The 2018 Midterms Game Day Thread of Angst, Worry, (and maybe some hope?)

So, she make residency on the date she is sworn in.

The gop already lost in court. They are just trying to cheat again.

I’m not sure at what point you should take up arms against the government, but I’m pretty sure I can see it from here.

Valdao concedes in the CA-21. 40 seat pickup officially, with the NC-09 a complete mess.

In your rear-view mirror you mean, right?

Projection… Intensifies!

Hey, if all that voter suppression and gerrymandering didn’t work, they must be cheating.

This year’s 8.6 point House popular vote win for the Democrats is the greatest on record for a minority party heading into an election. This dates all the way back to 1942, when the Clerk of the House started listing the House popular vote in its after-election statistics document. That is, the Democratic performance this year was better than the minority party’s in the previous 38 elections.

The Democrats won by a wider margin this year than Democrats did in 2006 or Republicans did in 1994 or 2010. They beat the previous record of 8.5 points Republicans won by in 1946. (Note: I’m assigning the Democrats and Republicans the votes for their candidates on other lines via electoral fusion.)

Importantly, Democrats didn’t just win because Republicans turnout was low. This year had the highest turnout for any midterm election at 50.1% in the last 100 years. Turnout was about 35 million more people than it was four years ago, when Republicans expanded their House majority.

That is some ballsy bullshit. Projection intensifies indeed.

Fixed that for you.

But the truth is so inconvenient.

It’s weird that the GOP picked up, what, 10-15 more seats in '94 with a lower vote percentage? Does gerrymandering account for this? (I have no idea to what extent and in which direction districts were gerrymandered in '94)

I don’t know the background, but why is the metric always “number of seats picked up”. Absent gerrymandering and distribution issues, a certain lead in the popular vote should predict a certain outcome in terms of house representation, not in the delta of house representation from one congress to the next.

Meaning the precise number of seats is too granular to be expected to align with a nationwide number?

It’s easier to win 50 new seats when you’re starting from 150 than when you’re starting from 200. What you need to look at is the total seats won, not the change in the number of seats won.

Let’s say once year Dems started with 30% of the seats but won the popular vote by 5% and ended up with 60% of the seats. Am awesome 30% gain! Another election the Dems start with 45% of the seats and win popular vote by 5% and end with 60% of the seats. Good, but only a 15% gain.

Why did 5% advantage in popular vote only lead to a 15% gain in the second case?

I’m not saying this is what happened, but it’s always seemed weird that people compare popular vote lead with CHANGE in representation.

Suppose that in the 2020 election, the House vote totals were exactly the same as in the 2018 election. Another outstanding performance by Democrats! Expected gain: zero seats.

That’s the difference between “the best” and “the most improved”.

The Reps won a total of 230 seats in 1994. It looks like the Dems will have 235 for the next Congress, is that right?

Yeah, in both cases the loser did slightly better than their national average: that is they earned 46-7% of the seats with about 45% of the vote.