He's running behind Hillary just about everywhere that counts, and his demographics are worse than Romney or McCain ever were. He can sell that narrative to his followers and declare himself the King of Debates all he wants but he's got to make gains somewhere. The debates are a much wider audience and he needs to look presidential to the people he needs, college educated whites (since it's a foregone conclusion he won't pick up any minority groups). For that to happen he has to keep his dick in his pants (I shouldn't have to say so, but I mean both figuratively AND literally) and look at least slightly respectable. That's the only way he really wins anything tomorrow night, regardless of who he or the pundits declare the victor.
-
Hey, guest user. Hope you're enjoying NeoGAF! Have you considered registering for an account? Come join us and add your take to the daily discourse.
PoliGAF 2016 |OT11| Well this is exciting
- Thread starter Crayons
- Start date
- Status
He's running behind Hillary just about everywhere that counts, and his demographics are worse than Romney or McCain ever were. He can sell that narrative to his followers and declare himself the King of Debates all he wants but he's got to make gains somewhere. The debates are a much wider audience and he needs to look presidential to the people he needs, college educated whites (since it's a foregone conclusion he won't pick up any minority groups). For that to happen he has to keep his dick in his pants (I shouldn't have to say so, but I mean both figuratively AND literally) and look at least slightly respectable. That's the only way he really wins anything tomorrow night, regardless of who he or the pundits declare the victor.
Suikoguy
I whinny my fervor lowly, for his length is not as great as those of the Hylian war stallions
Are the undecided numbers that much higher than normal? I've not seen it spelled out compared to past elections.
I just don't see more than half of those undecided voters going Trump, and it's my current goto theory of why the polling is fucked.
I just don't see more than half of those undecided voters going Trump, and it's my current goto theory of why the polling is fucked.
Gross
Next weekend at the earliest, but it depends on how the week and the coverage of the debate shakes out.
It won't be until Thursday or Friday. Wednesday we'll see some polls that include Monday night, but the full effects won't be felt until Thursday, at the earliest.
Friday to be sure, although most of us will know right away if the debate helped Clinton or not and we'll see instant polls about "who won" or live focus groups, etc.
eBay Huckster
Member
The Skull Kid is the federal government's institutional levers of power, Majora's Mask is the entire Republican Party, and if you need to know who the Moon is then I probably want to punch you.
I'm looking at 538 Florida. Not to beat a dead horse, but... everything is adjusted towards Trump. Even if the pollsters (like PPP or Siena) are marked as slight R biased pollsters.
And with all of that, he's only leading by 0.8%. HuffPollster has her up ~2%.
GhostTrick
Banned
Guys, do you have a link where I can follow the debate tmr/tonight, from France ?
Check closer to the beginning of the debate, it'll be hard to miss this debate because it'll be practically everywhere.
//edit
Twitter is live streaming it.
GhostTrick
Banned
did months of handwaving trump's success away during the primaries just break nate silver? it seems like he's going out of his way to show things positively for trump just because he doesn't want to be wrong again.
RastaMentality
Banned
User 463088
Banned
Do we have confirmation that they'll be doing snap polls afterwards? They usually do, but I haven't heard anything.
I really don't know. Punditry wise, he can do whatever he wants. But I just don't get the adjustments.
6/10 latest FL polls are adjusted 1 or 2 points towards Trump.
D
Deleted member 231381
Unconfirmed Member
Almost all the adjustment comes from trendline weighting, which means it will disappear after ~2 weeks of good Clinton polling. It probably would have shown Florida flipping to Clinton by the end of this week looking at how events are going. When you use trendlines, you necessarily introduce a little lag into your model because you can't use today's trend because you don't know what today's trend is without knowing tomorrow's data point. His adjustments towards Trump are just a hangover from Clinton's bad polling week.
shinra-bansho
Member
The other article I read earlier also says that the terrible Google thing is weighted heavily though because of sample size.
And if what another poster earlier wrote about the trend adjustment is correct it seems kind of... dumb.
...
Anyway I am curious how Trump's incoherent rambling plays. Maybe they'll medicate him.
And if what another poster earlier wrote about the trend adjustment is correct it seems kind of... dumb.
...
Anyway I am curious how Trump's incoherent rambling plays. Maybe they'll medicate him.
Aaronology
Member
I'm a Sound Sleeper (Hillary Realist).
shinra-bansho
Member
Either way you end up a mess.
IAmAnAgnostic
Member
New national Bloomberg poll.
D
Deleted member 231381
Unconfirmed Member
ouch, and that's a Selzer, too
ONLY ONE POLL THO
but ouch
I'm glad the debates are happening, really, or you feel like the media narrative would be "Clinton sinking / troubled presidential campaign / etc.". As it is, a good performance and the ship'll be righted with all this behind us.
