I'm simply saying that your sample size for any individual election is small, and when half come from a single organization, not very valid. It's even less valid when that organization has a well-known and well-earned rep for being a right-wing tool, as evidenced by its nonstop use right here on ET. Not exactly rocket science. Even the exit polls had Kerry leading going into 2004, and he lost. As I said before, don't count your chickens before they hatch. You're doing that twice - giving exaggerated importance to these polls, and then thinking you can go in a straight line from 2009 to the 2010 results. Real life doesn't work that way.
I agree that it would be stupid to bet money on the 2010 elections based on the polls of today. However I disagree that these current polls aren't significant indicators of the future. I also think it is idiotic to call Rasmussen a right wing tool when it favors Democrats in this group of polls more than the rest of the polling community. You think Rasmussen is a right wing tool because you don't understand the way they poll people. I am sorry to inform you but they only poll likely voters. EVEN if you completely ignored the Rasmussen polls the change in momentum among parties would be even worse for the Democrats based on the other polls. I am not outright calling the results of the 2010 elections. I am stating the change in momentum among parties. That is all I am doing. Unless you consider every single poll to be completely corrupt then you have no choice but to agree with this analysis. You can agree on this middle ground can't you?
I'd be with you if you hadn't called it "idiotic" to think Rasmussen had a right-wing bias. Let's take a look at the general election results compared to Rasmussen, which as you say polls likely voters, and NBC News/WSJ, which does the same: Obama: +7.3 RCP avg: +7.6 Rasmussen: +6 NBC/WSJ: +8 Looks reasonable? Not really. Rasmussen claims a margin of error - MOE column - of 2 points. The 7.3 margin is therefore uncomfortably close to the +8 which constitutes the tail end of their margin of error. NBC/WSJ claims a margin of error of 3.1. Their +8 nevertheless came closer to the actual result than Rasmussen did. Was this an isolated result? No. I went through all the polls labeled "Rasmussen Reports" on the page for last year's election, and compared it to all the polls done in the exact same time frame. The cumulative deviation from the mean for all of these polls was -4.53 in terms of measurement of the margin for Obama. Rasmussen was lower than the mean 3 out of 5 times. The highest over the mean was .8, the lowest under the mean was -2.33. That's pretty clear evidence of an anti-Obama bias. All except one of the polls Rasmussen was compared against for my little study were polls of likely voters, same as Rasmussen, so you can forget about claiming that made a difference. There's a reason Rasmussen is the only polling organization I've ever seen quoted in the P&R forum.
Rasmussen could favor Republicans overall. I just found it funny that you continued to call the firm a right wing tool even when it favored Democrats more than the rest of the polls combined. I called that idiotic because I have never spent my time trying to figure out which firms are fair and which are not because I don't look at just one poll. I believe firms have different ways of polling that create different results. Do I believe all of them are fair? Of course not. Let's theoretically replace all 19 of those Rasmussen polls with another polling firm. If there were a total of 40 polls and 19 of those polls were from just one firm than it would be agreeable to ignore a significant amount of those polls in order to find an average. Thats what I do when I look at polls. I look at the overall average. When I converse with people about polls I tend to assume other people have the ability of grasping the overall trend without jumping into a frenzy and ruining the entire conversation because of one polling firm in the group.
I was being extremely, extremely merciful to Rasmussen. Remember, in the case of a poll on an event that is going to happen in the real world, and therefore act as a real world check on your work, you'll try much harder to get it right than you would otherwise. The bias shown in my little study expands to a degree that can only be described as completely crazy when you look at the Obama approval page. Here, there is no check on Rasmussen's work. No one will ever know for certain just how off they are. All you can do is look at what everyone else reports, look at what they report, and shake your head and wonder just on what planet the folks they poll live on. Of course, what's really happening is that they will bias the folks they ask, bias the question, bias pretty much anything they can, to get the result they want. Once they're done with that, some clueless ET right-wing trailer park tool will pick it up and post it as evidence that some Obama policy/proposal is highly unpopular. Nothing of the sort will be true, of course, but it doesn't really matter. Truth and politics have a tangential relationship, but in the old days when Gallup was pretty much unquestioned in its dominance, we knew we had a pollster who was at least trying to get it right. With Rasmussen, no such thing is even contemplated. All they want is to confirm their own biases, and the biases of those who quote their polls, which are not just useless, but wrong. Demonstrably wrong.
Apparently you haven't picked up on my tone. My tone is that I could care less about one polling firm.
with over 10,000 comments from your handle alone, you are do props and honors, however, nothing can be as negative and blind sided as the Fox Business news networks coverage of current events... rose colored glasses would climb down off that persons face and simply walk away, given how they conduct themselves and lopsided news reporting... no ET not Fox.... not even close,