On the idea that the polls got the election wrong

[JohnStOnge]

Finally got around to looking directly at what polls were saying immediately before the election and, as I expected, the perception that the polls "got it wrong" is false. Any reasonable interpretation of the body of polling would have led to the conclusion that the election was too close to call.

See the RealClearPolitics poll summary from just before the election at http://www.realclearpolitics.com/epolls ... e_map.html . You can see by clicking on "Virginia" in the table that the last poll it includes was taken November 1 - 6. The RealClearPolitics people projected Clinton getting at least 203 electoral votes and Trump getting at least 164 electoral votes. But 171 electoral votes were up for grabs. Toss up.

The site had 18 jurisdictions indicated as being "solid" for one candidate. The candidate indicated as favored won all 18 of those jurisdictions.

It had 15 jurisdictions indicated as "likely" for one candidate. The candidate indicated as favored won all 15 of those jurisdictions.

It had 6 jurisdictions indicating as "leans" to one candidate. The candidate indicated as the one the State was leaning to won 5 of those jurisdictions.

When you see that margin of error thing with a poll, that's a 95 percent confidence interval. You expect the truth to be outside of the 95 percent confidence interval about 1 in 20 times over a large number of polls. The polls in aggregate were only "wrong" with respect to 1 jurisdiction out of 38. And, really, the 95 percent confidence thing probably only applies to the "solid" category. "Leans" is not a confident assertion.

Meanwhile, you can go to the RealClearPolitics average of polls for the popular vote just before the election at http://www.realclearpolitics.com/epolls ... -5952.html. It has Clinton winning the popular vote by 3.3 percentage points and she won it by 2.1 percentage points. The difference is well within what one should expect it to be.

Basically, if you were trying to anticipate what would happen based on polling results available on the night of November 7 you would say Clinton would win the popular vote but the electoral vote was too close to call.

And it WAS very close. Trump won Wisconsin, Pennsylvania, and Michigan by 0.8, 0.3, and 1.2 percentage points. It's not reasonable to expect polls to be able to pick the winner in States that close.

[AZGrizFan]

Except the electoral vote wasn't too close to call.



that is an electoral ass-kicking of epic, historic proportions.

Compare that to the results just 4 years earlier and you get a little bit of an idea of how BADLY America didn't want Hildabeast to become president.

[BDKJMU]

Finally got around to looking directly at what polls were saying immediately before the election and, as I expected, the perception that the polls "got it wrong" is false. Any reasonable interpretation of the body of polling would have led to the conclusion that the election was too close to call.

See the RealClearPolitics poll summary from just before the election at http://www.realclearpolitics.com/epolls ... e_map.html . You can see by clicking on "Virginia" in the table that the last poll it includes was taken November 1 - 6. The RealClearPolitics people projected Clinton getting at least 203 electoral votes and Trump getting at least 164 electoral votes. But 171 electoral votes were up for grabs. Toss up.

The site had 18 jurisdictions indicated as being "solid" for one candidate. The candidate indicated as favored won all 18 of those jurisdictions.

It had 15 jurisdictions indicated as "likely" for one candidate. The candidate indicated as favored won all 15 of those jurisdictions.

It had 6 jurisdictions indicating as "leans" to one candidate. The candidate indicated as the one the State was leaning to won 5 of those jurisdictions.

When you see that margin of error thing with a poll, that's a 95 percent confidence interval. You expect the truth to be outside of the 95 percent confidence interval about 1 in 20 times over a large number of polls. The polls in aggregate were only "wrong" with respect to 1 jurisdiction out of 38. And, really, the 95 percent confidence thing probably only applies to the "solid" category. "Leans" is not a confident assertion.

Meanwhile, you can go to the RealClearPolitics average of polls for the popular vote just before the election at http://www.realclearpolitics.com/epolls ... -5952.html. It has Clinton winning the popular vote by 3.3 percentage points and she won it by 2.1 percentage points. The difference is well within what one should expect it to be.

Basically, if you were trying to anticipate what would happen based on polling results available on the night of November 7 you would say Clinton would win the popular vote but the electoral vote was too close to call.

And it WAS very close. Trump won Wisconsin, Pennsylvania, and Michigan by 0.8, 0.3, and 1.2 percentage points. It's not reasonable to expect polls to be able to pick the winner in States that close.

When it comes to a lot of the swing state polls, which were the ones that really mattered, the perception that the polls got it wrong is true.

