Pope v. County of Albany

REENA RAGGI, Circuit Judge: Plaintiffs, black and Hispanic registered voters in Albany County, sue the County and the County Board of Elections (collectively “defendants”) in the United States District Court for the Northern District of New York (Lawrence E. Kahn, Judge) for enacting a redistricting plan for the Albany County Legislature (“Local Law C”) in response to the 2010 United States census that allegedly dilutes black and Hispanic voting strength in violation of Section 2 of the Voting Rights Act (“VRA”), 42 U.S.C. § 1973, by failing to provide for five majority-minority districts (“MMDs”). Plaintiffs here appeal from the district court’s denial of a preliminary injunction barring defendants from implementing Local Law C with respect to elections held in September and November of 2011. See Pope v. Cnty. of Albany, No. 11-cv-736 (LEKXDRH), 2011 WL 3651114 (N.D.N.Y. Aug. 18, 2011). For the reasons stated herein, we conclude that the appeal is not moot, even though the challenged elections have taken place. Nevertheless, it is without merit because the district court acted within its discretion in concluding that plaintiffs failed, on the record presented in support of their preliminary injunction motion, to demonstrate a likelihood of success on the third factor of a vote dilution claim identified in Thornburg v. Gingles, 478 U.S. 30, 51, 106 S.Ct. 2752, 92 L.Ed.2d 25 (1986), i.e., bloc voting by the white majority sufficient to defeat minority voters’ preferred candidate. At the same time, we identify a concern with the district court’s apparent demand for more than a simple majority of minority group members of voting age within the proposed districts to satisfy the first Gingles factor. Because the district court will have to apply the Gingles analysis at trial in ultimately resolving this case, we clarify the law applicable to the first factor at the same time that we affirm the district court’s denial of preliminary relief. 1. Factual Background A. Redistricting of the Albany County Legislature The Albany County Legislature consists of representatives elected every four years from 39 single-member districts. See Albany County Charter Art. 2, §§ 201, 206. After each decennial national census, the County legislature appoints a reapportionment commission charged with revising district lines to account for shifts in population. See id. § 207. The County’s past redistricting efforts have routinely triggered litigation. A challenge to the redistricting plan adopted after the 1990 census, which provided for only one district in which blacks would constitute a majority of the voting age population (“VAP”), resulted in the County entering into a consent judgment that prevented implementation of the plan and provided for the establishment of three MMDs in each of which blacks and Hispanics together constituted at least 63% of the VAP. See Consent J. & Decree, NAACP v. Albany Cnty., No. 91-cv-1288 (CGC) (N.D.N.Y. Nov. 13, 1991), ECF No. 2. When, following the 2000 census, the County enacted a redistricting plan that maintained three MMDs, the district court *568 enjoined implementation, concluding that Section 2 required the County to create a fourth MMD. See Arbor Hill Concerned Citizens Neighborhood Ass’n v. Cnty. of Albany, 281 F.Supp.2d 436 (N.D.N.Y. 2003); see also Arbor Hill Concerned Citizens Neighborhood Ass’n v. Cnty. of Albany, 289 F.Supp.2d 269, 276 (N.D.N.Y.2003) (stating that County’s plan “flagrantly violated the rights of minority voters”). To the extent the district court there did not think that it could order special elections based on the revised plan containing four MMDs, this court reversed and itself ordered such elections. See Arbor Hill Concerned Citizens Neighborhood Ass’n v. Cnty. of Albany, 357 F.3d 260, 263 (2d Cir.2004). After the 2010 census, the County enacted the redistricting plan challenged in this lawsuit, which maintains four MMDs. In doing so, the County followed the recommendation of its redistricting commission, which had retained a redistricting expert, John Merrill, to prepare a plan based on the new census information. The commission held public hearings, at which minority groups urged the creation of a fifth MMD. One such group, the Arbor Hill Environmental Justice Corporation, submitted a proposed redistricting plan to that effect (“AHE J Plan”). Merrill advised the Commission that, although he attempted to create a fifth MMD, the task was impossible because of the dispersion of the minority population, which he treated as including both blacks and Hispanics. See infra at 572 n. 5, 577 n. 11 (discussing issues presented by choice of minority group). Merrill deemed the AHEJ Plan unsatisfactory because, in his view, it violated a number of redistricting principles and constituted racial gerrymandering. 1 Accordingly, on May 19, 2011, the Commission adopted a final redistricting plan proposed by Merrill that retained four MMDs. The Legislature adopted the plan as Local Law C on May 23, 2011. B. Motion for Preliminary Injunction Plaintiffs filed this Section 2 challenge to Local Law C on June 29, 2011, and on July 15, 2011, they moved for a preliminary injunction to prevent defendants from “conducting the petition process, gathering signatures for elections, holding and conducting any further primary or general elections, or implementing or certifying elections in the districts under the current scheme implemented in 2004 or those created by Local Law C for any term of office commencing January 1, 2012.” Notice of Mot. at 1-2, Pope v. Cnty. of Albany, No. 11-ev-736, ECF No. 12. Plaintiffs maintained that they were likely to succeed on their claim of minority vote dilution whether the relevant minority population aggregated blacks and Hispanics or was limited to blacks alone. After evidentiary hearings and written submissions, the district court denied a preliminary injunction on August 18, 2011. Plaintiffs filed a timely notice of appeal and sought an expedited briefing schedule, which was granted, with *569 oral argument heard on December 16, 2011. In the interim, primary and general elections for the County legislature on the basis of Local Law C were held on September 13, 2011, and November 8, 2011, respectively. On November 14, 2011, defendants moved to dismiss the appeal as moot. On November 30, 2011, plaintiffs moved in this court for a preliminary injunction pending appeal to prevent defendants from “conducting elections or related processes, or qualifying and/or certifying individuals elected in the November 8, 2011 elections for Albany County legislature to take office.” Appellant Mot. for