Non-Gallup/Rasmussen tracking polls thread

[J. J.]

Sorry just woke up.

• Gallup: Obama 50%, McCain 43%, with a ±2% margin of error, compared to a 51%-42% Obama lead yesterday.

• Rasmussen: Obama 51%, McCain 45%, with a ±2% margin of error, compared to a 52%-45% Obama lead from yesterday.

• Hotline/Diageo: Obama 49%, McCain 41%, with a ±3.4% margin of error, compared to a 50%-40% Obama lead yesterday.

• Research 2000: Obama 53%, McCain 40%, with a ±3% margin of error, compared to a 52%-40% Obama lead from yesterday.

• Zogby: Obama 49%, McCain 43%, with a ±2.9% margin of error, compared to a 48%-44% Obama lead yesterday.

Fixed.

[J. J.]

Um, R2000 is not Zogby.   R2000 is a respectable firm with Democratic bias.  I would certainly consider R2000  to be better than the unknown Hotline poll.

Zogby has no consistency or reliability.

R2000 has some weighting problems.  Zogby is, well, Zogby.

[J. J.]

Um, R2000 is not Zogby.   R2000 is a respectable firm with Democratic bias.  I would certainly consider R2000  to be better than the unknown Hotline poll.

Zogby has no consistency or reliability.

R2000 has some weighting problems.  Zogby is, well, Zogby.

Well, fix the weighting and apply your own interpretation to it.

I'm not going to try to fix the weighting, except to note that it tends to be too pro Democratic.

[J. J.]

So?  They aren't a pollster to discard, unlike Zogby, who does not publish his internals and is not consistent.  Zogby is only posted because the media will observe it while I think R2000's results are in themselves interesting.

Yes, because the poll is too badly weighted.

If you want an A list, it's Gallup and Rasmussen.

The B list is Hotline.

The Z list are Zogby and Kos.  (Note that theae are both off, in opposite directions, than those other three).

[J. J.]

According to a certain supply-side pundit , the first iteration of the IBD/TIPP poll will come out today showing... Obama 45, McCain 43.

It sounds about right in terms of timing.  IBD/TIPP started doing their tracking poll in 2004 on October 12.

This poll has an excellent record (most accurate poll in 2004, within MOE in 2000 (Bush +1.9%)).  They do not push the leaners very much until the end.

It deserves a separate thread.

Is it out yet?

[J. J.]

According to a certain supply-side pundit , the first iteration of the IBD/TIPP poll will come out today showing... Obama 45, McCain 43.

It sounds about right in terms of timing.  IBD/TIPP started doing their tracking poll in 2004 on October 12.

This poll has an excellent record (most accurate poll in 2004, within MOE in 2000 (Bush +1.9%)).  They do not push the leaners very much until the end.

It deserves a separate thread.

Is it out yet?

Nope.  Hey, maybe this was psyops for Republicans.

Just confirmed it on CNBC.

[J. J.]

Here is a reference on the TIPP tracking:

Larry Kudrow:  Just off the press: Obama 45; McCain 43; and 13 percent unsure. The poll of 825 likely voters has an error margin of +/- 3.5 percentage points. Also, McCain has a 48-41 lead among investors with 10 percent not sure. All this is good news for McCain.

The only other one I found was here:

http://www.investors.com/editorial/IBDArticles.asp?artsec=8&issue=20081013

[J. J.]

13% undecided with three weeks to go?  Not sure why McCain's 9 point lead among investors is "good news" unless placed in the context of partisan identification...

The way they define "investors," $10K in the market, it basically includes suburbanites and those with 401K's. 

TIPP had good track record in 2000 and an even better one in 2004.

[J. J.]

13% undecided with three weeks to go?  Not sure why McCain's 9 point lead among investors is "good news" unless placed in the context of partisan identification...

The way they define "investors," $10K in the market, it basically includes suburbanites and those with 401K's. 

TIPP had good track record in 2000 and an even better one in 2004.

Oh, I'm not trashing the pollster.  I'm just not sure if that 9% is a good thing or bad thing for Obama.  If the investor number normally breaks 25% for the GOP and this year it's only 9%, that's a good thing, yaddamean?

Considering the market and the number of investors, I'm happy with it.

You must be logged in to read this quote.

Not great, but still good.  Still I'll need to see much more polling.

[J. J.]

13% undecided with three weeks to go?  Not sure why McCain's 9 point lead among investors is "good news" unless placed in the context of partisan identification...

The way they define "investors," $10K in the market, it basically includes suburbanites and those with 401K's. 

TIPP had good track record in 2000 and an even better one in 2004.

Oh, I'm not trashing the pollster.  I'm just not sure if that 9% is a good thing or bad thing for Obama.  If the investor number normally breaks 25% for the GOP and this year it's only 9%, that's a good thing, yaddamean?

Considering the market and the number of investors, I'm happy with it.

But what if that margin is normally 40%?  It's a very Republican-skewed demographic.  I'm not saying it's THAT skewed, but it actually be bad news in disguise or completely neutral news.

It shows a shift in the internals, but not an overall one.

[J. J.]

One advante is a four day sample; it won't be affected as much by a skewed sample.

Sam, how does the weighting look to you?

[J. J.]

Finally have the internals...

October 6-12, 825 LV

http://www.ibdeditorials.com/Polls.aspx?id=308790300872940#pollb

Small sample size per day, but there shouldn't be ridiculous movement, as this poll is highly weighted.  I think the party ID has always been 39D, 35R, 26I (2000/2004 as well).

Unless they're defining "Midwest" very strangely, I find it highly unlikely McCain is leading in the region. (Kerry won it under any reasonable definition.)

It might include KY and TN.  I'd really like to see a few days worth of polling.

[J. J.]

