Trailhead: A campaign blog.

The Popular Vote Chronicles: Don’t Forget Texas

Something we didn’t mention in our assessments of Hillary Clinton’s claim that she’s winning the popular vote: the Texas caucuses.

We and many other outlets have taken to using the Real Clear Politics popular vote count. The problem is, RCP factors in the Texas primary but not its caucuses. As a result, we end up underestimating Barack Obama’s popular vote tally. But by how much?

One way to estimate is to look at the results from the evening of March 4. Texas uses a “voluntary” reporting system, so only 41 percent of precincts ended up reporting their results on election night. Those numbers showed Obama winning the caucus by about 10 points. We can also look at the results of Texas’ county and state district conventions in late March, in which Obama won 58 percent of the vote to Clinton’s 42 percent. Again, that’s rough, but it’s the best we’ve got until the state convention in early June.

Based on those numbers, it looks like Obama won by anywhere from 10 to 20 points. (There are no official figures.) The Texas Democratic party estimates that turnout was roughly a million, which means that Obama probably netted anywhere from 100,000 to 200,000 votes—enough to cancel out Clinton’s 100,000-vote victory in the state’s primary.

You can see why outlets like NBC choose to ignore Texas entirely when counting votes. “We’re just all screwed up,” laughs Texas Democratic spokesman Hector Nieto. “We’re the only state with stripes on CNN.” There are other caveats: The March 4 caucuses were chaotic, with overflowing caucus sites and accusations of voter fraud. Also, Texas voters could vote in both the primary and the caucuses, and there’s no way to figure out exactly how many votes got counted twice. (Update 8:14 p.m.: Actually, there is: All of them got counted twice, since you had to vote in the primary in order to attend the caucus.)

But if you factor in this rough estimate of the Texas caucus results, Clinton is decidedly not winning the popular vote. RCP puts her ahead by 64,000 votes if you count Florida and Michigan and all the caucus states. But 100,000 votes from the Texas caucus would swing the advantage back to Obama.

As If March 4 Never Happened

Obscured in last night’s Mississippi results was the announcement by CNN that Obama won the Texas caucuses. The counting’s not done—we won’t have final results until March 29—but CNN still projected the likely delegate split:

After a comprehensive review of these results, CNN estimates that Obama won more support from Texas caucus-goers than Clinton. Based on the state party's tally, Obama's caucus victory translates into 38 national convention delegates, compared to 29 for Clinton.

And though Clinton won more delegates than Obama in the primary, 65 to 61, Obama's wider delegate margin in the caucuses gives him the overall statewide delegate lead, 99 to 94 — or once superdelegate endorsements are factored in, 109 to 106. [Emphasis added]

So … Obama won Texas? Depends on which count you think matters more—the popular vote or the delegate count. (There's plenty of debate over that.) It also depends on whose numbers you believe: MSNBC still has Texas as tied. But at the very least, if the final tallies on March 29 corroborate these numbers, Obama can make the case that he won Texas. (A case that, to be fair, his campaign has been making all along.)

Also, as First Read points out, this means that Obama’s victories in Wyoming and Mississippi do indeed cancel out Clinton’s March 4 victories. She netted about 15 delegates in the primaries that day, but Texas’ caucuses cut that number to six. In Wyoming, Obama netted two delegates and another five last night in Mississippi—thus erasing Clinton’s surge.

The math just gets uglier and uglier for Clinton.

The Texas Primary: Don't Mess With It

It’s fitting that a crazy, mucked-up Democratic primary season like this one comes down to the craziest, muckiest contest of all: Texas. The Lone Star State has never played a large role in a presidential primary—even Bill Clinton himself has said it’s all riding on March 4—so now campaign strategists and pundits alike are scrambling to master the state’s oddball system. (Hillary says the complex rules have “grown men crying” in the campaign.) In case you don’t spend your days and nights immersed in precinct-by-precinct analyses and delegate allocation procedures (PDF), here’s a quick primer on how it will all go down:

The Primary: Texas has 228 delegates in total, but only 126 of them are allocated like a regular primary. These pledged delegates are distributed among the state’s 31 senatorial districts—not, mind you, by congressional district. (This system has been in place for at least 20 years, so don’t think it’s the result of recent gerrymandering.) Each district gets between three and eight delegates, which it allocates proportionally based on the tally within that district.

But here’s the catch: Districts don’t receive delegates based on population or even registered Democrats. Delegates are allocated based on how many Democrats voted in the 2004 presidential election and the 2006 gubernatorial election. (A state party spokesman says the system is designed to “reward” participation.) For example, the state’s 14^th District, which includes Austin, receives eight delegates—the most of any district—because of its high turnout in previous races. At the same time, the 27^th and 28^th Districts, located on the southern border and West Texas, respectively, get only three delegates each because of low past turnout. 

So, who does this system benefit? Conventional wisdom says Obama. In previous elections, urban areas with high numbers of African-Americans and college students—Obama’s base—have turned out in droves. By contrast, rural areas with large white and Latino populations—groups that have favored Clinton—have participated less. (Hispanics leaned Republican in 2004, given Bush’s stance on immigration reform.) As a result, Obama’s ballot-happy constituencies start with a built-in advantage—their votes simply weigh more. As party leaders will remind you, however, that means nothing unless he can get them to turn out again. Plus, there’s always:

The Caucus: Starting at 7:15 p.m., right after polls close, Texas voters can show up to a district convention, also known as a caucus. This works a lot like other caucuses: Voters have to show up, stand in the same room, and be counted. (Unlike Iowa, however, there is no 15 percent threshold.) The main difference is that caucus-goers need to have voted in the primary earlier that day. So in a sense, caucus-goers get to vote twice. (And since fewer people turn out for caucuses than for primaries, a caucus vote has significantly more clout than a primary vote.) The state then allocates 67 delegates—one-third of total pledged delegates—based on who wins each district’s caucus. 

Again, Obama seems to benefit from the system. He has so far won every caucus state except Nevada (and maybe New Mexico). Theories abound as to why he fares so well in caucuses—possible factors include having a better field organization plus supporters with more free time and energy. If voters want to participate in the Texas caucuses, they’ll have to show up for two separate events, which won’t be easy for working parents. (As a result, many voters will probably show up late in the day—something to keep in mind when analyzing early results.) Both campaigns have strong ground organizations in Texas, but when logistics have been an issue in the past, Obama has typically fared better.

Superdelegates: Texas also has 35 superdelegates, composed of party leaders and elected officials. Among these, Clinton still leads Obama, 13-7, according to Politico’s superdelegate counter. But 14 are still undecided, and Obama has been picking them up at a faster rate recently. 

Early voting: Texas voters began casting early ballots on Feb. 19, and the numbers broke records from the start. More than 360,000 people have already voted, with some counties reporting 10 times the participation they had at this time four years ago. Early voting is expected to be a major part of the results—nearly 40 percent of voters in 2006 voted early. So far, turnout seems to be just as high in Clinton-friendly counties (Hidalgo, Galveston) as in those that should favor Obama (Travis, Williamson). 

Further reading: The Washington Post and Wall Street Journal on what Texas’ arcane system means for Clinton, the New York Times on how both campaigns are managing it, Slate’s "Election Scorecard" on the latest Texas polls, and Burnt Orange Report with an insanely fine-grained district-by-district analysis.

Trailhead thanks professor Bruce Buchanan of the University of Texas.