Looking at the 4 way polls in the 11 swing states:
-8 of the 11 underpolled Trump.
-There was only 1 state that had Trump ahead that Clinton won (NV). There was 3 states that had Clinton ahead that Trump won (PA, MI, WI)
-Of the 4 swing states that were off the most (WI, IA, OH, MI), and 6 of the top 7, all underpolled Trump

Underpolling Clinton
Clinton won Virginia 5.4/%. The RCP avg had Clinton up 5.0%, which = .4% off.
Clinton won Colorado by 4.9%. The RCP avg had Clinton up 2.9%, which = 2.0% off.
Clinton won Nevada by 2.4%. The RCP avg had Trump up .8%, which = 3.2% off.

Underpolling Trump
Clinton won New Hampshire by .6%. The RCP avg had Clinton up .3%, which = .3% off
Trump won Florida by 1.2%. The RCP avg had Trump up .2%, which = 1% off
Trump won Pennsylvania by .7%. The RCP avg had Clinton up 1.9%, which = 2.6% off
Trump won North Carolina by 3.7%. The RCP avg had Trump up 1.0%, which = 2.7% off
Trump won Michigan by .3%, yet the RCP avg had Clinton up by 3.4%, which = 3.7% off
Trump won Ohio by 8.1%. The RCP avg had Trump up 3.5%, which = 4.6% off
Trump won Iowa by 9.5%. The RCP avg had Trump up 3.0%, which = 6.5% off
Trump won Wisconsin by .7%. The RCP avg had Trump up 6.5%, which = 7.2% off

The swing state polls, by and large part, got it wrong.

[BDKJMU]

The day before the election:

[Skjellyfetti]

Underpolling Clinton
Clinton won Virginia 5.4/%. The RCP avg had Clinton up 5.0%, which = .4% off.
Clinton won Colorado by 4.9%. The RCP avg had Clinton up 2.9%, which = 2.0% off.
Clinton won Nevada by 2.4%. The RCP avg had Trump up .8%, which = 3.2% off.

Underpolling Trump
Clinton won New Hampshire by .6%. The RCP avg had Clinton up .3%, which = .3% off
Trump won Florida by 1.2%. The RCP avg had Trump up .2%, which = 1% off
Trump won Pennsylvania by .7%. The RCP avg had Clinton up 1.9%, which = 2.6% off
Trump won North Carolina by 3.7%. The RCP avg had Trump up 1.0%, which = 2.7% off
Trump won Michigan by .3%, yet the RCP avg had Clinton up by 3.4%, which = 3.7% off
Trump won Ohio by 8.1%. The RCP avg had Trump up 3.5%, which = 4.6% off
Trump won Iowa by 9.5%. The RCP avg had Trump up 3.0%, which = 6.5% off
Trump won Wisconsin by .7%. The RCP avg had Trump up 6.5%, which = 7.2% off

The swing state polls, by and large part, got it wrong.

Maybe 3 or 4 of those were outside the margin of error?

National polls were within 1% of accuracy.

The polls were wrong, but I'm not so sure their methodology was flawed. There was a huge surge for Trump in the final week and it surprised even their own campaign staff.

[93henfan]

You all keep debating this, but the bottom line is that even in "anonymous" polling, the person's husband, wife, boyfriend, girfriend, etc may still be in the same room when they're talking on the phone to the pollster, and as the protests on Friday and Saturday demonstrated, there are millions of whiny bitches in this country who would not let that person live it down if they said they supported Trump.

Millions of Trump voters had to bite their tongues while they were voting their conscience.

/story

[Skjellyfetti]

Millions of Trump voters had to bite their tongues while they were voting their conscience.

/story

National polls were off by about 1%

That's about 57,600 people.

[93henfan]

Millions of Trump voters had to bite their tongues while they were voting their conscience.

/story

National polls were off by about 1%

That's about 57,600 people.

My bad. I didn't realize only 5.7 million people voted.

[JohnStOnge]

When it comes to a lot of the swing state polls, which were the ones that really mattered, the perception that the polls got it wrong is true.

Looking at the 4 way polls in the 11 swing states:
-8 of the 11 underpolled Trump.
-There was only 1 state that had Trump ahead that Clinton won (NV). There was 3 states that had Clinton ahead that Trump won (PA, MI, WI)
-Of the 4 swing states that were off the most (WI, IA, OH, MI), and 6 of the top 7, all underpolled Trump

Underpolling Clinton
Clinton won Virginia 5.4/%. The RCP avg had Clinton up 5.0%, which = .4% off.
Clinton won Colorado by 4.9%. The RCP avg had Clinton up 2.9%, which = 2.0% off.
Clinton won Nevada by 2.4%. The RCP avg had Trump up .8%, which = 3.2% off.

