Pollster.com

That's right Ohio. 2.8 MOE is still too close for comfort.....

GOTV Ohio!!! 200k today certainly will help out this number. 50%+1 is enough to win OH

Even though the race looks tight this is the most important point about PPP - Ohio poll....

Obama banked a huge lead with early voters, who made up about 30% of the sample. He's up 65-34 with those folks. McCain's tightening the race to two points is predicated on his winning election day voters 54-44.

If Obama team in OH can get their people to vote to their utmost the 54-44 will be hard for McCain.

Glad the Sec of State and Gov are Dems.

I'm not that worried about OH. Only pollsters with a GOP house lean have shown McCain up within the last month.

FINAL WSJ/NBC National: OBAMA 51 MCCAIN 43.

So, what we got left to wait for tonight? Zogby? Is Drudge celebrating a one day bump, yet?

PlayingItStraight:
I look at the PPP poll here and other places and ask about selection. I say the same about Strategic Vision. I know both watch this site. They know the methodology, etc. So exactly who of party committees, candidate committees, 527s, etc. has paid each of you this cycle????

To both of you ... if you are on the up and up here and don't have a "dog in the fight" or "skin in the game" you'll disclose it. And someone might actually check the reports after the election.

I love this site. Its owners have their histories and biases just like all of us, but they do a really good job of trying to be objective, even if they do let their view obstructe their analysis at times. We all do. I have to admit, I have become really frustrated by the conversations of political cheerleading both right and left.

I wish the discussion were more directed at the "mathematics" than the opinions people express. Good God. The ignorance of the science (or art) of polling expressed by some is truly frustrating. At least understand a little bit of what polling is all about before shooting off your mouth about who is good and who is bad.

I think one of the great things about the authors of this site is that they truly seem to care about the "math" even if they are clearly for one candidate or another. Point is, if the math is wrong, understand why and how to do it better in the future.

Now to pontificate a little more ... and please feel free to tell me I am an idiot.

The biggest problem going with polling this cycle is the competition to call it right. Everyone seems, with a few exceptions, to want to win the "prize" for being closest. Being closest doesn't mean you did the best job of presenting the existing data. In fact, with some of the clowns doing polling these days it seems like they are more concerned with their models than the data.

I have a pretty extenstive background in using polls as stategic issue and marketing development. The firs thing you look for in hiring one of those people is their commitment to generate meaningful data. I never wanted the sugar coated approach. So ...

Hats off to a couple of people this cycle ...

DailyKos/R2K ... if I was the Obama campaign I wouldn't like them for anything other than their results ... why ... I think their model way overstated the reality of the electorate. Again, remember I said ... its accuracy ... not make me feel good.

GWU/Battleground ... at least once a week they give you extensive internals. You can make your own judgements. And they have an unaided ballot. In other words, data with no push.

Rasmussen ... sort of ... once a week if you pay for them you get crosstabs of some use. I'm a subscriber ... but I think their daily snapshot hides more than it reveals. I'd like to know a little more about the crosstab movement, but they like everyone that does party weighting ... well ... not showing the CTs daily hides more than it reveals. Oh, and one biggy that Rasmussen does give is "certain voters." That is a major missing detail from virtually everyone else except Rasmussen.

Mason Dixon -- I know they are criticized as leaning R. But truth be told, their polls are almost always accurate from a statistical MOE approach ... even when they pick the wrong winner. Part of the reason is that their polling approach does not push excessively to get people to "make a decision." Too many others do.

For this years biggest hall of shame pollster ... regardless of the results ... GALLUP. Show us some more data. Don't screw around with multiple models. Notice the convergence. Hello folks ... regardless of results ... can't you see them looking for validation for what they've been doing all these years.

Second place hall of shame ... zogby. Let's make dull boring data look newsworthy with hype rather than sticking to a straight statistical discussion.

Third Place Tie ... ABC, CBS/NYTimes, CNN. How about even a morsel of meaningful internals guys. If you believe in your work then lay it all out on the table.

