34 LOUISIANA v. CALLAIS Opinion of the Court The Robinson court erroneously concluded that the plaintiffs’ illustrative maps protected incumbents because the maps left all six Representatives “in the district where they currently live” and “could avoid incumbent pairing.” 605 F. Supp. 3d, at 830. That observation missed the point: An incumbent is not protected if he or she will lose re-election. And because the plaintiffs’ illustrative maps failed to protect all the incumbents that the State sought to shield, the plaintiffs did not meet their burden on this precondition. Nor did the plaintiffs meet their burden on the second and third Gingles preconditions. To show racially polarized voting, the Robinson plaintiffs offered evidence that black and white voters consistently supported different candidates, but their analysis did not control for partisan preferences. See 605 F. Supp. 3d, at 840. Even if the Robinson plaintiffs had met their burden on the Gingles preconditions, they still would have failed to show an objective likelihood of intentional discrimination based on the totality of circumstances. The Robinson court went through the nine Senate Report factors, but none of the evidence it cited showed even a plausible likelihood of intentional discrimination by the State. Much of the cited evidence—such as the low number of black Louisianans who have been elected to Congress in recent decades— failed to disentangle race from politics. See 605 F. Supp. 3d, at 845–846. Indeed, the court observed that black voters have been aligned with the Democratic party for decades and that issues discussed by that party appealed to black voters. Id., at 845. Those observations should have undercut, not strengthened, any showing of intentional racial discrimination because race and politics are so intertwined. The Robinson court also relied on the “ ‘sordid history’ ” of intentional discrimination by Louisianian officials in the decades before the Voting Rights Act’s passage. Id., at 846. And it cast aside as “irrelevant” the lack of evidence that