Texas conventions: We will finally do what to do with the Texas caucuses!

Two weeks ago, Iowa held its county conventions, resulting in Obama gaining a net 10 delegates. Today, it is Texas's turn to do the same. The process is simple: Thousands of delegates elected during the March 4th caucuses convene at conventions across the state and vote on the second stage of the attribution of 67 delegates!

You might remember that Texas Democrats reported only the results of 41% of the precincts, and we never got to know the rest of the results. Based on that, many in the media decided to go ahead and allocate 38 delegates to Obama and 29 to Clinton despite the fact that, in a number of districts, the candidates were teetering on the verge of the thresholds that would determine whether they would get an extra delegate with more than half of the ballots counted.

To make matters worse, the whole Texas voting process had bordered on complete chaos, with long lines, hours worth of delay, voting taking place in parking lots... As a result, the Clinton campaign sought to delay today's conventions and considered filing challenges, but ended up renouncing to do so. But this report in the Dallas Morning News should be enough to crush hopes that things would run smoother today. The process looks to be delayed by hours in many places, and challenges to the credentials of many delegates are apparently being filed, resulting in anger, frustration and shouting matches!

Besides the chaos and the shady nature of caucuses, today's voting is significant because we will finally know how to deal with the Texas caucuses, and how to assign their delegates. This second stage of the caucus process is in many ways as important as the first, and as much of a challenge with the two campaigns having to put together full scale turn out operations. And keep in mind that today's conventions will also elect delegates... which will then meet again to choose the final round of Denver delegates!

With 26% of precincts reporting, Obama is ahead 54% to 46%, but different parts of the state are reporting very unequally right now. As expected, Obama is getting some big numbers in hte conventions of Austin, Dallas and Houston (as a commenter points out, that includes a 272-52 in Harris County's 13th district portion). Hillary has her strongholds as well. In Webb County, in which Obama had gotten 21% of the primary vote, he failed to get 15% today at the county convention, which means that Clinton gets all the 51 delegates from there. In Hidalgo County (also on the border), it's 102-11 for Clinton.

Update: With 82% of the delegates now allocated, Obama leads 55.9% (3851 delegates) to 44.05% (2804 delegates). Only 55% of conventions have reported for now, and most of the remaining ones are from small counties that typically favor Clinton so the margin could still slightly tighten. We should know the rest of the results in the coming hours.

Caucus chaos: Obama gains in Iowa, Clinton contests Texas (Updated with new Iowa numbers)

So much for the pledged delegate count Campaign Diaries and most media groups have been keeping up. More than 13 thousand delegates elected on January 3rd convened today at Iowa's county conventions and they threw the delegate allocation numbers we had all relied on out of the window. (I explained how this was possible, and the stakes of today's convention, in this post yesterday). With all of the county conventions now completed, the Obama campaign can claim a big victory.

Based on the vote at the county conventions, the original delegate count out of Iowa (16 for Obama and 15 for Clinton) will change dramatically and reflect a significant gain on Obama's part: He can now claim 25 delegates out of Iowa, while Clinton lost a delegate and is now at 14 with Edwards losing eight to end up with 6 (these delegates will be up for grabs once again in the next round). With almost all delegates for the state convention selected, Obama obtained 52%, Clinton 32% and Edwards 16%, which more or less means that almost all Edwards's delegates who did not stay with their candidate switched to Obama.

This is obviously a huge gain for the Illinois Senator. His 10 delegate increase alone is superior to Clinton's edge out of Ohio! And this from a much smaller state like Iowa in which Obama got 9% more than his rival: With 38% of the vote, Obama can now claim 55% of the state's delegate, the biggest disparity we have yet seen in a system that looked so strictly proportional. Given that Clinton is already so far behind in the pledged delegate count and that we can hardly imagine her catching up in pledged delegates with all the remaining contests, this is obviously a blow to her chances.

But how is this different from superdelegates deciding to vote for whomever they want without respecting the will of voters? Iowa voted on January 3rd -- but the delegate allocation will not follow the will of that day's voters. And not only because Edwards delegates backed Obama, since Clinton appears to have lost a delegate! After all the vociferation about respecting the will of voters, this does not seem like the logical next step.

The caucus system is looking increasingly disastrous for Clinton. It means not one caucus but a whole series of caucuses, each further diluting Clinton's votes. Fortunately for the New York Senator, she has much less to fear in most other caucuses, for she has already reached such depths in most of them that she can't possibly go much further down (in Idaho, for example, she only has 3 out of 18 delegates). But there are two danger areas: Nevada, where she won the popular vote and has to defend 12 out of 25 delegates, and Texas.

The Texas caucuses are still reporting at 41%... which has prompted me to not include the delegate breakdown of that contest in my delegate count. The projection out of those 41% gives 38 delegates to Obama and 29 to Clinton, but both candidates are too close to the thresholds of one less/one more delegate in too many districts that have barely reported with a third of their precincts for a definite breakdown to make any sense (it is worth noting that some networks, like NBC, are also refusing to project final numbers).

Worried about facing the same problem in Texas that she did today in Iowa, Clinton is now openly contesting the March 4th caucuses as she had been threatening to do for a while. Her move today was to send a letter to the Texas Democratic Party, as reported by the Dallas Morning News, to ask that the March 29th county conventions (exactly what happened in Iowa today) be postponed. Why? Because she wants the identity of caucus-goers to be double-checked, and she cites a number of violations including votes being taken with head counts instead of written rolls.

It is true that the Texas caucuses were a stunningly chaotic affair, with officials having utterly failed at organizing them and preparing for the turnout. They were reports of caucuses starting hours late, voting taking place in parking lots in the dark and with flashlights because there was not enough room in the actual caucus location, etc. There was no evidence of fraud -- both Obama and Clinton threw such allegations at each other -- simply the incompetence of the caucus organizers and government officials. In this context, Clinton's move to contest the county conventions could be interpreted as a delaying tactic and one born out of genuine worry as to what the March 29th conventions could have in store.

To make matters worse, it's hard for us who cover the campaign to know what to expect and what is really going on in Texas when we have no access to the 59% of the results of a contest that is deciding a total of 67 delegates -- and I am very much afraid that most of the precincts which are not reporting are those where media reports say voting took place on parking lots.

Delegate breakdown, March 13th edition (still without the TX caucuses)

The post-March 4th delegate breakdown had Obama up 143.5 pledged delegates (without counting the Texas caucuses). A week later, Obama has won two contests -- Wyoming and Mississippi -- that help him make up some of the little ground he lost on March 4th.

I am still not including the delegate attribution out of the Texas caucuses. While that contest has been called for Obama now, only 41% of the precincts ended up reporting their vote publicly... the rest will just send delegates to the district-level convention. Most news agencies are now projecting a 9-delegate lead for Obama in the caucuses, but that's a very uncertain number: In most districts less than half of the precincts have reported, and in many of them Clinton and Obama are hovering around the threshold number (59%, 62% depending on how many delegates each district has) to get one more or one less delegate. This 9-delegate advantage could very easily change, and it does not seem logical to project the final results and include it in our totals before we get the final picture.

Frankly given how chaotic, unprepared and undemocratic the Texas caucuses ended up being I really see no reason why we should entertain the absurdity of their refusing to report results and pretend that they have.

Proof that delegate numbers can change with just a handful of votes being added to the totals: In our last delegate breakdown I reported that the California final allocation had changed once the provisional and absentee ballots had finished being counted, with Clinton losing 4 delegates and Obama gaining as many. This week, the same thing happened in Colorado. As the final voting total has been certified, Obama lost a delegate, and Clinton gained one: The new total is 35-20 instead of 36-19, and that change is now reflected in my delegate totals. And then we have the two states that voted this week:

Wyoming: 61% Obama-39% Clinton

• Obama: 7 delegates
• Clinton: 5
  

Mississippi: 61% Obama-37% Clinton

• Obama: 19 delegates
• Clinton: 14
  

In both races, Obama barely missed a better delegate breakdown -- just half a percentage point in Mississippi and he would have improved the statewide allocation from 1 to 3 delegates. And that brings us to the following totals of pledged delegates, including those from all the states that have already voted:

• Obama: 1368.5
• Clinton: 1223 delegates
• Outstanding: 67

The Texas caucuses exemplify all that is wrong with the process

Forget the New Mexico primary. The Texas caucuses have to be considered the cycle's most absurd election. For one, the state was entirely unprepared to face the massive turnout yesterday: Some caucuses started hours late, others were simply too crowded to allow everyone in the room. And now, 36 hours after the caucuses started, only 41% of precincts have reported their total, putting Obama ahead 56% to 44%.

Most media outlets are now ready to estimate the delegate split and give the caucuses to Obama by a large margin based on these early results, under the assumption that the 40% that are in are representative of the state at large. But there is no reason to believe that to be the case, quite the contrary: 76% of Travis County (Austin) has reported, and 48% of Dallas County, and 45% of Tarrant County (Fort Worth), all Obama strongholds and all above the statewide percentage of precincts reporting (41%). Austin's numbers are especially important, given how many voters there are there, so that 76% of precincts reporting will significantly distort the results.

To make matters worse, TPM is now reporting that we might never know the results of the Texas caucuses! Here's why: The county caucuses elect delegates for the district convention which elect delegates for the state convention which elect delegates for the DNC convention. Complicated enough, but that's pretty much the same as in any other caucus. But the difference is that the caucuses "weren't required to report to the state party headquarters. The results that have come in came from a voluntary system set up in order to help the media get an idea of what to expect the final delegate result to be!" So every caucus will just send its delegates to the district convention without ever telling the press or anyone who it is sending?!

