Google Ads


Site Stats

VA-05: Perriello Getting Crushed in Latest SUSA Poll

by: James L.

Thu Sep 02, 2010 at 12:19 PM EDT


SurveyUSA for WDBJ-TV Roanoke (8/31-9/1, likely voters,

Tom Perriello (D-inc): 35 (35)
Rob Hurt (R): 61 (58)
Jeff Clark (I): 2 (4)
Undecided: 2 (3)
(MoE: ±4%)

Brutal stuff. It's worth mentioning that last time, we mentioned some issues with the crosstabs, including Hurt's big support among younger voters. Those issues persist in this poll.

It's worth mentioning that Perriello fared more favorably in the most recent American Action Forum poll, trailing Hurt by 43-49 -- and that poll, conducted by a Republican firm, tested the top line match-up right after asking a question about the healthcare bill.

Could things really be this bad?

UPDATE: Dana has one more issue regarding the poll: a very optimistic voter turnout prediction.
James L. :: VA-05: Perriello Getting Crushed in Latest SUSA Poll
Tags: , , , , (All Tags)
Print Friendly View Send As Email

VA-05: Perriello Getting Crushed in Latest SUSA Poll | 40 comments
Yes
Things CAN be that bad and probably are.  I think SurveyUSA has a GOP house bias, but it's not all that strong.  I expect to see more and more of this sort of thing in house races as they get polled.

43 - Male - GOP/Libertarian - FL 22
No.
SurveyUSA has blown this race in the past many times...  I don't know why they keep making such errors, but the GOP bias overall makes Rasmussen look liberal.
SUSA Poll from mid-August '08 on this seat
Goode 64%, Perriello 30%.

23, male, Democrat, IA-02 (previously CA-26, PA-06, CA-06)
[ Parent ]
Again
For the millionth time, I offer my reservations for any poll of a rural House seat.  In practically every race this cycle, the polling of rural House seats has been pathetic.

That said, I think Perriello is down by 5 to 10 probably.

23, male, center-right cynical Republican, PA-7

[ Parent ]
Ryan is right
There have been some epic fails, particularly PA-12, regarding rural House seat polls.

That said - Critz was never down this much in any poll.  

[ Parent ]
Yeah, but two points
1). Perriello was probably still unknown when this poll was taken

2). It isn't 2008

That said - I'm convinced Perriello is not down by this much. He's probably down though, and for an incumbent congressman that isn't good.  

[ Parent ]
Perriello
was not the incumbent in that poll. I doubt his name recognition was great.

It's one thing to trail by double digits as a challenger. It's entirely another to trail by as much as a sitting member.

Even if the poll overstates Hurt's lead by 10 or 12 points, there is no way to spin it as anything but terrible news.

20, GOP, NH-02

[ Parent ]
good point ...
although the trajectory was going the other way back then in VA-5 ...

nevertheless, SUSA House polling in 2008 was a bit off in several races in 2008 (even in Oct./right before the election):

PA-11 Barletta 51, Kanjorski 45 (late October/November):
http://www.surveyusa.com/clien...

OH-15 Stivers 41, Kilroy 47 (late October):
http://www.surveyusa.com/clien...
End result in OH-15 was razor thin.

WA-8 Reichert 46, Burner 50 (October):
http://www.surveyusa.com/clien...

MN-3 Paulsen 43, Madia 46 (October):
http://www.surveyusa.com/clien...

MN-6 Bachman 44, Tinkelberg 47(October):
http://www.surveyusa.com/clien...

IL-10 Kirk 44, Seals 52 (October):
http://www.surveyusa.com/clien...

AL-2 Love 56, Bright 39 (August):
http://www.surveyusa.com/clien...

KS-2 Jenkins 43, Boyda 50 (August):
http://www.surveyusa.com/clien...

FL-21 Diaz-Balart 46, Martinez 48 (August):
http://www.surveyusa.com/clien...


[ Parent ]
Wow
One right out of that bunch.
[ Parent ]
Although they were all pretty close races in the end, and a lot of these were in the margin of error
Their big miss was definitely Bobby Bright's race.  
[ Parent ]
User LordMike: please present data on "overall" SUSA GOP bias
It's one thing to cite and question individual polls.

Before making an accusation of "overall" bias in polling, it's important to do statistical analysis.

[ Parent ]
Look above, silver spring offered quite a bit of good data on that point. I will say I don't trust either SUSA VA-05 poll......
As I said when their last VA-05 poll was released, it doesn't pass the laugh test that a scandal-free incumbent would be down by this big a margin before Labor Day, no matter the size of the wave.  The only precedent for something like this was incumbent Democrat Don Johnson's shocking 65-34 loss in Georgia in 1994.  SUSA is saying Perriello is looking at the same thing.  But the fact that a GOP-favoring independent group polled this same race between the two SUSA polls and gave Hurt a 49-43 lead, a mere 6 points, further makes me doubt SUSA.

No question Perriello is down.  But he's not down by THAT much.

