Per Curiam Opinion; Partial Concurrence and Partial Dissent by Judge WALLACE; Partial Concurrence and Partial Dissent by Judge MICHAEL DALY HAWKINS.
PER CURIAM:1Four Hispanic residents and registered voters of the City of Santa Maria challenge the City’s at-large city council election system under Section 2 of the Voting Rights Act of 1965, as amended, 42 U.S.C. § 1973. While the case was pending, two Hispanic candidates were elected to the city council. The district court dismissed the complaint as moot and, alternatively, granted summary judgment in favor of the City. We hold that the district court erred in dismissing the ease as moot and applied erroneous legal standards to the Plaintiffs’ vote dilution claim and, therefore, we reverse.
I. FACTS
Santa Maria is a California general law city located in Santa Barbara County. Since its incorporation in 1905, it has employed an at-large system for electing its mayor and city council members. The mayor, who is also a member of the city council, is elected by a citywide plurality vote to serve a two-year term. Council members are elected by a eitywide plurality vote for four-year terms. Their terms are staggered so that two council members are elected every other year. Each registered voter can vote for two candidates but is not prohibited from casting a single ballot (“single-shot” or “bullet” voting).
The City’s 1990 population of 61,284 has doubled since 1970. According to the 1990 Census, 46.4% of the City’s population is white, 45.7% is Hispanic, 1.9% is black, and 5.9% is Asian, Pacific Islander, or other.2 The Hispanic population is growing: Hispanics were 27% of the City’s population in 1970 and 33% in 1980. The parties’ experts agreed that Hispanic population, citizenship, and voter registration will continue to grow but disagreed at what rate. While Hispanies comprised nearly one-half of the City’s population in 1990, they were only 28.7% of the voting age citizen population.
Prior to 1994, no Hispanic had ever been elected mayor or council member. Twelve Hispanies ran and lost in city council and mayoral elections from 1972-1992 — all but two placed last. The parties’ experts studied four city council elections (1988, 1990, 1992, and 1994) and the 1990 mayoral election in depth and provided detailed statistical reports.3 These reports showed that while Hispanic candidates were the first or second preference4 of Hispanic voters in each election, Hispanic candidates were the least-preferred among non-Hispanic voters and came in last every time, except in 1994.
In 1988, four candidates ran for city council. Tom Urbanske, the incumbent who was white, was Hispanic voters’ first preference. A Hispanic candidate was Hispanies’ second preference but came in last overall, receiving only 15.7% of the non-Hispanic vote. Ur-banske and another white candidate were elected.
*547In 1990, the City’s long-time mayor, George Hobbs, was under fire from the Hispanic community for complaining about a “Mexican problem” in Santa Maria.5 He easily won the mayoral election, receiving 71.4% of the non-Hispanic vote but only 13.3% of the Hispanic vote. Another white mayoral candidate withdrew from the race two months before the election because the community was so polarized along racial lines. Nevertheless, he finished second overall, receiving more non-Hispanics’ votes than the two Hispanic candidates combined. The Hispanic candidates finished last and next-to-last, despite being Hispanic voters’ first and second preferences.
The Hispanic candidate did not fare any better in the 1990 city council election. The Hispanic candidate was the first preference of Hispanic voters, receiving 66.2% of Hispanics’ votes. Nonetheless, he received only 8.6% of the non-Hispanic vote and came in last out of five candidates. The two white incumbents, the first and second preferences of non-Hispanics while least-preferred by Hispanics, were re-elected.
In 1992, Urbanske, the white incumbent, was again Hispanic voters’ first preference out of seven candidates. Hispanics’ second preference, a Hispanic candidate, was least-preferred by non-Hispanic voters and came in last. Urbanske and Toru Miyoshi,6 the third preference of Hispanics and non-Hispanics, were elected.
Plaintiffs filed this action in August 1992, seeking a declaration that the City’s at-large election system impermissibly diluted the vote of Hispanics and an injunction creating a single-member district election system. Plaintiffs proposed a four- or five-district system in which Hispanics would comprise a majority of the voting age citizen population in one district. On the day before the July 1994 trial, the district court vacated the trial date because Plaintiffs had submitted new trial exhibits which contained statistical information from the June 1994 statewide primary elections and the City had no time to prepare rebuttal evidence.7 The district court continued the trial for an additional reason: “an informed ruling in this matter requires a review of the results produced by the upcoming Santa Maria election in November 1994.”
In the 1994 election, two Hispanic candb dates were elected to the city council. Joe Centeno, the city’s former police chief, finished first, as the second preference of Hispanics and non-Hispanics. Abel Maldonado, a young local businessman, won the second seat, as the first preference of Hispanics and the third preference of non-Hispanics. Non-Hispanic voters’ first preference, a white candidate, came in third. Thus, in 1994, Hispanic voters succeeded in electing their first and second preferences to the city council.8
In October 1995, the district court vacated the upcoming trial date and sua sponte issued an Order to Show Cause why the case *548should not be dismissed as moot. Plaintiffs argued that “special circumstances” surrounded the 1994 city council election, and thus the district court should ignore the results. Plaintiffs relied on four things to support their “special circumstances” argument:
(1) There was unprecedented crossover voting by non-Hispanics. From 1988-1992, non-Hispanic support for Hispanic city council candidates had declined, from 15.7% in 1988, to 8.6% in 1990, and to 4.7% in 1992.9 Yet in 1994, 35.2% of non-Hispanics voted for Centeno, and 26.1% voted for Maldonado. It was the first election in which Hispanic candidates were not the least-preferred among non-Hispanic voters.
(2) The two Hispanic winners received endorsements from community organizations and politicians, including Mayor Hobbs, who had never before endorsed a Hispanic candidate.10
(3) The winning Hispanic candidates received unprecedented financial contributions from non-Hispanic individuals and businesses. Prior to 1994, no Hispanic candidate had received more than $790 from non-Hispanics. In 1994, Centeno received $4,550 and Maldonado received $6,625 from the non-Hispanic community. Also, Maldonado spent approximately $32,000 on his campaign— three times the highest amount ever spent on a council election.
