Just a quick polling update here from SurveyUSA and Rasmussen - also a few points from ARG (for what they're worth) in North Carolina and Gallup.  Essentially, Obama is holding steady or improving by virtually any measure. No recent poll nationally, or in the upcoming states, has him suffering significantly from his 9 point loss in Pennsylvania.  Quick numbers:

As of this moment, about 9:25 PM Eastern Time, there are less than 15% of precincts reporting, a Clinton margin of 6%, and most networks have already called Pennsylvania for Clinton.  But I took a quick look at some critical numbers, and this thing could end up really, really, really close.

This is my final prediction. After analyzing many polls, I come up with the likely composition of the PA primary tomorrow. I entered the likely percentage for each candidate in a spreadsheet provided by Pollster.com and surprisingly Obama can win and will win the state.

PA PA PA.  That is all I hear!  Let's imagine Hillary Clinton does what many will expect, which is win PA by at least a high single-digit margin.  She'll probably stay in.  If not, she'll probably drop out.  I say there is a 60%-40% chance that Clinton will not achieve the high single-digit or double-digit margin.  Yet April 22nd isn't the date that matters, it's May 6th.   North Carolina and Indiana will vote on May 6th.

I'll go through Obama's hidden strength in Indiana and then highlight the ultimate reason for why this will be over by May 7th.

According to WCAU-TV, one of the sponsors of the SurveyUSA PA polls, Clinton is up 14.

No crosstabs yet, but my thoughts below the jump.

Some misguided individuals may write polling diaries on presidential race polls, like the poll showing Obama up 8 in Pennsylvania or down 17 in North Carolina.

But nobody writes about the most crucial issue facing us these days:

How many Americans have been rickroll'd?

SurveyUSA estimates that at least 18 million Americans have now been "Rickrolled," according to this latest national poll of 959 adults, conducted 04/08/08.

Unsurprisingly, the major difference in the crosstabs is by age group.

7% of 18-34 year olds have been rickrolled, while only 3% of those in the John McAncient demographic have been rickroll'd. Moreover, 66% of 18-34 know what rickrolling is, whereas a majority of those 35 and up have no idea.

Survey USA has produced polling for 14 concluded Democratic Primaries.

Below are the results of the last poll they conducted before each primary date.The actual primary results for each state follows the polling number.

This will make it easier to asses the accuracy of Survey USA in wake of their most recent Pennsylvania polling results.

LINK

Considering that every other pollster has the race within 5 points in the past few days, this seems like a major development. Obama himself said last week that he felt like he was still behind 20 in the state, even though polls were saying otherwise. I'm not sure what this means, whether this is just an outlier, whether SUSA is ahead of the curve in terms of predicting a dramatic shift, or whether every other pollster is just plain wrong. Thoughts?

A new SurveyUSA poll coming out today shows Clinton up by 18 points. Doesn't seem to match other polls, but maybe this is going to be another California. I guess a new Rasmussen poll for Pennsylvania is due out today as well, I'll be interested in seeing what that one shows.

It appears that SurveyUSA is going to come out with a new poll soon showing Clinton up by 12, down from a 19 point lead she had a few weeks ago. If you compare this with the Rasmussen poll it makes sense.
When Rasmussen had Obama down by 13, SurveyUSA had him down by 19. So, each poll shows him gaining 6-8 points.
It's not on the SurveyUSA website yet, but I found a link to the story from their site that was put out by a local Philly TV station website.

Much of the MSM throw NC into the Obama column without much thought, but a recent PPP poll had Clinton down by only a point, well within the +/- 4.3% margin of error.

"Some folks in the national media have been treating North Carolina as if winning it is a given for Barack Obama," said Dean Debnam, President of Public Policy Polling.  "The reality is that the state is pretty close, and either candidate could emerge with a victory."

Could a blowout win in PA create enough momentum to push her over the top in NC?

Tonight KSTP is reporting a new SurveyUSA poll: Coleman 51, Franken 41. Compared to an identical SUSA poll a month ago, this is a gain of 4 by Coleman and a loss of 5 by Franken.