Prelim. Inj. Pending Appeal at 1, Pope v. Cnty. of Albany, No. 11-3439-cv (2d Cir. Nov. 30, 2011), ECF No. 82. On December 28, 2011, this court summarily denied plaintiffs’ motion for a preliminary injunction pending appeal. II. Discussion A. Mootness Because the question implicates our jurisdiction, we first address defendants’ motion to dismiss the appeal as moot in light of the fact that the very 2011 elections that plaintiffs sought preliminarily to enjoin have now been conducted. “The occurrence of the action sought to be enjoined normally moots the request for preliminary injunctive relief because this Court has no effective relief to offer once the action has occurred.” Moore v. Consol. Edison Co. of N.Y., Inc., 409 F.3d 506, 510 (2d Cir.2005) (Sotomayor, J.) (internal quotation marks omitted). This principle does not apply, however, when “we do not lack the ability to offer effective relief” because we “can feasibly restore the status quo.” Id. at 509-10 (emphasis in original). This conclusion is a “natural, if not inevitable, extension of the well established principle that where a defendant with notice in an injunction proceeding completes the acts sought to be enjoined the court may by mandatory injunction restore the status quo.” Id. at 510 n. 4 (internal quotation marks omitted). In this case, defendants unquestionably had notice that plaintiffs were seeking to enjoin the 2011 elections. Indeed, the parties were engaged in an appeal from the denial of such relief when defendants proceeded with the 2011 elections. See id. (rejecting mootness challenge where district court denied preliminary injunction that sought to prevent plaintiffs termination and plaintiff was then terminated before appeal was heard). We do not fault defendants’ actions in this respect; no court order barred them from proceeding with the elections. We note only that defendants were on notice that plaintiffs were still pursuing preliminary injunctive relief. Thus, we will not dismiss the case as moot if we are satisfied that, should plaintiffs succeed on this appeal, it would be possible to restore the status quo. In considering that question, we are mindful of precedent recognizing that “[i]t is within the scope of [a federal court’s] equity powers to order a governmental body to hold special elections to redress violations of the VRA.” Arbor Hill Concerned Citizens Neighborhood Ass’n v. Cnty. of Albany, 357 F.3d at 262. Implicit in the power to order special elections is the authority to void elections conducted in violation of the VRA. It was not necessary to authorize such relief in Arbor Hill because the district court had enjoined elections under the challenged plan before we ordered special elections. See id. Nevertheless, Arbor Hill cites a case from a sister circuit holding that the equity power to order special elections applies even when the original election has already taken place. See Bell v. Southwell, 376 F.2d 659, 665 (5th Cir.1967) (holding that dis *570 trict court has power to void and order new elections for violations of VRA and Constitution). Other courts have similarly-ruled. See Stewart v. Taylor, 104 F.3d 965, 970 (7th Cir.1997) (stating that, despite holding of challenged election, court could order new election if plaintiffs motion for preliminary injunction has merit); Hamer v. Campbell, 358 F.2d 215, 222 (5th Cir.1966) (“[Hjaving concluded that the ... election should have been enjoined, we now must set it aside in order to grant appellants full relief in the same manner as if the said election had been enjoined.” (internal quotation marks omitted)). We agree with these cases, and thus conclude that this appeal is not moot because, if plaintiffs were to succeed on the merits, it would be possible for the court to restore the status quo by vacating the 2011 election. 2 Courts may understandably hesitate to void elections that have already been conducted as a form of preliminary equitable relief, preferring to take such action only upon a final determination that plaintiffs are entitled to permanent relief. That, however, is an argument for consolidating preliminary injunction proceedings with a trial on the merits where possible, see Fed.R.Civ.P. 65(a)(2), or for denying a preliminary injunction as a matter of equitable discretion, see Winter v. Natural Res. Def. Council, Inc., 555 U.S. 7, 32, 129 S.Ct. 365, 172 L.Ed.2d 249 (2008). It does not render the preliminary injunction motion, or the appeal from a denial of such a motion, moot. See Friends of the Earth, Inc. v. Laidlaw Envtl. Servs. (TOC), Inc., 528 U.S. 167, 193, 120 S.Ct. 693, 145 L.Ed.2d 610 (2000) (recognizing distinction between court’s power to order effective relief and court’s discretion to do so). Accordingly, we deny defendants’ motion to dismiss the appeal as moot, and we proceed to discuss the merits. B. The Preliminary Injunction Motion 1. Standard of Review Where, as here, a party seeks a preliminary injunction against government action taken in the public interest pursuant to a statutory scheme, a moving party must demonstrate that (1) he is likely to succeed on the merits of the underlying claim, (2) he will suffer irreparable harm absent injunctive relief, and (3) the public interest weighs in favor of granting the injunction. See Oneida Nation of N.Y. v. Cuomo, 645 F.3d 154, 164 (2d Cir.2011). 3 We review the denial of a preliminary injunction motion deferentially for abuse of discretion, which we will identify only *571 when the district court decision rests on an error of law or a clearly erroneous finding of fact. See id. On this appeal, we focus on the district court’s determination that plaintiffs failed to demonstrate the requisite likelihood of success on the merits of their Section 2 claim, a matter that implicates both questions of law and fact. 2. Analyzing Section 2 Claims by Reference to the Gingles Factors Section 2 of the Voting Rights Act, as amended, states in relevant part as follows: (a) No voting qualification or prerequisite to voting or standard, practice, or procedure shall be imposed or applied by any State or political subdivision in a manner which results in a denial or abridgement of the right of any citizen of the United States to vote on account of race or color.... (b) A violation of subsection (a) of this section is established if, based on the totality of circumstances, it is shown that the political processes leading to nomination or election in the State or political subdivision are not equally open to participation by members of a class of citizens protected by subsection (a) of this section in that its members have less opportunity than other members of the electorate to participate in the political process and to elect representatives of their choice. The extent to which members of a protected class have been elected to office in the State or political subdivision is one circumstance which may be considered: Provided, That nothing in this section establishes a right to have members of a protected class elected in numbers equal to their proportion in the population. 