If you were a bookie... Polls and bets on the 2004 elections

I don't trust anything with the Fox name on it.

It's as bad as worldnetdaily, commondreams, pick your poison.

Cycloptichorn

Don't get too excited Soz, for some reason that site's US map is using data from different polls on different states. For instance, it's using polls that have both Wisconsin and Florida going to Kerry. Those two alone would swing it the other way.

I recommend you check out Foxfire's RealClearPolitics link and check out what the variety of State Polls really say. It's searchable by state.

Bill,

They use the most recent polled data for each state on electoral-vote.com . .. It changes every day as more recent polls are used.

There's a nice explanation of the methodology inside if ya just look.

You may also note that they have Ohio going to Bush, when there are several polls out that show it the other way...

It's really too close to call right now.

Cycloptichorn

I've found Daly especially reliable/ bias-free. (Thanks nimh.) He has the current status 222 for Bush, 203 for Kerry, but 277 Kerry/ 254 Bush with tossups.

http://www.dalythoughts.com/ecb.htm

Too close to call to be sure.

John McIntyre of RealClearPolitics just said there's a possibility of a tie - 269 to 269. Yikes.

In which case each state would be given one vote and Congress would decide.

I've never bitten my nails....I may start before this is all over.

Wow.

I'm so curious about how the get-out-the-vote stuff will affect all of this. Pollsters are operating on a certain set of assumptions, and it seems possible that those assumptions will be upended. To simplify greatly, if they assume that 40% of voters are Democrats and 40% are Republicans, when in fact the significantly more new registered voters are Democrats and it turns out that say 45% of the voters are Democrats and 35% are Republicans, that skews a lot of things. (I doubt such a large discrepancy is possible, but when things are this close, any discrepancy could have a real effect.)

Sozobe - McIntyre has some thoughts on your exact question!

http://realclearpolitics.com/Commentary/blog_10_19_04_1148.html

Very interesting!

<Dang those pesky tossups!>

It wouldn't be the first time the polls (or the early vote count) were entirely wrong.

Hello. I was in Berlin for a festival/workshop for a few days, and at this time of the race, just 3 days off already gets you a backlog you cant catch up with. So forget it, you'll see all the new polls in my next graphs ;-).

I wanna make one exception tho, because it directly addresses Soz's point:


There was a Harris poll out this week (survey of Oct 14-17), and they took the unusual step of giving out numbers they arrived at using two different "likely voter" screens.

Using one definition of likely voters, they counted all 820 registered voters in the sample who said they were "absolutely certain" to vote. The result shows President Bush with a modest two-point lead:

Bush 48%
Kerry 46%

However, pollsters often go further in isolating likely voters in their samples. For example by excluding all those who were old enough to vote in the previous presidential elections yet did not do so. This because experience shows that those who didn't vote last time are much less likely to turn out this time even if they claim they are "certain" to vote. Harris notes that it's the stricter likely voter screen that "has proved more accurate in the past".

Applying it to its present sample, Harris came up with just 755 registered voters who said they were "absolutely certain" to vote and that (if old enough) they voted in 2000 - and the result has President Bush in a commanding eight-point lead.

Bush 51%
Kerry 43%

In the same write-up though, Harris does note that "there are some indications that in this election many people who did not vote in 2000" will turn out to vote - "in which case it would be wrong to exclude them".

755 or 820? Which number represents the true "likely voters" of these elections? Daly and another conservative blogger argued elaborately why it would be highly unlikely that all of those 820 (that is, the larger share of the electorate they represent) will turn out - unless turnout is higher than any time since 1972, is what the second guy calculated. Yet that there will be a substantial net extra number of people voting who didnt vote in 2000 seems to now be widely assumed. So will the truth turn out to be in the middle? (I'm going to use the average of the two Harris LV poll results above for my graph, in any case).

Bush will win with a 3-5% margin.

Kerry will win by 2%.

See, we can all make bold, useless predictions.

Cycloptichorn

At this juncture with the unusually large number of newly registered voters I doubt that anyone has a handle on how they will vote. They are the ones who will determine the shape of our next administration.

On another score, can I just say that I think Kerry stands little chance in Nevada - or even much-trompeted Colorado - and should get his ass back into West-Virginia and, who knows, perhaps even North Carolina?

In the past week, CO was polled four times. Bush had a lead of respectively 5%, 6%, 5% and 6%. NV was polled twice, showing solid Bush leads of 10% and 7% respectively.

On the other hand, West-Virginia was polled twice (Bush leads of 2% and 5%), Virginia once (Bush lead, 4%) and North Carolina twice: by Mason-Dixon (8% Bush lead) and SurveyUSA, which had Bush ahead by just 3%.

