Gallup Tracking Poll Thread [Obama vs McCain]

[Lunar]

I am treating them as independent events.  I'm not presupposing what the true answer is.

[Dan the Roman]

i think there is a simple explantation for this. Gallup does not weight by party id, while Rassmussen updates his weekly, and the rest use a static or who knows what system.

Gallup and Dailykos are both seeing a 9% Democratic advantage right now. Rassmussen saw a 6% one last week. The rest see a one or two point one. If the financial crisis has pushed independents into identifying as Democrats, this would be picked up by Gallup and Dailykos, while the other polls would see it as an actual gain for McCain, since they would not pick up the increased number of Democrats, but would notice that the "independents" in the sample had become marginally more Republican due to Democratic leaning ones changing their identification.

Midweek I would trust Gallup more than Rassmussen for general trends. If Ras is still showing tightening after he re-weights at the end of the week then I would probably trust him. But right now the differences are mostly explained by party id(and a bizzare pro-McCain sample yesterday in Ras).

[minionofmidas]

Bottom line of my argument:

I agree with Dan, ie it's the model not the poll (or the model and the poll). There's no way this difference is attributable to "a bad sample".
My maths excourse was mostly aimed at showing how absurd it is to try and solve these things through maths alone.

[emailking]

I am treating them as independent events.  I'm not presupposing what the true answer is.

You are presupposing the true answer. "...the idea that two polls showing different results would only have 1/400 chance of both being off and having the actual result be in the middle..." You are assuming where the answer is!

You are not treating them as independent events either. "Again, let's take my two poll, 8% off example." They're not independent if you specify where they are with respect to each other, as that statement does.

[muon2]

I am treating them as independent events.  I'm not presupposing what the true answer is.

You are presupposing the true answer. "...the idea that two polls showing different results would only have 1/400 chance of both being off and having the actual result be in the middle..." You are assuming where the answer is!

You are not treating them as independent events either. "Again, let's take my two poll, 8% off example." They're not independent if you specify where they are with respect to each other, as that statement does.

Statistically they are not independent events. They are independent measurements of the same event - the presidential preference. A comparison of MOE is inadequate, since these polls only use the statistical error to quote an MOE.

A better comparison would have the polls also quote a systematic error. The systematic error measures the expected measurement fluctuations based on the choice of modeling. Since no pollster directly uses a raw random sample, there is a systematic error associated with the poll. My guess is that the systematic errors are comparable to the statistical errors based on the dispersion of polling values seen on any given day.

[Nym90]

Since Obama's been 11 up in Gallup two days in a row, it's obviously not just one day of a bad sample. It would have to be at least two days worth.

[J. J.]

Since Obama's been 11 up in Gallup two days in a row, it's obviously not just one day of a bad sample. It would have to be at least two days worth.

The sample stays in for three days.  It does in Rasmussen as well.

That's why I've been saying we should wait until Saturday.  If either the 'bots or Gallup has a bad sample, it will drop out by then.

[??????????]

The race is probably Obama +4. This eleven point "lead" is absolutely unrealistic.

[Eraserhead]

The race is probably Obama +4. This eleven point "lead" is absolutely unrealistic.

I'd say 5-7.

[??????????]

The race is probably Obama +4. This eleven point "lead" is absolutely unrealistic.

I'd say 5-7.

Gallup has had some flatulence the past few days.

[Tender Branson]

Friday, October 10:

Obama: 51 (-1)
McCain: 41 (nc)

These results show little major change in the basic structure of the race, although McCain did somewhat better in Thursday night interviewing, suggesting the possibility that the race may have some fluidity in the days ahead.

[BlueSwan]

So a small summary of Obama's leads in today's tracking polls:

Gallup: Obama +10 (-1)
Rasmussen: Obama +5 (-)
Diego/Hotline: Obama +7 (+1)
Battleground: Obama +8 (+5)
DailyKos: Obama +12 (+2)
Reuters/Zogby: Obama +4 (-)

[Swing low, sweet chariot. Comin' for to carry me home.]

