• 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 2017 |OT6| Made this thread during Harvey because the ratings would be higher

Status
Not open for further replies.
The day-to-day Trump shitshow is so bad this week that we didn't nab a single 'Trump: We beat ISIS because I changed the military' post in this thread (that I saw this afternoon).
 
http://www.foxnews.com/politics/2017/10/17/fox-news-poll-alabama-senate-race-all-tied-up.html

1508270701163.png



A Fox News Poll also finds that among just the 53 percent of Alabama registered voters who are extremely or very interested in the race, Jones has a one-point edge over Moore (46-45 percent).

lol. i'd die if we won a seat in Alabama.
 
Aren't these polls anonymous? Why would people be "scared" of putting down their actual choice? The general election, I thought, showed clear tightening towards the end and especially after the FBI letter. They ended up being within the margin or error I thought? It's not like final polls had Clinton up 10 points then she lost in reality.
 

mlclmtckr

Banned
Aren't these polls anonymous? Why would people be "scared" of putting down their actual choice? The general election, I thought, showed clear tightening towards the end and especially after the FBI letter. They ended up being within the margin or error I thought? It's not like final polls had Clinton up 10 points then she lost in reality.

Phone polls regularly do underestimate racist candidates' support, the idea being that the respondent really doesn't want to admit they're racist out loud to a stranger over the phone

Actually in races like that asking "who do you think will win" produces a more accurate result than "who are you going to vote for."

Edit - I don't know about the polls right at the end there but if it had been within the margin of error why were the aggregators all putting Clinton at like 99%?
 
If it's an actual person asking the question they can't bring themselves to admit they're voting the way they are.
I see, fair enough.
That's FOX sounding an alarm to nationalize the race. I suspect Moore will take this as his cue to start banging the abortion and scary black people drum. Actually, those are the only two drums necessary in Alabama (and many other red states)

It's a senate race and, IIRC, Biden has already been participating. It's already national and Moore has already been saying dumb shit. I don't that changes much.
 
That's FOX sounding an alarm to nationalize the race. I suspect Moore will take this as his cue to start banging the abortion and scary black people drum. Actually, those are the only two drums necessary in Alabama (and many other red states)

Nope. Fox News polling is generally very good.
 

pigeon

Banned
Edit - I don't know about the polls right at the end there but if it had been within the margin of error why were the aggregators all putting Clinton at like 99%?

Because Clinton won the popular vote by a large margin and the polls accurately predicted that she would do so. Usually that means you'll win the election! The real problem here was on the individual state level, where the polls were not as good, and Clinton's vote gains were in states where she was either guaranteed to win (CA) or nearly guaranteed to lose (AZ, TX) while her vote losses were in swing states.
 

Wilsongt

Member
Excuse me as I roll my eyes again that On Point had Erick fucking Erickson on today. Yeah, he had a sob story, but fucker did a ton of irreversible damage.
 

mlclmtckr

Banned
Because Clinton won the popular vote by a large margin and the polls accurately predicted that she would do so. Usually that means you'll win the election! The real problem here was on the individual state level, where the polls were not as good, and Clinton's vote gains were in states where she was either guaranteed to win (CA) or nearly guaranteed to lose (AZ, TX) while her vote losses were in swing states.

Hmm in that case I'd say the problem is also with the models themselves, for assuming that a popular vote victory translated to an EC victory.
 

The Technomancer

card-carrying scientician
Hmm in that case I'd say the problem is also with the models themselves, for assuming that a popular vote victory translated to an EC victory.

A lot of 2016 showed that a lot of things were very screwed up but the biggest thing really does come back to how wrong everyone was about undecideds and how they broke
 

Y2Kev

TLG Fan Caretaker Est. 2009
NJ Gov:

Murphy (D): 47%
Guadango (R): 33%

Also Fox News

lol poor Kim
It’s not even rally her fault. I mean she’s awful. But Christie left such a stink behind him that it could be Jesus Christ on the ticket and he’d have no shot.

Kim must be a Nintendo Switch cause ha fans are port vote begging
 

pigeon

Banned
Hmm in that case I'd say the problem is also with the models themselves, for assuming that a popular vote victory translated to an EC victory.

