Correct. Gerrymandering means several different specific ways to cheat by drawing the districts, but one way is taking a district that’s going to be a blowout for you—say, you’re expected to reliably get 88% of the vote—and sharing that 88% with a nearby district, where you’re expected to get maybe 37%. If you draw the lines right, you can get two districts where you win with 66% of the vote, instead of winning one and losing one.

But why stop there? 88% is a huge lead, and in first past the post it doesn’t matter how much of the vote you get, so long as you get more than the next most popular candidate. It may require some truly unhinged district drawing, but what if you could get, say, five districts where you’re going to win with 46% of the vote, due to a strong (but not strong enough) third party spoiler candidate? Now you’ve spread out the voters in that 88% area and used them to bolster four other districts that you were going to lose (or were going to be competitive) into solid and reliable wins, or at least turn solid victory for the opposition into a competitive contest.

Except, oops, the guy at the top of the ticket is a literal supervillain except without any superpowers, and now it’s starting to weaken that original 88%. Now, instead of one blowout district, and instead of five solid wins, you’ve got, maybe, two competitive contests and three solid losses. If you’d left well enough alone, you might’ve still been able to win that blowout district with 58%, but because you got greedy you’ve lost everything.