Sponsored by:

vt.Buzz ~ a political blog

Political notes from Free Press staff writers Terri Hallenbeck, Sam Hemingway and Nancy Remsen



Poll by numbers

The last time that American Research Group Inc. of of Manchester, N.H. did a poll in Vermont, political strategists on both sides of the aisle pooh-poohed the results as way off base. Some of them are bound to be singing the same song again when they see ARG"s latest poll figures.

The new poll, done right after the primary was over on Sept. 13th and 14th, had independent Bernie Sanders ahead of Republican Richard Tarrant in the Senate race, but Tarrant just 15 peercentage points behind. Click here to see the whole poll.

In July, the last time ARG polled in Vermont, Sanders was said to be 21 percentage points ahead, a number that had Tarrant types smiling. Sanders folks back then all but said the poll was bogus.

"Our internal polling shows us up by 41 points," Sanders campaign manager Jeff Weaver said at the time. "I trust our polling more than this polling." Recently, the Sanders campaign has made a point of saying its internal polling was showing that Tarrant's negative ads were costing him support. The ARG poll, if it is to be believed, says otherwise.

In the House race, ARG had Democrat Peter Welch with 48 percent and Republican Martha Rainville with 45 percent. That's an improvement for Welch from the July ARG poll, where he trailed Rainville by a single percentage point, but it's not welcome news for Peter and Pepper the dog. They'd rather believe a recent Democratic Congressional Campaign Committee internal poll that has them up 10 points over Rainville.

Finally, the ARG poll had Gov. Jim Douglas whupping Democrat Scudder Parker by 27 percentage points, instead of the measly 11-point spread Douglas enjoyed in the July ARG survey. Back then, Team Douglas was all over the ARG poll, saying it was badly flawed. It's unlikely they will feel that way about this poll.

Of course, the only poll that really matters is the one that takes place on Election Day, Nov. 7. In the meantime, we're all ears on what you think about this ARG stuff.

-- Sam Hemingway

What is known about this polling organization? Do they have political leanings? An agenda? Do they have any company connections to any candidate?

The Bernie/Tarrant poll results indeed seem a little weird. I mean, the fact that Tarrant can be at a crowded event and not have a single person approach him is a very unscientific but very telling indicator.
Haik Bedrosian accurately predicted this poll in an article about Astroturfing. You can read Haik's comments Here
Hilarious how Sanders' internal poll numbers are taken as gospel, but these are immediately framed as part of some big conspiracy with zero evidence.
The American Research Group is located in New Hampshire. It has a very spotty history of inaccurate polls. They will not identify the source of their funding for these polls.

Besides Benie\\\'s polls, there are other independent polls that disagree with American Research Group. For example, the respected Rasmussen poll gives different results.
It's impossible to say whose polls are correct. When considering Bernie's "leaked" internal polls, however, their contention that Tarrant hasn't gained an inch since the beginning of the year is one that no rational person (or any other poll) would agree is likely.

BTW, Rasmussen also disagrees with Bernie's poll by quite a bit.
At least Bernie is clear that he is the source of the poll. The American Research Group won\'t identify the source of the funding for this so-called independent poll.

Who paid a New Hampshire polling organization to do a poll on the Vermont election, one day after the primary ended? My answer is Tarrant.
Remember, ten days ago, Haik accurately predictd that this polling company would publish this type of poll.
Rasmussen said that Tarrant gained ground as well. Guess they're funded by the GOP. Keep patting yourself on the back anonymously, Haik.
I guess I'd put more stock in the internal polling but hopefully they'll ingore them. Lest we forget Doug Racine's tepid campaign where the polling showed him leading and we ended up with Douglas. What a nightmare.
The ARG poll has absolutely no credibility. ARG will not say who funded the poll. They will not provide any details about their polling methods.

Three days before the 2004 Presidential election, ARG predicted Bush would win NH by 5 points. It didn't happen.

The Giuliani people also doubt these polls. ARG is able to produce polls that show McCain with an insurmountable lead.

ARG has absolutely no credibility.
We could all complain. We could all moan and groan about this poll. You're right, it's one poll. It's possible it's right... it's possible it isn't.

Of course Jeff Weaver will leak his own poll. My bet is they are a bit nervous. I can't think of a time they (Bernie and Jeff Weaver) did 9 commercials, ever.

We'll just have to wait and see if this poll is factually close, or not. We'll just wait until the next poll... what else is there to do?

