Mason-Dixon poll: Allen leads Webb in Virginia by 16 points; 20% undecided

[DeepModem Mom]

WP/AP: Poll: Sen. Allen Leading in Virginia
By BOB LEWIS
The Associated Press
Sunday, July 30, 2006

RICHMOND, Va. -- Republican Sen. George Allen has a 16-point lead over Democratic challenger Jim Webb in the latest independent statewide poll, published Sunday, but a fifth of the electorate is still undecided.

The election is closely watched nationally as an off-year referendum on the embattled Bush presidency because Allen, one of Bush's most reliable Senate allies, is preparing a 2008 presidential bid. Last year, Allen voted in support of the White House more than 95 percent of the time.

Forty-eight percent backed Allen and 32 percent supported Webb in the Mason-Dixon Polling & Research Inc. survey of registered voters likely to vote in the Nov. 7 election.

However, 20 percent of the 625 respondents surveyed statewide by telephone July 25-27 said they had not decided between Allen, a former governor seeking a second Senate term, and Webb, a former Republican who was President Reagan's Navy secretary....

***

Bush's job-approval rating is low even in Virginia, which last voted Democratic in a presidential election in 1964, the poll found. Forty-three percent rated Bush's performance as good or excellent while 56 percent judged it fair or poor. One percent of the respondents were undecided....

http://www.washingtonpost.com/wp-dyn/content/article/2006/07/30/AR2006073000201.html

[David Dunham]

1. Game Over.

[NCarolinawoman]

3. No, the game is not over.

Burr was about that far behind of Erskin Bowles in NC during the summer of '04. It was lack of name recongition. Then the ads started coming out in the fall and Burr ended up winning. ( I will say that Burr is a real stinker and his tv ads and mailers did not depict the real Burr, who is owned by the drug companies and other corporations). But the North Carolina sheeple believed the ads and he did end up winning.

There is still hope here.

[Doctor_J]

9. A Repub coming from way down in a red state is MUCH

different than a Dem coming from way down. This one is over. Too bad - my view of the citizens of VA is pretty bad right now.

[Orangepeel]

4. an incumbant under 50% is in some trouble

I'm not saying that it will be easy, or that it is even likely. But, conventional wisdom is that because the incumbant is better known, undecides tend to break heavily for the challenger.

[Hippo_Tron]

10. October surprises happen more often than you think

Honestly, I think that it will take one of those for Webb to pull this off. But if one does happen, Webb has positioned himself perfectly to win this thing.

[OzarkDem]

2. Ignore Mason Dixon polling

They're not reputable. They've delivered GOP-slanted poll results to the Cleveland Plain Dealer in the last 2 presidential elections that were off by 7-12 points.

[Awsi Dooger]

5. MyDD.com doesn't agree with you

http://mydd.com/story/2006/7/30/19949/8905

"Mason-Dixon, which is considered by many to be the most accurate pollster around, released new data on the Virginia Senate race between freshman Republican George Allen and Democratic challenger Jim Webb, and although the survey shows that Allen maintains healthy lead, he is nevertheless still under 50 percent at this point. Bob Lewis has the details for the AP."

I just looked at my file from 2004 and Mason Dixon nailed the 2004 Ohio result. They had Bush up by 1 in the next to last sample from October 15-18, then ahead by 2 in the final poll from October 27-29: http://www.realclearpolitics.com/Presidential_04/oh_polls.html

In 2000, it's very likely Mason Dixon, like other polling firms, had Bush leading by much more than the +4 in Ohio. The Gore team itself botched that one, thinking the state wasn't in play.

I will agree with you this is not an accurate measurement of the Virginia race. I think it's 10 points if they went to the polls today. I posted that when the margin was 5 in one poll, and I'll say the same thing when another sample says 16.

I like to look at history as a guide and also to weigh logical variables, to dispute poll numbers instead of accepting them at face value. I guess it's a natural spinoff of my sports background in Las Vegas. After all, if you think every pointspread is an accurate measure of the difference between the two teams, then why bet since every result should threaten to push the number? Admittedly, same as in sports, I overdo it sometimes in politics and try to alter a poll number that ends up being accurate.:)

[OzarkDem]

21. Mason Dixon didn't "nail" the Ohio result

they helped "create" the Ohio result.

