| So I've gotten the feeling that our prospects in the Senate have been sinking recently, even more so than during the last quarter of 2009. So I asked myself, "Why do I have that feeling?" And then I went back and looked. The answer in more cases than not is Scott Rasmussen.
I'm not saying Rasmussen is a bad pollster. In fact, he may just be ahead of the curve in terms of predicting what may be a dismal Democratic turnout in 2010. But he is an incredibly frequent pollster, and his polls have dominated the narratives in many of these races as a result of their sheer frequency.
Here are the races rated by Cook as Lean Retention or better for the challenger:
(1) ND-OPEN - Hoeven's dominance here has been tracked by several pollsters. Not a case in point.
(2) DE-OPEN - The proposition that Castle v. Coons is a washout is based on a single Rasmussen poll taken January 25 showing a 56-27 Castle lead. There is no other recent polling.
(3) AR-Lincoln - Ras is at least corroborated by PPP in showing Lincoln's sorry ass getting blown out.
(4) NV-Reid - Much like Arkansas, PPP corroborates Rasmussen's solid R leads.
(5) CO-Bennet - All of the gloom and doom in this race comes from two recent Rasmussen polls showing double-digit leads for Norton over Bennet. Research 2000 actually showed a small lead for Bennet only a month ago.
(6) PA-Specter - Again, the gloom and doom here comes from two recent Rasmussen polls showing 9-point leads for Toomey over Specter. Quinnipiac showed an even race on December 8.
(7) IL-OPEN - Once again, the gloom and doom here comes from a single Rasmussen poll showing Kirk up 6, which was directly contradicted by a PPP poll just a week prior showing Giannoulias up 8.
(8) MO-OPEN - Yet again, more gloom and doom exclusively from Rasmussen, showing Blunt up 7 and 6. Every non-Rasmussen poll has Carnahan ahead.
(9) OH-OPEN - Again, the narrative that Portman is winning comes from Rasmussen, although Quinnipiac had a 3-point Portman lead back in November.
(10) NH-OPEN - Several polls have corroborated Rasmussen's high single digit lead for Ayotte over Hodes, so this is not a case in point.
(11) KY-OPEN - Like New Hampshire, Rasmussen's polling showing high single single digit leads for Republicans is corroborated by other pollsters here.
(12) IN-Bayh - The only reason that this race is viewed as competitive as far as I can see is a Rasmussen poll that showed Mike Pence up on Bayh and John Hostettler within 3. Today, Research 2000 showed Bayh up 16 on Hostettler and 20 on Indiana-hating Dan Coats. Cook has now moved this from Safe D to Lean D, presumably based largely on Rasmussen.
(13) CA-Boxer - Kind of like Indiana. The main reason this race is viewed as competitive is Rasmussen's polling, starting in July when Ras showed a 4-point race with Fiorina while others showed 15 to 20 point leads. In fairness to Ras, a recent PPIC poll showed Tom Campbell within 4, giving some corroboration for Ras's take. But nobody else has had Fiorina closer than 8. Cook has had this at Lean D for some time, and I suspect that was partly based on the July Rasmussen poll.
(14) CT-OPEN - Ras shows a pretty solid Blumenthal blowout, although less so than other pollsters. Not a case in point.
I am using Wikipedia to track polling, and may be missing some polls. Please correct me if I am mischaracterizing anything.
Of these 14 races, I would say that Rasmussen has had a stranglehold on the recent gloom and doom narratives in 7: DE, CO, PA, IL, MO, OH, and IN. Put another way, I have been led to the subconscious belief that we are going to lose the first 6, and be in for a dogfight in IN, strictly based on Rasmussen polling. I would also put CA in pretty close to the same category as IN, although PPIC did recently confirm a close race with Campbell at least.
I do not think this is an accident. I do not remember this kind of frenetic pace from Rasmussen before Obama took office. SSP recently suggested Rasmussen has gotten so prolific that he could be called "spammy." My gut tells me Ras is getting as many polls out there as he can precisely so that he can dominate the narratives with his polls and their aggressive turnout model. Combine this with his right wing framing on issue polling, his inexplicable use of an aggressive likely voter screen for presidential approval three years before the election, his haste to poll Republican "dream" candidates, and his frequent yucking it up with conservative talking heads, and you've got yourself a Republican cheerleader trying to influence elections rather than study them. Again, his polls may be right. But his transparent efforts to drive the narrative seem very partisan to me.