US 2012 Presidential Election Polls

Manic Moran · 05-11-2012 10:23pm

I'm probably going to watch CNN or BBC. Either way, it'll be on the Internet.

nagilum2 · 05-11-2012 10:50pm

Inquitus wrote: »

A few primary polls versus a full election data set, you are comparing apples with delusional pumpkins.

This isn't 1996 when everyone knew Clinton would be relelected.

Anyone who thinks this is in the bag for either side is crazy. Everything says this is going to be the closest election since 2000. And there are real question marks with the underlying demographic sample that some of the polls are using as to whether there's a systematic oversampling of democrats.

The bottom line is no one will know until the day after tomorrow - perhaps not even then.

southsiderosie · 05-11-2012 10:51pm

nagilum2 wrote: »

The bottom line is no one will know until the day after tomorrow - perhaps not even then.

Yes, I wonder about this. I would be very surprised if they call the election tomorrow night.

That said, I'm putting my faith in Nate Silver, if for no other reason than to preserve my sanity for the next 48 hours.

05-11-2012 10:52pm

Ponster wrote: »

It wouldn't be the first time. In 1980 Carter the polls predicted a win by Carter although it was "too close to call" and "neck and neck". Reagan went on to win by a landslide taking 43 states.

That's odd. My recollection of the time was that a lot of us figured that Reagan was going to win ahead of the election. In fact, as I recall it, Carter started to fade badly in polls and "media perception" in the weeks after the disastrous attempt at a military rescue in the Iran hostage crisis.

But that clearly doesn't square with the story above - and as I look around on the web I see that the story has quite a degree of acceptance with people. It also seems to be a source of some comfort to Romney supporters.

And yet, until I saw the above post a short while ago and stuck some search terms into Google, I'd never have considered the 1980 election as a more modern version of "Dewey Defeats Truman". To repeat, as a teenager in 1980 with an interest in politics and elections, Reagan's win came as no surprise.

Maybe time has dulled my memory - I ain't as young as I used to be. But I did locate a link that supports, to some degree, what I thought happened. I offer it here as an alternative view.

http://themonkeycage.org/blog/2012/08/09/what-really-happened-in-the-1980-presidential-campaign/

Just saying.

nagilum2 · 05-11-2012 11:00pm

Deleted User wrote: »

That's odd. My recollection of the time was that a lot of us figured that Reagan was going to win ahead of the election. In fact, as I recall it, Carter started to fade badly in polls and "media perception" in the weeks after the disastrous attempt at a military rescue in the Iran hostage crisis.

But that clearly doesn't square with the story above - and as I look around on the web I see that the story has quite a degree of acceptance with people. It also seems to be a source of some comfort to Romney supporters.

And yet, until I saw the above post a short while ago and stuck some search terms into Google, I'd never have considered the 1980 election as a more modern version of "Dewey Defeats Truman". To repeat, as a teenager in 1980 with an interest in politics and elections, Reagan's win came as no surprise.

Maybe time has dulled my memory - I ain't as young as I used to be. But I did locate a link that supports, to some degree, what I thought happened. I offer it here as an alternative view.

http://themonkeycage.org/blog/2012/08/09/what-really-happened-in-the-1980-presidential-campaign/

Just saying.

I can't imagine that there could be a real landslide on either side this go around. The country is just too polarized and there doesn't seem to be great momentum or enthusiasm with either candidate. I'm sticking with my original projection of an EC - popular vote split, with Obama barely winning vie the electoral college.

Black Swan · 06-11-2012 12:23am

southsiderosie wrote: »

Yes, I wonder about this. I would be very surprised if they call the election tomorrow night.

If there is any substance to the current very close polling estimates, there is a very good chance that the presidential election will not be decided Tuesday. It's doubtful that there will be the huge delay as experienced during the 2000 election, but perhaps some delay will result due to election irregularities claimed by the losing party in one or more swing states (e.g., Ohio, etc.).

Jesus Wept · 06-11-2012 1:22am

Amerika wrote: »

So when Romney wins tomorrow, Paddy Power will have to pay out again?

They paying out??

Ozymandius2011 · 06-11-2012 1:30am

NV turnout has passed that of 2008. It is 72% compared to 67% in 2008. The Dems are holding their 7% lead there and on a 70% turnout, the GOP are holding there's in Colorado. In NV, the electorate is 54% Non Hispanic White, compared to 70% in CO.

joseph brand · 06-11-2012 1:52am

Paddy Power:
Obama 2/9
Romney 3/1

If I stick a tenner on Obama I stand to come away with €12.22 /rubs hands

nagilum2 · 06-11-2012 1:59am

Ozymandius2011 wrote: »

NV turnout has passed that of 2008. It is 72% compared to 67% in 2008. The Dems are holding their 7% lead there and on a 70% turnout, the GOP are holding there's in Colorado. In NV, the electorate is 54% Non Hispanic White, compared to 70% in CO.