Underpolling Trump
Clinton won New Hampshire by .6%. The RCP avg had Clinton up .3%, which = .3% off
Trump won Florida by 1.2%. The RCP avg had Trump up .2%, which = 1% off
Trump won Pennsylvania by .7%. The RCP avg had Clinton up 1.9%, which = 2.6% off
Trump won North Carolina by 3.7%. The RCP avg had Trump up 1.0%, which = 2.7% off
Trump won Michigan by .3%, yet the RCP avg had Clinton up by 3.4%, which = 3.7% off
Trump won Ohio by 8.1%. The RCP avg had Trump up 3.5%, which = 4.6% off
Trump won Iowa by 9.5%. The RCP avg had Trump up 3.0%, which = 6.5% off
Trump won Wisconsin by .7%. The RCP avg had Trump up 6.5%, which = 7.2% off

The swing state polls, by and large part, got it wrong.

No, they didn't. All of those States except Wisconsin were in the"Too Close to Call" category. And as far as the distribution whereby there are 11 States and the polls miss one way three times and the other way eight times goes:

If you were to assume that there's a 50:50 shot in each case that a poll could overshoot or undershoot in each case and you do 11 polls, the "margin of error" is that it you could end up with somewhere between being off 10 times one way to being off 10 times the other way. That's if the "margin of error" is a 95 percent confidence interval as it normally is. Being off 3 times one way and 8 times the other way is well within the realm of what could easily happen by random chance. It doesn't demonstrate, at all, that the polls were off.

[JohnStOnge]

Except the electoral vote wasn't too close to call.



that is an electoral ass-kicking of epic, historic proportions.

Compare that to the results just 4 years earlier and you get a little bit of an idea of how BADLY America didn't want Hildabeast to become president.

I'm kind of surprised you posted that because Trump's statement that he won by an electoral landslide was debunked early on. In terms of the electoral college margin Trump's victory ranked 46th out of 58 US Presidential elections.

[93henfan]

Just a couple of my favorite videos. I go back and watch these periodically just for stress relief:

phpBB [video]

phpBB [video]

[houndawg]

Except the electoral vote wasn't too close to call.



that is an electoral ass-kicking of epic, historic proportions.

Compare that to the results just 4 years earlier and you get a little bit of an idea of how BADLY America didn't want Hildabeast to become president.

I'm kind of surprised you posted that because Trump's statement that he won by an electoral landslide was debunked early on. In terms of the electoral college margin Trump's victory ranked 46th out of 58 US Presidential elections.

Jesus, Z, when you're getting your lunch eaten by JSO you probably should get some rest and come back in a few days...

[JohnStOnge]

Compare that to the results just 4 years earlier and you get a little bit of an idea of how BADLY America didn't want Hildabeast to become president.

And BTW Trump got a lower percentage of the overall vote than Romney did 4 years ago. I think the numbers should be pretty stable by now and right now the numbers at CNN are showing Trump getting 46.1% of the overall vote in 2016 vs. Romney getting 47% in 2012. He got a lower percentage of the overall vote than John Kerry did in 2004.

Just on the basis of the raw numbers there were clearly more people who did not want Trump to become President than there were who didn't want Hillary to become President. And I also think there was a problem with the pundits acting like they did. The polls did NOT support confidence in Hillary winning. I suspect that a lot of people thought Hillary was going to win. I think if everybody had gone into that day understanding that it was very close it very well could have made a difference. I think a lot of people either didn't vote or voted third Party because they thought Clinton was going to win regardless.

No way most people want Trump. That's pretty darned clear.

[JohnStOnge]

Jesus, Z, when you're getting your lunch eaten by JSO you probably should get some rest and come back in a few days...

I think you know very well that people get their lunch eaten by me all the time on this board. They just don't realize it.

But I DID forget to post a link supporting what I said in this case. Here is one:

https://www.nytimes.com/interactive/201 ... .html?_r=0

And it's not just that it was at the lower end of the spectrum as electoral college wins go. It was extremely close in key States. Give Trump's team credit, I guess, for getting the narrow wins in a number of key States. But they were very narrow wins. And in terms of the American People overall: More of them wanted Hillary Clinton than wanted Donald Trump in spite of Clinton's well known issues.

