Politics!

This forum is founded on discussions about T Campbell's work (alone and with artist partners).

Moderators: TCampbell, Gisele

Re: Politics!

Postby Valerie » Thu Nov 01, 2012 8:41 pm

I'd say the second and third debate helped Obama a lot more tham harmed him.

But mostly, I'm posting to tell you guys that trickle-down economics doesn't work. Duh.
Lia S wrote:Valerie is right.

As usual.


TCampbell wrote:Val has a harem, but it's chiefly structured online at the moment.
User avatar
Valerie
 
Posts: 3994
Joined: Wed Sep 10, 2008 11:18 pm

Re: Politics!

Postby thebitterfig » Thu Nov 01, 2012 9:29 pm

First debate was brutal. Others not so much. At least mild Team-Obama wins for all of them, with a fairly solid win at foreign policy. Granted, all they did was stop the bleeding from debate number one.

As to Intrade, I'm more inclined to trust the State Poll Aggregation models (fivethirtyeight most prominently, but many other aggregators are in about the same place), which are more 80/20. I'll take statisticians and SABRmetrics over the animal spirts of markets any day. Still, that's fairly close. Basically, flip two coins at once, if they both land tails, Romney wins.

Popular/electoral vote split is pretty unlikely too, from what I read. Nate Silver again is the go-to guy. A lot of polling shows Obama not doing doing as poorly relative to 2008 in non-swing states for a split to be likely. It'd need to be blowouts in red states, squeakers in blues, but that doesn't really look that likely. If things are close enough in blue states for Obama to lose the popular vote, he's probably losing Florida, Virginia, Ohio, Colorado, and Iowa, and thus the electoral college.

As to Kentucky... It's only 3 hours for me to get into Bourbon-country. If I wasn't well enough stocked, I'd have to go down, see if I couldn't get a few distillery-only bottlings.
The notes of this paradoxalist do not end here, however. He could not refrain from going on with them, but it seems to us that we may stop here. - Fyodor Dostoevsky, Notes from Underground (trans. C. Garnett)
User avatar
thebitterfig
 
Posts: 622
Joined: Mon Jan 09, 2012 4:25 pm
Location: Maine, where it's probably snowing.

Re: Politics!

Postby mindstalk » Thu Nov 01, 2012 10:06 pm

I think 538 is the only one to claim a probability... no, Princeton Election Consortium did too, I noticed today. Much higher ones, 90+, IIRC. PEC's tended to be optimistic, Silver to be conservative (in the cautious assumptions sense.) All the aggregators give the advantage -- expected # of electoral votes 270+ -- to Obama, because the data are numerous and not that complicated. Converting polls and their alleged margins of error and their history of being wrong into a probability seems tricky, though I guess it's what 538 is trying to do.

Either way, go vote. I did today, since I'm out of state next week.
Last edited by mindstalk on Thu Nov 01, 2012 10:28 pm, edited 1 time in total.
User avatar
mindstalk
 
Posts: 601
Joined: Mon Jan 18, 2010 10:02 am

Re: Politics!

Postby thebitterfig » Thu Nov 01, 2012 10:24 pm

Well, even if they don't often put probabilities on their charts, most state-poll aggregators have pretty similar Electoral College results.
The notes of this paradoxalist do not end here, however. He could not refrain from going on with them, but it seems to us that we may stop here. - Fyodor Dostoevsky, Notes from Underground (trans. C. Garnett)
User avatar
thebitterfig
 
Posts: 622
Joined: Mon Jan 09, 2012 4:25 pm
Location: Maine, where it's probably snowing.

Re: Politics!

Postby mindstalk » Thu Nov 01, 2012 10:41 pm

thebitterfig wrote:Well, even if they don't often put probabilities on their charts, most state-poll aggregators have pretty similar Electoral College results.


Well, yeah. If you add up all the states that lean one way or another you'll get the same results. Variety comes from discounting some polls, or assuming that not every weakly leaning state will go that way, or having a bigger "tied" category. electoral-vote counts every state as leaning unless they're in a dead tie, some other site had over 100 votes up for grabs, presumably where the poll difference was less than the margin of error or some 5% threshold or something. Which sounds sensible at first, however if you have 50 polls that all point one way, even within their individual margin of error, that probably tells you something; their aggregate margin of error is going to be less than the usual 3%.

(There's also how you do the average. electoral-vote averages the week's polls equally; I'd guess 538 does some decaying weighted average, e.g. 0.55*today + 0.45*yesterday for a dumb example.)
User avatar
mindstalk
 
Posts: 601
Joined: Mon Jan 18, 2010 10:02 am

Re: Politics!

Postby Mr. Brightside » Fri Nov 02, 2012 12:59 am

mindstalk wrote:I think 538 is the only one to claim a probability... no, Princeton Election Consortium did too, I noticed today. Much higher ones, 90+, IIRC. PEC's tended to be optimistic, Silver to be conservative (in the cautious assumptions sense.) All the aggregators give the advantage -- expected # of electoral votes 270+ -- to Obama, because the data are numerous and not that complicated. Converting polls and their alleged margins of error and their history of being wrong into a probability seems tricky, though I guess it's what 538 is trying to do.


