This case is on remand from the United States Supreme Court. See Collins v. City of Norfolk, — U.S. -, 106 S.Ct. 3326, 92 L.Ed.2d 733 (1986) (vacating Collins v. City of Norfolk, 768 F.2d 572 (4th Cir.1985)). The Court remanded the case for consideration in light of Thornburg v. Gingles, — U.S. -, 106 S.Ct. 2752, 92 L.Ed.2d 25 (1986) (“Gingles ”).1
The facts were developed succinctly in this court’s first opinion, see 768 F.2d at 573-75, and fully in the trial court’s opinion, see Collins v. City of Norfolk, 605 F.Supp. 377 (E.D.Va.1984). The case involves a challenge to the City of Norfolk’s at-large system of voting for city council members. The plaintiffs allege that this system violates the requirements of § 2 of the Voting Rights Act, 42 U.S.C. § 1973.2 The Senate Judiciary Committee Majority Report accompanying the bill that amended § 2 in 1982 noted seven “typical factors” that were probative of a § 2 violation. These factors were originally derived from White v. Regester, 412 U.S. 755, 93 S.Ct. 2332, 37 L.Ed.2d 314 (1973), and they form the framework for § 2 analysis under the amended statute. Interpretation of these factors and their application to the facts at hand are the central problems in this case:
1. The extent of any history of official discrimination in the state or political subdivision that touched the right of the members of the minority group to register, to vote, or otherwise to participate in the democratic process;
2. the extent to which voting in the elections of the state or political subdivision is racially polarized;
3. the extent to which the state or political subdivision has used unusually large election districts, majority vote requirements, anti-single-shot provisions, or other voting practices or procedures that may enhance the opportunity for discrimination against the minority group;
4. if there is a candidate slating process, whether the members of the minority group have been denied access to that process;
*9355. the extent to which members of the minority group in the state or political subdivision bear the effects of discrimination in such areas as education, employment and health, which hinder their ability to participate effectively in the political process;
6. whether political campaigns have been characterized by overt or subtle racial appeals;
7. the extent to which members of the minority group have been elected to public office in the jurisdiction.
S.Rep. No. 417, 97th Cong., 2d Sess. 28-29, reprinted in 1982 U.S.Code Cong. & Ad. News 177, 206-07 (emphasis added). The Report stresses that this list is neither comprehensive nor exclusive and that “there is no requirement that any particular number of factors be proved, or that a majority of them point one way or the other.” Id. at 29, 1982 U.S.Code Cong. & Ad.News at 207. The Report “espouses a flexible, fact-intensive test for § 2 violations,” Gingles, — U.S. at-, 106 S.Ct. at 2764, but it limits the proof of a § 2 violation in three ways.
First, electoral devices, such as at-large elections, may not be considered per se violations of § 2. Plaintiffs must demonstrate that, under the totality of the circumstances, the devices result in unequal access to the electoral process____ Second, the conjunction of an allegedly dilutive electoral mechanism and the lack of proportional representation alone does not establish a violation____ Third, the results test does not assume the existence of racial bloc voting; plaintiffs must prove it.
Id. The Gingles decision is essentially a gloss on these factors and limiting circumstances.
The essence of a § 2 claim, as characterized in Gingles, is that some electoral characteristic interacts with social and historical conditions to create an inequality in minority and majority voters’ ability to elect their preferred candidates. — U.S. at-, 106 S.Ct. at 2764. It is essential for plaintiffs to prove that minority and majority voters consistently prefer different candidates, and that the majority, by virtue of its numerical superiority, will usually defeat the choices of the minority. Id.
The minority must show three things as a threshold matter:
1. That it is sufficiently large and geographically compact enough to make a majority, if it were voting in a single-member district;
2. that it is politically cohesive; and
3. that the majority votes sufficiently as a bloc to enable it — in the absence of special circumstances, such as an unopposed minority candidate — usually to defeat the minority’s preferred candidate. Id. at---, 106 S.Ct. at 2766.
The implication of this gloss on § 2 is that, of the seven primary factors on the Senate Report list, two are typically the most important: the existence of racially polarized voting, which establishes both cohesiveness of the minority group and the power of white bloc voting to defeat the minority’s candidates, and the actual results of minority-preferred candidates in winning elections.
The opinions in Gingles, while not amounting to lucid explication of all features of the amended § 2, make clear that certain legal errors were committed in trying this case below. We summarize those errors here, and remand to the district court for reconsideration in light of Gingles.