42 U.S.C. § 1973; see also id. § 1973b(f)(2) (extending protection to language minority groups). The Supreme Court has recognized that a redistricting scheme may violate Section 2 by impermissibly diluting the minority vote. See Thornburg v. Gingles, 478 U.S. 30, 106 S.Ct. 2752. Gingles instructs that a viable Section 2 claim of vote dilution depends on plaintiffs’ preliminary showing of three factors: (1) that the relevant minority is “sufficiently large and geographically compact to constitute a majority in a single-member district,” (2) that the relevant minority is “politically cohesive,” and (3) that the majority “votes sufficiently in a bloc to enable it ... in the absence of special circumstances ... usually to defeat the minority’s preferred candidate.” Id. at 50-51, 106 S.Ct. 2752. A-though the Court in Gingles applied these factors to a challenged at-large multi-member districting plan, it has since applied the same factors to claims of vote dilution under single-member districting plans, such as the one at issue in this case. See League of United Latin Am. Citizens (LU-LAC) v. Perry, 548 U.S. 399, 425, 126 S.Ct. 2594, 165 L.Ed.2d 609 (2006). If a plaintiff satisfies the three preliminary Gingles factors, he must then satisfy the ultimate Section 2 inquiry by showing from the “totality of the circumstances,” that members of the identified racial group “have less opportunity than do other members of the electorate” to elect a candidate of their choice. Id. at 425-26, 126 S.Ct. 2594. 4 *572 3. The District Court’s Determination that Plaintiffs Failed To Demonstrate Likely Success in Establishing the Gingles Factors In denying a preliminary injunction of the application of Local Law C, the district court found that plaintiffs had failed to demonstrate likely success on all three preliminary Gingles factors. Insofar as plaintiffs had attempted to carry their burden by defining the relevant minority community to include blacks and Hispanics, the district court found the record evidence insufficient to establish the political cohesiveness of these two groups, the second Gingles factor. 5 Analyzing plaintiffs’ motion by reference to a minority group community denominated by the Department of Justice (“DOJ”) as “Non-Hispanic Black,” the district court found that plaintiffs had failed to show that group to be a sufficiently large percentage of the VAP in five MMDs to satisfy the first Gingles *573 factor. 6 In reaching this conclusion, the district court acknowledged that the proposed redistricting plans showed a DOJ Non-Hispanic Black YAP exceeding 50% in more than five MMDs. 7 Nevertheless, it concluded that more than the bare majority reflected in some of these MMDs was necessary to satisfy the first Gingles factor. Finally, the district court concluded that plaintiffs had failed to adduce sufficient evidence of bloc voting by the white *574 majority population in Albany County to satisfy the third Gingles factor. We identify legal error in the district court’s demand for plaintiffs to show more than a simple majority of the relevant minority group at the first step of Gingles analysis. At the same time, we conclude that the district court acted within its discretion in denying plaintiffs a preliminary injunction based on their failure to establish majority bloc voting. a. The First Gingles Factor Before the district court, defendants did not dispute the- minority VAP percentages shown in the five MMDs proposed in the AHEJ plan and plaintiffs’ Illustrative Plan 3. Rather, defendants argued that the bare majority DOJ Non-Hispanic Black VAPs reflected in the two plans failed to satisfy the “sufficiently large” requirement of the first Gingles factor. In support, they relied on the cross-examination testimony of plaintiffs’ expert, Dr. Baodong Liu, a professor of political science at the University of Utah, who testified that in the “academic world,” a “dominant black district” is defined as one in which blacks constitute at least 55% of the VAP. Liu Test., Hr’g Tr. at 103. [A 377] Defendants also pointed to the briefs of plaintiffs in the Arbor Hill litigation (challenging Albany County’s redistricting following the 2000 census), which questioned a remedial plan creating bare majority-minority districts in light of factors, such as lower voter turnout and registration rates in the minority communities at issue, which would render it impossible for minorities to elect candidates of their choice. See Pis.’ Objections to Report & Recommendation re: Remedial Redistricting Plan at 18, Arbor Hill Concerned Citizens Neighborhood Ass’n v. Cnty. of Albany, No. 03-cv-502 (NAMXDRH) (N.D.N.Y. Oct. 1, 2003), ECF No. 49. The district court agreed that “a bare-bones 50.44 percent” DOJ Non-Hispanic Black VAP, as indicated for proposed MMD 4 under the AHEJ plan, was insufficient to satisfy the first Gingles factor, “particularly where Plaintiffs’] own expert ... testified that a 55 percent minority population is required” for a district to be considered “ ‘dominant.’ ” Pope v. Cnty. of Albany, 2011 WL 3651114, at *3. While the district court acknowledged that the plurality opinion in Bartlett v. Strickland, 556 U.S. 1, 129 S.Ct. 1231, 173 L.Ed.2d 173 (2009), signaled that a bare majority could satisfy the first Gingles factor, see id. at 18, 129 S.Ct. 1231, the district court concluded that the opinion was non-binding and, in any event, irrelevant, as it pertained to “crossover districts,” which were not at issue in this case, Pope v. Cnty. of Albany, 2011 WL 3651114, at *3. We disagree and here clarify that no super-majority is necessary for plaintiffs to satisfy the first Gingles factor. In so ruling, we note that the first Gingles factor is only the initial step in a larger inquiry that ultimately asks (1) whether the totality of the circumstances evidences “an inequality in the opportunities” enjoyed by minority voters that prevents them from “electing] their preferred candidates,” Thornburg v. Gingles, 478 U.S. at 47, 106 S.Ct. 2752, and, if so, (2) how to “devis[e] a sound remedy” for that inequality, Johnson v. De Grandy, 512 U.S. 997, 1008, 114 S.Ct. 2647, 129 L.Ed.2d 775 (1994). The relative size of a minority group’s majority in a district may well be among the