PDA

View Full Version : Elections The polls mean next to nothing it this election. Here's why...


jAZ
07-27-2008, 10:30 PM
The polls in this election mean nothing in this election and here's why...

Turnout.

Aside from the wording of questions, turnout is the component of election polling models that is most more art than science. It's the part that skews the numbers up or down.

I say this at a moment where I'm not dismissing the latest results because they don't favor Obama (according to the Gallup Daily Tracking Poll, Obama is up now up 9% after his trip from 2% before).

That poll means nothing.

The key is will Obama be able to turn out his voters and will McCain be able to turn out his voters?

Pollsters try to predict that turnout, but that is the weakspot.

My gut tells me that Obama will have a Bush04 like turnout machine and McCain will have difficulty driving turnout.

The question will be, will McCain be able to drive himself to a large enough overall lead as to offset that machine that Obama and the Dems are building. That's an odd question given the machine that Bush built in 04. It carried him to an improbable victory. But McCain won't be able to do the same thing to the same degree. I'm almost certain of that fact. How close can he get though.

As an FYI, one thing I read that might be helpful for those who care to use in interpreting polls right now. Because turnout is an unknown, it's really hard to accurately predict a "likely voter" in a poll, so "registered" voter results are slightly more trustworthy for the time being.

Direckshun
07-27-2008, 10:36 PM
Hey if turnout's key, Democrats will sweep.

Guru
07-27-2008, 10:38 PM
That is what they said in 2004 too.

jAZ
07-27-2008, 10:40 PM
Hey if turnout's key, Democrats will sweep.

It's everything, IMO.

I think that's why strategists on the GOP side are quietly very nervous. McCain needs a big lead to overcome Dem enthusiasm. McCain needs Clinton voters, some of which he can get.

But this year, it's not about the swing voters as much as the turnout machines.

jAZ
07-27-2008, 10:41 PM
That is what they said in 2004 too.

And "they" were right, Bush won entirely based on his ability to turnout his voters.

Guru
07-27-2008, 10:45 PM
all I remember hearing was how the democrats felt that Kerry would run away with it.

jAZ
07-27-2008, 10:46 PM
all I remember hearing was how the democrats felt that Kerry would run away with it.

Then I'd withhold comment on a substantive thread like this. Not to be a dick, but you are admitting you didn't know what you were talking about when you posted. That's never good.

jAZ
07-27-2008, 10:50 PM
all I remember hearing was how the democrats felt that Kerry would run away with it.

Just a bit of background.

The polls showed Kerry up 4-5% the day before the election. Something significant like that. And there was a ton of backlash when Bush won by a modest/decent margin.

The reason even the last minute polls were wrong in 04 was because they got the turnout wrong. But turned out his base, Kerry didn't have the charisma couldn't strike up the enthusiasm beyond the anybody but Bush crowd (and the Dem base).

Guru
07-27-2008, 10:51 PM
Then I'd withhold comment on a substantive thread like this. Not to be a dick, but you are admitting you didn't know what you were talking about when you posted. That's never good.
:rolleyes::shake:

jAZ
07-27-2008, 11:00 PM
:rolleyes::shake:

Not my fault you weren't paying close attention in 2004.

markk
07-27-2008, 11:02 PM
people harp on turnout every time around. i don't think there are a whole lot of people who vote sometimes and not other times. i rather think there are some people who always vote, some people who never do, and relatively few who don't fall into one of those groups.

there's not really a good methodology to see how many people for example are registered in these registration drives who actually end up going out to vote. obviously someone who has never been registered at all wasn't too worried about voting in the first place. just one example

Guru
07-27-2008, 11:05 PM
Not my fault you weren't paying close attention in 2004.I wasn't referring to the result of the election but the feelings of everyone going into it.

You are an impossible individual. You go on and stay in your world of superiority. I'll leave your superior threads alone since there is no reason to ever doubt how advanced you are. Hell, I don't even dislike you but you still feel the need to talk down to everyone.

