Mea Culpa, Mea Culpa, Mea Maxima

m.rasmussenreports.com ^ | 9 Nov 2016 | By Larry J. Sabato, Kyle Kondik, and Geoffrey Skelley

[Billthedrill]

It's only a guess, but if you use last election's participation model too long after the nomination you don't pick up that +5 Dem to +5 Republican swing that changes the outcome so radically. Clearly clinging to a +8 Dem estimate after it became apparent that the distribution wasn't going to be like that resulted in some wildly mistaken estimates. It's why you gradually change from last election's distribution to registered voters to likely voters as the election approaches - you don't get what you really need, "actual voters", until afterward. Which, in turn, becomes the basis for next election's modeling.

The reason for that much filtering in the first place is to cut down the sample size to something manageable, but you aren't going to see a shift with that small a sample size because the old model is built in. And a lot of clients aren't going to pay for the wider survey.

There are, as well, polls that are deliberately skewed with an eye toward weaponization, but that isn't what we're talking about here. Citing those polls whose intention is to affect public opinion, as an honest measure of public opinion, is to believe a lie. Most pollsters aren't silly enough to make that error but it appears that many customers - the media, for example - are silly enough. That's the problem with taking sides and why it's unprofessional in the journalism biz.

I could be entirely wrong about this being the cause of Rasmussen "blowing it". They know their own methodology much better than I can guess it. But that's where I'd look.

[No_More_Harkin]

CO - hippies and granolas
VA - So many DC workers.
NV - Drunk moron club workers and gangsters.

[Qiviut]

My sympathies to Wahoos everywhere who are embarrassed by old Sabotage. I’m from a Hokie family so we’re grateful this morning that there are no poll putzes at VT (who probably would have been wrong as well). :-)

Should be a good Wahoo/Hokie football game coming up before too long!

[The Sons of Liberty]

Well, what can we say — we blew it.

I saw you eating a big dish of crow this morning Larry and the thought that kept running through my mind was "why are they paying this clown? and why is he still employed?" Allow me to say: "You're fired". Anyone, and I mean ANYONE could've wasted our time with the hot air and drivel you've put forth.

While we're on the subject of incompetency, let me relate a possible miracle I witnessed. Right before I went to bed last night around 11:15 PM, Karl (Marx) Rove was on explaining in his superior fashion and with obvious disdain for Donald Trump, that there was no way he could win. When I turned on the set this morning around 6:00 AM, soon Karl was on exclaiming how he had "known" President-Elect Trump would pull it out, all while talking like they were buddies from way back and like HE (Karl) was instrumental in the win. A conversion that is close to Biblical proportions.

[Arlis]

Dun - you guys are just too smart to realize your personal bias and prejudice color all of your poll methodology.......

[60Gunner]

Rasmussen didn't "blow it." They intentionally skewed it.

Never trust polls. EVER.

[peeps36]

How about the media pushing the Hillary is winning big and is up in every swing state. I got mad at people who called me in a panic at 7:30pm because of the media coverage. I told them to wait til votes are counted. The media stuck with that story until they couldn't hide the truth any longer. They are scum.

[generally]

>>this all happened while Clinton was winning Virginia by a larger margin than Obama did in 2012

That’s because her co-conspirator, McAuliffe, pardoned 60,000 felons last month so they could vote. Coincidentally, Clinton’s margin of victory was about 60,000 votes.

[William Tell]

One poster suggests that Trump won using a flanking movement through the rust belt. Perhaps we should be thankful for the biased pollsters who failed to warn Hillary of what was happening. Accurate polling may have allowed Hillary to win by helping her focus her efforts and craft issues that would appeal to that part of the electorate.

[Ken H]

I can't remember which pundit, but one made a comment that the dems biggest mistake was not cranking up their fraud machine in Wisconsin. They didn't think they'd need it.