Pollster.com

Articles and Analysis

 

Zogby Battleground States (9/9-12)

Topics: PHome

Zogby releases new McCain vs Obama numbers in eleven states among likely voters from 9/9 through 9/12 using internet panels.

 

Comments
rationalvoice:

Obama up by 2 in North Carolina, but down 8 in Virginia? Once again, these Zogby Interactive polls can be pretty much ignored.

____________________

RaleighNC:

To be fair to Zogby, SurveyUSA has McCain up 20 in NC, with no movement in VA. I'm sure the truth lies somewhere in the middle.

____________________

slinky:

I agree, it's too lopsided for McCain in Virginia and Pennsylvania to possibly be accurate. Just an outlier poll.

What's zogby drinkin' or thinkin'?

____________________

Lardo:

Take Likely Voter polls with a grain of salt for three simple reasons!!!

1. These polls are based on Likey Voters not actual Registered Voters.. Why does this matter?

2. Democrats registar lists in battleground states have increased by significant margins and by far out weigh Republican numbers in most key states.

3. Because this election will be like no other in history!! The turnout will be so massive (thanks to Palin & McCains dirty campaign)these polls will look like they were calculated by Greenspan on Crack!!

____________________

Ben:

New Hamshire, North Carolina and Pennsylvania seem way off to me. Zogby is the only one that has had McCain ahead in NH. You'd have to go back to a Rasmussen poll in April to find McCain ahead in NH. The NC result seems the most implausible. Also, I think the spread in NV is too big although I believe McCain is probably ahead.

____________________

change:

well this goes against all polls out there including the most respected ones like rasmusoon, survey usa and quinnipiac. how much longer is mccain gonna be able to kip the lid on: 154 lobbyists running his campaign, lies about palin's record that are now coming out, - this is a sleazy campaign for sure. don't vote mcsame if you got any brains! -mk

p.s please visit mclobbyist.com

____________________

Jacob S:

Picking random numbers would be about as good as Zogby. I wouldn't read too much into these polls.

____________________

Robi:

I think we all agree that these numbers are about as trustworthy as stillow's "facts".

____________________

JFactor:

If those results were true Obama would be toast. However, internet panels? Come on.

____________________

Robi:
____________________

John:

I would imagine that the response rate on an internet panel is highly dependent (much more so than phone polling) on the euthasism of the canditate's supporters and hence why these polls taken just after the republican convention give a (overly) strong showing for McCain. Earlier, just after the democratic primaries finished and it was the Obama supporters who were highly motivated, they showed the opposite and a (far too) strong showing for Obama (they had him winning in Arizona!). It would also explain Barr's high numbers in previous Zogby Internet polls.
Although quite why zogby interactive has shown Obama winning in NC for the last 3 polls is slightly beyond me.

____________________

Cicero:

Why are these Zogby polls always so obviously rubbish? I'm from the UK, and one of our pollsters - YouGov - does internet-only polls and gets results that are very good. They were the only company to accurately call the London Mayor election, and have always done well when it come to UK election results. Is it just that Zogby's weighting is poor whereas YouGov's is better?

____________________

boomshak:

I'm not sure why anyone should take an internet poll seriously.

That being said and considering Obama usually does MUCH better in internet polling since his base spends more time online and tend to be more activist, I am shocked by how well John McCain does here (since according to Obama, his supporters don't know how to use the internet).

THAT is the real story of these polls.

On every single poll you see coming out, look at the internals. John McCain takes the lion's share of partisan, non-partisan and independnet votes by 6 to 10 points. Team Obama better hope that Democrats can maintain a HUGE lead in party affiliation (despite all evidence to the contrary) or they are done.

____________________

boomshak:

In defense of Zogby:

Zogby was the first to come out with McCain +5 over Obama. Within a few days, Gallup concurred.

____________________

macsuk:

Boomshak

Your an idiot

____________________

boomshak:

macsuk,

You obviously have a lot of anger. If my candidate had every conceivable advantage and was still getting beaten by a 72 year old man and a soccer mom, I'd be angry too.