totality of circumstances that can inform the ultimate determination of vote dilution and the appropriate remedy. But it is simply the fact, not the size, of a minority group’s majority presence in a proposed district that permits a plaintiff to satisfy the first of Gingles’s preliminary requirements. *575 The first Gingles factor asks only whether the relevant minority group is “sufficiently large ... to constitute a majority” of the VAP. Id. at 50, 106 S.Ct. 2752. A majority is commonly understood as anything exceeding 50%. See Webster’s Third New International Dictionary 1363 (1986) (defining “majority” as “a number greater than half of a total”). As the Supreme Court has explained, this initial majority-minority inquiry serves a useful gate-keeping role by identifying, in broad terms, when a minority group has at least “the potential to elect representatives in the absence of the challenged structure or practice.” Thornburg v. Gingles, 478 U.S. at 50 n. 17, 106 S.Ct. 2752 (emphasis in original). Without such a potential, the minority group “cannot claim to have been injured by that structure or practice.” Id. Thus, a simple majority rule usefully serves at the outset to screen out cases in which there is no point in undertaking a full Section 2 analysis. Toward that end, it asks a preliminary question; it does not attempt to answer the ultimate one. Defendants nevertheless maintain that the district court permissibly demanded a super-majority showing to satisfy the first Gingles factor because a minority group that constitutes only a simple majority of the VAP may not realistically have the potential to elect representatives of its choice, particularly where minority voter registration rates and turnout are low. Such concerns may well warrant careful consideration when a court reviews the totality of the circumstances in deciding the ultimate Gingles inquiry. See id. at 45, 106 S.Ct. 2752 (holding that list of typical factors relevant to assessing totality of circumstances is not comprehensive and that such inquiry requires “searching practical evaluation of the past and present reality” (internal quotation marks omitted)); accord NAACP v. City of Niagara Falls, 65 F.3d 1002, 1008 (2d Cir.1995); see also Johnson v. De Grandy, 512 U.S. at 1012, 114 S.Ct. 2647 (stating that district court should analyze totality of circumstances “without isolating any ... arguably relevant facts from the act of judgment”). The concerns may also inform the appropriate remedy for a Section 2 violation. See Dickinson v. Ind. State Election Bd., 933 F.2d 497, 503 (7th Cir.1991) (observing that determination that super-majority is not necessary at “liability stage” is distinct from court’s ability to “consider, at the remedial stage, what type of remedy is possible based on ... minority voter registration and turn-out rates” (emphasis in original)); accord Bone Shirt v. Hazeltine, 461 F.3d 1011, 1019 (8th Cir.2006); see also Ketchum v. Byrne, 740 F.2d 1398, 1413 (7th Cir.1984) (identifying error in district court’s failure at remedial stage to provide for super-majority in light of minority group’s lower registration and turnout rates). 8 The first Gingles factor, how *576 ever, does not contemplate such a holistic review of the total evidence. The “ultimate end” of the first Gingles factor is simply “to prove that a solution is possible, and not necessarily to present the final solution to the problem.” Bone Shirt v. Hazeltine, 461 F.3d at 1019 (internal quotation marks omitted). Thus, the first Gingles question is straightforward and statistical: does the identified minority group form at least a simple majority of the relevant population in the proposed district? In urging us to conclude otherwise, defendants emphasize that the Supreme Court has described the first Gingles factor as requiring “a sufficiently large minority population to elect candidates of its choice.” Johnson v. De Grandy, 512 U.S. at 1008, 114 S.Ct. 2647 (emphasis added); accord LULAC v. Perry, 548 U.S. at 430, 126 S.Ct. 2594. We are not persuaded that the highlighted language transforms the first Gingles factor as defendants urge. We understand Supreme Court references to a “sufficiently large minority” to be reiterating, not departing from, Gingles’s observation that a minority group is “sufficiently large” when it “constitute^] a majority.” Thornburg v. Gingles, 478 U.S. at 50, 106 S.Ct. 2752. We do not understand these cases, or any other precedent, to require plaintiffs to show a super-majority to satisfy the first Gingles factor. Indeed, the reasoning of the plurality opinion in Bartlett v. Strickland, 556 U.S. 1, 129 S.Ct. 1231, 173 L.Ed.2d 173, even if not controlling, persuades us that it would be inappropriate to recognize such a requirement here. In Bartlett, the Supreme Court considered whether Section 2 required the creation of a district in which the identified minority group, though not a majority, was sufficiently large that, when coupled with reliable white “crossover” votes, it would be capable of electing the representatives of its choice. The plurality opinion — authored by Justice Kennedy, who also wrote those parts of the LULAC decision commanding a majority — espoused a “majority-minority” rule that required the minority to show that it was at least 50% of the VAP in the proposed district. In explaining its reasoning, the plurality provides one of the fullest discussions of the purpose of the first Gingles factor and, in the process, indicates that the factor demands no more than a simple majority to proceed with further Section 2 analysis: [T]he majority-minority rule relies on an objective, numerical test: Do minorities make up more than 50 percent of the voting-age population in the relevant geographic area? ... Where an election district could be drawn in which minority voters form a majority but such a district is not drawn, or where a majority-minority district is cracked by assigning some voters elsewhere, then— assuming the other Gingles factors are also satisfied — denial of the opportunity to elect a candidate of choice is a present and discernible wrong.... Not an arbitrary invention, the majority-minority rule has its foundation in principles of democratic governance. The special significance, in the democratic process, of a *577 majority means it is a special wrong when a minority group has 50 percent or more of the voting population and could constitute a compact voting majority but, despite racially polarized bloc voting, that group is not put into a district. Id. at 18-19,129 S.Ct. 1231. 