Trump campaign demands CNN apologize for poll that shows Biden leading

[*bfine32*]

(06-13-2020, 12:14 PM)Goalpost Wrote: Agree that the polls got the popular vote right.  They were off on the 'key' state totals though.  For instance 538 created a category of which states most likely could tip the election.  In their order it went...Florida, Pa, Mich, N Car, Va, Colorado, Ohio, Wisc, Minn, and Nevada.  On the day of the election, Hillary led, going in, in 9 of them, according to 538.   She only won 4 of them.  

Florida...predicted Hillary up .7,  trump won by 2
PA...predicted Hillary up 3.7,  Trump won by 1.
Mich...Predicted Hillary up 4.2, ended in a virtual tie which Trump won
North Carolina...predicted Hillary up by .7, Trump won by 4
Va..predicted Hillary up by 5.6, Hillary won by 5.
Colorado..predicted Hillary up by 4, Hillary won by 3.
Ohio...Predicted Trump by 1.9, Trump won by 9.
Wisconsin..predicted Hillary up by 5.3, Trump won by 1.
Minnesota..predicted Hillary up by 5.8, Hillary won by 1.
Nevada...predicted Hillary up by 1.2, Hillary won by 2.

538 was right.  These key states tipped the election.  But it seems they were maybe, what, about 4 or 5 points in range of error in average of these.  

Thanks, as I said The Polls predicted Hillary would be President

[*Nately120*]

(06-13-2020, 01:44 PM)bfine32 Wrote: So the polls are useless?

Maybe.  Either Trump won because of some hard to predict freak accident or the neo-cons have successfully finagled things so a minority of voters can beat the majority.

I'm no history buff, but are there polls where Romney or Dukakis or Bob Dole were expected to win by a healthy margin? I guess what I'm trying to figure out is if the polls are useless now because:

A. the world and polling methods have changed
B. the power of Trump throws stuff off
C. the polls were just flat-out wrong and/or conducted in a flawed manner


Are polls worthless now? If Trump were up big in the polls right now would Trump supporters still be saying "Oh, that means nothing!"

Personally, I'm of the mind Biden being up big means he MIGHT win. That's how slanted the playfield is at the moment.

[Belsnickel]

(06-13-2020, 12:14 PM)Goalpost Wrote: Agree that the polls got the popular vote right.  They were off on the 'key' state totals though.  For instance 538 created a category of which states most likely could tip the election.  In their order it went...Florida, Pa, Mich, N Car, Va, Colorado, Ohio, Wisc, Minn, and Nevada.  On the day of the election, Hillary led, going in, in 9 of them, according to 538.   She only won 4 of them.  

Florida...predicted Hillary up .7,  trump won by 2
PA...predicted Hillary up 3.7,  Trump won by 1.
Mich...Predicted Hillary up 4.2, ended in a virtual tie which Trump won
North Carolina...predicted Hillary up by .7, Trump won by 4
Va..predicted Hillary up by 5.6, Hillary won by 5.
Colorado..predicted Hillary up by 4, Hillary won by 3.
Ohio...Predicted Trump by 1.9, Trump won by 9.
Wisconsin..predicted Hillary up by 5.3, Trump won by 1.
Minnesota..predicted Hillary up by 5.8, Hillary won by 1.
Nevada...predicted Hillary up by 1.2, Hillary won by 2.

538 was right.  These key states tipped the election.  But it seems they were maybe, what, about 4 or 5 points in range of error in average of these.  

And with most of those, the differences was within the margin of error. Statistics aren't flawless because they are, by design, taking a sample and trying to extrapolate that data to make it fit the population. There is always a margin of error with these things to take that into account. For instance, let's take PA since that switched from what the poll had. The second to last poll (the document for the last one isn't there anymore) has Clinton over Trump 48/44. The MoE for that poll is 3% which means that the 48 for Clinton could actually be 45 or 51 and the 44 for Trump could actually be 41 or 47. Now, Trump out performed that margin with 48, but Clinton's number was right there at 47. When there is a specific number with a MoE given it means the answer is in that range. So for a lot of these the polls predicted the numbers correctly.

[Belsnickel]

(06-13-2020, 01:44 PM)bfine32 Wrote: So the polls are useless?

Nope.

[*bfine32*]

(06-13-2020, 01:47 PM)Nately120 Wrote: Maybe.  Either Trump won because of some hard to predict freak accident or the neo-cons have successfully finagled things so a minority of voters can beat the majority.

I'm no history buff, but are there polls where Romney or Dukakis or Bob Dole were expected to win by a healthy margin?  I guess what I'm trying to figure out is if the polls are useless now because:

A. the world and polling methods have changed
B. the power of Trump throws stuff off
C. the polls were just flat-out wrong and/or conducted in a flawed manner


Are polls worthless now?  If Trump were up big in the polls right now would Trump supporters still be saying "Oh, that means nothing!"

Personally, I'm of the mind Biden being up big means he MIGHT win.  That's how slanted the playfield is at the moment.

Simply saying all the 2016 polls predicted President Hillary. Polls are fine, but they're just that. 

FWIW No one earned the majority in 2016. But I'll bet you the vast majority of the polls had someone winning the majority. 

