About this Blog
About The Author
Email Me

RealClearPolitics HorseRaceBlog

By Jay Cost

« Clinton's "Moral Claim" | HorseRaceBlog Home Page | An Email from a Smart Reader (And an Update to Today's Column) »

How Clinton Won TX and OH

After the Wisconsin primary - there was evidence of pro-Obama momentum. There is no evidence of this from yesterday's two big contests in Ohio and Texas. In fact, Clinton not only regained ground she lost with her best groups, she made marked improvements among key portions of Obama's best groups.

To begin, let's look at Clinton's performance among her best demographic groups. The following chart reviews Clinton's net margin among these groups in non-Southern states through February 18th, Wisconsin, and Ohio.

Clinton in Ohio.gif

As this makes clear, Clinton came roaring back in Ohio last night - winning voters that she had won through February 18th (i.e. the contests through the Potomac Primary), but that she had lost in Wisconsin.

What about Obama? He under-performed in his key groups in Ohio last night. The following chart reviews Obama's net margin among his strongest demographic groups:

Obama in Ohio.gif

Clearly, Obama had the exact opposite experience in Ohio last night. He improved in Wisconsin relative to his prior results, but Ohio was a slide - not just relative to Wisconsin, but relative to his performances prior to it. It was Clinton, not Obama, who won white males, non-union households, the wealthy, and white Protestants. This was a change from past contests.

One constant we saw again in Ohio last night was that Clinton did well among late deciders. She won those who decided the day of the primary by 11 points, about what she won the entire vote by. She won those who decided three days before by 26 points. This is actually an improvement for Clinton. Normally, Obama performs better among voters who decide three days prior. Clinton also did better among voters who decided a week ago. Again, this is a change. Usually, Obama does better with them. All in all, it appears that Clinton again closed well. This time, she started closing a little earlier than she has in the past.

And what of Texas? We see the same basic narrative - although Texas did not favor Clinton as strongly. Let's take a look at the numbers for Clinton in her best groups.

Clinton in Texas.gif

These numbers are roughly consistent with what we have seen in the South to date. Clinton did roughly no better and no worse than she has in other states in the region.

As for Obama, here is how he performed in Texas:

Obama in Texas.gif

Typical for a southern contest - Clinton won white men (here Obama did better among white men than he has in other southern contests) and white Protestants. She did enjoy notable improvement in her standing among wealthier voters and Independents - two groups that Obama typically wins in the North or the South.

What of minority voters, namely Hispanics and African Americans? They performed largely as we have come to expect. In Texas, Obama won African Amerians 83-16, which is about what he has done time and again. In Ohio, he won them 87-13. Clinton won Hispanics 67-31 in Texas - again consistent with how she has performed to date.

What can we conclude from all of this? It should be clear that Texas and Ohio performed in a manner roughly consistent with the states prior to Wisconsin. From this, we might infer that any momentum that Obama developed after the Potomac Primary was not carried through yesterday. Wisconsin did not help him in Texas and Ohio - as Virginia, Maryland, and DC seemed to help him in Wisconsin. The states voting yesterday seemed to vote "normally." Over the next few days, I'll explore this in a bit more depth - making use of the vote totals as they become finalized.

Another point. Last night Clinton made only modest gains among the pledged delegates. As of this writing, no estimate for the delegate allocation in Texas was available, but through the other three states Clinton only netted 26 pledged delegates. This bodes well for Obama, and it is consistent with what we had expected.

However, with some votes left to be counted in all four states, Clinton netted about 330,000 votes on Obama. Those RCP popular vote counts have shifted. Clinton cut Obama's lead that excludes Florida and Michigan by more than 30%; she cut the lead that includes Florida by more than 50%; and (as of this writing) she seems to have erased Obama's lead in the count that includes Florida and Michigan.

This definitely puts her in striking distance of the popular vote lead that includes Florida. She has to win the remaining vote by about 6 points to draw even with him on that count. While it is far from assured that she will do this, it is quite plausible. [She'll have to win by about 10 points to draw even in the count that excludes Florida and Michigan - so that remains more difficult for her to achieve. I'll offer more precise estimates on all these figures after the votes have been fully tabulated.] If she does eliminate it, I think she will have an argument to take to the super delegates. That's not to say it is the more compelling argument; Obama will have a good one of his own. The point is that if she catches him in the vote count that includes Florida, she will have an angle on victory. She took a big step toward catching him last night.

Final point. The Obama campaign is proclaiming they won the Texas caucus by double digits. Indeed, that seems to be the case. Nevertheless, they need to be careful not to proclaim this too loudly. How will it look if Clinton wins a majority of the more than 2.5 million Texans who voted in the primary, but Obama wins the caucus in which about 100,000 people participated? [See My Update to this Point Here] That might help Clinton because it is evidence that the caucuses are not a good gauge of voter preferences. Obama needs to talk up his pledged delegate lead, without reminding people of how it is heavily dependent upon the caucuses. The Clinton camp is going to start attacking these caucuses.

-Jay Cost