Good night.

Mr. Kotter
07-27-2008, 11:07 PM
Just a bit of background.

The polls showed Kerry up 4-5% the day before the election. Something significant like that. And there was a ton of backlash when Bush won by a modest/decent margin.

The reason even the last minute polls were wrong in 04 was because they got the turnout wrong. But turned out his base, Kerry didn't have the charisma couldn't strike up the enthusiasm beyond the anybody but Bush crowd (and the Dem base).

If you are going to be an ass and call someone out for not knowing what they are talking about, you really should know what the fugg you are talking about....

:rolleyes:

Mr. Kotter
07-27-2008, 11:08 PM
I wasn't referring to the result of the election but the feelings of everyone going into it.

You are an impossible individual. You go on and stay in your world of superiority. I'll leave your superior threads alone since there is no reason to ever doubt how advanced you are. Hell, I don't even dislike you but you still feel the need to talk down to everyone.

Good night.

Too bad he's usually FOS too....like I just demonstrated. ;)

markk
07-27-2008, 11:10 PM
i don't think kerry ever had his campaign straight after that comedy of errors that was the democratic convention.

honestly, i didn't think Bush would be re-elected, but after John Edwards gave that chuckler with 'two americas', and then Kerry marched out on the stage with his salute and his "John Kerry reporting for duty" groaner of a line... that i think was the moment where i moved from thinking, 'Kerry's going to win' to 'by george they may find a way to fumble this one yet'

the whole thing was just very campy

jAZ
07-28-2008, 12:31 AM
If you are going to be an ass and call someone out for not knowing what they are talking about, you really should know what the fugg you are talking about....

:rolleyes:

Kotter's right here, sorry.

I was speaking of the exit polling discrepency numbers. Early results (not day before results) had Kerry up 4-5% then the results were 1% or so for Bush.

jAZ
07-28-2008, 12:34 AM
i don't think there are a whole lot of people who vote sometimes and not other times. ... relatively few who don't fall into one of those groups.
What (approx) do you mean by "relatively few" and "(not) a whole lot"? Just a general are? 1%? 5%? 10%? more?

If we agree but you think that's not much while I might.

DaneMcCloud
07-28-2008, 12:56 AM
That is what they said in 2004 too.

QFT.

I won't believe it until I see it.

Mr. Kotter
07-28-2008, 08:20 AM
....

I was speaking of the exit polling discrepency numbers. Early results (not day before results) had Kerry up 4-5% then the results were 1% or so for Bush.

Exit polls, epecially, are a joke.

Maybe that's because many folks are sick of journalists, polling organizations, and politicians....and other "pundits" telling them stuff, often, that may or may not be an accurate assessment of what they are thinking. In other words, we mess with talking heads and those we feel are trying to manipulate the process.

Put another way, we like to screw with those that screw with us....on a daily basis, through their distortions, their biases, and their obsessive over-emphasis on hypothetical polls and nuanced wording of silly assed surveys.

:hmmm:

Mr. Kotter
07-28-2008, 08:22 AM
QFT.

I won't believe it until I see it.

The Silent Majority still wields significant power....to be sure--despite the mainstream media's wishful thinking and premature obituary.

markk
07-28-2008, 08:38 AM
Kotter's right here, sorry.

I was speaking of the exit polling discrepency numbers. Early results (not day before results) had Kerry up 4-5% then the results were 1% or so for Bush.

different people vote at different times of the day.

markk
07-28-2008, 08:41 AM
What (approx) do you mean by "relatively few" and "(not) a whole lot"? Just a general are? 1%? 5%? 10%? more?

If we agree but you think that's not much while I might.

how much does turnout fluctuate in presidential election years, when adjusted for changes in population or total number of registered voters?

I remember hearing last time how amazing Kerry was and how the democratic base was so energized and they had registered more young people to vote than ever before and people were really going to turn out because the war was going so poorly and Bush was so hated. same things were said in 2000. it didn't happen.

