@Thomas Carter: "Polls predicted a Hillary Clinton landslide, also."

No, they predicted a 2/3 chance of a Hillary victory.

A Trump win was not out of the range of error, believe it or not. We've exhausted this topic before but the careful polling analyses had a decent chance for Trump (mostly because of the correlation within swing state voting).

the Princeton Election Consortium rates Clinton’s “win probability” at 97 percent; Daily Kos Elections sets it at 95 percent; HuffPost at 93 percent; The Upshot at 91 percent; and FiveThirtyEight at 88 percent (that’s FiveThirtyEight’s “polls-only” forecast; the one that takes fundamentals and election history into account gives HRC only a 85 percent probability of winning.

The HuffPost presidential forecast gives Democratic presidential nominee Hillary Clinton a 98 percent chance of winning the general election on Tuesday. That means we’re pretty darn certain that ― barring some major catastrophe, scandal or nearly every single poll being wrong ― Clinton will be elected.

A High Probability Of Hillary Clinton Winning Doesn’t Mean It Will Be A Blowout
A survey from the Princeton Election Consortium has found that Hillary Clinton has a 99 per cent chance of winning the election over Donald Trump.
Three days before the election, Ms Clinton has a projected 312 electoral votes, compared to 226 for Mr Trump. A total of 270 electoral votes are needed to win.

The probability statistic was found by the university’s statistical Bayesian model.

Polls are not used to find out what people are thinking. The polling establishment has been taken over by lefties to TELL people what they SHOULD be thinking.

@Toolsmith: "The predictions I heard on election night had 95% for a dem win. No one predicted rep win. NO ONE."

I didn't say that *every* pundit interpreted the polls correctly. Just because you saw one cable news network claim it was 95% doesn't mean every poll was incorrect.

538 predicted 2-1 odds in favor for Hillary (which I plainly stated on this site before the election), and even the betting sites only had 5-1 odds (if you'll recall, I recommended that you bet on Trump because of this exact discrepancy).

So saying "everyone thought it was 95%" is flat out incorrect.

Also, saying that, "no one predicted a Trump win" and "all the polls are wrong" just lacks any kind of understanding of statistics.

There is this thing, called, "the margin of error". All polls have some error, since they are sampling a sub set of voters, and people change their minds as time moves forward.

In this case, the margin of error of all the polls was greater than the individual polls, because the swing states tend to be correlated.

Also, the Comey letter changed a lot of voters' minds, which means there was an active trend right up until the election (meaning older polls were incorrect).

If you only take one thing away from this, it's that you can't rely on cable news for interpretation of statistics, because everyone that works at cable news is really really stupid.

"But here’s a stubborn and surprising fact — and one to keep in mind as midterm polls really start rolling in: Over the past two years — meaning in the 2016 general election and then in the various gubernatorial elections and special elections that have taken place in 2017 and 2018 — the accuracy of polls has been pretty much average by historical standards."

"You read that right. Polls of the November 2016 presidential election were about as accurate as polls of presidential elections have been on average since 1972. And polls of gubernatorial and congressional elections in 2016 were about as accurate, on average, as polls of those races since 1998. "

Trump won - at least for now - the immigration ban he proposed. It still has to go back to the lower court. The SCOTUS ruled and the Liberal/Pukes are saying the SCOTUS was wrong. Even one of the squirrely Justices had judicial constipation borne of skin color and ranted and slobbered about the injustice of the Justices.

Our judicial system is not based on laws...it is based on opinion. With so many 5 to 4 votes it is clear that the laws are not clear. It is the opinion of judges and justices as to what laws mean...not what the laws say.

Watcher said - "Our judicial system is not based on laws...it is based on opinion. With so many 5 to 4 votes it is clear that the laws are not clear. . . "

Here rises the question - With so many 5 to 4 votes, what role does the personal agenda of these justices play here?

We have, in point of fact, one justice even making public statements on some current issues.

@Tom C: "Point of fact, generally combining polls decreases the MOE because you have a larger sample size.

The number of polls predicting Hillary in the 90 percent range are more indicative of systemic bias than mere statistical error."

But you can't just "combine polls". There's a methodology, and how you do that makes a huge difference in the outcomes.

