2016 US Presidential Election

Seen something interesting in the news or on the intertubes? Discuss it here.

Moderators: Zamfir, Hawknc, Moderators General, Prelates

User avatar
Dauric
Posts: 3753
Joined: Wed Aug 05, 2009 6:58 pm UTC
Location: In midair, traversing laterally over a container of sharks. No water, just sharks, with lasers.

Re: 2016 US Presidential Election

Postby Dauric » Tue Oct 25, 2016 3:21 pm UTC

trpmb6 wrote:
It probably means nothing? If, in reality, Clinton was up by 5 points, you wouldn't expect to see every single pollster return a result of Clinton+5. You would expect to see a normal distribution of results centred around Clinton+5. You would expect to see that even if every pollster had the ability to take a true random sample of respondents and extract completely accurate information from them. In the universe we live in, polling is hard and some pollsters lean one way or the other, which means you get a few more weird outliers than you would expect if God was an actuary.


Correct, but when numerous models are showing a higher than usual turnout of democrats, some even higher than Obama's 2008 and 2012 turnout, you start to realize those polls are shifting the normal distribution to the left. Maybe these pollsters see something in the electorate I don't; it is their job afterall.. I just design airplanes. But I don't see the same level of enthusiasm for Clinton from my friends as I saw with Obama. Conversely, I see a much higher level of enthusiasm from my friends who support trump than I've ever seen for other candidates. And that's where I think the rallies show this.


You're looking at enthusiasm in support -for- Trump or Clinton. This year isn't really one where any candidate get support -for- their position, but people "voting for" one candidate or the other are actually voting -against- the other. Even Trump supports are hard pressed to point to any Trump policies they'd be better off under, they're mostly driven by a desire to keep Hillary out of office. Mostly the same can be said for Hillary, she's not really an inspirational candidate (definitely not the way Obama was), so a significant number of her votes will not be so much support for her candidacy, but an effort to prevent a Trumpster Fire in the Oval Office.
We're in the traffic-chopper over the XKCD boards where there's been a thread-derailment. A Liquified Godwin spill has evacuated threads in a fourty-post radius of the accident, Lolcats and TVTropes have broken free of their containers. It is believed that the Point has perished.

User avatar
ahammel
My Little Cabbage
Posts: 2135
Joined: Mon Jan 30, 2012 12:46 am UTC
Location: Vancouver BC
Contact:

Re: 2016 US Presidential Election

Postby ahammel » Tue Oct 25, 2016 3:26 pm UTC

trpmb6 wrote:
ahammel wrote:AFAICT the bias adjustment is done on the basis of historical accuracy of the polls. Can you give an example of what you mean?

But that is part of my point. This election is unlike any modern election. Others have already noted on this forum that Fivethirtyeight has repeatedly noted the complication of adding 3rd party candidates to their models
Sure, maybe there's a systematic bias one way or the other. That's why the 538 model has Clinton winning "only" ~85% of the time. (If they assumethatt polling errors among states are independent, Clinton wins >98% of the time.)

It probably means nothing? If, in reality, Clinton was up by 5 points, you wouldn't expect to see every single pollster return a result of Clinton+5. You would expect to see a normal distribution of results centred around Clinton+5. You would expect to see that even if every pollster had the ability to take a true random sample of respondents and extract completely accurate information from them. In the universe we live in, polling is hard and some pollsters lean one way or the other, which means you get a few more weird outliers than you would expect if God was an actuary.


Correct, but when numerous models are showing a higher than usual turnout of democrats, some even higher than Obama's 2008 and 2012 turnout, you start to realize those polls are shifting the normal distribution to the left. Maybe these pollsters see something in the electorate I don't; it is their job afterall.. I just design airplanes. But I don't see the same level of enthusiasm for Clinton from my friends as I saw with Obama. Conversely, I see a much higher level of enthusiasm from my friends who support trump than I've ever seen for other candidates. And that's where I think the rallies show this.

I just wish all the people who dislike both Trump and Clinton would actually unite under a common banner.
It wouldn't surprise me if the Dems got a turnout bump because of enthusiasm for not-Trump, even if their base isn't that enthusiastic about HRC.

TBH, I trust what the pills say more than I trust turnout at Trump rallies and what you say your friends think. But there's really no point in arguing about it, we're going to run the experiment for real in a few weeks and then we shall see what we shall see. I'd just note that every single election in my memory people have wrung their hands about biased polls, and every time numbers continue to be the best way to tell which of two things is bigger.

Ninja'd hard.
He/Him/His/Alex
God damn these electric sex pants!

User avatar
Xeio
Friends, Faidites, Countrymen
Posts: 5084
Joined: Wed Jul 25, 2007 11:12 am UTC
Location: C:\Users\Xeio\
Contact:

Re: 2016 US Presidential Election

Postby Xeio » Tue Oct 25, 2016 3:55 pm UTC

At this point we don't even have to purely speculate about turnout. We're already seeing Democrat turnout up significantly in several early voting states compared to 2012, and up beyond the 2008 early turnout as well.

It's not clear if that will end up translating to the final turnout or not, but it's interesting. Maybe this is a preview of the stronger ground game from the Clinton campaign.

Granted some of those large turnout increases are in places like Texas, which probably isn't in play this year even if Trump is polling comparatively badly there.

KnightExemplar
Posts: 5489
Joined: Sun Dec 26, 2010 1:58 pm UTC

Re: 2016 US Presidential Election

Postby KnightExemplar » Tue Oct 25, 2016 3:59 pm UTC

Xeio wrote:Granted some of those large turnout increases are in places like Texas, which probably isn't in play this year even if Trump is polling comparatively badly there.


Washington Post demoted Texas from "safe" to "leans Republican", because Hillary / Trump differential is less than 5% now.

