Two polls on Virginia's statewide campaigns were released within the last 48 hours: one, by the Democrat leaning polling firm, Public Policy Polling, and one by SurveyUSA for Roanoke television station WDBJ-TV. It is interesting to note that the PPP poll has received exponential media coverage, lasting well into the second day after it was released. The SurveyUSA/WDBJ poll was released last night but is hardly causing a blip on the Mainstream Media's radar screen. The most likely explanation is that the SurveyUSA/WDBJ poll seems out of whack when compared to other polls. While many have the governor's race in a four to seven point range, in favor of Republican Bob McDonnell, the SurveyUSA/WDBJ poll shows him up by double digits, as it does his running mates Lt. Governor Bill Bolling, running for re-election, and Senator Ken Cuccinelli (R-37, Fairfax), running for attorney general. Still, that's never stopped the notoriously out of proportion Washington Post polls from getting top billing across the state. 

In what has gone practically unmentioned in the frenzy of the tightening of PPP's poll, however, is that it also shows Lt. Governor Bolling and Senator Cuccinelli leading by eight and nine points, respectively. So, we have a Mainstream Media cherry picking news even from the one poll on which it has focused. Not only that, but by its own admission, the PPP poll's "internals" do not show bad news for McDonnell. Among the fndings (see PPP Blog here):

52% of voters say they're very familiar with the thesis and McDonnell actually has a 55-41 {lead} 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.

(Does this mean conservative voters are more informed, or just that they don't watch MSNBC?)

In another blog post,PPP's Tom Jensen writes that Democrat candidate Creigh Deeds leads among voters who were undecided a month ago by 35-32 percent, and lists this as an advantage for him. But it's within the margin of error and not enough to close the gap.

Back to SurveyUSA. Here's a link to its methodology and complete statistical breakdown. It survey 1,000 Virginians, 886 of them registered to vote, and filtered its responses to the 631 of them determined to be likely voters this November.