Rasmussen Tracking Poll [Obama vs McCain]

[Sam Spade]

I thought the sample that will fall off is O+10?

Ya, state polls tend to lag national polls a little.  Never quite figured out why.

[Sam Spade]

Ya, state polls tend to lag national polls a little.  Never quite figured out why.

i believe you, but that makes no sense

For months, I've been telling people to be patient.  In Rasmussen, there's like a +10 Obama sample that has to drop off, which will likely push the race closer by at least a couple of points.  Also, in Hotline, the poll coming off is a pro-Obama sample of some sort.

Right now, Obama is up about 2-3 points, and I'll lean towards saying it's more like 3 points.  But it's been that way for a week now and the debates will surely change things in some respect.  The state polls are simply reflecting that reality to a certain extent, with the usual weird outliers one would expect from bad firms.

[Sam Spade]

Unless the race has gravitated toward Obama in the last few days, I'm seeing a pretty solid Obama +3, maybe a bit under.

It was that way a few days ago (as I called it) - state polling takes a little time to catch up. 

Key is to try and catch the next movement. 

[Sam Spade]

Observationally, the movement over the past few days is rather interesting. 

Obama's gains (which are not as large as the McCain drops) have to do with movement of Indys in his favor and a slight movement towards him among GOPers.

McCain's losses, while having to do with Indy movement, also have to do with GOPers moving into the undecided column.

Democrats have basically not moved at all.

For example, even though Obama clearly won the night by 11-12 points, the number that dropped off (Obama +10%) had a greater raw Obama score than the number that went on the sample (by around 0.5%).

Tells me that McCain's move on Wednesday was not well-received with swing voters or his base.  So I was wrong in interpretation there.

[Sam Spade]

I assume yesterday's polling was prior to the debate, yes?

Probably.

The overwhelming majority of interviews for today’s report were completed before last night’s Presidential debate.

Rasmussen normally conducts interviews between 5 PM and 9 PM because, well, that's the time you're most likely to be home.

Since the debate occurred at 9 PM eastern, that would mean that a few interviews along the West Coast might have been conducted post-debate.  Probably not enough to be statistically significant, however.

[Sam Spade]

Unfortunately, your image is incorrect looking at the numbers.  It should be a missile crashing into the ground.  To help with some actual numbers:

Obama 49.81% (yesterday - 49.95%)
McCain 44.37% (yesterday - 45.16%)

[Sam Spade]

Unfortunately,

Oh but you just. 

Well, Obama did gain a point in the Gallup.

I'm not surprised.  There's a certain pattern this election is following - I'm waiting to see whether we break out of it.

Just observationally for Rasmussen to give the fuller picture:

           O+11-12    M+1       O+6       O+10    M+5
Obama 49.81% (49.95%/49.47%/49.40%/48.07%/48.16%)
McCain 44.37% (45.16%/45.96%/46.56%/47.79%/46.81%)

[Sam Spade]

It may also be that republicans, since they are for small government, are bailing on a McCain who seems to be going along with Bush and the bailout... just a guess though

Possibly, but I haven't heard McCain support the bailout plan as written.  I think the other explanation makes more sense, especially since I doubt small government Rs would move over to Obama.  Over to undecided, yes.

Generally, most people are against the plan in its present form, and I would advise both candidates to vote against it, so long as it passes.  I definitely wouldn't recommend not voting.

[Sam Spade]

One other minor point:  The gender gap has gotten quite large over the past couple of weeks.

[Sam Spade]

..though it is closer among men than it is among women.

Yes, but more importantly, here's the comparison to 2004:

Men (McCain +8 (-3))
Women (Obama +17 (-14))

[Sam Spade]

confused.  does that mean:

Men:  Bush +5
Women: Kerry +3

?

Men were Bush +11.

[Sam Spade]

Numbers-wise, today's sample was better than the one that went off for Obama.  For McCain it was roughly about the same.  I'll estimate it as Obama +7.68%.  A pro-McCain sample comes off tomorrow, so there well could be another bump.

[Sam Spade]

Numbers-wise, today's sample was better than the one that went off for Obama.  For McCain it was roughly about the same.  I'll estimate it as Obama +7.68%.  A pro-McCain sample comes off tomorrow, so there well could be another bump.

Last night estimate:
Obama +7.7%

But it is a bit of an improvement considering it was Obama +11 the previous night.

Actually, it was closer to Obama +12, but it's not a biggie.

[Sam Spade]

Taking a gander through the internals of the bailout numbers Rasmussen posted this morning, which should contain the same sample for his Prez election numbers, I'm willing to make the educated guess that the sample for today is *roughly* McCain +2.

So, expect the topline numbers to remain the same, since this is *roughly* the same sample as what dropped off.

