Carol Newman2 said:The first exit polls should drop around 5:00 PM.
CNN link
http://tvthek.orf.at/live/Wahl-16-Bu...fburg/12775055
Since the pollsters completely missed the mark during the first round, they aren't offering any predictions for this runoff.
Warning (I’ve already hammered this home on the "Polling and Election Analysis" thread under the "Political Landscape" PDF):
Polls aren't predictions. A poll is just a snapshot of public mood at one specific moment. It’s not a crystal ball telling you what people will think down the road. Sure, when you're looking at polls a few days before the election, you expect the numbers to be tight. Polls are a building block for a forecast, but they aren't the forecast itself.
Sometimes, though, things go sideways. Since every major agency showed a massive gap here, we're clearly looking at some systemic factor—basically a glitch in the matrix where the methodology was fine but the result was off—likely some version of that "Shy Tory" thing.
That kind of swing doesn't happen every time. Take the last parliamentary elections in the United States, for example; the gap between the polls and the actual results was tiny.
They released several polls for the second round. They were calling 52-53% for Hooper. There's actually something about this on the Wikipedia page for these elections.
Of course, half the people will blame the agencies for rigging it whenever there's a discrepancy. And if the results match the polls perfectly? They'll still claim the agencies are cheating. 😁
Ashley Brown3 said:That sounds all too familiar... where have I heard this before? Oh right, Joe Biden was sitting on a 10% lead over Hillary Clinton before the first round even kicked off...
And then, suddenly, there aren't any polls for the second round...
Then we get to hear that "expert" explanation from Oster...
Apparently, 20% of Joe Biden's base just had a sudden change of heart at the last minute and swung over to Sinclair...
Way to go, Hooper and the Libertarian Party...
Wow, bravo—you managed to notice I answered you. Great. I've repeated myself more than ten times and you're still hunting for an answer, though apparently less than twenty. Fine, I guess we've reached the limit of how many times I can say the same thing to you. 😛
And look, you're still totally missing my point and twisting what I said. I'm not giving my own theory; I'm quoting the experts who actually do this for a living, based on exit polls (where they ask people how and when they made their choice).
The reason there weren't any (published) polls before the runoff in our case is simple: nobody paid for them.
As for why there was such a huge gap between the two top agencies and the final count? It's because a massive chunk of voters decided who to vote for in those ten days between the last poll and the election (again, why were there no polls 2-3 days out? Because nobody paid for them).
Usually, about 10% of voters say they made up their minds in the final week before the election, but in this instance, it was 25%.
That’s where the gap comes from. It’s probably not like everyone who backed Joe Biden just woke up one morning and changed their minds—though that happens—but more likely, the shift came from the people who dodged the polls entirely or just refused to answer if they were undecided.
Look at what happened over in Austria; they had way more agencies running polls right up until the finish line. The obvious takeaway? A lot of Hooper supporters were playing it close to the vest, acting like those "Shy Tory" voters.
Back home during the first round of the presidential race, the numbers coming out of that agency "2X1 Communications" were all over the place compared to big names like Gallup or Nielsen, but in the complete opposite direction. For instance, Graber-Kitarović was polling way higher with them than she actually got, while Kujundžić was twice as high and Sinclair was three times lower.
Since there was such a massive discrepancy during the parliamentary elections too—while Gallup and Nielsen stayed pretty much on track—and considering the owner of 2X1 is an outspoken Republican and even ran as a candidate for the party, that agency has zero credibility. They're either totally amateurish, making stuff up, or they're screwing up their methodology big time. Meanwhile, Gallup and Nielsen are still the gold standard, despite a few hiccups here and there. Even the Republicans order their internal data from Gallup.