Almost done venting ... but here's the deal. When pollsters tell us its 52 -47 with an MOE of 2 it suggest the race will fall between 50 - 49 and 54 -45. But if you see the real data it says 47-43 certain voters ... or 49-41 with 10 left or 45 -45 with ten left. A very different picture. Point is that the pollsters are too concerned with projecting their opinions and views of likely outcomes than reporting meaningful AND COMPLETE DATA so we can all look at it and make our own assessments.

Tell us what the DATA SAYS ... NOT WHAT YOU THINK IT SAYS.

For instance, I'll pick on Morning Call in PA right now. I'm sure glad that you're not my pollster if I'm Obama. Or McCain for that matter. How on God's green earth can you have the size of the move that you've had in three days if you've got your methodology down? I try to look at the numbers --- with the little you get and it's pretty hard to find a statistical regime under which that happens unless you were Mark Foley in 2006.

OK enough ... I welcome slings and arrows based on a discussion of data and methodology which is what I thought this was all about. Oh. For hoots, lets all say where were from and where we are registered to vote. As for me ... registered voter in the Commonwealth of Virginia since 1978. Prior to that registered voter in New York State.

KMart Dad:
Glad the Sec of State and Gov are Dems.

----------------------------------------

Gee, I can't imagine why. I bet Cuyahoga County will have "disenfranchised voters" election night and certain carefully chosen polling precincts will be forced to stay open late.

@RaleighNC

Or maybe he meant that they can make sure there is no suppressing carefully chosen precincts like in the past?

I don't know what to make of Obama having a relatively high national vote lead, but he remains close in OH, FL and the other battlegrounds. Could they be overestimating the GOP GOTV effort?

Thank goodness all Obama needs is Kerry States + CO + NM or NV.

Finally, Ohio and Florida aren't the be all and end all of politics!!! And this is coming from an Ohioan. :)

This is a petition NOT to visit the zogby site tonight. His crappy polling along with trying to diss Nate Silver calls for a boycott! Who's in???

while these numbers are close, i think they're going to be very close to the actual results. I feel more confident about a 2 point lead now than a 5 point lead two months ago.

why does ibd have poll numbers for people who display flags and people who dont and who they support.thats the kind of stuff that makes wonder about ibd. if thats part of the over all poll numbers then it could be skewing the poll just a little. and it also tries to insuate something as well.

@ Playing it straight. OK : I'M from CT and a registered DEM. You are boring - plain and simple. Your post was way too long and obviously intentended to impress us trolls as to how smart you are. Bring some wit to the table and you will see that we are noelcowards.

Oh yeah, I forget about Iowa. He needs that too, but it's likely safe.

If things go well for Obama early on in the evening (and I'm planning to be in PA to make sure they do) OH will be fun to watch. McCain needs to kick ass there statewide tomorrow and win by 12 on election day returns to make up the deficit. Kind of like watching a race where the runners in the inside lanes start out far behind but everyone kind of enters the stretch even. I think OH is an important psychological race because of what happened here last time. Many Dems still feel that the election was literally stolen and there are still ongoing investigations regarding vote fraud (by Repubs in this case).

Anyway, this will be a great test of GOTV and enthusiasm in a state where Repubs used to rule; do they still have that fire in their bellies to get their people to the polls? I think we all know the answer regarding Dems...we'll find out about the 'pubs soon.

Just great to be able to think about these numbers and NOT be freaking out. I WOULD be freaking out if they were the final PPP numbers for PA I'd be in my car now, probably.

so what about this situation. Obama wins only the dark blue states plus Nevada with North Dakota or Montana putting him over 270?

+2 is a statistical tie. OH will be decided by one single factor - turnout.

