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

I'd give it fifty fifty odds, because those are in fact the odds when you consider the bias toward Bush in the media who pays the majority of the pollsters, and if you consider the small margin of the popular vote that comprised the popular win by Gore.

I know one friend who didn't vote last time who says he will vote for Kerry. Based on his past apathy I'd give a slight edge to Kerry. But a lot can happen between now and November. It would have to be something pretty dramatic to tilt the balance to the satisfaction of an odds maker.

If it was my "business" I might well tell you to take your bet elsewhere - like how about the voting booth?

0 Replies

Cant go to your voting booths, padmasambava ... alas ;-).

This one just for the record:

In 2000, "Zogby, CBS, Harris, Gallup, Pew and IBD/TIPP all had both candidates' totals correct within the margin or error."

From a blog post, mind you, didnt go back to check, but there it is.

Odd thing is that this time, they have such different views on the race. How likely is it that, say, Zogby - who was the second closest to the actual outcome in 2000 I believe, has it right again now, but CBS and Gallup, which were very close too, now are suddenly wildly off? Or vice versa?

0 Replies

I think that'd be referrin' to the final poll before the election, nimh ... most of 'em narrow the gap between perception and reality a bit as October wanes and November draws nigh. Another interestin' phenomenon is that in general, the Democratic Candidate's polling results, regardless of eventual winner, tend to be a bit higher than the actual Election Day Vote. Yet another interestin' phenomenon is that since the inception of the Gallup Poll, the Gallup Labor Day Leader has won 17 of the 19 covered contests, only Truman in '48 and Reagan in '80 marking exception to the rule. Incumbent Truman overcame a nearly 6 point Labor Day deficit to narrowly defeat media-favorite Dewey, while challenger Reagan and incumbent Carter were virtually tied 39/38 respectively in that yerar's Labor Day Poll, with 3rd-Party candidate Anderson polling in the low teens. Carter's approval rating at the time was the lowest Gallup had ever measured for a sitting president.

"For the record", History is not on Kerry's side.

0 Replies

timberlandko wrote:

Another interestin' phenomenon is that in general, the Democratic Candidate's polling results, regardless of eventual winner, tend to be a bit higher than the actual Election Day Vote.

Wasn't true last time with Gore tho, I think? He did better than most polls had him, I seem to remember ...

0 Replies

The 2000 election was exceptional in many particulars, nimh, that Gore's polling numbers did not conform to precedent being only one of several.

0 Replies

Ryan Lizza's take in TNR's Campaign Journal: "Bush is vulnerable, but Kerry is drowning."

Quote:

The new Newsweek poll shows Bush's bounce dissipating, but the new Time poll doesn't. Meanwhile, polling from Zogby, AP, and Democracy Corps show Bush up by two to five points. All the surveys below but Newsweek are for likely voters.

The three-way trial heats show an average Bush lead of six points. But note that the Time poll, which shows an eleven-point Bush lead and is already getting the most media coverage, seems to be an outlier.

<snipping data table cause I dont feel like complicated formatting: Newsweek, Time, Zogby, AP and Dem Corps>

The two-way trial heats show an average Bush lead of three points. (Time and the AP didn't report two-way numbers.)

<snipping data table with Newsweek, Zogby and Dem Corps>

Bush's gains have been led by women, independents, and younger voters. The general trend in most of the recent data is a move away from the economy and towards terrorism as the most important issue. There's also movement away from the importance of specific issues and towards intangibles like character and leadership, Bush's greatest strengths. That's the major reason for the Democratic attacks on Bush's Guard service.

Next week's polls will give us a better idea of how fleeting Bush's gains are, but history shows his lead should start to settle into the range of two or three points.

He seems to be holding on by the skin of his teeth. His average approval rating since the convention is 50.7 percent. To me, the race looks like two men wrestling in the ocean. Right now, Bush is treading water with his head barely above the surface. But he's got his hands wrapped around his opponent's neck, and Kerry can't seem to come up for air or yank Bush down. Bush is vulnerable, but Kerry is drowning.

posted 1:06 p.m.

0 Replies

So, apparently (I gather from the above), there's new Time and Newsweek polls out. Damn.

Those people are gonna mess up my graph system, do they know that? They had one just last week! What am I gonna do, huh? I dont wanna be taking averages of subsequent polls, like I've had to do one or two times in the job approval graph (the times this came up in the horserace thus far, it usually involved one poll that actually took place over the border of one period into the next, say: Aug 31-Sep 2, and I could "cheat" a little and shift it into the other period concerned.) But now?