In other words, we might never know how the remaining 59% of precincts voted, and as I said before we're talking about a smaller portion of an Obama stronghold like Austin.

Now, other reports are indicating that Texas Party is responding that it simply wants accurate reports and is taking its time -- not that it won't release the results at all. If anything, this just confirms how chaotic Tuesday night was at these voting places. And if TPM's report is true, the Texas Party won't even have to divulge its results publicly: their results could be as unreliable and chaotic as the process was, but there would be little we could to do hold them accountable.

For all it's worth, estimations have Obama gaining 7 delegates out of the caucuses based on the results we have right now, which would mean he gets 3 more delegates than his rival out of the entire Texas process. Overall, that would mean Clinton only got a net 8 delegates out of March 4th. But as I said, I want to see more of an indication of how Texas caucuses are shaping out.

Increasing the confusion, Democrats refuse to settle

Democratic voters do not want this race to end. Given the opportunity to hand the nomination to Barack Obama, they rallied behind Hillary Clinton and made sure the race would go on until -- at least -- April 22nd. So however burned out you feel, prepare for more debates, more Election Nights, more primary polls and more delegate counts. We are in this for the long haul.

There were two separate questions to be answered last night: First, would Clinton be able to survive and stay in the race? Second, would she be able to change the fundamentals of the race and win comfortably enough to get herself back in the nomination battle? The answer I offered yesterday to the latter -- more interesting -- question was that she needed "to triumph in Ohio by double-digits and win Texas comfortably enough to get a delegate lead (and probably win Rhode Island as well). A tall order for the New York Senator."

Incredibly enough, Hillary pulled that off: She won Ohio by 10%, posted an unlikely triumph in Rhode Island (by a margin bigger than in New York) and while she only won the Texas primary by 4%, she managed to pull out a delegate lead thanks to unexpectedly strong numbers in a handful of border districts where she managed to gain 3 out of 4 delegates. Boosted by those very strong results, the Clinton campaign powers on.

So this means that Clinton now has a stronger chance to become the Democratic nominee, right? Not so fast, things could never be that simple in this Democratic primary. Campaign Diaries's count prior to March 4th gave Obama a 148 pledged delegate lead. The compilation of yesterday's results shows that Clinton netted between 15 and 19 delegates -- without counting the Texas caucuses since results are still unknown at this hour. That's a much larger number than was expected, and credit goes to the Clinton campaign for having pulled that out. It is also the first Election Day in which Clinton nets a delegate gain (New Hampshire splits its delegates equally and Obama got one more in Nevada)! But those numbers also mean that Clinton was only able to cut into a very small fraction of Obama's overall lead.

The math is overwhelmingly against the Clinton campaign: They have to count on huge margins in all upcoming states (starting with Pennsylvania) and prevent falling behind in states that Obama is heavily favored in (such as Mississippi). Clinton simply does not have a path that gives her a pledged delegate lead right now. Now, the campaign will push for revotes in Florida and Michigan. Adding those states to the mix would give Clinton a reservoir of delegates to still be awarded, increasing her chances of cutting into Obama's lead.

This is what has made the Democratic contest into such a paradox: Mathematically speaking, Obama is the overwhelming favorite to win the nomination. But how can he convince superdelegates to join him and the party to coronate him without winning important occasions such as this? And on the other side we have Clinton, a candidate who is very unlikely to get the nomination but who can now press ahead with new-found momentum.

To make matters worse for Obama, yesterday's vote uncovered a certain number of weaknesses that the Clinton campaign will point to in the coming days in an effort to freeze the superdelegate migration towards the Illinois Senator. The exit polls showed that Clinton held her demographic group very well. She won the Latino vote in Texas nearly 2:1. In Ohio, the groups that had abandoned her in Wisconsin were back in her corner: She won by 18% among those with no college degree, triumphed among the female vote and regained a significant edge among white males. And this time, she cut into Obama's demographics groups: She won the male vote by 3% in Ohio, forced a tie among Republican and independent voters, as well as among the high-income electorate. Obama's demographic coalition was the one that trembled, and his weakness among blue-collar voters and Latinos was once again on full display. This will be Argument A in Clinton's case to superdelegates.

Finally, there is the question of how Clinton pulled this off. In New Hampshire, she had benefited from a last minute boost among the female vote. In Ohio and in Texas, her comeback was fueled by last-minute deciders as well, as she posted some massive leads among those who had decided in the past 3 days. The two stories that dominated the last few days of campaigning were first the questions about Obama's national security credentials (fueled by the red phone ad) and second the controversy surrounding Obama's stance on NAFTA.

Which of these factors was determinant in yesterday's vote? The one hint we have is that Clinton's victory was more overwhelming in Ohio than in Texas, pointing to the NAFTA story being more important. And the Obama campaign should use that as a lesson in crisis control, as they messed up their response to Canada's allegations, not settling on a coherent defense for long enough to make the story even more damning than it was.

The Clinton campaign, however, will probably point to the red phone ad and the spot attacking Obama's Senate record as factors as well. They will take comfort in the late decider breakdown and take that as encouragement to step up attacks on Obama's credentials. Last night's results could prompt Clinton to go much more negative than she has up to now. Given that the press is turning the spotlight increasingly on Obama, the Illinois Senator could be in for a rough 6 weeks.

And this is why Obama will regret yesterday's results for a long time to come. He is still the overwhelming favorite to win the nomination, and should only increase his delegate lead next Tuesday in Mississippi. But instead of turning his attention to John McCain, he will spent the next 6 weeks being questioned by the media and attacked by Clinton. And keep in mind that she is as trapped as her rival is: She must realize how unlikely her path to the nomination is, but she cannot possibly drop out with this kind of results.

Two months after Iowa, Democrats are continuing to go through the motions of a competitive race that could take the intra-party fight all the way to Denver in late August 2008. By all measures, it is difficult to identify a scenario in which Obama loses the nomination. But it is as difficult to see how he can put an end to this race any time soon.

The morning after: Postcripts on results and delegate count

Sorry for the late posting today. It was a late night and I am just getting to settle on an understanding of what happened yesterday. A full analysis will come soon but here are some pointers on the results and the delegate count.

In Rhode Island, it was unclear as of last night whether Clinton would get a 5 delegate lead or have to settle for a 12-9 split. Clinton got her 13th delegate after all, one of the only states (with Arkansas?) in which she really maximized the delegate potential. In Vermont, we already knew yesterday that Obama would have to settle for a 9-3 delegate split.

In Ohio, Clinton barely hang on to a double-digit victory when all of Cleveland was in (though there is still one outstanding county that has reporting no votes at all?!). Still unclear is how big a delegate lead Clinton will get here. My estimation of the delegate count -- going district by district -- gives her a 11 or 13 delegate lead, with a 5 delegate lead statewide and a 6 to 8 district-based delegate lead. The even-district rules hurt Clinton hurt, as a lot of the districts she did well in were 4 delegate districts. She got in the high 50s in a few of them, making those places a wash delegate wise. Other estimations have Clinton's lead a bit higher, so we will know more soon.

In Texas, Clinton has won the primary delegate count, something many people were doubting she could do with such a narrow victory. The count is 65-61 in her favor. As I indicated last night, she did so by getting a 3-1 split in border districts (the 19th, 20th, 21st in particular), places we were expecting to see 2-2 splits. And she had to split many districts 2-2 despite winning them. One big question mark is TX-26 where Clinton just barely missed the 62.5% mark. But in the early voting numbers, the count has Obama at 17,000 votes and Clinton at only... 1 vote, less than Joe Biden (who got 181). There is obviously a counting mistake there that could give Clinton an extra delegate once it is corrected.
In the Texas caucuses, the count appears to have been stuck at 36% reporting for almost 11 hours now, with Obama up 52% to 48%. I am not sure what is going on.

Overall, that means that Clinton scored a net 17 to 21 delegates (depending on what happens in TX-26 and the final Ohio count) over Obama -- prior to the Texas caucuses. The latter are unlikely to have such a big gap, so Clinton looks to have won the delegate count out of yesterday. Though not overwhelmingly, it's more than most people expected, even those who thought Clinton could pull off 3 victories.

In important congressional races, Denis Kucinich survived a tough primary challenge in OH-10 (approximately an 18% lead). In TX-14, Ron Paul survived a primary challenge as well. But the very interesting result is in TX-22, where Republicans selected Shelley Sekula Gibbs as their candidate against Rep. Lampson (in Tom Delay's former district). You might remember Sekula Gibbs was that infamous write-in candidate who then served in the House for 2 months, and attracted countless negative stories. She won this primary with only 29% of the vote, and it doesn't seem that she is going into the general in the strongest shape. (Update: I stand corrected from a very knowledgeable commenter, who points out that Texas has a runoff system in its primaries, so that Sekula Gibbs and Pete Olson will fight it out in an April 8th election for the right to take on Lampson.)

March 4th results thread: The race is shuffled... yet another time!