In 2000, they used bogus polls to demoralize the Gore campaign and convinced them to pull their campaign out of Ohio. The consensus w/in the Gore campaign at the time was that the Pee Dee's Mason Dixon poll was several points off their own internal polling. They knew the Mason Dixon numbers were bogus, but their opinion was "why stay here and deal with the pervasive lies and bogus poll results in Ohio; lets go somewhere else where the pollsters and news media aren't playing dirty tricks." We begged them not to fall for it, but they didn't listen...

[TheVirginian]

22. That sounds incredibly flaky.

I don't suppose you would have anything to back that up? Especially your claim that the Gore campaign would pull his campaign out of one of the most populous and most contentious states in the country?

[Awsi Dooger]

25. It's common knowledge Gore pulled out of Ohio early

He pulled his ads as much as six weeks early, primarily to focus on Florida. It also no doubt had impact on West Virginia, since those are bordering states and the local TV stations bleed into the other state. I'm not saying Gore could have won either state, but it was no doubt a strategic blunder to abandon one that was hardly impossible. Many local Ohio election strategists were miffed and outraged since they thought they had a great union ground game in that state. In 2000 our GOTV was more sophisticated than the GOP's.

Here, if you want a link, there are hundreds: http://www.csmonitor.com/2004/0917/p01s04-uspo.html

"Four years ago, with just a few weeks left in the presidential contest, Al Gore's campaign pulled out of Ohio, a move many Democrats came to regard as a tactical error after George W. Bush won the state by a surprisingly slim margin of four percent."

Otherwise, IMO it's flaky or worse to accept poll numbers at face value. It's funny how they can err cycle after cycle yet that's forgotten and the new numbers accepted as gospel. For example, in Georgia the Democrat is historically oversampled by considerable margin. That was true in Zell Miller's gov race in '94, Cleland's senate race in '96, Gore's campaign in '00. In other words, long before Diebold. I'm not saying I expected Barnes to lose in 2002, but I damn sure expected that race to be much closer than expected, and for Cleland to lose, since the polling had tightened considerably in that race to basically even, and the trend indicated the numbers would no doubt still be too favorable to the Democrat.

Alaska is another state we are routinely overstated. That was the case in the 2002 gov race between Frank Murkowski and Fran Ulmer polled very close then it turned into a 15 point rout. So in 2004 I always doubted Tony Knowles, despite his popularity, truly led Lisa Murkowski. He was beaten by several points and the exit polling indicated the early deciders favored Murkowski overwhelmingly, by 16%, then Knowles won a huge percentage of the late deciders. I think the polling was bullshit all along. That state is known as particularly difficult to poll.

In New York we are understated. Just look at the Schumer/D'Amato race in '98, the late polls in Hillary/Lazio in 2000, or the recent presidential campaigns.

This year I refuse to believe Brown is leading DeWine by 8 points, or anything close to that, in Ohio. It makes zero sense given the natural partisanship of that state, the political philosophies of the two men and how that fits with the typical preference of the voters in Ohio, plus the incumbency status and DeWine's poor but not abysmal approval rating. None of it adds up to an incumbent trailing by 8 points. You realize how rare that it, a senatorial incumbent blasted by big margin even though the partisanship of the state is somewhat in favor of his party? I know all about the Republican scandals in that state, but I'll trust my instincts and handicapping above polls.

[OzarkDem]

28. I know, I was working on the campaign

The bogus poll and subsequent barrage of news articles stating "Gore had no chance" , using the bogus poll numbers as "proof" Gore was "far behind" and "couldn't catch up" led to the pull out.

It takes more than one poll to give an accurate picture. Those that vary too far from the norm are usually not reliable - Mason Dixon often falls into that category.

I know it disappoints you terribly, but the reputable polls have been consistently showing again and again and again and again that Brown is several points ahead of DeWine.

