Can All the Polls Be Screwy?

Our friend Dafydd ab Hugh continues the conversation Paul and I have had over whether the polls that consistently show bad results for Republicans can all be wrong. Actually, Dafydd says, they can. I won’t try to summarize his reasoning; suffice it to say that I learned some things about polling from his post that I hadn’t understood before. Check it out, especially if you’re a Republican looking for a glimmer of hope.
Plus, Dafydd’s site is fun to look at; click to enlarge, then go there!

PAUL adds: It’s possible, I suppose, that all the polls are screwy (and certainly possible that public opinion, wherever it is now, will move). But I don’t think the polls by organizations like Rasmussen are “oversampling” Democrats the way the big news organizations are.
Consider Ohio. CBS/NYT has the Democrat, Brown, up by 14 points. Rasmussen has him up by 6 points. Rasmussen may have the more reliable poll, but it still doesn’t look good for DeWine. Similarly, in Maryland USA Today/Gallup has Cardin up by 15 points, compared to 9 points in Rasmussen’s poll. The MSM polls may be unreliable, but you probably wouldn’t bet on Steele.
When you get to Missouri and Tennesse, though, the distinction arguably starts to make a difference. In Missouri, USA Today/Gallup has the Dem up by 3 points; Rasmussen has Republican Talent up by 1 point. This is all within the margin of error, though. In Tennesse, USA Today/Gallup has Ford, the Dem, up by 5 points (more than the margin of error); Rasmussen has him up by only 2 points (less than the margin). And some non-MSM polls have Corker leading Ford.
What to conclude? That the White House/Mehlman/Winston thesis may provide some comfort in a few states (and some congressional districts), but the oversampling of voters who identify themselves as Dems probably does not explain away Republican woes.
MORE FROM PAUL: I shouldn’t blog while watching a ball game. Having read Dafydd’s post more carefully, I see that he covered the basic points I made (which shouldn’t come as a surprise).
I wrote about the sampling issues Dafydd discusses here. The links will take you to the counter-arguments of the MSM pollsters.
UPDATE: John Hawkins of Right Wing News has a good roundup of the Dems’ 20 best pickup opportunities in the House. Bottom line is, it’s going down to the wire.