"Teh momentum" is a media device used by Republicans and media throughout the past 20-30 years to depress Democratic will to vote (we are always less likely to go to the polls). We all know it exists in part from our personal experience with Bush v. Gore ("there is no difference, have a beer"). The very same thing is occurring in poll analysis and media narrative at this point in the Democratic primary. (No conspiracy, I mean each side has a will to see the results as they want to).

These single-buyer commissioned polls (like USA Today or CNN) have a say in what the likely voter model and sampling method is. Look at Rasmussen's report from today:

The current survey projects a slightly smaller turnout from Hispanic voters (26%, down from 31%) and a very slight increase among African-American voters (21% up from 19%). It is impossible to project exactly who will show up and vote in a Primary. Rasmussen Reports analyzed a variety of turnout models and found plausible results ranging from a two-point Obama lead to a five-point Clinton lead.

In part, the media outlets are choosing the polls to fit the narratives. So, I think we Clinton supporters can validly push back on "teh momentum" narrative by citing the national tracking polls at Gallup and Rasmussen where the likely voter model is consistent day to day. Also, at the state level, what the hell is going on with analysts projections for Hispanic voters, when one poll (CNN) says 30% (as it has been in SW exit polls), and another poll (SUSA) says 13% (following a 17% drop).

Lastly, SUSA says 25% of voters have voted, with Hillary at +5%--but projects that the remaining voters will go for Obama by +6-7% for the rest of the week and total. That is all predicated by the accuracy of their model for those who have not yet voted. Right now, they only win if Clinton's voters give up.

"Of the 25% of respondents who have already voted, it's Clinton 51%, Obama 46%."

A CNN poll released earlier today with roughly the same percentage difference (-4) changed their estimate of the Hispanic proportion from 31% to 26% of Democratic primary voters from one week to the next. But, that poll suggested that Hispanics still favored Clinton 2 to 1.

SUSA says more like 4 to 3, but does not mention their proportion of the likely Democratic primary voters. No state in the Southwest has demonstrated anywhere near the minor margin for Clinton by Hispanics that SUSA suggests, the exit polls from California to New Mexico are ~67%.

I think we all know that much can happen, in either direction, in a week.  But I also recall that just before Super Tuesday, the NYT (not that they deserve much respect these day for the McCain story, but they're still the NYT) had a 'hit piece' on BC on Thursday (re: how he got the money to help fight AIDS in Africa) and then the Excelon piece on BO that Sunday (Of course no one seemed to care about him lying about passing that bill in Iowa, nor his ties to big energy).  

obama now atop clinton

as part of the headline was a clue as to how seriously i should take it: not so much.

one interesting note: it showed obama with 85% of the AA vote. what it failed to note was how much of their sample was AA, and what % of texas's population is AA? presumably, they know the answer to the first, else how would they carve that % out of the total?

Hillary has to get tournament tough as they say in tennis.

Maybe she'll stay in just to prove she's her own women and  doesn't listen to all of her husband's advice :)

I think it's perfectly legitimate for Clinton to fight through March 4, and don't for a minute buy the arguments that she should bow out now. But having staked so much on these two states, if she loses one and can't pull off a decisive win in the other (Ohio, which I think she wins by less than 10), she should wrap it up. I think she will, because she's a good Democrat and team player.  And she'll also be thinking of her own legacy, which will be cemented by bowing out gracefully.

You telling me that you REAALY beleive he said that with a straigt face?

simply put, i'm beginning to think sen. obama has drunk from the endless well of his supporter's (you included) magic kool-ade.

i found the above very interesting, with regards to the texas primarycaucus: according to this hispanic female, her male counterparts are all basically neanderthals, who won't vote for sen. clinton solely because she's a she, and might somehow harm their personal macho image.

Most voters, those on the blogosphere notwithstanding, are not as wedded to a candidate as polls might suggest.  Polls have consistently shown that most voters would be happy with either candidate.  If this is truly case then, inevitability, momentum and "being on the winning team" matters.

Given the current polls, it appears likely that Hillary Clinton will lose TX outright and perhaps get a 1-2% win in Ohio - though recent polling seems to suggest an increasing rate of slippage.