Gallup Tracking Poll Thread [Obama vs McCain]

[Sam Spade]

Last three nights estimated Gallup results(most recent first):

Obama +3.0%
Obama +3.4%
Obama +8.6%

Thus Obama's best night will fall off the tracker, so McCain should gain a point or 2 tomorrow.

That doesn't fit with this remark:

Obama enjoyed one of his widest advantages over McCain of recent weeks in Thursday night's interviewing. It will be important to see whether the stock market's reaction today to aggressive government intervention in the crisis has an impact of the direction of the presidential race over the next few days.

So, how are you coming to this conclusion.  Once again, do you have access to internals?

[Sam Spade]

Last three nights estimated Gallup results(most recent first):

Obama +3.0%
Obama +3.4%
Obama +8.6%

Thus Obama's best night will fall off the tracker, so McCain should gain a point or 2 tomorrow.

That doesn't fit with this remark:

Obama enjoyed one of his widest advantages over McCain of recent weeks in Thursday night's interviewing. It will be important to see whether the stock market's reaction today to aggressive government intervention in the crisis has an impact of the direction of the presidential race over the next few days.

So, how are you coming to this conclusion.  Once again, do you have access to internals?

I don't know why that is written into their write-up. I have spoken to 2 other people who's nightly results are similar to mine(off by a point because of rounding), Gallup can write whatever they want in their write-up but the numbers don't lie at this point.

Wait till tomorrow, if McCain gets closer, my numbers will prove to be close to accurate.

Well, it would seem like the reverse of your numbers would better fit their actual talk, but you're right, we'll see...

[Sam Spade]

Extremely disappointed with these numbers.

You know, it is quite possible that your guesstimates are wrong here and that Gallup's statement from Thursday is correct (which translated to me into being at O+7 or better because there were a couple of samples during the DNC/aftermath which were greater than O+6 - after Hillary and Labor Day)

It would make more sense, also as compared to the rest of the tracking polls.

In other words, if Sunday is the mean, with McCain up 1-2, as I figured then, we see this type of pattern:

Monday:  Huge McCain dropoff
Tuesday:  McCain gain, but not to the level of Sunday
Wednesday:  Reversion of McCain back to Monday's levels
Thursday:  Another McCain dropoff, below Monday and Wednesday
Friday:  McCain gain, but not to the level of Tuesday

What this means in terms of actual numbers is your own guess...    Remember, I always have tended to believe that public perception in polling lags a day or so behind actual events, but that's just me.  Vorlon may disagree, actually.

[Sam Spade]

I'll agree, my numbers may be off here, I may have to relook at my calculations. But my Rasmussen numbers are indeed pretty precise.

Well, if you're getting them from a premium member, they ought to be. 

[Sam Spade]

Sunday, September 21, 2008

Obama - 49% (-1)
McCain - 44% (+1)

It's 49-45. 

[Sam Spade]

Obama's 'best individual night in weeks' drops off tomorrow.  What's up momentum change. 

It's not momentum change - it's erosion of the bump.  Big difference.

Hopefully by Monday, I'll have an idea where the race truly stands, but I might have to wait until Thursday.

[Sam Spade]

Obama's 'best individual night in weeks' drops off tomorrow.  What's up momentum change. 

Actually, according to the data, the best night in weeks actually dropped off today, given the largest shift (3 points) was September 15

Uh, that sample dropped off yesterday, not today.  And once again, as I did to RowanBrandon with his analysis, I point to this language:

Obama enjoyed one of his widest advantages over McCain of recent weeks in Thursday night's interviewing. It will be important to see whether the stock market's reaction today to aggressive government intervention in the crisis has an impact of the direction of the presidential race over the next few days.

Although I don't know what the last two days' samples have been, it seems clear that Thursday's sample (Friday) was more pro-Obama than Wednesday's sample (Thursday).

How much?  Who knows...

[Sam Spade]

Friday nights sample was at least a high single digit win for Obama, while Saturday was a small McCain win.

So, what do you think it was?

[Sam Spade]

Wednesday, September 24, 2008
Obama - 47% (nc)
McCain - 44% (nc)

If Rowan's numbers are correct, which is undeterminable, Obama had a very good night last night as he did with Rasmussen.

Actually no.  The last number he posted above would have dropped off yesterday.  Of course, that assumes that the numbers are, in any way, correct.

[Sam Spade]

Of course, today's state polling is all Obama so far.

Not really.  Lots of confusion.

[Sam Spade]

Remember that there will be a 1 in 20 sample every three weeks statistically, and between the two "main" tracking polls that means one every 10 days, or a week and a half.

Problem is, Gallup's language doesn't suggest such an event.

This update covers interviewing conducted Monday through Wednesday, and as such includes one night after McCain's announcement that he was suspending election campaigning and flying to Washington to help seek a bipartisan solution to the financial crisis. A night by night analysis of interviewing results, however, does not suggest that McCain had a dramatically better night against Obama on Wednesday. Instead, the data show that McCain has been doing slightly better for the last three days than he had in the previous week, and with some strong Obama days falling off of the rolling average, the race has moved to its current tied position.

We shall see.