9 Those of our sister circuits to have considered the question have agreed that nothing more than a simple majority is necessary to satisfy the first Gingles factor. See Bone Shirt v. Hazeltine, 461 F.3d at 1019 [ (8th Cir.) ] (rejecting need to show “super-majority status”); Westwego Citizens for Better Gov’t v. City of Westwego, 946 F.2d 1109, 1117 (5th Cir.1991) (concluding that 52.8% VAP is sufficient); Dickinson v. Ind. State Election Bd., 933 F.2d at 503 [ (7th Cir.) ] (concluding that 50.27% VAP is sufficient, and rejecting need for super-majority at threshold stage). 10 This precedent reinforces our own conclusion that the first Gingles factor can be satisfied by showing that an identified minority group forms a simple majority of the relevant population of a proposed district. Both the AHEJ Plan and Illustrative Plan 3 show DOJ Non-Hispanic Blacks to constitute a majority in five proposed districts for representatives to the County legislature. See supra at 573 n. 7. 11 Thus, plaintiffs should not have been denied injunctive relief on this ground. 12 *578 b. The Third Gingles Factor To satisfy the third Gingles factor, plaintiffs must demonstrate that the white majority “votes sufficiently as a bloc to enable it — in the absence of special circumstances, such as the minority candidate running unopposed — usually to defeat the minority’s preferred candidate.” Thornburg v. Gingles, 478 U.S. at 51, 106 S.Ct. 2752; accord LULAC v. Perry, 548 U.S. at 425, 126 S.Ct. 2594. In reviewing the district court’s determination that plaintiffs failed to show a sufficient likelihood of success on the merits of this factor to warrant preliminary injunctive relief, we apply de novo review to its identification and application of law, but we review factual determinations only for clear error. NAACP v. City of Niagara Falls, 65 F.3d at 1008. Here the latter review predominates because the law, which the district court correctly identified, recognizes the need for some flexibility. As the Supreme Court has observed, “no simple doctrinal test” applies to the third Gingles factor because racial bloc voting can “vary according to a variety of factual circumstances.” Thornburg v. Gingles, 478 U.S. at 58, 106 S.Ct. 2752. The factual circumstance that here prompted the district court to conclude that plaintiffs failed to carry their preliminary injunction burden was unexplained omissions from the electoral data provided to their expert witness, Dr. Liu. In reaching his conclusion that Albany County exhibited racial polarization in biracial elections from 2004 to 2010, Dr. Liu analyzed fourteen elections from that seven-year time frame: twelve single-member elections and two multi-member elections. Operating on the assumption that an election was racially polarized if a majority of black voters, but only a minority of non-black voters, supported the black candidate, 13 Dr. Liu determined that nine of the twelve single-member elections exhibited racial polarization. 14 Without explaining how he defined racial polarization for mul *579 ti-member elections, Dr. Liu reported that, in the precincts he could analyze, in 2005, two black candidates in a field of six obtained 44% of the black vote and 29% of the non-Hispanic white vote; while in 2010, two black candidates in a field of six obtained 65% of the black vote and 31% of the non-Hispanic white vote. 15 Defendants did not employ competing expert testimony to challenge Dr. Liu’s opinion. Rather, they pointed to his failure to consider a number of biracial County elections won by black candidates in the 2004-2010 period. 16 At the preliminary *580 injunction hearing before the district court, Dr. Liu seemed generally unaware of these elections, testifying that he had relied on election data provided by Aaron Mair, the executive director of the Arbor Hill Environmental Justice Corporation, who submitted his own affidavit in support of plaintiffs’ motion for a preliminary injunction. Nothing in the record explains what criteria determined the election data provided to Dr. Liu. Dr. Liu, however, agreed that his analysis would be improved by more data and that the omitted data were relevant. On this record, the district court found that it was not persuaded that plaintiffs were likely to succeed on the Gingles bloc voting factor because Dr. Liu formed his racial polarization opinion without considering several countywide and citywide elections that he acknowledged provided relevant data. The court considered the omissions significant because it found that “[w]hile the Arbor Hill decision correctly concluded that whites [in Albany County had] voted as a bloc before and during 2003, there has clearly been a change in the voting patterns of white residents of the County in the last eight years.” Pope *581 v. Cnty. of Albany, 2011 WL 3651114, at *6. The court observed that, whereas no minority candidate had ever been nominated, much less elected, to public office in the County through 2003, in the years thereafter, black candidates had won a number of countywide and local offices, and Barack Obama won a majority of Albany County’s votes in the 2008 general presidential election. See id. 17 In challenging the district court’s determination that they failed to demonstrate likely success on the third Gingles factor, plaintiffs essentially fault the court for not according more weight to their expert’s opinion. The question of what weight to accord expert opinion is a matter committed to the sound discretion of the factfinder, and we will not second guess that decision on appeal absent a basis in the record to think that discretion has been abused, which is not this case. See United States v. Yousef, 327 F.3d 56, 126 (2d Cir.2003) (reviewing weight that district court assigned to expert testimony for abuse of discretion). A factfinder may certainly consider the bases for an expert’s opinion and may accord the opinion less, or even no, weight if the record suggests that the bases are defective, incomplete, or questionable. See NAACP v. City of Niagara Falls, 65 F.3d at 1020 (concluding that district court was entitled to find plaintiffs’ expert unpersuasive because her analysis covered only portion of relevant timeframe and she relied heavily on interviews with plaintiffs). To the extent plaintiffs suggest that defendants bore the burden to offer contradictory expert testimony, we disagree. It is plaintiffs’ burden to establish the Gingles factors, see Thornburg v. Gingles, 478 U.S. at 50-51, 106 S.Ct. 2752, and plaintiffs failed to satisfy that burden here, where the district court justifiably declined to rely on their expert, see NAACP v. City of Niagara Falls, 65 F.3d at 1020. To be sure, in the context of election challenges, the law does