(4) It was well-known throughout Santa Maria and by city council members that Plaintiffs had filed a vote dilution lawsuit and that the district court had continued the trial to evaluate the results of the 1994 election. Five days before that election, Maldonado stated in the Santa Maria Times, “If I get elected, it will show that the city of Santa Maria is not racist. Some parties in this lawsuit don’t want to see a Hispanic on the City Council until after the suit is settled.”11
In September 1996, the district court dismissed the case as moot under Fed. R. Civ. Proe. 12(b)(1), stating that the election of two Hispanics to the city council precluded the court from granting effective relief to Plaintiffs. In the alternative, the court granted the City’s Motion for Summary Judgment, finding that Plaintiffs could not prove that a majority voting bloc usually defeats the Hispanic-preferred candidate, a necessary precondition to establishing a claim of vote dilution.12 This appeal followed.13
II. STANDARD OF REVIEW
We review de novo a district court’s dismissal of an action on the ground of mootness and grant of summary judgment. See Native Village of Noatak v. Blatchford, 38 F.3d 1505, 1509 (9th Cir.1994) (mootness); Valladolid v. City of National City, 976 F.2d 1293, 1295 (9th Cir.1992) (summary judgment).
III. MOOTNESS
Article III of the Constitution limits federal courts to adjudicating actual, on*549going controversies between litigants. See Deakins v. Mohaghan, 484 U.S. 193, 199, 108 S.Ct. 523, 98 L.Ed.2d 529 (1988). Generally, an action is moot “when the issues presented are no longer ‘live’ or the parties lack a legally cognizable interest in the outcome.” Lee v. Schmidt-Wenzel, 766 F.2d 1387, 1389 (9th Cir.1985) (internal quotations and citations omitted). “The basic question in determining mootness is whether there is a present controversy as to which effective relief can be granted.” Northwest Environmental Defense Center v. Gordon, 849 F.2d 1241, 1244 (9th Cir.1988). Claims for injunctive relief become moot when the challenged activity ceases if “subsequent events have made it clear that the alleged violations could not reasonably be expected to recur.” Chinese for Affirmative Action v. Leguennec, 580 F.2d 1006, 1009 (9th Cir.1978).
The district court recognized that the effective relief Plaintiffs sought was a district election system but stated that the real reason for bringing the lawsuit was to elect Hispanie-preferred candidates. Because this result was achieved in the 1994 at-large election, the district court reasoned that a district system would not be more “effective.” Therefore, the district court concluded it could not grant Plaintiffs any effective relief and declared the case moot.
Because the district court’s mootness analysis misapprehends the essence of a vote dilution claim, we reverse. Section 2 of the Voting Rights Act of 1965, 42 U.S.C. § 1973 (“Section 2”), prohibits any standards, practices, or procedures that deny or abridge the right of citizens to vote on account of their race, color, or membership in a language minority group. See Valladolid, 976 F.2d at 1295. A Section 2 violation is any political process leading to an election that is not “equally open to participation” by a minority group “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.” 42 U.S.C. § 1973(b). As amended in 1982, Section 2 “requires proof only of a discriminatory result, not of discriminatory intent.” Smith v. Salt River Project Agr. Imp. and Power Dist., 109 F.3d 586, 594 (9th Cir.1997). Nevertheless, the Voting Rights Act does not establish a right of a minority group to elect its representatives in proportion to its population size. See 42 U.S.C. § 1973(b). Rather, the focus of Section 2 is “whether the political process is equally open to minority voters.” Thornburg v. Gingles, 478 U.S. 30, 79, 106 S.Ct. 2752, 92 L.Ed.2d 25 (1986). Thus, Section 2 guarantees a fair process, not an equal result.
There is a present, live controversy in this ease — whether the City’s at-large election system impermissibly dilutes the votes of Hispanics. The district court concluded that Hispanics have the ability to elect their preferred candidates in the at-large system; that was a determination of the merits. The results of the 1994 election are certainly a factor relevant to whether Hispanics have the ability to elect their preferred candidates in the current system. Those elections results, however, do not affect the power of the district court to grant an effective remedy.
This lawsuit would be moot if, for example, Plaintiffs had moved out of Santa Maria or the City had already adopted a district election system. See Chinese for Affirmative Action, 580 F.2d at 1009 (claim for injunctive relief requiring multilingual ballots in a particular election was moot after the election was held, but sufficiency of the city’s current compliance with Voting Rights Act was a live controversy). But the presence of two Hispanics on the city council does not affect the power of the district court to grant effective relief if Hispanic votes are impermissibly diluted.
Moreover, the relevant inquiry in a vote dilution case is whether a majority bloc usually defeats the minority-preferred candidate, not whether a minority candidate is ever elected. In Gingles, 478 U.S. at 30, 106 S.Ct. 2752, the authoritative interpretation of a Section 2 claim, the Supreme Court stated that a district court may accord greater weight to minorities’ lack of success over the course of several recent elections rather than to their success in one election. See id. at 76, 106 S.Ct. 2752.
Because loss of political power through vote dilution is distinct from the mere ina*550bility to win a particular election, a pattern of racial bloc voting that extends over a period of time is more probative of a claim that a district experiences legally significant polarization than are the results of a single election.
Id. at 57, 106 S.Ct. 2752 (internal citation omitted). Rather than folding the 1994 election results into its analysis of the merits, the district court found those results dispositive as to its continuing jurisdiction. The 1994 election results, which plaintiffs allege “sporadically and serendipitously benefit[ed] minority voters,” id. at 76, 106 S.Ct. 2752, did not terminate the present, live controversy in this ease.
IV. GINGLES AND ITS THIRD PRONG
Gingles established a two-step inquiry for analyzing a vote dilution claim. In the first step, a minority group challenging an at-large election system must demonstrate that: (1) the minority group is sufficiently large and geographically compact to constitute a majority in a single-member district; (2) the minority group is politically cohesive; and (3) the Section 2 majority14 votes sufficiently as a bloc to enable it, in the absence of special circumstances, “usually to defeat the minority’s preferred candidate.” 478 U.S. at 50-51, 106 S.Ct. 2752. In the second step, the district court must decide, based on the “totality of the circumstances,” whether the challenged practice impermissi-bly impairs the ability of the minority group to elect their preferred representatives. Id. at 44-45, 106 S.Ct. 2752. The Senate Judiciary Committee produced a list of non-exhaustive factors a court should consider in examining the totality of circumstances.15
The key issue before us is the third Gin-gles precondition. The district court granted summary judgment for Santa Maria, in the alternative, because it found that Plaintiffs could not prove the third Gingles prong. Because four Hispanic-preferred candidates were elected in the last four city council elections (a success rate of 50%), the district court concluded that a white majority voting bloc did not usually defeat the Hispanic-preferred candidate. The district court also relied on the projected growth of the Hispanic population in Santa Maria and the availability of single-shot voting to support its conclusion.