[Sam Spade]

Taking a gander through the internals of the bailout numbers Rasmussen posted this morning, which should contain the same sample for his Prez election numbers, I'm willing to make the educated guess that the sample for today is *roughly* McCain +2.

So, expect the topline numbers to remain the same, since this is *roughly* the same sample as what dropped off.

I would find it really strange that, at a time when Obama appears to be around +5 or better in the trackers, and in a day just after the debate in which Obama was considered to have done better than McCain by the watching focus groups, McCain could have a +2 sample.

On the same point, I also find it really strange that the R2K tracker is producing consecutive daily figures in sequences such as O+5, O+6, O+7, O+5, O+4 etc., and yet you are claiming that Ras, with a sample size three times larger, is flicking between such extremes as, for example O+11 and M+2 on consecutive days.

I don't really trust the R2K/Daily Kos alliance, for good reason.  You may disagree, but I stand by my principles here.

As for the Ras numbers, flickering greatly between margins seems likely to occur, especially given the crisis and uncertainty about things right now.  But don't forget, there are other tracking polls out there, other numbers to be looked at, and other data to be examined.  Rasmussen's daily result today is but a data point within a data point.

Anyway, today's numbers...

Monday, September 29, 2008
Obama 50%
McCain 45%

[Sam Spade]

Last five days:

              M+3        O+8      O+12      M+1       O+6      O+10   
Obama 50.44% (50.43%/49.81%/49.95%/49.47%/49.40%)
McCain 45.05% (44.35%/44.37%/45.16%/45.96%/46.56%)

Sample today is *roughly* McCain +2.07.  Rowan may come in and correct me later on this number.

EDIT:  My sleepiness makes it not McCain +2.07%, but rather McCain +3.22%.  Ugh.

[Sam Spade]

I should mention that opposition to the bailout, at least through last night's numbers, fell quite evenly among both Democrats, Republicans and Indys.  In fact, there really is not a  statistically significant difference between their levels of support (1/3 for, 1/3 against, 1/3 not sure), although I would say that Republicans support it slightly more than Indys who support it slightly more than Dems.

[Sam Spade]

Look folks, so long as the ladder Rowan has provided is correct, the guesses as to the actual daily numbers will be correct.

[Sam Spade]

The only thing I would point out, muon2, is that I separately calculated the averages of the one-day samples using the internals of the separate one-day only polls used in:

1) Debate results (Saturday Prez sample)
2) Bailout numbers (Sunday Prez sample)

When Rasmussen gives these numbers, he also gives the internals of how they break down in GE w/leaners (McCain v. Obama).  

While not entirely accurate (he does not include the small number of undecideds), we can use these internals multiplied to the actual support of the other numbers to make an educated guess at the end number.  So, this adds a bit of a cross-check into the analysis.

For Saturday, I said Obama +7 to 8 (the formula has it at 7.70%)
For Sunday, I said McCain +2 (the formula has it at 3.22%).  Looking at the numbers again now, and doing the calculations, McCain +3 actually fits more accurately with the internals.

[Sam Spade]

The only thing I would point out, muon2, is that I separately calculated the averages of the one-day samples using the internals of the separate one-day only polls used in:

1) Debate results (Saturday Prez sample)
2) Bailout numbers (Sunday Prez sample)

When Rasmussen gives these numbers, he also gives the internals of how they break down in GE w/leaners (McCain v. Obama).  

While not entirely accurate (he does not include the small number of undecideds), we can use these internals multiplied to the actual support of the other numbers to make an educated guess at the end number.  So, this adds a bit of a cross-check into the analysis.

For Saturday, I said Obama +7 to 8 (the formula has it at 7.70%)
For Sunday, I said McCain +2 (the formula has it at 3.22%).  Looking at the numbers again now, and doing the calculations, McCain +3 actually fits more accurately with the internals.

I'm not saying your estimate is wrong, I'm just saying that the variation is suspiciously large. I recognize I only have 6 points to deal with, so this set can be a statistical fluke. However, I think the occurrence of three points outside the 95% range, when I expect less than one gives me reason to raise a question.

Nah, I understand your point.  My main concern has to do with whether the beginning of the sequence is incorrect or not.  If it is, then the entire sequence would be wrong.  All we could do is tell, vis-a-vis the three-day average, that McCain's numbers improved 2.09% (roughly) in comparison to yesterday's numbers.  We couldn't tell the *starting off point*, which would allow us to determine how this movement compares to the sample that dropped off (which would then be able to give us the actual daily sample).

This is one of the reasons why I'm starting to do the internal one-day checks on Rasmussen's other questions that he asks that are released separately in the polling database.  I want to see how close my guesstimates come to the ladder sequence.  So, for the last two days, these have been my guesstimates (based on the internals):

Sat (debate sample):  O53-M45
Sun (bailout sample):  M50-O47

They might be a tad lower on both sides (more undecideds), but considering the fact that Friday's sample was most certainly a high undecided sample (both candidates moved downwards in raw internals, clearly), I have to believe that this is a reasonable guess.

