41.4% - 35.5% - 23.1%

[The Vorlon]

If you add up all the state by state party identifications in ARG's 50 state poll-o-thon and weight according to each states population you get:

41.4% Democrats,
35.5% Republicans
23.1% Independents

A 5.9% Democrat advantage on Party ID

[MODU]

Hence why I use my stated 2% rule on ARG polls.    It brings the poll figures back to the norm.

[jacob_101]

If you add up all the state by state party identifications in ARG's 50 state poll-o-thon and weight according to each states population you get:

41.4% Democrats,
35.5% Republicans
23.1% Independents

A 5.9% Democrat advantage on Party ID

Good, then Bush is probably tied in MN!

[zorkpolitics]

A recent report claims that most pollsters have seen a shift to an increased Republican self identification after the RNC. A 6% excess of Democrats in the ARG polls would seem to go against that trend:
http://story.news.yahoo.com/news?tmpl=story&ncid=1963&e=22&u=/ap/20040924/ap_on_el_pr/leaning_gop_2&sid=96378798

[Pollwatch99]

Yes, MN does look tied according to Rasmussen.  He dropped NC as a daily tracking poll ( I think he wanted to monitor whether a Bush state was trending towards Kerry).  MN is now a daily tracking because the Kerry states are all moving towards Bush; some more than others ( IA, WI, MN, PA, ME, NM).  You don't see the Bush states trending towards Kerry

[Filuwaúrdjan]

Yes, MN does look tied according to Rasmussen.  He dropped NC as a daily tracking poll ( I think he wanted to monitor whether a Bush state was trending towards Kerry).  MN is now a daily tracking because the Kerry states are all moving towards Bush; some more than others ( IA, WI, MN, PA, ME, NM).  You don't see the Bush states trending towards Kerry

Not another partisan hack

[A18]

What Bush states are trending Kerry, then?

[Pollwatch99]

I cannot find any Bush states trending Kerry.  Anybod else see any?

[Filuwaúrdjan]

Your question makes no logical sense.

I'll answer something else: Bush seems to be doing worse in several states than he did in 2000, and he's also doing better in several states (notably Wisconsin and New Jersey) than he did in 2000.

[Sam Spade]

I would've said Colorado a few days ago, but now with two polls out showing Bush at the same lead as in 2000, I really can't say that either.

The only state that Bush won that's still trending towards the Dem. more than in 2000 is still NC, by probably 5-7 points.  But we haven't had a poll out of there lately, so that could be wrong also.

All the other Bush states are either even with 2000 are trending towards Bush right now as far as I can tell.  (Montana excluded, maybe, but who cares about Montana).

[Filuwaúrdjan]

I would've said Colorado a few days ago, but now with two polls out showing Bush at the same lead as in 2000, I really can't say that either.

The only state that Bush won that's still trending towards the Dem. more than in 2000 is still NC, by probably 5-7 points.  But we haven't had a poll out of there lately, so that could be wrong also.

All the other Bush states are either even with 2000 are trending towards Bush right now as far as I can tell.  (Montana excluded, maybe, but who cares about Montana).

There just aren't enough good polls (there's plenty of crap ones) to make a definate statement either way IMO.

[Pollwatch99]

Colorado looks fine for Bush.   Looking at NC is like looking for trends in NY.  Bush win will NC, Kerry will win
NY unless something really strange happens

C0 2000 Bush +8.4%
SUSA        9/21-9/23 626 LV Bush +8
Ciruli Assoc          9/14-9/18 600 LV Bush +12
Rasmussen            9/16          500 LV Bush +1

Democrats like to focus on FL.  3 others states were incredibly close in 2000 (NM, WI, and IA).  My remarks are based on poll numbers and the numbers are the numbers, they are not partisan.  They are moving towards Bush compared to 2000 results.

Here they are:

NM 2000 Gore +0.06%
Mason-Dixon        9/15-9/16 625 LV Bush +4
ARG                     9/14-9/16  600 LV Kerry +5

WI 2000 Gore +0.2%
Badger Poll      9/15-9/21 504 RV Bush +14
ABC News            9/16-9/19 625 LV  Bush +10
Mason-Dixon        9/14-9/16 625 LV  Bush +2
ARG                      9/12-9/15 600 LV  TIE

IA 2000 Gore +0.3%
SUSA                    9/20-9/22 784 LV  Bush +4
Fox News              9/21-9/22 800 LV  Bush +3
Research 2000       9/19-9/21 602 LV  Bush +2
CNN/USA/Gallup 9/16-9/19 631 LV  Bush +6

MN  2000 Gore +2.5%
MPR/Mason-Dixon  9/11-14 625 LV  Bush +2  
CNN/Gallup/USAT  9/11-13 675 LV  TIE
ARG                         9/10-12 600 LV   Kerry +2

PA 2000 Gore +4.2%
RASMUSSEN 7 day rolling average     TIE
Fox News         9/21-9/22 800 LV         Kerry +3
ARG                 9/15-9/19 600 LV         Kerry +1
Mason-Dixon | 9/14-9/16 624 LV          Kerry +1

ME 2000 Gore +5.1%
SUSA                   9/20-9/22 636 LV          Bush +1  
Critical Insights   9/10-9/23 600 LV          Kerry +3
 

[Sam Spade]

It is right of you to leave out Florida in the guessing equation, because I still believe it'll be another 2-3 weeks before we can get a trustworthy poll out of there (and I don't consider the ARG, Quinnipiac or Gallup polls trustworthy).