[JohnStOnge]

Just a couple of my favorite videos. I go back and watch these periodically just for stress relief:

phpBB [video]

phpBB [video]

He did get low support among Latinos and you too should laugh at the idea of him being elected President of the United States because the only alternative is crying. What just happened is absolutely ridiculous and it made this country the laughing stock of the planet. If somebody EVER wanted to make the argument that Americans are idiots this is Exhibit A.

[houndawg]

Compare that to the results just 4 years earlier and you get a little bit of an idea of how BADLY America didn't want Hildabeast to become president.

And BTW Trump got a lower percentage of the overall vote than Romney did 4 years ago. I think the numbers should be pretty stable by now and right now the numbers at CNN are showing Trump getting 46.1% of the overall vote in 2016 vs. Romney getting 47% in 2012. He got a lower percentage of the overall vote than John Kerry did in 2004.

Just on the basis of the raw numbers there were clearly more people who did not want Trump to become President than there were who didn't want Hillary to become President. And I also think there was a problem with the pundits acting like they did. The polls did NOT support confidence in Hillary winning. I suspect that a lot of people thought Hillary was going to win. I think if everybody had gone into that day understanding that it was very close it very well could have made a difference. I think a lot of people either didn't vote or voted third Party because they thought Clinton was going to win regardless.

No way most people want Trump. That's pretty darned clear.

Stop the presses.

[Baldy]

Jesus, Z, when you're getting your lunch eaten by JSO you probably should get some rest and come back in a few days...

I think you know very well that people get their lunch eaten by me all the time on this board. They just don't realize it.

Oh my GOD, JSO just made a funny.

You do eat a lot of shit sandwiches though.

[houndawg]

Jesus, Z, when you're getting your lunch eaten by JSO you probably should get some rest and come back in a few days...

I think you know very well that people get their lunch eaten by me all the time on this board. They just don't realize it.

But I DID forget to post a link supporting what I said in this case. Here is one:

https://www.nytimes.com/interactive/201 ... .html?_r=0

And it's not just that it was at the lower end of the spectrum as electoral college wins go. It was extremely close in key States. Give Trump's team credit, I guess, for getting the narrow wins in a number of key States. But they were very narrow wins. And in terms of the American People overall: More of them wanted Hillary Clinton than wanted Donald Trump in spite of Clinton's well known issues.

You need a new schtick, this one just isn't the same after your queef-fest during the election.

You can't just go back to what you were doing after you've were exposed like that. At least JBB had enough shame to disappear after he had his breakdown

[BDKJMU]

When it comes to a lot of the swing state polls, which were the ones that really mattered, the perception that the polls got it wrong is true.

Looking at the 4 way polls in the 11 swing states:
-8 of the 11 underpolled Trump.
-There was only 1 state that had Trump ahead that Clinton won (NV). There was 3 states that had Clinton ahead that Trump won (PA, MI, WI)
-Of the 4 swing states that were off the most (WI, IA, OH, MI), and 6 of the top 7, all underpolled Trump

Underpolling Clinton
Clinton won Virginia 5.4/%. The RCP avg had Clinton up 5.0%, which = .4% off.
Clinton won Colorado by 4.9%. The RCP avg had Clinton up 2.9%, which = 2.0% off.
Clinton won Nevada by 2.4%. The RCP avg had Trump up .8%, which = 3.2% off.

Underpolling Trump
Clinton won New Hampshire by .6%. The RCP avg had Clinton up .3%, which = .3% off
Trump won Florida by 1.2%. The RCP avg had Trump up .2%, which = 1% off
Trump won Pennsylvania by .7%. The RCP avg had Clinton up 1.9%, which = 2.6% off
Trump won North Carolina by 3.7%. The RCP avg had Trump up 1.0%, which = 2.7% off
Trump won Michigan by .3%, yet the RCP avg had Clinton up by 3.4%, which = 3.7% off
Trump won Ohio by 8.1%. The RCP avg had Trump up 3.5%, which = 4.6% off
Trump won Iowa by 9.5%. The RCP avg had Trump up 3.0%, which = 6.5% off
Trump won Wisconsin by .7%. The RCP avg had Trump up 6.5%, which = 7.2% off

The swing state polls, by and large part, got it wrong.

No, they didn't. All of those States except Wisconsin were in the"Too Close to Call" category. And as far as the distribution whereby there are 11 States and the polls miss one way three times and the other way eight times goes:

If you were to assume that there's a 50:50 shot in each case that a poll could overshoot or undershoot in each case and you do 11 polls, the "margin of error" is that it you could end up with somewhere between being off 10 times one way to being off 10 times the other way. That's if the "margin of error" is a 95 percent confidence interval as it normally is. Being off 3 times one way and 8 times the other way is well within the realm of what could easily happen by random chance. It doesn't demonstrate, at all, that the polls were off.

