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

[jsnell]

@Strollen

Looking at some 20th century elections I don’t think it would be as high as 260 closer to 240. But as you say there isn’t a way of calculating it.

That’s not what I said! It would be easy to calculate, but I’m not invested enough in this discussion to go and dig for the raw data :)

Here’s a random image search result for congress popular vote vs. seats (chosen just since they tried fitting the theoretically proper curve on it rather than a straight line):

https://www.researchgate.net/figure/Seats-votes-curve-in-congressional-elections-1972-2014_fig1_291016647

It would be nice to have the exact numbers instead of just reading a graph, but it’s clear that the intercept for 54% popular vote is somewhere in the range of 59-60% of the seats.

@Strollen

Gerrymander works best in close elections 51-52%, in purple state but at some point it backfires. We came very close to that in this election and argue we saw that in Orange County, CA.

But you’re just repeating the same mistake here as in the original post. You can’t use these elections as an example of “Gerrymandering works until it doesn’t” since Gerrymandering didn’t backfire. It won the Republicans 25 seats more than they should have expected.