____________________

LarryS:

Wow -- pollsters discovered the Internet exists and try to use it for their "scientific" polling -- I am not sure an accurate way of polling exists, quite frankly, in today's society. But one question Zogby -- do you ever look at your results and ay "does this make sense?" From the results you post, I doubt it.

____________________

boomshak:

An objective observer would stand back and look at the current political climate, the economic instability and worldwide challenges. They would look at the blatant favoritism given to one candidate (5:1 according to Rasmussen) by the MSM. They would take this all in and say Barack Obama cannot possibly lose.

And yet, here we are.

John McCain cements his lead in red state after red state while Barack Obama's once double-digit lead in blue states shrinks to within the margin of error (PA, NJ, MI, WA and MN now in play - inconceivable!).

Question: Since the conventions, in what blue state has Obama increased his lead? In what states has his lead shrunken dramatically.

It should be impossible, yet here we are.

Here we are.

____________________

thoughtful:

I don't know why these Zogby interactive polls are posted other than for curiosity value.

____________________

boomshak:

INTRADE UPDATE:

Have no idea if this means anything, but McCain is surging on Intrade this morning for some reason.

He was up by 1 at 6 am this morning. He is now up by 6. Is there some news that would be driving this? Who knows?

____________________

boomshak:

Many people may discount Intrade (as being about as reliable as a Zogby poll), but I have found that the people who trade there tend to hear news before the rest of us do.

Anyway, fun to talk about I suppose.

____________________

boomshak:

Oh, ok, this may be it from Rasmussen:

"The Rasmussen Reports daily Presidential Tracking Poll for Sunday shows John McCain reaching the 50% level of support for the first time since Barack Obama wrapped up the Democratic Presidential Nomination. McCain retains a three-point advantage for the third straight day, 50% to 47% (see recent daily results)."

3 days in a row +3 and now 50%. I suppose that is significant.

____________________

boomshak:

Ok, last post on this subject (I swear). Here's more from Rasmussen this morning. This is DEVASTATING for Obama. According to Rasmussen, he is losing the Enthusiasm Gap BADLY:

"John McCain's convention and his choice of Sarah Palin as a running mate have moved him from 18 points behind to a one-point lead over Barack Obama when voters are asked who is likely to win this year's presidential election, according to a new Rasmussen reports national telephone survey.

Thirty-five percent (35%) say they expect McCain to win, 34% expect an Obama victory, and 30% say the race is too close to call. (see crosstabs).

While McCain's 35% to 34% edge is statistically insignificant, it is a remarkable turnaround from a month ago when 46% expected an Obama victory versus 28% who thought the Republican would win (see trends).

The new survey also found that 45% of voters say the election will be exciting, up from 23% in mid-August. GOP voters are more excited about the election now than Democrats.

Fifty-three percent (53%) of McCain voters say it will be exciting while just 38% of Obama voters say the same. A month ago, just 20% of McCain voters thought the election would be exciting while 29% of Obama voters held that view."

Wow, just wow.

____________________

clesky:

"macsuk:
Boomshak

Your an idiot

Posted on September 14, 2008 7:46 AM"

A very well thought out and delivered point. You must be a speech writer. As the insults and name calling at those that do not believe in Obama increase, the number of blue states decreases...that could be part of the problem. In politics, the delivery of the message is more important than the message. Most Americans (see red states) are turned off by elitist attitudes, and do not enjoy being talked down to just because they decide not to drink the Kool-aid. Gore and Kerry learned this lesson, and if the current trend continues, so will Obama.

____________________

boomshak:

Bad news for Obama. Rasmussen has just announced today that he will adjust his party affiliation sampling from a 7.6% to a 5% advantage to Democrats (and he will update these weekly).

____________________

Tybo:

Boomshak , do you have a link?

THanks
T

____________________

slinky:

The regression line on this site has pulled ahead for McCain in Ohio. Everybody else could just stay home, because Ohio is going to decide this election.

____________________

Tybo:

obama raised 66 million in aug.

