People's Pundit Daily Presidential Tracking Poll (Trump 42 - Clinton 41)

People's Pundit Daily ^ | 8/25/2016 | People's Pundit Daily

[usafa92]

No article just the link to the poll.

[usafa92]

For those who have never heard of this firm, I believe that they are essentially LongRoom under another name. If you read their methodology, they weight polls by US Census data to state and county, which was what LongRoom claimed they did. Are they accurate? I don't know. I am including a link below from their website which claims they were the most accurate Internet based pollster in the 2014 mid terms. Also, Gravis seems to be affiliated with them in some manner.

https://www.peoplespunditdaily.com/news/politics/2014/11/06/ppds-big-debut-most-accurate-election-projection-model-of-2014-hands-down/

[usafa92]

https://www.peoplespunditdaily.com/news/politics/2014/11/06/ppds-big-debut-most-accurate-election-projection-model-of-2014-hands-down/

Working Link

[usafa92]

FYI

[pieceofthepuzzle]

There are multiple polls with these numbers, with the only outliers being Quinapiac, and Reuters (which is using a new methodology in which they ‘assign’ a vote to undecideds).

Consider this. In many social circles the most vocal people are leftists, and they have no problem attacking and even shouting down those who think differently. This makes admitting you are voting for Trump very awkward, and admitting voting for Hillary an admission that earns you brownie points. Thus, it is highly unlikely that there are many ‘hidden’ Hillary voters, and also very unlikely that those who have undeclared are going to break for Hillary. Further, many who are ‘afraid’ to admit they are voting for Trump say deflect the question and say ‘I can't believe the choice we have between these two’, and allude to potentially voting for a third party, or not voting. It would make sense that these people will wind up being included in the Libertarian or other candidate groups, or will be designated as not likely voters - when in reality they will be voting for Trump.

The point is that there are many minefields affecting the polls.

[xzins]

I looked and they seem to be separate

[Rokke]

Odd diversity in polling data recently. I believe that is primarily because there has been a rapid and accelerating change in how people respond (or don't) to pollsters. I read recently that only 10% of people actually respond to pollsters. That's down from 80% just a couple election cycles ago. So who are the 10% who actually respond? They are certainly not representative of a vast majority of the population as a whole. So is there opinion actually valid? The Brexit results would indicate they are not. Jury is still out on this election obviously. But I think the "science" of polling is in the midst of a massive evolution and other than trending data, I think the published numbers are extremely questionable.

The LATimes Poll shows some very strange variations in some of its samples. The whole purpose of a rolling average is to dampen out some of the noise of daily polling. So the relatively wild swings in the daily published number is odd.

[Theodore R.]

Are the Gophers going to be able to vote for Trump. Is he on the MN ballot?

[TexasFreeper2009]

Trump has A LOT of upside even at these numbers. Once he topedoes the no name Johnson, Trump will be over 50.

[entropy12]

That is the same point I made in a previous post today.

Don’t tell me D/R/I breakdown because if only 10% people polled respond, the whole model is statistically useless.

[SamAdams76]

Donald Trump is pleased by this news.

[Salvation]

TTTT!

[Rokke]

If you haven’t read it, this is an interesting article about current problems in polling.
http://www.nytimes.com/2015/06/21/opinion/sunday/whats-the-matter-with-polling.html?_r=1

[BlackFemaleArmyColonel]

Nice!!!

[entropy12]

That article just reinforces my thoughts on modern day polling. The results can be off by as much as 10%.