« Fox News Poll: WH Race in Statistical Tie | Main | Kerry Up 1 in NH See-Saw Race »

Kerry Gains on Bush in Time Poll

Bush leads Kerry 48-44 among nation-wide RV's in a new Time magazine Poll conducted 9/21-23. (Bush was up by 12 among RVs in Time's poll 2 weeks earlier).

Comments

It boggles my little mind. Who are they polling? I can't believe that Bush fatigue has simply vanished.

I don't believe that Nader has support as high as 5%, which makes me disbelieve the rest of the poll.

BTW, wasn't this Time poll something we were completely trashing a few weeks ago? Why would they be right now if they weren't then?

This time: Likely voters reported party identifications are: 36% Republican, 31% Democratic, 24% Independents. Registered voters party affiliations are: 36% Republican, 30% Democratic, 24% Independent.

The 9/10 poll: Likely voters reported party identifications are: 34% Republican, 35% Democratic, 22% Independents. Registered voters party affiliations are: 31% Republican, 32% Democratic, 26% Independent.


Am I missing something? More Republicans, fewer Dems and the margin is still cut in half.

Now the questions are: Will the media report this with the same zeal as they did the polling data of two weeks ago? What will the administration do to divert attention away from a "come back"-oriented story line in the press?

Also, an interesting story in the NYTimes about the onslaught of new voters registering in key states:

http://www.nytimes.com/2004/09/26/politics/campaign/26vote.html?hp

Now...will they make it to the polls? Will the polls accept them?

Hey, I have a question for Ruy.

Over the last few weeks, you have written about popular polls, such as Gallup/CNN and said that they have oversampled Rs as a percentage of their sample population.

In the recent days, these polls have been tightening showing a closer race with Bush still in the lead, does this mean that in a sample that is overly populated with Republicans, Bush is ahead only by a small margin? Or have they changed their samples to include more Ds?

I think the implication here is obvious, so Ruy please explain this before I get giddy and wrongly warm myself to the thoughts of an election day landslide for Kerry!

QUESTION FOR RUY (or anyone else)

What do you make of the NYTimes article in Sunday's pages that claims groups like ACT, etc., are kicking so much butt registering new Democratic voters. If that story is true -- and the Times seems to have done exhaustive number crunching -- then it would suggest that not only are the party ID weights misguided but that they ought to be weighted even more toward Demos than in 2000.

Is it possible that the 2004 election -- with all the newly registered voters, with the gazillions of Bush-haters who don't have a landline to answer phone polls, with so many young voters outraged after seeing Fahrenheit -- will completely change the polling paradigm? It's the TIVO election, in so many ways. People pay attention to what they want to pay attention to, only answer the phone if someone they know is calling and do not fall into any currently identifiable niches.

It just seems like the pollsters are using 20th Century methods to figure out the first race of the 21st Century.

Sep 25

Rasmussen today has Bush only 0.6 ahead in the horserace. FL Bush +1, OH +3, PA +1, MN and NV tied. Kerry leads in MI.

And the voter reg numbers in the NY Times story is very heartening indeed.

I don't see much data to suggest either giddyness or gloom. The election is still several weeks away. The dynamics will change in some way during this time. Perhaps multiple times. Be patient and effortful.

In that light, I do offer some promising information. I've been worrying about Pennsylvania. Using race2004.net for my state poll source, several polls right after the Republican convention mostly showed Bush in the lead. In polls dated at race2004 from 9/7-/9/15, [and not including polls from GOP or Dem organizations] Bush is up by 4% (Quinnipiac), 3% (TNS), 1% (Rasmussen), and 1% (Gallup). The race was tied per Keystone. Kerry was up in just 1 of these 6 polls (by 2%, Survey USA).

In the last 6 polls, 9/16-24, Kerry has been in the lead in 5. 3% for Opinion Dynamics and Zogby, 2% for Temple, and 1% for ARG and Mason Dixon. Bush led by 1% in the Rasmussen poll.

Kerry has to hold Pennsylvania.

And then there's Ohio. In three of the 6 polls from 9/7 to 9/18, Bush had leads from 7-11%. Overall, these 6 averaged (unweighted) to a 5.5% /Bush lead. The last three polls have been closer, with Bush up by 4% (Opinion Dynamics), 3% (Rasmussen), and 2% (ARG).

Kerry needs to find a way to pull out Ohio. If he can get Ohio and hold all the Gore 2000 states, that would give him a 280-258 win. Kerry could then withstand losing Wisconsin and pull out a 270-268 win. [If that one Maine district were also to go to Bush, Kerry would then need to pull out one other small state, such as New Hampshire.]

There are other options. And this "only pick up Ohio" option would require Kerry either rallying from behind (2-6% in polls of the last week) in Iowa or pulling out two other small states such as New Hampshire (tied in the only poll from last week), Nevada (down 2% and 9% in the two polls from last week), or Arkansas (no polls last week, Kerry down 3% or tied the week before).

