Some polls now have Romney ahead.

[First]

Very odd polling day in terms of contrast of state and national polls.

Nationally, among tracking polls, Obama lost 1 in Rasmussen's, 1 in Gallup's, 2 in IBD's and remained tied in PPP's.

At the state level it was a much different story, with Obama:

Up 1 in Virginia 49-48 based on PPP poll.
Up 3 in Colorado 50-47 based on PPP poll.
Up 6 in Michigan 52-46 based on EPIC-MRA poll.
Up 6 in Wisconsin 51-45 based on NBC/WSJ/Marist poll.
Up 8 in Iowa 51-43 based on NBC/WSJ/Marist poll.

Based on the Gallup polling of 6 and 7 point spreads, you would have to see undeniably strong movement in Midwestern states where Obama is strong because that suggest a national trend (states like Iowa, Michigan, Wisconsin), but you see exactly the opposite. This suggests Gallup is an outlier, but we won't know for a couple/few more days.

 

[First]

Ah, I didn't see that 538 article. I think that adequately explains it then, if their LV models are that wild in 4 different elections and no one else sniffs them within half a dozen % points, then it's likely wrong. But Obama did lose one point in Rasmussen and 2 in IBD today, so I wouldn't say for sure yet until just before the next debate.

 

[umbrella39]

Are we setting up for a Bush/Gore all over again (i.e. Obama wins with EC and Romney wins the popular vote)?

I don't think it will be that close on the PV either tbh...

 

[randomrogue]

538 updated their numbers:

291 (O) to 246 (R) on the EC

50.3% (O) to 48.8 (R) on the PV

Obama up to 70.4% chance of winning

 

[Engineer]

At the same time, the RCP EC map now favors Romney 206 to 201 for Obama.

It's getting tight and it might come down to whoever wins Monday's debate.

(I do understand that there might be a lag on Tuesday's debate on the polls and RCP EC map coverage so we will see how it shakes out).

 

[randomrogue]

At the same time, the RCP EC map now favors Romney 206 to 201 for Obama.

It's getting tight and it might come down to whoever wins Monday's debate.

(I do understand that there might be a lag on Tuesday's debate on the polls and RCP EC map coverage so we will see how it shakes out).

I don't believe that map even reflects their own poll numbers. They have a ton of states as contested when they aren't.

 

[Charles Kozierok]

The RCP map is very conservative about putting states in the "toss up" category.

The "no tossup states" map shows the bottom line: for Romney to win, even if he holds all the states they have for him now, he has to flip either Ohio or two of Wisconsin, Iowa and Nevada. And I don't think he's really had a lead in any of them in months.

 

[jpeyton]

At the same time, the RCP EC map now favors Romney 206 to 201 for Obama.

There are 538 electoral votes, not 407. You're looking at the wrong map:

http://www.realclearpolitics.com/ep...ctions_electoral_college_map_no_toss_ups.html

 

[shortylickens]

I never thought to ask but has Huffington Post been unbiased and objective in the past?
The only reason I ask is I like their map the best and I'd hate to have to change now that I'm used to it.

http://elections.huffingtonpost.com/2012/romney-vs-obama-electoral-map

There's is simple and effective.

 

[Charles Kozierok]

I never thought to ask but has Huffington Post been unbiased and objective in the past?

No. It's a very left-leaning site.

 

[Blackjack200]

No. It's a very left-leaning site.

That's okay though, I think Silver (538) leans left, and RCP leans right. I prefer 538 because I think they have the best analysis of what's driving the polls and why some polls show different results than others.

 

[Blackjack200]

There are 538 electoral votes, not 407. You're looking at the wrong map:

http://www.realclearpolitics.com/ep...ctions_electoral_college_map_no_toss_ups.html

I think you could actually look at it as bad news for Romney. If the race is this close you want to have a thin lead in many states, not a thick lead in fewer states.

 

[randomrogue]

More poll numbers in and Romney's momentum has stopped completely.

 

[Farang]

More poll numbers in and Romney's momentum has stopped completely.

He still has not broken through IA, OH, or WI, those three are a royal flush for Obama

Worse for Romney, Obama holds backups for all three in NV, VA, and CO, which combined we could call a coin toss at this point (some lean more than others)

 

[woolfe9999]

There are 538 electoral votes, not 407. You're looking at the wrong map:

http://www.realclearpolitics.com/ep...ctions_electoral_college_map_no_toss_ups.html

Even that number seems to be slightly overstating Romney's chances. It's a "no toss up" map, so look at how many states are within a single percentile. There are three: Virginia, which is called a straight tie but shows for Romney on the map; Colorado, which shows Romney up by .2%; and New Hampshire, which shows Obama up by .8%. Romney gets 22 EV's out of these states and Obama gets 4. The largest chunk is Virginia at 13 EV's but shows a dead statistical tie in the polling. It's questionable to call any of those 3 states for either candidate, but even when called that way, Romney still loses. You could even give Romney New Hampshire on top of Virginia and Colorado and he still loses.