2am: Contrary to what most reports are indicating, it looks like Clinton could win the delegate count of the Texas primary, despite the fact that she is only winning the state by 4% right now (a margin that could still decrease). Burnt Orange Rerport has posted a very very important breakdown by state senate districts, and it projects Clinton ahead 64 delegates to 62 delegates. Now, we were expecting Clinton to lose the delegate count with such a small victory, but here is why that is not happening: the 19th, 20th, 21st districts -- at the border -- all look like they went for Clinton with more than 62.5% of the vote, which means she got a 3-1 delegate in all of them. She got more than her total delegate lead just in those districts, and we were expecting the two candidates to split most of these 2-2 (In the 26th as of now, Clinton is barely missing the mark and has to settle for a very disappointing 2-2 split, though this could still vary). By building up huge margins in these Latino districts, Clinton is unexpectedly strong in the delegate count and she offsets the fact that delegate are disproportionaletly attributed in Obama-heavy districts. To add insult to inury, it seems that Obama missed the 6-2 split in Austin and has to settle for a 5-3 split instead in the biggest prize of the night! All these districts are stil counting votes, so the delegate allocation can still change -- and I will keep an eye on it.
With that, Clinton goes into the caucus part of Texas with a small cushion. Let's see if that's enough to get her a delegate lead out of March 4th. With 22% in, Obama is up 56% to 44% -- clearly this could be a big result for him.

1:40am: The new updated OH exit polls show that Clinton won by improving her stance in some Obama demographics: She won the male vote by 3%, and only trailed by 4% among college graduates (whereas she won by 18% among those with no degree). She even tied him among families with an income of more than $100,000. Finally, the exit poll suggests that Clinton forced a tie among the 30% of independents and registered Republicans who participated in this primary. She made no inroads whatsoever, however, among black voters, a group she lost 87% to 13%, but her strength among the white vote (64% to 34%) is surprising given how much she had weakened there in the past few weeks -- and especially among white men, a group she had serious trouble with as of late but among which she won 58% to 39%. Also determinant, Clinton has a 16% lead among those who decided over the past week (and even bigger among those who decided in the past 3 days.

1:20am: The official Vermont delegate count is 9-6. In Rhode Island, it is still unclear as to whether Clinton got a 3 or a 5 delegate lead. In Ohio, meanwhile, Cincinnati is now fully in (finally) but Cleveland is two thirds reporting. Obama is not creating any kind of big margin there (52% to 47%), so it now appears pretty much certain that Clinton will get a double-digit lead in Ohio (with 92% reporting, she is up 55% to 43%), which I had pointed at hours ago as the mark above which Clinton could hope to create a new narrative in this race ( I confess I did not expect her to meet the double-digit mark)!

1:00am: Texas called for Clinton! Incredible night for the New York Senator who won 3 out of 4 contests, two of them decisively.
With 77% reporting, she is up 51% to 47% -- and Austin, Dallas and Houstin will not be enough to change that. The delegate count is still unclear, but as I said earlier there is some hope for Clinton in these Southern districts in which she is posting some huge margins. In Houston and Dallas, however, Obama appears to be just above the threshold for that extra delegate, which should help him in the final count.
The caucuses, meanwhile, are starting to report. With 5% in only, Obama is up 56% to 44%.

12:10am: This could end up being a huge factor in the Texas primary: Clinton is getting some huge margins in Southern Texas, above 70% (even 80%) in many of the counties there. And they aren't small counties either: In Hidalgo County, for example, gives Clinton 50,000 votes to Obama's 17,000... with only 49% reporting (which also implies Clinton still has some vote reserves here). Why is this especially important? This is an area with 4 delegate districts which the candidates were supposed to split 2-2. If Clinton is able to open up a 3-1 delegate district in a few districts here, it could allow her to get an unexpected delegate lead in places we were expected to see a tie.
She is leading 51% to 48% statewide right now, more than 60,000 votes ahead. But Houston is still barely reporting.

11:50pm: Obama speaks in San Antonio, Texas, accuses McCain and Clinton of "echoing each other." Expect him to pick up this theme in the coming days. In Ohio, Clinton is still up 56% to 42% and Cleveland has started reporting! And with 16% reporting... Clinton is actually very narrowly ahead. It looks like Clinton will maintain a double-digit lead in this state. In Texas, meanwhile, Clinton has passed the million votes mark and is 50,000 votes ahead.

11:30pm: In an enthusiastic victory speech from Columbus, Ohio, Clinton repeated "Yes we will," vowed to press on and give millions of Americans who still haven't voted a chance to vote: "We're going on, we're going strong, and we're going all the way!" She is showing no intention of withdrawing, and unless Obama really turns things around in Texas, it's going to be hard to force her out of the race until... April 22nd.
In Texas, we have crossed the 50% precincts reporting mark -- and Clinton is up 30,000 votes. Austin is now 64% reporting and Dallas is at 54%, so they are both above the statewide number, not a good sign for Obama. In Houston, however, we are still at 14% reporting, so there still are a lot of votes to come from there. Clinton can count on El Paso -- 12% reporting -- where she is leading by a huge margin for now (69-25). No word from the caucuses yet.

11:15pm: Austin and Dallas suddenly reported a big chunk of their votes (they are up to 30% and 35% reporting respectively), but Clinton managed to cling to her lead, albeit a very small one (17,000 votes). In Ohio, it's still 57% to 41% with almost two thirds of the vote reporting, and here again Obama's strongholds are finally reporting: Cincinnati is now 20% in.

11pm: Ohio is called for Clinton! The networks are not waiting for Cincinnati and Cleveland to start reporting to call Ohio. The margin is so big at this point that even big Obama margins in those cities could close the gap. For now she is only increasing her lead, back in front 58% to 40%.
In Texas, Clinton is pulling ahead quickly (though the big cities are still out); she is now up 20,000 votes.

10:50pm: Clinton has overtaken Obama in Texas (by 1,100 votes), with 24% of precincts in. She has been rising very steadily since the early votes put her down by about 100,000 votes about 90 minutes ago. There are two conflicting factors are at play here: (1) The Obama-favoring cities have yet to report, and while El Paso and San Antonio haven't either Obama's strongholds have more voters. (2) Clinton's huge late momentum (61% among those who decided in the past 3 days) means that we have to expect that she gets a lead among those who voted today, which makes this steady catch-up game entirely logical.
In Texas, still no news from the caucuses but as much as the system is defensible in places like Iowa which know how to run a caucus, it makes no sense in a state that never seen anything like this and is finding itself entirely overwhelmed.

10:30pm: The margin in Texas keeps getting smaller, a 9,000 votes margin for Obama now (49-49) with 19% reporting! It's difficult to know exactly which districts are reporting faster here given the early voting numbers, but it does look that Austin, Dallas and Houston are reporting much slower -- which is good news for Obama -- but neither have El Paso and San Antonio. In Ohio, things look good for Clinton right now (56% to 42%) as Obama's margin in a place like Columbus is a bit smaller than it should be, but it is obviously impossible to know what is going on until we get an idea of the kind of margin in Cincinnati and Cleveland.
A few clues about exit polls: 61% of those who chose in the last 3 days in Texas went for Hillary Clinton, providing her a crucial last minute boost. The big debate, of course, will be what cause this last-minute surge. Also crucial to Clinton's win: (1) Latinos backed her heavily, 63% to 35% and (2) independents and Republicans (about 30% of the vote combined) gave a very small edge to Obama, within single digits, a much smaller margin than in some of the previous contests.

10:10pm: CNN is describing scenes of chaos in the Texas caucuses, for example a voting location at which there were so many people in the room as to create a fire hazard, leaving to hundreds of people being left out of the room... with no one knowing what to do with them. You can be sure the Clinton campaign is taking notes of all of this to mount a strongest than ever charge against caucuses in coming days (if they are still alive, of course, which is still unclear at this late hour). In Austin, meanwhile, one of the caucuses has not started yet... as they aren't able to get things organized. That means people have been waiting there for 2 hours.
In Rhode Island, Clinton is keeping her huge 61% to 38% lead (this is bigger than in New York right now), though this is one place where even-delegate districts could hurt her bid to get more delegates -- the most likely scenario right now is a 3 delegate lead, though this will obviously evolve.
In Ohio, 35% reporting still put Clinton up 57% to 41% but 0% of Cleveland and Cincinnati's counties are reporting, though we are at 45% of Columbus's counties (with Obama up by 10% there). In Texas, with 15% reporting now we are at a 27,000 vote margin.

9:55pm: McCain gives his victory speech, with a banner reading "1191" in the audience. He devotes a significant part of his speech to Iraq, and on national security more general, a clear signal of where is going to take the race over the next few months. He goes on to discussing the economy to some length at well, mocking the Democratic refusal of trade and globalization. He ends with, "The contest begins tonight."
Meanwhile, there is interesting movement in the Democratic side: In Ohio, we are now at 21% counted and Clinton is hanging on to a 20% lead. (Remember that she led by a huge margin Missouri for most of Super Tuesday until St. Louis reported. We will know more about Ohio as soon as the cities start reporting). In Texas, Obama is now suddenly up by only 51% to 48%, a 35,000 vote margin. But the surprise comes from Rhode Island, where Clinton is leading 62% to 36% with 28% reporting (with 70% in the Provide County!) -- this could give her more delegate than most of us expected if she holds on to that margin.

9:30pm: Rhode Island called for Hillary Clinton, her first win since February 5th, and breaking a 12 win-streak for Obama! Now let's see whether she can get any delegate advantage out of this. In Texas, we are now at a 53-46 margin, with more than 1 million votes counted. In Ohio, we have reached the 11% precincts mark, with a 24% Clinton lead (as I explained 5 minute sago, however, the Obama-heavy parts of the state are not reporting at all, including Cincinnati and Cleveland, but still).

