Trump Retakes lead in PPD Poll (+2)

People's Pundit Daily ^ | 10/20/2016 | PPD

[rb22982]

Trump: 43
Clinton: 41
Johnson: 7
Stein: 4
Undecided: 5

[alternatives?]

Clinton has a hard time getting over 41%. I think a lot of the remainder end up going to Trump.

[SoldierDad]

Can someone explain why Hillary isn’t being beaten by double digits?

[Irish Eyes]

Can someone explain why Hillary isn’t being beaten by double digits?

The media for one reason.

[rb22982]

Yes - MSM, GOPe

[SoldierDad]

I think there’s more to it than that.

[MMaschin]

Rasmussen went from +2 Trump to +7 Clinton back to +3 Trump in about a week so who knows.

To me, that gives the poll a sense of believably.

After the tape came out, I was mad. I questioned my support. But after a couple days, and hearing Hillary's voice, and thinking about what was at stake - I came back around. So, to me, that large drop, and slow pickup, reflect exactly what I was going through.

Curious thing, all the polls that show Trump currently in the lead, show that. The polls that show Hillary with a large lead, look linear. That seems unnatural, and wrong to me.

[rb22982]

I think the polls you are talking about will show a large hillary lead for as long as they can and then in the last day change it to relatively close so they won’t be “too far off” and will just say undecideds broke Trump. They want to discourage R turnout as much as possible.

[Fai Mao]

“1) Getting the right sample. Landlines were historically the best way but now only about 60% of the country has a landline, only 7% of people respond to calls and around 30% of people lie about their intention on whether or not they will vote
2) You can take your subgroups from #1 and then forecast an electoral mix by demographic based on response from #1 but that typically assumes turnout similar to the last couple of elections which may or may not be the case.
3) Internet polling - same issues - you have to have the right email database of a representative sample, get those people to agree to take the poll and then weight them proportional to your expected demographic turnout.”

And they call democratic places that can’t vote. I actually received a poll call two nights ago. I asked them if they knew where the 671 Area Code was and if they knew that citizens of US territories could not vote? They politely hung up on me.

[Sam Gamgee]

I have taken all the polls of the last 4 days that have over 1000 in number sampled. That is the Rasmussen, PPD, LA Times. I have adjusted the LA Times into 2 polls. One where I knock off 2% on Trump and the other by 5% as an economist has re-weighted the LAT to give Hillary plus 5.

I get an average of 1% for Clinton leading. I have also 1% on Clinton leading in FL. Donald has stopped the hemorrhaging, and now we need to see if he can get above HC in the polling.

[rb22982]

Interesting way to do it - out of curiosity, why would you change the results of the second most accurate poll in 2012 by 2% or 5%?

[Pearls Before Swine]

Hillary’s nuclear secret spill on national TV is going to have a huge impact on what is going on right now.

I hope so. But, common sense says the number had to be under 10 to 15 minutes, or else there would be no launch on warning capability. Then again, Obama has said, "we can absorb a strike" (if its from a friendly to him Muslim country). Also, the specific number is important, because it tells how far offshore missile submarines can loiter if they want to launch to degrade response capability.

What a stupid b***h!

[Sam Gamgee]

http://www.latimes.com/nation/politics/trailguide/la-na-trailguide-live-updates-an-alternative-look-at-the-usc-la-times-1476982671-htmlstory.html

This explains the 5%.

As for the 2% a month or so ago, the makers of the poll admitted their weighting could result in 1 to 2% over stating Trump.

[Sam Gamgee]

The tables are interesting. How can Trump lead Hilary 40 to 25 with independents, and she only get 30% of whites and he barely wins?

[wyowolf]

I hope you’re right. I voted yesterday in Ga, and if everyone thinks the AA vote is going to stay home, I’d say they are dead wrong. I figured there would be minimal people there... there were a TON! for the middle of the day and middle of the week I was shocked...

I got a really bad feeling about this...

[rb22982]

He has a different way of weighting the same data. Not a surprise. Doesn’t make him right. Like I said, this poll was dead on in 2012. There was an article posted on here the other day that had random polling data sent to 6 different pollsters and they measured anywhere from Trump +1 to Clinton +5 off the same made up data.

[rb22982]

Turnout weighting looks too high for both D and R is why - D/R/I: 42/37/20 - they might be including leaners in the D and R

[rb22982]

Also the la times poll is just a two way. Trump has consistently performed better in the four way by 1-2 Pts so even if it was off slightly there is built in cushion for trump in this poll

[rb22982]

Im hoping trump gets a lot of secret AA votes. Non college whites normally vote 10-25D but complete reversal this election. If trump gets 20% of AA it’s over

[Steve Van Doorn]

said, "Turnout weighting looks too high for both D and R is why D/R/I: 42/37/20"
that is D+6 which is the same as 2012. Hillary will not pull that many Democrats. independent are about 20 points to short. We need to take into account the corruption rate which this year will be high both parties do not want him to win.

[convoter2016]

I agree that Rasmussen's poll jumps from +2 Trump to +7 Clinton to +3 Trump around the time of the Billy Bush Donald Trump bus video means that Rasmussen is honestly trying to predict the election. Rasmussen may still be wrong but at least it's a credible poll. After the 2012 election Rasmussen admitted that using only respondents who answered land line phones might have biased the poll toward Republicans. Rasmussen now includes on-line voters in their polls. We have to wait to see if Rasmussen has improved their predictions.