***SPRAYER
07-28-2008, 08:45 AM
how much does turnout fluctuate in presidential election years, when adjusted for changes in population or total number of registered voters?

I remember hearing last time how amazing Kerry was and how the democratic base was so energized and they had registered more young people to vote than ever before and people were really going to turn out because the war was going so poorly and Bush was so hated. same things were said in 2000. it didn't happen.

But this time it's gonna be different.

:)

jAZ
07-28-2008, 08:55 AM
how much does turnout fluctuate in presidential election years, when adjusted for changes in population or total number of registered voters?

I remember hearing last time how amazing Kerry was and how the democratic base was so energized and they had registered more young people to vote than ever before and people were really going to turn out because the war was going so poorly and Bush was so hated. same things were said in 2000. it didn't happen.

2004 was the highest turnout on record since 1968, IIRC. Which turned around means that 1968 still has the highest turnout, even when the population was much smaller that every year before.

Turnout changes, and turnout of various population changes. And the variation is signicant when compared to the margin of victory in many elections. It's not 50%, but it's enough to make an impact.

In 2000, Bush found a way to sqweeze every bit out of the evangelical vote in a way that other GOP candidates before him hadn't been able to do. Kerry set a record for youth vote in 2004. It just wasn't enough to offset Bush's turnout.

Turnout matters in a huge way, in fact, it's the single most important aspect of the final stage of the campaign. It used to be the final day, but with mail in voting, it's now the final month.

It's why people like Bill Clinton were going to heavy Dem areas in the final days. It's not to make up people's minds, it's to energize like-minded voters, to ensure they turnout.

mlyonsd
07-28-2008, 08:59 AM
I agree polls mean nothing at this point.

It's all about the debates.

markk
07-28-2008, 09:30 AM
2004 was the highest turnout on record since 1968, IIRC. Which turned around means that 1968 still has the highest turnout, even when the population was much smaller that every year before.

Turnout changes, and turnout of various population changes. And the variation is signicant when compared to the margin of victory in many elections. It's not 50%, but it's enough to make an impact.

In 2000, Bush found a way to sqweeze every bit out of the evangelical vote in a way that other GOP candidates before him hadn't been able to do. Kerry set a record for youth vote in 2004. It just wasn't enough to offset Bush's turnout.

Turnout matters in a huge way, in fact, it's the single most important aspect of the final stage of the campaign. It used to be the final day, but with mail in voting, it's now the final month.

It's why people like Bill Clinton were going to heavy Dem areas in the final days. It's not to make up people's minds, it's to energize like-minded voters, to ensure they turnout.

right... until i see some kind of indication that turnout has ever varied more than a couple of percent i'll remain unconvinced.

jAZ
07-28-2008, 09:43 AM
right... until i see some kind of indication that turnout has ever varied more than a couple of percent i'll remain unconvinced.

Normally, I'd endorse such a statement as being a reasonable standard for judgement. But you seem to have formed your existing opinions without any similar evidence to support it. If I'm not missing something that would seem rather arbitrary.

Mr. Kotter
07-28-2008, 10:29 AM
Normally, I'd endorse such a statement as being a reasonable standard for judgement. But you seem to have formed your existing opinions without any similar evidence to support it. If I'm not missing something that would seem rather arbitrary.

Voter turnout numbers have fluctuated 4-6% generally since 1968. However, "turnout" numbers prior to 1968 were highly suspect due to "vote early, vote often" exhortations of political bosses and....the number of "dead people" voting. Vote fraud and corruption have been substantially reduced since, making comparisons of turn-out today to time prior to 1968 unreliable, at best.

Hence, since 1968...turn-out has ebbed when voters are "energized" by some major issue or controversy (1972, Watergate and Vietnam; 1992, Clinton's appeal for "Change" and the economic doldrum of 1991 and early 1992; and 2004, hatred of W. and the war in Iraq) and it tends to decline in times of peace and relative prosperity (1988 and 1996 are clear examples.)