One advante is a four day sample; it won't be affected as much by a skewed sample.

Sam, how does the weighting look to you?

Four-day sample?  Looks like a 7-day sample to me.

I know that party ID is somewhere between D+2 to D+4.  I'm pretty sure it's D+4, but I'm willing to be corrected.  As for the rest of the weights, they look ok to me (with the usual MOE caveats), but I don't have exact percentages.

You are correct, I thought it was a seven day sample.

Plus four over Democrats?

[J. J.]

Drudge leaked tomorrow's Zogby numbers (if anyone actually cares)...

Obama 49%
McCain 43%

Zogby certainly does...

Now if Dudge would leak Rasmussen's numbers, I'd be interested.

[J. J.]

Hotline 10/14/08

Obama 48 (nc)

McCain 42 (nc)

[J. J.]

• Gallup*: Obama 53%, McCain 43%, with a ±2% margin of error, compared to a 53%-43% Obama lead yesterday.

• Rasmussen: Obama 50%, McCain 45%, with a ±2% margin of error, unchanged from yesterday.

• Hotline/Diageo: Obama 48%, McCain 42%, with a ±3.4% margin of error, unchanged from yesterday.

• Research 2000: Obama 52%, McCain 41%, with a ±3% margin of error, compared to a 52%-40% Obama lead from yesterday.

• Zogby: Obama 49%, McCain 43%, with a ±2.9% margin of error, compared to a 48%-44% Obama lead yesterday.

Fixed

[J. J.]

• Gallup*: Obama 53%, McCain 43%, with a ±2% margin of error, compared to a 53%-43% Obama lead yesterday.

• Rasmussen: Obama 50%, McCain 45%, with a ±2% margin of error, unchanged from yesterday.

• Hotline/Diageo: Obama 48%, McCain 42%, with a ±3.4% margin of error, unchanged from yesterday.

• Research 2000: Obama 52%, McCain 41%, with a ±3% margin of error, compared to a 52%-40% Obama lead from yesterday.

• Zogby: Obama 49%, McCain 43%, with a ±2.9% margin of error, compared to a 48%-44% Obama lead yesterday.

Fixed

Stop that.  They're a respectable poll with a slight democratic bias.  In fact, you CITE their numbers in your arguments in other threads!  I get how you feel about it, but you're not contributing anything by doing that repeatedly.

As has been pointed out from the start, they have a very bad weighting, even though they showed a gain for McCain.

[J. J.]

The Daily Kos poll is questionable in my view, and will always be questionable.

Indeed, it is consistently the most favorable poll to the Dems by a large margin.  But it's a reasonable respectable institution, unike Zogby.  I reserve the strike-out for proven idiocy only.

You know, Zogby, although he's an idiot, has gotten things correct before.  We might call it luck, but a broken clock is right twice a day.

I can't legitimately call Daily Kos a reasonably respectable institution, even though I know that's not what you're talking about.

If you knew what I was talking about, why'd you say that then?   

Zogby does get some things right, his 6% today could be right on.  I still don't think his numbers are useful for us.  R2000 is what a Democratic surge would look like, roughly...

Then why use it, if it isn't accurate.  Crap is still crap, even if it shows the Democrats going downward.

[J. J.]

I'm not "using" it, but the firm has at least an average track record and publish all of their crosstabs.  It's not Zogby.

Well, if you are going to strike through Zogby, strike through R2000.

[J. J.]

ZOGBY

WEDNESDAY

OBAMA 48.2%, MCCAIN 44.4%... NOT SURE 7.4%... DEVELOPING...

Fixed. 

[J. J.]

TIPP Tuesday 10/13/08

Obama     45 (nc)

McCain     42 (-1)

[J. J.]

What is the party ID weighting for this Zogby poll?

Weighting? Zogby doesn't need weighting.

The claim that they do, but I couldn't find it.

I did stumble on to their secret:  www.llewellyn.com/free/tarot.php

[J. J.]

• Gallup*: Obama 52%, McCain 44%, with a ±2% margin of error, compared to a 53%-43% Obama lead yesterday.

• Rasmussen: Obama 50%, McCain 45%, with a ±2% margin of error, unchanged from yesterday.

• Hotline/Diageo: Obama 49%, McCain 41%, with a ±3.4% margin of error, compared to a 48%-42% Obama lead from yesterday.

• Research 2000: Obama 52%, McCain 41%, with a ±3% margin of error, unchanged from yesterday.

• Zogby: Obama 48%, McCain 44%, with a ±2.9% margin of error, compared to a 49%-43% Obama lead yesterday.

from TPM

Fixed.

[J. J.]

TPM's summation:

• Gallup*: Obama 51%, McCain 45%, with a ±2% margin of error, compared to a 52%-44% Obama lead yesterday.

• Rasmussen: Obama 50%, McCain 46%, with a ±2% margin of error, compared to a 50%-45% Obama lead from yesterday.

• Hotline/Diageo: Obama 49%, McCain 41%, with a ±3.4% margin of error, unchanged from yesterday.

• Research 2000: Obama 52%, McCain 41%, with a ±3% margin of error, unchanged from yesterday.

• Zogby: Obama 49%, McCain 43%, with a ±2.9% margin of error, compared to a 48%-44% Obama lead yesterday.

Fixed, and post the Battleground results as well (and consider drawing a line through it as well).

[J. J.]

Why don't you cross out the DailyKos poll?

Because it's a legitimate firm with just some consistent weighting issues, but makes them all available by fully publishing all of its crosstabs.

Battelground I don't post because it's not on TPM, obv.  I'm curious what JJ is doing by doing this every day, everyone knows his opinion on the poll already, it seems infantile.

Because you misrepresent when you discount Zogby and include Kos without it.