not require that expert opinions be supported by an exhaustive analysis of elections within the relevant period. “The number of elections that must be studied in order to determine whether voting is polarized will vary according to pertinent circumstances.” Thornburg v. Gingles, 478 U.S. at 57 n. 25, 106 S.Ct. 2752. Indeed, flexibility is particularly warranted when comprehensive data are not available. See id. (“[T]he fact that statistics from only one or a few elections are available for examination does not foreclose a vote dilution claim.”). Further, even when omitted data are available, a court may properly consider whether they pertain to exogenous elections or the ones specifically at issue. See Goosby v. Town Bd., 180 F.3d 476, 497 (2d Cir.1999) (concluding that exogenous elections “are less probative than elections involving the specific office that is the subject of the litigation” (internal quotation marks omitted)). Nevertheless, a factfinder’s ability to excuse omitted data in considering an expert opinion does not mean that it is compelled to credit the opinion or to find that it demonstrates the sufficient likelihood of success on the third Gingles factor to support a preliminary injunction. In making *582 a factual assessment of bloc voting on a limited record, the trial court could properly consider not only that exogenous election data were omitted from an expert’s analysis, but also the content of the data, as well as the reasons for their omission. Here, the omitted data, showing the success of black candidates in various elections conducted in Albany County during the relevant period, may have been more favorable to defendants than the data provided to Dr. Liu for his analysis. Further, the record provides no insights as to why these data were not provided to Dr. Liu, while data for similar elections were. We do not predict what Dr. Liu’s opinion would be if he were to include the omitted data in his analysis. Much less do we predict whether plaintiffs will be able to prove bloc voting or the other elements of a Section 2 claim when they make a full evidentiary presentation to the district court on final, rather than preliminary, presentation of their case. Evidence of black candidates’ electoral successes may be probative of “a general willingness of white voters to vote for black candidates.” Abrams v. Johnson, 521 U.S. 74, 93, 117 S.Ct. 1925, 138 L.Ed.2d 285 (1997) (internal quotation marks omitted). But such evidence does “not necessarily negate” a finding of bloc voting, particularly if “elections are shown usually to be polarized” or the success of minority candidates in particular elections can be explained by “special circumstances, such as the absence of an opponent [or] incumbency.” Thornburg v. Gingles, 478 U.S. at 57, 106 S.Ct. 2752. Further, where, as here, a Section 2 challenge is to a single-member redistricting plan, a court considering exogenous elections does so mindful that the ultimate “inquiry into the existence of vote dilution caused by submergence ... is district specific.” Id. at 59 n. 28, 106 S.Ct. 2752 (faulting district court for aggregating racial polarization numbers across several independently challenged multimember districts). These matters, however, warrant no further discussion here; they are for the district court to consider in any subsequent proceedings. We here conclude only that the district court acted within its discretion in finding, based on the record then before it, that plaintiffs failed to carry their burden as to the third Gingles factor because of questions raised by omitted data as to their expert’s bloc voting analysis. III. Conclusion To summarize, we conclude as follows: 1. Because it would be possible to restore the parties to the status quo if plaintiffs were to prevail on this appeal, the appeal is not moot. 2. Because the first factor for a Section 2 claim identified in Thornburg v. Gingles does not require plaintiffs to show that the minority group is a super-majority in a proposed MMD, plaintiffs satisfied that factor by showing that the DOJ Non-Hispanic Black VAP in five proposed MMDs exceeded 50%. 3. Nevertheless, because the district court did not clearly err in finding that plaintiffs had failed to demonstrate likely success on the third Gingles factor, white bloc voting, it acted within its discretion in denying a preliminary injunction of the 2011 election for the Albany County Legislature. Defendants’ motion to dismiss the case as moot is denied, and the judgment denying plaintiffs’ motion for a preliminary injunction is aefiRmed. The parties shall bear their own costs. Footnotes [1] The Supreme Court has recognized that traditional redistricting factors, including "making districts compact, respecting municipal boundaries, preserving the cores of prior districts, and avoiding contests between incumbent[s],” may inform a legislature's redistricting choices unless, of course, they are superseded by the mandates of the Constitution or federal law. Karcher v. Daggett, 462 U.S. 725, 740, 103 S.Ct. 2653, 77 L.Ed.2d 133 (1983); accord Rodriguez v. Pataki, 308 F.Supp.2d 346, 363 (S.D.N.Y.) (applying Karcher in context of redistricting of state legislature), aff'd, 543 U.S. 997, 125 S.Ct. 627, 160 L.Ed.2d 454 (2004); see generally Favors v. Cuomo, No. 11-cv-5632 (RR)(GEL)(DLI)(RLM), 2012 WL 928223, at *4-7 (E.D.N.Y. Mar. 19, 2012). [2] Our decisions in Freedom Party of New York v. New York State Board of Elections, 77 F.3d 660 (2d Cir.1996), and Independence Party of Richmond County v. Graham, 413 F.3d 252 (2d Cir.2005), are not to the contrary. In those cases, elections were conducted pursuant to injunctions entered by the district court. In appealing therefrom, defendants did not seek to have those elections vacated. They invoked appellate jurisdiction to review issues that they contended were capable of repetition but would otherwise evade review. Not persuaded by this argument, we dismissed the appeals as moot, see Freedom Party of N.Y. v. N.Y. State Bd. of Elections, 77 F.3d at 662; Independence Party of Richmond Cnty. v. Graham, 413 F.3d at 256, without any need to decide, as we do here, that our ability to void a conducted election that did not comport with the VRA supports our jurisdiction to review the denial of a motion to grant a preliminary injunction preventing that election from going forward. [3] In Winter v. Natural Resources Defense Council, Inc., 555 U.S. at 20, 129 S.Ct. 365, the Supreme Court also required the plaintiff to show a fourth factor, that the balance of equities tips in his favor. Accord Salinger v. Colting, 607 F.3d 68, 80 (2d Cir.2010). Because this appeal