2008 will see another ebb (with Obama's cult of personality, economic difficulties, and moonbat hatred of W and Reps in general)....which should benfit Obama, unless the "silent majority" asserts itself again.

Direckshun
07-28-2008, 10:30 AM
It seems the Obama campaign agrees with jAZ here. They've dedicated themselves to registering 10 million new voters this year.

Calcountry
07-28-2008, 11:38 AM
The polls in this election mean nothing in this election and here's why...

Turnout.

Aside from the wording of questions, turnout is the component of election polling models that is most more art than science. It's the part that skews the numbers up or down.

I say this at a moment where I'm not dismissing the latest results because they don't favor Obama (according to the Gallup Daily Tracking Poll, Obama is up now up 9% after his trip from 2% before).

That poll means nothing.

The key is will Obama be able to turn out his voters and will McCain be able to turn out his voters?

Pollsters try to predict that turnout, but that is the weakspot.

My gut tells me that Obama will have a Bush04 like turnout machine and McCain will have difficulty driving turnout.

The question will be, will McCain be able to drive himself to a large enough overall lead as to offset that machine that Obama and the Dems are building. That's an odd question given the machine that Bush built in 04. It carried him to an improbable victory. But McCain won't be able to do the same thing to the same degree. I'm almost certain of that fact. How close can he get though.

As an FYI, one thing I read that might be helpful for those who care to use in interpreting polls right now. Because turnout is an unknown, it's really hard to accurately predict a "likely voter" in a poll, so "registered" voter results are slightly more trustworthy for the time being.One fly in your ointment. How many Kerry voters are going to vote for McCain this time?

Heck, they may even be telling pollsters that they will vote for Barry, no way they want to be seen as racist.

Then, in the privacy of the booth.....

markk
07-28-2008, 11:43 AM
Normally, I'd endorse such a statement as being a reasonable standard for judgement. But you seem to have formed your existing opinions without any similar evidence to support it. If I'm not missing something that would seem rather arbitrary.

i didn't claim there was statistical evidence and I didn't offer any. neither did you. i explained what I thought was the truth, and you replied with a lot of words but nothing in the way of reason to believe turnout varies more than a couple of percent.

i think if you did a regression you could very reasonably predict turnout through a simple fitting of population increases and the number of registered voters each cycle, with some kind of modifying factor for a re-election, where a sitting president is on the ballot.

Mr. Kotter
07-28-2008, 12:36 PM
It seems the Obama campaign agrees with jAZ here. They've dedicated themselves to registering 10 million new voters this year.

Registering them is one thing; getting them to the polls is another.

Obama's critics will be highly energized to defeat him though too; it will be a battle between Obamaniacs and the old-school "silent majority"....who succeeds best will likely win.

And we won't know until November. Polling is particularly vulnerable in this election because of the unknown impact of the Bradley Effect. Obama needs to be up by at least 4-6 points consistently through October and early November for his supporters to feel "safe."

jAZ
07-28-2008, 01:12 PM
Voter turnout numbers have fluctuated 4-6% generally since 1968. However, "turnout" numbers prior to 1968 were highly suspect due to "vote early, vote often" exhortations of political bosses and....the number of "dead people" voting. Vote fraud and corruption have been substantially reduced since, making comparisons of turn-out today to time prior to 1968 unreliable, at best.

Hence, since 1968...turn-out has ebbed when voters are "energized" by some major issue or controversy (1972, Watergate and Vietnam; 1992, Clinton's appeal for "Change" and the economic doldrum of 1991 and early 1992; and 2004, hatred of W. and the war in Iraq) and it tends to decline in times of peace and relative prosperity (1988 and 1996 are clear examples.)

