There is going to be a ton of analysis about this election, like there always is.
The polls showed some things that led to certain expectations. Some didn't happen and others were a bit closer than expected.
While not to the extent predicted, the polls did generally get the presidential election right. In both Texas and Florida they seemed off on predictions about Latino voting.
When all is said and done the eventual electoral colllege vote is going to be in the predicted range.
While the popular vote is pretty lopsided it is a little below the predicted level.
The biggest disappointments are in the down ballot races and the only conclusion I can come to is that we had more split ballots this year than usual. More republicans who couldn't vote for Trump still supported the rest of the slate.
It should be noted, as the old saying points out, all politics are local. So in each race the reasons for the outcome could simply be locl reasons.
What is always important to remember is that polls are based on statistics. You sample a few thousand people and extrapolate to millions. In order to do that you make a lot of mathematical assumptions. Assumptions don't vote, people do.
When you have a candidate who is untraditional and divisive, I think it is very had to get those assumptions correct.
Pollsters are effectively excellent mathematicians but people aren't as predictable as numbers are.
No comments:
Post a Comment