Texas is still likely to vote Republican, but... daaaaammmmmmnnnnn, we're talking about Texas now? Lol.
First Strike +1/+1 and Indestructible.

User avatar
Dauric
Posts: 3753
Joined: Wed Aug 05, 2009 6:58 pm UTC
Location: In midair, traversing laterally over a container of sharks. No water, just sharks, with lasers.

Re: 2016 US Presidential Election

Postby Dauric » Tue Oct 25, 2016 4:05 pm UTC

Xeio wrote:At this point we don't even have to purely speculate about turnout. We're already seeing Democrat turnout up significantly in several early voting states compared to 2012, and up beyond the 2008 early turnout as well.

It's not clear if that will end up translating to the final turnout or not, but it's interesting. Maybe this is a preview of the stronger ground game from the Clinton campaign.

Granted some of those large turnout increases are in places like Texas, which probably isn't in play this year even if Trump is polling comparatively badly there.


Colorado has been a battleground state this year.Colorado Public Radio has been tracking information from the Colorado Secretary of State as far as which parties have been returning ballots and putting it into a graphic chart.
We're in the traffic-chopper over the XKCD boards where there's been a thread-derailment. A Liquified Godwin spill has evacuated threads in a fourty-post radius of the accident, Lolcats and TVTropes have broken free of their containers. It is believed that the Point has perished.

User avatar
Soupspoon
You have done something you shouldn't. Or are about to.
Posts: 2359
Joined: Thu Jan 28, 2016 7:00 pm UTC
Location: 53-1

Re: 2016 US Presidential Election

Postby Soupspoon » Tue Oct 25, 2016 4:16 pm UTC

Talking of TX, there's a lot of mumbling from places I don't actually give much credance to that Texas voting machines are biased towards Clinton.

I'm inclined to believe it's part wilful obfuscation and part genuine but unfounded hysteria by Trump supporters who have taken his pre-accusations 'seriously' and want to get their retalliation in against a Hillary win in first. There might be a genuine problem, of course. The likelihood that there's serious fraud (and yet it's been made so obvious to the voters) is pretty much hard to credit.

User avatar
trpmb6
Posts: 272
Joined: Tue Jul 09, 2013 6:27 pm UTC

Re: 2016 US Presidential Election

Postby trpmb6 » Tue Oct 25, 2016 4:24 pm UTC

Dauric wrote:You're looking at enthusiasm in support -for- Trump or Clinton. This year isn't really one where any candidate get support -for- their position, but people "voting for" one candidate or the other are actually voting -against- the other. Even Trump supports are hard pressed to point to any Trump policies they'd be better off under, they're mostly driven by a desire to keep Hillary out of office. Mostly the same can be said for Hillary, she's not really an inspirational candidate (definitely not the way Obama was), so a significant number of her votes will not be so much support for her candidacy, but an effort to prevent a Trumpster Fire in the Oval Office.


I think this is probably the best response I've seen. I'm in the bucket of, I am a republican, who can't stand Hillary, but detest who Trump is and will not vote for either group. My wife is the same. I've simply been trying to correlate what I see in polls with what I see in rallies. Your comment brings me to another observation. Trump's rallies show he does have a base of enthusiastic supporters. Clinton has her own base but they don't seem to turn out in support of her in quite the same numbers. In past elections (I know I've been saying this election is unlike any other election we've had) the candidate who was being propped up by voters who simply detest "the other guy" have a harder time of winning. Mainly because these type of voters put a lower priority on getting out and voting (more on that in a second). Clinton and Trump both have historically low approval ratings.

The questions are then: Which candidate can overcome their low approval ratings and draw more of the "I'm for the other guy" vote, who can keep their base enthusiastic and actually get out the vote and how much of a difference does the increase in ballots by mail make.

I think 2 of those 3 favor Clinton, and one of them (base enthusiasm) favors Trump (my opinion). I think whoever takes 2 of the 3 takes the election.

User avatar
CorruptUser
Posts: 8648
Joined: Fri Nov 06, 2009 10:12 pm UTC

Re: 2016 US Presidential Election

Postby CorruptUser » Tue Oct 25, 2016 4:28 pm UTC

KnightExemplar wrote:
Xeio wrote:Granted some of those large turnout increases are in places like Texas, which probably isn't in play this year even if Trump is polling comparatively badly there.


Washington Post demoted Texas from "safe" to "leans Republican", because Hillary / Trump differential is less than 5% now.

Texas is still likely to vote Republican, but... daaaaammmmmmnnnnn, we're talking about Texas now? Lol.


If Texas goes Clinton, no points for correctly guessing that it'll be blamed on all the Mexicans that moved back home illegally immigrated.

User avatar
Soupspoon
You have done something you shouldn't. Or are about to.
Posts: 2359
Joined: Thu Jan 28, 2016 7:00 pm UTC
Location: 53-1

Re: 2016 US Presidential Election

Postby Soupspoon » Tue Oct 25, 2016 4:30 pm UTC

CorruptUser wrote:If Texas goes Clinton, no points for correctly guessing that it'll be blamed on all the Mexicans that moved back home illegally immigrated.
Mexican voting machines. *nodnod*

GodShapedBullet
Posts: 686
Joined: Mon Nov 26, 2007 7:59 pm UTC
Location: Delaware
Contact:

Re: 2016 US Presidential Election

Postby GodShapedBullet » Tue Oct 25, 2016 5:35 pm UTC

trpmb6 wrote:The questions are then: Which candidate can overcome their low approval ratings and draw more of the "I'm for the other guy" vote, who can keep their base enthusiastic and actually get out the vote and how much of a difference does the increase in ballots by mail make.

I think 2 of those 3 favor Clinton, and one of them (base enthusiasm) favors Trump (my opinion). I think whoever takes 2 of the 3 takes the election.


