Zogby Interactive: This is officially the funniest sh**t ever.

[Robespierre's Jaw]

LOL. Thank God for Zogby. It proves that even so called "experts" can get polls catastrophically wrong. I wonder who the people at Zogby question regarding this years Presidential Election? That question will have to be answered when Zogby does a real poll, when they question real people, whenever that might be....

[minionofmidas]

People are exaggerating in this thread. Some of these are well off, sure. Bound to happen with fifty polls at once, not to mention with internet polling.

I want another SUSA 50-state sweep now.

[Brittain33]

Or the UK:

"Conservatives set to landslide Wales"

"Conservatives trail badly in Cities of London and Westminster"

"Tories winning the Northeast"

"Labour winning southern England"

You forgot "Labour, Conservatives tied in Northern Ireland."

"12-point swing to Respect in Tunbridge Wells"

[Brittain33]

What's truly sad is that this is the future of polling, so we'd better learn to like it.

[Democratic Hawk]

Third, from other unpublished surveys I have seen, I suspect that the "undecided" category is likely to break favorably for McCain.

Hopefully not should Obama cross the acceptability threshold. The task ahead for Obama is to reassure those who are 1) looking for a change of direction and 2) want to believe in him and his vision but remain sceptical between now and November - and only he can do that

Dave

[Wiz in Wis]

What's truly sad is that this is the future of polling, so we'd better learn to like it.

Betamax...

Laserdisc...

Altavista...

Interactive polling

[Josh/Devilman88]

Like many have said, more polls in the next two election cycle will be done like this. With cell phones taking over land-lines, phone polls are getting harder to do.

[Albus Dumbledore]

Zogby is the worst.

[Eraserhead]

Ummm... wow. Just wow. I mean... wow. I'm speechless.

[Josh/Devilman88]

It would be funny if the election turns out like this.

[Fmr President & Senator Polnut]

It'd be funny ... but for some reason unnervingly so.

[Eraserhead]

Oh yeah and lol @ CARLHAYDEN for trying to push this poll because Barr is doing well in it. You go Carl, let your freak flag fly buddy.

[kevinatcausa]

Apparently a map based on the actual polling data would have been too absurd even for Zogby, so he had to make up his own criterion for the colors.  For example:


Florida: McCain +4 (2.5x MOE), McCain gets the state
Iowa: Obama +4 (1x MOE), Obama gets the state
Texas: McCain+3 (1.76x MOE), McCain gets the state
Colorado: Obama+2 (0.56x MOE), Obama gets the state
Oklahoma: McCain+5 (1x MOE), McCain

Virginia: Obama+5 (1.78xMOE), Too Close to Call
North Carolina: Obama+9 (3.3xMOE) Too Close to Call

Alaska has a "Single Digit" lead for McCain (with likely a huge margin of error), but he just calls the state for McCain based on conventional wisdom instead of giving the actual data (I wouldn't be surprised if this is what happened with Vermont too--the actual results were absurdly close with a huge margin of error, so he just chose not to present them.)

[??????????]

[○∙◄☻¥tπ[╪AV┼cVê└]

Since Obama has a 3 point lead in Arizona, shouldn't it be solid blue like New Hampshire? LOL.

[kevinatcausa]

Ooh...I missed New Hampshire from my list above!   I think Zogby's methods were:

Look at my polling results, and look at what everyone else's results were.  If they match, color the state accordingly.  Otherwise, make it purple regardless of the actual polling results. 

[CARLHAYDEN]

Apparently a map based on the actual polling data would have been too absurd even for Zogby, so he had to make up his own criterion for the colors.  For example:


Florida: McCain +4 (2.5x MOE), McCain gets the state
Iowa: Obama +4 (1x MOE), Obama gets the state
Texas: McCain+3 (1.76x MOE), McCain gets the state
Colorado: Obama+2 (0.56x MOE), Obama gets the state
Oklahoma: McCain+5 (1x MOE), McCain

Virginia: Obama+5 (1.78xMOE), Too Close to Call
North Carolina: Obama+9 (3.3xMOE) Too Close to Call

Alaska has a "Single Digit" lead for McCain (with likely a huge margin of error), but he just calls the state for McCain based on conventional wisdom instead of giving the actual data (I wouldn't be surprised if this is what happened with Vermont too--the actual results were absurdly close with a huge margin of error, so he just chose not to present them.)

Of the states you listed, on the surface the map is not supported by the data in: Colorado, North Carolina, Oklahoma and Virginia.  However, Zogby may realize that his sample in those states is not representative of likely voters, and he is "weighing" the results in the map.

[○∙◄☻¥tπ[╪AV┼cVê└]

Wow, Obama is actually up in all of the purple states except for IN, and there's no poll for MT and ND. He actually does better in purple AZ than blue CO. No idea how Zogby chose the numbers.

Anyways, he didn't poll some of the smaller states. I guess not enough people signed up who said they are from those states.

[Hash]

Anyways, he didn't poll some of the smaller states. I guess not enough people signed up who said they are from those states.

And Zogby made up its crappy commentary.

[The Mikado]

Before you guys laugh too much, consider this.  Zogby is reaching people who callers do not.  Young people of the internet age as opposed to older folks who are more likely to answer the phone in their homes and are more likely to vote for McCain.  Could this be the unreached and motivated younger crowd who have reached voting age at a time when they hate current conservative politics?   Maybe?  Seems like a streach, but Dems have made huge in roads with this group of voters and they could be that 5% or so who tip the election in ways that we have never seen before.  Remember, even a state like SC for example, has a huge amount of Democrats, they just don't win the state.  Could the cell phone and Facebook crowd that is not being reached by polsters be getting reached by Zogby?  An interesting thought.

This isn't a scientific poll.  People can sign up for it and there's no randomization.  It's a straw vote, not a poll, and it's worse than useless.

I'd have no problem, in theory, with internet polling, as long as it obeyed the laws of polling: random sample, no self-selection, etc.  But anyone who's ever opened a statistics textbook knows that John Zogby has produced the worst statistical piece of human excrement since the 1936 Literary Digest poll, and that would be true even if the election turned out exactly like Zogby's numbers indicate.