US 2012 Presidential Election Polls

nagilum2 · 15-10-2012 3:57pm

I'm beginning to think more and more we might see Obama win an electoral college squeaker while losing the national popular vote by about a million votes.

If Obama just holds Ohio and Pennsylvania, plus either Iowa or Nevada, he wins with 271 electors barring a shocker in another state. That is true even if Romney sweeps North Carolina, Virginia, New Hampshire, Colorado and Florida. Romney's path to 270 is much tougher.

THIS:
http://elections.huffingtonpost.com/2012/custom-presidential-election-map#nkammnjannjajenje

OR:
http://elections.huffingtonpost.com/2012/custom-presidential-election-map#nkamdnjannjjjenje

Amerika · 15-10-2012 4:09pm

Hmmm… The alternative universe of the Washington Post/ABC poll, in which the D+9 live, is almost laughable at this point. And yet Obama's lead is only 3 points, within the margin of error. I’m now beginning to think RCP throws these unrealistic polls in which Democrats will descend on the voting booths curiosity of some loaves and fish miracle, just to get people talking about RealClearPolitics.

Permabear · 15-10-2012 5:12pm

This post has been deleted.

Amerika · 15-10-2012 5:39pm

Gary Johnson could very well play the spoiler for Romney in Colorado, Nevada, New Hampshire, Ohio, Virginia, Wisconsin, and maybe even Florida.

Ozymandius2011 · 15-10-2012 6:32pm

Permabear wrote: »

This post had been deleted.

Well I have heard on a Colorado news channel report I saw online that Johnson was taking away more Obama votes there....Many libertarians despise the GOP's finger-wagging brand of Christianity just as much as the Dem's perceived penchant for tax-and-spend Big Government statism.

Duck Soup · 16-10-2012 10:46am

Joe Biden seems to have done the job. Nate Silver is reporting today that out of the 10 weekend polls, 5 have Obama gaining, 1 has Romney up and 4 are unchanged.

http://fivethirtyeight.blogs.nytimes.com/2012/10/15/oct-15-distracted-by-polling-noise/

On the Five Thirty Eight forecast, Obama's chances of winning have gone up from 61.1% on Friday, 62.9% on Saturday, 63.4% on Sunday to 66% yesterday.

All eyes on the debate tonight.

nagilum2 · 16-10-2012 6:45pm

Real Clear Politics has two new tracking polls out today that show Romney extending his lead a little.

http://www.realclearpolitics.com/epolls/2012/president/us/general_election_romney_vs_obama-1171.html

Rasmussen Tracking:
10/13 - 10/15 1500 LV 3.0 49 47 Romney +2

Gallup Tracking:
10/9 - 10/15 2700 LV 2.0 50 46 Romney +4

EDIT: Daily Kos has a poll out today that also shows Romney leading 50-46.

But critically, Obama still holds slim leads in Ohio, Iowa, and Nevada. Barring unexpected results in another state, Obama needs only Ohio plus one of the other two to win the electoral college 271-269, regardless of the popular vote totals. Also in the Daily Kos poll, Romney is winning Red States by an average of 16 points, while Obama is winning Blue states by only 7. I think it's becoming increasingly likely that Obama will lose the popular vote. His strategy at this point should be to concentrate nearly everything he's got on Ohio and Iowa. If he gets those two states, he almost certainly wins.

Also, these sites show nice rollups of the state-by-state polls and more granular data for each state:
http://electoral-vote.com/
http://www.270towin.com

Atlantis50 · 16-10-2012 10:36pm

Duck Soup wrote: »

Joe Biden seems to have done the job. Nate Silver is reporting today that out of the 10 weekend polls, 5 have Obama gaining, 1 has Romney up and 4 are unchanged.

http://fivethirtyeight.blogs.nytimes.com/2012/10/15/oct-15-distracted-by-polling-noise/

On the Five Thirty Eight forecast, Obama's chances of winning have gone up from 61.1% on Friday, 62.9% on Saturday, 63.4% on Sunday to 66% yesterday.

All eyes on the debate tonight.

Nate Silver's 'model' is a pathetic joke.

Despite the fact that most polls and seasoned political analysts call the race a tie and too close to call, Silver's 'model' has never had Romney's chances over 40%.

It's obvious there's a bias or 'house effect' of at least 15% in favour of Obama in Silver's 'model'.

In other news, Gallup's poll today shows Romney beating Obama by 4% among likely voters, 50 to 46%.

The liberal website The Dail Kos show a Romney lead of 4% nationally, much to the dismay of the readers: http://www.dailykos.com/story/2012/10/16/1144982/-Daily-Kos-SEIU-State-of-the-Nation-poll-Romney-s-best-numbers-of-the-week

Duck Soup · 16-10-2012 11:00pm

Atlantis50 wrote: »

Nate Silver's 'model' is a pathetic joke.

Silver correctly forecast the winner of 49 out of the 50 states in the 2008 election. I'd tend to take him seriously.

