Skip to comments.Is Rasmussen using D+3, or D+6?
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.)
Depends on where he’s polling. I think he uses 2008 and 2010 turnout averages, and then leverages it against enthusiasm factors to get a proper sample figure.
Rasmussen uses a D+3 (39D/36R/25I).
Then yesterday, people were surprised that a couple people posted he had changed even that to D+6.
I'd love to know what he's using at the state level, but I'd settle for being sure about his national model.
I'm in a daily conversation lifting the spirits of many Romney people (to the truth, as far as I'm concerned), and making fun of my liberal friends, who will eat their words for every time in the past 3 months they have called me nuts, crazy, lying etc.
thanks for all your posts. you have really helped a lot of us analyze and stay informed.
quick question: do you feel “pretty good” about things?
I do, personally. I go back and forth, but the lows are not as low and the highs somehow don’t feel high, but just really well...rational.
I feel pretty good.
Agree? Or no?
Any idea what the talk of D+6 switch from D+3 was yesterday?
I can probably find the thread, but it didn’t sound like people were just tossing it out - sounded like they were paid subscribers, at least 2 people stated it as fact.
I can’t give you a link but we did have a Raz subscriber post a few days back that it was definitely D+3. He sounded pretty authoritative so I took him at his word.
Also, a 3 point jump from D+3 to D+6 would cost Romney nearly 3 points in the tracking poll over a 3 day period. We saw no such jump. So I’m sticking to the D+3 theory.
if Rasmussen uses D+6, and in reality there are 4% more registered Republicans than democrats, in ‘12, doesn’t that mean Rasmussen has over sampled democrats by 10? so, his 4 point lead for R/R should be 14?
Ras isn’t weighting the pollsthe reality is that there are just more Ds.
It’s probably D+3, and there’s a good bet that the GOP and DEMs will be within 1% of each other nationally.
You are like the lying Kenyan repeating the bs 5 trillion dollar tax cut that Mitt supposedly has proposed. Prove me wrong!
If turnout is key, shouldn't we weight the poll enough to give the average D voter enough confidence to say "Pookie" will do it, while scaring enough Americans to set 5 alarm clocks the day of the election? In that case, D+4 sounds about right.
I’m agreed - it would make no sense not only because of that, but his numbers didn’t change.
I’ll see if I can dig it up. Whoever the 2 posters were, they were pretty clear about the D+6, appropriately shocked, and I don’t think they were ... skewing around ... so to speak.
The talk, I believe, was based on Romney winning the Indies by 20+ but still only up by 3-4. It did not seem possible.
But I think many missed that Romney was scoring 86-88% base support, not the historical 90-93. Every extra 3% from base (based on 36% Rs) adds a full point.
He did firm up his base today, and stands at 90% for Republican support. But dropped Indie support.
Also, Rasmussen only rounds off the toplines numbers. So when it shows, 50-47 for example, it could be closer to 51-46 when you dig into the data (decimal points), e.g. 50.46-46.50 will read 50-47 as the topline, but it’s actually almost 4 point lead.
that was NHWingut. See his post number 3 above.
Also see his posts in the DAILY RAS thread from yesterday and today.
NH does awesome work with the RAS Internals
Here’s where the posts are ...
Gman, Ravi, Lacey2 refer to it ...
I’m reviewing to see if its refuted later in the thread.
His Florida state poll from Friday was 750 likely voters 43R/39D/18O.
His Virginia poll from Thursday was 750 likely voters 38R/37D/24O.
His Ohio poll from Wednesday was 750 likely voters 39R/38D/23O.
Here’s a bit copied and pasted from that thread, on Friday ...
To: The G Man
Yeah he did change it to a D+6 model. That irritates me. I think he is scared of showing a romney blowout.
10 posted on Fri Oct 26 2012 09:48:49 GMT-0400 (EDT) by Ravi
Got it - so the D+6 was probably a reference from a state poll?
Assuming so, thanks PJ
why does RAS weight like that then? If his own party ID split is different? Not challenging you, but you have provided some pretty insightful information. Thanks in advance....
You’ve provided specific information, duplicated by “Political Junkie Too”, and yet some don’t seem to want to hear the answers.
Gallup doesn't ask their party affiliation.
Thanks for posting. This needed to be flushed out.
Once it gets out on Twitter it’s hard to bring it back.
But I am positive that Rasmussen is using a D+3.
I’ve been under the impression that most of the legit pollsters simply report whatever the party representation of the sample happened to be.
Thanks NH ...
Certainly makes sense that it’s still D+3
And probably it was a serious discussion about D+6, but referring to an individual state, or just quoting a bad source. Not going to investigate the entire thread.
I just found the North Carolina poll from Saturday was 500 likely voters 34R/41D/24O.
The Pennsylvania poll from Thursday was 500 likely voters 37R/43D/20O.
The Michigan poll from Friday was 500 likely voters 34R/37D/29O.
The Massachusetts poll from Tuesday was 500 likely voters 17R/43D/40O.
I could go on.
I was wrong. Retweeted @numberscruncher who was incorrect. Nhwingnut’s explanation is exactly correct. Romney was winning only 85-88% of republicans up until today so that’s why he wasn’t cleaning pbama’s clock even though he was cleaning up wit indies.
95D/-10R/15O (or at least it seems that way having lived here for over 50 years)
Some do, some don’t.
It depends on the sample size. Gallup does a massive sample (2500 plus) so they will get a good read of the electorate and will not weigh.
But smaller samples have to be weighted or else they would be useless.
Where in the heck is he polling these Republicans? I do not know of one Republican in CA where I live who is voting for Obama. Squirrelly, if you ask me.
Cool, thanks. Gallup seems the better barometer to me in that case.
Not according to the guy who runs Gallup. They do not ask party at beginning of poll, they do ask party at end of poll.
That’s been my explanation to friends.
Rasmussen has a reputation to uphold in terms of accuracy, but he also has a business presence & narrative.
No one can ever prove you were wrong 2 weeks before the election.
If I’m Rasmussen, I’m thinking ‘stay close enough to the truth that I can veer towards it a few days prior to election, but stay within the pack, giving me the most options.’
‘Hedge’ and ‘conservative’ are the exact two words I’ve probably used 10 times in emails to explain to friends in the past week ...
if ras is not sampling properly....+3 or +6D, when really it’s + something for R, why is he any better than the rest?....because he’s not +13D? why bother to sample @ all if your not being honest, like the lib polls?
Yes, but I also thought I heard him say they don’t use it for weighting.
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.