October 02, 2010 07:10 PM UTC

October Strategy: Base Suppression, Fooling "Likely Voter" Models

•  
• by: Colorado Pols

From today’s McClatchy report on their new poll of key Senate races, including Colorado’s:

In Colorado, Republican Ken Buck leads incumbent Sen. Michael Bennet by 50 percent to 42 percent among likely voters…

In each state, Republicans are benefitting from an enthusiasm gap, where their supporters are much more eager to turn out and vote on Nov. 2. In Colorado, for example, the Democrat leads 41 percent to 40 percent among registered voters. However, the contest flips among those most likely to vote, who give the Republican an 8-point lead.

“We define likely voters as registered voters who report they have an excellent or good chance of voting in November and express a high to moderate degree of interest in the elections,” said Lee M. Miringoff, director of the Marist Institute for Public Opinion at Marist College in Poughkeepsie, N.Y., which conducted the polls…

In Colorado, likely voters are in a sour mood.

They think the worst is yet to come for the economy by a margin of 49 percent to 41 percent. That anxiety is worst among those making less than $50,000 a year, a group that usually favors Democrats.

Just 39 percent of likely voters approve of Obama’s job performance, while 56 percent disapprove, his worst margin in the three states. While that score is tilted a bit by 93 percent disapproval among Republicans, even 54 percent of the state’s independents, a key voting bloc, disapprove of the president.

The Senate race is beginning to narrow into a few key themes: one of the biggest of which is an emerging gap between modeled “likely voters,” which for a variety reasons are tilted toward Republicans as this poll shows, and actual registered voters–in Colorado, in this poll, it’s the difference between Michael Bennet narrowly leading versus losing ground to Ken Buck. It is worth nothing that the spread between “likely” and actual registered voters in this poll is larger for Colorado than the other states surveyed: in Wisconsin, the poll of registered voters still doesn’t have Russ Feingold ahead.

What this means is that October is going to be defined by Bennet’s campaign doing everything it can to drag less-enthusiastic Democratic base voters back to the polls, principally by driving up Buck’s negatives–while Buck’s campaign’s #1 goal is to suppress interest in Bennet among those same unenthusiastic and, in many cases, low-information voters. You’ll recall that key to Bennet’s primary victory in August was his ability to turn out votes that weren’t adequately reflected in “likely voter” polls showing a broad swing to Romanoff. That performance, and the acknowledged expertise of Bennet’s campaign staff, gives hope to Democrats that they can pull a victory right out from under these assumptive poll models.

Working against Bennet is the growing disparity of independent expenditures in this, as with other races around the country. There has been significant press this past week about reduced giving by many high-level Democratic funders this year, whereas conservative groups like Karl Rove’s “Super PAC” American Crossroads have invested truly unprecedented amounts of money in nationally strategic races–including Colorado. There is a risk that deteriorating narratives elsewhere could result in unilateral wholesale disarmament everywhere by major Democratic funders, which would prove disastrous for yet-winnable races like this one. That hasn’t happened yet, but neither has giving to match Republican-leaning independent efforts.

In the meantime, strategists at the Democratic Senatorial Campaign Committee and the White House absolutely do consider Colorado a winnable race, and there is no indication whatsoever that resources for this race under their control are anything other than top priority–in fact, watch for the DSCC to start cutting others loose to reinforce here, for all the reasons stated above.

And get ready for a month of major combat, folks.