This is a pretty good take on some important issues the campaigns are facing.

The "whoever takes 2 of the 3 takes the election" is not strictly necessary, because none of the questions are binary. If you really dominated getting out the vote of your adoring base, it might not matter if you lose out in mail-in voting or attracting undecideds.

It'll also be interesting to see how the supposed superiority of Clinton's ground game affects get out the vote efforts. Could this mitigate the supposed enthusiasm gap?

A lot of interesting things to unpack once the election is over.

User avatar
ahammel
My Little Cabbage
Posts: 2135
Joined: Mon Jan 30, 2012 12:46 am UTC
Location: Vancouver BC
Contact:

Re: 2016 US Presidential Election

Postby ahammel » Tue Oct 25, 2016 5:41 pm UTC

trpmb6 wrote:I think this is probably the best response I've seen. I'm in the bucket of, I am a republican, who can't stand Hillary, but detest who Trump is and will not vote for either group. My wife is the same. I've simply been trying to correlate what I see in polls with what I see in rallies.
Why do you expect what you see in the polls to be particularly correlated with what you see in the rallies?
He/Him/His/Alex
God damn these electric sex pants!

morriswalters
Posts: 6885
Joined: Thu Jun 03, 2010 12:21 am UTC

Re: 2016 US Presidential Election

Postby morriswalters » Tue Oct 25, 2016 5:45 pm UTC

Zamfir wrote:
morriswalters wrote:The bar in using bribery in that fashion is so low as to be non existent. In common use it begs the comparison to illegal behavior, AKA, Crooked Hillary. That definition would also apply to the billion or so Hillary will raise in this current election cycle.

Thankfully, the Trump campaign has read our discussion and decided they want none of this vague stuff. "is it bribe or not". If you call them to ask if they can be bribed, they say sure, we'll send you our preferred procedure.
http://www.telegraph.co.uk/news/2016/10 ... r-fundrai/

Our non conversation covered past and sitting Presidents, whom you seemed to say, either accept or have accepted bribes, for vague actions unspecified in order to build Presidential Libraries, Foundations and of course the enrich the Bush Oil companies.

Did that investigation uncover a bribe to Trump? It consisted of various people saying they could get to The Donald to get him to listen. Not that The Donald was asking for the money or even was aware of the conversation. The felony is taking illegal campaign contributions from a foreign national. A specific crime. This type of investigation reminds me of that fool who investigated Planned Parenthood.

User avatar
The Great Hippo
Swans ARE SHARP
Posts: 6829
Joined: Fri Dec 14, 2007 4:43 am UTC
Location: behind you

Re: 2016 US Presidential Election

Postby The Great Hippo » Tue Oct 25, 2016 5:59 pm UTC

Zamfir wrote:Another thing - about that "legacy" thing. That might imply that the Clinton's only benefit here is some abstracted contribution to charity, prhaps for the associated praise. If that is what you're saying, then I think it underestimates the value of a large personal charity fund.
Right; every dollar that goes to the Clinton Foundation furthers the political ambitions of the Clinton family. Along with providing charitable services, it creates opportunities for the Clintons to build relationships and goodwill with various groups and individuals. Obviously, they didn't create the Clinton Foundation as a political capital factory -- but charitable organizations do produce political capital, and it's naive to presume that the Clintons aren't going to benefit from that.
ahammel wrote:Why do you expect what you see in the polls to be particularly correlated with what you see in the rallies?
This confuses me a lot too; it's a sentiment I've seen expressed numerous times. It seems on the scale of confusing anecdotal evidence with empirical evidence; a lack of correlation between the amount of support you physically see and the amount of support we actually measure seems to me to be not even slightly unusual.

I'm also fascinated with trying to understand the Trump base -- I have a strong (purely feeling-based) notion that one of the primary fears driving their support of Trump is fear of irrelevance. It makes sense to me when I think about where a lot of them come from: white (confronted with growing populations of 'not-white'), middle-class (which is shrinking), uneducated (makes it harder to keep up with the accelerating pace of technology), older voters (closer to death).
Last edited by The Great Hippo on Tue Oct 25, 2016 6:03 pm UTC, edited 1 time in total.

User avatar
Soupspoon
You have done something you shouldn't. Or are about to.
Posts: 2359
Joined: Thu Jan 28, 2016 7:00 pm UTC
Location: 53-1

Re: 2016 US Presidential Election

Postby Soupspoon » Tue Oct 25, 2016 6:03 pm UTC

(On the subject of support at rallies.) Obviously it's just a rallying cry (quite literally, as well), but when watching a rally of whichever candidate (because I'm interested, even though not involved) the up-play of the featured candidate and down-play of their opponent(s) by cheers and boos to obviously loaded statements and then taking that near-guaranteed sense of agreement to justify "and that's why we're gonna win!" always strikes me as a annoying. Regardless of whether the currently featured candidate is the one I'd support if I could...

But it's a different system, a little bit alien to me. (For indirect proof, observe Nigel Farage's appearance at his first Trump rally. That guy always annoyed me, here in the UK, with his presentation style, but he seemed almost sane alongside Trump. He also didn't mesh too well into the US style. His delivery speed was atuned to the British stump-speed, he didn't have the same sense of making cheer-worthy statements and pausing for the 'Merkin-style adulation, but ploughed onward, overtalking the opening smatterings of vocal support by the crowd until it ended up with a non-plussed audience refraining from firing their supportive salvos, half of whom had probably started off thinking "UK" was a typo for "OK", not really knowing much more than apparently this Limey Prime Minister(!) was endorsing Donald and that was proably a good thing, right? ;) Well, maybe that. But it was a distinct culture clash, even if not a Borat-scale one..)

