Romney, Obama, Rasmussen

ANTPogo said:
They already started...their last couple of national polls and their recent Ohio poll have all shifted two or so points towards Obama, and are listed as "tie".

In the latest roundup of national polls, in fact, Romney is not leading in a single one - there are 9 showing Obama ahead, and four (including Rasmussen) showing a tie.
Click to expand...

Foolmewunz said:
Gallup got back on line in time to have their last seven embarrassing polls adjusted. They and Rasmussen come in at Romney +1.

I've been mentally conceding VA to Romney for the past month. Some believable-sounding talking head from a University convinced me of that when I was half awake one morning. I didn't see Obama covering +300, but apparently Silver's count and the RCP average have it trending that way.
Click to expand...

All the polls have been trending Obama's way, Rasmussen is just picking up what everyone else is, a late bounce for Obama probably due to the better part of the week being hurricane coverage, Rasmussen is still most likely to be correct in my view. :)

Because of this unexpected bounce I have to revise my estimate. Bounces are (generally) driven by events, the next is the election and I expect that to serve to diminish most of the rest of it, thus in the end it will be Romney by three, 51-48. The other polls will be off due to overestimating the democrat turnout, and Rasmussen and Gallup handicapped by too much of the respondent population being dominated by the 'Honey Boo-boo vote' who are the ones home just waiting for their pollster call so they can say 'Obama.' Also I think a fair amount of the late bounce is a kiss good-bye, Americans not disliking Obama and willing to rally in times of turmoil as usual, but when they get into the voting booth they will vote their pocketbook.

I think Romney sweeps the south, winning by about 6 in Florida and about the same in North Carolina, and by about 2-3 in Virginia as Goode voters go into the booth and realize they'd rather have their vote count. Romney picks up New Hampshire and possibly Maine's second district, but I cannot get enough info to regard that as anything more than an unlikely surprise. Colorado goes Republican, the GOP leading in early voting there and it's generally a state friendly to Republican moderates. Nevada however goes to the Reid machine by about three points.

That leaves the industrial Midwest, and here it is revealed that the map is not as friendly to Obama as was thought by many. That was predicated on Obama possibly taking something in Dixie like he did in '08, if that doesn't happen and he doesn't sweep the two competitive ones in the west he then goes into the Great Lakes needing to take basically everything. I suspect one of Ohio, Pennsylvania or Michigan to go Romney, and possibly even Minnesota, but information at this juncture suggests he may well fall short in the big three, and there's not much information out of Minnesota to rely on. However all Romney has to do is take Wisconsin and he will barely push past the post with 271, and I suspect he'll pick up Iowa as well, to make it 277 minimum, with a possibility of picking up one of the other three Great Lakes states, to near 300, and it's not all that unlikely he'll win two of them for a more resounding victory.
 
Last edited:
So the polls - all the polls, systematically - are being skewed in Obama's favor by all those people who are sitting around at home, eagerly awaiting calls from polling firms just so they can say they're voting for Obama (without, of course, going on to actually vote for Obama)? And there's a chance Romney might win in Minnesota, a state where the last four polls taken since October 28 have Obama up by 7, 9, 8, and 11?!
 
ANTPogo said:
So the polls - all the polls, systematically - are being skewed in Obama's favor by all those people who are sitting around at home, eagerly awaiting calls from polling firms just so they can say they're voting for Obama (without, of course, going on to actually vote for Obama)?
Click to expand...

That pretty much sums it up! :)

No electorate has ever been so abused by robo-callers and polling. You guys might think I'm kidding, but I'm not! For all the polling attempts, here in Wisconsin I've gotten at least one robo-caller (virtually) every single day for about four months now, often multiple attempts per day. I ought to see if I can get my ATT package to list all the calls I've gotten here since July, you guys wouldn't believe how much politicking we've endured here. I don't think the resulting pool is especially representative of the population that will actually vote and the GOP will have a turnout advantage.

ANTPogo said:
And there's a chance Romney might win in Minnesota, a state where the last four polls taken since October 28 have Obama up by 7, 9, 8, and 11?!
Click to expand...

I did say possibility and not probability. I looked at those polls and was unimpressed by their methodology, (also the ones that showed it a toss-up as well) I think they overstate Obama's lead. I think it's more along the lines of 3-5, which is what Obama will probably win by if he takes it. However that leaves open a non-zero chance that Romney will sneak it out there.
 
Kaosium said:
All the polls have been trending Obama's way, Rasmussen is just picking up what everyone else is, a late bounce for Obama probably due to the better part of the week being hurricane coverage, Rasmussen is still most likely to be correct in my view. :)

Because of this unexpected bounce I have to revise my estimate. Bounces are (generally) driven by events, the next is the election and I expect that to serve to diminish most of the rest of it, thus in the end it will be Romney by three, 51-48. The other polls will be off due to overestimating the democrat turnout, and Rasmussen and Gallup handicapped by too much of the respondent population being dominated by the 'Honey Boo-boo vote' who are the ones home just waiting for their pollster call and so they can say 'Obama.' Also I think a fair amount of the late bounce is a kiss good-bye, Americans not disliking Obama and willing to rally in times of turmoil as usual, but when they get into the voting booth they will vote their pocketbook.