Get over it, Brown stands an excellent chance of winning and continues to pull ahead in spite of the millions DeWine is spending on dirty campaigning.

[Awsi Dooger]

29. I'll go along with Brown several points ahead

But not anything close to 8 or more. I think the Rasmussen poll of Brown +2 is a logical barometer of the race right now.

Look, I want to win as much as anyone but I've been thru two cycles on DU in regard to senate races. How we were going to win in red states, one after another. It's exceptionally difficult, especially against an incumbent.

I'm very impressed with the race Sherrod Brown has run. Better than I expected. This cycle favors us and the continued Republican ethics probems in Ohio should give Brown an added boost. That email about Strickland being gay is just another example of how crooked and desperate they are.

But I'm still worried about two things:

* Is Brown too liberal to defeat an incumbent senator in Ohio?

* Will the national security theme again provide a small benefit to Republican candidates nationwide in federal races, specifically senate?

[OzarkDem]

38. Those are GOP talking points

No and no and Brown is ahead by 8 points in two reputable polls and has had a similar margin for a few months now.

Why do you keep repeating GOP talking points and promoting the GOP "Brown isn't that far ahead" meme? You're starting to sound like a GOP troll. If so, you're wasting your time.

By all means, don't vote for him, he doesn't need GOP votes.

[Awsi Dooger]

44. Very typical around here, label and dismiss if it's a counter argument

I'm not disputing Brown has been ahead by 8 points or more in several polls. A clapping seal can subtract one number from another.

I question the accuracy of those polls, based on historical knowledge of the statewide voting trends. Just like in 2004 I severely questioned the poll showing Kerry ahead by 8 in Ohio. Or the polls with Kerry barely ahead in Hawaii and New Jersey.

Sorry, but I'm not a slave to poll numbers. If the election were held today, I'd gladly bet you and anyone else on this board that the margin wouldn't be anywhere near 8 points. And I'd have much the best of it. Actually I found out the hard way that man-to-man wagers on message boards don't pay off, since you never get the money.

Start naming senatorial incumbents who lost by 8 points or more. Especially in states favoring the partisanship of their side. I think Ohio will be a 50/50 party ID state this year, not 40-35 GOP like in 2004, but that's still a huge leap from an 8 point advantage. We beat Hutchison by 8 points in Arkansas in 2002. So it can be done. But we had edges there we don't have this time, namely our nominee being the son of a popular former governor and senator.

Admittedly, the statewide voting tendency is not 100% applicable to senate races. In gov races, people tend to vote for the person. In senate races they vote for a combo of the person and the party. In presidential races they vote for the party. That is extreme generalization but the basic truth is there and you'll come out way ahead if you understand it. It's the reason a Strickland can have a huge lead in Ohio for governor, but a Democrat running for president has no chance to have that type of lead in that state. In a senate race it's feasible, but extremely unlikely, especially against an incumbent. And, IMO, since the challenger is apparently more liberal than the state prefers for its statewide office holders. We have GOP scandals going for us in Ohio. I'd rather have a steady partisan drift in our favor, like Virginia and Colorado. There is nothing like that in Ohio and I'm not pretending otherwise.

I'm not a cheerleader around here. That is true. The cheerleaders routinely get it wrong. Yesterday I saw a wonderful +40-50 estimate for our House net. So I'm a troll, and that guy is a valuable DUer? Fine. BTW, every college football team will go undefeated this year. I'm also not a Diebold defeatist. They are the most pathetic of all. On election night you can count on me to be rooting in one race after another, all the threads and obscure races. Into the wee hours, long after others have gone to sleep, or started threads saying we were robbed in any race we lost by narrow margin. I'll have spent that day driving a GOTV van, just like several recent cycles. The difference is this year favors us, so we'll come out ahead especially in gov races and I have high hopes for the House. If I had to put a number on us retaking the senate, it would be less than 5%.

And one thing I almost forgot, the Republicans potentially benefitting from irrational national security fear in those senate races is hardly an empty talking point. The Strategic Vision polls shout that in one obscure question, "do you expect a homeland terrorist strike within 6 months?" By all probability rights the answer should be overwhelming No. But that result is between 71 and 86% Yes in every state polled to date. The No is always 10-15% with the rest undecided. If we don't think that is potentially significant in the close senate races, we're kidding ourselves.