Tyndmyr
Posts: 10119
Joined: Wed Jul 25, 2012 8:38 pm UTC

Re: 2016 US Presidential Election

Postby Tyndmyr » Tue Oct 25, 2016 6:47 pm UTC

trpmb6 wrote:I'd like to know if those here actually believe the ABC/Washington post poll that was released showing Clinton with a 12 point lead in a 4 way race. The demographic sampling was 36-27-31, Democrats / Republicans / Independents. Democrats didn't even come close to that high (+9) of an advantage over republicans for either of Obama's wins in 2008 and 2012. And that was with Obama's huge rallies he had back in the lead up to the 2008 election (Something I might note, is quite similar to the rallies Trump has been holding for the past year and a half, see Tampa rally from yesterday). Couple this with the recent podesta wikileak email dump showing talking points about how to manipulate polling with oversampling to discourage republican voters from turning out.

I have no doubt that the polling this election is extremely volatile and unpredictable. Ranging from the rise of two 3rd party candidates (and even a 3rd in Evan who is making Utah a potential swing state) to the day to day scandal driven media. But a 9 point democrat advantage for a poll seems way off base to me.


Why? Look, I wouldn't base my assessment on a single poll so much as the general trend of polling. And overall, polling has Clinton with a significant lead over Trump. That doesn't mean any given poll is perfect, but it gives you a pretty good idea of the plausible range of actual preferences. Volatility doesn't actually seem to be unusual. Third parties, overall, are not that huge of an influence. Utah, sure. But the rest of the country, not really.

It seems likely that Clinton does actually have a healthy lead. The exact number might be off a touch, but right now, it's not especially close, so uncertainty isn't a huge factor. If anything, aggregates like Fivethirtyeight may be undervaluing certain factors. For instance, they're showing MN as more purple that it should be due to lack of polling. In practice, people aren't polling MN because it always goes blue for the president by a healthy margin. So, the state isn't *actually* in play in any real way. It's going to break for Clinton in any scenario short of a Trump blowout(which is wildly improbable).

Thus, Trump's position may actually be slightly worse than it is commonly portrayed. At this point in the game, he needs a rather large break.

trpmb6 wrote:
ahammel wrote:AFAICT the bias adjustment is done on the basis of historical accuracy of the polls. Can you give an example of what you mean?

But that is part of my point. This election is unlike any modern election. Others have already noted on this forum that Fivethirtyeight has repeatedly noted the complication of adding 3rd party candidates to their models.


It's a modeling issue, sure. In part because so much stuff is centered around the top two parties. So, there's issues like folding in the polls that have the third parties and those that don't. It's a complication for them, but it's not a particularly important thing for Trump's odds. Most areas are simply not seeing a lot of third party support. It's only interesting because fivethirtyeight is trying to get a lot more accuracy and detail than a lot of historical poll references have.

User avatar
Quizatzhaderac
Posts: 1510
Joined: Sun Oct 19, 2008 5:28 pm UTC
Location: Space Florida

Re: 2016 US Presidential Election

Postby Quizatzhaderac » Tue Oct 25, 2016 7:06 pm UTC

Zamfir wrote:The problem that Zohar describes , is that coalitions make it hard to distinguish from the outside when a party made an empty promise, and when they genuinely tried to achieve a goal but had to give it up in coalition negotiations. It's a genuine downside.
So it's a transparency issue? (assuming the opaqueness isn't hiding other issues)

Maybe publish the collation agreements into public record in parliamentary systems?
For presidential systems, it would be nice if the legislature's voting records also showed what log rolling was going on.
The thing about recursion problems is that they tend to contain other recursion problems.

KnightExemplar
Posts: 5489
Joined: Sun Dec 26, 2010 1:58 pm UTC

Re: 2016 US Presidential Election

Postby KnightExemplar » Tue Oct 25, 2016 7:08 pm UTC

trpmb6 wrote:I'd like to know if those here actually believe the ABC/Washington post poll that was released showing Clinton with a 12 point lead in a 4 way race.


Since Washington Post leans liberal, its reasonable to expect that the poll would lean left in some assumptions. That doesn't mean that the methodology is bad or that the poll is useless. But it means that we should change our expectations of individual polls.

For example, most discussion around here revolves around RealClearPolitics or FiveThirtyEight, which are polls-of-polls.

For example: http://www.realclearpolitics.com/epolls ... -5491.html

There are dozens and dozens of polls in America, conducted by many news organizations. Instead of arguing over which ones are good or bad, its a good idea to just average them all together.

In the vast majority of cases, individual polls are very good at tracking, since individual publications tend to keep similar methodologies every few months when they conduct a poll. (Ex: I bet you that the WashPo has used the same 36-27-31 split you dislike the whole year). Indeed, there are some polls out there that are explicitly designed to track shifting opinions and NOT designed to actually predict the election. As long as the methodology is consistent, then the poll is not about the precise number they arrive at... but instead about how the number moves throughout the year.

EDIT: It appears that the ABC / Washington Post poll you're criticizing is a tracking poll, and not intended to actually predict the election.

In the past month, all polls have shifted towards Clinton's favor. That's the main story here. Of course we won't know the precise election results until Nov 8th... but that's why we hold elections. To officially count things. All the other polls are just estimations and are subject to bias.

The polls-of-polls model of RealClearPolitics and FiveThirtyEight are likely superior. It seems like averaging all the polls together into one model has been successful. Add some bias towards polls which have historically been useful, ignore the polls that are easily manipulated or have had poor correlations to results (ex: ignore internet polls. They've been junk for decades)
Last edited by KnightExemplar on Tue Oct 25, 2016 7:19 pm UTC, edited 3 times in total.
First Strike +1/+1 and Indestructible.

