Houston Democrats

I've worked in market research for many years and I can tell you that when pollsters start weighting their results to match what the demographics are, there's a very good chance that their demographic sampling itself is skewed--that is, there's too many people in one geographic zone, or too many of one race or one age group--and this sampling may be providing distorted results.

When that's the case, the margin of error is going to be bigger than the statistical probability models would give you for a truly random survey. That said, the other thing I've found out about the science of statistical sampling is to trust the randomness. You can really sloppy in methodology in telephone sampling and still count on getting results that are a pretty close representation of the universe.

In other words, even if Bell is a bit above 15%, he ain't a lot above 15% and we've got our work cut out for us this year.

Looking closer at the demogs, you can get a good feel for where the candidates' bases lie. Perry and Friedman are stronger among men. Bell and Strayhorn do their best among women (but the Hair is still pulling a plurality of women's votes here--thanks, ladies!)

Perry is also strong among Hispanics, earning 45% of the polled Hispanics' support. Bell leads only among black voters, but only with 40% of this key Democratic base. Neither Bell nor Strayhorn's support correlates to levels of education; there's little drift in support for them as voters slide up and down the key socioeconomic scales of education and income. For education, Strayhorn and Bell supporters tend to show the same balanced profile.

In contrast, Friedman is more likely to get Texans' support based on how much schoolin' they got. He has 21% support among people with post-baccalaureate degrees. One sign of this survey's bad demographic skew is shown by the fact that 24% of respondents had graduate degrees--hardly representative of Texas voters. Perry has the opposite problem. He has 47% of the vote among people who've never been to college. As education levels rise, support tends to drop away from Mr Handsome. Go figure.

But if this survey is undersampling people with lesser education levels and not correcting for that in the weighting, then Perry's support may be stronger than the 39% that Survey USA is reporting.

Party and ideology carry few surprises. One of those surprises is that the survey found Bell with 32% among Democratic voters vs Strayhorn's 30% among Democrats. Friedman has a slight plurality of 32% among independents. Strayhorn also has 30% among self-identified moderates (who make up 35% of the electorate, according to Survey USA).

From the looks of this data, I'd say that Bell's real race right now is against Ms Strayhorn, not Kinky Friedman and not Rick Perty. She's soaking up the rural, moderate, and North Texas votes that a Democrat would normally pick up. She's pulling a statistical tie with Bell among our own party. She's sucking up the money that would normally go to our nominee.

Friedman, whom more progressives are worried about, is mostly pulling new voters into the system; he's not robbing us. He's doing okay among liberals (25% compared to 28% of liberals who are supporting Bell), but that is a self-declaration and may reflect social liberals who are apathetic or uninformed about the role that government plays in providing economic opportunities and growth. Friedman's campaign still looks like an ego trip, a cutesy marketing campaign for his current overstock of unsold CDs, a bid for a regular gig at Branson, Missouri, or perhaps the next FarmAid concert--as if the yuppie dilettantes who think Kinky's gravely mumbles are real country music give a rat's patootie about family farms. And his books suck, too.

But I digress.

The point I'm staggering toward is that Strayhorn is the candidate holding Bell down in the teens. Maybe people are nostalgic for an Ann Richards type of governor. Maybe Bell is projecting too urban an image. Despite the fact that Texas is a large, mostly urban state (moreso than California or New York, where small cities form counterweight voting blocks to offset their metropolises), Texans still want to think of themselves as small town, rural folk. That's our mythos and that's a base, this survey suggests, that Chris Bell is not reaching.

November's a long ways away, of course, and the survey, I suspect, overweights Friedman's base. But the results are sobering and a challenge to all of us this fall. Bell's message must be tailored to the voters he's not reaching and Strayhorn's substance-free feel-good campaign must be exposed for the empty shell that it is. We can all make history here, but the clock is running on us.