[TheVirginian]

45. That is certainly uncalled for.

Mr. Dodger is perhaps one of the most knowledgable posters when it comes to pre-election polls, odds, and outcomes. I've payed attention to every polling thread that has been made on this board for the last six months, and his posts most consistently provide the best insight as well as the most objective and rational viewpoint of how to analyze each individual race. Just because he has an opinion that you don't like doesn't make him a GOP troll. That's an unfair extremist accusation.

Dodger raises good points: Ohio voted for George Bush twice in the last two elections (disregarding voting fraud on whatever level you believe it exists), has two Republican Senators, and a Republican Governor. It is most definately a swing state, but there's no denying that the state leans Republican. While everybody accepts the toxic environment it is for an incumbent Republican in Ohio, there are still factors that are very important to consider:

1) The majority of Ohio voters lean Republican;

2) Mike DeWine is a moderate Republican, running against a liberal Democrat;

3) Mike DeWine is an incumbent who has committed no major fault that precipitates being kicked out of office (like Conrad Burns or Bob Taft);

4) Mike DeWine and Sherrod Brown are even-up in terms of fundraising and in-state and national support.

So there are very good reasons to question whether or not Brown is truly up by as much as some polls suggest.

Secondly, its very important to remember that a pre-election poll is not an exact science, and it represents a snapshot of the day the poll was conducted, not a narrative for the entire race. Mr. Dodger, myself, and some others have taken to looking at all of the polls that are released over a multi-month period and extrapolating from that a trend of how the race is going, usually in the results of a margin (i.e., Brown up on DeWine by 2 points, Allen up on Webb by 10 points, etc.). So even if two seperate polls (in this case, the Columbus Dispatch and Zogby) give Brown an 8 point lead, there are other polls to consider, and there are other "snapshots" of time to consider, which is why we reach a different conclusion.

Finally, you are at your liberty to interpret polling however way you like. If you see a poll that has Brown up by 8 points in late July, then you can go ahead and write a letter to Mr. Brown offering your services as a campaign manager for his 2012 re-election campaign, if that's how sure you are he'll win. Other people have different ways of looking at polling, and different reasons for doing so. As Mr. Dodger has stated time and time again, he's been on this board for both the 2002 and 2004 elections, and is approaching these elections from a smarter perspective that is less likely to lead to disappointment and wild claims of voting fraud. If you disagree with him, that's perfectly fine, too, but there is absolutely no need to turn your guns on him, or anybody else, for seeing something different in the numbers.

I venture to say that there are a small handful of posters on this board who are committed to a rational, objective, and thorough analysis of the lead-up to November, with an emphasis on pre-election polling. Without a doubt, Mr. Dodger is one of them. And whatever your beliefs and principles are, I know without a doubt that rationality and objectivity are in far too short supply around here. How sad it would be to shut out one of the most mature voices on this board and baselessly accuse him of being a troll just because he has a more sophisticated way of analyzing polling data.

[Clark2008]

40. Well, they've never been accurate in Tennessee.

EVER.

Funny they should be called the Mason-Dixon poll and they can't call Southern states worth a shit.

[Awsi Dooger]

46. I've charted polls since '96 and Mason-Dixon is excellent

Yes, they botch some. Perhaps routinely Tennessee. I have those numbers on an elusive disk. And I remember a race in New England in maybe 2002 that they missed by 10 or 12 polls despite a poll sample late in October. But overall, I trust them as much as anyone. I have no idea how Mason-Dixon is ridiculed around here. Nor Rasmussen. It's apparently from posters who don't like the pre-election polls, more from an analysis of how the polls compared to actual results.