User avatar
sardia
Posts: 5763
Joined: Sat Apr 03, 2010 3:39 am UTC

Re: 2016 US Presidential Election

Postby sardia » Tue Oct 25, 2016 7:12 pm UTC

Tyndmyr wrote:
Why? Look, I wouldn't base my assessment on a single poll so much as the general trend of polling. And overall, polling has Clinton with a significant lead over Trump. That doesn't mean any given poll is perfect, but it gives you a pretty good idea of the plausible range of actual preferences. Volatility doesn't actually seem to be unusual. Third parties, overall, are not that huge of an influence. Utah, sure. But the rest of the country, not really.
.

Minor quibble, the volatility is higher than the 2012 races, which both lowers Clinton's chance of winning and increases her chance of a Texas sized landslide. Third parties are influential... In how volatile the race is. A third party vote is the same as a undecided vote to the model. So a high third party voter size means the election is extra uncertain(or the model is extra cautious with its error bars).

User avatar
ahammel
My Little Cabbage
Posts: 2135
Joined: Mon Jan 30, 2012 12:46 am UTC
Location: Vancouver BC
Contact:

Re: 2016 US Presidential Election

Postby ahammel » Tue Oct 25, 2016 7:30 pm UTC

KnightExemplar wrote:
trpmb6 wrote:I'd like to know if those here actually believe the ABC/Washington post poll that was released showing Clinton with a 12 point lead in a 4 way race.


Since Washington Post leans liberal, its reasonable to expect that the poll would lean left in some assumptions.
Maybe, but bear in mind that the LA Times poll, for example, is strongly Trump-leaning despite the fact that the paper itself is endorsing HRC. I don't think it's clear that Dem-leaning institutions tend to publish Dem-leaning polls in general (or vice-versa). The skews are more likely related to things like methodology1 than anybody fudging numbers in favour of their preferred candidate.

1. E.g., if you call landline phones, you tend to reach older folks, who tend to vote Republican.
He/Him/His/Alex
God damn these electric sex pants!

User avatar
trpmb6
Posts: 272
Joined: Tue Jul 09, 2013 6:27 pm UTC

Re: 2016 US Presidential Election

Postby trpmb6 » Tue Oct 25, 2016 7:33 pm UTC

The Great Hippo wrote:I'm also fascinated with trying to understand the Trump base -- I have a strong (purely feeling-based) notion that one of the primary fears driving their support of Trump is fear of irrelevance. It makes sense to me when I think about where a lot of them come from: white (confronted with growing populations of 'not-white'), middle-class (which is shrinking), uneducated (makes it harder to keep up with the accelerating pace of technology), older voters (closer to death).


I certainly think you're onto something there. But I also think the Republican party helped create the Trump base. It's important to note that many of the things Trump stands for, particularly immigration, are things that the republicans have promised to the base for the last few election cycles but failed to follow through on. It shouldn't be a surprise that white, uneducated (I hate that word, it is demeaning) middle class Americans support a candidate against free trade, who talks about bringing jobs back to America. They've seen manufacturing dwindle during their lifetime and have no place to go in the new tech driven American Economy. A lot of them finally have a candidate that they truly believe will follow through on the things they believe to be important. Most of them held their nose and voted for McCain and Romney (or as I suspect, even stayed home).

This is at the heart of my original post about polling and the connection to enthusiasm. There is a perceived enthusiasm gap between Clinton and Trump due to the large rallies he holds. Whether or not that actually materializes in the election is what I am most interested to see. Will the voters who don't like either hold their ground and do as I am doing and vote 3rd party? Or will they cave and vote for the Republican or Democrat? These are all questions that have to be answered by the pollsters when they form their sampling. The pollsters clearly think there is a larger democrat base this cycle. Some saying the largest we've ever seen. We'll find out on the 8th if that's really the case.

User avatar
ahammel
My Little Cabbage
Posts: 2135
Joined: Mon Jan 30, 2012 12:46 am UTC
Location: Vancouver BC
Contact:

Re: 2016 US Presidential Election

Postby ahammel » Tue Oct 25, 2016 7:38 pm UTC

trpmb6 wrote:It shouldn't be a surprise that white, uneducated (I hate that word, it is demeaning) middle class Americans support a candidate against free trade, who talks about bringing jobs back to America. They've seen manufacturing dwindle during their lifetime and have no place to go in the new tech driven American Economy.
Interesting, people who live in areas where the manufacturing jobs have gone away are less likely to support Trump; not more, and Trump supporters are richer than average, not poorer.
He/Him/His/Alex
God damn these electric sex pants!

KnightExemplar
Posts: 5489
Joined: Sun Dec 26, 2010 1:58 pm UTC

Re: 2016 US Presidential Election

Postby KnightExemplar » Tue Oct 25, 2016 7:40 pm UTC

ahammel wrote:
KnightExemplar wrote:
trpmb6 wrote:I'd like to know if those here actually believe the ABC/Washington post poll that was released showing Clinton with a 12 point lead in a 4 way race.


Since Washington Post leans liberal, its reasonable to expect that the poll would lean left in some assumptions.
Maybe, but bear in mind that the LA Times poll, for example, is strongly Trump-leaning despite the fact that the paper itself is endorsing HRC. I don't think it's clear that Dem-leaning institutions tend to publish Dem-leaning polls in general (or vice-versa). The skews are more likely related to things like methodology1 than anybody fudging numbers in favour of their preferred candidate.

1. E.g., if you call landline phones, you tend to reach older folks, who tend to vote Republican.


I don't necessarily mean it in a way of "bad liberal bias media bad", but that a liberal paper would likely take methodologies that favor liberal candidates.

Ex: a Republican-leaning paper may be more likely to call landline phones, because there are more older folks who write in the paper and they trust landline polls better. A liberal publication may use SurveyMonkey instead, because the younger, liberal crowd is more accustomed to the internet.

