You’ve done a great job laying out evidence of a nation split into separate echo-chamber media bubbles. You’ve also nicely explained how the extremes have evolved so the overall picture looks more like a horseshoe. I’d like to argue that picture is incomplete. It’s missing the rise of a populist nationalist movement that had coalesced out of segments disaffected with the Progressive Left capture of the Democrats. Rather than only seeing the electorate as split into immutable warring camps, the broader perspective captures dynamic themes and sees factions realigning. For an example consider the phenomenon of voters who pulled the lever for Obama but who refused to do so for Hillary. We don’t know enough about why they switched, but they did make a switch and did not stay in their original bubble. Beyond that you may want to consider that significant segments of the white, Asian, and Latin populations are moving away from the Democrats. Hillary’s loss of white women in particular was no fluke: it would have been much worse but for Trump being such a misogynistic pig. Whether the Russiagate campaign will appeal to those who defected in 2016 is not clear, but there’s a good likelihood it only rouses those still in the bubble.
I speculate that fixed bloc thinking is what caused so many pollsters to get it wrong in 2016. Yes the bubbles and horseshoe do capture parts of the picture, but there are other parts, parts in which people are switching sides, that are easily missed if you stare too hard at the bubbles.