Here's a mathematical breakdown of the most accurate pollsters in 2004, based on the presidential battleground states: http://www.realclearpolitics.com/Commentary/blog_11_8_04_1018.html

1) Mason-Dixon
Failed to Project Winner: 6.2% | Average Error = 1.9

A final Minnesota poll showing a one-point Bush win is the only blemish on Mason-Dixon's otherwise perfect scorecard this year. Not only did Brad Coker project the correct winner in 15 out of the 16 battleground states we looked at, he did so with amazing accuracy. Three states were dead on the final number and the overall difference between Mason-Dixon's final polls and the actual election results was a minuscule 1.8 points. Furthermore, if you look down through the list of Mason-Dixon's projections it's impossible to detect any consistent leanings toward one candidate or another.

2) Rasmussen
Failed to Project Winner: 6.2% | Average Error = 2.1

Rasmussen's battleground state polling this year was extremely solid and a close second to Mason-Dixon. He was dead on in PA and just missed in IA. Average error was a very good 2.3 points, with NJ and AZ the only states where he wasn't within three points of the final spread. No partisan trends either way.

[TheVirginian]

47. Really?

Four polls were done the third week of July in Tennesee.

Zogby had Ford a point up on Corker.

Mason Dixon had Corker with a 12-point lead on Ford.

Rasmussen had Corker with a 13-point lead on Ford.

The University of Tennesee had Corker with a 12-point lead on Ford.

Which one of those seems out of place?

[TheVirginian]

8. Ignoring polling data is not sound advice.

[OzarkDem]

20. It is if its Mason Dixon

I know for a fact that they did rigged polling in the Ohio presidential races for the Plain Dealer in 2000 and 2004 by oversampling GOP voters.

They're a "for hire" pollster who will give you the results you want.

[TheVirginian]

23. "Know for a fact". How do you know for a fact?

Every poll can be analyzed as under-sampling and over-sampling demographics, because every poll uses different methodologies in determining its sample, and its impossible to tell what the most current demographics are.

Sticking your head in the sand is not an advisable way to win an election.

[OzarkDem]

27. Spoke w/ the person

who paid for the poll. He was feeling quite good about himself for having pulled a dirty trick.

Polls can be very helpful in campaigns if they're well done. Dishonest ones are used to destroy good campaigns and candidates. Given all the other dirty tricks that Republicans use to win races, why are you surprised they would pay to rig polls?

[Ken Burch]

6. Well, Webb could still pull it out...

But this does put the lie to the DLC myth that he was electable and his primary opponent wasn't.

It's hard to feature the other guy being in worse shape than this at this stage of the game.

[TheVirginian]

7. If Miller won the primary, this would be another Nevada.

Allen would be above 50% in every poll, and Miller would be lucky to break 40%. I agree with Mr. Dodger... Webb is 10-12 points down. That's not going to win it, but its a lot better than Miller could have done.

[Ken Burch]

11. If Miller WERE to do that much worse

It would be due as much to the determination of DLC types to make sure he did worse (like their political forbears made sure McGovern got creamed in '72)just to prove that a non-DLC Dem couldn't be competitive as it would be due to any inherent fault in Miller himself.

The point is, Webb's pathetic showing proves the hardball campaign to stop Miller was unjustified.

[TheVirginian]

13. I don't know where you're getting this from.

[Ken Burch]

14. Democratic party history and every account of that race.

The Webb campaign's central arguement was that Webb, and only Webb could win, that it was intolerable to even consider nominating Miller. It was extreme hardball from everything I've read about it. Now this poll pretty much discredits the whole rationale.

I damn well hope Webb can pull this out, but he has to make up to the Miller people and his campaign owes a hell of a lot of apologies.

[TheVirginian]

15. No, Webb's argument was that Webb had a better chance.

It was never that Webb would win or that Miller would lose, it was that Webb was more likely to win and Miller was more likely to lose, and frankly, I haven't seen anything that disproves that thus far. You can't prove that Miller would be doing better than Webb right now; I personally believe that he'd be doing a lot worse. At any rate, I think this animosity against Jim Webb on behalf of Harris Miller is misplaced.

[Ken Burch]

16. Sixteen points down at this stage ain't all that electable.

And if I were in VA I'd be working hard for Webb. I doubt, however, that if Miller had been nominated that Webb backers would be helping Miller that much. Witness how non-McGovern Dems treated the '72 nominee.
A choice that was proven incredibly stupid by history.