[Sam Spade]

Friday, September 26, 2008
Obama 48% (+2)
McCain 45% (-1)

[Sam Spade]

Multiple Kossacks are saying a Gallup guy was on tv reporting that today's poll will show Obama +8. It's unclear if it's the tracking poll or the regular Gallup poll, but would they release a regular poll on a Sunday?

I heard the same thing myself.

I don't think that USA Today/Gallup can put out a poll this quick - they would at least have to finish the interviews sometime this afternoon, most likely.  So it has to be the tracking poll.

Considering that the sample which drops off today is actually pro-McCain (it had to be at least McCain by a couple to get the Thursday result), my guess is that Rasmussen and Gallup had a similar sample today - Obama +6 to 8 or so.  If things continue as they should, Rasmussen will probably bump up to where Gallup is tomorrow.

[Sam Spade]

Multiple Kossacks are saying a Gallup guy was on tv reporting that today's poll will show Obama +8. It's unclear if it's the tracking poll or the regular Gallup poll, but would they release a regular poll on a Sunday?

btw, the Gallup guy was Frank Newport.  He is the guy at Gallup.

[Sam Spade]

Monday, September 29, 2008
Obama 50% (nc)
McCain 42% (nc)

[Sam Spade]

You must be logged in to read this quote.

No, StatesRights is exactly right. Those on the left vastly underestimate the absolute hatred that this election has generated for the media.

Many of us now regard the media as nothing more than political terrorists. They are vile scum, liars of the worst order.

There are consequences to everything. And the media has yet to see the consequences of their decision to become the "Pravda of Amerika."

The fallout from this will be massive.

rofl!  You obviously have your finger on the pulse of America.  Or not.

Pulse of America, no.  Pulse of right-wing Republicans in this country (e.g. those who post on FR), possibly.

[Sam Spade]

So Sam, does this mean that there were two really good McCain samples, one really good McCain sample, or something else?

When the chart of the last 7 days is (from beginning to end):

                           (D1)  (D2)  (D3/BO)   (today)
+/-0, O+3, O+5, O+8, O+8, O+6,        O+4

It stands to reason that the samples of the last two days were better than the two that fell off.  In fact, it is probable that this extends to all three days.

Much more than that, I can't say.

[Sam Spade]

Personally, I think Rasmussen has an outlier sample in its mix (from yesterday), but whatever.  I also don't think Obama is up by 11.

[Sam Spade]

Important note in today's update:

Obama's current advantage is slightly less when estimating the preferences of likely voters, which Gallup will begin reporting on a regular basis between now and the election. Gallup is providing two likely voter estimates to take into account different turnout scenarios.

The first likely voter model is based on Gallup's traditional likely voter assumptions, which determine respondents' likelihood to vote based on how they answer questions about their current voting intention and past voting behavior. According to this model, Obama's advantage over McCain is 50% to 46% in Oct. 9-11 tracking data.

The second likely voter estimate is a variation on the traditional model, but is only based on respondents' current voting intention. This model would take into account increased voter registration this year and possibly higher turnout among groups that are traditionally less likely to vote, such as young adults and racial minorities (Gallup will continue to monitor and report on turnout indicators by subgroup between now and the election). According to this second likely voter model, Obama has a 51% to 45% lead over McCain. -- Jeff Jones
As Vorlon has pointed out, it is now at this point in the race (second week of October or so) when Gallup's LV model starts becoming useful to us. 

Hopefully, they will post this every day...

[Sam Spade]

All tracking polls for today:

Gallup: Obama +10 (+3)
Rasmussen: Obama +5 (-1)
R2000: Obama +12 (-1)
Battleground: Obama +8 (-2)
Hotline: Obama +6 (-2)
Zogby: Obama +4 (-2)

So Obama loses ground everywhere, except in Gallups poll where he gains 3 points and yet again the polls are all over the place. The only constant is that Obama leads by at least 4 points.

If Gallup only looks at likely voters Obamas lead is cut to 7 points.

Battleground was unchanged.

[Sam Spade]

10/14/08

Obama 51 (nc)
McCain 42 (+1)

is this LV or RV?

We really need to post all three:

RV
Obama 51
McCain 42

LV (new voter formula)
Obama 53
McCain 43

LV (traditional)
Obama 51
McCain 45

[Sam Spade]

LV (new voter formula)
Obama 53
McCain 43

LV (traditional)
Obama 51
McCain 45

which one is better?

Since they really don't describe how they come to a determination, it is hard for me to say what exactly the method is and why one should be better.  My question for Vorlon to ask Gallup is which of the 13 questions do they remove in the new model.

I note that the traditional determination (for today) produces a 76.6% RV turnout, whereas the new voter model produces an 82.0% RV turnout.

RV turnout in 2004 was 72.9%.  Therefore, I would tend to go with the traditional model.

[Sam Spade]

October 15, 2008

RV
Obama 50 (-1)
McCain 43 (+1)

LV (new voter formula)
Obama 52 (-1)
McCain 44 (+1)

LV (traditional)
Obama 49 (-2)
McCain 46 (+1)

[Sam Spade]

I personally suspect the sample dropping off tomorrow is a pro-Obama sample, but with the Dow doing as it is today, it may be replaced by a similar one.

[Sam Spade]

My guess is that the 2nd likely voter model has the party identification as 37-37 D-R like in 2004.

And that is based on?