Which goes back to my original point that every poll is basically going to be biased one way or the other because of methodology. The only 100% legitimate poll is the election itself. Still, it seems like averaging all of these polls together, combined with analyzing trends leads to useful data.
First Strike +1/+1 and Indestructible.

Tyndmyr
Posts: 10119
Joined: Wed Jul 25, 2012 8:38 pm UTC

Re: 2016 US Presidential Election

Postby Tyndmyr » Tue Oct 25, 2016 7:48 pm UTC

sardia wrote:
Tyndmyr wrote:
Why? Look, I wouldn't base my assessment on a single poll so much as the general trend of polling. And overall, polling has Clinton with a significant lead over Trump. That doesn't mean any given poll is perfect, but it gives you a pretty good idea of the plausible range of actual preferences. Volatility doesn't actually seem to be unusual. Third parties, overall, are not that huge of an influence. Utah, sure. But the rest of the country, not really.
.

Minor quibble, the volatility is higher than the 2012 races, which both lowers Clinton's chance of winning and increases her chance of a Texas sized landslide. Third parties are influential... In how volatile the race is. A third party vote is the same as a undecided vote to the model. So a high third party voter size means the election is extra uncertain(or the model is extra cautious with its error bars).


Higher than 2012, perhaps, but plenty of races have had more interesting back and forths, I think. Trump hasn't ever had a lead really, he's merely wobbled between about even and a solid Clinton lead.

And, in the end, the total number of undecided and third party votes STILL wouldn't be enough to put Trump at par, even if every single one was going to him. And, they're not all going to him. Even if it's more than half, there's going to be some split, and some will stay home.

He also needs a significant swing, or polling error or something. Even assuming a gradual regression to the mean, as per the previous two Clinton-favoring trends, there simply isn't enough time for that to get Trump enough ground at this point. It's likely that the final score will be somewhat closer than polling sits right now, but a fat enough gap in polling lets you not worry overly much about that. All Clinton needs to do right now is hang tight and let Trump flail.

I mean, sure, on some level, third party candidates do increase uncertainty, it's just that this factor is unlikely to actually save Trump.

trpmb6 wrote:I certainly think you're onto something there. But I also think the Republican party helped create the Trump base. It's important to note that many of the things Trump stands for, particularly immigration, are things that the republicans have promised to the base for the last few election cycles but failed to follow through on.


I think there's a good case that the republican party has created many of it's own problems by banging drums it doesn't really care about, yeah. Get the voters out, and not really care about their issues because...who else are they gonna vote for? Definitely some problems involved in taking your base for granted.

User avatar
ahammel
My Little Cabbage
Posts: 2135
Joined: Mon Jan 30, 2012 12:46 am UTC
Location: Vancouver BC
Contact:

Re: 2016 US Presidential Election

Postby ahammel » Tue Oct 25, 2016 7:51 pm UTC

KnightExemplar wrote:I don't necessarily mean it in a way of "bad liberal bias media bad", but that a liberal paper would likely take methodologies that favor liberal candidates.

Ex: a Republican-leaning paper may be more likely to call landline phones, because there are more older folks who write in the paper and they trust landline polls better. A liberal publication may use SurveyMonkey instead, because the younger, liberal crowd is more accustomed to the internet.
I agree that's totally plausible, but is that really what's going on? Like does 538 tend to adjust liberal newspapers' polls towards Trump and conservative organizations' polls HRC-ward? Honestly asking, I don't know.
He/Him/His/Alex
God damn these electric sex pants!

Dark567
First one to notify the boards of Rick and Morty Season 3
Posts: 3662
Joined: Thu Jun 25, 2009 5:12 pm UTC
Location: Everywhere(in the US, I don't venture outside it too often, unfortunately)

Re: 2016 US Presidential Election

Postby Dark567 » Tue Oct 25, 2016 8:04 pm UTC

ahammel wrote:Interesting, people who live in areas where the manufacturing jobs have gone away are less likely to support Trump; not more, and Trump supporters are richer than average, not poorer.
To some extent you have to unpack this. Trump voters are richer overall, but they are much more white, who tend to be richer than minorities. Now, within the white population, Trump supporters are poorer. As far as the manufacturing thing goes, it's not so much manufacturing vs. other jobs, but rural vs. urban manufacturing.


ahammel wrote:I agree that's totally plausible, but is that really what's going on? Like does 538 tend to adjust liberal newspapers' polls towards Trump and conservative organizations' polls HRC-ward? Honestly asking, I don't know.
Some what yes. They have polling adjustment biases so you can see for yourself how they handle it. http://projects.fivethirtyeight.com/201 ... t/updates/
Last edited by Dark567 on Tue Oct 25, 2016 8:06 pm UTC, edited 1 time in total.
I apologize, 90% of the time I write on the Fora I am intoxicated.


Yakk wrote:The question the thought experiment I posted is aimed at answering: When falling in a black hole, do you see the entire universe's future history train-car into your ass, or not?

User avatar
Thesh
Made to Fuck Dinosaurs
Posts: 5436
Joined: Tue Jan 12, 2010 1:55 am UTC
Location: Colorado

Re: 2016 US Presidential Election

Postby Thesh » Tue Oct 25, 2016 8:05 pm UTC

ahammel wrote:Like does 538 tend to adjust liberal newspapers' polls towards Trump and conservative organizations' polls HRC-ward? Honestly asking, I don't know.