Interesting. The conventional wisdom is that if the democrats aren't leading early voting then they're unlikely to win the state, which would move Colorado to Romney. I'm not sure how realistic that is though.

I think the NV numbers sync with both candidates not really contesting NV too much in recent days. You can tell where both candidates are worried by where they are travelling.
Romney in Florida and Virginia on the last day tells me the republican internals say those states may not be quite the lock everyone thinks they are. Obama in Wisconsin and Iowa on the last day tells me the the Democratic internals tell them the same thing about those two states.

It's almost like the candidates themselves don't even quite know what to believe, which says we're in for a long night or maybe days.

dttq · 06-11-2012 3:21am

What day and time will be know who has won the US election?

Ozymandius2011 · 06-11-2012 3:23am

Fivethirtyeight.com now placing Florida in the Obama column with a 52.5 probability.

Dttq in answer to your question we may or may not know for weeks depending on recounts, lawyers etc. If it's a blowout we should know early hours of morning. But there are likely to be many "Provisional (unverified) ballots" and in OH, they can't be counted until November 17th. In OH a recount follows automatically if the margin is 0.025% or less. 1% and a candidate can ask for one.

nagilum2 · 06-11-2012 3:37am

Ozymandius2011 wrote: »

Fivethirtyeight.com now placing Florida in the Obama column with a 52.5 probability.

Dttq in answer to your question we may or may not know for weeks depending on recounts, lawyers etc. If it's a blowout we should know early hours of morning. But there are likely to be many "Provisional (unverified) ballots" and in OH, they can't be counted until November 17th. In OH a recount follows automatically if the margin is 0.025% or less. 1% and a candidate can ask for one.

Is Silver off his rocker on FL? The RCP average (esp the most recent FL polls) do not seem to support that at all.

nagilum2 · 06-11-2012 3:54am

joseph brand wrote: »

Paddy Power:
Obama 2/9
Romney 3/1

If I stick a tenner on Obama I stand to come away with €12.22 /rubs hands

I thought they already paid out? Paddy Power confuses me.

Ozymandius2011 · 06-11-2012 3:58am

nagilum2 wrote: »

Is Silver off his rocker on FL? The RCP average (esp the most recent FL polls) do not seem to support that at all.

Well the Registered Dems are around 4% ahead in the early vote on a 52% turnout. Around 4.5 million people have voted early.

4,469,393
Party Reg
Dem 42.9%
Rep 39.1%
None/Oth 18.0%
Method
In-person 54.0%
Mail 46.0%
52.9%

nagilum2 · 06-11-2012 4:16am

Ozymandius2011 wrote: »

Well the Registered Dems are around 4% ahead in the early vote on a 52% turnout. Around 4.5 million people have voted early.

4,469,393
Party Reg
Dem 42.9%
Rep 39.1%
None/Oth 18.0%
Method
In-person 54.0%
Mail 46.0%
52.9%

That's a 3.8 spread - weak early voting figures for democrats.

In 2008, that spread for Obama was 9 points. That's Obama performing at -5 what he was four years ago, in a state that he only won by 2.8 points.
LINK: http://www.washingtonpost.com/blogs/the-fix/wp/2012/11/05/final-early-vote-numbers-suggest-a-very-close-race/

I think Silver is a great number cruncher, but he will get this one wrong. Bank on it.

If Paddy Power offered me odds on Romney taking Florida, I'd put A LOT on it.

southsiderosie · 06-11-2012 4:46am

nagilum2 wrote: »

That's a 3.8 spread - weak early voting figures for democrats.

In 2008, that spread for Obama was 9 points. That's Obama performing at -5 what he was four years ago, in a state that he only won by 2.8 points.
LINK: http://www.washingtonpost.com/blogs/the-fix/wp/2012/11/05/final-early-vote-numbers-suggest-a-very-close-race/

I think Silver is a great number cruncher, but he will get this one wrong. Bank on it.

If Paddy Power offered me odds on Romney taking Florida, I'd put A LOT on it.

The early voting period in Florida was shorter this election cycle than it was in 2008.

Ponster · 06-11-2012 4:47am

nagilum2 wrote: »

If Paddy Power offered me odds on Romney taking Florida, I'd put A LOT on it.

I think Silver is wrong here but I think that it's perhaps a lot closer in FL than you think. According to almost most polls Romney is leading FL by less than Obama is leading OH.

nagilum2 · 06-11-2012 4:48am

southsiderosie wrote: »

The early voting period in Florida was shorter this election cycle than it was in 2008.

The length of the period should not affect the ratio, only the volume, assuming uniform voting distribution throughout the period in question.

nagilum2 · 06-11-2012 5:23am

Ponster wrote: »

I think Silver is wrong here but I think that it's perhaps a lot closer in FL than you think. According to almost most polls Romney is leading FL by less than Obama is leading OH.