This is a very close thing this year. Keep focused and keep at it. And keep your fingers crossed for Kerry to do well this Thursday. Does anyone know how his cold is?

Reminder on Posting Policy

EDM does not host comments that attack other participants as individuals, rather then ideas. This is true even if the personal material represents only part of the content.

Posts that do not appear can be revised to remove such personal material and resubmitted.

This comments section is open to all points of view so long as the commenters discuss issues and ideas and express themselves in civil form.

EDM is currently evaluating other forum systems that allow for the creation of specific communities of interest (e.g. activist democrats). Look for more information after the election.

EDM Staff

Time Poll Sept. 21-23, 2004. N=1,014 registered voters nationwide, 877 likely voters.

I will now demonstrate how flawed this poll is.

According to the Time Poll, Bush has an edge of 54-42 among likely voters.

The pool of "likely voters" is 877 voters. 54% of that number is 474 "likely Bush Voters."

Compare the pool of 1014 "registered voters." 48% of that number is 487 "registered Bush Voters."

So, we now know that under the Time poll, 474 likely Bush voters exist among 487 registered Bush voters.

In other words, Time is reporting that 97.3% of the registered Bush voters will vote.

Now compare the way they downsize the Kerry numbers:

Kerry has 44% of 1014, which is 446 registered voters.

But he only gets 42% of the 877 "likely voters" or 368 voters.

In other words, Time is reporting that only 82.5% of registered Kerry voters will vote.

To summarize, using their "likely voter" spin, they eliminated only 13 Bush voters, but threw out 78 Kerry voters to create a 54-42 split. That isn't science, it's skullduggery.

As was posted more extensively under the CBS poll story, the question will be how many of these new voter registration forms will be accepted by the Republican secretary's of state in these battleground states, and how many will be thrown out for various technicalities like failing to check a box that you are over 18, even though you have provided your date of birth on another line. I think the Republican operative quoted in the NYT registration story has already previewed the upcoming Republican line for when these new voters are turned away at the polls on Nov 2: that the progressive voter registration groups did "sloppy work" which required all those nice new voter registrations to be thrown out by the poor secretarys of state, what a shame, etc. etc. Our opponents are quite determinedly anti-democratic and will stop at nothing to win.

BTW, wasn't this Time poll something we were completely trashing a few weeks ago? Why would they be right now if they weren't then?

No Gollum,
Ruys prime directive is to compare "apples to apples" - even if those apples have worms in them.

It's an interesting analysis Gabby. I would tend to buy high 80s or even low 90s for the Bush voters. Somewhere in the neighborhood of 80-90% of registered voters do turn out to vote in a typical presidential election and it's not uncommon to see slightly higer turnout for Republicans than Democrats (otherwise Democrats would win every time). But 97% does seem a little over the top.

While we're on the subject, does anyone have any idea why Marist College's LVs always seem to number <70% of their RVs?

It would appear that Scott Rasmussen and Ruy Tiexera agree in principle of the RV/LV - 2-way/3-way thing.

http://www.capitolhillblue.com/artman/publish/article_5283.shtml

Nice observation by Gabby Hayes, sidekick to Roy Rogers whom we all love and remember dearly. It's clear from analyses here and elsewhere that these media-bought polls are cooking the books. It's beginning to look like deja vu all over again vis a vis 2000 with polls showing Bush up several points just days before the election. Anyone for an October surprise?
Now we have destroyer fleets of the coast of N Korea. Dare anyone suggest Bush/Rove might be manufacturing a "crisis" or "imminent threat of ICBMs from NK" ?
Film at 11.

To me the most important part of that poll was not that Kerry cut Bushes lead in half, it was two of the other findings; A majority think that Iraq has made the world less safe and 60% of likely voters find Kerry likable.

Those two things tell me that we can win.

There is a HUGE increase in new voters in both Ohio and Florida according to the New York Times. I live in Ohio and local news media is also reporting a huge increase in new voters in Democratic areas of Ohio. Will polls catch those new voters? How do polls handle situations where there is volatility in the numbers of registered voters?

Ethan, you hit upon something that I'd been mulling over myself for a while. If the popular Time/Gallup polls now show the race tightening (which they do) and they still oversample Republicans (which they do) then there can be only one conclusion:

Kerry pulls off an upset in a few "red" states.

Not to jinx it, but look for Kerry wins in almost all the "true battleground" states, plus maybe even a Kerry upset in a "close red" state (like VA, NC, or AR).

Plus the NY Times had an article showing registration in Dem-leaning counties was 6 times the registration jump in Rep-leaning counties in OH, and with a similar Dem registration lead in FL.

And Bush is now on the defensive, having to explain his rosy Iraq comments to an increasingly skeptical public.

Sep 26
Rasmussen has a Bush only +1 in the horserace.
FL Kerry +1; MI K +4; MN tied; OH Bush + 4; NV tied.
A trend? We'll see. Rove is dragging his feet on debate #1. Trying to raise expectations? Who would imagine?

Regarding the comment above about the GOP skullduggery in rejecting new voter registrations....