9:25pm: Speaking to his supporters in Irving, Texas, Mike Huckabee drops out and announces his support for McCain and need to unite the party.
Meanwhile, 8% of precincts are reporting in Rhode Island and show Clinton up 53% to 46%. In Ohio, we're at 6% reporting and Clinton is up 60% to 38% -- but the parts of the state that are reporting are where she is expected to do well. And keep in mind that Cleveland, one of Obama's strongholds, will report late since polls were opened much longer.

9pm: McCain becomes the GOP nominee. Polls have closed everywhere now, and Rhode Island and Texas are called for McCain which CNN calculates with grant fanfare gets the Republican above the 1,195 delegate mark.
Among Democrats, RI and TX are both too close to call. Exit polls in RI show a very close race, with Clinton about 3% up once again. The Texas exit polls have Clinton up extremely narrowly as well (though early polls are not included here, I believe). The exit polls also show that there is a big educational gap (with Clinton winning ahead among those without a college degree, Obama among those who do) but there is almost no income gap here, with all income groups (except those earning less than $15,000 and more than $200,000) splitting their vote. There is, however, a clear gender gap in TX, +13% among men for Obama and +17% among women for Clinton.

8:55pm: A new batch of early votes have been counted in Texas and the margin is now 85,000 votes between the two candidates so a slight improvement for Clinton which puts her down 54-45 with almost 1 million votes counted (which means we probably now have about a quarter of the votes reporting).
Another interesting point: Unless Obama crosses the 65% mark in Vermont (which he is far from with 37% of the votes in), the delegate split will be 9-6, so a 3 delegate advantage.

8:45pm: The Texas caucuses are underway, and they look to be as chaotic as we expected. The Clinton campaign is charging that the caucuses are marked by irregularities and that Obama supporters have taken control of some caucuses locations. In Ohio, the polls in Cleveland were kept open for by a judge because of high turnout.
Ohio is still not reporting its numbers, while Obama is up 46% to 36% with more than 800,000 votes counted (almost all are early votes, but this will represents about 20% of the final turnout, so Clinton has digged herself in a hole).

8:20pm: Ohio polls have been closed for 50 minutes now but almost no votes are being released. An interesting exit poll result is that Clinton got a much higher percentage of the white vote than usual -- 61%.
The Texas primary is shaping up, though, with Obama blowing out Clinton among early voters. 1% of precincts only is reporting, but early voters could amount to between 30% and 50% of voters and those numbers are coming in fast: Obama is up 58% to 41% right now, 430,000 votes versus 300,000 votes! This could a huge result going forward, as it is going to be very difficult for Clinton to overcome this huge a margin in early voting. (Wolf Blitzer does not appear to understand that the 700,000 votes already reporting are not coming from that 1% of precincts but are early votes, since he announced after reading those numbers that we should get some numbers results from early voting soon.)

8pm: Most polls have now closed in Texas as well (except in the El Paso region) and it looks like the networks are going to wait to release exit polls, but the first votes are already being reported (it's still very early). The results in Ohio are also coming in very slowly. We are already at 13% in Vermont, and the numbers are 58-40 for Obama. (Can someone explain why Wolf Blitzer insists that "we don't know where these votes are coming from" when CNN's results page has a very clear map of which counties are reporting what?)

7:35pm: Ohio polls have now closed. McCain is called the winner, and the Democratic race is too close to call. The CNN exit polls right now are showing Clinton up by a 3-4% margin -- way smaller than what she needs. If these numbers are confirmed, Clinton will not get the margin she needs in Ohio. The income and education gaps are back in full force here, with Clinton much stronger among lower-income and less-educated voters.

7:10pm: The polls have closed in Vermont, and the state was quickly called for Barack Obama and John McCain. Exit polls show a big Obama victory (probably above 60%), but VT only awards 15 delegates so the split will likely be somewhere around 10-5 (which could end up being meaningful).

6:45pm: More exit polls hint, still from the second wave of exit polls, i.e. these numbers could still shift dramatically, particularly the internals, so don't rely on them. They have Clinton winning Texas women 54%to 46% (which is good for her, but definitely underwhelming). But some very good news for Clinton supporters is her huge lead among late-deciders, 66% to 34% (unclear whether late-decider means decided today or over the past 3 days, but it confirms that Clinton has had some momentum over the past few days, as the last polls were suggesting). In Ohio, Clinton is up for now among union voters (56-43) as well as among those who say that NAFTA has taken jobs away (51-48). Meanwhile, MyDD posts some overall numbers that have Obama up 2% in OH and TX (essentially an exit-poll tie, and it is unclear what those numbers rely on).
I will post updated and refined numbers as they become available through the night.

6pm: As always, take all exit polls before polls close with a huge grain of salt. These hints come from second wave of exit polls, and they were wildly off in many states on February 5th. That said, the first leaked exit polls look to be showing close races in Ohio, Texas and Rhode Island, and a clear Obama lead in Vermont (I have not seen exact numbers, which is why I am going to leave it at "close races"). The AP is also reporting that about 1/5 Ohio voters are independents and 1/4 in Texas, two indications that are pretty much in line with what we expected (ARG modeled 23% and 27% respectively in its latest poll release).

Also, CNN is reporting that Texas Latinos are going for Clinton 64-35 and blacks going for Obama 84-16. This is roughly what we expected -- though if confirmed this means that Clinton has made no inroads among blacks and Obama among Hispanics -- which means that the determining factor will be the relative turnout of blacks and Hispanics. And the Washington Post offers us a hint of that: "In Texas, Hispanic turnout was up from 2004, when Latino voters made up about a quarter of all Democratic primary voters." That would obviously be a significant boost to Clinton's chances.

Original post: Yet another results thread -- perhaps the last one of the 2008 primary season, perhaps the prelude to many more to come. And it could be a long night, since we will have to wait all the way to the end of the Texas caucuses to get an idea of the shape of the race. The first polls close at 7pm in Vermont, followed at 7:30pm by Ohio. Most of Texas will close at 8pm ET, though the El Paso region will be voting until 9pm. (If past election nights are any indication, votes should start coming in at 8pm and not wait until 9pm). Rhode Island closes late, at 9pm as well.

Since there are no exit polls for now, read these guidelines of what to watch for tonight.

March 4th guidelines: Will this be the end of primary season?

The 2008 campaign has been going on much longer than most of us expected already, and this is only the second Election Day with a candidate in a position to bury the opposition. The only comparable occasion was New Hampshire nearly two months ago, when Hillary Clinton had her back against the wall but came storming back. Will she manage to do the same today, and keep herself in the game?

That's the fundamental question of the night. Unfortunately, answering it will be much more difficult than it was on January 8th, since it is very unclear what a win means for Clinton tonight -- as I explained earlier today. To put it simply, Clinton needs to win Ohio, Rhode Island and Texas; only letting go of Vermont would be acceptable for her. And then we will have to watch for the Texas caucuses, the results of which could differ from the results of the primary.

Right now, the two campaigns are throwing around accusations of voting irregularity, though it does seem that both sides have enough legitimate complaints that we can blame the election system -- especially in Ohio, where voting machines are once again proving they cannot be relied on. Another controversy today is Rush Limbaugh urging his (very conservative) listeners to cross-over to the Democratic side and vote for Hillary Clinton in order to prolong the Democratic primaries and get Barack Obama more bloodied up before he gets to turn attention to McCain (Kos had launched the inverse campaign in Michigan when he had urged people to vote for Romney).

There is anecdotal evidence of strong Republican cross-over, and that would usually mean a big advantage for Obama, who is often much weaker among registered Democrats. Limbaugh's call could make it a bit more difficult to assume that a higher proportion of GOP voters will help Obama as it usually does, though it is doubtful how much weight such calculations really carry. We will surely get an idea of this when looking at exit polls and trying to determine the breakdown of Republican voters who migrated to the Democratic primary.

Another interesting indication to watch for in exit polls will be the breakdown of blue-collar voters who are usually Clinton's base but who deserted her in Wisconsin. Obama has been hurt by the the NAFTA and Canada controversy of the past few days and that might be helping Clinton recover some of her strength.

There are other races to watch for today:

Republican primaries: In the week following Super Tuesday, John McCain looked like he was in for a series of tight races against Mike Huckabee; essentially assured of winning the nomination he lost a string of states the first week-end and then got into a tight race in Virginia. Since then, the GOP primaries have fallen off the media's radar, and McCain is posting huge leads in Ohio and Texas poll. Huckabee had some hope for a while of embarrassing McCain again in Texas, so it is worth keeping an eye on the results there, but it does not look like McCain will have to break a sweat.

Congressional primaries: All eyes will be on OH-10 tonight, where Dennis Kucinich is facing a strong primary challenge by Cleveland Councilman Joe Cimperman, who was endorsed by the Cleveland Mayor. The main complaint against Kucinich is that he is abandoning his district during his presidential runs.

Other congressional primaries to watch for include OH-02 and TX-22, with contested Democratic and Republican primaries to take on vulnerable incumbents (especially in TX-22).

When a win isn't necessarily a win

Today's results could be the most difficult to handicap, and this in a context that has seen 11 Election Days already. Indeed, we are in the very difficult position of asking what it would mean for either candidate to get a "win" out of tonight.