I think Romney sweeps the south, winning by about 6 in Florida and about the same in North Carolina, and by about 2-3 in Virginia as Goode voters go into the booth and realize they'd rather have their vote count. Romney picks up New Hampshire and possibly Maine's second district, but I cannot get enough info to regard that as anything more than an unlikely surprise. Colorado goes Republican, the GOP leading in early voting there and it's generally a state friendly to Republican moderates. Nevada however goes to the Reid machine by about three points.

That leaves the industrial Midwest, and here it is revealed that the map is not as friendly to Obama as was thought by many. That was predicated on Obama possibly taking something in Dixie like he did in '08, if that doesn't happen and he doesn't sweep the two competitive ones in the west he then goes into the Great Lakes needing to take basically everything. I suspect one of Ohio, Pennsylvania or Michigan to go Romney, and possibly even Minnesota, but information at this juncture suggests he may well fall short in the big three, and there's not much information out of Minnesota to rely on. However all Romney has to do is take Wisconsin and he will barely push past the post with 271, and I suspect he'll pick up Iowa as well, to make it 277 minimum, with a possibility of picking up one of the other three Great Lakes states, to near 300, and it's not all that unlikely he'll win two of them for a more resounding victory.
Click to expand...
I like to listen to a lot of conservative talk radio, and the variety of rationalizations offered for why the polls are all wrong has been very entertaining today. There was a bit of "Dewey Defeats Truman" update, with "The polling methodology is no longer viable -- it assumes people have land lines and no caller ID". I'd actually expect mostly Republican senior citizens to be in that demographic, so I'm not sure how it would lead to oversampling Democrats, but that was the claim.

Then there was the talk show host who seemed to lack an understanding of statistics (not surprising, since he routinely attacks universities as places where nothing but lies are taught), who proudly announced that, in a survey of 1000 people, a difference of as few as five voters would be enough for the pollster to give the race to Obama.

You seem to think the Democrats are all 47%ers, sitting at home waiting for their government check to arrive, and thus more available when the pollster calls. At least that theory has the advantage that, if the premise was correct, the conclusion might be valid.

I think it will be interesting to see where the pollsters get it wrong, but I still think their overall call will be correct. I guess come Wednesday morning, one of us will be right.
 
Kaosium said:
That pretty much sums it up! :)

No electorate has ever been so abused by robo-callers and polling. You guys might think I'm kidding, but I'm not! For all the polling attempts, here in Wisconsin I've gotten at least one robo-caller (virtually) every single day for about four months now, often multiple attempts per day. I ought to see if I can get my ATT package to list all the calls I've gotten here since July, you guys wouldn't believe how much politicking we've endured here. I don't think the resulting pool is especially representative of the population that will actually vote and the GOP will have a turnout advantage.
Click to expand...

So basicly a sample size of 1 then huh?

;)

I just don't understand how you can be fooled by what Rasmussen is doing here. This is exactly what they did in 2008. They polled multiple points to the right of the rest of the polls right up to the last polling cycle, at which point they pulled closer to the middle in a effort to say "look, we were pretty accurate with our final number."

This was predicted by others in this very thread months ago I believe.
 
Over on the Titanic, O'Reilly is starting the wake. I'm not sure if I heard correctly, but he came as close to agreeing with 290-248 for Obama as is humanly possible. He actually had two Dem pundits on at the same time and I swear I heard him agreeing with them and using the word "prediction", and then he said, "Karl Rove will be along shortly to tell me why I'm wrong."

I can't find a transcript on line and the Fox site doesn't quote him in those discussions. I'll be at my coffee house when it's re-run and don't get Fox there, so I'll have to rely on the internet to check it out.

He and Krauthammer (who's still calling a Romney win but is really subdued and seems to be just doing his duty) spent five minutes discussing how Mittmentum got stopped in its tracks by Sandy, and then Christie's man-hug was the kiss of death.

A very VERY subdued O'Reilly. Luckily for the frothing at the mouth brigade, Hannity's on now and got his cheerleader outfit on. He's doing a montage of the "Forgotten Victims of Obama's Katrina". Desperation.
 
Unabogie said:
Nate Silver has Romney down to 8.6% chance of victory.

http://fivethirtyeight.blogs.nytime...o-polls-show-benefit-of-auto-rescue-to-obama/

Obama up to 314 EVs.
Click to expand...

I have to say that Silver sticks by his numbers. He's got FL as a tossup but with Obama now slightly favored 52%-48% probability. I don't think anyone's giving Obama a chance in FL.

And his 314 EV is based on his algorithms on likelihood. I don't see a mathematical path to 314, and I'm sure that's not what it means anyway.
 
Foolmewunz said:
I have to say that Silver sticks by his numbers. He's got FL as a tossup but with Obama now slightly favored 52%-48% probability. I don't think anyone's giving Obama a chance in FL.

And his 314 EV is based on his algorithms on likelihood. I don't see a mathematical path to 314, and I'm sure that's not what it means anyway.
Click to expand...

Right, I think that means the simulation, run X number of times, oscillated between something like 302 and 330. I haven't crunched numbers but that seems about right.
 
Does Silver factor things like Rick Scott's sliminess into his simulations when determining that Obama has a 52% chance to take Florida?
 
maxpower1227 said:
Does Silver factor things like Rick Scott's sliminess into his simulations when determining that Obama has a 52% chance to take Florida?
Click to expand...

I dunno, but when I look there, it's showing Romney at 55% to win Florida. What am I missing?

ETA: Refresh is your friend, note to self.

Woot!
 
Last edited:
You must log in or register to reply here.

Back
Top Bottom