OK, I agree that the GOP will stop at nothing to win this. But, at least in Pennsylvania, when you register to vote it's not a mystery whether or not your application has been accepted or not. You get a voter registration card in the mail. So it's not like you show up on election day and they can say, "Oops, sorry. You're not on the rolls."

OK...they're Republicans...they cheat...so they COULD say that. But people who have just registered need to follow-up to make sure their applications went through and thus minimize the likelihood.

My biggest concern is the one someone mentioned: The lengths to which Republicans will go to reject Democratic voter registration. Republicans are notorious for doing such things to gain advantage, since they know they can't win heads up.

El Guapo, you were great in THE THREE AMIGOS.

To win, Bush must carry Ohio. Similarly, Kerry must take PA. to win. The question then might be, is Ohio more within the reach of Bush than PA is within the reach of Kerry?

Seems to me that Kerry must put more resources into PA to assure a win. Once he's sure that the Keystone State is safely locked up, he can make his move on Ohio. The sooner he locks PA up the better his chances overall.

CAL D said:
Somewhere in the neighborhood of 80-90% of registered voters do turn out to vote in a typical presidential election...

ACTUALLY
It may be true some places, but not in most states. In 2000, the US composite mean was ~67.5% of registered voters turing out to vote.

2000 HIGHS
Wyoming 97.1% (¿error?)
Oregon 78.9
Connecticut 77.9
Maryland 76.7
Colorado 76.6

2000 LOWS
Indiana 54.5%
Oklahoma 55.3
D.C. 57
Mississippi 57.1
Hawaii 57.7

SOURCE: http://www.fec.gov/pages/2000turnout/reg&to00.htm

So those numbers are a little fishier than you were givin' 'em credit for, CAL D.

NEWSFLASH: BATTLEGROUND POLL - GWB 51%, KERRY 45%, GWB JOB APPROVAL 53%...

http://www.tarrance.com/Week2charts.pdf

Kerry is also behind 20% in fighting terrorism and Iraq; It's tough for Kerry to win facing numbers like that.

I see many posters here are getting excited about the new Dem registration being reported by the NY Times in FLA & OH, and how these new Dems are not being captured in surveys. I have news for you all: The NY Times, like CBS, is in the tank for Kerry, and is trying desperately to salvage his candidate, especially since all their Abu Gharib, Richard Clarke, etc., stories couldn't do the trick.

They also have reported very little on all the stories of possible vote fraud being perpetrated by Dems - ie 20,000 with dual NY & FLA voter registration reported by the NY Daily News a few weeks back, dead Dems in WIS with voter registration cards, et al.

They also do not fully explain how this phenomena is not the typical voter registration churn, or capture the intensity and enthusiasm likely to occur among Reps this year because of anger over the CBS fraud, the over the top hatred of President Bush, Kerry's trashing and unpopularity with the military, etc.

My advice to Kerry supporters: Believe the NY Times at your peril.

The Time poll is the best news I've read in awhile! 36R - 30D among RVs and the gap is closing. More important is Kerry's attack on Bush's Iraq policy is clearly hitting. The idea of a Bush-Iraq credibility gap is a powerful one. Hopefully Kerry can drive that point home in the debates.

BTW, this Will Saletan article on debate tactics is great. Saletan is a centrist who probably understands swing voters better than most of us.

http://slate.msn.com/id/2107141/

Elrod,

Sorry to disappoint you but insofar as GWB is concerned, Saletan is no centrist - Indeed, he is a card carrying member of the Michael Moore (FahrenHype 911), Whoppi Goldberg (Kiss my ___), Terry McCaulliffe wing of the Dem party - In other words, he's a Bush Hater in sheep's clothing.

As for your supposition that Kerry's attacks are resonating, reasonable people will disagree - Especially when you consider the battleground poll released last night showing he's trailing GWB by 6 points.

Frankly, any benefit from his anti-Iraq War speeches was mitigated by his condescending and haughty attack on the very courageous PM of Iraq; Who couldn't even thank us without having to see Kerry's mug & gait flailing away on a subject which he knew very little about other than what sees on CBS and in the NY Times.

Oh, the selected prime minister of iraq comes here and in effect tries to interfere with the election. he lies about what is happening in Iraq and the cons are outraged at Kerry? Excuse me, boys! The one to be outraged at are the liars, bush and alawi. And as i recall newsweek and cnn both use gallup poll numbers. that has been shown to be out of touch with reality. so where is the real discussion? where is the discussion about over 1000 dead in iraq? the WMD? the missing osama? the deficit? What in the world is wrong with you conservatives(not anappropirate name) more like big spenders? You attack the truth teller and applaud the liar. what is is the night gallery in real time? bush is a liar and a loser. he is losing the election. they lied to you about iraq. the media kissed up and carried it along. now why is there any question that the media is not doing it again. i would assume they were without any real indication they had reformed. duh!

Reference the New York Times article on voter pick up in Ohio: the same information is also being reported by media outlets in Ohio, not a hotbed of liberalism.

Nice Blog!