ROGERS, Circuit Judge.

ARGUED: David Martin, University of Virginia School of Law, Charlottesville, Virginia, for Appellants. Renata Y. Staff, Office of the Ohio Attorney General, Columbus, Ohio, for Appellees. ON BRIEF: David Martin, Stephen L. Braga, University of Virginia School of Law, Charlottesville, Virginia, Alphonse A. Gerhardstein, Jennifer L. Branch, Gerhardstein & Branch, Cincinnati, Ohio, for Appellants. Renata Y. Staff, Ryan L. Richardson, Office of the Ohio Attorney General, Columbus, Ohio, for Appellees.

OPINION

ROGERS, Circuit Judge.

Ohio elects its state judges through a hybrid process. Judicial candidates are first selected through partisan primary elections. On the general-election ballot, however, their names show no partisan affiliation, even though judicial candidates may affiliate with political parties throughout the entirety of their campaigns. The plaintiffs—the Ohio Democratic Party, three individuals who were candidates for state-court judgeships in the 2010 election, and a statewide labor organization—brought suit below challenging the constitutionality of the Ohio law that precludes the inclusion of judicial candidate party affiliations on the general-election ballot. While the plaintiffs argue that Ohio's electoral system burdens their First and Fourteenth Amendment rights, the burden is minimal and is outweighed by Ohio's interest in minimizing partisanship in judicial elections. Consequently, the district court correctly granted the Ohio Attorney General's motion for summary judgment.

Judicial candidates in Ohio must first compete in partisan primary elections in which they are listed on their political party's primary election ballot, known as an “office type ballot.” Ohio Rev.Code § 3505.03. However, unlike their legislative and executive counterparts, the winners of the judicial primary elections must then participate in a nonpartisan general election. Ohio Rev.Code § 3501.01(J). Ohio Rev.Code § 3505.04—the statute which the plaintiffs challenge—precludes judicial candidates from being associated with their political parties on the general-election ballot:

No name or designation of any political party nor any words, designations, or emblems descriptive of a candidate or his political affiliation, or indicative of the method by which such candidate was nominated or certified, shall be printed under or after any nonpartisan candidate's name which is printed on the ballot.

Ohio's justification for this two-tiered approach is that it wishes to minimize reliance on political parties in judicial elections while still establishing a formal role for political parties in the nomination process. Ohio explains that its system “leaves to political parties the right to place candidates for judicial offices in nomination” while also “withdraw[ing] candidates for judicial offices from partisan politics” in recognition of the fact that judges “can serve no party, promulgate no partisan theories of government, encourage no partisan economic measures.” State ex rel. Weinberger v. Miller, 87 Ohio St. 12, 99 N.E. 1078, 1085 (1912). While Ohio has been electing its judges under this system since 1851, id. at 1086, few other states employ a similar method.

On July 28, 2010, the plaintiffs filed a complaint in the district court against various Ohio state officials, alleging that § 3505.04 violates their First and Fourteenth Amendment rights to freedom of expression and association. The plaintiffs also requested a temporary restraining order and preliminary injunction that would permit judicial candidates to list their political party affiliations on the 2010 general election ballot. The Ohio Attorney General later intervened to defend the statute.

At the hearing for the temporary restraining order, testimony established that many voters are uninformed about judicial races. Additionally, many voters abstain from voting in nonpartisan judicial races despite having cast a vote in other races on the same ballot, a phenomenon known as “voter drop off.” For example, in the 2008 general election, about one percent of voters failed to vote for President, but approximately thirty percent did not cast a vote in a race for state Supreme Court Justice. Overall, voter drop off in Ohio judicial elections ranged between 20–32% in the 1980s, 1990s, and 2000s. Evidence also established that voter drop off is greater in nonpartisan judicial races than in other “low-information” partisan races—that is, races in which voters are just as likely as in judicial races to be uninformed about the candidates. An expert witness concluded that voter drop off in Ohio's judicial races is therefore caused at least in part by the elections' nonpartisan nature. The district court denied the plaintiffs' request for a temporary restraining order and preliminary injunction. We affirmed the district court's order. Ohio Council 8 Am. Fed'n of State, Cty. & Mun. Emps., AFL–CIO v. Brunner, 462 Fed.Appx. 557, 559 (6th Cir.2012) (per curiam). While the appeal of the denial of the preliminary injunction was still pending, Ohio filed a motion for summary judgment in the district court. Following our affirmance of its order denying the preliminary injunction, the district court granted Ohio's motion for summary judgment. Ohio Council 8 Am. Fed'n of State, Cty. & Mun. Emps., AFL–CIO v. Brunner, 24 F.Supp.3d 680, 683 (S.D.Ohio 2014).