ONLY ONE POLL THO
but ouch
I'm glad the debates are happening, really, or you feel like the media narrative would be "Clinton sinking / troubled presidential campaign / etc.". As it is, a good performance and the ship'll be righted with all this behind us.
IAmAnAgnostic
Member
Didn't check, but here is the methodology.
http://assets.bwbx.io/documents/users/iqjWHBFdfxIU/rf1VkxwwhH2c/v0
D
Deleted member 231381
Unconfirmed Member
No. It just uses "definitely vote" as the LV screen. I've talked about this before, and that's a mildly Dem-leaning screen. The main shift seems to be coming from falling enthusiasm amongst younger voters and minorities again, and a poor showing amongst women (so I'd guess something going on with young women?).
D
Deleted member 231381
Unconfirmed Member
That's the demographics before weighting, see:
That means that's before both weighting and the LV filter; Selzer are only asking it to see how good their contact techniques are - an unbiased contact method should have all demographics within the MoE of the population, which is indeed the case here. It doesn't have a bearing on the poll itself, other than that heavily weighted samples (because they were undersampled originally) will have a correspondingly larger MoE.
That means that's before both weighting and the LV filter; Selzer are only asking it to see how good their contact techniques are - an unbiased contact method should have all demographics within the MoE of the population, which is indeed the case here. It doesn't have a bearing on the poll itself, other than that heavily weighted samples (because they were undersampled originally) will have a correspondingly larger MoE.
User 463088
Banned
The M/F split is completely off. The last time there was close to equal M/W was in 1980. That would be a 6 point swing towards men in an election with the first female nominee of a major political party. Those non white numbers seem a bit off too, but you know, we've seen that in polling a lot lately.
D
Deleted member 231381
Unconfirmed Member
adam's post is why people need to read my post.
EDIT: To be more precise, every pollster wants their unweighted contact sample's demographics to look the same as the demographics of the electorate, *not likely voters*. If it doesn't, it means your contact method is missing out a particular niche, which could affect the accuracy of the poll. So every pollster's *unweighted* sample should be pretty close to White 72, Black 13, Hispanic 11, Asian 2 etc. If you had a result like White 75, Black 14, Hispanic 8, Asian 2, you'd know you were probably missing out on a portion of Hispanic voters in some way - probably because you only poll in English and they only answer in Spanish - and that means even if you reweight the poll, it won't be accurate because a demographic isn't being contacted properly. Spanish-speaking Hispanics vote differently as a group to English-speaking ones, so your exclusion of them introduces a bias to your result.
However, Selzer's demographics are basically bang on; all within the MoE. So that means their contact methods are good. Then they apply likely voters screens and so on. Because women are more likely to vote than men, this is where the expected gender split should appear - NOT in the original unweighted sample because women are an equal portion of the electorate itself; they're only an outsized portion of likely voters.
EDIT: To be more precise, every pollster wants their unweighted contact sample's demographics to look the same as the demographics of the electorate, *not likely voters*. If it doesn't, it means your contact method is missing out a particular niche, which could affect the accuracy of the poll. So every pollster's *unweighted* sample should be pretty close to White 72, Black 13, Hispanic 11, Asian 2 etc. If you had a result like White 75, Black 14, Hispanic 8, Asian 2, you'd know you were probably missing out on a portion of Hispanic voters in some way - probably because you only poll in English and they only answer in Spanish - and that means even if you reweight the poll, it won't be accurate because a demographic isn't being contacted properly. Spanish-speaking Hispanics vote differently as a group to English-speaking ones, so your exclusion of them introduces a bias to your result.
However, Selzer's demographics are basically bang on; all within the MoE. So that means their contact methods are good. Then they apply likely voters screens and so on. Because women are more likely to vote than men, this is where the expected gender split should appear - NOT in the original unweighted sample because women are an equal portion of the electorate itself; they're only an outsized portion of likely voters.
User 463088
Banned
As a policy, I try not to.
Kidding!
.I hadn't yet taken a look inside the poll. I would still like to see their crosstabs, and I hate when pollsters don't release them.
That question doesn't sit well with me. It really, really doesn't.
CrankyKong
Member
Meanwhile, in the upside down.....
http://sitsshow.blogspot.com.au/201...dation-scandals-finish-the-DC-crooks-off.html
http://sitsshow.blogspot.com.au/201...dation-scandals-finish-the-DC-crooks-off.html
nature boy
Member
Trump +1 in Colorado, Clinton +1 in PA
http://edition.cnn.com/2016/09/26/p...linton-colorado-pennsylvania-polls/index.html
Clinton +6 4-way, +10 H2H in Virginia, Wason Center poll
http://edition.cnn.com/2016/09/26/p...linton-colorado-pennsylvania-polls/index.html
Clinton +6 4-way, +10 H2H in Virginia, Wason Center poll
Honestly, Canada taking over the US is not the worst outcome that could happen this election.
am I on a list now for reading that
D
Deleted member 231381
Unconfirmed Member
Trump's figures amongst whites without college degrees are incredible. He's regularly creeping over the 70% margin now. That Colorado poll is some spooky business.
