HomeDiagramsDatabaseMapsForum About
     

Go Back   SkyscraperPage Forum > Regional Sections > Canada


Reply

 
Thread Tools Display Modes
     
     
  #41  
Old Posted Nov 7, 2016, 6:42 PM
geotag277 geotag277 is offline
Registered User
 
Join Date: Nov 2013
Posts: 5,091
Quote:
Originally Posted by kwoldtimer View Post
Haven't all pollsters long since adjusted their polling methods to adapt to the technology?

I have wondered whether Trump supporters are not being undercounted because they refuse to/are too embarrassed to/are too paranoid to identify themselves to pollsters as Trump voters. We'll find out tomorrow.
The major ones have, but the big issue is that the polling cannot be automated, therefore making it extremely expensive.

Even the ones that have, the era and demographics of cell phone users are changing rapidly. I'm not convinced pollsters are keeping up with their methodologies. Things have gotten progressively worse since 2008 in terms of polling.

Due to high unfavorability ratings, there may be large "silent majorities" either way.

If there was ever an election where polls are off, it would probably be this one.
Reply With Quote
     
     
  #42  
Old Posted Nov 7, 2016, 6:45 PM
hipster duck's Avatar
hipster duck hipster duck is offline
Registered User
 
Join Date: Oct 2014
Location: Toronto
Posts: 4,109
This thread really asks 2 questions:

1) Should Canada prepare for destabilization after this election?

2) Should Canada prepare for the destabilization of the United States in the future, whether in the near or very distant future?

re 1: No. Not only is the alt-right relatively small and not only are the chances that they will take up arms very low, but the people who ardently support Trump don't have demographics on their side. Not just ethnic demographics, but age demographics. Trump's supporters are mainly old, and older people, men included, generally don't incite or participate in violent acts, especially if they're likely to be outnumbered.

Now, if Trump supporters were unemployed young men, and if the United States' age pyramid resembled a Middle Eastern or Sub-Saharan African country where there were far more people under 30 than over 50, I'd be a bit more scared.

re 2: It wouldn't hurt us to try to reduce our dependence on the US, even when our relationship is good, and even if the likelihood of the US unraveling is very slim.

For starters, we really should diversify who we export to, even though the US continues to be the world's biggest single market and one that is on (for now) good economic footing.
Reply With Quote
     
     
  #43  
Old Posted Nov 7, 2016, 7:16 PM
eternallyme eternallyme is offline
Registered User
 
Join Date: Mar 2009
Posts: 5,243
Quote:
Originally Posted by blueandgoldguy View Post
I think at worst there will be a few protests with a few skirmishes thrown in for good measure as a few extreme supporters of Trump become emboldened once he loses.

The Koch brothers will do their darnedest to ensure their chosen politicians are elected so that the Repubs have controlled of the house and/or the senate...leading to further gridlock as it was for much of the Obama administration..so business as usual.
The Koch Brothers haven't given a cent to the Trump campaign, and have openly opposed him.
Reply With Quote
     
     
  #44  
Old Posted Nov 7, 2016, 7:21 PM
lubicon's Avatar
lubicon lubicon is offline
Suburban dweller
 
Join Date: Jun 2003
Location: Calgary - our road planners are as bad as yours Edmonton
Posts: 5,047
Any government that is doing its job should be preparing for these situations all the time, if not then they are not doing their main duty which is keeping their citizens safe. That said, this elections isn't going to be any different than others and nothing is going to happen. I recall countless elections recently where people were predicting chaos, mass exodus' to Canada etc. if so and so won and it doesn't come to pass.
__________________
Two things are infinite: the universe and human stupidity; and I'm not sure about the the universe.

Albert Einstein
Reply With Quote
     
     
  #45  
Old Posted Nov 7, 2016, 7:25 PM
eternallyme eternallyme is offline
Registered User
 
Join Date: Mar 2009
Posts: 5,243
Regarding US regional differences, they tend to be overrated. Even in blue states, rural areas are usually very conservative (with the possible exception of New England). Meanwhile, even in red states, major cities are usually quite liberal (good examples being Dallas, Houston, Atlanta, New Orleans, Nashville and St. Louis).
Reply With Quote
     
     
  #46  
Old Posted Nov 7, 2016, 7:28 PM
eternallyme eternallyme is offline
Registered User
 
Join Date: Mar 2009
Posts: 5,243
Quote:
Originally Posted by lio45 View Post
Exactly. Couldn't agree more.

