Trump drops below 40% in an NBC poll

[Dashinka 24,110]

13 minutes ago, CanAm1980 said:

"Skewing' the sample can be achieved many ways but that doesn't make the results wrong. They are just numbers.

 

The most useful thing about polls us to measure shift's in public opinion which you can do by using the same methods over a period of time. One would expect a lot if volatility right after that debate.

 

The polls are not meaningless, just another tool. 

Polls are also used to shape public opinion which is something our MSM is apt to do.

[Voice of Reason 10,298]

2 hours ago, CanAm1980 said:

"Skewing' the sample can be achieved many ways but that doesn't make the results wrong. They are just numbers.

 

The most useful thing about polls us to measure shift's in public opinion which you can do by using the same methods over a period of time. One would expect a lot if volatility right after that debate.

 

The polls are not meaningless, just another tool. 

50 years ago, I would have agreed with you.  But today, polls are not used to indicate a shift in opinion; rather, they are used to try and convince us that opinion has shifted, and ours should as well.  It's not very effective for those with sharper minds, IMO.

[Burnt Reynolds 5,058]

6 hours ago, CanAm1980 said:

"Skewing' the sample can be achieved many ways but that doesn't make the results wrong. They are just numbers.

 

The most useful thing about polls us to measure shift's in public opinion which you can do by using the same methods over a period of time. One would expect a lot if volatility right after that debate.

 

The polls are not meaningless, just another tool. 

The understood point of polls is for them to reflect reality, especially when this sort of data is done for decision making. It's been proven from prior elections they don't reflect reality. From a data perspective, 800 people where there's well over 100 million registered voters is a garbage sampling for the entirety of the US (and that's not including the stratified sampling necessary to reflect realistic voting which isn't the NPV), so playing with what's clearly awful data doesn't make it more accurate (hence the MS Paint analogy). And like I said at the very start of the thread, those who really believe the pointless national popular vote is going to see Trump get under 40%, raise your hand.. and justify it. No one did, and that's because the whole point of this is to try and demoralize Trump supporters, it's not about reflecting reality. We know Trump has his own team of analysts, and their success from the last election. Guaranteed, while the Biden team is playing up the public one for PR purposes, unless they're certain they're going to lose, they have their own team too and are't relying on this pointless public polling.

Edited October 6, 2020 by Burnt Reynolds

[Crtcl Rice Theory 2,432]

2 hours ago, Burnt Reynolds said:

The understood point of polls is for them to reflect reality, especially when this sort of data is done for decision making. It's been proven from prior elections they don't reflect reality. From a data perspective, 800 people where there's well over 100 million registered voters is a garbage sampling for the entirety of the US (and that's not including the stratified sampling necessary to reflect realistic voting which isn't the NPV), so playing with what's clearly awful data doesn't make it more accurate (hence the MS Paint analogy). And like I said at the very start of the thread, those who really believe the pointless national popular vote is going to see Trump get under 40%, raise your hand.. and justify it. No one did, and that's because the whole point of this is to try and demoralize Trump supporters, it's not about reflecting reality. We know Trump has his own team of analysts, and their success from the last election. Guaranteed, while the Biden team is playing up the public one for PR purposes, unless they're certain they're going to lose, they have their own team too and are't relying on this pointless public polling.

That's not the way statistics work.

 

 

 

[Burnt Reynolds 5,058]

2 minutes ago, CanAm1980 said:

That's not the way statistics work.

That's not the way rebuttals work.

[Crtcl Rice Theory 2,432]

2 hours ago, Burnt Reynolds said:

That's not the way rebuttals work.

A poll is a snapshot and not a forecast. 1004 is the optimal accepted sample size. Increasing the sample size will not necessarily increase the precision or confidence in your snapshot. Taking more Polls and insuring the sample size 

is random may increase the accuracy, but only for that point in time, which leads to the stability of the public's views.

 

Joe Biden has had a  consistent lead  when polled against Trump for three years, with the popular vote. When you get to 7% lead nationally, there is very little margin to overcome the popular vote nation wide in the swingstates to pull an inside straight with electoral votes.

 

Stagecraft and boat parades will not overcome how steady the Biden lead is.  The President will not be reelected and if the NBC and Fox Polls holdup, you could see Ohio, Ga and Texas fall in Biden's column.

 

The die is cast, there are very few undecided voters to sway.

 

 

 

 

 

[laylalex 7,282]

I have to say, I do find it interesting though how little coverage there has been regarding the IBD/TIPP poll showing little space between Biden and Trump nationally. I must search for it two or three times a day, and come up with little outside the usual suspects covering it. IBD gets an A/B rating at 538, so I don't know why it isn't cooked into the metrics.

[Burnt Reynolds 5,058]

5 minutes ago, CanAm1980 said:

A poll is a snapshot and not a forecast. 1004 is the optimal accepted sample size. Increasing the sample size will not necessarily increase the precision or confidence in your snapshot. Taking more Polls and insuring the sample size 

is random may increase the accuracy, but only for that point in time, which leads to the stability of the public's views.

 

Joe Biden has had a  consistent lead  when polled against Trump for three years, with the popular vote. When you get to 7% lead nationally, there is very little margin to overcome the popular vote nation wide in the swingstates to pull an inside straight with electoral votes.

