May 14, 2016
SteveS wrote:I have very little experience following pollsters--never paid much attention to political polls, but this year I have been following Jim's posts with 538 as a possible future gambling interest, and noticed Silver had Hillary a 99% chance of winning Michigan and Silver himself stated Michigan was a huge upset with all the combined Polls he accumulates for 538. I interrupt this to mean Silver was "shocked" to the outcome in Michigan. Are you saying Michigan is unique to normal polling processes compared to other states and this is the reason 538 failed with its combined polls forecast in Michigan?
Yes, I think it was a huge upset, and yes, I disagree with JSAD, the pollsters were taken by surprise. That's why I said the event had psychological impact. But it was
only psychological impact. JSAD is right that there was almost no
practical impact of that win. (He's also right that the news media had a huge run with it.)
Bernie had a great day. He had a Harry Truman moment: He went to bed early, assuming like the rest of the world that he'd come in 10-20 points behind Hillary in a state she was widely expected to win. They had to wake him up to tell him he'd won, and then he hopped out there, hair even more disheveled than usual, and gave a roaring speech. OTOH, I don't think it was as big event for Hillary. Losing is never fun, and she'd worked fiercely for the Flint, MI people in particular and had a leg up, but her trackers knew Bernie had been gaining. Like I said a few minutes ago in another thread about the Obamacare court case, this was surely
disappointing to her, but not
surprising the way it was for TV viewers.
And you can't think of this win-lose in the same sense as, say, a football game. This was a football game where both teams won, and nobody got any practical advantage except for enthusing their cheering sections. (But that makes it a matter of mundane astrology - the story of the cheer block, not of the candidates.) At the end of the day, Bernie got 67 delegates and Hillary got 63, almost the same.
Also, I think that "99%" figure keeps throwing people. A 99% prediction that Hillary would win does not at all mean that she was expected to get nearly all the vote. "Winning" is defined as getting more than the other person, so this was a statistical projection that it was 99% likely Hillary would get at least
one more vote than Bernie. As it turned out, he got 49.8% of the vote (not even half) and she got 48.3%. That's close! He got 595,222 votes, she got 576,795 votes.
That "99% likely to win" prediction would have been correct if she'd gotten 18,428 more votes. That's not a lot.
So... I think you're right that there is something to track down here, but it's not an
objective difference that matters. Bernie, with lousy transits, got a really nice surprise (but not a life-altering surprise - just something on the level of, say, having a favorite daughter show up on the campaign trail when he thought she was out of the country). The public got startled and got revved up to the excitement of a last minute touchdown game. (It was very much like a last-minute touchdown by the team that was slightly behind - that analogy is very close.) So mundane astrology is where I'd expect to see the most, I wouldn't expect much past routine frustration and grumbles for Hillary, and I'd expect a nice surprise in Bernie's charts.
, no different when Trump was elected Prez the Left (Dems) had a meltdown. But if the Senate remains Rep majority, we will see the same so-so—gridlock. But, if Bernie happens to get a majority Senate. 