Turnout is the big problem with any poll. Where polls vary widely between firms, it's because they're making different assumptions regarding turnout of different demographics. You can't really get an honest answer whether someone plans to vote, let alone whether they actually will. Race is an issue: on the one hand, there is a general upward trend, but Obamamania in '08, Voter ID laws, and the mobilizations of the Tea Party, and let's face it, OWS, may add up to a backslide this cycle. Also, Republicans are more enthusiastic than in 2008, and Democrats less so, albeit more than in 2010. Because of those numbers, margins of error aren't enough - basically, conservative estimates of these effects mean an Obama win, whereas a reasonable nudge to the slider gives it to Romney, without changing one respondent.

He's doing better than he was before Sandy, though.
(There has never been a signature.)
Mr. Brightside
 
Posts: 2026
Joined: Thu Jan 11, 2007 2:44 am

Re: Politics!

Postby Zanosuke Kurosaki » Fri Nov 02, 2012 1:24 am

Mr. Brightside wrote:
mindstalk wrote:I think 538 is the only one to claim a probability... no, Princeton Election Consortium did too, I noticed today. Much higher ones, 90+, IIRC. PEC's tended to be optimistic, Silver to be conservative (in the cautious assumptions sense.) All the aggregators give the advantage -- expected # of electoral votes 270+ -- to Obama, because the data are numerous and not that complicated. Converting polls and their alleged margins of error and their history of being wrong into a probability seems tricky, though I guess it's what 538 is trying to do.


Turnout is the big problem with any poll. Where polls vary widely between firms, it's because they're making different assumptions regarding turnout of different demographics. You can't really get an honest answer whether someone plans to vote, let alone whether they actually will. Race is an issue: on the one hand, there is a general upward trend, but Obamamania in '08, Voter ID laws, and the mobilizations of the Tea Party, and let's face it, OWS, may add up to a backslide this cycle. Also, Republicans are more enthusiastic than in 2008, and Democrats less so, albeit more than in 2010. Because of those numbers, margins of error aren't enough - basically, conservative estimates of these effects mean an Obama win, whereas a reasonable nudge to the slider gives it to Romney, without changing one respondent.

He's doing better than he was before Sandy, though.


Also not helped by the rather strong possibilities that the Republican party is now attempting to cheat, and seem to not care one bit that they could get caught at it. :(
Stand tall and shake the heavens.

Beep beep, I'm a jeep.
User avatar
Zanosuke Kurosaki
 
Posts: 1588
Joined: Tue Dec 07, 2010 7:32 pm
Location: Texas

Re: Politics!

Postby Alice Macher » Fri Nov 02, 2012 7:24 am

Via Facebook:


Image
"Life doesn't wait forever." --Lisa Winklemeyer
User avatar
Alice Macher
 
Posts: 4715
Joined: Wed Jun 09, 2010 4:57 pm
Location: Toronto

Re: Politics!

Postby mindstalk » Fri Nov 02, 2012 11:44 am

Mr. Brightside wrote:
mindstalk wrote:I think 538 is the only one to claim a probability... no, Princeton Election Consortium did too, I noticed today. Much higher ones, 90+, IIRC. PEC's tended to be optimistic, Silver to be conservative (in the cautious assumptions sense.) All the aggregators give the advantage -- expected # of electoral votes 270+ -- to Obama, because the data are numerous and not that complicated. Converting polls and their alleged margins of error and their history of being wrong into a probability seems tricky, though I guess it's what 538 is trying to do.


Turnout is the big problem with any poll. Where polls vary widely between firms, it's because they're making different assumptions regarding turnout of different demographics. You can't really get an honest answer whether someone plans to vote, let alone whether they actually will. Race is an issue: on the one hand, there is a general upward trend, but Obamamania in '08, Voter ID laws, and the mobilizations of the Tea Party, and let's face it, OWS, may add up to a backslide this cycle. Also, Republicans are more enthusiastic than in 2008, and Democrats less so, albeit more than in 2010. Because of those numbers, margins of error aren't enough - basically, conservative estimates of these effects mean an Obama win, whereas a reasonable nudge to the slider gives it to Romney, without changing one respondent.

He's doing better than he was before Sandy, though.


Polls are mostly of likely voters at this stage, and determining those is more sophisticated than just asking "will you vote" though even that's better than nothing. http://www.gallup.com/poll/111268/How-G ... -work.aspx for one example. Past behavior does tend to predict future behavior for humans if not for stock markets, so people who have voted, especially frequently, are good candidates.

I think Silver said he has a fudge factor for voter ID laws, though I'm not sure.