Employing the “flexible” Anderson–Burdick balancing test, the district court first concluded that § 3505.04 does not impose a severe burden on the plaintiffs' First and Fourteenth Amendment rights. Brunner, 24 F.Supp.3d at 686–98 (citing Anderson v. Celebrezze, 460 U.S. 780, 103 S.Ct. 1564, 75 L.Ed.2d 547 (1983); Burdick v. Takushi, 504 U.S. 428, 112 S.Ct. 2059, 119 L.Ed.2d 245 (1992)). The district court then reasoned that any burden that the statute does place on the plaintiffs is justified “if the statute serves an important regulatory interest of the State.” Id. at 698. Ohio claimed as its only interest “minimizing partisanship in judicial elections.” Id. The district court described Ohio's “Solomonic approach” to carrying out its stated interest as “half-hearted at best,” noting that Ohio “steeps the judicial selection process in partisan politics everywhere but in the voting booth on election day.” Id. at 699–701. Nonetheless, because the district court concluded that the burden that § 3505.04 places on the plaintiffs is not severe, the court held that Ohio's interest in diminishing reliance on political parties in judicial elections is legitimate and that it outweighs the plaintiffs' interests in using the general-election ballot as a forum for their expressive and associative activities. Id. at 701. The plaintiffs appeal.

Although the 2010 election is long over, this case is not moot. The “capable of repetition yet evading review” doctrine saves a case from mootness when “(1) the challenged action was in its duration too short to be fully litigated prior to its cessation or expiration, and (2) there was a reasonable expectation that the same complaining party would be subject to the same action again.” Weinstein v. Bradford, 423 U.S. 147, 149, 96 S.Ct. 347, 46 L.Ed.2d 350 (1975) (per curiam). Challenges by former candidates and their party sponsors are “typically capable of repetition because [a candidate] ‘retains the right to run for judicial office again,’ ” Platt v. Bd. of Comm'rs on Grievances & Discipline of Ohio Supreme Court, 769 F.3d 447, 453 (6th Cir.2014) (quoting Carey v. Wolnitzek, 614 F.3d 189, 197 (6th Cir.2010)), and because “it is likely that the [party] will once again seek to place candidates on the ... ballot.” Libertarian Party of Ohio v. Blackwell, 462 F.3d 579, 584–85 (6th Cir.2006). The plaintiffs' claims regarding the 2010 election evaded review, yet are capable of repetition, because the former candidates may run again for judicial office in Ohio, the Ohio Democratic Party will likely seek to again place judicial candidates on the general-election ballot, and the members of the statewide labor organization will again vote in Ohio judicial elections.