Plaintiffs argue that: (1) the election of a Hispanic-preferred white candidate is not as probative as the defeat of a Hispanic-preferred Hispanic candidate; (2) the use of single-shot voting and the changing demographics of Santa Maria are irrelevant to a Gingles prong three analysis; and (3) special circumstances surrounded the 1994 election. We agree.
A. Who is the “minority’s preferred candidate”?
1.
The Supreme Court has not defined the term “minority’s preferred candidate.” See *551Gingles, 478 U.S. at 51, 106 S.Ct. 2752. Gingles created confusion on this issue because the Supreme Court split on the relevance of the candidate’s race in finding racially polarized voting. Racially polarized voting, one of the' most important Senate factors, is the combined effect of the second and third Gin-gles preconditions (minority political cohesion and majority bloc voting). See id. at 56, 106 S.Ct. 2752; Romero, 883 F.2d at 1423. Justice Brennan, writing for a plurality, defined racially polarized voting as a correlation between the race of the voters and the selection of candidates. See Gingles, 478 U.S. at 74, 106 S.Ct. 2752. He rejected other factors such as the race of the candidates or voters’ racial animosity toward minority candidates as primary determinants of voter behavior. See id. at 67-72, 106 S.Ct. 2752. Although he noted that minority and majority voters often select members of their own race as their preferred representatives, Section 2 was concerned with “the status of the candidate as the chosen representative of a particular racial group, not the race of the candidate.” Id. at 68, 106 S.Ct. 2752 (emphasis in original).
Five Justices did not agree. Justice O’Connor wrote for a plurality and Justice White wrote separately to state that the race of the candidate could be relevant in a vote dilution inquiry. See id. at 100-01, 106 S.Ct. 2752 (O’Connor, J.); id. at 83, 106 S.Ct. 2752 (White, J.). Their position was based, however, on concern that ignoring the race of the candidate as well as other evidence concerning voting preferences would overlook the relevance of partisan affiliation in determining voter behavior.
Although helpful, these various arguments do not address the issue before us: whether a “minority’s preferred candidate” may be a non-minority. We only briefly mentioned the issue in Valladolid, where we stated that plaintiffs “too easily equate the presence or absence of minority individuals on the council with the victory or defeat of minority-preferred candidates.” 976 F.2d at 1297; but see Gomez v. City of Watsonville, 863 F.2d 1407, 1416-17 (9th Cir.1988) (discussing the success rate of “Hispanic candidates” in the third Gingles prong analysis).
Most circuits have rejected Justice Brennan’s position on the lack of importance of a candidate’s race16 and hold that a minority preferred candidate may be a non-minority. See Lewis v. Alamance County, 99 F.3d 600, 607 (4th Cir.1996), cert. denied, — U.S. -, 117 S.Ct. 1820, 137 L.Ed.2d 1028 (1997); Sanchez v. State of Colorado, 97 F.3d 1303, 1320 (10th Cir.1996), cert. denied, — U.S. -, 117 S.Ct. 1820, 137 L.Ed.2d 1028 (1997); Uno v. City of Holyoke, 72 F.3d 973, 988 n. 8 (1st Cir.1995); NAACP v. City of Niagara Falls, N.Y., 65 F.3d 1002, 1016 (2d Cir.1995) (“Niagara Falls ”); Clarke v. City of Cincinnati 40 F.3d 807, 812 (6th Cir.1994); Nipper v. Smith, 39 F.3d 1494, 1540 (11th Cir.1994) (en banc); Jenkins v. Red Clay Consol. Sch. Dist., 4 F.3d 1103, 1125 (3d Cir.1993); League of United Latin American Citizens v. Clements, 999 F.2d 831, 864 (5th Cir.1993) (en banc) (“LULAC”). Conversely, a candidate is not minority-preferred simply because the candidate is a member of the minority. See Lewis, 99 F.3d at 607; Harvell v. Blytheville School Dist. # 5, 71 F.3d 1382, 1386 (8th Cir.1995); Jenkins, 4 F.3d at 1125.
We join our sister circuits in rejecting the position that the “minority’s preferred candidate” must be a member of the racial minority. To hold otherwise would, in the words of Judge Cabranes, provide judicial approval to “electoral apartheid.” See Niagara Falls, 65 F.3d at 1016. “Such an approach would project a bleak, if not hopeless, view of our society—a view inconsistent with our people’s aspirations for a multiracial and integrated constitutional democracy.” Id. The minority community may prefer a white candidate just as the white community may prefer a minority candidate. See Lewis, 99 F.3d at 607.
2.
We must next determine which candidate in a given election is a minority-preferred *552candidate. Although this is an issue of first impression in the Ninth Circuit, other Circuits have reached the issue, resulting generally in two very different approaches.
The Second, Fourth, and Sixth Circuits use an objective test that considers which candidate receives the most votes from minority voters. See id. at 614; Niagara Falls, 65 F.3d at 1018-19; Clarke, 40 F.3d at 810 n. 1.
The Third, Eighth, Tenth, and Eleventh Circuits, on the other hand, use variations on a totality of the circumstances theory and require district courts to conduct election specific analyses of anecdotal evidence such as whether minorities mobilized to support the candidate. See Sanchez, 97 F.3d at 1321; Harvell, 71 F.3d at 1386; Nipper, 39 F.3d at 1540; Jenkins, 4 F.3d at 1129.
We agree with the Second Circuit that a bright-line rule is the better test. The other approach of “evaluating whether a person is, ‘as a realistic matter,’ minority-preferred— based on subjective indicators such as ‘anecdotal testimonial evidence’—is a dubious judicial task, and one that can degenerate into racial stereotyping of a high order.” Niagara Falls, 65 F.3d at 1018. In addition, many of the extrinsic factors relied upon by the courts adopting the totality of the circumstances analysis do not necessarily bear a correlation with how all minority voters feel about a candidate, only how activist groups feel. Whether minority voters mobilized to support a white candidate, a factor considered relevant by the Third Circuit, Jenkins, 4 F.3d at 1129, only indicates how those minorities willing to become politically active feel about a candidate; not all minorities may have the time or inclination to take such steps, even though they support that candidate. A bright-line rule, on the other hand, is based on the premise “that the ballot box provides the best and most objective proxy for determining who constitutes a representative of choice.” Niagara Falls, 65 F.3d at 1019.