[TheVirginian]

17. Again, I don't know where you're getting this from.

What makes you think that Webb backers wouldn't support Miller? Especially if the prime reason for selecting Webb in the first place was to get rid of George Allen, what makes you think that they wouldn't put forward the same effort to reach the same goal?

[Ken Burch]

18. Because the right wing of the party is committed to proving

that ONLY right wing Dems like Webb and Warner, the kind of Dems who don't disagree with Republicans on anything but trivial side issues, can be elected.

[TheVirginian]

19. And you're only proof of this is from 34 years ago?

[sandnsea]

33. Webb is the netroots darling

Don't even pretend that he was the one chosen by the "right wing" of the party, that is just ludicrous. Webb is the one that all the blogs and grassroots got behind. Miller was the liberal Dem choice.

[Ken Burch]

35. The netroots were backing the more conservative candidate for the Ohio

Senate nomination. I've never regarded the bloggers as infallible.

[Guaranteed]

12. Midterms are all about turnout. nt

[Moloch]

24. This is very discouraging news :(

I wanted to volunteer for the Webb campaign, but all they've asked me to do is send them contributions (which I understand they badly need, but I am not in the position to donate a lot of money) I really hope we can pull this off in Va.

[formernaderite]

26. Look, for alot of people the perception is two republicans are

running. Allen wins. Just because Webb was the anti-war candidate, doesn't give him democratic credentials, and it won't gain him any repub traction....in the end, why would they change horses?

[sandnsea]

34. Then why did the blogs support Webb?

If he's just another Republican?

Where the hell is this shit coming from. :shrug:

[Ken Burch]

36. God knows what the webroots were thinking here.

And I hope some people can understand why some of us are suspicious of the idea of giving a Reagan/Bush appointee a Democratic Senate nomination.

[sandnsea]

37. To blame the "right wing" Democrats is just false

I wasn't in support of Hackett or Webb because I didn't see them as very progressive either. But the fact is that DUers and Kossacks and those who identify themselves as the "true liberals" or the "Democratic wing" or whatever the fuck, DID support them. So don't turn around now and dump Webb on the "right wing" of the Democratic Party, that's just not at all what happened.

[Ken Burch]

49. Hey, I did acknowledge the role of the webroots here.

It's just that this poll demonstrates that the main case for nominating Webb(the idea that he was inherently more electable)was bogus.

It's still a good idea to campaign hard for Webb, but those who ganged up on Miller and opposed him do have some apologies to make.

[Hippo_Tron]

30. Wait for the debates, Felix's stupidity will be shown loud and clear

George Felix Allen Jr. does a damn good job reading from a script but make no mistake, he's dumb as a post.

[TheVirginian]

32. They've already had their first debate.

[Hippo_Tron]

41. Really? that's news to me, how did it go?

[Catchawave]

42. The audio is on the campaign site:

http://www.webbforsenate.com/media/

It wasn't televised, even in Virginia, or on C-SPAN !

[Hippo_Tron]

43. What's the point if it wasn't televised?

[TheVirginian]

48. It was reported.

[LibDemAlways]

31. Allen has more skeletons in his closet than the average

cemetary. I hope to God Webb isn't squeamish about getting down and dirty and exposing the lying racist, abusive bastard for the creep that he is. Trying to run a campaign on the issues against a lowlife like Allen would be an exercise in futility. Allen knows he can't run on his sorry record, so all he can do is use a bunch of swiftboat surrogates to attack Webb. Webb has to go after Allen in kind. I was very dismayed when Ryan Lizza's devastating piece about Allen came out in April in the New Republic and Webb's response was that it hit "below the belt." That's exactly where Allen needs to be kicked, and hard.

[politicasista]

39. I think Gov. Warner should have ran against Felix n/t

[rep the dems]

50. He probably could have won

but I don't think many people go from Governor to Senator. Plus if he wants to run for Governor again in 2009 (assuming he doesn't become President or Vice President in 2008)running for senate wouldn't make much sense.