Gallup Trump & Clinton favorability numbers megathread

[Erich Maria Remarque]

I'm basically saying that Gallup is the best leading indicator of all the polls that follow (national and state)

Evidence?

[dspNY]

Look at the Gallup favorables graph and transpose it onto either the RCP average or Pollster and you'll see the leading indicator from Gallup

[Erich Maria Remarque]

Look at the Gallup favorables graph and transpose it onto either the RCP average or Pollster and you'll see the leading indicator from Gallup

Lol, of course, they are correlated, but you said it is the best indicator. Look at LA Times and 538/RCP average, and you'll see it is a good indicator as well. All the polls are correlated — surprice, surprice! But all the polls have noice as well.

[‼realJohnEwards‼]

Look at the Gallup favorables graph and transpose it onto either the RCP average or Pollster and you'll see the leading indicator from Gallup

Lol, of course, they are correlated, but you said it is the best indicator. Look at LA Times and 538/RCP average, and you'll see it is a good indicator as well. All the polls are correlated — surprice, surprice! But all the polls have noice as well.

So a trend over 2/3 days which point towards Clinton? While Reuters (junky as it is on its own) corroborates? That's noise? ok then whatever you say

[Hammy]

Look at the Gallup favorables graph and transpose it onto either the RCP average or Pollster and you'll see the leading indicator from Gallup

Lol, of course, they are correlated, but you said it is the best indicator. Look at LA Times and 538/RCP average, and you'll see it is a good indicator as well. All the polls are correlated — surprice, surprice! But all the polls have noice as well.

LA Times pollsters admitted it had a Republican bias, and they use the same pre-selected sub-sample, which as others have pointed out, isn't necessarily demographically representative of the electorate.

[Maxwell]

today Clinton is at 40% favorability for the first time since August, while Trump, after a couple of days in 35 and up, is back down to 34% favorable.

[Erich Maria Remarque]

Look at the Gallup favorables graph and transpose it onto either the RCP average or Pollster and you'll see the leading indicator from Gallup

Lol, of course, they are correlated, but you said it is the best indicator. Look at LA Times and 538/RCP average, and you'll see it is a good indicator as well. All the polls are correlated — surprice, surprice! But all the polls have noice as well.

So a trend over 2/3 days which point towards Clinton? While Reuters (junky as it is on its own) corroborates? That's noise? ok then whatever you say

Not me, Nate
It might be noise.

[ProgressiveCanadian]

I'm basically saying that Gallup is the best leading indicator of all the polls that follow (national and state)

Well the did so well in 2012

[‼realJohnEwards‼]

Also, Gallup's sample sizes are 5 times that.

[Erich Maria Remarque]

Look at the Gallup favorables graph and transpose it onto either the RCP average or Pollster and you'll see the leading indicator from Gallup

Lol, of course, they are correlated, but you said it is the best indicator. Look at LA Times and 538/RCP average, and you'll see it is a good indicator as well. All the polls are correlated — surprice, surprice! But all the polls have noice as well.

LA Times pollsters admitted it had a Republican bias, and they use the same pre-selected sub-sample, which as others have pointed out, isn't necessarily demographically representative of the electorate.

I will say it slowly. Treeeeeends. We are talking about treeeeeends. Gallup treeeeeends vs LA Times treeeeends.
Just check who is doing better vs average on RCP/Upshot/538 etc.

[Maxwell]

Gallup was pretty bad in 2012 and I assume they're pretty bad this year.

BUT MUH DATA POINTS

[Erich Maria Remarque]

Also, Gallup's sample sizes are 5 times that.

Didn't know that. There is nothing about sample size on their page.

[Erich Maria Remarque]

BUT MUH DATA POINTS

Lol, I agree

[‼realJohnEwards‼]

Also, Gallup's sample sizes are 5 times that.

Didn't know that. There is nothing about sample size on their page.

Here you go:

You must be logged in to read this quote.

http://www.gallup.com/poll/101872/faqs.aspx#!mn-us

[Ebsy]

Looks like getting sick might have actually helped Clinton's favorables.

[KingSweden]

Looks like getting sick might have actually helped Clinton's favorables.

Counterintuitively, but so far the limited evidence may suggest that...

[Erich Maria Remarque]

Also, Gallup's sample sizes are 5 times that.

Didn't know that. There is nothing about sample size on their page.

Here you go:

You must be logged in to read this quote.

http://www.gallup.com/poll/101872/faqs.aspx#!mn-us

I see, but the margin of error is still pretty big anyway. That's why one should look at average over few polls. If the trend is real, it will be revealed by all the polls eventually

[Erich Maria Remarque]

Looks like getting sick might have actually helped Clinton's favorables.

Counterintuitively, but so far the limited evidence may suggest that...

Lol, no. Even if trend is real, there is no evidence at all that it was her illness.

As I said in some other thread. I'm more happy about slow, but steady growth, then large bump caused by news. Slow growth usually sticks, while news go away. We need one-two week of polling before we know for sure.

[GeorgiaModerate]

Somewhat surprisingly considering the latest polls, Clinton has opened up her biggest lead in quite a while in Gallup favorability numbers.  Sample period is 9/19-25:

Clinton 41/55 (-14)
Trump 32/63 (-31)

Clinton has been slowly gaining and Trump slowly declining over the last several days. 

http://www.gallup.com/poll/189299/presidential-election-2016-key-indicators.aspx?g_source=ELECTION_2016&g_medium=topic&g_campaign=tiles#pcf-image

[heatcharger]

Somewhat surprisingly considering the latest polls, Clinton has opened up her biggest lead in quite a while in Gallup favorability numbers.  Sample period is 9/19-25:

Clinton 41/55 (-14)
Trump 32/63 (-31)

Clinton has been slowly gaining and Trump slowly declining over the last several days. 

http://www.gallup.com/poll/189299/presidential-election-2016-key-indicators.aspx?g_source=ELECTION_2016&g_medium=topic&g_campaign=tiles#pcf-image

Hopefully this lasts past the debate.

[GeorgiaModerate]

Current sample (Sep 27-Oct 3) is:

Trump 33/63 (-30)
Clinton 42/54 (-12)

The oldest sample on their page (Sep 3-10) is Trump 35/5060 (-25), Clinton 38/57 (-19).  The net change since then is Trump -5, Clinton +7.

[dspNY]

Current sample (Sep 27-Oct 3) is:

Trump 33/63 (-30)
Clinton 42/54 (-12)

The oldest sample on their page (Sep 3-10) is Trump 35/50 (-25), Clinton 38/57 (-19).  The net change since then is Trump -5, Clinton +7.

Those numbers would give Clinton a 7-8 point national win

[egalitt]

Gallup was pretty bad in 2012 and I assume they're pretty bad this year.

BUT MUH DATA POINTS

Right, Trump will win anyways, notwithstanding polling results.

[GeorgiaModerate]

Gallup was pretty bad in 2012 and I assume they're pretty bad this year.

BUT MUH DATA POINTS

Right, Trump will win anyways, notwithstanding polling results.

These are just favorability numbers.  Gallup doesn't poll the Presidential race anymore.

[HillOfANight]

https://twitter.com/PollsAndVotes/status/783383784360022016
The Marquette pollster did a nice graph of this.



Within their parties, Clinton up, Trump down.