In our view, however, the requirement imposed by the Second Circuit that a candidate receive 50 percent or more of the votes cast by a minority group to qualify as minority-preferred can be too restrictive, particularly in a city such as Santa Maria, where all candidates routinely receive fewer than 50 percent of the Hispanic vote. See Appendix. Therefore, we hold that a candidate who receives sufficient votes to be elected if the election were held only among the minority group in question qualifies as minority-preferred. See Lewis, 99 F.3d at 614. We also agree with the Second Circuit that “a court need not treat the candidate as minority-preferred when another candidate receiving greater support in the primary failed to reach the general election.” Niagara Falls, 65 F.3d at 1019.
B. Evidentiary weight of election of minority-preferred majority candidate
Plaintiffs do not argue that the district court erred in considering the elections of Urbanske in its third Gingles prong analysis. Rather, they contend the district court should have placed less evidentiary weight on Hispanic voters’ ability to elect Urbanske, a white candidate, than on their inability to elect a Hispanic candidate.
Few courts have addressed this issue. Our sister circuits have grappled with the question of whether, in a Gingles prong three analysis, an election involving only non-minority candidates is relevant and, if so, whether it should receive equal evidentiary weight as an election in which there is a contest between a minority and non-minority candidate. Most courts hold that a fully non-minority election may be relevant and is admissible to determine whether there is a voting bloc of sufficient power that usually defeats a minority’s preferred candidate. An election pitting a minority against a non-minority, however, is considered more probative and accorded more weight. See Lewis, 99 F.3d at 608, 610 n. 8 (assuming that white-white elections are entitled to less weight); Uno, 72 F.3d at 988 n. 8; Niagara Falls, 65 F.3d at 1017; Nipper, 39 F.3d at 1540; Jenkins, 4 F.3d at 1128; LULAC, 999 F.2d at 864.
The general reasoning behind this conclusion is that non-minority elections do not provide minority voters with the choice of a minority candidate and thus do not fully *553demonstrate the degree of racially polarized voting in the community.
[T]he most probative evidence of whether minority voters have an equal opportunity to elect candidates of their choice is derived from elections involving [minority] candidates. Certainly, where only white candidates are competing for office in a particular jurisdiction, it is virtually unavoidable that certain white candidates would be supported by a large percentage of ... [minority] voters. Such elections, however, may reveal little about the issue to be determined: the capacity for white bloc voting usually to defeat [minority] candidates of choice. Particularly where voting is extremely polarized by race in elections in which [minority] candidates participate, white-on-white elections in which a small majority (or a plurality) of [minority] voters prefer the winning candidate seem comparatively less important.
Nipper, 39 F.3d at 1540 (internal quotations and citation omitted;.
The reasoning behind this rule provides persuasive guidance on the issue before us. For just as a minority vs. non-minority election is more probative of racially polarized voting than a non-minority vs. non-minority election, the inability of Hispanic voters to elect a Hispanic candidate is more probative in a Girigles prong three analysis than the ability of Hispanic voters to elect a non-minority candidate.
The Sixth Circuit alone has addressed this issue. In Clarke, the court held that the district court should have examined whether white bloc voting was targeted against black candidates (electoral success rate of 47%), not just against black-preferred white candidates (electoral success rate of 74%). See 40 F.3d at 812. The court reasoned:
When white bloc voting is “targeted” against [minority] candidates, [minority] voters are denied an opportunity enjoyed by white voters, namely the opportunity to elect a candidate of them own race. If [minority] voters nevertheless are able to elect many or most of their preferred candidates who are white, a court that refuses to consider candidate race will be unable to conclude that the white majority votes sufficiently as a bloc to enable it ... usually to defeat the minority’s preferred candidate. ... But the Act’s guarantee of equal opportunity is not met when, in the words of Judge Richard Arnold, “[Candidates favored by [minorities] can win, but only if the candidates are white.”
Id. (some internal quotations and citations omitted) (quoting Smith v. Clinton, 687 F.Supp. 1310, 1318 (E.D.Ark.) (three-judge panel), aff'd, 488 U.S. 988, 109 S.Ct. 548, 102 L.Ed.2d 576 (1988)); but see Lewis, 99 F.3d at 610 n. 8 (opining that candidate’s race was relevant only in determining whether racial polarization exists, not whether minority-preferred candidates were usually defeated).17
The Clarke court relied on Jenkins, 4 F.3d 1103, and Citizens for a Better Gretna v. City of Gretna, 834 F.2d 496 (5th Cir.1987), for support. See Clarke, 40 F.3d at 812. In Jenkins, the court observed that “if the minority community would prefer a minority candidate or believes that a minority candidate would best represent its interests, then § 2 establishes a right on the part of those citizens for their preferred minority candidate to have an equal opportunity to be elected.” 4 F.3d at 1128 n. 22. Gretna addressed a situation nearly identical to ours:
In a multiple seat election such as Gretna’s aldermanie race, the minority necessarily will have more than one preferred candidate. In the Gretna elections studied, blacks exercised their right to vote to fill all at-large positions, but only one available candidate was black. Thus, it was virtually unavoidable that certain white candidates would be supported by a large percentage of Gretna’s black voters. Significance lies in the fact that the black candidate preferred by the minority was defeated by white bloc voting. That blacks also support white candidates acceptable to the majority does not negate instances in which white votes defeat a black preference.
834 F.2d at 502.
We believe this reasoning is persuasive and hold that, in a multi-seat, multi-vote *554election, the ability of the minority to elect a non-minority candidate is not as probative in a Gingles prong three analysis as the inability of the minority to elect a minority candidate. To hold otherwise would doom any Section 2 claim in which white candidates, acceptable to minorities and non-minorities alike, are regularly elected but in which minority candidates, preferred by minorities but unfavored by whites, are consistently defeated. Our rule furthers the Voting Rights Act’s goal of protecting the minority’s equal opportunity to “elect its candidate of choice oh an equal basis with other voters.” Voinovich v. Quilter, 507 U.S. 146, 153, 113 S.Ct. 1149, 122 L.Ed.2d 500 (1993). The Act means more than securing minority voters’ opportunity to elect whites.18
The district court erred in applying a simple mathematical approach—counting the number of successful Hispanic-preferred candidates divided by the number of elections—in its Gingles prong three analysis. This mechanical approach failed to fulfill the district court’s duty to make “a searching practical evaluation of the past and present reality” with “a functional view of the political process.” Gingles, 478 U.S. at 45, 106 S.Ct. 2752 (internal quotations omitted) (citing S.Rep. No. 97-417, at 30 (1982), reprinted in 1982 U.S.C.C.A.N. 177, 208). The reality facing Hispanic voters in Santa Maria in the 1988,1990, and 1992 elections was a choice of one Hispanic candidate and several non-Hispanic candidates for two council seats. In those elections, Hispanic voters preferred the lone Hispanic candidate and, unavoidably, one of the remaining non-Hispanic candidates. The Hispanic candidate came in last while Urbanske, a Hispanic-preferred candidate, won in 1988 and 1992. See Gingles, 478 U.S. at 60-61, 106 S.Ct. 2752 (district court properly considered “the very different order of preference blacks and whites assigned black candidates”). The success of a Hispanic-preferred candidate like Urbanske does not necessarily demonstrate that Hispanics have an equal opportunity to participate in Santa Maria city council elections. Rather, it may simply show that, at least historically, Hispanics have been relegated to casting a veto between majority-preferred white candidates.
