Wait, what?

So, I called the House for the Republicans at 8:43 PM Eastern time. CNN was nearly in tears, Nate Silver had lowered the odds of Republicans holding the House from 1 in 15 to 1 in 2, and there had been an eight-point turnaround from the pre-election polls favoring the Democratic candidate for Florida governor. Of the two key early House races involving vulnerable Republican incumbents, the one in Virginia went Democrat, the one in Kentucky held.

Game over. Right? It looked like my scenario of the Republicans losing a few seats, but not their House majority had proven correct. So, I called it and turned in.

Then I wake up this morning to reports of +34 Democrats in the House, +3 Republicans in the Senate.

WHAT. THE. HELL?

Now, I don’t mind being wrong, which I obviously was, but I do like to know why. And this combination of being correct about a few things while getting the larger element wrong is puzzling. How could most of the early metrics I’d chosen as indicators favor the Republicans and still produce end results like this? My first stab at explaining the dichotomy:

  1. Trump turned the most dangerous areas with his campaigning. Where did he campaign the most heavily? Indiana and Florida. Where did Republicans seriously outperform the polls from the day before, by as much as eight percent in the case of the Florida governer’s race? Indiana and Florida. I should have known to discount the Trump effect elsewhere.
  2. The non-incumbency factor. 40 Republican incumbents retired and Democrats took 34 seats. Due to the nature of American politics, it’s always easier for an incumbent to hold his seat than for a newcomer to claim it, even in a favorable district. The numbers don’t match up perfectly, as some of the flipped seats were weakly held where new incumbents swept in on Trump’s 2016 coattails, but I doubt that synchronicity is entirely coincidental.
The strangest thing is the way that Republicans gained three seats in the Senate, which of course demonstrates that although the Democrats took the House, there was no Blue Wave of the sort long predicted by the media. And as a bonus, let me observe that the primary lesson of the election appears to be that identity trumps even economic self-interest for the diverse tribes of not-America.

Blacks voted 89.9 percent Democrat. The “natural conservatives” voted 72 percent Democrat.

For once, Bill Kristol is correct.

I’ve always disliked the phrase “demography is destiny,” as it seems to minimize the capacity for deliberation and self-government, for reflection and choice. But looking at tonight’s results in detail, one has to say that today, in America, demography sure seems to be destiny.

It is becoming increasingly evident that there is no such thing as a non-white America any more than there is a Jewish Palestine. Whatever it is, whatever its benefits may be, whatever it may become, it simply will not be “America” as Americans have known it for 200 years.

UPDATE: The Senate is looking even better now at 55-45.