Ozymandius2011 · 16-10-2012 11:10pm

If Obama loses the popular vote but wins the E.C., I wonder would enough of his Electors stay with him when their state delegations vote to get him over the line given it could be close and how bitterly the Dems opposed what happened in 2000 when Bush got in that way? So-called "faithless electors" who go against the vote in their state have been a feature of US Presidential elections since the US was founded.

Black Swan · 16-10-2012 11:18pm

The RCP average (7 to 15 October) slightly in favour of Romney before tonight's 2nd presidental debate (i.e., Town meeting):
Romney = 47.4
Obama = 47.0
Spread = +0.4 Romney

Given the confidence intervals associated with the polls listed and RCP averaged, this would appear to be a tie for all practical purposes.

Atlantis50 · 16-10-2012 11:31pm

Duck Soup wrote: »

Silver correctly forecast the winner of 49 out of the 50 states in the 2008 election. I'd tend to take him seriously.

So did RealClearPolitics in their poll of polls for the states.

2008 was an easy year for prediction in any event as an Obama victory was never in doubt in the 6 weeks before the election and his winning margin in most swing states was relatively comfortable.

And touting a 49/50 accuracy rating is unwarranted because only about 10 of those 50 states swing between the Dems and the Reps. There will be no kudos for predicting Obama will win California and Romney will win Kentucky for example.

You have also failed to address the central point of why Silver's model has never had Romney above 40% despite polls and political analysts calling the race a tie.

Do you believe Silver's model or your lyin' eyes?

Ozymandius2011 · 16-10-2012 11:43pm

Well Silver's model doesn't only rely on the polls but also on "state fundamentals" and economic factors. And it's the Electoral College that matters in the end not the popular vote. The model has them within 1% on the popular vote.

nagilum2 · 16-10-2012 11:45pm

Ozymandius2011 wrote: »

If Obama loses the popular vote but wins the E.C., I wonder would enough of his Electors stay with him when their state delegations vote to get him over the line given it could be close and how bitterly the Dems opposed what happened in 2000 when Bush got in that way? So-called "faithless electors" who go against the vote in their state have been a feature of US Presidential elections since the US was founded.

More and more I think this is a very realistic scenario. Not likely, but quite possible. And if it happens, Obama will be president for a second term without question.

This time the democrats would talk about the the legitamacy of the EC and it would be the republicans screaming about how Obama won illegitimately.

How many people complain about a questionable penalty that goes for their side instead of against them? The view of such things is always colored by the lens of ones own immediate circumstances.

nagilum2 · 16-10-2012 11:55pm

Duck Soup wrote: »

Silver correctly forecast the winner of 49 out of the 50 states in the 2008 election. I'd tend to take him seriously.

Not only that, he is a sabremetrician who knows statistics very well and built a company around forecasting baseball player performance. He applied his knowledge to the elections and ended up being closer than pretty much anyone.

He didn't work for the nyt back then. They hired him based on performance.

He publishes his methodology right on the site and provides a lot more info about where his number come from than many other polling and poll aggregation organizations.

EDIT: Supporting information - best and worst pollsters of 2008:

The prize for accuracy goes -- not surprisingly -- to Nate Silver of fivethirtyeight.com, who underestimated Obama's margin of victory by only four-tenths of one percent, coming within three-tenths of one percent of Obama's actual number, and within one-tenth of one percent of McCain's actual number.

(btw - the most accurate actual poll in 2008 that wasn't an aggregation was Rasmussen - they were within 5 tenths. The RCP average of polls was off by 11 tenths )

Brian? · 17-10-2012 1:02am

Ozymandius2011 wrote: »

If Obama loses the popular vote but wins the E.C., I wonder would enough of his Electors stay with him when their state delegations vote to get him over the line given it could be close and how bitterly the Dems opposed what happened in 2000 when Bush got in that way? So-called "faithless electors" who go against the vote in their state have been a feature of US Presidential elections since the US was founded.

The dema weren't fighting on the basis of the popular vote in 2000, they were taking issue with how Florida was "won" by Bush.

Ozymandius2011 · 17-10-2012 1:05am

Brian? wrote: »

The dema weren't fighting on the basis of the popular vote in 2000, they were taking issue with how Florida was "won" by Bush.

Well Hillary was introducing bills to abolish the E.C. I think afterwards.

Scuba Ste · 17-10-2012 3:27am

Ozymandius2011 wrote: »

If Obama loses the popular vote but wins the E.C., I wonder would enough of his Electors stay with him when their state delegations vote to get him over the line given it could be close and how bitterly the Dems opposed what happened in 2000 when Bush got in that way? So-called "faithless electors" who go against the vote in their state have been a feature of US Presidential elections since the US was founded.

Faithless electors have been a feature not a deciding one though. I'm not sure but are you suggesting that this election could be affected by faithless electors? It won't.