So a small summary of Obama's leads in today's tracking polls:

Gallup: Obama +10 (-1)
Rasmussen: Obama +5 (-)
Diego/Hotline: Obama +7 (+1)
Battleground: Obama +8 (+5)
DailyKos: Obama +12 (+2)
Reuters/Zogby: Obama +4 (-)

thanks for the summary, I was search for the others but am having trouble finding them since they are no longer stickied

[Lunar]

The race is probably Obama +4. This eleven point "lead" is absolutely unrealistic.

It's unrealistic, but your Obama +4 is far more so.  When a credible pollster shows a large lead, you just don't throw it out, you average it out.

Let's look at the polls yesterday (today's are about the same, but you posted this before Gallup's new one came out).

Gallup: Obama +11
Rasmussen: Obama + 5
Hotline: Obama +6
R2000: Obama +10
Battleground: Obama +5
Zogby: Obama +4

...so you go with the Zogby?  Even if you didn't know that you were, the fact that you ARE going with the Zogby when *all* the other polls show something different should be worrying

[Likely Voter]

Maybe someone should do a new post with a daily summary of all the tracking polls with the differences from the previous day?

[J. J.]

I suspect there is a bad sample (possibly Tuesday's sample) that was overly pro-Obama.

[Tender Branson]

I suspect there is a bad sample (possibly Tuesday's sample) that was overly pro-Obama.

Which probably means Obama is down to 6-8 tomorrow.

[J. J.]

I suspect there is a bad sample (possibly Tuesday's sample) that was overly pro-Obama.

Which probably means Obama is down to 6-8 tomorrow.

On Gallup, yes.  If there is drop of 1-3 points on Gallup for Obama tomorrow, I won't be celebrating.

[Tender Branson]

Saturday, October 11:

Obama: 51 (nc)
McCain: 42 (+1)

[J. J.]

Saturday, October 11:

Obama: 51 (nc)
McCain: 42 (+1)

Expected, maybe even a slight uptick for Obama, maybe.

[Tender Branson]

Saturday, October 11:

Obama: 51 (nc)
McCain: 42 (+1)

Expected, maybe even a slight uptick for Obama, maybe.

According to the RCP average, Obama is 0.1% away from 50% ...

[BlueSwan]

All of the tracking polls:

Zogby: Obama +4 (-)
Rasmussen: Obama +7 (+2)
Gallup: Obama +9 (-1)
Hotline: Obama +10 (+3)
R2000: Obama +12 (-)
Battleground: Sorry couldn't find new data for today - yesterday they had Obama +8

[Lunar]

Saturday, October 11:

Obama: 51 (nc)
McCain: 42 (+1)

Expected, maybe even a slight uptick for Obama, maybe.

Expected?  You said you half-expected a drop of 4+ points.

Instead, the two polls are beginning to converge, Rasmussen up and Gallup down, like I said.

[J. J.]

Saturday, October 11:

Obama: 51 (nc)
McCain: 42 (+1)

Expected, maybe even a slight uptick for Obama, maybe.

Expected?  You said you half-expected a drop of 4+ points.

Instead, the two polls are beginning to converge, Rasmussen up and Gallup down, like I said.

I suspect there is a bad sample (possibly Tuesday's sample) that was overly pro-Obama.

Which probably means Obama is down to 6-8 tomorrow.

On Gallup, yes.  If there is drop of 1-3 points on Gallup for Obama tomorrow, I won't be celebrating.

I expected a 1-3 point drop, just do to a skewed sample.  It happened.

Note I'm saying an Obama trend upward, maybe because his numbers have not dropped up the upper edge of this range.

Boy, Lunar, I'm saying that Obama might be doing better than the polling shows, and you get upset with me.

[Lunar]

My bad sir.  I guess I misunderstood, I thought you meant by "not celebrating" that you meant about your own prediction!