In fact, it's SurveyUSA who had Bush up by 11% in Arizona and 7% in Nevada, and only 3% in NC, 4% in Virginia, 5% in Arkansas and 6% in Missouri (with Kerry up 1% in Florida, 2% in Ohio and 6% in Pennsylvania).

There's nothing wrong with trying to turn NH and ME into blue states once and for all. But the "Western strategy" the Kerry team is now chasing (NV + CO), while mostly giving up on the margins of the South (WV, VA, NC, MO, AK) is IMHO very tricky.

Apart from being tricky, it implies a vision of the party's future I dont much like. Ceding the populist South to the right, while instead carving out a new blue-state majority by 'annexing' states that border existing blue states in the north and west. Shouldn't cede the culture war.

Welcome back nimh! Missed ya.

Some are bolder - and more useless (i.e. less supported by the data reported here) than others.

It all depends on the data you look at, George.

The current polls showing Bush ahead are flawed in many, many ways. We could argue back and forth about it, but why bother? You'll know one way or the other in two weeks.

Boy that date sure snuck up on us...

Cycloptichorn

PolySays offers an interesting observation on the relationship between election outcome and Gallup's polling figures:



This year, as reported by USA Today, the Sept 24-26 call was Bush 52%, Kerry 44%. The October 14-16 call was the same, Bush 52%, Kerry 44%.

Not much joy for The Kerry Camp today at RealClearPolitics, where both the 3-way and the Head-to-Head show Bush +2.9, or at PollingReport

(click graphic for details.)

Edited to correct PollingReport Graphic link; they changed it this afternoon

A strident minority: anti-Bush US troops in Iraq

Though military personnel lean conservative, some vocally support Kerry - or at least a strategy for swift withdrawal.

By Ann Scott Tyson | Correspondent of The Christian Science Monitor

WASHINGTON – Inside dusty, barricaded camps around Iraq, groups of American troops in between missions are gathering around screens to view an unlikely choice from the US box office: "Fahrenheit 9-11," Michael Moore's controversial documentary attacking the commander-in-chief. "Everyone's watching it," says a Marine corporal at an outpost in Ramadi that is mortared by insurgents daily. "It's shaping a lot of people's image of Bush."

The film's prevalence is one sign of a discernible countercurrent among US troops in Iraq - those who blame President Bush for entangling them in what they see as a misguided war. Conventional wisdom holds that the troops are staunchly pro-Bush, and many are. But bitterness over long, dangerous deployments is producing, at a minimum, pockets of support for Democratic candidate Sen. John Kerry, in part because he's seen as likely to withdraw American forces from Iraq more quickly.

"[For] 9 out of 10 of the people I talk to, it wouldn't matter who ran against Bush - they'd vote for them," said a US soldier in the southern city of Najaf, seeking out a reporter to make his views known. "People are so fed up with Iraq, and fed up with Bush."

With only three weeks until an Oct. 11 deadline set for hundreds of thousands of US troops abroad to mail in absentee ballots, this segment of the military vote is important - symbolically, as a reflection on Bush as a wartime commander, and politically, as absentee ballots could end up tipping the balance in closely contested states.

It is difficult to gauge the extent of disaffection with Bush, which emerged in interviews in June and July with ground forces in central, northern, and southern Iraq. No scientific polls exist on the political leanings of currently deployed troops, military experts and officials say.

To be sure, broader surveys of US military personnel and their spouses in recent years indicate they are more likely to be conservative and Republican than the US civilian population - but not overwhelmingly so.

A Military Times survey last December of 933 subscribers, about 30 percent of whom had deployed for the Iraq war, found that 56 percent considered themselves Republican - about the same percentage who approved of Bush's handling of Iraq. Half of those responding were officers, who as a group tend to be more conservative than their enlisted counterparts.

Among officers, who represent roughly 15 percent of today's 1.4 million active duty military personnel, there are about eight Republicans for every Democrat, according to a 1999 survey by Duke University political scientist Peter Feaver. Enlisted personnel, however - a disproportionate number of whom are minorities, a population that tends to lean Democratic - are more evenly split. Professor Feaver estimates that about one third of enlisted troops are Republicans, one third Democrats, and the rest independents, with the latter group growing.
Pockets of ambivalence

"The military continues to be a Bush stronghold, but it's not a stranglehold," Feaver says. Three factors make the military vote more in play for Democrats this year than in 2000, he says: the Iraq war, Defense Secretary Donald Rumsfeld's tense relationship with the Army, and Bush's limited ability as an incumbent to make sweeping promises akin to Senator Kerry's pledge to add 40,000 new troops and relieve an overstretched force.

