McDonnell by 14

Bolling by 13

Cuccinelli By 11

Yet Public Policy Polling has Deeds “closing” and down by 5.

What’s Going On?

In part, it’s the quirkiness of doing polls on the cheap these days.

But most of the explanation can be found in the diametrically conflicting models of the likely electorate found inside the poll numbers.

Public Policy Polling has an implicit turnout model heavily tilted toward the Democrats- in their survey, 37% of the elctorate is made up of Democrats and only 29% is Republican. 34% is Independent. It looks, I think, a lot like the presidential electorate last year.

Survey USA has a likely voter model that is tilted in the exact opposite direction.  37% is Republican, 32% Democratic and 29% Independent.

Here’s the difficulty.

In a gubernatorial election, less than 50% of registered voters will actually show up.

The fundamental issue for pollsters, then, is to determine who among this “less than 50%” is really going to vote.

You can imagine how important the issue is for candidates.

So here we are.

Same day.

Two polls.

Two entirely different electorates.

The bad news for Democrats is not simply that they’re behind in both.

But that they’re even trailing in one that has turnout tilting heavily toward¬†the D’s.


There’s over a month left.

And both could be off-base.

Up until SurveyUSA, Deeds supporters felt that the polls were trending in the right direction.

Democrats hope that the poll is an outlier.

GOP strategists think it may be a momentum killer.


  1. Polls are polls, not unerring prophecy. Straws in the wind– until the wind shifts. The weather on election day sometimes makes a diffference, deterring more supporters of one candidate than the other.
    And the considerations you call attention to in terms of the voter sample
    are not always easy to weed out, even when one is disinterested. One just has to listen to a candidate taking apart a poll favoring his rival.

  2. I notice in the last 3 PPP polls the ratio of Democrats to Republicans has tilted more to the Democrats, while the SurveyUSA poll sample has remained almost constant.

    I also note that the 2008 McCain - Obama voters tend to vote the same way for Governor.

    I am not sure either poll has the voter turnout pegged, but the SurveyUSA poll does seem to be a good indicator, given the makeup of the last 3 samples being almost identical, that there has been no real movement in the race. If anything McDonnell is pulling ahead by a bit, but within the margin. If you are interested in my analysis, I have the SUSA poll at and the PPP poll here

  3. [...] 14, Bolling leading Wagner by 13, and Cuccinelli leading Shannon by 11. The ink was barely dry on Bob Holsworth’s analysis of this poll versus the PPP one when Rasmussen released their poll showing McDonnell up by 9. [...]

Leave a Reply