The latest Zogby/Wall Street Journal/Battleground States Poll, released today, describes a governor’s race in which the major party candidates have broken away from the independents. The results:

Perry 37.5%
Bell 26.2%
Friedman 13.2%
Strayhorn 13.0%
Werner (L) 3.9%

The first question that should be asked about any Zogby poll is: Is it credible? Zogby does not use random samples; its respondents come from a list of people who have signed up to be polled. Candidates’ Web sites–Barbara Radnofsky’s, for instance–urge supporters to sign up. Zogby has consistently had higher numbers for Bell than other polls, which indicates to me that people who identify with major parties are more likely to sign up for Zogby than independents, whose interest in politics, one can infer, is less intense than partisans’.

Compare Zogby with Opinion Analysts’ poll for Texans for Insurance Reform that was released yesterday:

Perry 34%
Strayhorn 19%
Bell 19%
Friedman 9%

Except for Friedman, this is almost a carbon copy of the most recent poll by Rasmussen:
Perry 34%
Strayhorn 21%
Bell 19%
Friedman 18%

I think that both Strayhorn and Friedman’s numbers in Zogby are too low, and the reason is that independents are underrepresented in the sample. I think Bell’s numbers are too high, and perhaps Perry’s as well, because major party supporters are overrepresented. Bell was at 15% in the October 2 Dallas Morning News poll. Performing well on the debate gave him a bump, and so did getting his spots on TV–but an 11 point bump? No way. Something is askew in the poll.