If you live in Ohio and want Obama to win, then you MUST do MORE than just vote for Obama. When you go in to vote for Obama, bring a bunch of your friends. Tell them it'll be fun if you go together. On the day of the election, call your friends, call your relatives, call your neighbors, call everyone you know and tell them to go vote for Obama. And tell them you aren't gonna stop pestering them until they actually gone and have cast their vote.

this is driving me freaking nuts I wish we could just go ahead and vote allready

Regarding PPP's OH poll: overwhelming advantage among the early voters (for Obama), and a rather large deficit among those who whill vote on the election day....

Ummm............ This projection will only hold IF the reason why Obama has such a wide margin among the early voters is because Obama voters are voting early, while McCain voters are simply waiting for 11/4.

however, what if the real reason for the early voter advantage is due to the aggressive and highly effective GOTV advantage Obama has. If that's the case, then we should also reasonably project that the same GOTV advantage will also benefit Obama on the election day - meaning, those voters who were "dropped" by the LV voter model would should up, and other Obama voters will be pushed to show up and vote proportionally more than the McCain supporters....

In that case, the projection that McCain will win over Obama among the voters that vote on 11/4 is not a done deal.

I remember reading an interview with Axelord that his internal analysis of the early voters confirm their strategy: that, many first time and sporadic dem leaning voters who would have been dropped by the LV type poll model indeed SHOWED UP as early voters.

Thoughts?????

Regarding PPP's OH poll: overwhelming advantage among the early voters (for Obama), and a rather large deficit among those who whill vote on the election day....

Ummm............ This projection will only hold IF the reason why Obama has such a wide margin among the early voters is because Obama voters are voting early, while McCain voters are simply waiting for 11/4.

however, what if the real reason for the early voter advantage is due to the aggressive and highly effective GOTV advantage Obama has. If that's the case, then we should also reasonably project that the same GOTV advantage will also benefit Obama on the election day - meaning, those voters who were "dropped" by the LV voter model would should up, and other Obama voters will be pushed to show up and vote proportionally more than the McCain supporters....

In that case, the projection that McCain will win over Obama among the voters that vote on 11/4 is not a done deal.

I remember reading an interview with Axelord that his internal analysis of the early voters confirm their strategy: that, many first time and sporadic dem leaning voters who would have been dropped by the LV type poll model indeed SHOWED UP as early voters.

Thoughts?????

Actually, if PPP is correct about these numbers in Ohio McCain needs a 57%-43% break on the day of Election.

The 54%-44% difference gets you to that 2 point win by Obama. That would not be enough.

According to the poll 30% of the respondents say they have voted. So that's presumably 10% M and 20% Obama with 70% remaining.

McCain would need another 40% to win. Obama would need a little more than 30% to cross the 50% threshhold.

40/70=57.15% 30/70=42.9%

Gonna be very tough for McCain.

Of course maybe people are inflating their early voting actions.

Actually, if PPP is correct about these numbers in Ohio McCain needs a 57%-43% break on the day of Election.

The 54%-44% difference gets you to that 2 point win by Obama. That would not be enough.

According to the poll 30% of the respondents say they have voted. So that's presumably 10% M and 20% Obama with 70% remaining.

McCain would need another 40% to win. Obama would need a little more than 30% to cross the 50% threshhold.

40/70=57.15% 30/70=42.9%

Gonna be very tough for McCain.

Of course maybe people are inflating their early voting actions.

@ dbrot81

Speak for yourself. I voted last Tuesday. :)

@common sense

Not trying to impress anyone. We all get on pins and needles and vent. Sorry if I annoyed anyone or ruined the fun.

dont overreact, they are called battleground states for a reason, one of them is bound to be close, however small victories in each one add up to a decent national victory as polls are showing

you can and should....if you live in OH, and then, yes...convince all your friends to vote early too...then sit back and enjoy seeing PA and VA go blue and the rest is just for show;

11,700 Obama volunteers in the field in VA...kinda getting close to Civil War Battle numbers now. That means about ..what....750,000 to 800,000 people contacted on just one day? I called one person today in VA who had just had someone knocking at her door and another who was also a volunteer back from canvassing.