43, male, Indian-American, Democrat, VA-10

[ Parent ]
Actually, DC, I think tietack was talking about the accusations of right wing bias
Which I think is an unfair accusation against SUSA, especially if you look at what sliver spring came up with. They were wrong in favor of the Democract six out of eight times.

But your point about Don Johnson's loss rings absolutely true. Perrillo is not behind by this much.  

[ Parent ]
Thanks, I misunderstood tietack on that one. Regarding SUSA, I'm pretty sure owner/founder Jay Leve is a DEMOCRAT......
Seriously, I recall reading that somewhere years ago when SUSA first broke out into the scene.

FWIW, SUSA's methodology is, this close to an election, to use a likely voter screen that I don't know the metrics for, but from what I vaguely recall is a pretty reasonable and conventional screen even if not always accurate.  But the kicker with SUSA is they do NO WEIGHTING, meaning they let the results fall where they may, no matter the demographics.  If their screen turns up an electorate that is only 10% black when 20% of the registered voters are black, or that is 28% Democratic when Democrats have never been lower than 35% in a given state or district, they just lay that out there and let readers make what they will of it.

43, male, Indian-American, Democrat, VA-10

[ Parent ]
Appreciate the understanding
I was certainly less than 100% clear in my previous note. User LordMike said this of SUSA:
the GOP bias overall makes Rasmussen look liberal.

So while individual polls certainly can raise questions, it is important to provide a complete statistical analysis before making an accusation of GOP bias overall, especially in comparison to Rasmussen.

Backing up such a statement would require a statistical analysis of --all-- polls of both SUSA and Rasmussen over a given time period. Only with such data in hand can such a comparison be made.

Only then can user LordMike's assertion be verified.

[ Parent ]
And I'll just say it...
Given what Silver Spring provided us and some of the other evidence, SUSA has not shown a right wing bias in the past. So I don't think you can say that's true with this poll (as opposed to it being a bad poll).  
[ Parent ]
I think the more appropriate question is one of methodology
Does SUSA have a methodology problem with House races? Anecdotal evidence suggests that all of the major pollsters do.

If nothing else, this poll (and the previous one) of VA-05 suggests that SUSA is consistent with what they do. If the election proves SUSA wrong, I would hope that they will reassess their methodology.

I suggest a possible exception to the methodology problem associated with major national pollsters: a few pollsters who specialize in 1 or 2 states. Some seem to be more reliable w/r/t individual House races in their states. Siena, Elway, Hibbits, and Field come to mind.

But this is all conjecture, based on selective observations. I present no data here.

[ Parent ]
No bias
But there polls are leaning Republican this year without a doubt. Maybe now, as in 2008, they are picking up a bigger wave than is really there. In other words the "bias" is with the party advancing.
[ Parent ]
Even Hostettler managed 39% in '06
And Periello is a much, much better Congressman and campaigner than Hos was, albeit in an unfriendlier district for his party. I think he'll lose, but he'll be in at least the mid-40's.

20, CD MA-03/NH-01/MA-08
[ Parent ]
Nate Silver did have SUSA as one of the best pollsters in 2009
And Kos did look at them pretty seriously to replace R2K.

Not that this poll isn't off. But I don't by the "right wing bias" argument either.  

[ Parent ]
just for the record ..
I still think SUSA is one of the best pollsters, BUT they just happened to be wrong on some of these particular races (PA-11 was just dead wrong, as that poll was done a day or so before the election; some of the other ones were prob. within margin of error, and can't really judge the pollster on ones done in early Oct. or earlier ..) -- which tells us that even the best pollsters can sometimes be way off ...

another aspect to this is the timing.  August & Sept. polls can often be very different from October & there are always outliers -- here's a good example from SUSA:

North Carolina: McCain 58, Obama 38 (September 2008 SUSA poll):

http://abclocal.go.com/wtvd/st...

[ Parent ]
Agree 100 percent

[ Parent ]
I think you misundserstand my statement...
I wasn't implying that they were skewing anything or questioning their integrity in any way, only that their GOP house effect has been very strong this cycle (and this cycle only).  Perhaps it is due to their automated methodology... I don't know.  But, most of their poll that they've put out has shown a greater lead for GOP candidates than other polls... often significantly greater leads.

They also have a history of being wrong in this district, as has been pointed out below.  This poll should be taken with a grain of salt until confirmed otherwise.

[ Parent ]
Once again, I request data
to back up the following assertion

their GOP house effect has been very strong this cycle (and this cycle only)

If you want to limit the data to SUSA polls this cycle, that makes sense -- but --

it's important to include ALL SUSA polls from this cycle in such an analysis.

Only then can your assertion of a GOP house effect be verified.

As Nate has them as 538's #1 pollster (of more than 40 races), they've earned the benefit of the doubt.

