Pollster.com

Rasmussen gave you an answer on statewide primaries vs. presidential primaries. However, he didn't give an answer (that I saw) on the recent actual special elections for House seats, in NY, PA and Hawaii, or even for the MA Senate seat. Those were certainly interesting.

I have another gripe with him as well. He also seems to poll statewide races with aggressive comments about the issue of

(second half of post seems to have gotten truncated)...What I said was "He also seems to poll statewide races with aggressive comments about the issue of incumbents getting

I want to hear Glen Beck sing Crazy Train by Ozzy on his next program. The song defines him quite well.

So now the "Rasmussen problem" is that their final poll was 5 days (perhaps 7 days) ahead of a primary, when you wanted them to wait until 4 days before? Nothing actually wrong with the data?

Assume Rasmussen's model is incorrect. Assume that Scott Rasmussen knows it and realizes it's too late to correct it without ruining himself. Assume that the industry standard mechanism for grading pollsters involves comparing final weekend polls to election results. Why would Rasmussen poll on the last weekend and put his reputation any more on the line than it already is? There is absolutely no incentive to poll the last weekend...

Here is a little good news on the economy. We have a long way to go, but there has been an improvement. http://www.gallup.com/poll/122840/Gallup-Daily-Economic-Indexes.aspx

, June 01, 2010 (Courtesy not of MSNBC or the New York Times, but of Rasmussen Reports)

The Rasmussen Employment Index rose just over a point in May to reach its highest level since September 2008. At 74.8, the monthly measure of worker confidence is up eight points from a year ago but still down 13 points from September 2008.

Now, 19% of workers report their firms are hiring, up one point from last month and the highest level reached since September 2008. However, 26% of working Americans say their firms are laying workers off, down slightly from last month. May was the 20th straight month during which the number of workers who report their firms are laying off was higher than the number who said their firms were hiring.

Among private sector managers, just 21% say their firms are hiring while 26% say their firms are laying people off, marking an improvement from last month.

Typically, an increase in the Employment Index suggests that the government’s official report on job creation will be stronger than the month before.

Markos Moulitsas fuming about polling results? Is this some sort of joke? The polls he posts on his site are, without a doubt, the most inaccurate-to-the-point-of- being-ludicrous polls I have ever seen. I can't think of anyone who takes the DailyKos/Research2000 polls seriously as their conclusions are more dubious than Newsweek polls.

It is hard to say what is creepier, the left's obsession with Sarah Palin or its obsession with Rasmussen polls.

"I want to hear Glen Beck sing Crazy Train by Ozzy on his next program. The song defines him quite well."

And the award for "Most Irrelevant Comment" of the year goes to...

"Markos Moulitsas fuming about polling results? Is this some sort of joke? The polls he posts on his site are, without a doubt, the most inaccurate-to-the-point-of- being-ludicrous polls I have ever seen."

Oh come now, what about Zogby Interactive internet polls?, or some of the ARG democratic presidential primary polls, both of which had 'misses' of over 20%.

Rasmussen currently has the greatest volume of polls and a noticable and significant trend of being more favourable to the republicans than the poll average. Rasmussen may be right in their poll projections, but by not polling just before an election they are making it very difficult to tell whether their model is accurate. It is reasonable to ask why this is.