There are two separate questions at play here: (1) What will it take for Clinton to fundamentally alter the state of the race and get herself back in the running? (2) What will it take for Clinton to survive today's votes, i.e. have a rationale to go on with her campaign tomorrow morning? The second question would seem to be dependent on the first, but the Clinton campaign has been giving enough clues as of late that there is a pretty significant gap between the two.

The answer to the first question is rather simple: To have a chance at being the nominee, Clinton needs to triumph in an Ohio by double-digits and win Texas comfortably enough to get a delegate lead (and probably win Rhode Island as well). A tall order for the New York Senator -- albeit one that the last batch of polls suggest she could pull off.

But if Clinton wins, say, Ohio by 6% and Texas by 4%, these victories will be insufficient to boost her in the delegate chase and would make Obama even more favored to win the nomination -- but they would be good enough that Clinton would probably not drop out.

This would truly be a paradoxical situation: The very same results would both confirm that Clinton has very little chance of becoming the nominee and change the momentum in her favor. This would be an almost unprecedented situation in the history of modern primaries, as a candidate who is increasingly unlikely to get the nomination would press ahead with new-found determination.

This possibility would create a huge headache for Obama: The next opportunity to get Clinton out of the race won't come until April 22nd, and that would mean that Obama would have to fight the primary fight -- and face a lot of new scrutiny since the media appears to have decided to turn the spotlight on him -- in a race that he would have essentially won.

Of course, we aren't there yet. While the in-between scenario (Clinton survives but does not change the fundamentals of the race) is the most likely, I already pointed out that Clinton could potentially make herself relevant if she gets a huge night tonight. And then there is the very plausible scenario that Obama gets to eliminate Clinton tonight: He has a huge opportunity.

What would constitute a knock-out victory by Obama? If he wins Ohio or the popular vote in Texas, it would be very difficult for Clinton to stay in the race and even if she does Obama would have beaten her at her own games, in the states Clinton herself had pointed to as her firewall.

Can Obama hope that Clinton drops out if the decision is more muddled (small wins by the NY Senator in OH and TX)? The campaign knows that its now or... in 6 weeks, and they are pressuring Clinton very heavily to drop out in such a scenario. It looks like they have prepared a major offensive in the coming days: Tom Brokaw is now reporting that Obama has 50 superdelegates ready to endorse him and that have yet to come forward, and many observer are now pointing out that the Obama campaign has still not released its February fundraising figure that was supposed to be so overwhelming. TNR's Mike Crowley is suggesting that they will release all this information in the coming days to force Clinton out of the race.

And this is where things get really complicated: It's going to be very difficult to figure out just how muddied the results are because of... the Texas delegate allocation rules which make it a huge possibility that Obama would get more delegate out of the state than Clinton even he loses by less than 5-6%. So if Clinton gets a 5% popular vote win but Obama gets the delegate lead, both campaigns will claim to have won Texas... Clinton would have a more logical claim to victory (as she did in Nevada), but considering that her challenge today is to cut Obama's delegate lead it will be difficult to cover that as anything than a loss.

And I am not even accounting here for the Texas caucuses that will start immediately after primary polls close. 67 delegate are at stake here... That's almost as much as the entire state of Minnesota, where Obama had triumphed in the caucuses and gotten a 24 delegate lead. Even if Clinton wins the primary vote by a few points, Obama could very well crush her in the caucuses and get a delegate lead big enough to undermine whatever Clinton gets in the Texas primary and in Ohio. Sounds unlikely? There was a 30% gap in Washington between the primary results and the caucus results... If even a third of that happens in Texas, it could become Exhibit A of the Clinton campaign's charge against caucuses, and it's stuff like that they need to provide cover to superdelegates.

Complicated? Certainly. What is most terrifying that the most confusing scenarios outlined here (Clinton gets a narrow TX primary win, loses caucuses, loses delegates) are not the most far-fetched at all, but are more likely than simpler routs

Can Clinton pull off a comeback?

Hillary Clinton knows she needs to make something big happen tomorrow in Ohio and Texas if she wants to stay alive. We are only a few hours from polls opening, and the New York Senator is getting as favorable news as she could have hoped for. She has been left for dead for most of the past week, in striking similarity to the run-up to New Hampshire, but the latest polls open the possibility of that analogy playing out all the way and it is very possible that Clinton could come back from the dead once again tomorrow.

A significant comeback is still very unlikely but it now appears possible, unlike a few days ago. And by comeback I do not mean narrow wins in Ohio and Texas, but also a potentially strong victory in Ohio, again something that looked out of the question a few days ago

• ARG just released its last polls. In Texas, Clinton has moved ahead 50% to 47%, a 10% swing in 4 days and the first time Clinton is up in an ARG Texas poll since Super Tuesday.
• But the real story is in Ohio, where ARG's poll (in the field Sunday and Monday) shows Clinton up 56% to 42% -- including a 60% to 38% lead among registered Democrats.


• Insider Advantage, meanwhile, apologetically shows Clinton leading in Texas, 49% to 44%. Matt Towery, IA's CEO, appears afraid that "this is a lonely and difficult spot for a pollster. However, we have polled thousands of voters in Texas and have yet to show Sen. Obama leading." Such disclaimers are somewhat rare.
• He adds that this result is obtained in positing the most positive turnout model for the Illinois Senator: "Our weighting is designed to discount Hispanic turnout in areas where historically their turnout has been lighter, and to maximize potential African-American turnout. Even with this model, which we believe to be accurate, though obviously not a perfect situation for Clinton, she leads."
  

• Update: Yet another data point demonstrating Clinton's progress comes from Zogby's latest tracking poll. In Texas, Clinton surges by 6% in one day, transforming a 3% deficit into a 3% lead -- her first lead in Texas since Zogby started tracking.
• In Ohio, the race is now tied at 44% while Obama was up 47% to 45% yesterday.
  

Combine this with the polls I reviewed this afternoon, with PPP and Univ. of Cincinnati both showing a 9% Clinton lead in Ohio and Suffolk a 12% lead, and other institutes confirming her Texas climb. And then there is Rasmussen's daily tracking poll in which Clinton has regained the advantage nationally for the first time in 3 weeks (albeit only 46% to 45%).

We won't know until tomorrow if Clinton can ride this movement, we can start offering some explanation as to what moved the numbers over the week-end. Mark Penn has been saying that Clinton will recover by Tuesday because of the red phone ad that, he claims, finally raised questions in voters mind about Obama's experience on national security. Clinton might even have made the point a bit too strongly today when she said: "I think that I have a lifetime of experience that I will bring to the White House. I know Senator McCain has a lifetime of experience to the White House. And Senator Obama has a speech he gave in 2002."

A new ad by Clinton (you can watch it here) repeats a charge Clinton made at last Tuesday's debate -- that Obama was too busy running for president to conduct a single committee hearing on Afghanistan. And it is an effective spot because it uses Obama's own words at the debate. As soon as Obama answered, I thought that his response was weak (he defended himself saying that he became president of the Committee at the beginning of 2007 when he was starting his presidential run) sounded particularly weak, since Clinton's criticism is precisely that he started running too early.

But Obama will certainly not be undone, and he will not let himself go down because of ads. He has the money to respond to whatever Clinton tries against him, and the rapidity of his counter-offensive in that red phone affair displayed that. There must be more at play, and that something could simply be that there has been a much tougher spotlight than usual on Obama as of late, starting with the NAFTA and Canada story is hurting Obama -- and the fact that Clinton's momentum is particularly visible in Ohio supports this explanation.

TNR's Mike Crowley reviews media coverage today and concludes that "this might be the best news cycle she's had in a month." First Read has more details about the rough questions the media subjected Obama to today, in a much more confrontational tone than usual. Overall, it looks to have given Hillary an opening to strike back and keep herself going through March and April. We will know tomorrow evening whether Clinton managed to take advantage of this opening or whether this momentum back towards her is only momentary movement fueled by negative stories that will have faded from voters' consideration by tomorrow.

In last day of campaigning, Clinton appears to have stopped the bleeding

Polls in Ohio and Texas will open in less than 24 hours in Ohio and Texas (and Rhode Island and Vermont, not that anyone really cares at this point) and after showing a steady movement away from Hillary Clinton over the past two weeks, the latest surveys suggest that Clinton might have stopped the bleeding and perhaps even regained some of the lost ground.

Naturally, not all polls agree -- and whatever bump some of them are registering is very slight, and in no way guarantees that Clinton will have a good day tomorrow. The point here isn't that Clinton is strengthening her position but that her downward spiral in polls might have stopped, at least in Ohio.

• First, Rasmussen: In Ohio, it shows Clinton up 50% to 44%, an uptick from the previous poll which had her up by 2% only.
• Similarly in Texas, Obama now has a 1% edge (48% to 47%) but he was up 4% in the previous poll -- and he was gaining ground in every poll.


• Public Policy Polling shows similar improvement for Clinton: In Texas, Clinton is now leading 50% to 44% -- up from a 48% tie in the previous poll. She does so by remaining very strong among Hispanics (67%) and regaining strength among white voters (58%).
• In Ohio, PPP shows Clinton up 51% to 42% -- a marked improvement from her previous 50-46 lead.


• The University of Cincinnati survey shows the same margin in Ohio, with Clinton up 51% to 42%.