I. RACIALLY POLARIZED VOTING
The legal standard for the existence of racially polarized voting looks only to the difference between how majority votes and minority votes were cast; it does not ask why those votes were cast the way they were nor whether there were other factors present in contested elections, such as “white backlash.” The court should inquire separately into minority and majority voting, to see whether members of the minority usually vote for the same candidates — if so, there is the requisite minority political cohesiveness — and to see whether the majority vote is sufficiently homoge*936nous to cancel the minority vote plus the cross-over vote. These inquiries together determine whether there is a “pattern of racial bloc voting that extends over a period of time,” a critical factor in a claim of vote dilution under § 2.
The district court in this case used an erroneous definition of racially polarized voting. The plaintiffs’ definition, offered to the court, was that “racially polarized voting occurs when the majority of one racial group votes contrary to the majority of another racial group.” The court rejected this in favor of the defendants’ definition, which contained three elements:
First, the presence or absence of “white backlash,” that is, whether white voters turn out in greater numbers than usual in response to the potential election of black candidates. Second, the voting patterns of black and white voters over a period of years. Finally, whether whites attempt to limit the field of candidates.
605 F.Supp. at 386. This choice was wrong, as Gingles makes clear. The trial court should not inquire into the first and third elements in determining the existence of polarized voting.
There is no consensus in Gingles on the proper role of causation evidence concerning voting patterns, except that such evidence cannot be used to rebut the claim that there is racially polarized voting.3 The district court’s definitional use of the concept of “white backlash” and the lack of white attempts to limit the field of candidates cannot be defended on the basis of any of the Gingles opinions. On remand, the court should pursue the question solely through the analysis of voting patterns, without seeking to explain why those patterns exist.
The district court in this case also rejected plaintiffs’ studies of voting patterns due to methodological problems. The problems included, but apparently were not limited to, the lack of data from the mostly white Norfolk naval station residents. See 605 F.Supp. at 387. It is unclear from the opinion whether these methodological problems, leading to rejection of the plaintiffs’ polarization studies, also included the failure to isolate variables other than race in the voting behavior analysis. It is within the trial court’s discretion to find that methodological flaws undercut the probative value of a study so deeply as to render it inadmissible. See, e.g., United States v. 25.406 Acres of Land, 172 F.2d 990, 993 (4th Cir.), cert. denied, 337 U.S. 931, 69 S.Ct. 1496, 93 L.Ed. 1738 (1949); Fed.R. Evid. 403. But the failure to include causative factors that purport to explain voting patterns cannot be a basis for rejecting plaintiffs’ studies on the existence of polarized voting. And given that this case was tried before a judge, the studies should have been admitted unless their methodological flaws were so severe as to render them irrelevant or totally misleading.
The racially polarized voting factor is one of the two cardinal factors identified by Gingles. The district court’s error in defining it cannot be discounted. It is crucial to the legitimacy of the political system that questions about its fairness be answered *937unambiguously. Cf. United States v. Carolene Products Co., 304 U.S. 144, 152 n. 4, 58 S.Ct. 778, 783 n. 4, 82 L.Ed. 1234 (1938) (“prejudice against discrete and insular minorities may be a special condition ... curtailing] the operation of those political processes ordinarily to be relied upon to protect minorities, and may call for a correspondingly more searching judicial inquiry”). Overly restrictive definitions of elements required to be proven by civil rights plaintiffs can act as blinders on justice, and so must be avoided.
II. MINORITY ELECTORAL SUCCESS
The Court in Gingles reversed the part of the district panel decision that found a § 2 violation in North Carolina House District 23 (Durham). The fact that blacks in District 23 had enjoyed proportional representation since 1973 mandated judgment for the defendants.4
A black citizen has been elected to the Norfolk City Council, comprised of seven members, in every election since at least 1972. 605 F.Supp. at 405. The district court found further that “white candidates who have been the choice of a majority of the black electorate have been elected to the Council consistently over the years.” Id. at 405. The district court concluded: “blacks and candidates of choice of the black community have been elected to public office in Norfolk.” Id. at 405.
The district court did not find that the black community in Norfolk had achieved proportional or nearly proportional representation. Gingles makes clear only that a minority cannot prove a § 2 violation in the face of continued and proportional success at the polls. The myriad numbers before this panel seem to indicate a consistent level of political success by Norfolk blacks that nevertheless is not as conspicuous as that achieved by the black community in North Carolina District 23.5 On remand, the district court must determine whether, as a factual matter, black electoral success in Norfolk has been so consistent and nearly proportional as to overshadow any racially polarized voting and other factors tending to dilute minority political access.
