It may be a little premature to start writing Michael Bennet's political obituary...PPP's newest Colorado poll finds him holding onto a 46-45 lead over Ken Buck.
Both candidates have their party base pretty much completely locked up: Bennet is winning 85% of Democrats and Buck is winning 84% of Republicans. Bennet's slight edge comes because of something that is pretty unusual for Democratic candidates across the country this year- he's ahead 48-38 with independent voters.
Bennet's lead with independents is not because they like him- in fact they don't. His approval rating with them is a negative 37/45 spread. Rather it appears to be the price to pay for Republicans nominating a candidate with limited appeal to the center. Independents see Buck unfavorably by an even wider spread, 31/50.
A closer look at moderate voters, who comprise the majority of independents in Colorado, shows a lot of trouble for Buck. They see him negatively by greater than a 2:1 margin, 27/57. They're not real high on Bennet, giving him only a +7 approval rating at 43/36. But when it comes to the horse race they give the incumbent a 24 point lead despite their tepid feelings toward him personally because of their greater animosity toward Buck.
Bennet continues to be, as we've found in all of our polling for the whole cycle, an unpopular Senator. His approval rating is just 35% and 49% of voters disapprove of the job he's doing. If a moderate Republican had been nominated against him this race might be over. But Buck's unpopular too at a 41/46 favorability spread and that's keeping this race close.
It should be noted that these poll numbers are predicated on an electorate that split its votes evenly between Barack Obama and John McCain, suggesting a pretty large drop in Democratic turnout from 2008 given that Obama actually won the state by 9 points. This is one race where if there is any closing of the enthusiasm gap over the last four weeks it could put Bennet over the top.
Expect this to be one of the closest races in the country.
Full results here