Section 3505.04 is constitutional because at most it minimally burdens the plaintiffs' rights to freedom of expression and association, and because the state's interest is sufficient to outweigh that minimal burden. Supreme Court precedent requires that we apply just such a balancing test, often referred to as the Anderson–Burdick test. A court considering a challenge to a state election law must weigh “the character and magnitude of the asserted injury to the rights protected by the First and Fourteenth Amendments that the plaintiff seeks to vindicate” against “the precise interests put forward by the State as justifications for the burden imposed by its rule,” taking into consideration “the extent to which those interests make it necessary to burden the plaintiff's rights.” Anderson, 460 U.S. at 789, 103 S.Ct. 1564; see also Tashjian v. Republican Party of Conn., 479 U.S. 208, 213–14, 107 S.Ct. 544, 93 L.Ed.2d 514 (1986). “[T]he rigorousness of our inquiry into the propriety of a state election law depends upon the extent to which a challenged regulation burdens First and Fourteenth Amendment rights.” Burdick, 504 U.S. at 434, 112 S.Ct. 2059. If a state imposes “severe restrictions” on constitutional rights, then the state law must pass strict scrutiny to survive, meaning that it must be “narrowly drawn to advance a state interest of compelling importance.” Id. (quoting Norman v. Reed, 502 U.S. 279, 289, 112 S.Ct. 698, 116 L.Ed.2d 711 (1992)). If, however, the regulations are minimally burdensome and nondiscriminatory, a less-searching examination closer to rational basis applies, Green Party of Tenn. v. Hargett, 767 F.3d 533, 546 (6th Cir.2014), and “the State's important regulatory interests are generally sufficient to justify the restrictions.” Burdick, 504 U.S. at 434, 112 S.Ct. 2059 (internal quotation marks omitted). Further, if the regulation falls somewhere in between the two extremes, “the burden on the plaintiffs [is weighed] against the state's asserted interest and chosen means of pursuing it.” Green Party of Tenn., 767 F.3d at 546.

The burden on the plaintiffs' First and Fourteenth Amendment rights is minimal because political parties and judicial candidates remain free to provide, and voters remain free to receive, a plethora of information regarding whether a given candidate affiliates with or is endorsed by a particular political party. The plaintiffs contend that § 3505.04 burdens a political party's right to champion its nominees and educate voters, leading to large amounts of voter drop off; judicial candidates' right to express their qualifications and associate with their political parties; and voters' rights to associate, receive information, and cast meaningful votes. While the plaintiffs' asserted interests are not trivial, the expansive ability of the plaintiffs to exercise their associational and expressive rights in other ways during the general election demonstrates that any burdens that § 3505.04 places on their constitutional rights are minimal.

First, § 3505.04 only minimally burdens a political party's rights because a political party has no First Amendment right to designate its nominee on the general-election ballot and because a party has many other opportunities to champion its nominee and educate voters. The Supreme Court has twice concluded that political parties do not have a First Amendment right to party designation of their nominees on a ballot. Wash. State Grange v. Wash. State Republican Party, 552 U.S. 442, 453 n. 7, 128 S.Ct. 1184, 170 L.Ed.2d 151 (2008); Timmons v. Twin Cities Area New Party, 520 U.S. 351, 362–63, 117 S.Ct. 1364, 137 L.Ed.2d 589 (1997). In Timmons, a minor political party challenged Minnesota's “fusion ban,” which prohibited an individual from appearing on the ballot as the candidate of more than one party. 520 U.S. at 354, 117 S.Ct. 1364. The Court reasoned that although the fusion ban “prevents the New Party from using the ballot to communicate to the public that it supports a particular candidate who is already another party's candidate,” the New Party's constitutional rights were not severely burdened, because political parties do not have a right to use the ballot to convey a particularized message. Id. at 362–63, 117 S.Ct. 1364. The Court further explained that “[b]allots serve primarily to elect candidates, not as forums for political expression.” Id. at 363, 117 S.Ct. 1364. Likewise, in Washington State Grange, the Court reaffirmed its holding in Timmons. 552 U.S. at 453 n. 7, 128 S.Ct. 1184 (citing Timmons, 520 U.S. at 362–63, 117 S.Ct. 1364). The Court upheld a Washington initiative permitting candidates to identify themselves on the ballot by their party preference regardless of whether they were endorsed by, or were even members of, that party. Id. at 447, 128 S.Ct. 1184. In doing so, the Court explained that it was “unexceptionable” that political parties in Washington could no longer indicate their nominees on the ballot, reiterating that “[t]he First Amendment does not give political parties a right to have their nominees designated as such on the ballot.” Id. at 453 n. 7, 128 S.Ct. 1184 (citing Timmons, 520 U.S. at 362–63, 117 S.Ct. 1364).