2008 will see another ebb (with Obama's cult of personality, economic difficulties, and moonbat hatred of W and Reps in general)....which should benfit Obama, unless the "silent majority" asserts itself again.
Your break has served yo well. Good stuff.

jAZ
07-28-2008, 01:16 PM
i didn't claim there was statistical evidence and I didn't offer any. neither did you. i explained what I thought was the truth, and you replied with a lot of words but nothing in the way of reason to believe turnout varies more than a couple of percent.
I asked you to explain what you meant by (paraphrase) "not much". Because what you have in mind and what I have in mind might be the exact same thing, but you think a few % isn't much and I think it's a lot when elections are won and lost by 1-2% anymore.

As Kotter pointed out, turnout has swung by 4-6% over the years. And that's 4-6x the margin of victory in 2004 and even more than that in 2000. And greater than what I expect the margin of victory to be in 2008 for either McCain or Obama.

4-6% feels about right to me and my guess is that's about what you were thinking too. And that as I said before, we were talking about the same thing all along.

jAZ
07-28-2008, 01:37 PM
Registering them is one thing; getting them to the polls is another.

Obama's critics will be highly energized to defeat him though too; it will be a battle between Obamaniacs and the old-school "silent majority"....who succeeds best will likely win.

And we won't know until November. Polling is particularly vulnerable in this election because of the unknown impact of the Bradley Effect. Obama needs to be up by at least 4-6 points consistently through October and early November for his supporters to feel "safe."

I don't think we will feel safe at all. But aside from that, you and I seem to be in agreement.

markk
07-28-2008, 02:17 PM
I asked you to explain what you meant by (paraphrase) "not much". Because what you have in mind and what I have in mind might be the exact same thing, but you think a few % isn't much and I think it's a lot when elections are won and lost by 1-2% anymore.

As Kotter pointed out, turnout has swung by 4-6% over the years. And that's 4-6x the margin of victory in 2004 and even more than that in 2000. And greater than what I expect the margin of victory to be in 2008 for either McCain or Obama.

4-6% feels about right to me and my guess is that's about what you were thinking too. And that as I said before, we were talking about the same thing all along.

If you're asking how many people I think vote in a given election that have never voted before, I don't know. It'll be a single digit percentage, I think, probably in the low single digits.

Many of these people will have just reached voting age for this election and would have voted regardless in other years, or they had it last time but were away at college or something, were registered elsewhere, thus not voting. There's no need to look for a reason why those people voted.

The 'registered by a campaign registration drive' voter is not a major force in elections, except maybe in a few key areas. Although ACORN and their registration of pets and cadavers comes to mind.

Turnout is not equal to margin of victory. There's no way for you to correlate "turnout" in 2000 to margin of victory. There's no data available to you about who voted what way back then and if they were new or not, or how they have voted before or since. What meaningful way is there to study it other than notoriously flawed methods like exit polling? It's voodoo.

BucEyedPea
07-28-2008, 02:52 PM
It's everything, IMO.

I think that's why strategists on the GOP side are quietly very nervous. McCain needs a big lead to overcome Dem enthusiasm. McCain needs Clinton voters, some of which he can get.

But this year, it's not about the swing voters as much as the turnout machines.

Not mention Barr hurting Mac, especially in Georgia.

'Hamas' Jenkins
07-28-2008, 03:14 PM
Just keep pinning your hopes on Gay Divorce as a wedge issue, RWers.

SBK
07-28-2008, 03:25 PM
For all the polls jAZ always posts I found this thread to be funny when it's right above Jake's McCain up poll.

We're still 100 days out, nothing matters today. A LOT is going to happen between now and then......

jAZ
07-28-2008, 04:05 PM
For all the polls jAZ always posts I found this thread to be funny when it's right above Jake's McCain up poll.

We're still 100 days out, nothing matters today. A LOT is going to happen between now and then......

Just be be clear, check the dates/times. I'm not posting this as political cover. In fact, I posted it in the wake of good news for Obama.