"The military as a whole supports the Iraq war," Mr. Feaver says, noting a historical tendency of troops to back the commander in chief in wartime. "But you can go across the military and find pockets where they are more ambivalent," he says, especially among the National Guard and Reserve. "The war has not gone as swimmingly as they thought, and that has caused disaffection.

Whether representing pockets of opposition to Bush or something bigger, soldiers and marines on Iraq's front lines can be impassioned in their criticism. One Marine officer in Ramadi who had lost several men said he was thinking about throwing his medals over the White House wall.

"Nobody I know wants Bush," says an enlisted soldier in Najaf, adding, "This whole war was based on lies." Like several others interviewed, his animosity centered on a belief that the war lacked a clear purpose even as it took a tremendous toll on US troops, many of whom are in Iraq involuntarily under "stop loss" orders that keep them in the service for months beyond their scheduled exit in order to keep units together during deployments.

"There's no clear definition of why we came here," says Army Spc. Nathan Swink, of Quincy, Ill. "First they said they have WMD and nuclear weapons, then it was to get Saddam Hussein out of office, and then to rebuild Iraq. I want to fight for my nation and for my family, to protect the United States against enemies foreign and domestic, not to protect Iraqi civilians or deal with Sadr's militia," he said.

Specialist Swink, who comes from a family of both Democrats and Republicans, plans to vote for Kerry. "Kerry protested the war in Vietnam. He is the one to end this stuff, to lead to our exit of Iraq," he said.

'We shouldn't be here'

Other US troops expressed feelings of guilt over killing Iraqis in a war they believe is unjust.

"We shouldn't be here," said one Marine infantryman bluntly. "There was no reason for invading this country in the first place. We just came here and [angered people] and killed a lot of innocent people," said the marine, who has seen regular combat in Ramadi. "I don't enjoy killing women and children, it's not my thing."

As with his comrades, the marine accepted some of the most controversial claims of "Fahrenheit 9/11," which critics have called biased. "Bush didn't want to attack [Osama] Bin Laden because he was doing business with Bin Laden's family," he said.

Another marine, Sgt. Christopher Wallace of Pataskala, Ohio, agreed that the film was making an impression on troops. "Marines nowadays want to know stuff. They want to be informed, because we'll be voting out here soon," he said. " 'Fahrenheit 9/11' opened our eyes to things we hadn't seen before." But, he added after a pause, "We still have full faith and confidence in our commander-in-chief. And if John Kerry is elected, he will be our commander in chief."

Getting out the military vote

No matter whom they choose for president, US troops in even the most remote bases in Iraq, Afghanistan, and elsewhere overseas are more likely than in 2000 to have an opportunity to vote - and have their votes counted - thanks to a major push by the Pentagon to speed and postmark their ballots. The Pentagon is now expediting ballots for all 1.4 million active-duty military personnel and their 1.3 million voting-age dependents, as well as 3.7 million US civilians living abroad.

"We wrote out a plan of attack on how we are going to address these issues this election year," says Maj. Lonnie Hammack, the lead postal officer for US Central Command, an area covering the Middle East, Central Asia, and North Africa, where more than 225,000 troops and Defense Department personnel serve.

The military has added manpower, flights, and postmark-validating equipment, and given priority to moving ballots - by Humvee or helicopter if necessary - even to far-flung outposts such as those on the Syrian and Pakistani border and Djibouti.

Meanwhile, voting-assistance officers in every military unit are remind- ing troops to vote, as are posters, e-mails, and newspaper and television announcements. Voting booths are also set up at deployment centers in the United States.

"We've had almost 100 percent contact," says Col. Darrell Jones, director of manpower and personnel for Central Command, and 200,000 federal postcard ballot applications have been shipped.

"We encourage our people to vote, not for a certain candidate, but to exercise that right," he said, noting that was especially important as the US military is "out there promoting fledgling democracy in these regions." Many of the younger troops may be voting for the first time, he added.

"The military continues to be a Bush stronghold, but it's not a stranglehold," Feaver says.

"The military as a whole supports the Iraq war," Mr. Feaver says, noting a historical tendency of troops to back the commander in chief in wartime.

A recent survey showed military 4:1 for Bush over Kerry.

http://www.usatoday.com/news/politicselections/nation/president/2004-10-03-bush-troops_x.htm

Advantage: Bush

Marist Poll, Oct 17-19, compared to two weeks ago:

Likely voters:
Bush 49% (no ch.)
Kerry 48% (+2)

Registered voters:
Bush 47% (no ch.)
Kerry 47% (+1)

Bush job approval
Approve 49
Disapprove 48

Link

AP/Ipsos poll, Oct. 18-20, compared to two weeks ago

Likely voters
Bush 46% (no ch.)
Kerry 49% (-1)

Bush job approval 47%

Link