Gotta rethink.

Meanwhile, here's the numbers, sorry. Note that they starkly disagree about whether Bush's post-Convention bounce is solidifying or being halved:

Time Poll , Sept. 7-9, 2004 (compared to last week):

2-way, likely voters:
Bush 54 (+1)
Kerry 42 (-1)

2-way, registered voters:
Bush 51 (+1)
Kerry 39 (-3)

<gasp - Thirty-nine percent? Note also that Bush is leading by as much among registered as among likely voters now.>

3-way, likely voters:
Bush 52 (no ch.)
Kerry 42 (no ch.)
Nader 3 (no ch.)

3-way, registered voters:
Bush 50 (+1)
Kerry 39 (-1)
Nader 4 (-1)

Newsweek Poll, Sept. 9-10, compared to last week):

2-way:
Bush 50 (-4)
Kerry 45 (+2)

3-way:
Bush 49 (-3)
Kerry 43 (+2)
Nader 2 (-1)

The Newsweek poll has specifics.

- Kerry gets just 4% of Republicans, and Bush only 7% of Democrats. Independents lean to Kerry, 45% to 39%, with 7% (!) for Nader. (Wonder how many of those will go which way before the elections ...)

- Bush has an identical 49% to 43% lead among both men and women (that must be a first, all I've seen until the Convention was a huge gender gap).

- Non-military households are evenly divided with 46% for each, while military households prefer Bush 55% to 36%.

- A bit of a seesaw in the age groups. Kerry holds on to a 3% lead among the 18-29s and the 50-64s are evenly divided at 46%. But those aged 30-49 go to Bush by 51% against 41% and those over 65 by 50% to 40%.

0 Replies

Anyhow, this is what the horserace graph (only counting two-way race polls) looked like before these latest Time and Newsweek polls:

http://home.wanadoo.nl/anepiphany/images/bush-kerry_sept2.gif

Note how the eight post-Convention polls seem to divide up into two equal groups of four:

- Time, Newsweek, CNN/Gallup and CBS have a Bush lead of 7-11%;
- Fox, Zogby, ICR and Democracy Corps have one of merely 1-3%.

The CNN/Gallup poll drastically falls from the first to the second category if you look at their registered voters numbers. But the 3-way-only ABC/WaPo poll goes up in the first category, while the 3-way AP-Ipsos poll falls either right in the middle (likely voters) or also up in the first category (registered voters).

OK, now hold on to your seats here, folks, here's the average graph of up to last night:

http://home.wanadoo.nl/anepiphany/images/bush-kerry_average_beginsept.gif

Now thats what I call a bounce ...

0 Replies

And this is the current look of the Bush job ratings graph ... if thats gonna be a star on top of the Christmas tree, its a mighty big star ...

0 Replies

Nimh - thanks for all the hard work. This is the first thread I check. Just wanted you to know it's appreciated. Smile

0 Replies

No problem, JW.

Next, a surprise, boys and girls. New poll by Investors Business Daily / Christian Science Monitor / TIPP. Sept 7-12. They're one of the polls that had the 2000 "totals correct within the margin or error."

And they have Kerry in the lead.

To be more exact (I admit posting the above for effect), in their likely voter sample they have Bush and Kerry tied, at 47% each. In the registered voter sample they have Kerry in the lead by 46% to 44%.

Amazingly though, this means that the latter shows a bounce (if only a 2% one) for Kerry since their last poll on August 17-23, before the Republican Convention. (No bounce for the likely voter sample cause they didnt specify one before.)

Why the unusual result? One partial explanation could be that the IBD/CSM poll doesn't push those who do not immediately indicate a preference for an answer as much. So, in the Newsweek, AP-Ipsos, Time, ABC/WaPo, CNN/Gallup (etc) polls, you have a follow-up question to those who dont know whom to vote for yet, asking which candidate they lean to, at least. Like Fox (which also showed only a small Bush bounce), IBD/CSM doesn't ask this follow-up question. So they often have a higher percentage of undecideds.

It suggests that part of the Bush bounce in the other polls was thanks to people who still haven't made their mind up, but after the Convention at least leaned to Bush. Thats good news, cause they can more easily be swayed back again too.

Now, if Kerry gets his campaign in order still ...

0 Replies

Quote:

They're one of the polls that had the 2000 "totals correct within the margin or error."

Which others did, as well? Just wondering.