 

[Farang]

Even that number seems to be slightly overstating Romney's chances. It's a "no toss up" map, so look at how many states are within a single percentile. There are three: Virginia, which is called a straight tie but shows for Romney on the map; Colorado, which shows Romney up by .2%; and New Hampshire, which shows Obama up by .8%. Romney gets 22 EV's out of these states and Obama gets 4. The largest chunk is Virginia at 13 EV's but shows a dead statistical tie in the polling. It's questionable to call any of those 3 states for either candidate, but even when called that way, Romney still loses. You could even give Romney New Hampshire on top of Virginia and Colorado and he still loses.

Agreed.. all the blue on that map is where Romney has not, ever, been able to break through (exception: NH). That map is Obama's firewall. Then everything else that's been swinging back and forth, or is a tie, is red--VA, CO, FL. What that map shows is that Romney is playing defense on a losing map

 

[woolfe9999]

Agreed.. all the blue on that map is where Romney has not, ever, been able to break through (exception: NH). That map is Obama's firewall. Then everything else that's been swinging back and forth, or is a tie, is red--VA, CO, FL. What that map shows is that Romney is playing defense on a losing map

Yup, the only thing I'm confused about is the PV polling right now. RCP shows Romney up .7% but fivethirtyeight shows Obama up by 1.2%. I've never seen the 2 disagree by that much. I'd really like to know why there is nearly a 2 point gap there. For the moment, I'm going to assume the real number is the average of the two, putting Obama up by barely a hair.

 

[Blackjack200]

Yup, the only thing I'm confused about is the PV polling right now. RCP shows Romney up .7% but fivethirtyeight shows Obama up by 1.2%. I've never seen the 2 disagree by that much. I'd really like to know why there is nearly a 2 point gap there. For the moment, I'm going to assume the real number is the average of the two, putting Obama up by barely a hair.

I think Silver puts his polls through an algorithm, I think it's a weighted average, not a simple average. I don't know if RCP does the same, but it could explain the difference with the weird Gallup numbers.

 

[woolfe9999]

I think Silver puts his polls through an algorithm, I think it's a weighted average, not a simple average. I don't know if RCP does the same, but it could explain the difference with the weird Gallup numbers.

I think they both do. Differences in weighting methodology may explain it, but then I am still wondering why they were much closer together in '08 and indeed, even just 2 weeks ago.

 

[Engineer]

Yup, the only thing I'm confused about is the PV polling right now. RCP shows Romney up .7% but fivethirtyeight shows Obama up by 1.2%. I've never seen the 2 disagree by that much. I'd really like to know why there is nearly a 2 point gap there. For the moment, I'm going to assume the real number is the average of the two, putting Obama up by barely a hair.

RCP now shows Romney up by 0.1

 

[Farang]

I think they both do. Differences in weighting methodology may explain it, but then I am still wondering why they were much closer together in '08 and indeed, even just 2 weeks ago.

Not really surprising, Nate Silver uses a much more complex method than I think RCP uses. Correct me if I'm wrong but RCP's method seems arbitrary at times.. I remember reading that the guy thought Gallup's number was an outlier, so he decided to weight it down 2/3rds. Compare that to Silver who basically built a forecasting machine before the election, and now just feeds numbers into it and explains why it reacts certain ways.

 

[First]

Obama now up a smidgen in the RCP national average:

Hartford Courant/UConn has him up 3
Latest IBD tracking has him up 2, a 2-point gain since yesterday.
Latest Rasmussen tracking has it tied, also gaining 2 since yesterday.
Gallup going from +7 to +6 for Romney.

Two good state polls for Romney today from Rasmussen though. Shows him up 3 in Virginia and 5 in Florida.

 

[woolfe9999]

Obama now up a smidgen in the RCP national average:

Hartford Courant/UConn has him up 3
Latest IBD tracking has him up 2, a 2-point gain since yesterday.
Latest Rasmussen tracking has it tied, also gaining 2 since yesterday.
Gallup going from +7 to +6 for Romney.

Two good state polls for Romney today from Rasmussen though. Shows him up 3 in Virginia and 5 in Florida.

I think these polls you're seeing today are probably the first reflecting the small second debate bounce that is expected for Obama. If it continues like this for a few more days, Romney will probably be in a win or die situation Monday night.

 

[buckshot24]

Hartford Courant/UConn has him up 3

Their poll has a split of 46.4% Democrat vs 38.7% Republican and only 14.9% independent. If the vote that turns out is that split then Obama wins, hands down.

Latest IBD tracking has him up 2, a 2-point gain since yesterday.

They don't explicitly state the number of Dems vs Republicans (I don't feel like doing the math). For those who voted for Obama in 08' only 81% is voting favor him this year while those who voted for McCain, 91% favor Romney. 4% of McCain voters favor Obama while 10% of Obama voters favor Romney.

Assuming that the exact same voters turnout this year this poll indicates that Romney would get 50.6% (compared to the total votes between both candidates) of the vote while Obama has 49.4%

Not a good poll for Obama.

 

[Todd33]

buckshot24 has this all figured out, they should just call the election now.