Well, at some level, it does, right? Like, you can't lose the popular vote by that much and still win the election. Trump was pretty much on the knife's edge.

The models that did state-by-state modeling also failed in general, mostly because they handled undecided voters poorly.

I personally think that 538 was once again the best performing model, 70/30 was about the correct odds for that election, and Trump was lucky to win. We live in a shitty timeline.
 

Wilsongt

Member
While Trump attacks McCain...

Barack Obama
@BarackObama
I'm grateful to @SenJohnMcCain for his lifetime of service to our country. Congratulations, John, on receiving this year's Liberty Medal.
6:59 PM · Oct 16, 2017
 
Sam Wang (Princeton) gave Clinton a 99% chance, Nate Cohn (NYT) gave her 85%, and Nate Silver (538) gave her 70%. All told, I think Silver did quite well. Any sensible model should have had Clinton ahead, but his model got two key things correct: one was the uncertainty indicated by having a high number of undecideds, and another was the correlation of state level results.

On a national level the polling error wasn’t that bad (late polls were mostly showing Clinton+3 and Clinton+4, actual result was Clinton+2). Obviously the polls were way off in MI/PA/WI.
 

Y2Kev

TLG Fan Caretaker Est. 2009
Sam Wang (Princeton) gave Clinton a 99% chance, Nate Cohn (NYT) gave her 85%, and Nate Silver (538) gave her 70%. All told, I think Silver did quite well. Any sensible model should have had Clinton ahead, but his model got two key things correct: one was the uncertainty indicated by having a high number of undecideds, and another was the correlation of state level results.

On a national level the polling error wasn’t that bad (late polls were mostly showing Clinton+3 and Clinton+4, actual result was Clinton+2). Obviously the polls were way off in MI/PA/WI.

What day is it? Why does it still hurt so much
 

studyguy

Member
MIC reporting green card holders might be banned from enlisting in the military for now?

https://mic.com/articles/185297/exc...-a-move-that-may-break-federal-law#.l4P7STVPO
In the past, green card holders were allowed to enlist and ship off to basic training so long as a background check had already been started. Reports said the new policy would delay green card holders who enlisted in the military from shipping out to basic training for at least a year, thanks to a backlog of background checks.

However, the email from Williamson states green card holders cannot even get to the background check process, as they are now barred from enlisting “until further notice.”

Barring green card holders from enlisting in the military is against federal law, which states that an “alien who is lawfully admitted for permanent residence” may be enlisted in “any armed force.”

Seems dumb. Lost on transgender military issues, guess it's time for another L
 

UraMallas

Member
Well, at some level, it does, right? Like, you can't lose the popular vote by that much and still win the election. Trump was pretty much on the knife's edge.

The models that did state-by-state modeling also failed in general, mostly because they handled undecided voters poorly.

I personally think that 538 was once again the best performing model, 70/30 was about the correct odds for that election, and Trump was lucky to win. We live in a shitty timeline.

Agreed. I was shitting on him for stating what I see in hindsight as some data-driven truths. Definitely humbled me a bit because I was just so sure. Silver was like "hold up there, bud" and I wasn't having it.
 

DTC

Member
I don't believe that Doug Jones is anywhere better than -10 vs Moore. Moore and Donald Trump have such high favorables in most Alabama polls. And it's Alabama.
 
Well, at some level, it does, right? Like, you can't lose the popular vote by that much and still win the election. Trump was pretty much on the knife's edge.

Practically, yes. In theory, you could win the popular vote 500K to 50, but if all of your votes were in Wyoming while your opponent got one vote in the other 49 states (plus DC) then you'd lose the EV 535-3.

2000 raised the question, but since it was essentially a tied race, people let it go that one vote fell one way and the other vote fell the other way. 2016 kinda brought the question back though. How far apart could a potential election be between popular vote and EC? Further than we thought in 2000.
 

Y2Kev

TLG Fan Caretaker Est. 2009
I mean if anything it gave us a good benchmark for 2020. If the panhandle closing shows massive racist fuck turnout then we can turn it off early. Better pick an Attractive White Man.
 
Status
Not open for further replies.
Top Bottom