A stark finding from part of a project I'm working on (more TK), but it looks like Democrats' electoral college-popular vote divide increases as turnout approaches 70-75%, then explodes in the opposite direction as FL, AZ, NC, and TX flip. 1/3
This should probably worry Dems more than it excites them. Turnout is unlikely to be high enough to solve all of their issues without courting swing voters. In the meantime, whites in WI, MI and PA will have a disproportionate say in deciding the election. 2/3
3/3 In some ways, the electoral college acts like a time machine for American politics. Progressives and liberal activists like to dream of a day when their moral and policy "superiority"—which is often a majority, mind you—can win elections, but America is not there yet.
Bonus/3 Very similar to stuff that @Nate_Cohn has been working on for a while, btw -- but hopefully with a bit more complexity, and updated data, in scenario generation. nytimes.com/2019/07/15/ups… 
This is right. When I talk about these things, it's usually in response to bad conventional wisdom among the media or pundit class, not among actual practitioners (who have usually been having these debates for years by the time analysts take sides)