Yes, it both adjusts and weights polls based on past performance. You can see all poll ratings here:

http://projects.fivethirtyeight.com/pollster-ratings/

And if you click on a state it will give you the weights and adjustments of individual polls:

http://projects.fivethirtyeight.com/201 ... t/florida/
Honesty replaced by greed, they gave us the reason to fight and bleed
They try to torch our faith and hope, spit at our presence and detest our goals

User avatar
ahammel
My Little Cabbage
Posts: 2135
Joined: Mon Jan 30, 2012 12:46 am UTC
Location: Vancouver BC
Contact:

Re: 2016 US Presidential Election

Postby ahammel » Tue Oct 25, 2016 8:15 pm UTC

Sorry, yeah I know that 538 does skew adjustments. What I'm asking is, do polls by liberal organizations tend to lean Democrat and polls by conservative organizations tend to lean Republican, or is it independent?
He/Him/His/Alex
God damn these electric sex pants!

KnightExemplar
Posts: 5489
Joined: Sun Dec 26, 2010 1:58 pm UTC

Re: 2016 US Presidential Election

Postby KnightExemplar » Tue Oct 25, 2016 8:31 pm UTC

ahammel wrote:Sorry, yeah I know that 538 does skew adjustments. What I'm asking is, do polls by liberal organizations tend to lean Democrat and polls by conservative organizations tend to lean Republican, or is it independent?


Personally speaking, I can't say for sure.

But my level of cynicism leans towards this direction, if you get what I'm saying.
First Strike +1/+1 and Indestructible.

Tyndmyr
Posts: 10119
Joined: Wed Jul 25, 2012 8:38 pm UTC

Re: 2016 US Presidential Election

Postby Tyndmyr » Tue Oct 25, 2016 8:41 pm UTC

ahammel wrote:Sorry, yeah I know that 538 does skew adjustments. What I'm asking is, do polls by liberal organizations tend to lean Democrat and polls by conservative organizations tend to lean Republican, or is it independent?


So, I scanned my state, MD, on the basis of I'd probably know which organizations are liberal and which are not, but it quickly breaks down.

Is google consumer surveys liberal or conservative, for instance? It biases slightly red, looks like.

Some, such as Goucher College, appear in the state list, but not the overall list.

So, not really sure.

User avatar
ahammel
My Little Cabbage
Posts: 2135
Joined: Mon Jan 30, 2012 12:46 am UTC
Location: Vancouver BC
Contact:

Re: 2016 US Presidential Election

Postby ahammel » Tue Oct 25, 2016 8:48 pm UTC

Tyndmyr wrote:
ahammel wrote:Sorry, yeah I know that 538 does skew adjustments. What I'm asking is, do polls by liberal organizations tend to lean Democrat and polls by conservative organizations tend to lean Republican, or is it independent?


So, I scanned my state, MD, on the basis of I'd probably know which organizations are liberal and which are not, but it quickly breaks down.

Is google consumer surveys liberal or conservative, for instance? It biases slightly red, looks like.

Some, such as Goucher College, appear in the state list, but not the overall list.

So, not really sure.
I was going to go figure it out myself based on newspaper endorsements, but, hilariously, Trump hasn't received a single endorsement from a newspaper that commissioned a poll this cycle.
He/Him/His/Alex
God damn these electric sex pants!

Dark567
First one to notify the boards of Rick and Morty Season 3
Posts: 3662
Joined: Thu Jun 25, 2009 5:12 pm UTC
Location: Everywhere(in the US, I don't venture outside it too often, unfortunately)

Re: 2016 US Presidential Election

Postby Dark567 » Tue Oct 25, 2016 8:58 pm UTC

Tyndmyr wrote:
So, I scanned my state, MD, on the basis of I'd probably know which organizations are liberal and which are not, but it quickly breaks down.

Is google consumer surveys liberal or conservative, for instance? It biases slightly red, looks like.

Some, such as Goucher College, appear in the state list, but not the overall list.

So, not really sure.
538 does it by results vs. polling. So if a poll repeatedly gets states wrong by overestimating GOP or Dem support, eventually 538 will introduce an adjustment in future projections. If you actually go through it, the bias seems to be more related to methodology than bias of the parent org.
I apologize, 90% of the time I write on the Fora I am intoxicated.


Yakk wrote:The question the thought experiment I posted is aimed at answering: When falling in a black hole, do you see the entire universe's future history train-car into your ass, or not?

User avatar
The Great Hippo
Swans ARE SHARP
Posts: 6829
Joined: Fri Dec 14, 2007 4:43 am UTC
Location: behind you

Re: 2016 US Presidential Election

Postby The Great Hippo » Tue Oct 25, 2016 9:03 pm UTC

trpmb6 wrote:It shouldn't be a surprise that white, uneducated (I hate that word, it is demeaning)
Non-college graduates? I can't think of any unjumbled way to clearly and effectively say 'people who did not attend or graduate college'.

I do agree that it's probably not the best term to address them, as it carries very unnecessary implications; I do think being a non-college graduate can lead to technology and science appearing quite mysterious (if only because college teaches you the terminology with which to digest technological and scientific knowledge). And I think that as we become more reliant on technology and science -- and as less-tech-savvy people watch more-tech-savvy people perform feats that appear to be indistinguishable from dark sorcery -- those less-tech-savvy people begin to feel increasingly like they're being left behind.

Fear of irrelevancy even explains the hysteric, apocalyptic tone of Trump and his base. When you're feeling irrelevant, what's a quick way to gain some relevance back? Talk about the end of the world; talk about last elections, biblical plagues -- cats and dogs, living together.

User avatar
Soupspoon
You have done something you shouldn't. Or are about to.
Posts: 2359
Joined: Thu Jan 28, 2016 7:00 pm UTC
Location: 53-1

Re: 2016 US Presidential Election

Postby Soupspoon » Tue Oct 25, 2016 9:09 pm UTC

ahammel wrote:but, hilariously, Trump hasn't received a single endorsement from a newspaper that commissioned a poll this cycle.
Don't you know that The Media is biased against him..? All the media! Everywhere!!

