This article tells you pretty much all you need to know about the "known unknowns" a week before the election (and really, right through election day).
https://www.realclearpolitics.com/articles/2018/10/30/uncertainties_loom_as_midterms_enter_final_stretch.htmlThe most interesting and, I think important, part of Trende's analysis is this: The RCP polling aggregates currently predict Dems will net approximately 11 seats among the races leaning one way or another, with an additional 30 races rated as "toss ups." But it would be foolish to assume that "toss up" means the statistical equivalent of a coin flip, and that these races will break roughly 50/50. They are more likely to break significantly one direction or the other.
And the truth is, none of us have any idea. Which is what makes predictions like Nate Silver's 83.7% so absurd. It's meaningless. If he's wrong, he'll just shrug and say "16.3%".
Trende thinks the toss ups are more likely to break for the Dems. I think they're more likely to break for the GOP. My sitting at a bar reasoning for this is (1) recent events, on net, seem to be favoring the GOP, and (2) scant data on early voting seems to be in line with the older, whiter demographic typical of midterms, which may indicate that pollsters are slightly overestimating the Dems' enthusiasm advantage in their sampling.
Elaborating a bit more on the first point, several recent issues have played to GOP strengths: the hot economy (highlighting GOP tax cuts and polling indicating higher trust in GOP on managing the economy), Kavanaugh smear blowback (highlighting the Dems' mob tactics and disregard for due process), and the caravans (highlighting Dems' weakness on immigration and American's desire for stronger border enforcement).
So we'll see. I think the Dems pick up 19 in the House and the GOP picks up 4 in the Senate. But I wouldn't be shocked if the Dems pick up 30 in the House. I
would be shocked if the GOP picks up fewer than 2 Senate seats.
