KY: Rasmussen: Paul dominates Conway

[Small Business Owner of Any Repute]

Lots of more interesting stuff about why today's poll is very flawed, to say the least.

Again, there is no proof that the poll is flawed. You may think it is flawed, but until you see a general election result (or even other polls in this space), you have no reason or ability to say that for sure.

Yeah... 

Apparently you missed the fact that the whole point is that Rasmussen DOESN'T POLL close to actual elections for some nefarious reasons.

Yes. Clearly, the reasons are nefarious. The fact that they don't do a poll on the literal eve before the election can only be explained by ill intent and evil scheming.

[Landslide Lyndon]

Lots of more interesting stuff about why today's poll is very flawed, to say the least.

Again, there is no proof that the poll is flawed. You may think it is flawed, but until you see a general election result (or even other polls in this space), you have no reason or ability to say that for sure.

Yeah... 

Apparently you missed the fact that the whole point is that Rasmussen DOESN'T POLL close to actual elections for some nefarious reasons.

Yes. Clearly, the reasons are nefarious. The fact that they don't do a poll on the literal eve before the election can only be explained by ill intent and evil scheming.

Or perhaps all the other polling outfits that do exactly that are dummies and Rasmussen is genius.
I wonder what's more possible?

[Small Business Owner of Any Repute]

Lots of more interesting stuff about why today's poll is very flawed, to say the least.

Again, there is no proof that the poll is flawed. You may think it is flawed, but until you see a general election result (or even other polls in this space), you have no reason or ability to say that for sure.

Yeah... 

Apparently you missed the fact that the whole point is that Rasmussen DOESN'T POLL close to actual elections for some nefarious reasons.

Yes. Clearly, the reasons are nefarious. The fact that they don't do a poll on the literal eve before the election can only be explained by ill intent and evil scheming.

Or perhaps all the other polling outfits that do exactly that are dummies and Rasmussen is genius.
I wonder what's more possible?

The last-minute poll is hardly a universal occurrence in the polling world, and it's not as if Rasmussen is AWOL for months -- their last in Kentucky was two weeks before and their last in Massachusetts was one week before. Their poll of Pennsylvania's Democratic primary was a week prior to the election, and the numbers look damn good (Sestak +5).

The standard Rasmussen MO is to react to news that has already happened, not to news that will soon occur. From a pollster's standpoint, there's more value and less risk involved in the former than the latter. You may not agree with what ultimately comes down to a business decision, but I see no reason to question their methodology or motive. Personally, I'm glad that someone is out there doing a post-election poll on Kentucky, a post-scandal poll on Connecticut, and a post-party-switch poll on Florida. Even if both of those polls are done in turbulent waters, it's nice to have at least some kind of snapshot of where each race is or could be going.

And thanks to Rasmussen's committment to regular monthly polling of key races, we'll know soon how short term those bumps in Kentucky and Connecticut are.

I still don't understand what's so nefarious about them not doing a poll the day before the election. And I don't understand what's genius or "dummy"-like about it either. It's nothing more than a time point. No one is saying that Rasmussen is the gold standard of polling, simply that they're generally accurate (and have been quite good at polling past cycles) and that there's no reason to believe that is no longer the case.

What I do understand is why you (and Daily Kos) would be interested in a partisan hit job on them, of course...

[Landslide Lyndon]

I can't believe that you didn't understand that I used the word nefarious humorously. Your obsession with it amazes me.

As for the rest, what you're saying is that Rasmussen don't poll close to elections because they want to cover their ass in case of a mistake.
Fair enough, but don't expect me to take them seriously when they are AWOL in races like PA and KY where there is a national interest.  PPP, Quinnipiac, Suffolk, SurveyUSA, went into the field during the last days for that reason.

P.S. If there are any doubts about Rasmussen's agenda/leanings, just look at today's Arkansas poll.

[Dan the Roman]

Mr. Moderate may well be right. And in the case of Massachusetts it may have been a one-time thing. The problem here is that Rasmussen is a narrative pollster, as you mention he is able to get a poll out on CT within a day or two of a major event. This indicates not that he was close to the primary and correct in PA or the first one to pick up movement in MA, but that he simply chose not to repoll when it might have been tested. Nefarious? Maybe, maybe not. It is worht pointing out that with 3-hour polls nothing stops them from doing this, and without clients its entirely their choice. So we are left with the fact that for some reason, that can not possibly be lack of time or money, Scott chooses not to poll. Maybe Zenu told him not to?

What I will say is that there is nothing to indicate he is a reliable pollster this year. As Strategic Vision showed, it is possible to make up semi-plausible results out of thin air. Nothing Rasmussen has showed this year, is something one of us on this board could not have come up with by randomly guessing where one of these races were. I myself, while in Turkey over Christmas, when asked where I thought the MA Senate race was, gave the number 51-40. And it was pretty hard to be much off in PA, given everyone had the race in a similar place. 47-42 is easily generated. I really think it would be interesting to create a daily random number thread and see if it is just as plausible because I have little doubt it would be. Strategic Vision was also a very accurate pollster in 2004, 2006, and 2008, at least in their political polls. Thats why they took so long to be caught.

I understand why people don't like Rasmussen's results, and I was in your boat to a large extent last year. But the build up of evidence that Rasmussen is far more POS than independent has been continuous, that they have a political agenda overwhelming, and that they no longer take clients for their polling, in my view, far more damming than the fact 1 in 20 for them seems more like 2 in 5. It was one thing a few months ago when PPP used to stand with them. But given the comparative performance of the two firms recently(and PPP's performance in MA was vastly superior to Rasmussen's they did a poll two days before that was 52-46, Ras was 48-47 a  week out) I don't see how anyone can seriously choose to stick it out with Rasmussen over PPP.

[Badger]

Somebody posted the internals over at Swing State project and they are a thing of beauty.

Democrats give Paul 53% favorability, and even liberals give him 42% favorability.

Under "race," Rasmussen offers choices of white, black, and "other," and "other" favors Paul over Conway 80-7.
And Rasmussen has Paul winning 25% of black voters.

Even if these numbers are accurate, they sure won't hold come November. Not even close.