(Or people who get paid to think seriously about this kind of thing have thought seriously and whatever leaning they might have they still can't come up with enough good news stories in his favour... One or the other.)

User avatar
ahammel
My Little Cabbage
Posts: 2135
Joined: Mon Jan 30, 2012 12:46 am UTC
Location: Vancouver BC
Contact:

Re: 2016 US Presidential Election

Postby ahammel » Tue Oct 25, 2016 9:09 pm UTC

Soupspoon wrote:
ahammel wrote:but, hilariously, Trump hasn't received a single endorsement from a newspaper that commissioned a poll this cycle.
Don't you know that The Media is biased against him..? All the media! Everywhere!!

(Or people who get paid to think seriously about this kind of thing have thought seriously and whatever leaning they might have they still can't come up with enough good news stories in his favour... One or the other.)

Or he has a habit of threatening lawsuits against people who write bad things about him. Editorial boards love that shit.
He/Him/His/Alex
God damn these electric sex pants!

Tyndmyr
Posts: 10119
Joined: Wed Jul 25, 2012 8:38 pm UTC

Re: 2016 US Presidential Election

Postby Tyndmyr » Tue Oct 25, 2016 9:14 pm UTC

Soupspoon wrote:
ahammel wrote:but, hilariously, Trump hasn't received a single endorsement from a newspaper that commissioned a poll this cycle.
Don't you know that The Media is biased against him..? All the media! Everywhere!!

(Or people who get paid to think seriously about this kind of thing have thought seriously and whatever leaning they might have they still can't come up with enough good news stories in his favour... One or the other.)


Sometimes bias is earned.

User avatar
ahammel
My Little Cabbage
Posts: 2135
Joined: Mon Jan 30, 2012 12:46 am UTC
Location: Vancouver BC
Contact:

Re: 2016 US Presidential Election

Postby ahammel » Tue Oct 25, 2016 9:20 pm UTC

Very timely 538 article about voter enthusiasm.

tl;dr, the number of Clinton voters who say they're voting for her (as opposed to against Trump) is about average at 56%. On the other hand, the proportion of Trump voters who say they're voting for him as opposed to against Clinton is a record low at 41%.
He/Him/His/Alex
God damn these electric sex pants!

User avatar
gmalivuk
GNU Terry Pratchett
Posts: 25739
Joined: Wed Feb 28, 2007 6:02 pm UTC
Location: Here and There
Contact:

Re: 2016 US Presidential Election

Postby gmalivuk » Tue Oct 25, 2016 9:35 pm UTC

trpmb6 wrote:There is a perceived enthusiasm gap between Clinton and Trump due to the large rallies he holds. Whether or not that actually materializes in the election is what I am most interested to see.
As we saw with Bernie's rallies during the primaries, rally enthusiasm often doesn't translate to better election performance.

I suspect it has something to do with the fact that people enthusiastic enough to go to rallies don't get any extra votes for doing so.
Unless stated otherwise, I do not care whether a statement, by itself, constitutes a persuasive political argument. I care whether it's true.
---
If this post has math that doesn't work for you, use TeX the World for Firefox or Chrome

(he/him/his)

User avatar
CorruptUser
Posts: 8648
Joined: Fri Nov 06, 2009 10:12 pm UTC

Re: 2016 US Presidential Election

Postby CorruptUser » Tue Oct 25, 2016 11:32 pm UTC

trpmb6 wrote:It shouldn't be a surprise that white, uneducated (I hate that word, it is demeaning)


If we are going down the route of college vs non-college educated, can we please also split out the STEM and non-STEM degrees? Because I feel like someone with a degree in Engineering is going to have an opinion that is further away from someone with a degree that has the words "studies" or "arts" in them than a person with only a high school diploma.

User avatar
sardia
Posts: 5763
Joined: Sat Apr 03, 2010 3:39 am UTC

Re: 2016 US Presidential Election

Postby sardia » Wed Oct 26, 2016 12:03 am UTC

CorruptUser wrote:
trpmb6 wrote:It shouldn't be a surprise that white, uneducated (I hate that word, it is demeaning)


If we are going down the route of college vs non-college educated, can we please also split out the STEM and non-STEM degrees? Because I feel like someone with a degree in Engineering is going to have an opinion that is further away from someone with a degree that has the words "studies" or "arts" in them than a person with only a high school diploma.

Do you think the difference is solely based on their income or something? Can you google o a source? I've heard rumors it's the experience of going through college itself that causes the schism.

morriswalters
Posts: 6885
Joined: Thu Jun 03, 2010 12:21 am UTC

Re: 2016 US Presidential Election

Postby morriswalters » Wed Oct 26, 2016 12:35 am UTC

The Great Hippo wrote:Non-college graduates? I can't think of any unjumbled way to clearly and effectively say 'people who did not attend or graduate college'.
greater or lesser educational attainment.

Educational attainment refers to the highest level of education completed (e.g., a high school diploma or equivalency certificate, an associate’s degree, a bachelor’s degree, or a master’s degree). Between 1995 and 2015, educational attainment rates among 25- to 29-year-olds increased. The percentage who had received at least a high school diploma or its equivalent increased from 87 to 91 percent, with most of the change occurring between 2005 and 2015. The percentage who had completed an associate’s or higher degree increased from 33 percent in 1995 to 46 percent in 2015. Similarly, the percentage who had completed a bachelor’s or higher degree increased from 25 percent in 1995 to 36 percent in 2015, and the percentage who had completed a master’s or higher degree increased from 5 percent in 1995 to 9 percent in 2015.


Return to “News & Articles”

Who is online

Users browsing this forum: Majestic-12 [Bot] and 14 guests