Good number.
He's now off the stated goal of 450 over 4 months (dnc and Obama combined) by over 129 million

____________________

ndirish11:

I am no Obama supporter by any means but these polls make no sense at all. McCain with a big lead in Penn? I thought this was supposed to be democratic pollster. Unless there are other polls that back this up, I would disregard them. If this were true McCain would win the election in a landslide.

The only polls that seem right to be me might be missouri and florida. Everything else is to lopsided to McCain. And shouldn't the internet polling favor Obama? I didn't think McCain's main voting based knew how to work a computer

____________________

ndirish11:

I am no Obama supporter by any means but these polls make no sense at all. McCain with a big lead in Penn? I thought this was supposed to be democratic pollster. Unless there are other polls that back this up, I would disregard them. If this were true McCain would win the election in a landslide.

The only polls that seem right to be me might be missouri and florida. Everything else is to lopsided to McCain. And shouldn't the internet polling favor Obama? I didn't think McCain's main voting based knew how to work a computer

____________________

djneedle83:

Magic 8 ball says.. Zogby sucks

The entrance to become a polling company in America is easy and so many of these new companies illustrate the inabilities to survey voters in an incredibly changing demographic in this country.

____________________

Tybo:

Thanks Boom!

____________________

boomshak:

Sure Tybo,

For those that care about such things, McCain has surged to 7 points ahead on Intrade this morning after being only a point ahead.

People are likely looking at the internals of that Rasmussen Poll and thinking Obama is in trouble.

P.S., I equate the MSM and Obama attacks on Sarah Palin to blowing on a fire.

If the fire is small (like a match), blowing on it puts it out. If it is big (like a forest), blowing on it makes it burn hotter.

____________________

thoughtful:

"For polling data released during the week of September 14-20, 2008, the targets are 38.7% Democratic, 33.6% Republican, and 27.7% unaffiliated. For the previous week, the targets were 39.7% Democrat, 32.1% Republican, and 28.2% unaffiliated" Rasmussen

50 - 47 McCain with Rasmussen weighting of prior August 31 this poll would be 48 -48!

____________________

boomshak:

Thoughtful,

Nice try, but if you read the article carefully, you will see that Rasmussen is NOT going to apply the new weightings until polling starting TODAY.

That means that since this poll is a 3 day moving average, the 50-47 numbers reflect 100% the old weightings. (11-13th)

If all three days had been weighted by the new numbers, McCain would be up by almost 5.

Again, nice try, but read the article more carefully next time :)

____________________

boomshak:

Well, we can throw out the new Newsweek Poll showing this thing tied. They are giving an 8 point sampling edge to Democrats.

Next.

____________________

John:

@Boomshak
Firstly, I am not sure we should throw any poll out unless there is a very good reason and secondly the Newsweek poll had a +6 party id margin among RVs, the +8 was among all adults. Rasmussen over the same period had 7.6% and it has now moved to 5.1% among LVs. Conventional wisdom (and in the past) suggests that the party id advantage should be greater among RVs than LVs.

____________________

slinky:

The justification for the Rasmussen shifted weighting is not clear to me.

What changed in party affiliation in the past 6 weeks, or several months for that matter, that motivated a shift?

As a scientist and statistician, it is clear to me that the fact that no pollster releases raw data makes it impossible to judge the veracity of any of this.

____________________

zotz:

slinky-
I have been wondering about that too. It is a funamental question and I haven't heard an adequate explanation for it. Are they just guessing? If not where is the data.

____________________

KipTin:

Robi refers to "stillow's facts" yet Robi never uses any of his own. Admit that you are biased for your candidate and quit denigrating other posters here who do not agree with you. It is really tiresome and RUDE.
----------

Yes, Rasmussen is a respected poll that has McCain at 50% today and still 3 points ahead of Obama (47%).
----------

____________________

cinnamonape:

Boomshak: " Bad news for Obama. Rasmussen has just announced today that he will adjust his party affiliation sampling from a 7.6% to a 5% advantage to Democrats (and he will update these weekly)."