On remand, the district court should consider the elections of Urbanske in 1988 and 1992 in determining whether racial bloc voting exists in Santa Maria. The defeat of Hispanic-preferred Hispanic candidates, however, is more probative of racially polarized voting and is entitled to more evidentia-ry weight. The district court should also consider the order of preference non-Hispanics and Hispanics assigned Hispanic-preferred Hispanic candidates as well as the order of overall finish of these candidates. Only then can the district court determine whether a Section 2 majority bloc voting usually defeats Hispanic voters’ candidates of choice.
C. The relevance of other factors to Gin-gles’ third precondition
The district court also considered the use of single-shot voting and the growth of the Hispanic community in Santa Maria in its grant of summary judgment for the City on Gingles prong three. That consideration is misplaced.
Under the City’s present at-large election system, a Hispanic voter may vote for one or two candidates in a multi-seat election; the district court agreed with the City that this system “provides the greatest opportunity” for Hispanics to elect a preferred candidate. While we question the factual accuracy of this statement,19 we hold *555that the mere ability to bullet vote does not eliminate a triable issue regarding whether minority-preferred candidates are usually defeated.
The ability of minority voters to bullet vote does not remedy a vote dilution injury. When a court finds an inference of vote dilution because the plaintiffs proved the three Gingles preconditions, the court may consider the presence of an “anti-single-shot” provision in its “totality of the circumstances” inquiry. See Gingles, 478 U.S. at 37, 106 S.Ct. 2752 (quoting Senate Report factor three). Although we have recognized that bullet voting concentrates minority voting power on select candidates and withholds minority votes from majority candidates, see Badillo v. City of Stockton, 956 F.2d 884, 890 (9th Cir.1992), we have never stated that vote dilution is solved by the absence of an anti-single-shot provision. Indeed, no court has affirmatively required minority voters to bullet vote to elect their preferred candidates.
Whatever relevance the existence of an “anti-single-shot” provision has in a “totality of the circumstances” inquiry, it is simply insufficient to sustain summary judgment on Gingles prong three for the City.20
The district court also erred in including the growing Hispanic population in its Gingles prong three analysis. First, the third Gingles prong asks a predominantly historical question—that is, whether the Section 2 majority bloc usually defeats the minority’s preferred candidates. See Gingles, 478 U.S. at 51, 106 S.Ct. 2752. This question is answered by evaluating past election results. Neither the growth of the minority community during these past elections nor its projected growth is relevant to finding racially polarized voting in a jurisdiction. Dramatically changing demographics become relevant only when the court finds the three Gingles preconditions met and considers the totality of the circumstances.
Second, a finding of vote dilution “depends upon a searching practical evaluation of the past and present reality.” Gingles, 478 U.S. at 45, 106 S.Ct. 2752 (internal quotations omitted). . Nevertheless, “[t]he ultimate question in any section 2 case must be posed in the present tense, not the past tense.” Uno, 72 F.3d at 990. Past elections may be less probative of vote dilution than more recent elections. See id.; see also Aldasoro v. Kennerson, 922 F.Supp. 339, 373 (S.D.Cal.1995) (Hispanics presently have the ability to elect at-large). At the same time, the court must decide whether the current system actually dilutes the minority vote, not whether it may potentially aid minority voters. See Jenkins, 4 F.3d at 1123. A court may not refuse to fashion a remedy after a claim of vote dilution is established based simply on its belief that the vote dilution injury will be solved in the future by changing demographics.
On remand, if the district court finds the three Gingles preconditions met and engages in a “totality of the circumstances” analysis, the proper inquiry is whether changing demographics demonstrate that Hispanics presently have the ability to elect at-large, not whether they will have this ability in the future.
D. The “special circumstances” doctrine
Plaintiffs argue that the success of two Hispanic candidates in 1994 was attributable to special circumstances and therefore should have been discounted or ignored by the district court in its Gingles prong three analysis. The district court found that Plaintiffs *556failed to submit any evidence that non-His-panies voted for, endorsed, or financially supported the Hispanic candidates in an effort to defeat the lawsuit, and thus the special circumstances doctrine did not apply. Plaintiffs concede that they have no direct evidence of a conspiratorial intent by whites to elect Hispanic candidates to thwart the lawsuit.21
The “special circumstances” doctrine was enunciated in Gingles, where the Court stated that minority success in elections infected by special circumstances should not be counted or emphasized in a prong three analysis. 478 U.S. at 51, 106 S.Ct. 2752. This is because isolated minority success is distinct from the “usual predictability” of majority success present in a vote dilution claim. Id. Examples of special circumstances are: the absence of an opponent; incumbency; utilization of bullet voting; or the pendency of the instant lawsuit. See id. at 57, 106 S.Ct. 2752. When those circumstances are present, voters may behave differently than they would otherwise — as such, that election is not probative of whether a Section 2 majority voting bloc usually defeats the minority’s preferred candidate.
The Gingles Court devoted the greatest attention to the special circumstance of minority success in an election during the pen-dency of litigation challenging the existing system. In such a situation, the district court “could properly consider to what extent the pendency of this very litigation [might have] worked a one-time advantage for [minority] candidates in the form of unusual organized political support by white leaders concerned to forestall single-member district-ing.” Id. at 76, 106 S.Ct. 2752 (internal quotations and citation omitted). A plain reading of Gingles, however, does not compel the conclusion that plaintiffs must submit evidence that white leaders manipulated an election with the intent to thwart a vote dilution lawsuit. The Court stated that a district court should cautiously view minority electoral success achieved after a vote dilution lawsuit is filed. See id. The Court then stated several reasons why these minority successes might be unrepresentative of usual voting patterns: white politicians “might” apprehend that the election of a minority candidate would be politically expedient or “might” believe that the election of a minority candidate would thwart the lawsuit. Id. at 76 n. 37, 106 S.Ct. 2752 (quoting Zimmer v. McKeithen, 485 F.2d 1297, 1307 (5th Cir.1973)). The Court did not hold that, to invoke the special circumstances doctrine, plaintiffs must prove that white politicians are motivated by an intent to thwart the lawsuit or other racist reasons.
