Posted on 10/28/2012 7:49:43 PM PDT by HannibalHamlinJr
Yesterday morning I believe, in a thread, at least twice it was posted that Rasmussen has moved from a D+3 to D+6 partisan model. I took it that the second post was from a Rasmussen paid subscriber. Do we have confirmation or denial of this? ... Wouldn't normally interrupt into main threads, but this is a pretty important factoid, question about veracity came up on another thread tonight. (I searched ... but without searching body of threads, can't locate it to see if it was confirmed later in the thread.)
This is where Ras may have some issues with a robopoll and people lying. That’s just a hunch on my part. Some enterprising dems may i repeat may be claiming to be republicans voting obama. That’s why romney’s share of republican voters is slightly lower in ras compared to other pollsters from what I’ve seen.
I think there is one remaining caveat.
Rasmussen has his party id published at the end of every month.
As stated, in September it came in at R+2.6 (after a rough September for Romney).
I’m wondering if it went back up to around the R+4 range (like it was in August), would Rasmussen then adjust his weighting for the last week to reflect a more realistic turnout? Not so much a weight of R+4 but say R+1 like Gallup.
It should be an interesting week.
Poll ping discussion.
If D/R party ID is even on election day, and Romney is really winning Indys by RCP average of 15, then this is a blowout of epic proportions.
Seems like every pollster is ignoring the 2010 elections. Maybe they never factor mid-terms into the presidential race. But the ground swell for Repubs in 2010 has not abated. The reasons for the massive swing in 2010 is even more urgent now. The people who were mad in 2010 are even angrier now. USSC upheld Obamacare. Economy still stuck in neutral. More government intervention and regulation.
I have been saying since August, R53/O47 and 330 EVs.
I think you meant to say “D+2.6” for September.
I've wondered about that, myself. What happened in 2010 doesn't fit their "narrative."
I have been saying since August, R53/O47 and 330 EVs.
I hope you are right!
Yes. Republicans historically get 92-93% of their base, while Democrats secure 88-89% of their base.
Both Romney (90%) and Obama (86%) are underperforming with their base.
I find it hard to believe that Romney won’t get 94% or so of Republican voters on 11/6.
No, the party id survey came in at R+2.6... But he uses a D+3 sample.
OK, thanks for clarifying.
I’m agreed -
Which is why I’ve told friends that Ras is no longer a Bible.
That said, I still think he’s the benchmark in terms of trend. His D+3 is a sin, but a predictable sin. Many of the other polls - I doubt they’re even consistent WITHIN their own model.
But this year, I’ve trusted the polls that don’t sauce it up, plus adjusting Rasmussen’s polls for the skew. Looks like from his intra-state polls - FL, VA etc ... that he’s not applying as much sauce.
Some of the smaller intra-state pollsters are the most accurate because of this - they’re more a reflection of what’s on the ground.
So I supect the Michigan poll from a couple days ago is pretty close, as is the Minnesota poll calling a 3 point race. (It’s probably about 1 though - but Marist would probably say it’s Obama +45)
Susquehanna, soon after releasing it’s PA polls showing it even and then 3 weeks ago with Romney up, defended itself in the face of a lot of flack by simply identifying all the things we’ve been talking about here. That it’s not a 2008 wave (might be the opposite), that blacks and yoots aren’t turning out, that indy’s are breaking to Romney ... and about 5 other things ... they just said ‘hey, we’re just being rational, not saucy.’
And remember - Suffolk pulled out of FL, VA and NC about 2 weeks ago, stating that they were a lock for Romney. That was about 10 days before people starting saying ‘well, looks like FL is leaning strong to Romney, NC too, VA looking scary for Obama.’
Suffolk called all that 2 weeks ago.
Ras though, whatever his reasoning, is I belief reliable as long as you know the adjustment.
But even with adjustments, it doesn’t all add up. I think we’ve accounted for 70% of the inaccuracies. The rest - who knows ... you can tweak these numbers any way you want with plausible deniability.
If you listen to the news - they take little pieces from each poll and stitch it together into the story they want to tell.
The best poll is “where are the campaigns visiting and advertising”
Which is why it’s so great that Romney has his “Expand the Map” push going on.
Thanks SE...
BTW: I was looking at the Gallup and Rasmussen Party Id surveys tonight.
It’s amazing how close they match.
Rasmussen Party Id: 36.8R/34.2D/29I
Gallup Party Id: 36R/35D/29I
If these are true (and I don’t doubt they are), Obama will lose in a landslide.
At first I discounted Gallup - too volatile in the past ... but they’ve been pretty reasonable and stable since 10 days or so ago when they first came out with RR way up, and they and Ras have been converging.
Still, my spidey senses tell me this is going to be RR +7’ish come election day, and that’s without Benghazi getting any worse.
Gallup’s finding Obama dropping 7 points in 3 days on Approval ... not sure what’s collapsing there - if it’s ‘end of the bandwagon’/’start of avalanche’ or Benghazi sinking in or what.
Lot of pissed of Dems who went to the mat for Obama and are just now seeing he ain’t got their backs and never did.
Scott was on the tube the other day. He said: he is not manipulating the D-R number one way or the other. He says they simply make the contacts and, based on the answers to questions about past voting history, intention to vote and enthusiasm, he classifies them as “likely voters” or not. Then he asks about party ID. Specifically, he asks “as of today, do you consider yourself R, D or I.” If his contacts reveal a party advantage in the likely voter model, it’s simply the fact that that is what he is seeing according to his model.
Romney by 7?! Cmon, put down the bong and step away.
This is a close race. Every voter needs to get out and vote. No one will win by over 5 points. No election has ever been that out of whack with poll averages. All averages have romney by 1. I think he may win, if everything goes well for him, by 2-3.
You most likely are right, but tell that to the Battleground poll out tomorrow that has Mitt +5.
I’ll email myself your username, and ping you on the 8th :-) You will be happy to be wrong in this instance.
There are qualitative things happening that can’t be measured.
Everyone here is going to vote, I’m not concerned about talking about numbers around people who will claw their way through hell to vote. I’ve already had to go twice to get an absentee after the first one disappeared from my mailbox.
But at this point, Romney is focused on mandate, he’s past winning. That’s what ‘expand the map’ is all about.
Amen!
Romney by 7?! Cmon, put down the bong and step away.
This is a close race. Every voter needs to get out and vote. No one will win by over 5 points. No election has ever been that out of whack with poll averages. All averages have romney by 1. I think he may win, if everything goes well for him, by 2-3.
So much conflicting info .... I just assume he’s a businessman and in his business, he’s the PR arm of the business, he’s not sworn to uphold the absolute truth, so he bends where it benefits him and where he doesn’t feel it hurts anyone else.
I’m really really really ready for today to be November 8. (Leaving the 7th for the O campaign and media to focus test protesting the result.)
Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.