This is a dynamic to keep an eye on for November. At a certain point, if Trump keeps flailing, a bunch of "safe" Republican seats could be put at risk—all at the same time. (Quick thread.)
There are two competing imperatives when you put together a gerrymander—you want to keep your seats safe by building in a solid majority of your voters, but you want to get as many of them as you can. It's a balancing act.
Let's say I define "safe" as a 55-45 majority for my party, because even in a bad year, that'll be enough to keep my incumbents in office. Once I've got that number, I'll build every district to give me about that same majority.
(If I make my districts 70-30 instead of 55-45, they're theoretically "safer," but I'm wasting voters that could be packed into other districts, giving my party fewer seats. Again, it's a balancing act.)
Think of each 55-45 district as a dam, designed to keep the village below safe. Now think of a historically unpopular presidential candidate's reelection campaign as a hundred-year flood.
And crucially for this analogy, EVERY DAM IS THE SAME HEIGHT.
If all your districts are 55-45, and there's a four-point swing away from your party, all of your candidates still win. But if there's a six-point swing, they all lose. In almost every scenario, you're fine, but in almost all the ones where you're not fine, you're devastated.
Because when all the dams are the same height, a storm that breaches one of them is going to breach them all.
Now, obviously, some districts are going to wind up stronger than 55-45, even if that's your goal, and some candidates are going to have greater incumbency advantages than others, and Trump will do better in some demographics than others, and so on.
So it's not literally the case that the GOP would actually lose all their seats in any of these states, even in a worst-case Trump scenario.
But where you have really aggressive gerrymandering, either for House districts or in a state legislature, you pretty much never have a mediocre year. You almost always get what you planned for, but when you don't, things get real ugly real quick.
And 2020 is a census year, which means that the legislators who are going to be responsible for redistricting are going to be the ones who get elected this November.
Which in turn means that if a state's gerrymander gets breached this year, it'll likely be dismantled next year.
We're still a long way off. But keep an eye on it. Things could get real interesting in a couple of months.
This is one of the reasons why gerrymanders are such a growing problem for democracy—we have the tools to make them much more effective now than we did a couple of decades ago.
(But "effective" also means "fragile, once the floodwaters rise.")
It's similar how the twin towers could be taken down by a jet plane strike, but the Empire State Building would have just wound up with a big hole in its side. Before you had computers, you had to build your skyscrapers with a much bigger margin of error.
Good question. The issue with gerrymandering is that you wind up with a lot of districts with essentially the same margin, rather than some 52-48, some 54-56, some 59-51, and so on.
In a non-gerrymandered map, different districts will have different degrees of lean, and the parties' standing in the legislature will tend to rise or fall pretty smoothly on the basis of their statewide share of the vote.
But in a gerrymandered map, my party will win about the same number of seats whether we win the statewide vote by three points or lose it by three points, because each of my party's seats has been given a cushion against a bad year.
And of course you accomplish this by shoving all of the opposing party's voters into a small number of districts—if the state is 50/50 Dem/Republican, and all the Dem seats are 90/10 while all the GOP seats are 55/45, you're going to wind up with a lot more GOP seats than Dem.
Imagine that there are 20 million voters in your state, and 20 legislative districts—one million voters per district.
If you put 600K GOP and 400K Dem voters in each GOP district, but 800K Dem voters and 200K GOP voters in each Dem district, you wind up with (if I did my math right), 15 GOP seats and only FIVE Dem seats, even though the state is evenly divided.
And even if the Dems win statewide by two million votes, the GOP still wins each of their 15 seats 550K to 450K.
(This is basically what happened in the state legislature election in Wisconsin in 2018—the Dems won the popular vote 53-45, but the GOP still garnered a nearly 2-to-1 advantage in seats held.)
And remember, the Supreme Court said just last year that this is totally fine. (5-4 decision, Roberts wrote for the majority.) washingtonpost.com/politics/court…
(Read Kagan's dissent, BTW. It'll make you angry, and scared.) supremecourt.gov/opinions/18pdf…
"The practices challenged in these cases imperil our system of government."
I like Dems' chances in the coming decade on a level playing field. I think gerrymandering sucks, AND I think the Dems don't need it, so I'd be happy to see both state and federal action against it.