I have no idea about enthusiasm. Romney's still a Mormon and the guy who didn't so much win the nomination as fail to lose it to anyone else. Senate race troglodytes are whipping up liberal enthusiams.
User avatar
mindstalk
 
Posts: 601
Joined: Mon Jan 18, 2010 10:02 am

Re: Politics!

Postby Mr. Brightside » Fri Nov 02, 2012 12:00 pm

It's funny you should cite Gallup, since they're sort of just the ones I was alluding to: although they haven't done a state-by-state, and haven't polled since Sandy, Gallup gives Romney a 51-46 popular lead, well ahead of pretty much any prediction that has Obama winning the EC, and the furthest gap even before Sandy.
(There has never been a signature.)
Mr. Brightside
 
Posts: 2026
Joined: Thu Jan 11, 2007 2:44 am

Re: Politics!

Postby mindstalk » Fri Nov 02, 2012 12:22 pm

Mr. Brightside wrote:It's funny you should cite Gallup, since they're sort of just the ones I was alluding to: although they haven't done a state-by-state, and haven't polled since Sandy, Gallup gives Romney a 51-46 popular lead, well ahead of pretty much any prediction that has Obama winning the EC, and the furthest gap even before Sandy.


OTOH, http://www.dailykos.com/story/2012/10/1 ... -a-history

Meanwhile Rasmussen, which does likely voter screens before everyone else in the cycle, and allegedly has a pro-GOP bias, finds a dead heat: http://www.realclearpolitics.com/epolls/latest_polls/
and slightly positive net job approval for Obama.
Or see http://www.realclearpolitics.com/epolls ... -1171.html
Gallup's a big outlier.
User avatar
mindstalk
 
Posts: 601
Joined: Mon Jan 18, 2010 10:02 am

Re: Politics!

Postby Pink Freud » Fri Nov 02, 2012 1:34 pm

Yeah, Seth Meyers pretty much summed up my opinion on polls last week: "...people who still have a land line phone and pick-up calls from "Unknown Caller" on caller ID."
User avatar
Pink Freud
 
Posts: 716
Joined: Sat Feb 19, 2011 4:13 am
Location: here

Re: Politics!

Postby thebitterfig » Fri Nov 02, 2012 7:18 pm

We get to get rid of daylight savings this weekend? Fucking hell yes. They really ought to abolish it, stick with winter time all year long. Nothing I hate more than driving to work in the dark.

*edit* Not quite true... Fundamentalists are worse...
The notes of this paradoxalist do not end here, however. He could not refrain from going on with them, but it seems to us that we may stop here. - Fyodor Dostoevsky, Notes from Underground (trans. C. Garnett)
User avatar
thebitterfig
 
Posts: 622
Joined: Mon Jan 09, 2012 4:25 pm
Location: Maine, where it's probably snowing.

Re: Politics!

Postby mindstalk » Fri Nov 02, 2012 9:23 pm

If there are 6 domain experts opining about their domain, and 5 say X while 1 says Y, and you have no reason to believe any are bought out or deliberately biased, who do you bet on, the 5 or the 1?

Does you answer change if you want to believe the 1?

***

To get a 3% margin of error in a poll, you need a random sample of about 1000 people. It doesn't matter what you're sampling *from*, whether as small as Wyoming or as big as the USA (or the world!) You need 1000 people.

So a national poll is sampling 1000 people across the country. A poll about Ohio sampled 1000 people in Ohio. Ditto for all the other state polls. If there are 12 toss-up states, then in a polling cycle their polls have 12,000 people, vs. 1000 for the whole nation. And their predictive power is focused on "who will win", since the electoral college goes by state, while the national poll is 'wasting' time asking people in Texas or California how they'll vote, which is irrelevant for EC purposes; we know how those states are going to go.

Add the fact that Gallup is an outlier even among national tracking polls.

It's possible that polling is totally borked this year, across half a dozen or more companies, and Gallup is uniquely right. But it's possible that it's Gallup that's borked. Which is more likely? They're an old and familiar name in polling, but I don't know of anything that says they're uniquely qualified.

(And, actually, if polling is totally borked this year, I'd think it's more likely that it'd be borked for *everyone*, *including* Gallup, and we wouldn't know anything for sure, than that there's some horrible mistakes one company is uniquely avoiding.)
User avatar
mindstalk
 
Posts: 601
Joined: Mon Jan 18, 2010 10:02 am

Re: Politics!

Postby Adrishiana » Fri Nov 02, 2012 9:27 pm

Pink Freud wrote:Yeah, Seth Meyers pretty much summed up my opinion on polls last week: "...people who still have a land line phone and pick-up calls from "Unknown Caller" on caller ID."


Sometimes the caller ID actually says "political call" or, recently, "REPUBLICANS." I laughed (but did not answer the phone).
No more zoos!
User avatar
Adrishiana
 
Posts: 1457
Joined: Thu Jan 11, 2007 8:37 am

PreviousNext

Return to General Discussion

Who is online

Users browsing this forum: Google [Bot] and 5 guests

cron