If you are bad at statistics and just assume the polls in each state are independent variables, then you will highly underestimate the chance that all the swing states move together (which they are already known for doing).

If you look at the existing correlation, you know that the each of the swing states tends to vote the same way. So variability in Ohio + variability in Florida does not reduce in size, since you're actually looking at the same variability.

This is the same reason that Collateralized Debt Obligations became so toxic - they (the ratings agencies) assumed the chance that a home in California might default was independent of the chance that a home in Florida might default, so they assumed the overall risk was much lower than it actually was.

And, asking people how they will vote isn't nearly as complicated. How could all of the polls be independently incorrect? Do you really believe all they did was just miss a huge swing on the preference of the voters the last day or so? That's a cheap excuse.

And, separately, you can't say that if one poll gives you 95%.and a different independent poll also gives you 95% that the confidence in the result shouldn't be increased. That's the reasoning for excluding outliers.

But these polls did not behave that way,. They included in total, and likely independently, a very large bias.

Polls predicted a Hillary Clinton landslide, also.

Polls slant far left. Trump's prediction of a "red wave" may not be far off the mark.

@Thomas Carter: "Polls predicted a Hillary Clinton landslide, also."

No, they predicted a 2/3 chance of a Hillary victory.

A Trump win was not out of the range of error, believe it or not. We've exhausted this topic before but the careful polling analyses had a decent chance for Trump (mostly because of the correlation within swing state voting).

@Tom C: "Polls slant far left. Trump's prediction of a "red wave" may not be far off the mark."

This is not true, some slant left, some slant right, and the slant is measurable since you can compare the predictions against actual vote tallies.

The predictions I heard on election night had 95% for a dem win. No one predicted rep win. NO ONE.

Clinton has 90 percent chance of winning: Reuters/Ipsos States of the Nation

the Princeton Election Consortium rates Clinton’s “win probability” at97 percent; Daily Kos Elections sets it at95 percent; HuffPost at93 percent; The Upshot at91 percent;and FiveThirtyEight at88 percent(that’s FiveThirtyEight’s “polls-only” forecast; the one that takes fundamentals and election history into account gives HRC only a85 percentprobability of winning.Some People Are Now 100 Percent Sure Hillary Clinton Will Win

The HuffPost presidential forecast gives Democratic presidential nominee Hillary Clinton a98 percentchance of winning the general election on Tuesday. That means we’re pretty darn certain that ― barring some major catastrophe, scandal or nearly every single poll being wrong ― Clinton will be elected.A High Probability Of Hillary Clinton Winning Doesn’t Mean It Will Be A Blowout

.A survey from the Princeton Election Consortium has found that Hillary Clinton has a

99 per centchance of winning the election over Donald TrumpThree days before the election, Ms Clinton has a projected 312 electoral votes, compared to 226 for Mr Trump. A total of 270 electoral votes are needed to win.

The probability statistic was found by the university’s statistical Bayesian model.Survey finds Hillary Clinton has ‘more than

99% chance’of winning election over Donald TrumpConsistently leaning FAR left.Polls are not used to find out what people are thinking. The polling establishment has been taken over by lefties to TELL people what they SHOULD be thinking.

Thanks Tom. Thought I remembered that correctly.

@Toolsmith: "The predictions I heard on election night had 95% for a dem win. No one predicted rep win. NO ONE."

I didn't say that *every* pundit interpreted the polls correctly. Just because you saw one cable news network claim it was 95% doesn't mean every poll was incorrect.

538 predicted 2-1 odds in favor for Hillary (which I plainly stated on this site before the election), and even the betting sites only had 5-1 odds (if you'll recall, I recommended that you bet on Trump because of this exact discrepancy).

So saying "everyone thought it was 95%" is flat out incorrect.

Also, saying that, "no one predicted a Trump win" and "all the polls are wrong" just lacks any kind of understanding of statistics.

There is this thing, called, "the margin of error". All polls have some error, since they are sampling a sub set of voters, and people change their minds as time moves forward.

In this case, the margin of error of all the polls was greater than the individual polls, because the swing states tend to be correlated.

Also, the Comey letter changed a lot of voters' minds, which means there was an active trend right up until the election (meaning older polls were incorrect).