Well that is interesting. I would think that Rasmussen would be more concerned about State-by-State registration figures rather than a "set number".

Also this "pre-setting" of party affiliation figures (and Rasmussen has been reducing Democratic numbers for some time vs. Republican) actually says more about why his POLL has shown a tightening race than perhaps what is actually happening.

One would think that even if he was attempting to model actual claimed affiliation that would have to come FROM the poll responses. So the adjustment would be announced at the time of the poll, not beforehand. But the problem with using that method is that it has no independent cross-check (i.e. is your sampling biased and selecting people who are unrepresentative of the actual affiliation).

____________________

slinky:

Yes, the weight on weighted variables should come from outside the quantities you're measuring. So, Rasmussen should release the data that provides the basis for affiliation weighting. He hasn't done that, ergo his polls are suspect.

Look, you know, I can cook numbers too and release them. Because it was once a respected poll or polling org. doesn't mean squat. In order for these observations to be believed, they must be validated and reproduced. You can't do that if they don't share data or methodology.

Anybody can say anything, but it needs to be checked out. No ones gonna convince me that this is the best we can do in America.

Why the hell don't the Democrats commission a poll and release all the raw data?

What possible harm?

____________________

KipTin:

FACTS from Rasmussen... Go to the following LINK:

"New Rasmussen Reports Party Weighting Targets: 38.7% Democrat 33.6% Republican--

However, from now through Election Day, we are making two adjustments to that process. First, due to potentially changes in the political environment, we will now base our targets upon survey interviews conducted over the preceding six weeks, rather than three months. Second, we will update the targets weekly, rather than monthly. Each weekly update will be reported here.

At the moment, this shift will have little impact on the daily Presidential Tracking Poll. However, it could provide a significantly more reliable measure of the race as Election Day draws near."

http://www.rasmussenreports.com/public_content/politics/election_20082/2008_presidential_election/new_rasmussen_reports_party_weighting_targets_38_7_democrat_33_6_republican

____________________

cinnamonape:

Boomshak: "Thirty-five percent (35%) say they expect McCain to win, 34% expect an Obama victory, and 30% say the race is too close to call. (see crosstabs)."

But didn't you just say that Rasmussen is increasing the component od Republicans vs. Democrats and independents in their polling. In July, Democrats had a 5% advantage over Republicans in their survey...in August Rasmussen started shifting this to independents and then they moved the indy's into the Republican column.

This despite all evidence that new registrations of Democratic voters at levels between twice to ten times the number of Republicans. So why would Rasmussen increase Republican voters when RV numbers for the other party continue to go up?

Perhaps because it would make the campaign APPEAR to be a "horse race" and that the enthusiasm levels and support responses for McCain (and Palin) would increase in the surveys?

One would think that surveys would hold their criteria constant from poll to poll. In Science it is called a "controlled" experiment...play with the variables that are not to be measured and you'll get results that differ merely from the unrelated increase.

____________________

zotz:

This is why I am confused. Either Ras is right or the NYT is right. They both can't be right.

"In several states, including the traditional battlegrounds of Nevada and Iowa, Democrats have surprised their own party officials with significant gains in registration. In both of those states, there are now more registered Democrats than Republicans, a flip from 2004. No states have switched to the Republicans over the same period, according to data from 26 of the 29 states in which voters register by party. (Three of the states did not have complete data.)"
http://www.nytimes.com/2008/08/05/us/politics/05flip.html

____________________

slinky:

Well, unless you have good reason for changing the weighting of the variables.

But, since Rasmussen hasn't shared any good reason, we can just call his results 'biased'.

____________________

slinky:

It's kind of Republican accounting that Rasmussen is doing.

You know, like, let's not talk about the CIA budget, or Social Security, or bailouts of Fannie Mae or Freddie Mac,

Let's just provide a budget that says how nice and happy everything is.

No need to deal with reality, right?

Rasmussen appears to poll like that: But, I don't yet see that he's alot lot worse than his competitors. They are all less than supportive of a real Democracy.