I'm sure it's close - Romney wouldn't be there on the last day otherwise - but I'd still play those odds above much else in this election.

I thought the silver model at 65% Obama was about right. He now has it at a 92% chance, though, and that feels like a far too aggressive probability to me.

If someone gave me Romney at 9/1, I'd definitely take that at this point!

southsiderosie · 06-11-2012 5:52am

nagilum2 wrote: »

The length of the period should not affect the ratio, only the volume, assuming uniform voting distribution throughout the period in question.

I don't think the volume is comparable because IIRC a Sunday was dropped - therefore, black churches missed an opportunity to do a 'Souls to the Polls' event, which is a disproportionately heavy voting day.

Duck Soup · 06-11-2012 12:51pm

Without veering into the voter suppression thread territory, the whole reason Republican legislatures - and only Republican legislatures - have been trying to curtail early voting is because it's used disproportionately by Democratic voters. And some days and times within the early voting periods are more heavily Democratic-loaded than others.

For example, if you cut off the weekend for early voting, lower income voters who cannot afford to take a weekday (or part thereof) off to vote may not be able to vote on the Tuesday. Plus as Rosie said, the Souls To The Polls drive on Sundays, used predominantly by African-American churches, is taken out of the equation.

You can also take more low income voters out of mix by curtailing voting during after-work hours on weekdays. So if, for example, you allowed early voting 10am-10pm in 2008 and now only allow it 10am-6pm, you will lose low income voters wanting to vote but not lose wages.

In a nutshell, if it didn't work at curbing the opposition's vote, they wouldn't do it.

Duck Soup · 06-11-2012 1:05pm

The final national polls before voting starts today.

seligehgit · 06-11-2012 6:10pm

Inquitus wrote: »

GOP tears are delicious, if you aren't a right wing religious whackjob, watching Obama win on Fox is tasty.

Any other right wing news stations you'd recommend that are on freesat?Is Fox available on line?

Amerika · 06-11-2012 6:19pm

seligehgit wrote: »

Any other right wing news stations you'd recommend that are on freesat?Is Fox available on line?

Actually a lot of us GOPers are planning on watching MSNBC tonight in order to see their pundits go into full mental meltdown.

Black Swan · 06-11-2012 7:41pm

Duck Soup wrote: »

The final national polls before voting starts today.

Most of these polls fall within a confidence interval of plus or minus 3, so if they fail to predict the election, later they may claim that they were not off, given that the reported numbers fell within this interval.

Ozymandius2011 · 07-11-2012 12:28am

CNN exit poll:
Democrats 37%
Republicans 34%
Independents 29%

Dems: Obama 92, Romney 7
Reps: Romney 92, Obama 5

Conservatives: Romney 85, Obama 15

Urban people: 62-38 for Obama
Rural people: 62-38 for Romney.

Independents: 60-40 for Romney.

---
The exit poll debunks Rasmussen polls that say there are 1% more Reps than Dems.

---
CNN (Ohio) ezit poll:
Getting better 37%
Worse 33%

--
Better on economy:
Romney 48
Obama 47

---
Virginia:
Economy: romney 53, Obama 45
Satisfied+Enthusiastic = 50%
Angry+ Dissatisfied = 48%

TwoShedsJackson · 07-11-2012 1:55pm

So let's see how the only pollsters Amerika trusts, Rasmussen, got on with their polling predictions:

Nationally, Rasmussen polled at 49%-48%. The actual result was (so far) 50%-49% Obama, the reverse of Rasmussen's poll.

In Colorado, Rasmussen polled at 50%-47% for Romney. The actual result was 51%-47% for Obama, the reverse of Rasmussen's poll.

In Florida, Rasmussen polled at 50%-48% for Romney. The actual result was 50%-49% for Obama, the reverse of Rasmussen's poll.

In Iowa, Rasmussen polled at 49%-48% for Romney. The actual result was 52%-47% for Obama, the reverse of Rasmussen's poll, doubled.

In New Hampshire, Rasmussen polled at 50%-48% for Romney. The actual result was 52%-47% for Obama, the reverse of Rasmussen's poll.

In Ohio, Rasmussen polled at a 49%-49% tie. The actual result was 50%-48% for Obama, a two-point swing.

In Virginia, Rasmussen polled at 50%-48% for Romney. The actual result was 50%-48% for Obama, the reverse of Rasmussen's poll.

In Wisconsin, Rasmussen polled at a 49%-49% tie. The actual result was 52-47% for Obama, a six-point swing.

In other words, in all the races that mattered, Rasmussen got it egregiously wrong. They didn't call a single battleground state right except for North Carolina, and even there it appears that they overestimated the margin of Romney's win.

Duck Soup · 07-11-2012 2:34pm

Rasmussen relied solely on automated calls to landlines. They might want to revisit their adherence to that model.

cristoir · 07-11-2012 2:51pm

Silver was spot on.

US 2012 Presidential Election Polls

Comments