Shoddy reporting on mediocre polling

I was surprised by this headline:

"California poll shows strong leads for McCain and Clinton"

Because I'm still not seeing a lot of "Hillary" banners around here. More Obama. But this is not LA.

And then I read:

On the Democratic side, the California poll brought troubling news for Barack Obama, who trailed Hillary Clinton by double digits, 49% to 32%. That gap only amplifies the importance of Obama's new backing from senator Edward Kennedy, whose visit to the west coast this week will give him an opportunity to help Obama make inroads with sceptical Latino voters.

Not only is he behind, it's a big margin? Surprising.

Then:

Yet the poll's Democratic numbers were based only on voters affiliated with the party, when California Democrats hold what is known as an open primary, independents are permitted to cast ballots.

Dear God. Any sixth grader understands that the Democratic primary in CA is open, and the Republican one is not. Ergo most independents will be voting in the Democratic primary. And most of Obama's support is from independents.

So: This poll is kind of useless for predicting election outcome, given that a very decisive voting block wasn't sampled.

And also: This headline is a bit of a distortion, in that, while this poll shows a lead among registered Democrats, it does little to predict the likely outcome of the election.

At least they're honest with the facts of what the poll is and is not in the body of the article, just wish they had been honest with their headline. Of course, it worked. It got me to read the article.