As you may be aware, GOPers have been charging for the past week that all national (and presumably state) polls showing a deep plunge in the fortunes of Romney-Ryan are wrong—for one reason or another but mainly because they are “skewed” and “over sample” Democrats. I’ve pointed to the fallacies in this in recent days, with GOPers relying on outdated numbers for the percentage of self-identified Republicans in the US. Many who once called themselves GOPers now tell pollsters they are Democrats or Independents since they are disgusted with the Republican brand and plan to vote for Obama, and maybe other Dems, this time.
In the fantasy land of the right, however, these polls are all “skewed” because they “under-sample” proud Republicans. I haven't seen this much panic among GOPers since Upton SInclair was nearly elected governor of California in 1934 as a real Socialist-Democrat, not the Obama variety.
But the conservatives’ chief defense? The good old Rasmussen Poll, long “skewing” to the right and a favorite of Fox News and Red Staters, and now, reliably, the only poll showing an edge to Romney. This morning’s daily tracker finds an even split but a 2 percent edge for Mitt when the “leaners” are added. That’s been the case in recent days.
Now, it’s true, that Rasmussen’s final poll in 2008 came closer to the final tally than nearly anyone else’s. But the differences were not great and none had called the race for McCain. The problem for Rasmussen today appars two-fold: it continues to “weigh” the tally based on GOP registration figures—not the current mood of GOP voters—and unlike many polls does not reach those who only use cellphones.
Now, we all know polls can be “wrong” or at least somewhat off. That could even be the case now even without obvious “skewing.” But the Rasmussen Poll not only flies in the face of Gallup and Bloomberg (which now find a 6 percent Obama lead) but other major polls that find an even wider gap—and dozens of state polls which seem to reveal Obama opening up not just leads but big leads in major states such as Ohio, Florida, even Virginia. See a full rundown on recent polls here.
That’s on top of double-digit landslides in large states such as New York, PA, Massachusetts, Illinois and New Jersey. And a 20 percent or more lead in easily the largest of all, California. If anything, one wonders why a 6 percent national lead isn’t larger.
So: are all these state polls—which come from numerous sources, not usually “Democrat”-hired and some, in fact, that have always leaned right—also wrong? Do they all sample too many Dems? If they are close to being right, then, for Rasmussen to find an even race, it must mean Romney is winning Texas by, oh, 91 percent to 9 percent— in fact, he would have to be carrying the entire red South by about those numbers to make up for the multimillion-voter Obama edge just in New York and California. But, of course, polls in those states, show no such thing.
But perhaps we should thank Rasmussen for its methods and orientation that have led to the current state of GOP complaceny. The Romney camp seems to not feel it needs to shake up its team or change focus since it is, as the candidate says, “tied” with Obama and “it’s still early.” So, thank you, Scott Rasmussen.
One more thing: The other GOP delusional and oft-repeated claim is that all the polls had Carter beating Reagan in 1980. See my new posting here about how, in fact, virtually every major poll had Reagan winning as Election Day neared that year, some by as much as 5%.
Greg Mitchell’s books and e-books on influential American campaigns include Tricky Dick and the Pink Lady, The Campaign of the Century (on Upton Sinclair’s 1934 race, recently highlighted at The New Yorker ) and Why Obama Won. He also blogs daily at Pressing Issues.