Regardless of the exact makeup of Congress for the next two years and then for the next other two years, and regardless of who of Trump or HRC is POTUS for the next four years, that POTUS will definitely NOT have an easy time with Congress -- he or she is already very disliked, either very intensely by the GOP Reps and Senators in her case (a bloc that is basically guaranteed to control the House), and by probably nearly everyone (including lots within his own party) in his.

So, expect the status quo, and political gridlock. "Business as usual" is correct.

(For the record, I'm pretty fine with the status quo, generally speaking.)
Even if the Democrats regain the Senate (House is almost certain to stay in GOP hands), it will probably flip back to the Republicans after the 2018 midterms. The Democrats have 25 Senate seats up for election in 2018, of which several are in quite red states. The Republicans have only 9 such seats, which are mostly in very red states. That means from 2018 to 2020, the GOP should control Congress again, possibly by a solid margin.
Reply With Quote
     
     
  #47  
Old Posted Nov 7, 2016, 7:32 PM
whatnext whatnext is offline
Registered User
 
Join Date: Feb 2009
Location: Vancouver
Posts: 22,111
Quote:
Originally Posted by geotag277 View Post
There are several good reasons to believe the polls this year are way off compared to previous elections as well.

There are a high number of undecided voters.

Both candidates have very high unfavorability ratings.

As more people switch to cell phone usage exclusively, traditional polls will be less and less accurate. Even in 2012 Obama far outperformed his polling. It wasn't that accurate.

Tomorrow will be interesting.
IMHO a lot of people polled are not going to want to admit voting for Trump.
Reply With Quote
     
     
  #48  
Old Posted Nov 7, 2016, 7:59 PM
jmt18325's Avatar
jmt18325 jmt18325 is offline
Heart of the Continent
 
Join Date: Apr 2009
Location: Winnipeg
Posts: 7,284
Quote:
Originally Posted by geotag277 View Post
Of course, the correct way to interpret the 538 results here is that 538 gives her a chance slightly better than a coin flip of winning those states...
Sure, but like 308 here, they were accurate last time to almost the electoral college vote. Besides, Clinton can win without all of the states. Trump...not so much.
Reply With Quote
     
     
  #49  
Old Posted Nov 7, 2016, 8:02 PM
jmt18325's Avatar
jmt18325 jmt18325 is offline
Heart of the Continent
 
Join Date: Apr 2009
Location: Winnipeg
Posts: 7,284
Quote:
Originally Posted by geotag277 View Post
There are several good reasons to believe the polls this year are way off compared to previous elections as well.

There are a high number of undecided voters.

Both candidates have very high unfavorability ratings.

As more people switch to cell phone usage exclusively, traditional polls will be less and less accurate. Even in 2012 Obama far outperformed his polling. It wasn't that accurate.

Tomorrow will be interesting.
Most of that (other than the sudden Trudeaumania) could have been said about the last election here - of course 308 proved that polls, in the aggregate, can largely be trusted.
Reply With Quote
     
     
  #50  
Old Posted Nov 7, 2016, 8:03 PM
jmt18325's Avatar
jmt18325 jmt18325 is offline
Heart of the Continent
 
Join Date: Apr 2009
Location: Winnipeg
Posts: 7,284
By the way, as more polls are released today, it's confirmed my hypothesis - according to fivethirtyeight, Clinton has an almost 70% of winning now, up 5% since yesterday.
Reply With Quote
     
     
  #51  
Old Posted Nov 7, 2016, 8:05 PM
jmt18325's Avatar
jmt18325 jmt18325 is offline
Heart of the Continent
 
Join Date: Apr 2009
Location: Winnipeg
Posts: 7,284
Quote:
Originally Posted by whatnext View Post
They also thought Remain would win in the UK.

I don't think a poll aggregate was used in the case.

The individual polls are largely useless on their own.
Reply With Quote
     
     
  #52  
Old Posted Nov 7, 2016, 9:56 PM
eemy's Avatar
eemy eemy is offline
Closed account
 
Join Date: Aug 2006
Posts: 4,456
This article suggests that there is something wonky about Nate Silver's methods, likely his use of t-distributions over normal distributions. It's a worthwhile read, though the author is clearly not a neutral observer.

What's Wrong With 538?
Reply With Quote
     
     
  #53  
Old Posted Nov 7, 2016, 10:43 PM
geotag277 geotag277 is offline
Registered User
 
Join Date: Nov 2013
Posts: 5,091
Probability is not an easy thing to wrap your head around. I took several university level courses in statistics, and while I don't know everything about the field, I understand that if you don't have a serious background in modelling and statistics you won't be able to say much intelligent about criticizing these types of probabilistic models. I didn't see much qualifications for this Evan Cohen guy, besides the fact he is a CFA and seems to think he knows more about statistics than he thinks he does.