[ Parent ]
WA and CA-Sen
Heinrich.
[ Parent ]
Yes and No
It's probably not this bad, but no one knows for sure. One thing: Perriello needs to get an internal out there pronto that disputes this poll, or he's going to start having trouble raising any more cash.  
Can you prove it that he can't raise any cash? In fact you're probably wrong because...
...SUSA's last poll was just as bad, and still Perriello didn't release any private polling.  So I imagine the money kept rolling in as needed.

43, male, Indian-American, Democrat, VA-10
[ Parent ]
He did reply to the poll....
Don't know if he released private poll numbers, but he did call out the SUSA poll as being inconsistent with what he's seeing.
[ Parent ]
To Help Perriello
Give a max donation to Clark. More bang for your buck.
[ Parent ]
Nope, I don't play that game. In fact I gave $500 to Perriello, an amount I've given only...
...a couple times before in a single check, the first time to Kumar Barve's short-lived Congressional bid in MD-08 in 2002, and the second time to Jim Webb in 2006.  Of course I gave Obama over $1200 in 2007-08 and $600 to Ashwin Madia in 2008, but those were split up in a bunch of contributions over time.

I won't give to a right-winger and/or a Republican.

43, male, Indian-American, Democrat, VA-10

[ Parent ]
Just Saying
That even if this poll is off. Perriello's best path to victory is division on the Republican side.  
[ Parent ]
Things are bad...
I am not sure things are THAT bad, but they are bad for sure.  

I think we might be passing the point of no return in regards of the House.  Without any markable improvement, anything below 35 in terms of Republican net gains is nearly impossible.  

23, male, center-right cynical Republican, PA-7

Maybe, but...
See conspiracy's point in the thread about Sabato in the Daily Digest below. I think the Dems money advantage might start making a difference. Also, I do think the Gallup poll will shift next week, give the improvements in the President's numbers this week.

I give Dems a month to start to turn things around before I start predicting apocalypse.  

[ Parent ]
Yes

I find this to be too extreme
This is a huge gap, I could buy 10 points, but not 26.

24, male, African-American, CA-24, Democrat. Chair of the SSP Black Caucus.
A joke
I never thought any outfit could make Rasmussen look good, but here it is.
Larger Question
Is Perrriello past the point of no return? His position is a lot different as an incumbent than a challenger, and the underfunded Clark's vote will continue to deteriorate as Election Day approaches.

My instincts tell me that this poll could be way off and Perriello is still in Rick Santorum in 2006 territory.

Thoughts from VA...
There is no way this poll will be proven right by November but Hurt is definitely up by somewhere between 5-10 points let's be honest. Perriello has no major scandals that put him at the ire of many voters but he is in a very conservative gerrymandered district and barely won with the wind at his back, hence 5-10 points is about right. Full disclaimer in that I live in VA but losing him would be a major dagger. He has great convictions and doesn't shy away from what he believes in. He is even willing to go to Tea Party events and defend himself. I can't see Robert Hurt going anywhere where he won't get any softball questions. He's had more town halls than any other Congressional member out there. He is young and would have a political future even if he loses. I could see him running for Governor in 2013 (especially if he gets to run against Ken Cuccinelli) as I think that district will be even more gerrymandered next time where a rematch would be tough as they will dilute Charlottesville even more.  

28, Male, Democrat VA-08  
Look at the internals in this poll, they're screwy too
http://www.surveyusa.com/clien...

The district is 25% African-American -- SUSA shows only 11% of the voters being black. (And, in the converse, they say 86% of voters will be white, in a district where only 70% of the population is).

They also show 21% of black voters and 18% of self-described "liberals" voting for Hurt -- completely unrealistic numbers.

And I'm sure that the poll failed to capture the sizable student population at UVA.

If you take these things into account and re-weight the sample to reflect a more realistic black portion of the electorate, you end up with a much smaller lead for the Republican and a very close race overall.

Odds are still that Perriello will have a hard time holding the seat, but these numbers are completely wacked out... expect a close race up to the very end.

It's not like this is an R+16 AL-02-esque district, it's R+5
You could read that as making the poll seem more suspicious, but you could also read it as a very bad sign for Democrats that they're doing very poorly even in districts that aren't that bad for them.
VA-05: Perriello Getting Crushed in Latest SUSA Poll | 40 comments

Copyright 2003-2010 Swing State Project LLC

Primary Sponsor

You're not running for second place. Is your website? See why Campaign Engine is ranked #1 in software and support among Progressive-only Internet firms. http://www.mediamezcla.com/

Menu

Make a New Account

Username:

Password:



Forget your username or password?

About the Site

Publisher: DavidNYC

Senior Editor: James L.

Senior Editor: Crisitunity

Contributing Editor: Jeffmd

Search the SSP archives:

Sign up to receive a daily e-mail whenever the SSP is updated:

SSP Resources

Presidential Vote by CD

SSP Document Collection

Race Tracker Wiki

Dave's Redistricting App

Blogroll

Daily Kos Elections
DemConWatch
Five Thirty Eight
Governing
Hotline On Call
Political Wire
Pollster.com
Prop*50
Public Policy Polling
The Thicket
Powered by: SoapBlox