With a similar (I think slightly downsized) sustained effort in OH I believe Obama will get within 7% or 8% of McCain tomorrow and win by 4-5% due to his "banked" votes, unless the Republicans have the equivalent of Barack's army waiting in their "armory"; we'll see what kind of weapons they've got tomorrow.

@ liz from NJ

Good point! And I read a great analysis that said getting lots of early votes for your candidate is a great advantage, but not for the reason that most people think. It's very probable that all the people who voted early were enthusiastic people who were gonna vote for their candidate regardless of whether they voted early or on election day. However, once you've voted early and the candidate's campaign system knows that you've already voted, that means they don't have to waste any GOTV effort on you, so they can aim their entire GOTV resources on the people who are less enthusiastic and who might not have gotten out to vote without being spurred on by GOTV encouragement.

But putting this one along with the Columbus Dispatch poll, in combination with the remarkable national numbers now breaking for Obama, makes me feel cautiously confident about OH for Obama.

liz what part of NJ?

Liz...I would tend to agree with your thinking. I suspect that there is a synergistic effect. Early voters will participate in GOTV activities and the ability to target likely Obama voters on the day of the election thereby increases.

At the same time it's been argued that at some point the early Obama voters "tap out" a pool of likely voters and thus reduce the pool available on the day of the election.

Most pollsters hold that about 60% of the electorate will come out. But this election may prove that is wrong...or if it's true there may be a "depression" by the Republicans and a "enthusiasm" factor (noticed in most polls) by the Democrats (and other Obama supporters). I don't know of any pollsters that include that partisan factor in their likely voter model though many ask the question.

@Straight

No way Zogby doesn't take the prize for being the worst pollster in years.

I keep seeing poll results - both national and regional - that tell the story of VERY LARGE margin for Obama among the early voters.

Then, the pollsters go on and say that the end results will be a much narrower win for Obama since the remaining voters are not breaking with the same/similar margin in favor of Obama.

I think this type of conclusion is only possible if the early-voter-for-obama is solely due to the enthusiam of the Obama voters who can't wait till 11/4.

If we treat Obama's GOTV as an "independent variable" with a "causal relationship' with the Obama win among the early voters, then we could resonably assume that the GOTV advantage will also produce "un-calibrated" advantage for Obama that goes beyond the voter's stated preference for Obama.

My personal belief is, the Obama advantage among the "late voters" may not be as large as that among the early voters, but larger than what the pollsters are projecting..... I am basing this on the assumption that Obama's GOTV has a direct impact on both early voters AND late voters.

Liz and cinnamonape:
I agree with you guys.
In SUSA as well, Obama has a huge lead in early voting but they are tied among others:
http://www.surveyusa.com/client/PollReport.aspx?g=8247ab3a-7593-421d-8100-61e6e00ebc0b

BTW The Unions in Ohio (and elsewhere) have told all their members that they need to vote early and have strongly encouraged GOTV efforts. This includes getting vehicles to drive people to the polls, "buddy systems" to make sure that the buddy goes to vote, and a host of other strategies.

On campuses there will be people outside of dorms and dining commons, hitting up cafes, bus stops, senior centers, working hard in certain zip codes. Then there is the cell-phone message thingee!

Just read NBC / WSJ final poll will be O + 8 tomorrow. Not bad!

any news from zogby?

Considering the usual position of PPP amongst the polls, I'd say it's about time to GET OUT THE VOTE!!!, and forcefully so!

@ Pro-America

His polling is so bad it crashed his site. Anyone know what numbers he pulled out of his hat tonight?

ZOGBY:

Obama +7

Obama also leading in Zogby's new ALL 8 battleground state polls!!!

Obama up 7 in Zogby

My 17 year old son went to PA (from NJ) to do canvassing for Obama over the weekend.

Both days, he was assigned to a predominantly black community.