I found that article to really lack an understanding of what Nate's model is really telling you. Calling a model "irrational" and saying at a high level "the model thinks there is a 1% chance of such and such happening, therefore it is absurd" is not really understanding the very basics of probabilistic modeling.

While everyone was happy in 2012 that a lot of probabilistic poll aggregators correctly predicted the results (in all 50 states no less) - if you look at the details, their margins of error were very high, and in every case a number of states fell out of their model's margin of error (2 in the case of 538).

Given what the model is telling us right now, Trump could win, and there wouldn't even be a state out of the margin of error - that means even if every poll is accurate to within their margin of error, the aggregator is telling you a Trump victory is still possible. And it's highly unlikely that the polls are within their margin of error, and it's very unlikely in this election cycle that 538 will be within it's margin of error in every state.
Reply With Quote
     
     
  #54  
Old Posted Nov 7, 2016, 11:03 PM
SteelTown's Avatar
SteelTown SteelTown is online now
It's Hammer Time
 
Join Date: Mar 2004
Location: Hamilton
Posts: 19,872
Kinda reminds me of recent South Park episode, Canada built a wall to keep the Americans out


http://balanceoffood.typepad.com/can...he-us-out.html
Reply With Quote
     
     
  #55  
Old Posted Nov 7, 2016, 11:23 PM
1overcosc's Avatar
1overcosc 1overcosc is offline
Registered User
 
Join Date: Jun 2013
Location: Kingston, Ontario
Posts: 11,452
Quote:
Originally Posted by jeremy_haak View Post
The discussion this morning on CBC was that the Latin-American vote in the advance polls in Nevada is far beyond what pollsters were predicting leading into the election. Nevada has been predicted as solidly behind Trump, but if the advance polls are a reflection, the guest on CBC was saying that it would be almost impossible for him to carry the state. I suspect that the unusual nature of this election may make polling less reliable than in previous elections; I guess we'll know for sure Monday morning though!
Advance polling data in Nevada tends to be very predicative of how the state will go. Because Nevada has full in-person voting every day for weeks before the election, the majority of Nevada's ballots are cast before the election. And they release detailed data of advance poll turnout which makes it easy to more or less predict who won Nevada even before the election happens.

Nate Silver, the 538 guy, noted in a commentary article that his model's inability to factor in advance polling data means that his model predicts Nevada as a tight; but given advance polling data, Clinton has probably already won Nevada... and his model noted that Trump's chances of winning without Nevada are only 9%.

Hispanic turnout is also a factor to watch in Arizona. The large Hispanic population in Arizona, combined with the high turnout and anti-Trump anger among Hispanics, may very well allow Clinton to win Arizona. People have been predicting for the past decade that Arizona is destined to transform from a red state into a blue state in the future due to growing Hispanic population (Arizona is expected to become majority Hispanic within a few decades--Hispanics are already a majority among babies born in the state) and migration to the state from the Northeast. It appears the arrival of Trumpism may have hastened that transition.
Reply With Quote
     
     
  #56  
Old Posted Nov 7, 2016, 11:29 PM
eternallyme eternallyme is offline
Registered User
 
Join Date: Mar 2009
Posts: 5,243
Quote:
Originally Posted by 1overcosc View Post
Advance polling data in Nevada tends to be very predicative of how the state will go. Because Nevada has full in-person voting every day for weeks before the election, the majority of Nevada's ballots are cast before the election. And they release detailed data of advance poll turnout which makes it easy to more or less predict who won Nevada even before the election happens.

Nate Silver, the 538 guy, noted in a commentary article that his model's inability to factor in advance polling data means that his model predicts Nevada as a tight; but given advance polling data, Clinton has probably already won Nevada... and his model noted that Trump's chances of winning without Nevada are only 9%.

Hispanic turnout is also a factor to watch in Arizona. The large Hispanic population in Arizona, combined with the high turnout and anti-Trump anger among Hispanics, may very well allow Clinton to win Arizona.
Arizona is going to be one of the toughest states to predict. It has competing factors - the retirees there are VERY pro-Trump on nationalist grounds. But the Hispanic population and the younger people (there aren't that many young white liberals there though) go head to head with them a lot. The oddity in AZ is that the most populated area (around Phoenix) is the most conservative, which is entirely due to the seniors that live there. Tucson is far more liberal - in fact there have been movements to break off SE Arizona into a separate state (which would likely be blue leaning).