Electors don't vote as such, it really is a formality and many states have punishments for electors that change votes.

It's an anachronism, they should abolish it.

And as another poster explained 2000 was about Florida with even a Supreme Court Justice questioning the validity the election.

Ozymandius2011 · 17-10-2012 3:36am

Scuba Ste wrote: »

Faithless electors have been a feature not a deciding one though. I'm not sure but are you suggesting that this election could be affected by faithless electors? It won't.

Electors don't vote as such, it really is a formality and many states have punishments for electors that change votes.

It's an anachronism, they should abolish it.

And as another poster explained 2000 was about Florida with even a Supreme Court Justice questioning the validity the election.

Wikipedia says one election was decided by it in the 1800s. I think it was John Quincy Adams' election.

Brian? · 17-10-2012 6:34am

Ozymandius2011 wrote: »

Well Hillary was introducing bills to abolish the E.C. I think afterwards.

Which should be done in all fairness. The EC is pointless these days.

jank · 17-10-2012 3:40pm

Brian? wrote: »

Which should be done in all fairness. The EC is pointless these days.

That is debatable, it means the smaller states still have a say with the EC. Don't forget the US is a republic not a democracy.

nagilum2 · 17-10-2012 4:31pm

jank wrote: »

That is debatable, it means the smaller states still have a say with the EC. Don't forget the US is a republic not a democracy.

I agree. This is a critical distinction. The electoral college system ensures that the campaign strategy doesn't simply become one of turning out base voters in areas where one party is highly dominant, but instead tends to focus on winning states that have the most heterogeneous mixture of political views.

matthew8 · 17-10-2012 5:03pm

Republican (the adjective not the party) Democracy is not an oxymoron. They are not mutually exclusive whatsoever, and infact are pretty much the opposite.

As for Nate Silver's model, I trust him a hell of a lot more than the mainstream media. After all, who the heck would watch the debates if they didn't think the race was close. It's in the media's interest to say the race is close. It's probably in Silver's interest too, but he's not as reliant on the race being close.

Brian? · 17-10-2012 5:56pm

jank wrote: »

That is debatable, it means the smaller states still have a say with the EC. Don't forget the US is a republic not a democracy.

Yes, but I would like it to be more democratic.

Scuba Ste · 17-10-2012 6:43pm

Ozymandius2011 wrote: »

Wikipedia says one election was decided by it in the 1800s. I think it was John Quincy Adams' election.

Nope, the 1824 election was decided by congress because no candidate won a majority of the electoral vote. Faithless electors have never affected a presidential result. They did affect a VP election in 1836 but the 'winner' elected by congress anyway. There hasn't been more than one faithless elector in a Presidential election since the 1800's. It's an anachronism; it will not affect this years outcome.

Amerika · 17-10-2012 6:46pm

17, October – Gallup - Likely Voters Trial Heat

Romney – 51%
Obama – 45%

Still too early for any happy thoughts from my side, as a lot can still happen in 3 weeks.

http://www.gallup.com/poll/157817/election-2012-likely-voters-trial-heat-obama-romney.aspx

(And Paddy Power just put Romney's odds at 2/1... OH MY!)

Scuba Ste · 17-10-2012 7:23pm

jank wrote: »

That is debatable, it means the smaller states still have a say with the EC. Don't forget the US is a republic not a democracy.

What? It is a democracy. That's just being pedantic.

Smaller states may have a say because of the electoral college system but the majority of states are largely irrelevant because of it. The electors are now irrelevant and yet they still meet to cast their vote after election day. It's a silly system that was a poor compromise in 1787.

The only valid argument I've heard to support it is that elections could potentially be more expensive if candidates had to visit more states in different regions of the US. Even this though is a poor reason to keep the system.

southsiderosie · 17-10-2012 7:32pm

Because the presidency is not decided by popular vote, at this point in the game, I don't think focusing on the overall 'likely voter' scenario makes sense - you have to get down to the state level, since that's how the votes are actually tallied.

Amerika · 17-10-2012 7:40pm

southsiderosie wrote: »

Because the presidency is not decided by popular vote, at this point in the game, I don't think focusing on the overall 'likely voter' scenario makes sense - you have to get down to the state level, since that's how the votes are actually tallied.

That is a correct statement at any point in the game, as well as for page 1 through page 24 of this thread. I just find it interesting if that’s what the general consensus will be now that Romney is pulling ahead in the polls.

Brian? · 17-10-2012 7:43pm

southsiderosie wrote: »

Because the presidency is not decided by popular vote, at this point in the game, I don't think focusing on the overall 'likely voter' scenario makes sense - you have to get down to the state level, since that's how the votes are actually tallied.

I completely agree. I only ever look at the swing state polls myself:

http://www.rasmussenreports.com/public_content/archive/2012_electoral_college_scoreboard

Statistically pretty much a tie with a lean towards Obama.

US 2012 Presidential Election Polls

Comments