Tuesday, September 29, 2009

Deeds continues to close in

Creigh Deeds has now pulled to within five points of Bob McDonnell in the race to be Virginia's next Governor, and most of the remaining undecided voters are Democrats.

McDonnell leads 48-43. A month ago his lead was 49-42 and the month before that it peaked at 51-37.

McDonnell owes his continuing advantage to two primary factors: a 53-37 lead among independents and a much higher degree of unity within his party than Deeds enjoys. He is winning 96% of the Republican vote while Deeds is at 82% of the Democratic vote.

Deeds appears to have more room to grow. 53% of the remaining undecideds are Democrats while only 7% are Republicans. Although the fact that Deeds has not locked up those votes yet does show some degree of lukewarmness toward his campaign, those voters are still more than likely going to end up 'coming home.'

The increasingly negative tone of the race appears to be hurting both candidates' standing with the voters. McDonnell's net favorability has dropped from +22 (53/31) a month ago to now just +5 (47/42). Deeds has seen a similar although less dramatic decline from +12 (47/35) to +1 (43/42).

Bob McDonnell's thesis is having a mixed impact on the race. Only 2% of people who say they supported him a month ago now say they're going to vote for Deeds, so the extent to which the thesis is changing people's minds is limited. But it may be playing a role in increasing Democratic turnout. In our last poll those planning to vote this year had voted for John McCain by a 49-45 margin. Now the likely electorate voted for Barack Obama by a 48-45 margin, indicating intended Democratic turnout is now pushing closer to what it was last year. The thesis may not have turned McDonnell votes into Deeds votes, but it looks like it is helping to turn non-voters into Deeds votes.

52% of voters say they're very familiar with the thesis and McDonnell actually has a 55-41 with that group, reflecting the fact that Republicans are more engaged this year and following the campaign more closely. Deeds is up 56-41 with the 29% of voters who claim moderate knowledge of the thesis.

There's been little movement among women over the last month. What was a nine point lead for Deeds with them is now an eight point advantage.

This race is being defined largely by rigid partisanship. McDonnell is winning only 8% of the Obama vote and Deeds is winning only 7% of the McCain vote so by and large however you voted last year is the predictor for how you're going to vote this year. That means this is all going to come down to turnout. If the electorate looks the same way it did last November Deeds will probably pull this one out by the skin of his teeth. But if Republicans continue to be more energized (or Democrats are more disinterested) McDonnell will continue the lead he's held throughout the general election right through the finish line.

Down ballot Bill Bolling leads Jody Wagner 43-35 for Lieutenant Governor and Ken Cuccinelli is up 43-34 on Steve Shannon for Attorney General.

Full results here


Anonymous said...

Looks like you have too many Democrats and Indepedents in the sample and not enough Republicans.

D: 37%
R: 29%
I: 34%

Virginia's exit polls last year were:

D: 39%
R: 33%
I: 28%

Tom Jensen said...

We also have an electorate that voted for Obama by 3 when he actually won the state by 6. Must not have enough Obama voters in the sample.

You can play that game both ways.

Anonymous said...


We'll see. SurveyUSA should have their Virginia numbers out tomorrow. Obama was at 42/54 in its last VA poll, though he might have improved since then. We'll see.

Rasmussen had Obama at 47/49 in its last Virginia poll.

If you've got Obama in the positive side, you may be the only one of your IVF counterparts to show such a result.

So yes, the game can be played multiple ways.

Either way, Deeds appears to be toast.

sjc said...

Party ID is not a fixed value. It is what people say it is at the time of the survey.

Stephanie said...

I love the explanations you give for the data here; I also love the fact that PPP usually polls candidates and opinions about individuals, not such nebulous, easy-to-influence questions as "what do you think of Obama's plan for sand and gravel"? I'm curious whether you've seen David Moore's book about the polling industry, and if so, what you think. I'm also hoping you'll stop using "disinterested" to mean "uninterested" ("disinterested" once meant "impartial," "without a vested interest in an outcome," "able to judge fairly," though it gets used in other ways so often that it might just be time to avoid the word). But it's a small complaint. I enjoy the blog.

Anonymous said...

"Rasmussen had Obama at 47/49 in its last Virginia poll."

Red herring and credibility alert! Rasmussen is a right-wing Republican pollster who skews hard to the Republican's favor. Just look at pollster.com to compare his results to all the others. He is far out of the mainstream re: Obama approval/favorable. A Republican relying on a skewed-to-the-Republicans poll is quite disingenuous. You can safely add several points to any Democratic numbers from any Rasmussen poll to get a more accurate assessment. Which is shown by looking at similar polls. Republicans love to quote Rasmussen because he is so skewed...but doesn't reflect reality. The short of it...extremist and sexist McDonnell will be toast come November. Deal with it.

Anonymous said...

"Must not have enough Obama voters in the sample."

Uh no, you have an electorate that can't remember.

You have the voters saying they voted for Kaine over Kilgore by 12% (not by 5%-6%) back in 2005.

2008 was the best year for Democrats in the past decade and to have more Democrats in this sample than last year . . . it's just silly.

Ryan said...

How come we can get McDonnell's thesis but not Obama's? Just sayin.

Anonymous said...

Rasmussen is one of the most accurate polls out there.

They had McDonnell leading Deeds by just 2% two weeks ago and now PPP comes out with a poll showing McDonnell with a 5% lead.

I guess that makes PPP a Zionist, neo-con poll.

Anonymous said...

"Red herring and credibility alert! Rasmussen is a right-wing Republican pollster who skews hard to the Republican's favor."

Notice I also mentioned SurveyUSA.

All the pollsters are in a conspiracy against Obama.

Web Statistics