• Next comes SUSA's Ohio poll, which has Clinton leading 54% to 44%, fueled by a 27% lead among women. Clinton is ahead among those who have already voted by 13%. The trendline of the SUSA poll is very interesting: Clinton had a 17% lead, then a 9% lead, then a 6% lead -- and has now reverse this decline.
• In Texas, SUSA shows Obama up 1 point 49% to 48% -- but just as in Rasmussen's, it is Clinton that has slightly improved, up from a 4% deficit last week. SUSA estimates that half of the electorate has voted early, and puts Clinton up 2% among early voters. She is leading the Hispanic vote with 64%.
  


• Quinnipiac shows Clinton up 49% to 45% in Ohio, which is a sharp decline from her 11% lead last week (she was leading by 21% two weeks ago). But the poll was in the field starting on Wednesday, so it is unlikely to register any Clinton stabilization over the past few days. She is enjoying a 32% point gender gap, and is up 32% among white women.


• Worth noting that the ARG polls that I reported yesterday also had good movement for Clinton, who progressed by 2% in Ohio and 7% in Texas.


• That leaves us with Zogby, which shows Obama strengthening his positions! He is ahead 47% to 44% in Texas, and Zogby tells us to look at East Texas tomorrow.
• In Ohio, Zogby shows Obama taking the lead for the first time, 47% to 45%. I believe this might be the first survey by any polling group to show Obama ahead.

If Clinton confirms tomorrow that she has stabilized herself and manages to win comfortably in Ohio and get a victory in Texas, Obama will have lost a huge opportunity to put an end to the Democratic contest. He would still remain the overwhelming favorite to win the nomination given how difficult the math of a Clinton comeback looks, but the next context in which Obama would have an opportunity to get Clinton out of the race is... Pennsylvania, on April 22nd. And let's not forget that he might be in for expensive and grueling races in Florida and Michigan if the DNC agrees to do-overs.

One last poll on this Monday morning, as Rasmussen released a survey for New Jersey's general election:

• Obama loses against McCain, 45% to 43%, while Hillary crushes him 50% to 39%.

This gets us back to the old pattern of Clinton doing significantly better than her rival in the Northeast, with McCain waging strong contests in these must-win states for Democrats. She also used to poll much better in Southern states but Obama put an end to that pattern in the past few weeks.

Update: Here's one more poll bringing good news to Clinton, as she is up 52% to 40% in the Suffolk poll from Ohio.

Sunday polls: Ohio and Texas are too close too call

A new series of poll confirms that both Ohio and Texas are too close to call at this point. That is in and of itself a huge victory for Barack Obama, as it seems guaranteed that, even if Clinton wins both states, she will not do so by significant margins and fail to cut into Obama's delegate lead. And given how dire her delegate situation is, that could mark the end of her campaign. In other words, Ohio and Texas are both too close to call when we ask who will win them, but the Illinois Senator is at least 8-10% above where he has to be.

Zogby, MSNBC/McClatchy and ARG all released two polls, with sensibly similar results

• In Ohio, MSNBC shows Clinton ahead 47% to 43%. The pollster adds that the NAFTA issue does not seem to be helping either candidate.
• Zogby's tracking poll shows the slightest of upticks for the New York Senator, who edges out Obama 47% to 46% after the two were tied at 45% yesterday.
• Finally, ARG also shows a slight improvement for Clinton, who is up 51% to 44% (she was up 5% two days ago).


• In Texas, MSNBC shows Obama edging out Clinton 46% to 45%. Hispanics are voting for Clinton 2:1 and black voters are going for Obama 86% to 6%. It looks like the proportion of each group in the final electorate will be decisive in deciding who comes out with the win.
• ARG shows very good movement for Clinton in this state, as she closes a 7% gap to tie Obama at 47%. They have been a couple of polls in recent days showing slight Clinton movement in Texas.
• Zogby was one of them yesterday, but now rolls back Clinton's progress, as she is back down 4%, 47% to 43%.

Considering that the Texas primary will be followed by caucuses -- which appear to have never been called at all -- we might be in for a long night in Texas.

Update: Well, just when polls were looking like they were in agreement, the Columbus Dispatch has released a poll with Clinton up 56% to 42%, the kind of margin she had 10 days ago. Now polls are not disagreeing over who will win Tuesday states but also about whether she can win this big! However, there are two huge caveats: (1) The Columbus-Dispatch poll is (in)famously a mail-in poll, in which readers have to return slips, a methodology often criticized. (2) The responses came in from February 21 to February 29th... a very large time window at a tie where Obama was making major inroads and cutting into Clinton's lead.

Assessing Clinton's chances, with delegate calculations and the latest polls

Zogby is back to releasing tracking polls -- something they had not done since the run-up to Super Tuesday. And they bring mixed news for Hillary Clinton who now leads in neither Ohio nor Texas:

• In Ohio, Clinton and Obama are now tied at 45%. Among the GOP, despite an 8% in Huckabee's favor, McCain is still ahead by... 33%!


• In Texas, Clinton has cut Obama's lead by 4% and now "only" trails by 2%, 45% to 43%. Zogby adds that Clinton led Obama by double-digits in the Friday numbers alone, and we will know in the coming days if that is an outlier or the beginning of a trend.


• The BELO tracking poll out this morning confirms that there might be the slightest of trends in Clinton' s favor who takes back the lead, albeit by a small 46% to 45%.


• Finally, we got a rare poll from Rhode Island showing Clinton up 49% to 40%, thanks to a massive 32% lead among women -- for a 50% gender gap!

I have been reminding you for a few days now that the important metric has never been whether Clinton can win Ohio and Texas but whether she can win them big and get the kind of delegate lead she need to cut into Obama's advantage. When the question becomes whether she is going to win Ohio and Texas at all, it becomes hard to imagine how she will storm back to significant victories.

To illustrate this point and how crucial big wins are for Clinton, let's turn to some delegate calculations. Forbes has a very useful delegate calculator online that allows you to project the delegate attributions of the upcoming states by predicting the percentages of both candidates. Marc Ambinder conducted the same exercise yesterday and discovered just how dire Clinton's situation is, definitely take a look at his calculations as well, but here is my take on this:

The Campaign Diaries pledged delegate count right now stands at 1195.5 versus 1048, approximately a 150 delegate gap. Now, let's leave out all the upcoming states that Obama is very strongly favored to win: Vermont, Wyoming, South Dakota, Oregon, Montana and Mississippi. I am not even including North Carolina, Kentucky (which holds caucuses) and Indiana (where a recent poll had Obama way ahead) for the sake of making an argument (I grant that there is an argument to be made for Oregon staying competitive, since Obama only won the Washington primaries by 3%).

That leaves us with Ohio, Texas, Rhode Island, Pennsylvania, West Virginia, Puerto Rico, North Carolina, Indiana and Kentucky. It would take Clinton winning every single one of these contests with 58% to get a 150 pledged delegate lead out of them and barely catch up with Obama. If we take don't count North Carolina and Kentucky, that percentage rises to 62%.

And then we would still have to throw in Obama's victories in places like Mississippi and Vermont in the calculation and the fact that Texas also holds caucuses. Clinton's one hope is Puerto Rico, where there is a possibility of delegates being attributed winner take all which would give her a huge 55 delegate lead out of that contest, which could definitely help her make up some of the difference.

Now if Clinton managed to gather enough momentum to start winning these states with 55%, she would not close the delegate gap entirely but she would get very close and she would suddenly have momentum and all the buzz, which could push some superdelegates in her corner. Also, if Clinton managed to get fairly close she would then make a renewed argument for sitting Florida and Michigan. And while she would not necessarily succeed (after all, if she doesn't have the votes to become the nominee it is unlikely she can control the committee that will decide on sitting FL and MI), it could provide cover to superdelegates who want to vote for Hillary.

But these numbers point to how difficult it will be for her to either get a pledged delegate back or even get that close to Obama, whose February sweep really did wonders.

Obama airs response to Clinton's red phone ad

Talk about rapid response. It's only been a few hours since Clinton's red phone ad was first revealed, and the Obama campaign is already up on air with his answer. The ad (which you can view here) looks very similar to Clinton's: It starts with children sleeping and a phone ringing, with a hurried off voice informing us that "It's 3am and a phone is ringing in the White House." But it then dramatically changes gears.

While Clinton's ad went on to say, "Your vote will decide who answers that call. Whether it's someone who already knows the world's leaders, knows the military, someone tested and ready to lead in a dangerous world," Obama's invokes the war: "When that call gets answered, shouldn't the president be the one -- the only one -- who had judgment and courage to oppose the Iraq War from the start? Who understood that the real threat to America was Al-Quaida in Afghanistan."

On the one hand, the response Obama chose is predictable. At the Ohio debate, Clinton accused Obama of always and systematically referring back to 2002 and Iraq when challenged on national security inexperience, and emphasized her point by explaining how Obama then refused to live on to his anti-war days. Obama's automatic referral to the early call on Iraq in this ad is an illustration of Clinton's point.

On the other hand, Obama's ad admirably distorts Clinton's. Considering that the New York Senator did not just accuse Obama of being inexperienced but did so by invoking imagery that is now increasingly associated with Republican tactics (despite its association with Johnson and Mondale), Obama's reminder of their respective position in 2002 looks like more of a diss than usual.

Then there is the truly remarkable speed of delivery (though the ad's production does look a bit hurried). After all, one of the chief criticisms Clinton's campaign voices against Obama is that he is not prepared to fight and respond to attacks, making him too easy a target for Republicans in the general. By coming around within a few hours, Obama's team is hoping to show it is ready to fight back and that its war room is as good as any.

