A new post-primary Rasmussen survey of likely Colorado voters puts GOP U.S. Senate candidate Ken Buck ahead of Democratic Sen. Michael Bennet, 46 percent to 41 percent. The results almost exactly mirror the results of Rasmussen surveys conducted at the beginning and end of July.
Although many who see a consistent right bias in Rasmussen’s methods will question the survey, Rasmussen trends right intentionally this year based on an assumption that doing so will produce results that will more accurately reflect actual voting. Polling analysts at Five Thirty Eight write that this Rasmussen “house effect” is the product of Rasmussen’s theoretical “model of what the 2010 election is going to look like.” Rasmussen believes this year the electorate will be more conservative. As Five Thirty Eight explains, the accuracy of the Rasmussen 2010 model will be tested in November.
The latest Rasmussen Colorado poll found 5 percent of likely voters prefer neither Bennet nor Buck and that 7 percent are still undecided.
A July 27 poll saw Bennet pulling down only 42 percent of the vote against Buck.
In that same late July poll, Rasmussen leaned heavily toward Norton in the GOP primary match-up she lost to Buck. Rasmussen reported Norton leading Buck 48 percent to 39 percent. Buck appears to have won the primary by roughly 4 points.
Today’s survey results are based on August 11 polling of 750 likely voters in Colorado. The margin of sampling error is +/-4 percentage points.
* Correction: I read the July 27 Rasmussen poll incorrectly. Rasmussen had Norton leading Bennet not Buck 48 percent to 39 percent. I regret the error. Thanks to Debra Falk at Rasmussen and the careful readers who pointed out the mistake. –JT