Survey USA poll.

Franken's support among women and in the Twin Cities has weakened since the last poll.

(From the diaries -- kos)

Measuring Size of Obama’s (Relative) Coattails on Congressional Races

This is a long diary to read so I will tell you upfront what my conclusion is after examining the recent SurveyUSA 50-state polling:  Yes, Barack Obama may indeed have huge coattails for Senate and Congressional races in the fall.  Now, to the long explanation --

The recent SurveyUSA polling indicated a relatively even match between Obama and Clinton re. who would have the most electoral votes when facing McCain.  Obama’s total came to 280 EV, while Hillary’s was 276 EV – a seemingly even match:

http://www.surveyusa.com/...
http://www.surveyusa.com/...

SurveyUSA released 50-state Obama-McCain and Clinton-McCain matchups. It's the most definitive showing yet that polling results offer no clues as to which candidate is the most electable.

The reigning questions from objective analysis seems to be 1) How to make this fair taking into account MI and FL. 2)Will Clinton cede the nomination or take it into the mud. and 3)who is the most likely winner.
As of today, it is Obama. Despite the spin from the Clinton and the media. Her wins are a lot less than astronomical She needed huge wins to narrow the gap substantially and didn't get it. The other spin has been about her offer to Obama. The timing seems suspicious at best. First of all, despite her "gaining momentum", basic projections put the delegate nomination well out of her grasp. She would need to win every state by more than 25% to achieve that. Not to mention she is down in the popular vote by over 500,000. For her to ask the front runner to give up his position is bizarre to say the least. Tantamount to encouraging political suicide. The other strange echo has been her claiming to have one NY and CA somehow preclude Obama from winning it in the general. According to the SurveyUSA poll, both candidates handily win those 2 states. NJ is closer for Obama, but should go his way.

Today, SurveyUSA released a general election polling matchup for both Barack Obama vs. John McCain and Hillary Clinton vs. John McCain. From a cursory look at the math, it would seem as though there's not much of a difference between the two candidates: Obama only scores 4 more electoral votes than Clinton. However, as one can tell from looking at the maps of the Electoral College, Obama has a much better chance of expanding the playing field than Clinton does. Follow me below the fold for a more complete analysis...

SurveyUSA has just released their 50-state head-to-head polling results:

While the results look pretty good 280 Obama 258 McCain, 276 Clinton 262 McCain, it's the actual state-by-state numbers in some of those states that are solid red that look amazing (especially for Obama, but also for Clinton in some places)

Alaska: McCain 48-Obama 43; McCain 56-Clinton 34
Idaho: McCain 52-Obama 39; McCain 63-Clinton 27
Indiana: McCain 50-Obama 41; McCain 53-Clinton 36
Kansas: McCain 50-Obama41; McCain 51-Clinton 42
Montana: McCain 47-Obama 39; McCain 53-Clinton 33
Nebraska: McCain 45-Obama 42; McCain 57-Clinton 30
North Dakota: Obama 46-McCain 42; McCain 54-Clinton 35
North Carolina: McCain 47-Obama 45; McCain 49-Obama 41
Oklahoma: McCain 57-Obama 34; McCain 50-Clinton 42
South Carolina: McCain 48-Obama 45; McCain 48-Clinton 42
South Dakota: McCain 47-Obama 43; McCain 52-Clinton 40
Texas: McCain 47-Obama 46; McCain 49-Clinton 42
Utah: McCain 50-Obama 39; McCain 65-Clinton 27
Wyoming: McCain 54-Obama 35; McCain 62-Clinton 28

Full 50-state side by side comparison coming below (give me a sec ...)

So, I've been a fair amount of work on how to weight polls from different pollsters and from different periods of time.  It beats what I consider to the the depressing tenor of the Democratic campaign in recent days, and it's part of a project I'm working on to build some prediction models for the general election, which I hope will turn into an off-site project with frequent updates to Daily Kos.  You ever wanted a version of the RCP averages that is smart enough to weight a Survey USA poll more than an ARG poll?  That's what I'm going to try and do.