[Sam Spade]

The only thing I would point out, muon2, is that I separately calculated the averages of the one-day samples using the internals of the separate one-day only polls used in:

1) Debate results (Saturday Prez sample)
2) Bailout numbers (Sunday Prez sample)

When Rasmussen gives these numbers, he also gives the internals of how they break down in GE w/leaners (McCain v. Obama).  

While not entirely accurate (he does not include the small number of undecideds), we can use these internals multiplied to the actual support of the other numbers to make an educated guess at the end number.  So, this adds a bit of a cross-check into the analysis.

For Saturday, I said Obama +7 to 8 (the formula has it at 7.70%)
For Sunday, I said McCain +2 (the formula has it at 3.22%).  Looking at the numbers again now, and doing the calculations, McCain +3 actually fits more accurately with the internals.

I'm not saying your estimate is wrong, I'm just saying that the variation is suspiciously large. I recognize I only have 6 points to deal with, so this set can be a statistical fluke. However, I think the occurrence of three points outside the 95% range, when I expect less than one gives me reason to raise a question.

Nah, I understand your point.  My main concern has to do with whether the beginning of the sequence is incorrect or not.  If it is, then the entire sequence would be wrong.  All we could do is tell, vis-a-vis the three-day average, that McCain's numbers improved 2.09% (roughly) in comparison to yesterday's numbers.  We couldn't tell the *starting off point*, which would allow us to determine how this movement compares to the sample that dropped off (which would then be able to give us the actual daily sample).

This is one of the reasons why I'm starting to do the internal one-day checks on Rasmussen's other questions that he asks that are released separately in the polling database.  I want to see how close my guesstimates come to the ladder sequence.  So, for the last two days, these have been my guesstimates (based on the internals):

Sat (debate sample):  O53-M45
Sun (bailout sample):  M50-O47

They might be a tad lower on both sides (more undecideds), but considering the fact that Friday's sample was most certainly a high undecided sample (both candidates moved downwards in raw internals, clearly), I have to believe that this is a reasonable guess.

The best test would be an extended time correlation analysis. If data over the last 3 weeks shows a strong three day cycle then the initial data is likely off. The good part of this type of analysis is that it can pick up cycles independent of an underlying long-term trend.

I'm not saying your estimate is wrong, I'm just saying that the variation is suspiciously large. I recognize I only have 6 points to deal with, so this set can be a statistical fluke. However, I think the occurrence of three points outside the 95% range, when I expect less than one gives me reason to raise a question.

but this is statistics, clustering is always a possibility.

Clustering is a statistical possibility, but a persistent three-day cycle would indicate a systematic bias in the process. This assumes that short term effects should be statistical, and there is no external environmental effect to provide voter shifts in a three-day pattern.

I agree with your conclusion, muon.  I really don't have the time to implement it though.  So, this is the best I can do unless Rowan provides us proof for his sequence, which is the same third-party that provided us the secrets to begin with. 

[Sam Spade]

For some reason Rasmussen is being lazy about posting the daily snapshot today so that I can tell you more, but IMO, the ladder Rowan's been posting has to be incorrect, examining the other internals for the inexperienced and old polls conducted today.

In sum, they really don't make any sense, especially considering the undecideds have to be so low, unless today's sample is roughly about Obama +4 to Obama +6, maybe Obama +7.

[Sam Spade]

Chart:
                                         samp falls off tom      samp falls off today
Obama 50.85% (50.44%/50.43%          /           49.81%          /           49.95%)
McCain 44.89% (45.05%/44.35%          /           44.37%          /           45.16%)

As I said before, I now highly doubt Rowan's sequence that he provided, because one of the internal samples in the "inexperienced/old" questions made it abundantly clear that there was pretty much no way Obama's sample could be much over 51% today (definitely under 52%) for the number to work out internally.

Similarly, one of the other internals there make it highly unlikely that McCain's sample could be much under 45% (definitely over 44%).  And I really suspect 46% is the more likely answer.

I may review the past couple of days' polls and see if I can find a number in the past few days which I can use to isolate the result.  If so, we may have something.

[Sam Spade]

Still trying to isolate the McCain number, but I'm still 99.9% sure the sequence is incorrect.

[Sam Spade]

Still trying to isolate the McCain number, but I'm still 99.9% sure the sequence is incorrect.

Well, that has seemed likely for a while now. Why not check Alcon's thread. https://uselectionatlas.org/FORUM/index.php?topic=84114.0

Ya, but Alcon's sequence is incorrect also because it doesn't end up with the right internal.