0 Replies

nimh wrote:

Add another one, only 3-way numbers I'm afraid (so it wont show up in my graph):

AP-Ipsos poll, out today (done Sept 7-9):

Among likely voters (no comparative previous numbers available):
Bush 51%
Kerry 46%
Nader 1%

Among registered voters (compared to early August):
Bush 51% (+6)
Kerry 43% (-5)
Nader 2% (-1)

(Odd, in this poll Bush does better when a wider sample is used; usually (see above) it's the other way around.)

This one needs to be corrected. On realclearpolitics.com it still says 51 - 46, but since Dales mentioned other numbers I went back to pollingreport.com and there it now says:

Sept 7-9

Among likely voters:
Bush 52
Kerry 43
Nader 2

Among registered voters:
Bush 51 (+6)
Kerry 43 (-2)
Nader 2 (-1)

Dunno what went wrong there earlier.

0 Replies

JustWonders wrote:

Quote:
They're one of the polls that had the 2000 "totals correct within the margin or error."

Which others did, as well? Just wondering.

Zogby, CBS, Harris, Gallup and Pew, apparently.

0 Replies

A whole page of self congratulation.

A Time poll. Hmmm. That ought to be really unbiased.

Lou Harris lives in the same building as the holdout juror in one of the recent anti trust suits. . .

Coincidence? Maybe, but I'd guess it's a "meaningful coincidence."

Don't count your chickens before they're hatched.

0 Replies

For a bit of polling fun, Kerry narrowly beats Rush Limbaugh, ties Martha Stewart and Joseph McCarthy:

Quote:

... In last week's Washington Post-ABC News Poll, John F. Kerry was viewed favorably by 36 percent of registered voters, down 18 points over the past six months. But just how low Kerry's standing has fallen cannot be appreciated fully without comparing his standing with that of other household names in Gallup polls over the years. Kerry finds himself in a dead heat with Martha Stewart, and behind Herbert Hoover -- although he narrowly beats O.J. Simpson.

Michael Jordan: 83 (2000)

Tony Blair: 76 (2003)

Pope John Paul II: 73 (2003)

Democratic Party: 54 (2004)

John Ashcroft: 49 (2003)

Michael Dukakis: 47 (1988)

Prince Charles: 45 (2003

Herbert Hoover: 43 (1944)

Jesse Jackson: 38 (2003)

Vladimir Putin: 38 (2003)

John Kerry: 36 (2004)

Martha Stewart: 36 (2004)

Joseph McCarthy: 35 (1954)

Rush Limbaugh: 34 (2003)

Pete Rose: 34 (2004)

O.J. Simpson: 29 (1995)

Osama bin Laden: 1 (2001)

0 Replies

Touche Timberland. Cant help but grin at the bizarreness of that. How did Bush rank in this list, by the way?

From the same article, by the way (off-topic), in the paragraphs directly below:

Quote:

"In an unprecedented show of support for the Republican Party and President Bush, 18 percent of the 2004 delegation identify themselves as veterans or active military personnel," the Republican National Committee proclaimed on Aug. 16. It added that "approximately 140 delegates identify themselves as currently serving in the U.S. military."

That would indeed be an unprecedented show of support. Pentagon Directive 1344.10, dated Aug. 2, 2004, states that active-duty military personnel may not attend political conventions "unless attending a convention as a spectator when not in uniform." Serving as a voting delegate would appear to go beyond the "spectator" role that is permitted and into the prohibited realm of "participation."

An RNC spokesman said the "active military personnel" actually referred to reservists and National Guardsmen, who are not on active duty.

0 Replies

In contrast to nimh, I find the implication of Timber's list clear and straightforward. Tony Blair should run for president of America. This whole independence thing was a silly idea to begin with.

0 Replies

Overall the list speaks quite well of the discretion of those polled. Too bad Chirac was not included on the list.

If the Democrats chose Blair the Republicans would have to choose between Michael Jordan and the Pope.

Perhaps the revolution was a good idea after all.

0 Replies

Good news - bad news - good news - bad news ... the hot/cold bath threatment continues.

ICR/International Communications Research poll. Sept. 8-12, 2004. Numbers compared to September 1-5 (that was - during or directly after the convention?)

2-way, likely voters
Bush 52 (+4)
Kerry 44 (-3)

2-way, registered voters
Bush 49 (+3)
Kerry 45 (-2)

3-way, likely voters
Bush 51 (+5)
Kerry 44 (-2)
Nader 3 (-1)

3-way, registered voters
Bush 48 (+5)
Kerry 44 (-1)
Nader 3 (-2)

0 Replies

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

Related Topics

Quick Links

My Account

able2know