Recall #InsideTheNumbers when I showed you how badly data must be manipulated to find that.
On that episode of #InsideTheNumbers, we discussed the results of Ohio in the poll we conducted for @EpochTimes and crunched live raw data on air from responses we separately gathered after @FoxNewsPoll Biden +3.
It was incompetence. A lie. Or both.
Watch it. 👆. You'll see
Worth noting, after the @FoxNewsPoll showing Biden +3 in Ohio, the Tiny Crystal Balls moved Ohio to "Toss Up".
Why? Because demographics and other predictive indicators suggested it was one?
No, folks.
Because they have tiny crystal balls and poll-readers. Bad ones at that.
If you watched the above episode, then you knew Ohio was not close. Because we commit cardinal sins and show you raw, unprocessed data and files.
They think you're just too stupid to understand that information. Laura and I, do not. We believe you can and should knot this stuff.
• • •
Missing some Tweet in this thread? You can try to
force a refresh
Thread on Exit Polls and Voter Analysis: Neither the right or the left should put much, if any stock, in either.
Over the years, after witnessing one horrible blunder after another and the damage they leave in their wake, I truly believe media should abstain from using them.
We often ignore the years when exit polls correctly called the winner, though were still showing serious issues due to at least non-response bias.
Veteran exit pollster Murray Edelman has spoken a lot about this.
Here are some notable, egregious misses.
In 2004, the Kerry Campaign was elated. The Bush Campaign was in shock re Florida. The latter was confident they over-performed exit polls. By night's end, a Kerry exit poll win turned into a historically comfortable Bush win in a razor thin state he barely won 4 years before.
Another tidbit from #Wisconsin: Overall, 68.9% of voters "believe there is a significant number of 'shy' voters who do not want to share that they are voting for @realDonaldTrump," the second highest state yet.
Remember, #Michigan posted the highest percentage of voters (37.6%) who say they are uncomfortable being truthful to pollsters.
Suburban voters (20%) were much more likely than rural and urban voters (14.2% each) to say they are "very" uncomfortable. docs.google.com/spreadsheets/d…
Not surprisingly, suburban voters in Michigan were the most likely (70.4%) to "believe there is a significant number of 'shy' voters who do not want to share that they are voting for @realDonaldTrump."
The Bobulinski interview on @TuckerCarlson is proving w/o a doubt @JoeBiden played a direct role in lucrative deals with hostile foreign entities while serving as VP.
He said claims he did not “are a blatant lie. I almost stood up and screamed liar and walked out” of the debate.
“I want to simply this for the American people as much as I can.”
Bobulinksi said the May 13 email details how the Bidens will divvy up the money from the deal and that “the Big Guy” is @JoeBiden, who’s 10% cut was funneled through his brother Jim Biden’s 20% cut.
Bobulinksi: “I remember looking at Jim Biden and saying, ‘How are you guys getting away with this?’” and he “chuckled” and said, “plausible deniability.”
That would explain lambasting those who used his name instead of “the Big Guy” in messages and emails.
More than 1/5 (22.7%) of the 605 voters we interviewed last night in Michigan told us they are NOT "familiar with the revelations" in "recent news reports" surrounding Joe Biden.
FTR, the percentage who are familiar was closer to 65% than 80% w/ those who said they were “unsure”. It’s plausible to think they just don’t want to say “No” and that others who chose “Yes” don’t want to seem uninformed.
But in 2016, it was much higher for reports on Clinton.
If that’s the geo participation in the NYT Poll, then at least a significant part of Biden’s lead in them is in no small part nothing more than response bias and an artifact of the polling mode.
1. I'd be embarrassed to tweet maps that show such a distribution.
2. Quinnipiac is not a serious poll, hasn't called an election correctly in years, and wrong in Florida for 5 straight cycles, often grossly. So, please stop asking me what I think about that vs. our Trump +1.6.