In the mean time... watch for Bernie's man: Jeff Weaver to release another internal poll showing them up by 80%... cause the ads are "SOOO NEGATIVE". Anyone read Bernie's blog lately? Eesh, welcome to Vermont.
ARG is funded by media outlets that subscribe for early access to their results, as described on their website. It's very simple. No one has "refused to reveal the results of their funding." The poll was one simple question, asked of 600 people. Calling it fixed or fake smacks of desperation.
You can't avoid two facts.

1. Haik Bedrosian accurately predicted this poll on his blog ten days ago.

2. This polling organization has a terrible track record.

3. Giuliani's supporters also criticize ARG.

Okay, I know that was really three facts, but ARG would probably count them as two.
"This polling organization has a terrible track record."

According to who? Certainly not all of the major media outlets, who subscribe to their results. Any idiot could have predicted Tarrant would continue to gain ground, Rasmussen said the same thing.
Well, Haik predicted the actual polling organization and the timing of the poll.

With regard to the terrible track record, you can't miss the fact that ARG was just plain wrong about New Hampshire's vote in the 2004 Presidential election.

If ARG is indeed legitimate, why don't they reveal who is funding their polls? Almost all other polls tell you if a media organization (like Washington Post or CBS) has paid for a poll.
Predicting the polling organization and the timing of the poll isn't exactly magic. ARG has done a number of polls on the race; chances were excellent they would do another before the race was over. Directly after the primary, when you actually have the official candidates, is a logical time to do a poll. Simply because Haik realized that after the primary a poll would probably take place, and the polling organization that had been releasing numbers for several months would probably conduct that poll, does not mean he's uncovered some type of conspiracy.
"why don't they reveal who is funding their polls?"

They do "reveal" that information - it's anyone who subscribed to the results. That's their business model, it's described on their site. If you're that curious about who the subscribers are, try emailing the local media outlets and asking them if they subscribed.
Absolutely fascinating. This poll was supposed to be independent. Yet, there is a very passionate defense of ARG from Tarrants supporters. If Tarrants campaign had nothing to do with this poll, why are they so passionate in their defense of ARGs tactics?
How can you claim that Tarrant supporters are passionate in the defense of ARG. That would mean that the 40% would all know that the poll was fixed.

Who's Haik? People keep mentioning him. Is he a member of the media?
I think it's interesting that people seem to take Sanders' "internal poll" numbers as gospel, especially when you KNOW who paid for it. If you're so sure Tarrant's camp is paying for the ARG poll, and that's why the results are suspect, why doesn't the Sanders' camp's numbers come under the same scrutiny?

And Haik's a blogger who seems committed to uncovering some vast conspiracy behind the Tarrant campaign, particularly when things happen that paint Sanders in a less-than-stellar light.
Interestingly, the Giuliani campaign is also highly critical of ARG. To read how Giulianis people view ARG, click Here

Keep in mind, ARG first published a Tarrant poll at the same time as a McCain poll.
"there is a very passionate defense of ARG"

Every criticism of this poll has a simple, obvious response. Providing that response is not a "passionate defense" of anything.
I don't find Giuliani's response particularly interesting as much as completely predictable. When you're not doing as well as you had hoped in a poll, of course you'll take issue with it. There's nothing else you really can do. If, somehow, contrary to all the other polls out there, the ARG showed Giuliani ahead of McCain, do you really think he'd take exception to it? No, of course not - he'd trumpet it as a huge success, despite the fact it didn't match up with other polls. He's wouldn't be honest and come out and say "I'm really not that good" - so, my question is, why should I believe him when he comes out and says he's really not that bad? I'd be willing to bet that ANY poll that came out that showed Sanders to not have as large a lead as he'd like, he'd immediately decry as "unfair," and try to discredit that poll as well. It's just the nature of politics.
Wow. You guys have an answer for everything.

*There is very consistent evidence that ARG produces polls with results that are inconsistent with other polls.

*They have a track record of favoring the front-running Republican.

*They have missed by as much as 20 percentage points.

On top of all of this, Haik Bedrosain accurately predicted that this polling company would have this result. It is obvious that ARG is part of Tarrants astroturfing strategy.
It's amazing that the Bernie Brigade, who are apparently so confident in their guy's chances for victory, are expending so much energy and using such weak "arguments" in trying to discredit a legitimate poll from a legitimate polling organization with 21 years' experience. Especially while propping up Bernie's internal polling, which makes the laughable statement that Tarrant has gained no ground since the beginning of the year.
It's time that American Research Group (ARG) is exposed. Here are the facts:

* In the 1996 NH Primary, ARG had Dole winning by 7 points. Buchanan actually won.