fails on the likelihood of success factor, we need not consider the balance of equities further. [4] Among the totality of circumstances appropriately considered are those set forth in the Senate Report on the 1982 amendments to the Voting Rights Act: (1) the history of voting-related discrimination in the State or political subdivision; (2) the extent to which voting in the elections of the State or other political subdivision is racially polarized; (3) the extent to which the State or political subdivision has used voting practices or proce *572 dures that tend to enhance the opportunity for discrimination against the minority group; (4) the extent to which minority group members bear the effects of past discrimination in areas such as education, employment, and health, which hinder their ability to participate effectively in the political process; (5) the use of overt or subtle racial appeals in political campaigns; (6) the extent to which members of the minority group have been elected to public office in the jurisdiction; (7) evidence demonstrating that elected officials are unresponsive to the particularized needs of the members of the minority group; and (8) evidence that the political subdivision's use of the contested practice or structure is tenuous. See Thornburg v. Gingles, 478 U.S. at 44-45, 106 S.Ct. 2752 (citing S.Rep. No. 97-417 (1982), reprinted in 1982 U.S.C.C.A.N. 177, 206). This list "is neither comprehensive nor exclusive.” Id. at 45, 106 S.Ct. 2752; accord NAACP v. City of Niagara Falls, 65 F.3d 1002, 1008 (2d Cir. 1995). Another relevant consideration identified by the Supreme Court is "whether the number of districts in which the minority group forms an effective majority is roughly proportional to its share of the population in the relevant area.” LU-LAC v. Perry, 548 U.S. at 426, 126 S.Ct. 2594. [5] The circuits are split as to whether different minority groups may be aggregated to establish a Section 2 claim. Compare Nixon v. Kent Cnty., 76 F.3d 1381, 1393 (6th Cir.1996) (era banc) (holding such "coalition suits” impermissible), with Concerned Citizens of Hardee Cnty. v. Hardee Cnty. Bd. of Comm’rs, 906 F.2d 524, 526 (11th Cir. 1990) (approving aggregate Section 2 claims) and Campos v. City of Baytown, 840 F.2d 1240, 1244 (5th Cir. 1988) (same). In Bridgeport Coalition for Fair Representation v. City of Bridgeport, 26 F.3d 271, 275-76 (2d Cir.), vacated on other grounds, 512 U.S. 1283, 115 S.Ct. 35, 129 L.Ed.2d 931 (1994), this court identified a Section 2 violation in a challenge that aggregated blacks and Hispanics, without specifically ruling that such aggregation was permissible. Meanwhile, the Supreme Court has expressly reserved decision on the issue, while at the same time indicating that a plaintiff relying on an aggregated minority group would have to demonstrate that its members are politically cohesive. See Growe v. Emison, 507 U.S. 25, 41, 113 S.Ct. 1075, 122 L.Ed.2d 388 (1993); see also Badillo v. City of Stockton, 956 F.2d 884, 890-91 (9th Cir.1992) (affirming dismissal of Section 2 claim because plaintiffs failed to show sufficient black/Hispanic cohesion, without discussing whether aggregation was in fact permissible). In finding plaintiffs not to have demonstrated the political cohesiveness of blacks and Hispanics in Albany County, the district court did not, as plaintiffs argue, hold that, as a matter of law, the burden needed to be satisfied through statistical evidence. See Thornburg v. Gingles, 478 U.S. at 57 n. 25, 106 S.Ct. 2752 (stating that where applicable election data is not available, "courts must rely on other factors” to determine whether Section 2 claim has been proved); see also Brewer v. Ham, 876 F.2d 448, 454 (5th Cir.1989) (noting that "statistical evidence is not a sine qua non to establishing cohesion”). Rather, the court found that the particular socioeconomic data and anecdotal evidence advanced by plaintiffs to demonstrate the aggregated group’s political cohesiveness were not factually convincing. We express no view on this factual determination because plaintiffs' injunction motion fails, in any event, at the third step of Gingles analysis, majority bloc voting. See infra at II.B.3.b. [6] Because the parties argue by reference to VAP, we do not here consider the alternative possibility of employing citizen voting-age population ("CVAP”) in evaluating a Section 2 claim. See Reyes v. City of Farmers Branch, 586 F.3d 1019, 1023 (5th Cir.2009) (analyzing claim by reference to CVAP); Negron v. City of Miami Beach, 113 F.3d 1563, 1569 (11th Cir.1997) (same); cf. Garza v. Cnty. of Los Angeles, 918 F.2d 763, 779-86 (9th Cir.1990) (Kozinski, J., concurring in part and dissenting in part) (discussing relative advantages of VAP and CVAP in Fourteenth Amendment context). But see Bartlett v. Strickland, 556 U.S. 1, 18, 129 S.Ct. 1231, 173 L.Ed.2d 173 (2009) (plurality opinion) (framing question as whether minority group constitutes greater than 50% of "voting-age population”); Nathaniel Persily, The Law of the Census: How To Count, What To Count, Whom To Count, and Where To Count Them, 32 Cardozo L.Rev. 755, 780-81 (2011) [hereinafter The Law of the Census ] (advocating that courts avoid reliance on CVAP rather than VAP because of significant data limitations). [7] To demonstrate that it was possible to draw five MMDs, plaintiffs submitted the AHEJ plan as well as three illustrative plans prepared by their expert, redistricting consultant William S. Cooper. We focus on the AHEJ plan and Illustrative Plan 3, as these are the plans pressed by plaintiffs on appeal as demonstrating a Section 2 violation. The following table illustrates the demographics for the MMDs proposed by these two plans: [[Image here]] [[Image here]] The "MMD” column references the number assigned to the district in each plan. The remaining columns show the percentage of the minority VAP in the given district under three different measures: (1) "Any Part Black,” (2) "DOJ Non-Hispanic Black,” and (3) "DOJ Non-Hispanic Black and Hispanic.” "Any Part Black” counts any individual identifying as "black” on the census form, including those who marked “black” as well as any other race or ethnicity, including Hispanic. "DOJ Non-Hispanic Black” counts individuals who identify as black, but excludes those who additionally identify themselves as Hispanic or any other minority. "DOJ Non-Hispanic Black and Hispanic” combines "DOJ Non-Hispanic Black” with anyone who identifies as Hispanic. See generally Georgia v. Ashcroft, 539 U.S. 461, 474 n. 1, 123 S.Ct. 2498, 156 L.Ed.2d 428 (2003) (recognizing DOJ's use of “DOJ Non-Hispanic Black” in analyzing claims under