The plaintiffs' arguments here fail for the same reason that the New Party's arguments in Timmons failed: a political party has no First Amendment right to use the general-election ballot for expressive activities. The plaintiffs attempt to distinguish Timmons by arguing that “[t]he burdens imposed by Ohio's prohibition on all political parties is self-evidently larger than Minnesota's fusion ban in Timmons, which burdened the associational rights only of minor political parties.” However, an election law that treats political parties differently raises more constitutional concerns, not fewer, than a regulation that treats all political parties equally. Cf. Rosen v. Brown, 970 F.2d 169, 175 (6th Cir.1992). Thus, because a political party has no right to use the ballot itself to educate voters, § 3505.04 does not significantly impede a political party's ability to spread its message to voters who do not know the winner of the party's primary.

The plaintiffs infer from the fact that the Timmons Court proceeded to engage in Anderson–Burdick balancing that the inability to designate nominees on a ballot imposes at least some burden on political parties' rights. See 520 U.S. at 363, 364–69, 117 S.Ct. 1364. Even granting such an inference, any burden that § 3505.04 does place on political parties is minimal because of the extensive remaining ways that a party may convey its message. Indeed, in arguing that § 3505.04 does nothing to further Ohio's stated interest in minimizing partisanship in judicial elections, the plaintiffs themselves have highlighted how much latitude Ohio's election scheme gives political parties to champion their judicial nominees and educate voters after the primaries—and therefore how little § 3505.04 actually burdens them. The plaintiffs explain that political parties may publicly endorse and campaign for their primary winners before the general election and distribute model ballots at polling places for voters to take with them into the voting booth, and that Ohio's county boards of elections list candidates' party affiliations on the boards' websites. Because of the numerous ways by which political parties may still involve themselves in the general election, the inability to educate voters on the ballot itself imposes at most a minimal burden on political parties.

The “voter drop off” phenomenon also fails to demonstrate that § 3505.04 burdens First and Fourteenth Amendment interests of political parties—at least in any way that is not inherently balanced by achievement of the nonpartisan goal of the statute. Instead of working to show how much parties' rights are burdened, the argument works to show how much § 3505.04 does, in fact, further Ohio's interest in minimizing partisanship in judicial elections. The plaintiffs complain that “uninformed voters simply do not vote in judicial races ... when there is no partisan identifier to rely upon.” However, the voter drop off of which plaintiffs complain may just as readily indicate that Ohio's election scheme is fulfilling its stated purpose: Fewer votes are cast solely in reliance on the judicial candidates' party affiliations than would be cast if the elections were partisan, thereby reducing partisanship in judicial elections. The votes that the plaintiffs complain that they are losing are almost by definition votes that would have been based on only the candidates' party affiliations—the very thing that Ohio is attempting to minimize. In short, a political party has many other means at its disposal to educate its members besides relying on uninformed voters to cast party-line ballots.

Moreover, § 3505.04 does not prevent voters from voting based on party affiliation or from casting a party-line ballot. As detailed above, political parties may endorse candidates and campaign on their behalf throughout the general election, and candidates may publicly express their affiliations with their political parties. This case is therefore not like Eu v. San Francisco Cty. Democratic Cent. Comm., 489 U.S. 214, 223, 109 S.Ct. 1013, 103 L.Ed.2d 271 (1989), in which a total ban on primary endorsements by political parties “directly hamper[ed] the ability of a party to spread its message and ham[strung] voters seeking to inform themselves about the candidates and the campaign issues.” Here, unlike in Eu, voters remain able to ascertain a judicial candidate's political party affiliation from a variety of different media and even to rely solely upon a party's endorsement of a candidate while casting their votes. Section 3505.04 only inhibits a voter's ability to rely exclusively on the general-election ballot to learn about the candidates' party affiliations. Such a burden is minimal, and is further alleviated by voters' ability to bring sample ballots prepared by the political parties with them into the voting booth.