* In the 2000 NH Primary, ARG had Bush winning by two points. They missed by a whopping 20 points as McCain won New Hampshire.

* In the 2004 Presidential Election, they had Bush beating Kerry by 5 points in New Hampshire. Of course, Kerry won New Hampshire.

They consistently favor the republican who is supported by the New Hampshire Republican hierarchy. They are also consistently differnent than all other polls, and they are consistently wrong.
Amazing how disproven points are regurgitated again and again in the very same thread. This poll has really thrown Bernie's folks into a frenzy.
Polls change daily. People change there minds.
What's with the "anonymous" thing? It's impossible to follow these comments if you're all chickens cackling. BTW, thanks, Simms.
For a moment, consider the components (Dem/Rep/Ind) of the poll are legit. Does anyone think that the ratio between Dem/Rep/Ind that go to the polls will be 29/28/43? The number of Dems has to be more than a slight advantage, so the poll favors the Republicans.
"In the 2004 Presidential Election, they had Bush beating Kerry by 5 points in New Hampshire. Of course, Kerry won New Hampshire. "

Of course, Kerry won New Hampshire with 340,511 votes, compared to Bush's 331,237 - a 1% difference. Not exactly the huge difference you'd have expected from the comment.

And, of course, in the 1996 NH primary, Buchanan won by... one percentage point. Yeah, it's unfortunate the ARG polls (or any other polls, for that matter) aren't always going to be 100% accurate - heck, then we wouldn't even have to go out and vote.
Here are some more facts, taken from this site: http://www.realclearpolitics.com/

"American Research Group
Failed to Project Winner: 50% | Average Error = 2.0

ARG got a bit unlucky this year. They called for a 1-point Bush win in NH and the result was Kerry +1. They also projected a 1-point Kerry win in New Mexico and the result was Bush +1. The big miss, however, came in Florida where ARG's last poll had Kerry up two. ARG did offset these misses by nailing tight outcomes in IA and WI giving them an excellent score in overall average error."

Hmm.. ARG had Kerry up 2 in Florida, and up 1 in New Mexico - yet you don't see anyone bringing those facts to the table.
ARGs track record of being correct in only 50% of the elections is unimpressive. You could do just as well by flipping a coin.
"ARGs track record of being correct in only 50% of the elections is unimpressive."

Nothing like making something up and just throwing it out there. Pathetic.
I suppose I should have put this in some context:

"CNN/USA Today/Gallup
Failed to Project Winner: 67% | Average Error = 5.8 "

"FOX News/Opinion Dynamics
Failed to Project Winner: 50% | Average Error = 4.5"

"Strategic Vision
Failed to Project Winner: 44% | Average Error = 2.4 "

In 2004, ARG predicted a 48-48 split for Bush and Kerry; CBS predicted a 49-46 win for Bush; the actual numbers were about 51-48. Interestingly, the CNN polls were the closest to ARG's, with a 49-49 tie.

If someone wants to make the argument that ARG doesn't have the best polls, go ahead. But to continue to try to argue that it's part of some huge Republican conspiracy simply because the numbers aren't what you'd like to believe, well, that's ignorance at its worst.
Interestingly, the Sanders campaign just released information from a poll that says Sanders has a 66-22 lead over Tarrant. This poll will probably get its own entry soon - but I think it will be interesting to see if Sanders supporters take exception to this poll as well. After all, it has clear political leanings (the Garin Hart Yang Research Group, which conducted the poll, states on its website "Garin-Hart-Yang has a strong record in helping Democratic candidates win in the most difficult circumstances"). Clearly not an impartial source by any means. If Sanders supporters truly have a problem with ARG because of the apparently unknown funding of the poll, anything less than a problem with the Garin-Hart-Yang Research Group amounts to outright hypocrisy. But then, I wouldn't be suprised; "I do not accept corporate PAC money... except from the Fanjuls."
I have no problem with ARG, it puts Sanders in a huge lead.
Post a Comment

<< Home

Recent Posts

Recent Comments


June 2006   July 2006   August 2006   September 2006   October 2006   November 2006   December 2006   January 2007   February 2007   March 2007   April 2007   May 2007   June 2007   July 2007   August 2007   September 2007   October 2007   November 2007   December 2007   January 2008   February 2008   March 2008   April 2008   May 2008   June 2008   July 2008   August 2008   September 2008   October 2008   November 2008   December 2008   January 2009   February 2009   March 2009   April 2009   May 2009   June 2009   July 2009   August 2009   September 2009   October 2009   November 2009   December 2009   January 2010   February 2010   March 2010   April 2010