Nevertheless, several courts seem to have interpreted Gingles that way. In Valladolid, we found that plaintiffs’ special circumstances argument failed because plaintiffs had submitted no other election results in support of the third Gingles requirement. 976 F.2d at 1297-98. We noted, in dictum, that the special circumstances doctrine did not apply because plaintiffs had “adduce[d] no evidence that the pendency of litigation in 1988 led to unusual support for minority-preferred candidates by whites seeking to maintain the at-large system.” Id. at 1298; see also Aldasoro, 922 F.Supp. at 376 (no evidence that Hispanic candidate won because of unusual, organized white support or endorsement or because whites wanted to secure the dismissal of the lawsuit). The Second Circuit rejected plaintiffs’ attempt to portray a minority candidate’s election — due to the candidate’s outstanding credentials and popularity — as surrounded by special circumstances. See Niagara Falls, 65 F.3d at 1021 n. 22. The court found this argument “absurd” and noted the absence of evidence indicating an attempt by white voters to thwart the pending lawsuit. See id. at 1021. Recently, the Eleventh Circuit concluded that individual decisions to appoint minorities to the city commission or Board of Education were not attempts “to avoid the reach of the Voting Rights Act or to manipulate the instant case in any way.” Askew v. City of Rome, 127 F.3d 1355, 1382 (11th Cir.1997). The court emphasized, however, that voting behavior, not racist voting motivation, was *557the central inquiry in a Section 2 case. See id.
The Fourth Circuit has directly ruled on this issue. “[T]he absence of a conspiracy or an intent to moot this litigation does not end the district court’s inquiry. The court should probe further to determine whether the [minority] candidate’s success ... while this action was pending, resulted from unusual circumstances.” Collins v. City of Norfolk, 816 F.2d 932, 938 (4th Cir.1987).22 In Collins, two black city council members were elected for the first time in city history. The mayor had, for the first time, supported a second black candidate and had publicly stated that the election might moot the issue of black representation. See id. The court held that the proper focus was the voting patterns that resulted from the mayor’s conduct and statement, not the mayor’s intent or a conspiracy by whites. See id. Thus, the Fourth Circuit requires proof of unusual circumstances, such as unusual white support for a minority candidate, not that the unusual circumstances resulted from an intent to thwart the lawsuit.
We are convinced this approach is sound and consistent with the overall purpose of the Voting Rights Act. An intent requirement has been rejected in other contexts of the Act. In City of Mobile v. Bolden, 446 U.S. 55, 100 S.Ct. 1490, 64 L.Ed.2d 47 (1980), the Supreme Court departed from its prior eases and held that plaintiffs must prove purposeful discrimination in a vote dilution claim under the Act. See id. at 66-67, 100 S.Ct. 1490. Congress swiftly amended Section 2 with the express purpose of repudiating the “intent test” because it was “unnecessarily divisive [in that] it involve[d] charges of racism on the part of individual officials or entire communities,” it placed an “inordinately difficult burden of proof on plaintiffs,” and it asked “the wrong question.” Gingles, 478 U.S. at 44, 106 S.Ct. 2752 (internal quotations omitted) (quoting S. Rep. at 36, 1982 U.S.C.C.A.N. at 214). Instead, the “results test” asks the right question—whether minorities have equal opportunity to participate in the political processes or to elect their chosen candidates as a result of a challenged practice or structure. See id. (quoting S. Rep. at 28,1982 U.S.C.C.A.N. at 206).
Also, we rejected an “intentional discrimination” standard of proof when a district court considers the first Senate factor—the history of official discrimination in a state or political subdivision. See Gomez, 863 F.2d at 1418. We noted that Congress amended Section 2 to relieve plaintiffs of the burden of proving such intent. Although intentional discrimination may be relevant to a vote dilution claim, we held that plaintiffs need only focus on the results of official discrimination. See id. Thus, it seems clear that requiring proof of intentional discrimination or racist motives in any part of a vote dilution claim “diverts the judicial inquiry from the crucial question of whether minorities have equal access to the electoral process to a historical question of individual motives.” S. Rep. at 18, 1982 U.S.C.C.A.N. at 177.
We hold that proof of groupwide or individual discriminatory motives has no part in a vote dilution claim. To invoke the special circumstances doctrine regarding an election that occurred after a Section 2 lawsuit is filed, plaintiffs must show that a particular election was surrounded by unusual circumstances. Those unusual circumstances must demonstrate that the election was not representative of the typical way in which the electoral process functions. The focus is voter behavior, not voter motivation.
This rule parallels the policies set forth in Gingles for rejecting the “intent test.” First, it removes the requirement that plaintiffs not only brand but prove that white voters, organizations, or leaders are racists or motivated by racist ambitions. “Such inquiries can only be divisive, threatening to destroy any existing racial progress in a community”—a result clearly antithetical to the purpose of a statute meant to engender racial harmony. Gin-*558gles, 478 U.S. at 72, 106 S.Ct. 2752 (internal quotation omitted) (quoting S. Rep. at 36, 1982 U.S.C.C.A.N. at 214 (testimony of Dr. Arthur S. Flemming, Chairman of U.S. Comm’n on Civil Rights)). Second, it eliminates an inordinately difficult burden of proof from plaintiffs. As this case demonstrates, it is virtually impossible to find evidence that white politicians or community groups conspired to manipulate an election to ensure the victory of a minority candidate. Rarely would these individuals admit to such a motive. We cannot even begin to imagine how plaintiffs could prove that thousands of individual voters cast their ballots intending to thwart the lawsuit. A thousand voters may vote for a particular candidate for a thousand different reasons; a discriminatory motive may only be a subconscious thought of the voter. Third, our rule asks the right question—whether a particular election was surrounded by unusual circumstances such that its results are not reliable indicators of typical voting behavior. Only minority electoral success in typical elections is relevant to whether a Section 2 majority voting bloc usually defeats the minority’s preferred candidate.