[Belsnickel]

(06-13-2020, 06:10 PM)bfine32 Wrote: Simply saying all the 2016 polls predicted President Hillary. Polls are fine, but they're just that. 

FWIW No one earned the majority in 2016. But I'll bet you the vast majority of the polls had someone winning the majority. 

Actually, no. The vast majority had no one winning the majority. The majority may have been within the MoE, but the numbers they reported were consistently below it.

[*bfine32*]

(06-13-2020, 06:20 PM)Belsnickel Wrote: Actually, no. The vast majority had no one winning the majority. The majority may have been within the MoE, but the numbers they reported were consistently below it.

So do we consider the MoE or not? 

[Belsnickel]

(06-13-2020, 08:44 PM)bfine32 Wrote: So do we consider the MoE or not? 

You consider the MoE. The majority acknowledged that winning a majority of the popular vote was a possibility as it fell within their confidence interval, but very few saw it as the likely scenario, meaning they weren't predicting it.

[*bfine32*]

(06-13-2020, 08:55 PM)Belsnickel Wrote: You consider the MoE. The majority acknowledged that winning a majority of the popular vote was a possibility as it fell within their confidence interval, but very few saw it as the likely scenario, meaning they weren't predicting it.

You're just kinda arguing in circles now with the new "predicting" qualifier; as this is what I said:

"I'll bet you the vast majority of the polls had someone winning the majority." You consider MoE and that's exactly what we had. 

So can we say this:

"The vast majority of the polls had Hillary winning the White House"?

To be honest I don't understand you objections to the simple facts.

It's not a partisan thing: Polls are flawed and they were in 2016. Hell, Biden might win in a landslide, but I'll be damned if I put my money on a poll. 

[hollodero]

(06-13-2020, 09:16 PM)bfine32 Wrote: It's not a partisan thing: Polls are flawed and they were in 2016. Hell, Biden might win in a landslide, but I'll be damned if I put my money on a poll. 

lol... I mean, that's kinda true. There are some false narratives around regarding how far off polls were in 2016, but that shouldn't take away from the fact that polls are quite off in many instances. This is true in quite some states in 2016, and it is true around the world.

[Dill]

(06-13-2020, 06:00 PM)Belsnickel Wrote: And with most of those, the differences was within the margin of error. Statistics aren't flawless because they are, by design, taking a sample and trying to extrapolate that data to make it fit the population. There is always a margin of error with these things to take that into account. For instance, let's take PA since that switched from what the poll had. The second to last poll (the document for the last one isn't there anymore) has Clinton over Trump 48/44. The MoE for that poll is 3% which means that the 48 for Clinton could actually be 45 or 51 and the 44 for Trump could actually be 41 or 47. Now, Trump out performed that margin with 48, but Clinton's number was right there at 47. When there is a specific number with a MoE given it means the answer is in that range. So for a lot of these the polls predicted the numbers correctly.

You know he did not follow this, right?

[Dill]

(06-13-2020, 06:20 PM)Belsnickel Wrote: Actually, no. The vast majority had no one winning the majority. The majority may have been within the MoE, but the numbers they reported were consistently below it.

Who did most of the polls have winning he popular vote?

[GMDino]

Polls give a percentage with a margin of error.

The vast majority are within that margin of error.

The few that were not were close but in states that swung the EC and Trump had a surprise win.

And people who don't defend Trump and say they didn't vote for Trump keep trashing polls they don't understand to support the idea that he's going to win again.

Pretty much society in general right now except most people openly admit the support Trump instead of hiding behind the facade that they are just trying to be "non political".

[*Nately120*]

(06-13-2020, 09:16 PM)bfine32 Wrote: You're just kinda arguing in circles now with the new "predicting" qualifier; as this is what I said:

"I'll bet you the vast majority of the polls had someone winning the majority." You consider MoE and that's exactly what we had. 

So can we say this:

"The vast majority of the polls had Hillary winning the White House"?

To be honest I don't understand you objections to the simple facts.

It's not a partisan thing: Polls are flawed and they were in 2016. Hell, Biden might win in a landslide, but I'll be damned if I put my money on a poll. 

Now that you mention it Biden may have a much better chance than even the polls think since Biden could get votes from people who don't want to admit they are voting for a pervy asshole.

[*bfine32*]

(06-14-2020, 02:00 AM)Dill Wrote: Who did most of the polls have winning he popular vote?

Who did the polls have wearing the most expensive shoes? 

[BrownAssClown]

Polls are for suckers. Didn't we learn anything from 2016. Nobody knows how the vote is going to go, Biden could win or Trump could win re-election. If anybody thinks it's a slam dunk for any of these old farts, they're sadly mistaken and IMO full of beans. I'll wait for the votes to be counted and I hope it's a fair(LOL) election and those folks that are eligible to vote are allowed to vote and not impeded from doing so.

Now, it's not so much that I want to see Biden win as I want to see Trump lose. I could write 30 pages as to why, but the meltdown that would ensue  from the president would be YUGE. Remember Trump could still face charges from the state of New York because the statue of limitations would still be in effect in late January 2021. If Trump wins re-electon I believe the charges would expire because you can't charge a sitting president. If Trump loses he will be back to a regular citizen and could face charges in NY. It would be the best ending to a reality show in the history of television.

141 days...11 hours....39minutes.