Ohio judicial candidates are also entirely free to associate themselves with the parties of their choice and express their party affiliations publicly in forums other than the general-election ballot. As the district court explained, candidates are free to associate with political parties by “running in partisan primaries, attending political gatherings, participating in fund-raising activities sponsored by political parties, appearing on political parties' sample ballots, and saying that they are a member[ ] of or are endorsed by a political party.” Brunner, 24 F.Supp.3d. at 692. Unlike in Carey, in which we held unconstitutional a judicial ethics canon that precluded judicial candidates from publicly announcing their political party affiliations, 614 F.3d at 201–03, § 3505.04 does not prevent candidates from announcing their party affiliations to voters. Instead, § 3505.04 only prohibits candidates from having their party affiliations announced by the state in one place: on the general-election ballot. Thus, unlike the content-based restriction in Carey, § 3505.04 merely limits the forums in which judicial candidates may express themselves. Indeed, it stretches the idea of “candidate expression” to include the way that the state identifies a candidate on the ballot. “Ballots serve primarily to elect candidates, not as forums for political expression.” Timmons, 520 U.S. at 363, 117 S.Ct. 1364.

Our decision in Rosen does not support the idea that there is a substantial burden on the plaintiffs in this case. In Rosen, we held that an Ohio voting regulation that barred nonparty legislative candidates from being designated as “Independent” on the general-election ballot, while allowing Republican and Democratic candidates to be identified by their parties on the ballot, was unconstitutional. 970 F.2d at 171. While we noted that “party identification is the single most important influence on political opinions and voting,” id. at 172, we also noted that a state could bar all political designations on ballots, id. at 175. Thus, Rosen stands for the limited proposition that although a state may refuse to include party labels on ballots entirely, once it chooses to allow them, it must do so in a nondiscriminatory manner. See id.

In short, because of the various means other than the general-election ballot by which political parties and judicial candidates may communicate party-affiliation information to voters, and because of the practical nature of ballots versus other sources of information, § 3505.04 places at most only minimal burdens on the First and Fourteenth Amendment rights of political parties, judicial candidates, and voters in Ohio judicial elections.

The minimal burden is outweighed by § 3505.04's advancement of the important interest of reducing partisanship in judicial elections. A minimally burdensome and nondiscriminatory election law such as § 3505.04 is constitutional if it advances “important regulatory interests,” Burdick, 504 U.S. at 434, 112 S.Ct. 2059, a form of review akin to rational-basis review. Green Party of Tenn., 767 F.3d at 546. Under such review, plaintiffs “bear[ ] a heavy constitutional burden” to demonstrate that a state's election law is unconstitutional. Schrader v. Blackwell, 241 F.3d 783, 790–91 (6th Cir.2001).

Ohio's stated interest of minimizing partisanship in judicial elections is an important one. In light of the different role that judges must play from that of their legislative and executive counterparts, we have recognized that minimizing partisanship in judicial elections can be a compelling state interest. Carey, 614 F.3d at 201. Likewise, in recognizing that judges must exercise neutrality and independence in the performance of their duties, the Supreme Court held that the closely related interest of maintaining public perception of judicial integrity and impartiality is “a state interest of the highest order.” Williams–Yulee v. Fla. Bar, ––– U.S. ––––, 135 S.Ct. 1656, 1666, 191 L.Ed.2d 570 (2015) (quoting Caperton v. A.T. Massey Coal Co., 556 U.S. 868, 889, 129 S.Ct. 2252, 173 L.Ed.2d 1208 (2009)). Plaintiffs do not really deny the importance of this state interest. Instead, they contend that § 3505.04 does nothing to further that interest. On the contrary, § 3505.04 does advance Ohio's interest by discouraging purely party-line voting and signaling to voters that judges are responsible only to the law and not to their political parties.

First, § 3505.04 furthers Ohio's interest in minimizing partisanship in judicial elections by working to reduce the number of votes cast solely in reliance on party affiliation. As explained above, the very fact that voter drop off is greater in judicial races than in other partisan low-information races supports the conclusion that fewer individuals in judicial elections rely solely on party affiliation to cast their votes than if the candidates' affiliations were designated on the general-election ballot. The disparity in the number of voters who “drop off” in the nonpartisan judicial races compared to those who drop off in the partisan low-information races suggests the existence of a significant number of voters who vote in low-information races based only on party designation. The fact that these individuals do not vote in low-information judicial races indicates that Ohio's system is working: Judicial elections are probably decided based solely on partisan politics to a lesser degree than if the party designations were included on the ballot.