He came back and told us that as a kid who grew up in an upper middle class, predominantly white community he had a gut level feeling that people might get mugged in a community like that. But, he felt so welcomed by the people, and the folks opened their doors and arms when they heard that this scrawny teenager from NJ was there for Obama, he said "mom, I could have walked around $1M in cash, and I would feel completely safe among these folks".

He went to PA to work for Obama, but I think more than Obama, he helped himself.

I am so glad that he did it.

all we're left with from Zogby isthe teaser that the result is "NOT GOOD for McCain"...site seems to be down, or at least won't load completely; are millions of light-sleeping Dems trying to logon to Zogby right now?

Just a small note. A 5% popular vote lead with 130,000,000 expected turn out (if not more) is a win by 6,500,000 for Obama.

Correction on Zogby state polls.. Obama leads in 6 out of 8 states... McCain leads by 1 in NC, and 5 in IN

Everyone go to reuters for latest Zogby polls

Zogby state polls. Details! We need details!

/can't get zogby's site to load
//yes, I know zogby is full of crap, but I need my poll fix

Obama up 11 in PA
Obama up 2 FL
Obama up 6 in VA and OH
Obama up 8 in NV
Obama up 1 in MO

Here's the link to today's Zogby numbers:

http://www.reuters.com/article/newsOne/idUSN3134134020081103

All good news based on previous set of polling. What Nate over at 538 says about Zogby is that, while his numbers are all over the place it's the trends that are meaningful (just far too dramatic for his or anyone's taste). Having said that, if Obama is moving upward Nationally, and in PA, OH , and VA and NV that lets me breathe far easier. It's not that I think Obama is up 11 in PA. No way; but I do think the downward pressure may have abated, and that's what we've all been (secretly perhaps) a bit worried about, though Axelrod and Co didn't seem too concerned. We all have to sometimes fight with our own inner concern troll I guess. This and PPP PA, plus Ras moving back towards Obama (plus a superior GOTV effort) makes me now feel pretty confident about PA and VA

alright:
All Obama
National by 7
Penn by 11
Virginia by 6
Ohio by 6
Florida by 2
Missouri by 1
Nevada by 8
NC by 1
Indiana by 5

These realy are great numbers for Obama (if you have any faith in Zogby).

I would also like to comment I have little faith in Zogby!!!

Zogby three day tracking
50.9 Obama
43.8 McCain
5.3 Not Sure / Other

Zogby has realized that he can't play partisan with Drudge and still keep his reputation. So now he's made major revisions to his analyses so that he can be "in the game".

It's ridiculous!

I wonder if he just makes up numbers for convenience sake. Does anyone audit these guys?

How is Obama polling among PA Catholics?

@ ModerateGuy

No, Zogby has McCain leading in NC and IN.

Since we know the numbers from Friday (Obama 47, McCain 48) and Saturday (Obama 52, McCain 42) we know what the numbers were on Sunday:

Obama 54
McCain 41

Anyone got the state numbers, not just the spread?

Oops would like to correct my last stament McCain leads in Indiana and NC. Sorry for the typo

remember its zogby..

I'll be more interested to see what he (zogby) has to say tomorrow. Probably Barr or Nader by 3.

Except for PA these numbers do all look pretty realistic, though I think Obama has already done enough to possibly win in NC, due to massive AA early voting; just a huge margin to have to overcome.

I think IN is maybe 2-3 points closer but no biggie if McCain wins there. Would be nice to see those ultra-hard-working Obama Hoosier (and more than a few IL) volunteers have something HUGE to celebrate. If Obama is really -2 now GOTV can actually win it in IN since there's no comparable McCain presence on the ground.

FL is a true tossup with a slight Obama lean....1-2 points.

OH is a better pickup opportunity now based on the early voting gap (which is much smaller thus far in Florida)

does Zogby provide numbers for early voters in states that do early voting?

For instance, given the party ID and AA turn out among the NC early voters, I have to assume that Obama is doing very well among them.