That would be true if the 8 RCP swing state averages that were that off on Trump and the 3 RCP swing state averages that were off on Hillary were off on average the same amount. But the 8 RCP swing state averages that were that off on Trump were off by an avg of about 3.8%, twice what the 3 RCP swing state averages that were off on Hillary, an average of about 1.9%.

Also, according to
http://worldpopulationreview.com/states/
those 8 swing states that under polled on Trump have a total population of about 71 million, an avg of close to 9 million per state.

For the 3 swing states that under polled for Hillary have a total population of about 17 million, an avg of about 5.7 million per state.

[AZGrizFan]

I'm kind of surprised you posted that because Trump's statement that he won by an electoral landslide was debunked early on. In terms of the electoral college margin Trump's victory ranked 46th out of 58 US Presidential elections.

Jesus, Z, when you're getting your lunch eaten by JSO you probably should get some rest and come back in a few days...

like I give a shit enough to look that crap up.

[AZGrizFan]

Jesus, Z, when you're getting your lunch eaten by JSO you probably should get some rest and come back in a few days...

I think you know very well that people get their lunch eaten by me all the time on this board. They just don't realize it.

But I DID forget to post a link supporting what I said in this case. Here is one:

https://www.nytimes.com/interactive/201 ... .html?_r=0

And it's not just that it was at the lower end of the spectrum as electoral college wins go. It was extremely close in key States. Give Trump's team credit, I guess, for getting the narrow wins in a number of key States. But they were very narrow wins. And in terms of the American People overall: More of them wanted Hillary Clinton than wanted Donald Trump in spite of Clinton's well known issues.

Yeah, that's why they're called "key states'....

[kalm]

Jesus, Z, when you're getting your lunch eaten by JSO you probably should get some rest and come back in a few days...

like I give a shit enough to look that crap up.

[Grizalltheway]

Jesus, Z, when you're getting your lunch eaten by JSO you probably should get some rest and come back in a few days...

like I give a shit enough to look that crap up.

Or you could have just looked at the second graphic you posted which shows that Trump's margin of victory was significantly smaller than Obama's 4 years ago.

[AZGrizFan]

like I give a shit enough to look that crap up.

Or you could have just looked at the second graphic you posted which shows that Trump's margin of victory was significantly smaller than Obama's 4 years ago.

Alright, motherfuckers. Do you REALLY care what the margin of victory was for George Washington, John Adams and Thomas Jefferson?

In my lifetime:

Ronald Reagan (1984): 97.6%
Richard Nixon (1972): 96.7%
Ronald Reagan (1980): 90.9%
Lyndon Johnson (1964): 90.3%
George HW Bush (1988): 79.2%
Bill Clinton (1996): 70.4%
Bill Clinton (1992): 68.8%
Osama Bin Laden (2008): 67.8%
Osama Bin Laden (2012): 61.7%
Donald Trump (2016): 56.9%
John Kennedy (1960): 56.4%
Richard Nixon (1968): 55.9%
Jimmy Carter (1976): 55.2%
George W. Bush (2004): 53.2%
George W. Bush (2000): 50.4%

OK.








Turns out 2016 wasn't that epic.

[Grizalltheway]

Or you could have just looked at the second graphic you posted which shows that Trump's margin of victory was significantly smaller than Obama's 4 years ago.

Alright, motherfuckers. Do you REALLY care what the margin of victory was for George Washington, John Adams and Thomas Jefferson?

In my lifetime:

Ronald Reagan (1984): 97.6%
Richard Nixon (1972): 96.7%
Ronald Reagan (1980): 90.9%
Lyndon Johnson (1964): 90.3%
George HW Bush (1988): 79.2%
Bill Clinton (1996): 70.4%
Bill Clinton (1992): 68.8%
Osama Bin Laden (2008): 67.8%
Osama Bin Laden (2012): 61.7%
Donald Trump (2016): 56.9%
John Kennedy (1960): 56.4%
Richard Nixon (1968): 55.9%
Jimmy Carter (1976): 55.2%
George W. Bush (2004): 53.2%
George W. Bush (2000): 50.4%

OK.








Turns out 2016 wasn't that epic.

All's fair in love, war and trolling JSO.