Second, removing party designations from the general-election ballot furthers Ohio's interest in minimizing partisanship in judicial elections because it communicates to voters that judges may not be “concerned with party platforms or party expediency” or “depart from the plain provisions [of the law], no matter how much [they] may be opposed to the principles or purposes of it.” Weinberger, 99 N.E. at 1085. Divorcing judicial candidates from their party affiliations in the voting booth may enhance the public's perception of judicial integrity by conveying the impression that judges are, at the end of the day, bound by the law rather than by their political ideologies. Minimizing this specific type of partisanship—voters' association of judicial candidates with their political parties at the climactic moment of choice in the same way that voters associate executive or legislative candidates with their parties—may also decrease the chance that voters will blur the distinctions between officials who are beholden to certain constituencies and those who are required to be neutral and independent. Elected judges may feel freer to rule in good conscience against their former fellow-partisans when the citizen actions that finally gave them their judicial role were formally nonpartisan. Furthermore, by denying partisan cues on the ballot, Ohio's electoral system might encourage voters to inform themselves about the qualifications of judicial candidates prior to stepping into the booth on election day. While the efficacy of such a policy might be up for debate, the Constitution does not “require elaborate, empirical verification of the weightiness of the State's asserted justifications.” Timmons, 520 U.S. at 364, 117 S.Ct. 1364.

Contrary to the plaintiffs' claims that the ability of political parties to remain involved in the general election creates a hole in the efficacy of the law, Ohio's two-tiered regime for electing judges is not internally inconsistent at all. Parties naturally want their members to fill government positions. A very logical judicial election scheme could accordingly enlist parties to come up with candidates best suited to win a nonpartisan election. Such a scheme would take advantage of the obvious ability of parties to recruit, cull, and vet candidates, while incentivizing the parties to come up with candidates who can win a nonpartisan race. The ultimate purpose would be to promote the reality and appearance that judges are motivated by an allegiance to law rather than to party. Ohio's scheme is roughly consistent with such an approach, and there is nothing internally inconsistent about it. 14 The plaintiffs argue that § 3505.04 does not further Ohio's interest in any way, because Ohio has not completely eliminated partisan politics from its judicial elections. However, “[a] State need not address all aspects of a problem in one fell swoop,” Williams–Yulee, 135 S.Ct. at 1668, especially where, as is the case here, a state's law is reviewed under a standard similar to rational-basis review, cf. Williamson v. Lee Optical of Okla., Inc., 348 U.S. 483, 487–88, 75 S.Ct. 461, 99 L.Ed. 563 (1955). Indeed, even when strict scrutiny applies, laws can be constitutional even though they “conceivably could have restricted even greater amounts of speech in service of their stated interests.” Williams–Yulee, 135 S.Ct. at 1668. Despite the plaintiffs' claims, their own evidence of voter drop off and the message that nonpartisan ballots communicate to voters show that Ohio's interest in minimizing partisanship in judicial elections is being furthered, at least to some degree.

Additionally, the fact that it is still possible for Ohio voters to rely on nothing more than a judicial candidate's party affiliation to cast their votes does not fatally undermine the reasonableness of the scheme. Totally eliminating such reliance would require First Amendment infringements that would not pass muster under strict scrutiny. For instance, in Republican Party of Minn. v. White (White I ), 536 U.S. 765, 122 S.Ct. 2528, 153 L.Ed.2d 694 (2002), the Supreme Court found unconstitutional a judicial ethics canon that prevented judicial candidates from “announc[ing] his or her views on disputed legal or political issues,” id. at 770, 779–80, 122 S.Ct. 2528, and in Carey, we held unconstitutional a judicial ethics canon that prevented candidates from publicly announcing their political party affiliations. 614 F.3d at 202. But the fact that the First Amendment permits candidates to campaign, and citizens to vote, on partisan bases in formally nonpartisan elections cannot mean that such formally nonpartisan elections are unconstitutional.

In sum, Ohio's interest in minimizing partisanship in judicial elections is sufficient to justify any minimal constitutional burden imposed by § 3505.04. The judgment of the district court is accordingly affirmed.