It's called Capitalism. And, they prefer it to Democracy.

____________________

slinky:

Pollsters feel they have no obligation to the electorate.

They're wrong of course.

As wrong as the media is to think that.

But, that's the prevailing view.

____________________

KipTin:

Do not compare apples to oranges......

FACT: Rasmussen polls for LIKELY voters, NOT REGISTERED voters, and registered voters do favor Democrats. Likely voters even the field among political affiliation.

____________________

KipTin:

Look to the right column on "Latest Developments"....TIE in Minnesota by Star Tribune (45%).

Also -- SurveyUSAObama 49/McCain 47.

Minnesota will likely turn light blue soon on Pollster. Now toss-up on RCP.

____________________

zotz:

Ah ha, so it's the filter we are talking about!

Take for example the Marist NJ poll.
"Obama may be ahead by 7% in Marist New Jersey poll but it shrinks to 3% when restricting the sample to likely voters and leaners. Theoretically, as the change candidate that is bringing new voters into the fold, Obama's voters are less likely to be considered a "likely voter" in these polls. And it could be possible that his level of support is actually much higher than reported."
http://www.usaelectionpolls.com/


____________________

KipTin:

Yes, likely voters are the ones who will actually vote in November. If states had a high percentage of registered voters actually vote, then there would be no need to even discuss "likely" but that is not the case.

Here is some info from the 2004 election:

"More than 125 million Americans -- 64 percent of those ages 18 and older -- went to the polls in last year's presidential election, according to data scheduled to be released today by the Census Bureau."

Women 65%
Men 62%
Non-Hispanic White 67%
African Americans 60%
Hispanics 47%
Asians 44%

"The agency also found that turnout rates were closely correlated to a voter's age"--

65-74 73%
55-64 73%
45-64 69%
35-44 64%
25-34 56%
18-24 47%

http://www.washingtonpost.com/wp-dyn/content/article/2005/05/25/AR2005052501965.html

____________________

zotz:

MN is still blue on RCP.
PA has become a toss up (zogby?)

____________________

zotz:

If you are saying those percentages will apply to this election I have to say I do not agree.

____________________

KipTin:

What I am saying is that in the past...1972, 2000, 2004) there were HUGE efforts to register youth, minorities, and anyone else Democratic-leaning.

And the best the youth vote ever got was 1972 with 55%, AND that was the FIRST YEAR that 18-year olds could vote for President. Also that year there was a DRAFT for the Vietnam War.

2008 may be better than 2004 (Iraq War already in progross), but in reality the numbers will not change too much.

____________________

KipTin:

Someone at RCP is flip-flopping today. Look again at RCP Electoral Map

State by State Changes--
09/14 Minnesota Leaning Obama»»»Toss Up

____________________

KipTin:

Pennsylvania has been a toss-up on RCP since 9/7.

____________________

zotz:

I think Palin's "war with Russia" comment got a lot of young people's attention. If I were Obama I would be playing that on a running loop. (but that's just me) As far as the "nigras" are concerned everybody knows they only vote when they are paid to. (NOT!)

____________________

Independium:

The range that we see in daily polling underlines the wisdom of, among other reasonable approaches, the use of rolling averages. Taken at face value, these polls do not look good for a U.S. citizen who announces that, finally, there is a reason to take pride in this country and wishes to maintain the meager reason to have such pride.

____________________



Post a comment




Please be patient while your comment posts - sometimes it takes a minute or two. To check your comment, please wait 60 seconds and click your browser's refresh button. Note that comments with three or more hyperlinks will be held for approval.

MAP - US, AL, AK, AZ, AR, CA, CO, CT, DE, FL, GA, HI, ID, IL, IN, IA, KS, KY, LA, ME, MD, MA, MI, MN, MS, MO, MT, NE, NV, NH, NJ, NM, NY, NC, ND, OH, OK, OR, PA, RI, SC, SD, TN, TX, UT, VT, VA, WA, WV, WI, WY, PR