Utah is the hardest to predict - the question is did the LDS voters bite the bullet and stick to Trump just to avoid Clinton winning on a 3-way vote split? That has shades of recent Canadian elections there...
Reply With Quote
     
     
  #57  
Old Posted Nov 7, 2016, 11:41 PM
geotag277 geotag277 is offline
Registered User
 
Join Date: Nov 2013
Posts: 5,091
Quote:
Originally Posted by 1overcosc View Post
given advance polling data, Clinton has probably already won Nevada...
That's not even close to true. All early voting data tells you is that registered Democrats have voted +5% over registered Republicans. It doesn't tell you who they voted for, and I remind you that both Clinton and Trump have historically high unfavorability ratings. There is no guarantees about how registered Democrats or registered Republicans are actually voting.

Furthermore, compared to 2012, registered Democrats were +7% in Nevada above registered Republicans. This year, they are only +5%, which could signal relative voter apathy among Democrats, and may also signal relative dissatisfaction with Clinton as the nominee.

Finally, a full 20%+ of the early vote is independents, and we have no idea how they are voting.

There are some who want to claim early victory in Nevada for Clinton, but that is far from the case. We won't really know until tomorrow.
Reply With Quote
     
     
  #58  
Old Posted Nov 8, 2016, 12:30 AM
1overcosc's Avatar
1overcosc 1overcosc is offline
Registered User
 
Join Date: Jun 2013
Location: Kingston, Ontario
Posts: 11,452
Quote:
Originally Posted by geotag277 View Post
That's not even close to true. All early voting data tells you is that registered Democrats have voted +5% over registered Republicans. It doesn't tell you who they voted for, and I remind you that both Clinton and Trump have historically high unfavorability ratings. There is no guarantees about how registered Democrats or registered Republicans are actually voting.

Furthermore, compared to 2012, registered Democrats were +7% in Nevada above registered Republicans. This year, they are only +5%, which could signal relative voter apathy among Democrats, and may also signal relative dissatisfaction with Clinton as the nominee.

Finally, a full 20%+ of the early vote is independents, and we have no idea how they are voting.

There are some who want to claim early victory in Nevada for Clinton, but that is far from the case. We won't really know until tomorrow.
Party affiliation data is one thing but there's also calculations made (don't know if that's direct-from-data or an extrapolation from data by municipality) that Hispanic turnout in Nevada is way up from 2012, which is a strong sign of Clinton strength. And Nevada has a history of polls underestimating the impact of Hispanic votes--a similar thing happened in the 2010 Senate election when Democrat Harry Reid won Nevada against expectations.
Reply With Quote
     
     
  #59  
Old Posted Nov 8, 2016, 12:32 AM
1overcosc's Avatar
1overcosc 1overcosc is offline
Registered User
 
Join Date: Jun 2013
Location: Kingston, Ontario
Posts: 11,452
Quote:
Originally Posted by eternallyme View Post
Utah is the hardest to predict - the question is did the LDS voters bite the bullet and stick to Trump just to avoid Clinton winning on a 3-way vote split? That has shades of recent Canadian elections there...
The 3-way race that has emerged in Utah is interesting and it will be cool to see how factors like strategic voting possibly play out.. that's something Americans are not used to dealing with.
Reply With Quote
     
     
  #60  
Old Posted Nov 8, 2016, 1:17 AM
SignalHillHiker's Avatar
SignalHillHiker SignalHillHiker is online now
I ♣ Baby Seals
 
Join Date: Jul 2012
Location: Sin Jaaawnz, Newf'nland
Posts: 34,658
Latest poll of Canadians shows 76% would vote for Clinton.

Regionally, the prairies are the most pro-Trump.

http://www.ipsos-na.com/news-polls/p...e.aspx?id=7459
__________________
Note to self: "The plural of anecdote is not evidence."
Reply With Quote
     
     
This discussion thread continues

Use the page links to the lower-right to go to the next page for additional posts
 
 
Reply

Go Back   SkyscraperPage Forum > Regional Sections > Canada
Forum Jump


Thread Tools
Display Modes

Forum Jump


All times are GMT. The time now is 11:01 PM.

     
SkyscraperPage.com - Archive - Privacy Statement - Top

Powered by vBulletin® Version 3.8.7
Copyright ©2000 - 2024, vBulletin Solutions, Inc.