The factfinder must use great care in assessing what counts as a “minority electoral success.” The mere election of a candidate who appears to have received votes from more than fifty percent of minority ballots does not count as a minority electoral success, when each ballot may contain votes for more than one candidate. In such a situation, if there were other candidates, preferred by a significantly higher percentage of the minority community, who were defeated in the same election, then it cannot fairly be said that the minority community has successfully elected representatives of its choice.6 Each such situation must be reviewed individually to determine whether the elected candidates can be fairly considered as representatives of the minority community. The presumption must be that they cannot, if some other candidate has received significantly more minority votes.
*938Similar difficulties arise in assessing “special circumstances” that may explain minority electoral success. The three-judge district panel in Gingles v. Edmisten found that the pendency of the § 2 lawsuit reduced the weight to be accorded black electoral success in the 1982 election. In contrast, the district court in this case found that the 1984 election of a second black to the Norfolk City Council was not a result of a “conspiracy” to moot this lawsuit. 605 F.Supp. at 393. The Supreme Court in Gingles reaffirmed the importance of investigating beyond the surface outcome of an election when evaluating minority success; a trial court must consider, for example, whether that success was prompted by an attempt to forestall Voting Rights Act litigation. See — U.S. at -, 106 S.Ct. at 2780; see also Zimmer v. McKeithen, 485 F.2d 1297, 1307 (5th Cir.1973), aff'd sub nom East Carroll Parish School Board v. Marshall, 424 U.S. 636, 96 S.Ct. 1083, 47 L.Ed.2d 296 (1976).
Moreover, the 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 black candidate’s success in 1984, while this action was pending, resulted from unusual circumstances. Never before in the history of the city had two black councilmen served simultaneously. Nevertheless, the mayor supported a second black candidate for the first time. He coupled his unprecedented support with the public statement, “After the election, the issue of black representation may become a moot point.”
The mayor’s intent, or whether he and other white leaders conspired, is not dispositive. A proper inquiry must examine the result of the mayor’s conduct and statement. If voting patterns show unusual white support for the black candidate in 1984, the legal significance of his success should be diminished. See — U.S. at- --, 106 S.Ct. at 2778-81. As long as that particularized investigation is made, however, the trial court’s findings should not be disturbed on appeal unless they are clearly erroneous.
The presence of either of these two cardinal factors — voting that is severely polarized along racial lines or sustained minority electoral success that achieves proportional or nearly-proportional representation— would weigh very heavily in the ultimate determination that minority voters do not or do have equal access in the polity. But that ultimate determination still must be made on the basis of the “totality of the circumstances.” Once the ultimate determination is made, the “clearly erroneous” test of Rule 52(a) is the appropriate standard for appellate review. See — U.S. at -, 106 S.Ct. at 2780; Fed.R.Civ.P. 52(a). However, the definitions and legal criteria — such as of “racially polarized voting” —must accord with law, and will be reviewed as legal questions.
III. OTHER SENATE REPORT FACTORS
The district court’s definition of “a candidate slating process” was overly restrictive, and should be abandoned. The district court found no group of white citizens in Norfolk acting as a formal or informal slating organization which controlled access to the ballot. The district court accepted a definition of “slating group” proposed by one of the defendants’ experts that put important emphasis on the fact that such a group proposes candidates for all available seats. Thus, two white businessmen who ran together for City Council in 1976 were not viewed as slated candidates, because they had no third running mate for the other vacant seat. Gingles did not change or refine the understanding of a “slating group” or “slating process.”
There seems to be no consensus in federal law or in political science texts on a definitive meaning of the phrase “slating group.” But there is no support in the law for the restrictive definition chosen by the trial court. In White v. Regester, 412 U.S. 755, 766-67, 93 S.Ct. 2332, 2339-40, 37 L.Ed.2d 314 (1973), and Whitcomb v. Chavis, 403 U.S. 124, 150-51 & n. 30, 91 S.Ct. 1858, 1872-73 n. 30, 29 L.Ed.2d 363 (1971), two important sources for the phrase that ultimately made its way into the Senate *939Report, the Court viewed “slating” as essentially involving the endorsing of candidates. The Court did not explore whether the slating groups involved in those cases always filled all slots on the ballot. It strains common sense to suppose that an organization could regularly put up candidates for ninety-nine of one hundred slots on a ballot, yet escape being labeled a “slating organization” for purposes of § 2 analysis. On remand, the district court should avoid the use of highly artificial definitions to restrict the Senate Report inquiry. While the fact that the alleged slating group in Norfolk sometimes endorsed candidates without giving endorsements for all available seats might be a factor in finding that it was not a slating organization that denied access to minorities, that fact does not contravene the Senate Report factor as a matter of law.
