Bennet Favored, GOP Gov. and Sen. Primaries Up in the Air

With one day to go until all the votes are in, it is anybody’s guess who will win the Republican primaries for governor and Senate, and even incumbent Democratic Senator Michael Bennet could sweat out election night before he finds out whether he is nominated for his first election to the seat he now holds as an appointee.
Bennet leads Andrew Romanoff, 49-43, with 9% still claiming to be undecided.

Romanoff has gained 9 points on Bennet since PPP’s last poll of the race in mid-May,
when Bennet was ahead, 46-31. In the suddenly most dynamic race of the three, Dan Maes has closed the gap with previous favorite Scott McInnis following numerous charges of plagiarism against McInnis. Both Maes and McInnis have also been assailed by Republican Tom Tancredo, who has entered the general election race for governor on a third-party ticket. In May, McInnis led a five-way contest 50-15 over Maes, with others barely registering. But in a simple two-way, Maes has gained a remarkable 34 points since then, and now trails by only one point, 41-40. 19% are still undecided. McInnis’ favorability rating has declined a stunning 43 points, from 54-15 to 36-40.

In a nine-candidate Senate test in May, Jane Norton led Ken Buck, 31-26, with the other seven in low single digits. While there has been considerably less movement toward the more conservative challenger compared to the gubernatorial primary, Norton now tops Buck only 45-43, with 12% unsure of how they will vote. Both candidates are equally well liked compared to PPP’s previous poll.

“There have been a lot of developments and scandals in Colorado that have rapidly
changed the shape of all three races, and turnout is going to be key for whoever ends up on top Tuesday night,” said Dean Debnam, President of Public Policy Polling.
PPP surveyed 448 likely Colorado Democratic primary voters, with a margin of error of +/- 4.6%, and 767 likely Republican primary voters, with a margin of error of +/- 3.5%, from August 7th to 8th. Other factors, such as refusal to be interviewed and weighting, may introduce additional error that is more difficult to quantify.