Finally, our rule avoids an argument implicit in proof of intent to thwart a pending lawsuit—that somehow the victorious minority candidate was unqualified and elected only because of white manipulation. We strenuously reject any invitation to disparage the credentials of any candidate running for public office.23 That is what happened in this case, in which the City argued in its brief that the Hispanic candidates who lost before 1994 “were weak candidates who would have lost even if they were white.”24 It would be a disservice to both the individual candidates and the goals of the Voting Rights Act for federal courts, years after an election, to scrutinize the qualifications of minority candidates who run for public office in jurisdictions with historically white-only officeholders. Arguments regarding a candidate’s qualifications for office are saved for the voters in an election. We reject any rule that would require the parties to cast aspersions on the qualifications of the victorious or defeated minority candidates.
Here, Plaintiffs submitted evidence that several aspects of the 1994 city council election were unprecedented: the percentages of non-Hispanic crossover voting for two Hispanic candidates; the endorsements of several white politicians and predominantly-white organizations for those candidates; the amount of financial support contributed by non-Hispanic businesses and individuals to those campaigns; and the amount that Maldonado spent on his campaign. It was well-known throughout Santa Maria that the district court was awaiting the results of that election. Days before the election, Maldonado told a local newspaper that his victory would prove “Santa Maria is not racist.” Plaintiffs have raised a triable issue of fact in whether the 1994 city council election was representative of typical voting behavior in Santa Maria. At trial, if Plaintiffs prove that unusual circumstances surrounded that election, the district court must discount the election results in its Gingles prong three analysis.
V. STRICT SCRUTINY
Finally, the district court based its grant of summary judgment for the City on its determination that any relief provided to Plaintiffs would not survive strict scrutiny. If the district court determines that Santa Maria’s at-large election system impermissi-bly dilutes the Hispanic vote, it may order the creation of a district system. The creation of new districts must be reasonably compact, “taking into account traditional dis-tricting principles such as maintaining communities of interest and traditional boundaries-” Bush v. Vera, 517 U.S. 952, 977, 116 S.Ct. 1941, 1960, 135 L.Ed.2d 248 (1996). As the Supreme Court recently emphasized, strict scrutiny applies if race is the predominant motive in redrawing district lines. See *559Abrams v. Johnson, 521 U.S. 74, —, 117 S.Ct. 1925, 1936, 138 L.Ed.2d 285 (1997). The Court has assumed that compliance with Section 2 can be a compelling state interest. See Bush, at 978, 116 S.Ct. at 1960-61; Miller v. Johnson, 515 U.S. 900, 921, 115 S.Ct. 2475, 2490-91, 132 L.Ed.2d 762 (1995).
The district court’s consideration of the scrutiny required for a possible future remedy was premature. Before the court concerns itself with strict scrutiny, it first must find that Plaintiffs have proved the three Gingles preconditions, that the totality of the circumstances shows that the Hispanic vote is diluted, that a district system would remedy the injury, and that race is the predominant factor in drawing district lines. Raising the strict scrutiny issue in its analysis of the third Gingles requirement was erroneous.
VI. CONCLUSION
Accordingly, we reverse the dismissal of the case as moot, the grant of summary judgment for the City on the third Gingles precondition, and the denial of summary judgment for Plaintiffs on the first Gingles precondition. .We take no position as to whether Plaintiffs may establish any of the three Gingles requirements. Plaintiffs, of course, are not required to prove any of the three preconditions in opposing a motion for summary judgment; they need only show a genuine dispute of material fact. The district court must consider a motion for summary judgment or the merits at trial under the legal rules established here.
REVERSED AND REMANDED.
APPENDIX
Estimated Percentages of Votes Cast by Non-Hispanic and Hispanic Voters in Santa Maria Elections
H — Hispanic candidate
E — Elected
Overall finish Nonr-Hispanic Hispanic
1988 City Council Election
Urbanske (E) 61.4 49.5
Firth (E) 46.7 4.2
Snyder 34.6 27.4
Martel (H) 15.7 45.6
1990 Mayoral Election
Hobbs (E) 71.4 13.3
Rodenberger 16.9 15.8
Lopez (H) 6.6 42.6
Castillo (H) 5.1 28.3
1990 City Council Election
Tunnell (E) 49.0 12.8
Orach (E) 48.4 8.7
Engel 27.2 23.1
Droegemeier 26.3 20.5
Guevara (H) 8.6 66.2
1992 City Council Election
Urbanske (E) 50.4 38.4
Miyoshi (E) 35.5 31.5
Firth 36.4 4.7
Varni 28.3 12.8
Anderson 12.1 22.1
Jerger 13.3 1.1
Castillo (H) 4.7 35.9
199k City Council Election
Centeno (E)(H) 35.2 39.8
Maldonado (E)(H) 26.1 65.4
Orach 42.2 1.1
Benedetti 24.2 4.0
Minami 9.4 22.3
Chenoweth 11.4 1.6
Guevara (H) 1.9 31.6
Hollister 10.4 1.2
Hosepian 5.5 3.1
Walker 3.2 10.6
Hunt 3.6 2.8
Watson 1.9 4.2
. Judge Trott joins in all of the Opinion. Judge Wallace joins in all but parts IV.B and IV.D. Judge Hawkins joins in all but part IV.A.2.
. Wc recognize that labels such as Hispanic, Hispano, Mexicano, Latino, white, Anglo, black, African-American, or Asian have different meanings to various groups. For purposes of simplicity and convenience, we attach no particular significance to any label and use broad terms such as "white,” "Hispanic,” and "black.”
. One expert performed a "bivariate ecological regression analysis” and a "homogeneous precinct analysis” for all five elections. Another conducted an exit poll analysis for the 1990 city council and mayoral elections.
. In other words, the first-preferred candidate among Hispanies received the most Hispanic votes; the second-preferred candidate received the second most votes from Hispanic voters.
. " ‘At this time in Santa Maria, we have a Mexican problem,’ Hobbs claimed. ‘We have a difficulty with scads of illegal aliens that have come across the border, and they’ve made our neighborhoods look not like Santa Maria neighborhoods. In certain streets, people (are) gathered around in groups drinking beer, smoking cigarettes. It's a formidable experience for a lot of the older people who have been here for a long time.’ ” Jeanne Sparks, Hobbs: Illegal alien situation out of hand, Santa Maria Times, July 17, 1990, at 5. Hobbs stated he was not referring to " ‘our Santa Maria Mexicans that have been here forever.’ ” Id. Hobbs proposed that U.S. tax money " ‘that’s wasted up here in programs and welfare’ " be used to build colonies in Mexico to " ‘make it so good for them that they would be willing to slay there.' ” Id.