The margins are expected. The enthusiasm of the group, however, is incredible. And a bit scary.
Plinko
Wildcard berths that can't beat teams without a winning record should have homefield advantage
Welp, another one confirming the blown lead in Colorado and tight race in PA. Here's hoping she wins the debates.
48% of respondents were independents. That seems high, but who knows. Good news there is that Johnson has 13%, and that will most likely drop.
Hillary's trustworthy numbers are tanking her campaign. She is fine with most categories.
Souldriver
Member
Trying not to get too involved in the election and poll numbers, but looking at the current situation...damn, how did it get so bad so quick?
Sometimes I don't understand the volatility of the voting populace or the polling. How do such swings between both candidates still exist?
Anyway. I'm still hoping this is a low point for Hillary, but I'm preparing for the worst. This year has proven that xenophobia and nationalism is the new hip thing in the Western world, so President Trump is within the realm of posibilities if the polls are close.
Sometimes I don't understand the volatility of the voting populace or the polling. How do such swings between both candidates still exist?
Anyway. I'm still hoping this is a low point for Hillary, but I'm preparing for the worst. This year has proven that xenophobia and nationalism is the new hip thing in the Western world, so President Trump is within the realm of posibilities if the polls are close.
D
Deleted member 231381
Unconfirmed Member
Honestly, even the margins aren't that expected. Nobody has ever won that group by that much in postwar American history.
As late as 1980's demographics, Trump would have won a 50 state sweep. That's... terrifying.
EDIT: In fact, looking closely, it'd have been enough as late as '92 maybe - hard to tell, but New York might have held out.
nature boy
Member
National polls out today show a tightening, YouGov, Bloomberg, Boston Herald.
And she ain't no Obama in terms of enthusiasm.
Although CNN/ORC has a big problem sampling people under 50.
And she ain't no Obama in terms of enthusiasm.
Although CNN/ORC has a big problem sampling people under 50.
Plinko
Wildcard berths that can't beat teams without a winning record should have homefield advantage
But she will have Obama out there and on the airwaves supporting her. He needs to get the message across that she is trustworthy.
User 463088
Banned
Anyone want to hazard a guess as to how you run a poll in Pennsylvania, but can't get an adequate sample out of Philadelphia? The fact that they couldn't explains how they couldn't get a decent non-white sample too.
I can't quite get my head around losing whites by 9, but only being up 1 in PA. Obama lost whites by 15 in 2012. She's also winning white women in that poll, something Obama didn't do in 2012.
I can't quite get my head around losing whites by 9, but only being up 1 in PA. Obama lost whites by 15 in 2012. She's also winning white women in that poll, something Obama didn't do in 2012.
I think it's probably time to start worrying quite a bit... Time is running out to make a difference other than through the debates. If we're being honest, September was a complete and utter waste after sitting out much of August after the successful convention. The campaign blew that lead and even after all those Trump fuckups and even after all those stupid comments/controversies, he's only slightly behind or even outright leading now.
Not saying to panic, but we got a tight toss-up race on our hands now. These state polls say as much. We can only hope that having that on-the-ground infrastructure advantage can shift a point or two in HRC's direction come election day.
And let's say Trump gains an advantage through tonight's debates or even subsequent debates, what then? The campaign needs to be impeccably on-the-ball from here on to November. The tide has turned and they have yet to blunt his building wave over the past few months since the disposal of Manafort.
Not saying to panic, but we got a tight toss-up race on our hands now. These state polls say as much. We can only hope that having that on-the-ground infrastructure advantage can shift a point or two in HRC's direction come election day.
And let's say Trump gains an advantage through tonight's debates or even subsequent debates, what then? The campaign needs to be impeccably on-the-ball from here on to November. The tide has turned and they have yet to blunt his building wave over the past few months since the disposal of Manafort.
cartoon_soldier
Member
OK that Colorado and PA polls giving me sweats
D
Deleted member 231381
Unconfirmed Member
I think it's pretty plausible that Republicans voting Libertarian were just there for the moral high ground, and are keener to return under cover of the secret ballot, whereas Democrats voting Libertarian are innate 'fuck the system' voters who won't leave come hell or high water. That explains why Johnson draws more from Clinton when provided, but why his dips mostly go Trumpwards.
- Status
Similar threads
- Locked
- Locked