Brief for State of South Dakota et al. as Amici Curiae 20-21. If encouraging citizens to vote is an important state interest, see Jones, supra, at 587 (Kennedy, J., concurring), then Oklahoma is entitled to protect parties’ ability to plan their primaries for a stable group of voters. Tr. of Oral Arg. 26.
Third, Oklahoma has an interest in preventing party raiding, or “the organized switching of blocs of voters from one party to another in order to manipulate the outcome of the other party’s primary election.” Anderson, 460 U. S., at 788-789, n. 9; Jones, supra, at 572. For example, if the outcome of the Democratic Party primary were not in doubt, Democrats might vote in the LPO primary for the candidate most likely to siphon off votes from the Republican candidate in the general election. Or a Democratic primary contender who senses defeat might launch a “sore loser” candidacy by defecting to the LPO primary, taking with him loyal Democratic voters, and thus undermining the Democratic Party in the general election.4 Storer, 415 U. S., at 735. Oklahoma has an interest in “tempering] the destabilizing effects” of precisely this sort of “party splintering and excessive fac[597]*597tionalism.” Timmons, 520 U. S., at 367; cf. Davis v. Bandemer, 478 U. S. 109, 144-145 (1986) (O’Connor, J., concurring in judgment). Oklahoma’s semiclosed primary system serves that interest by discouraging voters from temporarily defecting from another party to vote in the LPO primary. While the State’s interest will not justify “unreasonably exclusionary restrictions,” Timmons, 520 U. S., at 367, we have “repeatedly upheld reasonable, politically neutral regulations” like Oklahoma’s semiclosed primary law, id., at 369 (internal quotation marks omitted).
I — I HH 1 — <
Beyond their challenge to Oklahoma’s semiclosed primary law, § 1-104, respondents have expanded their challenge before this Court to include other Oklahoma election laws. Respondents contend that several of the State’s ballot access and voter registration laws, taken together, severely burden their associational rights by effectively preventing them from changing their party affiliations in advance of a primary election. Brief for Respondents 15-18 (discussing the joint operation of Okla. Stat. Ann., Tit. 26, §§ 1-108, 1-109, 1-110, 4-112, and 4-119 (West Supp. 2005)).
Though the LPO has unsuccessfully challenged one of these provisions before, see Rainbow Coalition of Okla. v. Oklahoma State Election Bd., 844 F. 2d 740 (CA10 1988) (rejecting First Amendment challenge by LPO and other political parties to Oklahoma’s ballot access provision, § 1-108 (West 1981 and Supp. 1987)), respondents raise this argument for the first time in their brief on the merits to this Court. Before the District Court and the Court of Appeals, the only associational burden of which respondents complained was that imposed by § 1-104 (West 1997), i. e., the need to disaffiliate from one party in order to vote in another party’s primary. See, e. g., Appellants’ Opening Brief in No. 03-6058 (CA10), pp. 5, 8-10, 30 (challenging only § 1-104 as applied to respondents); Plaintiffs’ Amended Trial Brief [598]*5989-25, Record Doc. 65 (same); Amended Complaint 6-9, id., Doc. 23 (same). As a result, there is virtually no evidence in the record on how other electoral regulations operate in tandem with § 1-104, whether these other laws actually burden respondents’ associational rights, and whether these laws advance important or even compelling state interests. We ordinarily do not consider claims neither raised nor decided below, Cooper Industries, Inc. v. Aviall Services, Inc., 543 U. S. 157, 168-169 (2004) (citing Adarand Constructors, Inc. v. Mineta, 534 U. S. 103, 109 (2001) (per curiam)), and respondents have pointed to no unusual circumstances that would warrant considering other portions of Oklahoma’s electoral code this late in the day, see Taylor v. Freeland & Kronz, 503 U. S. 638, 645-646 (1992). We therefore decline to consider this aspect of their challenge.
* * *
Oklahoma remains free to allow the LPO to invite registered voters of other parties to vote in its primary. But the Constitution leaves that choice to the democratic process, not to the courts. The judgment of the Court of Appeals is reversed, and the case is remanded for further proceedings.
It is so ordered.