. Interestingly, Miyoshi, an Asian candidate, campaigned on his “[c]ommitment to address the ‘Mexican problem.’ ” He promised he would "address the problem with sensitivity and fairness.”
. At that time, the City's defense to Plaintiffs' vote dilution claim was Plaintiffs could not create a district in Santa Maria in which Hispanics comprised a majority of the voting age citizen population (dubbed the "Thornburg district”). Plaintiffs' new evidence showed that Hispanics were 59% of registered voters in the proposed Thornburg district in July 1994.
. Also, the Hispanic vote seems to have been a decisive factor in the defeat of George Hobbs for mayor in 1994. The record does not contain, however, a statistical report of the Hispanic and non-Hispanic voting percentages for that election.
. In the 1990 mayoral election, two Hispanic candidates received 6.6% and 5.1% of the non-Hispanic vote.
. Endorsements were received from: Curtis Tunnell, long-time city council member; Casey Kyle, former mayor; Harrell Fletcher, former county supervisor; The Committee to Improve North County, informal political action committee of the Santa Maria Chamber of Commerce; and Santa Maria Firefighters Political Action Committee.
. Tim Epperson, Council Candidate gets Threatening Letter, Santa Maria Times, November 3, 1994, at C.
. Plaintiffs had moved for partial summary judgment, claiming that Hispanics were sufficiently large and geographically compact to constitute a majority of the voter-eligible population in a single-member district. The district court denied the motion, finding it was mooted by the 1994 election results.
. The City requests that we take judicial notice of the 1996 mayor and city council elections results, pursuant to Fed.R.Evid. 201(b)(2). In that election, Maldonado was elected mayor, and Miyoshi and a white candidate were elected to city council. We deny the request for two reasons. We may not rely on subsequent election results in reviewing the district court’s judgment. See Romero v. City of Pomona, 883 F.2d 1418, 1420 n. 1 (9th Cir.1989), overruled in part on other grounds, Townsend v. Holman Consulting Corp., 914 F.2d 1136 (9th Cir.1990) (en banc). Also, those election results do not tell us if any of the winning candidates were Hispanic-preferred, the relevant issue in this appeal.
. Gingles' reference to a "white majority” voting bloc must be read in the context of the facts presented to the Court. Because Section 2 protects both racial and language minorities, we will refer to the relevant majority voting bloc as the "Section 2 majority.”
. The so-called "Senate factors” are:
(1) the extent of any history of official discrimination in the state or political subdivision affecting the right of a member of a minority group to register, vote, or participate in the democratic process; (2) the extent to which voting in government elections is racially polarized; (3) the extent to which the state or political subdivision has used voting practices or procedures that tend to enhance the opportunity for discrimination against the minority group (for example, unusually large election districts, majority vote requirements, prohibitions against bullet voting); (4) exclusion of minorities from a candidate slating process; (5) the extent to which minority group members in the state or political subdivision 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; (6) the use of overt or subtle racial appeals in political campaigns; (7) the extent to which minorities have been elected to public office in the jurisdiction; (8) the extent to which elected officials are unresponsive to the particularized needs of minorities; and (9) the policy reason for the jurisdiction's use of the challenged practice. See S.Rep. No. 97-417, 97th Cong., 2d Sess. (1982), reprinted in 1982 U.S.C.C.A.N. 177, 206-07. Factors two and seven are the most important. See Gingles, 478 U.S. at 48 n. 15, 106 S.Ct. 2752.
. In fairness lo Justice Brennan, he did not suggest that only minority candidates could be considered minority-preferred. Rather, his discussion of the irrelevance of a candidate’s race was in the context of defining racial polarization.
. The Clarke court ultimately concluded that the third Gingles prong was not met because black candidates were not "usually” defeated. See Clarke, 40 F.3d at 813.
. Our rule does not contravene the Voting Rights Act’s prohibition against proportional minority representation. See 42 U.S.C. § 1973(b). The election of minority-preferred white candidates is relevant to determining the degree of racially polarized voting in a community. We recognize "the ultimate right of § 2 is equality of opportunity, not a guarantee of electoral success for minority-preferred candidates of whatever race.” Johnson v. De Grandy, 512 U.S. 997, 1014 n. 11, 114 S.Ct. 2647, 129 L.Ed.2d 775 (1994). When non-minority voters may elect their chosen representatives who are non-minorities, but minority voters cannot elect their chosen minority representatives, equal opportunity to elect is not achieved.
. As of 1994, Hispanics comprised only 28.7% of the voting age citizen population in Santa Maria. Even if all Hispanic eligible voters voted for the same Hispanic candidate, those total *555votes could conceivably not be enough to ensure election. See Sanchez, 97 F.3d at 1319 (Gingles does not require an absolute monolith in the white or Hispanic bloc vote). The 1990 city council election is illustrative. The Hispanic candidate received 66.2% of the Hispanic vote but finished last in a field of five. Only a combined total of 21.5% of Hispanics voted for the two winning white candidates. Each of the third- and fourth-place candidates received more Hispanic votes than the two winners. Obviously, bullet-voting would not have been effective.
. The City argues that bullet-voting bestows as much voting power to Plaintiffs in an at-large system as in a district system because, under a district system, Plaintiffs would be entitled to cast only one vote every four years. By bullet voting under the current system, Plaintiffs may cast one vote every two years. Plaintiffs’ claim is not concerned with the number of votes cast. Rather, Plaintiffs' claim focuses on the equal voting of all Santa Maria voters. A district system would require all voters to cast an equal number of ballots in the same period.
. Plaintiffs deposed most of the principal politicians or local leaders who endorsed or campaigned for Centeno and Maldonado. All professed reasons for their actions unrelated to the plaintiffs' pending lawsuit.
. Contrary to the City’s argument, Collins has not been implicitly overruled. In Collins v. City of Norfolk, 883 F.2d 1232 (4th Cir.1989) ("Collins II"), the court affirmed the district court's finding that the recent re-election of black candidates was attributable to special circumstances. See id. at 1243. Lewis affirmed Collins II’s central holding regarding the inquiry to be undertaken to determine who is the minority-preferred candidate but declined to extend it. Lewis, 99 F.3d at 612, 614.
. See, e.g., Baird v. Consolidated City of Indianapolis, 976 F.2d 357, 361 (7th Cir.1992) (minority electoral success after pending litigation "could lead white voters to throw the minority a bone”).
. Plaintiffs made similar arguments by labeling Maldonado as unqualified.