There are myriad reasons that Sen. Clinton has done rather poorly of late, not all of it the fault of her campaign. No one can honestly say the coverage of her has been fair. Every mistake magnified and all references to her faults in gender specific terms. It seems to have been the usual Washington insider game, the villagers are restless. What has disturbed me the most about this campaign is the amount of vitriol heaped on Sen. Clinton from many in the blogosphere, like her or not she isn't the devil as some would have it. I believe the Obama love is misplaced. He is not really very progressive (he did afterall support Joe Lieberman) and his stance on Social Security needs serious scrutiny. I hope for all our sakes "the skies open up, the light shines down and the celestial choirs sing" cause we're going to need all this harmony.

Will the last one out of Clinton campaign HQ please turn off the lights?

I live in Pennsylvania and we haven't even gotten started campaigning here yet--I'm going to my first volunteers for Obama meeting tonight! (I might add that I'm a middle-aged white woman...) Oh, I was also one of the million donors, the first time I've ever given money to any candidate.

How things go Tuesday will largely determine how things go in Pennsylvania come April. So if I may return to Texas for a moment, why in blazes are people so confident that Obama will win there? I hear it spoken of as a virtual fait accompli. But the SurveyUSA Texas poll released night-before-last shows that the 25% who have already voted in Texas have gone for Hillary 51% to 46%. Someone in here said that this reflects only the "early" voting and is therefore nothing to be concerned about given other positive news for Obama. But is that how SurveyUSA explains Hillary's first-quarter win? The data-gathering for the poll started Saturday, and the early voting started just three days before that. I'm assuming that the data for the early voting is distributed over the entire three days and not concentrated in, say, the first day. In other words, Hillary's win in the "early" vote doesn't sound so early to me, but something quite current. I'm not good at interpreting polls - in fact I'm rotten at it - so would someone explain to me all the confidence I'm hearing about Obama taking Texas? How much does he have to win of the remaining three-quarters of the vote in order to overcome Clinton's upfront lead of 5% in the first quarter? And given that pretty much all the other polling in Texas shows the race to be a dead heat, is it plausible, obvious, likely (you choose the word) that he'll be able to make up the difference in time? Please help me think through this to a more optimistic frame of mind.

My feeling is that if Obama doesn't win the popular vote in at least two of Tuesday's contests (TX and VT being the most likely), Hillary will be able to raise more money, continue on into PA, and have a good chance of winning there as well. In that event, even if Obama manages to hold Hillary at bay in the delegate count, he will arrive at the convention in a disheveled state, with a cloud of dubiety hanging over him. In other words, there will be a knock-down and drag-out fight in Denver which will bode ill for the general election. Obama has to win at least TX and VT outright and keep it close in OH in order to ease Hillary out of the race soon. A slight delegate gain won't affect the psychology of the race, which is media drive as well all know. And no one has convinced me yet that Hillary won't actually win Texas. That would give her the upper hand with the Supers and maybe the inside track to the nomination. I did say maybe.

I hope someone will soon show me that I'm a complete fool as a political prognosticator. I promise I'll sit in a corner with a cone hat on my head and giggle gleefully for the remainder of this election.

I have to question the basis for the assumption people seem to have that Obama will do better in the Texas Caucus than the Primary there.

I never bought the Clinton claim that caucuses somehow favor Obama, and I don't really see a reason why he should do better in one half of the "Texas Two Step" than in the other.

I'm amazed no one's noticed ... that this is GREAT NEWS!!1!! for HILLARY!!!1!1!!

I'm certainly no polling expert, so what I am about to say may make me appear to be a fool, a charlatan, or both. I think the fact that SUSA's latest poll has Clinton up by 5% in early voting is a product of SUSA's likely voter screen, and that, in fact, it portends a huge Obama advantage in early voting. There have been a number of reports (I unfortunately do not have a link handy) that early turn-out in Texas has been mind-blowing, sometimes exceeding 2004 by a factor of ten. The turn-out differential is biggest in areas that are expected to go heavily for Obama. However, I don't think SUSA's polling model likely takes this increased turn-out into account, but instead uses a likely voter model based in part on previous voting patterns. In other words, there are likely a ton of early Obama voters not being factored in by SUSA. For this reason, I fully expect Obama to out-perform the polls when it comes to early voting in Texas.

Here's one recent explanation of their methodology:

26.02.2008

A Peek at SurveyUSA's Methodology

After the California primary, I mentioned that what set SurveyUSA apart from all the polls (and prognosticators) predicting a better Obama showing was that SurveyUSA had a great likely voter screen that foresaw the huge surge in Latino turnout. Mark Blumenthal over at pollster.com has some information on how it works (click through for more details):

SurveyUSA's approach to likely voter modeling is comparable to that used by Iowa's Ann Selzer, in that they do not make arbitrary assumptions about the demographic composition of the likely electorate. As SurveyUSA's Jay Leve explains, they "weight the overall universe of Texas adults to U.S. census" demographic estimates, then they select "likely voters" based on screen questions and allow their demographics to "fall where they may." So some of the demographic variation from survey to survey is random, but large and statistically statistically significant variation should reflect real changes in the relative enthusiasm of voters.

But arbitrary assumptions are what make the process so much fun! Anyway, it's anybody's guess whether this method will work as well in Texas as it did in California, but it results in a 49–45 Obama lead, thanks in large part to exceptionally high predicted black turnout and relatively lower Latino turnout (although, perhaps surprisingly, Obama does slightly better among Latinos than he does among whites).

--Josh Patashnik