Section 5 of the VRA). [8] Indeed, low voter registration and turnout rates raise questions that go beyond the merely statistical because, as the Supreme Court has recognized, such circumstances may sometimes be "traceable, at least in part, to [a] historical pattern of ... official discrimination.” Thornburg v. Gingles, 478 U.S. at 39, 106 S.Ct. 2752; see also S. Rep. 89-162, reprinted in 1965 U.S.C.C.A.N. 2508, 2552 (suggesting that VRA was prompted by state discriminatory actions that served to achieve "low registration and voting”). For this reason, the law allows plaintiffs to challenge legislatively created bare majority-minority districts on the ground that they do not present the "real electoral opportunity” protected by Section 2. LULAC v. Perry, 548 U.S. at 428, 126 S.Ct. 2594 (noting in different context that "it may be possible for a citizen voting-age majority to lack real electoral opportunity”); see Salas v. Sw. Tex. Junior Coll. Dist., 964 F.2d 1542, 1556 (5th Cir.1992) (concluding that no per se rule bars minority group that is registered voting majority in district from demonstrating vote dilution where totality of circumstances might show low turnout stemmed from past discrimination); cf. Clark *576 v. Calhoun Cnty., 88 F.3d 1393, 1398 (5th Cir.1996) (noting, in response to defendant's argument that lack of black electoral success was explained by few blacks seeking electoral office, that voting rights cases should not “be defeated at the outset by the very barriers to political participation that Congress has sought to remove”). Thus, while a defendant’s creation of MMDs with simple majority representation does not necessarily close the gate to a potential finding of vote dilution, plaintiffs need not show more than a simple majority at the first step of the Gingles analysis to open the gate to further Section 2 inquiry- [9] In concluding that a showing of a simple majority is sufficient, we express no view regarding whether showing a majority is always necessary, a question that the Supreme Court has not resolved in a majority opinion. See Bartlett v. Strickland, 556 U.S. at 15-17, 129 S.Ct. 1231 (plurality opinion) (identifying no Section 2 violation in failure to produce district in which minorities were less than 50% of population but could elect candidates of their choice with help of reliable white crossover voting); LULAC v. Perry, 548 U.S. at 445-46, 126 S.Ct. 2594 (plurality opinion) (rejecting claim regarding district in which minority was less than 50% of population but was capable of influencing election outcomes). [10] Insofar as Judge Tjoflat, dissenting in Thompson v. Glades County Board of County Commissioners, 493 F.3d 1253, 1271 (11th Cir.), vacated, 508 F.3d 975 (11th Cir.2007), thought a bare majority rule "contravenes the searching factual analysis the Supreme Court set in place in Gingles,’’ we think that concern is adequately assuaged by permitting the district court to consider the size of the majority among the totality of circumstances informing the ultimate Gingles inquiry into vote dilution and the appropriate remedy, see Johnson v. De Grandy, 512 U.S. at 1011-12, 114 S.Ct. 2647. [11] Because plaintiffs satisfy the first Gingles factor for DOJ Non-Hispanic Blacks, we need not here consider whether the relevant minority group might more appropriately be identified as "Any Part Black,” for which the minority VAP percentages are even higher. See supra at 573 n. 7. Insofar as the parties take different positions on that question — with plaintiffs arguing for “Any Part Black,” while defendants urge "DOJ Non-Hispanic Blacks” (even though their consultant appears to have developed Local Law C by reference to a minority group consisting blacks and Hispanics) — we note that in Georgia v. Ashcroft, a case arising under Section 5 of the VRA, the Supreme Court observed that DOJ Non-Hispanic Black may be a relevant group in cases involving "a comparison of different minority groups.” 539 U.S. at 474 n. 1, 123 S.Ct. 2498. But where "only one minority group's effective exercise of the electoral franchise” is at issue, "it is proper to look at all individuals who identify themselves as black.” Id. (emphasis in original). In any event, the parties must be consistent. They cannot argue one Gingles factor by reference to a particular minority group, only to recast the minority group in arguing another factor. See Persily, The Law of the Census, 32 Cardozo L.Rev. at 772 (recognizing that while choice of more expansive minority group may assist plaintiffs in satisfying first Gingles factor, it may also add to their burden in demonstrating political cohesion required for second factor). [12] The first Gingles factor further required plaintiffs to show that their proposed MMD was "reasonably compact,” a question not *578 addressed by the district court. LULAC v. Perry, 548 U.S. at 430, 126 S.Ct. 2594. We express no view on how this question should be resolved in any further consideration of plaintiffs’ Section 2 claim. [13] Dr. Liu testified that he lacked sufficient data to perform a comparable analysis of Hispanic voting patterns in Albany County. [14] Dr. Liu opined that racial polarization was evidenced in the following elections: *579 [[Image here]] The following elections did not exhibit racial polarization, according to Dr. Liu: [[Image here]] The record does not reveal the winners in either the single- or multi-member elections analyzed by Dr. Liu. [15] The following table shows Dr. Liu's analysis of the results of the multi-member elections for the Board of Education: *580 Year Position Scope Type Black Candidate 2004 City Court Judge City Primary Heath-Roland 2004 City Court Judge City General Heath-Roland 2009 Common Council City Primary McLaughlin 2004 District Attorney County Primary Soares 2004 District Attorney County General Soares 2005 State Committee County N/A Barnett 2008 District Attorney County General Soares [16] Albany County Election Commissioner Matthew Clyne testified that black candidates were elected in the following biracial contests not analyzed by Dr. Liu: [[Image here]] [[Image here]] [17] We note that Dr. Liu was of the opinion that President Obama's winning of Albany County in 2008, supported rather than undermined the "conclusion that there is racial polarization in Albany.” Liu Test., Hr’g Tr. at 138-39 (explaining that, although 2010 census indicates that Albany County is 78% white, and that President Obama received approximately 64% of County’s votes, Obama received only 33% of County’s white vote while receiving 52% of white vote of New York State). We express no view one way or the other on the matter.