It would be way overstating the case by a lot to use this ad to dismiss Clinton's contention that Obama is not prepared to fight the GOP machine. The inexperienced charge has been leveled at Obama for months now -- especially on foreign policy -- and Obama has always invoked his opposition to the Iraq War. The question has always been whether he would immediately fight back against a new GOP smear or whether he would fold and wait (like Kerry did in August of 2004). If anything, the campaign's slowness to respond to the controversy over Michelle Obama's comments a few weeks ago is more revealing of how prepared they are. But a few days from March 4th, this rapid turn-around will certainly allow them to boast.

Meanwhile, two new poll were released from Ohio and Texas by Fox News:

• In Ohio, Hillary Clinton gets her biggest lead of the day, 46% to 38% on the strength of white women and blue-collar voters. Worth noting that the poll gives Clinton a slight lead among independents, which does not accord with most measures we have seen.

• In Texas, Obama is slightly ahead, 48% to 45%. He has tied his rival among white voters, but Clinton maintains a very clear edge among Latinos -- enough to keep her very competitive.
  

Friday polls: Obama where he wants to be in Ohio and Texas, McCain looking strong in Florida

Whether or not the ad war is any reason for this, Obama is continually improving his position in March 4th polls. Keep in mind, Clinton will have to prove these polls wrong by winning both states and do so by big margins. First, two polls from Ohio:

• A new Zogby poll shows a toss-up, in one of the closest polls from the state we have seen yet: Clinton is only ahead 44% to 42%.

• ARG confirms that the gap is rapidly closing, showing a 50% to 45% margin in a poll taken over the past two days. Five days ago, ARG showed a 49% to 39% gap.

• Update: Rasmussen confirms that Clinton has lost her Ohio lead, showing her ahead 47% to 45%. That's down from a 5% lead on Sunday night and a 9% lead last week.
  

If Clinton has to even worry about Ohio at this point, that leaves her even less time to focus on Texas, where her position looks even more dire:

• Zogby shows Obama relatively comfortably ahead, 48% to 42%.

• ARG has Obama up 51% to 44%, which is actually a one point improvement for Hillary (which is saying a lot). Clinton's lead among women (+9) and Hispanic (55-41) is much smaller than it should be, going a long way towards explaining her Texan troubles.
  


• Meanwhile, the BELO Texas tracking poll shows Obama taking the lead for the first time, albeit only 46% to 45%.

In all of these polls, John McCain hardly looks to be in any danger against Mike Huckabee, which is a semi-surprise given that Huckabee was expected to maintain some semblance of competitiveness in Texas. But with the media having given up on the GOP race, it is not surprising that Huckabee is finding no space to continue making his case.

McCain got even better news with the release of a Mason-Dixon general election poll from Florida:

• McCain leads Obama 47% to 37% and he leads Clinton 49% to 40%.
• McCain posts strong numbers among Republicans, while both Democrats (particularly Obama) are much weaker among independents.

If these numbers are confirmed (and Florida hasn't been very kind to Dems over the past few months), it could spell trouble for the Democratic nominee who can't afford to give up on the Sunshine State's 27 electoral votes. (I am not suggesting that Democrats need to WIN Florida to get the White House, but that they need to force McCain to defend those 27 electoral votes, otherwise he would have too much time to spend not only on places he really can't afford to lose, like Ohio, but also testing the Democratic nominee in blue-leaning states). One possible reason for McCain's strength could be that he is likely to be stronger than other GOPers would have been among the Cuban community.

Obama blankets OH and TX, Clinton responds with controversial ad

Continuing his spending dominance, Barack Obama is taking additional steps to blanket Ohio and Texas with advertisements. His campaign has bought time for a two-minute ad in ever media market of Ohio and Texas (and there are many, especially in Texas) the day before the election. This is a tactic that both major campaigns used in previous elections, particularly in Iowa, but it had not been tried yet in such big states. What is even more noteworthy is that the Clinton is for now showing no intention following suit, yet another indication of how massively she is being outspent in these two must-win states. Driving around in Texas, Politico's Jonathan Martin is also amazed by the extent of Obama's advertisement.

And, as I pointed out yesterday, this spending gap is not necessarily explained by the fundraising gap. Clinton's February $35 million -- by far her record since the start of the campaign -- should be more than enough to answer Obama. If anything, Clinton's campaign is probably thinking far less about saving money all the way to the convention right now than Obama's is.

Facing with this ad onslaught, it is looking like an increasingly impossible task for Clinton to get what she needs on Tuesday -- big enough victories to cut into Obama's delegate lead in any meaningful way. And the New York Senator is trying something new today: A controversial new ad that makes the preparedness for national security more explicitly than ever before in this race. With pictures of children sleeping and the sound of a phone ringing (watch the ad here), an announcer reads the following script:

It's 3am and your children are safe and asleep. But there's a phone in the White House and it's ringing.

Something's happening in the world. Your vote will decide who answers that call.

Whether it's someone who already knows the world's leaders, knows the military, someone tested and ready to lead in a dangerous world.

It's 3am and your children are safe and asleep. Who do you want answering the phone?

The ad naturally ends with the customary "I'm Hillary Clinton and I approve this message" (which should dispel any possibility that people think that this is, say, an Obama ad). The ad is reminiscent so many GOP ads over the past few cycles, emphasizing the danger of this world and the need for a strong lead. What came to my mind first was Bush's wolves ad from 2004 -- the suggestion of danger being almost stronger than showing danger itself. But, via Ben Smith, this ad truly comes from, an 1984 spot called "Red Phone" ran by Walter Mondale against Gary Hart, questioning the latter's preparation to deal with the "issues of our time" (watch this ad here).

What is most striking about Clinton's ad is how weak a contrast it draws -- it is truly the Democratic version of such ads, in that it remains very soft, so soft that I can't help but wonder whether it can really have any effect. The wolves ad, for example, was a direct attack on Kerry; Clinton's ad only draws implicit contrasts, and it is unclear whether the first-time viewer even notices them before being told who the ad is being ran by.

If Obama is the Democratic nominee, expect much much more brutal versions of this argument, since McCain has already made Obama's national security inexperience a central feature of his campaign. If this issue continues to dominate over the next few months, will it force Obama's hand when it comes to a VP pick, and will have have to choose someone based first and foremost on foreign affairs/national security credentials (Webb and Richardson?).

Obama outraises Clinton in February, continues posting strong poll numbers

News came yesterday that Obama was massively outspending Clinton in the March 4th state, which accounts for why it is now basically impossible for Clinton to get what she needs next Tuesday. Massive outspending does not necessarily win an election, but it certainly is enough to prevent a blow-out defeat, i.e. it prevents Clinton from achieving what she needs.

Now, the campaigns are releasing their February fundraising information: Clinton is announcing that she has raised around $35 million. Obama has not released definite numbers yet, but his campaign is indicating they have raised "considerably more" than $35 million. Estimations over the past few days have put Obama's fundraising somewhere around $60 million, which means about $2 million raised every day. Both candidates are raising much more money than they did in January (when Clinton had come out with $13 million).

But this does not explain why Clinton is being outspent so massively. She might have raised much less than her rival, but $35 million is not a low number by any means (it means more than $1 million a day, pretty much where Obama was in January). It is more than enough to get Clinton out of financial trouble, prove that she is financially viable -- and to allow her to blanket Ohio and Texas with advertisements and match whatever Obama spends. After all, there is so much that the Illinois Senator can spend without saturating the airwaves. So where are those $35 million going?

The situation is dire for Clinton in most polls that are now being released:

• The Texas Rasmussen poll out today shows Obama leading 48% to 44% -- the first time he has taken the lead in this state in Rasmussen.


• A People Calling People poll has Obama leading 40% to 33% -- a very high number of undecideds -- though Clinton is narrowly up among early voters (a statistic that is confirmed by the SUSA poll, and that could prove an interesting development given that it helped Clinton in California).


• Finally, the Belo poll has Clinton maintaining the tiniest of leads, 46% to 45% -- but the trendlines are in Obama's favor, as Clinton was leading by 3% yesterday.


• The Pennsylvania Rasmussen poll points an equally dramatic number, as Clinton is leading by only 4% -- 46% to 42%. PA doesn't vote until April 22nd (and if the OH and TX numbers hold for the next 6 days, Pennsylvanians won't really get to weigh in), but her drop there exemplifies the drop she has experienced nationally -- and speaks of her difficulties in neighboring Ohio.

As I explained yesterday, the big March 4th question was not whether Clinton would win and hold serve. Obama took a huge delegate lead in the February elections, forcing her to score big in Ohio and Texas. Now, we are looking at Texas polls showing her a little bit down or a little bit up, as if the interesting question is whether she will win the Lone Star State. All these polls show Clinton way under the true threshold she has to cross. When the question becomes whether she is going to win Ohio and Texas at all, it becomes hard to imagine how she will storm back to significant victories.

Update: Hope for Clinton? A new Insider Advantage poll shows a 5 point swing in her favor, as she now leads 47% to 43% in a poll conducted yesterday night, versus a 1% deficit in the Monday night poll. The pollster does insist that hte internals are pointing out towards a shift in momentum, and to the Latino vote solidifying Clinton's way. This is not confirmed by other polls, but yet again this is the most recently conducted survey.

But as I said above, this gives Clinton supporters hope that their candidate might win the state, not that she might win it big. And that is the key metric out of March 4th.