If McCain is still winning by 1 point, it means that the "late voters" are all breaking for McCain.

Re: my earlier post - I don't think this will hold. I believe that the GOTV advantage that benefited Obama among early voters will also have a positive impact among the late voters that the polling based on voter intention is not accounting for.

The zogby/reuters numbers are:
PA
O: 53.7
M: 40.0

OH
O: 50.2
M: 43.9

VA
O: 50.7
M: 44.6

NV
O: 50.7
M: 42.9

FL
O: 47.5
M: 46.2

MO
O: 47.4
M: 45.7

NC
O: 47.7
M: 49.3

IN
O: 43.9
M: 49.1

Zogby Battleground polls with numbers

@ RussTC3

That assumes that Zogby's sample size was exactly the same on each day, which is not necessarily true.

actually florida has a 4 point obama leab, as per RCP average

Zogby (nov 3)

Pennsylvania (21)
Nov. 2
Obama
53.7%

McCain
40.0%

Other/Not Sure
6.3%

Ohio (20)
Nov. 2
Obama
50.2%
McCain
43.9%
Other/Not Sure
5.9%

Virginia (13)
Nov. 2
Obama
50.7%
McCain
44.6%
Other/Not Sure
4.7%

Nevada (5)
Nov. 2
Obama
50.7%
McCain
42.9%
Other/Not Sure
6.4%

Where there are two numbers, top is most recent

Pennsylvania (21)
Nov. 2

Obama
53.7%

McCain
40.0%

Other/Not Sure
6.3%

Ohio (20)
Nov. 2
Oct. 26

Obama
50.2%
49.7%

McCain
43.9%
45.1%

Other/Not Sure
5.9%
5.3%

Virginia (13)
Nov. 2
Oct. 26

Obama
50.7%
52.0

McCain
44.6%
44.8

Other/Not Sure
4.7%
3.2

Nevada (5)
Nov. 2
Oct. 26

Obama
50.7%
48.2

McCain
42.9%
44.0

Other/Not Sure
6.4%
7.8

Florida (27)
Nov. 2
Oct. 26

Obama
47.5
49.7

McCain
46.2
46.4

Other/Not Sure
6.4%
3.9

Missouri (11)
Nov. 2
Oct. 26

Obama
47.4%
48.2%

McCain
45.7%
45.7%

Other/Not Sure
6.9%
6.2%
North Carolina (15)
Nov. 2
Oct. 26

McCain
49.3%
46.4%

Obama
47.7%
49.7%

Other/Not Sure
3.1%
3.9

North Carolina (15)
Nov. 2
Oct. 26

McCain
49.3%
46.4%

Obama
47.7%
49.7%

Other/Not Sure
3.1%
3.9

Indiana (11)
Nov. 2
Oct. 26

McCain
49.1%
50.2

Obama
43.9%
44.0

Other/Not Sure
7.0%
5.8

@ Playing it straight : You're OK. I'm the one with the problem. Obama's election is actually something that matters to me. My life has basically come down to the health of my family and the outcome of this election.

when a butterfly flaps its wings it sounds like a grenade in Zogby's polling but....there really WAS that butterfly. He picks up trends, (or else he pretends to at least). If you look very closely at the fluctuations in his polls they generally do parallel similar (though much smaller) ones in poll averages. So...maybe he's the canary in the coal mine. Main thing is he does need the business and, for his final polling numbers it's in his interest to do everything in his power to get it right.

Bottom line is that...if we see Ras tomorrow at 6 I might go out on a limb and say that Zogby, deservedly so or not, may have nailed the result this year, along with Kos/R2K. That would really be a hoot after all the abuse they've both taken, and perhaps the abuse is correct and they just both ended up in the right place (Obama 6.5% to 7% says I).

Of course, if Ras is down at 4 or 3 then I'll just throw up my hands; it's rare to have such a huge disparity between respected pollsters one day out from a presidential election This is the money shot for everyone so I really want to see where Ras comes in.