If you only take one thing away from this, it's that you can't rely on cable news for interpretation of statistics, because everyone that works at cable news is really really stupid.

Please read this:

https://fivethirtyeight.com/features/the-polls-are-all-right/

"But here’s a stubborn and surprising fact — and one to keep in mind as midterm polls really start rolling in: Over the past two years — meaning in the 2016 general election and then in the various gubernatorial elections and special elections that have taken place in 2017 and 2018 — the accuracy of polls has been pretty much average by historical standards.""You read that right. Polls of the November 2016 presidential election were about as accurate as polls of presidential elections have been on average since 1972. And polls of gubernatorial and congressional elections in 2016 were about as accurate, on average, as polls of those races since 1998. "@Tom C:

From your last link: "Other polls were less bullish on a Clinton victory. FiveThirtyEight said the likelihood Ms Clinton would win was 65 per cent"

Edit: the fact that they called 538 a "poll" is indicative of how stupid the article writer really is.

Point of fact, generally combining polls decreases the MOE because you have a larger sample size.

The number of polls predicting Hillary in the 90 percent range are more indicative of systemic bias than mere statistical error.

Trump won - at least for now - the immigration ban he proposed. It still has to go back to the lower court. The SCOTUS ruled and the Liberal/Pukes are saying the SCOTUS was wrong. Even one of the squirrely Justices had judicial constipation borne of skin color and ranted and slobbered about the injustice of the Justices.

Our judicial system is not based on laws...it is based on opinion. With so many 5 to 4 votes it is clear that the laws are not clear. It is the opinion of judges and justices as to what laws mean...not what the laws say.

“Even one of the squirrely Justices had judicial constipation borne of skin color”

I guess we know what you’re thinking because of your skin color, eh, watcher? Or is it just the dark folks who are easy to read?

You’re a turd.

you're a racist.

How do you figure?

I disagree with you, which makes you a racist. And a sexist and homophobe too.

Because you mock stupid old white men. Apparently, that's "reverse racism".

Let's add in ageist and microagressor. And oppressor. And mono-culturist, and anti-diversitarian.

I’m not mocking watcher. I’m pointing out his racist assumption, and calling him a piece of excrement because of it.

There’s a difference.

Udall is just stupid, apparently.

Well, that too.

Not to mention insulting to the handicapped.

Watcher said -

"Our judicial system is not based on laws...it is based on opinion. With so many 5 to 4 votes it is clear that the laws are not clear. . . "Here rises the question - With so many 5 to 4 votes, what role does the personal agenda of these justices play here?

We have, in point of fact, one justice even making public statements on some current issues.

Which should, it seems, be grounds for impeachment.

@Tom C:

"Point of fact, generally combining polls decreases the MOE because you have a larger sample size.The number of polls predicting Hillary in the 90 percent range are more indicative of systemic bias than mere statistical error."But you can't just "combine polls". There's a methodology, and how you do that makes a huge difference in the outcomes.

If you are bad at statistics and just assume the polls in each state are independent variables, then you will highly underestimate the chance that all the swing states move together (which they are already known for doing).

If you look at the existing correlation, you know that the each of the swing states tends to vote the same way. So variability in Ohio + variability in Florida does not reduce in size, since you're actually looking at the same variability.

This is the same reason that Collateralized Debt Obligations became so toxic - they (the ratings agencies) assumed the chance that a home in California might default was independent of the chance that a home in Florida might default, so they assumed the overall risk was much lower than it actually was.

The error in valuing CDOs included systemic bias.

And, asking people how they will vote isn't nearly as complicated. How could all of the polls be independently incorrect? Do you really believe all they did was just miss a huge swing on the preference of the voters the last day or so? That's a cheap excuse.

And, separately, you can't say that if one poll gives you 95%.and a different independent poll also gives you 95% that the confidence in the result shouldn't be increased. That's the reasoning for excluding outliers.

But these polls did not behave that way,. They included in total, and likely independently, a very large bias.

Brunswick area "mob-sters" were carrying protest signs on their way to the downtown mall this morning.

"Community organizing" against selected targets is alive and well. Sometimes it takes more than 8 years to realize something has been going on.