Unlike the situation in the disputed North Carolina districts at issue in Gingles, see — U.S. at---, 106 S.Ct. at 2759-62, black voter registration and turnout have recently increased in Norfolk. The district court acknowledged that fact, see, e.g., 605 F.Supp. at 402, 404, as did this court in its first opinion, see 768 F.2d at 574-75. This factor is probative, but not dispositive, of black electoral participation, and there is no reason to believe that the district court viewed it as inordinately important.
The district court found no special electoral devices used in Norfolk that “enhanced the opportunity for discrimination against the minority group.” 605 F.Supp. at 403. Many particular features of an electoral system, such as the lack of district residency requirements and the presence of staggered terms of office, may under some circumstances make it structurally more difficult for minorities to be elected. Cf. Gingles, — U.S. at-, 106 S.Ct. at 2761 (“procedures may operate to lessen ... opportunity”) (emphasis added). The question whether those procedures do so operate is a factual one, to be reviewed on the same basis as other factual matters. There is no indication that the district court clearly erred in finding no discriminatory devices in Norfolk.
Finally, the district court considered the plaintiffs’ allegations of unresponsiveness to the needs of the minority community and found that the defendants had rebutted most of these allegations. The district court ruled that the plaintiffs had to prove that the removal of 1,800 families, most of whom were black, for redevelopment for housing for middle-income families, was racially motivated. This was the wrong legal test. Proof of discriminatory intent, as Gingles noted, is not required by § 2 of the Act, and it is not an element of proof of the factors mentioned in the Senate report. On remand, the district court should determine whether the removal of these black families by city officials was unresponsive to their needs without requiring proof that the officials were racially motivated.
On remand issues may arise which are not addressed by this opinion. In this event, the district court should follow the mandate of the Supreme Court and consult Gingles for guidance.
REVERSED AND REMANDED.
. The Gingles decision affirmed in part and reversed in part a decision of a three-judge panel, Gingles v. Edmisten, 590 F.Supp. 345 (E.D.N.C.1984), which had held that a redistricting plan of the North Carolina General Assembly diluted black citizens’ votes in a number of challenged districts.
. Those requirements, as amended in 1982, read 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, or in contravention of the guarantees set forth in section 1973b(f)(2) of this title, as provided in subsection (b) of this section.
(b) A violation of subsection (a) 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, as amended, 96 Stat. 134 (emphasis added).
. Section III(c) of Justice Brennan’s opinion for the Court is a plurality section, joined by three other Justices. It says that evidence of other causes for minority bloc voting, such as economic status, party affiliation, incumbency, and the race of the candidate preferred by the minority, is irrelevant to the whole § 2 inquiry. Justice White wrote separately to say that the race of the candidates is an important additional factor that could-be considered in the inquiry into the existence of racially-polarized voting. One cannot infer from this separate opinion, as plaintiffs suggest, that Justice White agrees that other factors, such as socio-economic status, have no place in the entire § 2 case. Justice O’Connor, joined by the Chief Justice and Justices Powell and Rehnquist, concurred separately and said that these other factors should not come in on the question of the existence of racially polarized voting, but should be admitted on the question of whether white bloc voting will consistently defeat minority candidates.
It is difficult to square the plaintiffs’ assertion that causative factors are entirely irrelevant with the need to consider special circumstances that might explain particular minority successes, such as white reaction to the filing of a § 2 lawsuit. Those special circumstances are, in essence, other causation factors. For this reason, the separate concurrence of Justice O'Con-nor is more persuasive in explaining that causation evidence may play a role, albeit not in defining "racially polarized voting.”
. Justices Brennan and White thought that the district panel had erred as a matter of law in ignoring this degree of success, — U.S. at-, 106 S.Ct. at 2780, whereas the Chief Justice and Justices Powell, Rehnquist and O’Connor thought that the holding was "clearly erroneous” because insufficient weight was given to the black voters’ success, id. at-, 106 S.Ct. at 2794. Justice Stevens, joined by Justices Marshall and Blackmun, dissented from the holding concerning District 23, believing that the district court finding of vote dilution was supportable despite the continued election of one out of three representatives by black voters. Id. at -, 106 S.Ct. at 2796.
. One black (14%) was on the Norfolk City Council from 1968 until 1984, when two blacks (28%) were elected. Blacks constitute somewhere between 30 and 35% of the relevant Norfolk population. The question is not, however, so simple as how many blacks versus whites were elected. The court must examine the parties’ studies of voting preferences to determine which were the preferred candidates of the majority and minority communities.
. This may have been the case in the 1974 and 1980 elections in Norfolk, when black candidates apparently received votes from more than 80% of black voters, but were not elected, while one or more white candidates received votes from 56 to 74% of black voters, and were elected.