Oregon has also been polling very weirdly of late (as it usually does), but my gut feeling tells me that it is where it was in 2000, which would leave it leaning Kerry were Kerry to regain any of what he's lost.

[Filuwaúrdjan]

Colorado looks fine for Bush.   Looking at NC is like looking for trends in NY.  Bush win will NC, Kerry will win
NY unless something really strange happens

C0 2000 Bush +8.4%
SUSA        9/21-9/23 626 LV Bush +8
Ciruli Assoc          9/14-9/18 600 LV Bush +12
Rasmussen            9/16          500 LV Bush +1

Democrats like to focus on FL.  3 others states were incredibly close in 2000 (NM, WI, and IA).  My remarks are based on poll numbers and the numbers are the numbers, they are not partisan.  They are moving towards Bush compared to 2000 results.

Here they are:

NM 2000 Gore +0.06%
Mason-Dixon        9/15-9/16 625 LV Bush +4
ARG                     9/14-9/16  600 LV Kerry +5

WI 2000 Gore +0.2%
Badger Poll      9/15-9/21 504 RV Bush +14
ABC News            9/16-9/19 625 LV  Bush +10
Mason-Dixon        9/14-9/16 625 LV  Bush +2
ARG                      9/12-9/15 600 LV  TIE

IA 2000 Gore +0.3%
SUSA                    9/20-9/22 784 LV  Bush +4
Fox News              9/21-9/22 800 LV  Bush +3
Research 2000       9/19-9/21 602 LV  Bush +2
CNN/USA/Gallup 9/16-9/19 631 LV  Bush +6

MN  2000 Gore +2.5%
MPR/Mason-Dixon  9/11-14 625 LV  Bush +2  
CNN/Gallup/USAT  9/11-13 675 LV  TIE
ARG                         9/10-12 600 LV   Kerry +2

PA 2000 Gore +4.2%
RASMUSSEN 7 day rolling average     TIE
Fox News         9/21-9/22 800 LV         Kerry +3
ARG                 9/15-9/19 600 LV         Kerry +1
Mason-Dixon | 9/14-9/16 624 LV          Kerry +1

ME 2000 Gore +5.1%
SUSA                   9/20-9/22 636 LV          Bush +1  
Critical Insights   9/10-9/23 600 LV          Kerry +3

"Lies, Damn Lies and Statistics"

The actual "leads" are (IMO) meaningless. Look at the actual numbers, the number of undecideds, have a look at the MoE and the pollsters past record, have a lot at demographic info (if any), THEN (and only then) compare to the 2000 results.

[Pollwatch99]

Yes, numbers are in MOE in many cases.  However, with 3 or 4 polls, a MOE deviation should show Kerry tied or leading in one.  Trend to Bush.

[Filuwaúrdjan]

Yes, numbers are in MOE in many cases.  However, with 3 or 4 polls, a MOE deviation should show Kerry tied or leading in one.  Trend to Bush.

*sigh*
Read what I said...

[badnarikin04]

My question is....


What the hell is up with these independents? Doesn't independent mean you're not voting Republicrat? So how come we don't see that in the results?

[CARLHAYDEN]

After a lot of though, poring through mounds of data, and a lot of number crunching (not to mention coin flipping), I believe the correct weighting of the electorate at this time should be:

Democrats          38%

Republicans         37

Others                 25

[Pollwatch99]

If you're right; Bush is headed for a decent sized victory

[agcatter]

Exactly.

[A18]

My question is....


What the hell is up with these independents? Doesn't independent mean you're not voting Republicrat? So how come we don't see that in the results?

Independent means you aren't part of any political party.

[CARLHAYDEN]

What many polling firms are doing is to assume that 2004 is a repeat of 2000.

In that election, the Democrats, due in large part to an excellent GOTV drive (and corresponding lack of it by the Republicans) had an unusually high percentage of the actual vote.

Several pollsters were mortified by the result and resolved not to underestimate the Democrat vote again and they therefor changed their weighting to a four (or more) point advantage for the Democrats.

Gallup has the two parties as even in adherents and PEW has the Democrats up 2 (if memory serves me correctly).

My own analysis splits this difference.

If the Republicans do as good a job this year at GOTV (primarily early voting) as they say they will do, and if the Kerry campaign continues to stumble and bumble, then the actual turnout could be even more favorable for the Republicans than I project.

[Pollwatch99]

I'm beginning to fear that the problem with 2000 republican turnout was with the christian evangelical turnout which was 4MM less than 1996.  

I never though much of the Christian Coalition and Ralph Reed but maybe I was wrong.

We need these 4MM conservative voters to come to the polls to increase turnout

[CARLHAYDEN]

Actually the key demographic vote segment in this election is white women.

They favored Gore in 2000 as they saw him as a moderate who would continue the Clinton policies.

Since they, they have seen Gore act like a left wing nut (outdoing Howard Dean).

They like the way Bush handled the CBS smear campaign.

They don't like the way Kerry has acted in the past couple of weeks with his strident rhetoric.

They are the one segment of the electorate that looks at the wives of the candidates.

They like Laura Bush and dislike Heinz-Kerry.

They figure that the woman behind the man tells a lot about the man.

[A18]

And to that I say, the Massachusetts Supreme Court may have done more good than they know.