 

Stagecraft and boat parades will not overcome how steady the Biden lead is.  The President will not be reelected and if the NBC and Fox Polls holdup, you could see Ohio, Ga and Texas fall in Biden's column.

 

The die is cast, there are very few undecided voters to sway.

A poll, for say, presidential approval rating, is a snapshot because it represents something specific for that time. Election polling is an intent poll about how someone is going to vote in a future date. Might be a good idea to understand the difference.

 

Also, I applaud your certainty here in the bold, pretending like national polls mirrors state votes in light of how it didn't just this last presidential election. I definitely want it to remain come November, so, don't let this hinder you:

 

Clinton Holds 11-Point National Lead Over Trump: NBC/WSJ Poll

 

Quote

The NBC/WSJ poll was conducted Oct. 10-13 of 1,000 registered voters – via both cell phones and landline phones – and it has an overall margin of error of plus-minus 3.1 percentage points. Among the poll’s 905 likely voters, the margin of error is plus-minus 3.3 percentage points

https://www.nbcnews.com/politics/first-read/clinton-holds-11-point-national-lead-over-trump-nbc-wsj-n666986

 

11 point lead, margin of error +/- 3.3 (7.7 - 14.3).. actual result, 2.1. 

 

 

[Crtcl Rice Theory 2,432]

17 minutes ago, Burnt Reynolds said:

A poll, for say, presidential approval rating, is a snapshot because it represents something specific for that time. Election polling is an intent poll about how someone is going to vote in a future date. Might be a good idea to understand the difference.

 

Also, I applaud your certainty here in the bold, pretending like national polls mirrors state votes in light of how it didn't just this last presidential election. I definitely want it to remain come November, so, don't let this hinder you:

 

Clinton Holds 11-Point National Lead Over Trump: NBC/WSJ Poll

 

https://www.nbcnews.com/politics/first-read/clinton-holds-11-point-national-lead-over-trump-nbc-wsj-n666986

 

11 point lead, margin of error +/- 3.3 (7.7 - 14.3).. actual result, 2.1. 

 

 

So you learned nothing.

 

What was the average polling gap between Clinton and Trump the week before the election? How did that vary from the actual popular vote?  Nice Cherry picking tho.

[Burnt Reynolds 5,058]

10 minutes ago, CanAm1980 said:

So you learned nothing.

 

What was the average polling gap between Clinton and Trump the week before the election? How did that vary from the actual popular vote?  Nice Cherry picking tho.

In case it wasn't obvious with the evidence presented, I was saying it's you who's learned nothing, and clearly are learning nothing again. It's why you're citing polls in the same time frame early October (with the same pollsters who were so far outside the margin of error last time it was comical) but telling me about polls on election day (which this isn't) while simultaneously making conclusions about polls in early October. 

 

Keep trying to jam that square into the circle hole. 

[Crtcl Rice Theory 2,432]

Just now, Burnt Reynolds said:

In case it wasn't obvious with the evidence presented, I was saying it's you who's learned nothing, and clearly are learning nothing again. It's why you're citing polls in the same time frame early October (with the same pollsters who were so far outside the margin of error last time it was comical) but telling me about polls on election day (which this isn't) while simultaneously making conclusions about polls in early October. 

 

Keep trying to jam that square into the circle hole. 

Oh... there will be jamin'

Did you look up the election week polls or are you just going to let that data point slide?

 

[Burnt Reynolds 5,058]

Just now, CanAm1980 said:

Oh... there will be jamin'

Did you look up the election week polls or are you just going to let that data point slide?

 

"let me try a triangle instead" 

 

Clinton +5 with a +/- of 2.7 (2.3 - 7.7), actual result 2.1. Once again, outside the margin of error. And once again, incorrectly decided the election.

https://www.google.com/amp/s/www.nbcnews.com/news/amp/ncna678611

 

Like I said.. doesn't learn. 

[Crtcl Rice Theory 2,432]

10 minutes ago, Burnt Reynolds said:

"let me try a triangle instead" 

 

Clinton +5 with a +/- of 2.7 (2.3 - 7.7), actual result 2.1. Once again, outside the margin of error. And once again, incorrectly decided the election.

https://www.google.com/amp/s/www.nbcnews.com/news/amp/ncna678611

 

Like I said.. doesn't learn. 

1 This is not Clinton v Trump redux.

The margins are different

The stability of polls are different 

 

But go ahead and bet the mortgage money on Trump, the las Vegas odds are good. Maybe they will give you points.

 

[Voice of Reason 10,298]

Just now, CanAm1980 said:

1 This is not Clinton v Trump redux.

The margins are different

The stability of polls are different 

 

But go ahead and bet the mortgage money on Trump, the las Vegas odds are good. Maybe they will give you points.

 

You can argue every day until your fingers get tired, and it won't matter.  Only the election results will.  Let's discuss this after Trump gets re-elected, eh?

[Crtcl Rice Theory 2,432]

4 minutes ago, Voice of Reason said:

You can argue every day until your fingers get tired, and it won't matter.  Only the election results will.  Let's discuss this after Trump gets re-elected, eh?

That's no fun , don't you want to gloat once Trump wins?