Don't look now, but Romney is tanking in Rasmussen's tracking poll...

[Drunken Irishman]

Went from +8 to +4 in a day.



More proof Rasmussen shouldn't be taken seriously. That wild of a swing in a day proves just how ridiculous their numbers are.

[ChairmanAgnostic]

1. On the other hand, he is pandering to Liberty U

And that has to turn off a lot of muddle of the roaders.

[1StrongBlackMan]

2. I have a theory ...

Rasmussen IS a very good polling outfit with a brilliant business model.

See if one tracks Rasmussen's polling results over time, one would find that they are gop/conservative-leaning outlyers early in the cycle. Then, their polling results trends closer to the polling pack. By the election, their polling results are actually fairly accurate.

So for 40 weeks out of the year, they serve as a propandga outlet producing psuedo-scientific; but competely biases, polling information. This earns them a steady revenue stream from gop/conservatives as Rasmussen produces polling information supportive of gop their talking points.

Then for the remaining 12 weeks in the election cycle, they produce increasingly accurate polling data, so by the election they are spot on; thus earning them a reputation for being a legitimate polling outfit, but without jeopardizing the revenue stream from their gop/conservative clients.

[railsback]

3. That's always been my theory

Batshit crazy polling to work up the base, and then as the election closes in, realistic numbers to make sure their base doesn't get complacent.

[LiberalFighter]

4. They have it more accurate

because it gives them credibility even if they are way off base in the beginning.

[Jackpine Radical]

6. I think you've got it.

[fujiyama]

5. I think tracking polls are useful for longer term trends

over a several week stretch but it's very difficult to take them seriously for single days. They just seem way too volatile.

Other polls are OK for snapshots at a time, but things can change drastically this far out. Oddly enough though Rasmussen was dead on in their final prediction for 2008. http://blog.chron.com/txpotomac/2008/11/the-list-which-presidential-polls-were-most-accurate/

[1StrongBlackMan]

9. That's the exact point ...

They ARE a very good polling outfit; but their business model has them serving dual purposes.

Their "final Prediction" polling data was release about a week prior to the election. At that point, they had fully shifted from their propoganda function to their legitimate polling function.

[Indykatie]

7. The +8 Lead was BullSh*t to Begin With n/t

[Cali_Democrat]

8. Did Drudge blare his siren upon hearing this news?