Massively outspent, Hillary Clinton increasingly unlikely to score the big victories she needs next week

I apologize for the very slow posting today. Either my blogging or my hosting service appears to have major technical issues and refused to upload this post.

I wrote last week that the Democratic race could still drag on for weeks to come: If Hillary Clinton wins both Ohio and Texas, she would still be the heavy underdog but she would have survived to fight another day. Since then, Barack Obama has taken the lead in a few Texas polls and cut her Ohio edge to mid single-digits. Now, for Clinton to comeback on March 4th will allow her not to post blow out victories but to simply win both states -- by any margin at all.

Unfortunately for the Clinton campaign, this is no longer early January and the expectation game is no longer relevant to analyzing the campaign. Hillary's 2 point victory in New Hampshire was a giant triumph that entirely changed the state of the race. 2 point victories in Ohio and Texas would be very different: Clinton does not need to win, she needs to catch up Obama's massive pledged delegate lead. She can no longer battle herself back to a draw as she did on January 8th, she needs to change the fundamentals of the race.

And it is looking increasingly unlikely she will be able to achieve that next week. The Clinton campaign is now holding on to polls that show them still in the lead in Ohio and some even in Texas -- but the issue was never whether Clinton would be able to win Ohio and Texas but whether she would be able to win them big. And when the question becomes whether she is going to win them at all, it becomes hard to imagine how she will storm back to significant victories.

Complicating matters is the fact that Clinton is being heavily outspent in those decisive states, a dynamic that started playing out in the run-up to February 5th. Clinton has spent $14 million to Obama's $23 million over the past 30 days. And First Read now suggests that she is being outspent nearly 4:1 in Ohio once we take into account the independent expenditures that are playing on Obama's behalf (for instance SEIU). Who would have thought that Clinton could lose the election because of fundraising problems?

It is in this context not surprising that Clinton is unable to sustain a double-digit lead. Such massive outspending does not necessarily win an election, but it certainly is enough to prevent a blow-out defeat, i.e. it prevents Clinton from achieving what she needs. And the latest poll from Pennsylvania out today by Quinnipiac suggests that the contagion is reaching later states: Clinton is ahead only 49% to 43% down from a 16% lead two weeks ago (52% to 36%). That's an important drop that is reflected in the Texas and Ohio numbers.

Clinton has done plenty to try and change the dynamics of the race over the past two weeks, most notably her switch in tone over the past week-end. The debates were her best chance to reverse momentum, but now that the two debates have passed we can see how daunting that task was: Those were the 19th and 20th debates of the cycle, which means that (1) most attacks Clinton could launch had already been aired and that (2) Obama knew exactly what was coming and was prepared, simply because he had already heard it all. Given how difficult it is to knock an opponent out at a debate in a first confrontation, it is nearly an impossible task after a year of discussion.

In fact, the only debate development that truly changed the race were due to multi-candidate dynamics: In Philadelphia, Clinton stumbled because the entire field was against her and harassed her for the driver licenses answer (starting with Chris Dodd). In New Hampshire, it was the tag-teaming of Edwards and Obama that led to women rallying around Hillary. And in South Carolina, Obama benefited because he was the one in the middle, with Edwards and Clinton firing shots at him. In these two candidate debates, there were no such gang-ups (nor backlashes against them).

Now with 7 days to go until March 4th, it is unclear how many more occasions Clinton will have to regain her footing. And do so decisively.

Obama's train starts leaving the station, as Clinton's support erodes

Chris Dodd became the first former Democratic presidential candidate to endorse today, as he rallied behind his party's frontrunner. The reasons he gave were not necessarily that he wants Obama to be president more than Hillary, but rather that the train is leaving the station and slowing it down will only hurt Democrats in the fall: "It is now the hour to come together. I believe the hour has come now for us to make that choice – to stand up and say we’re going to get behind this candidacy... I don't want a campaign that is only divisive here, and there’s a danger of it becoming that."

Chris Dodd's endorsing Obama will have no significance in Ohio and Texas and the only state it could have mattered at all (CT) voted weeks ago. But the language he uses and the timing of his announcement exemplify why Clinton is so weak right now. Not only is she trailing in the delegate count, but she has to fight Obama's growing inevitability argument to prevent too many people from rallying behind him thinking that the primary is over. This is the position that she was in back in the fall, when she was picking up endorsement after endorsement. And now the roles have reversed, making it very difficult for Hillary to hope for a comeback.

To make matters worse, polls are continuing to show an erosion of her support in March 4th states, both in Ohio and in Texas. Yesterday, Obama took his first leads in Texas while Clinton kept a high single-digit advantage in Ohio. Today, three new polls underscore how close Clinton is to being forced out of the race:

• In Ohio, Rasmussen has Clinton up 48% to 43%. A few days ago, Clinton's lead was 8%, and that was already a drop in support. It seems that Obama's campaigning on NAFTA is working, as an overwhelming majority think that he is opposed to NAFTA while the verdict is split about Clinton's position.


• In Texas, meanwhile, it seems safe to say that Clinton is no longer the favorite. In the latest PPP survey, the two candidates are tied at 48%, though Hillary leads 52% to 44% among registered Democrats and is not weakening at all among Hispanics (68%). But Obama is getting strong results among Republicans and independents who, PPP notes, "plan to vote in the Democratic primary because of John McCain's status as the presumptive nominee."


• Finally, SUSA gives the Texas lead to Obama, 49% to 45%, a 9 point swing in his favor in a week. Obama is picking up grounds among most groups, and holds Hillary to 52% among Hispanics.

• Update: Add one more poll showing Clinton's support eroding. SUSA came out with its latest Ohio poll, and shows Clinton up 50% to 44%, holding firm thanks to a 22% lead among women (and a 37% gender gap). Two weeks ago, Clinton led by 17%. Last week, she led by 9%.

As I often note, don't forget that Obama will likely get more delegates out of Texas than his percentage would suggest, so even a 1% lead would give him a significant delegate lead -- and that's not even accounting for the caucuses. Clinton has to change the dynamics of this campaign in the final week in a dramatic way, and things like Chris Dodd's endorsement of the Illinois Senator underscore the challenge she faces that she first needs to explain how she can still win the nomination and try to slow down the Obama coronation.

Tonight's debate (the last one?) is one of her last chances to change the direction of the campaign.

Clinton loses her Texas lead, but what exactly would constitute "defeat"?

Polls released this morning suggested Clinton was able to stay afloat in March 4th states, but I did warn that that batch did not contain any numbers from Texas, which is shaping to be the most difficult state for Clinton. Now, new polls released this afternoon suggest a very troubled road away for the New York Senator in the Lone Star State, and don't forget that a loss here could spell the end of her career.

To make matters worse for Hillary Clinton, the unions that have newly endorsed Obama seem to be determined to boost the Illinois Senator. UFCW has a new ad running on his behalf in Ohio, and Marc Ambinder reports that SEIU looks ready to spend more than $700,000 on helping him win Ohio and Texas. Given that one of Obama's main weaknesses has been blue-collar voters, union efforts could help him make inroads in Clinton's constituencies. Though the Clinton campaign is mocking Obama for welcoming this help less than two months after he blasted Edwards for receiving independent help by unions, which Obama dismissed as special interest "Washington money." (Clinton has done a very good job of making it difficult for Edwards to jump in the race and endorse Obama.)

Two Ohio polls, first, paint a contrasting picture:

• PPP shows a very tight race, 50% to 46% in Clinton's favor. PPP notes that GOP and independent voters are flooding the Democratic primary now that the GOP contest is over and supporting Obama 80% to 13% and 64% to 33% respectively. Clinton is ahead among registered Dems 56% to 40%.


• ARG, meanwhile, shows Clinton ahead 49% to 39%, in line with what we saw this morning.

In Texas, however, Clinton is losing ground:

• Rasmussen shows a toss-up, 46% for Hillary and 45% for Obama.


• ARG shows Obama way ahead, 50% to 42% -- it is worth noting that ARG was the only pollster to have Obama up already a week ago (48-42), which suggests that their model turnout significantly differs other pollsters'. ARG results in most states have looked like outliers.


• And finally, in what is perhaps the most important poll of the day, CNN shows Obama up 50% to 46% in Texas, an improvement from Clinton's 50-48 lead last week. While Obama's lead is within the margin of error, the trendline in his favor is confirmed by the other polls and he has one more week to build on his momentum. CNN does point out that Clinton's support among Latinos is holding firm.

Keep in mind that the primary vote only decides 2/3rds of the delegates, and that most estimations hold that Clinton would have to win by a significant margin to get a delegate lead in those 2/3rds given the Texas allocation rules. A large primary win for Hillary in Texas is looking increasingly unlikely... And then Clinton will also have to face the caucus results, which could be disastrous for her in Washington state is any example (she lost the primary by 3% and the caucus by 36%), getting Obama a significant portion of the third of delegates awarded through the caucuses. Bill Clinton is already pushing back on the caucuses (in a strategy the CLinton campaign started in the days following Iowa): "The doors open at 7 and they close at 7:15. It would be tragic if Hillary were to win this election in the daytime and somebody were to come in at night and take it away."

What constitutes a loss for Clinton in Texas, or what constitutes a victor? Does she need a primary win, a primary delegate lead or an overall delegate lead? Will she lose if she get the first but not the rest? Will she be deemed the winner but not by enough to continue, and would she use a win on points to get out of the race gracefully?