Gallup Tracking Poll Thread [Obama vs McCain]

[Alcon]

I don't understand why you didn't check RCP for yourself before correcting someone, then.

[J. J.]

I don't understand why you didn't check RCP for yourself before correcting someone, then.

Because we've been discussing it right before I posted, it 4 hours ago.  At that point, there was a slight closing.

[TheGlobalizer]

He's always been a troll, but the troll is right.

It's also now +5.8.  Somehow I'm trusting realclearpolitics.com more on the realclearpolitics.com average than something Peper11 posted 10 hours ago (call me zany, call me nuts.)

I don't view the RCP average as inerrant truth.  The 7 day trend is towards closing, even if a few outlier polls are bumping the average up a bit in the short term.

[Alcon]

He's always been a troll, but the troll is right.

It's also now +5.8.  Somehow I'm trusting realclearpolitics.com more on the realclearpolitics.com average than something Peper11 posted 10 hours ago (call me zany, call me nuts.)

I don't view the RCP average as inerrant truth.

Muy excelente

The 7 day trend is towards closing, even if a few outlier polls are bumping the average up a bit in the short term.

I generally agree, but I don't see much evidence that it's anything beyond regression toward the mean, which is rarely linear.  And even if it were, McCain would be tied at best Election Day -- and would therefore almost certainly lose.  McCain needs something to happen.

[J. J.]

He's always been a troll, but the troll is right.

It's also now +5.8.  Somehow I'm trusting realclearpolitics.com more on the realclearpolitics.com average than something Peper11 posted 10 hours ago (call me zany, call me nuts.)

I don't view the RCP average as inerrant truth.  The 7 day trend is towards closing, even if a few outlier polls are bumping the average up a bit in the short term.

It was closing last week, but it stopped Friday or Saturday.  Some of it was pro-Obama sample dropping out of Gallup.

[freedomburns]

You must be logged in to read this quote.

Could you be bothered to refresh me on what regression to the mean means?  I seem to remember some of this from General Ed statistics at UCSC twenty years ago...

[muon2]

You must be logged in to read this quote.

Could you be bothered to refresh me on what regression to the mean means?  I seem to remember some of this from General Ed statistics at UCSC twenty years ago...

Regression toward the mean is the effect of looking at outliers of a sample and finding them to be closer in the next measurement. This effect is strongest when the samples and the method to collect them are identical and dominated by statistical error. In that case one expects that an outlier is really just a statistical fluke and the next measurement will have high probability of landing closer to the mean.

This does not apply if there are systematic differences in the samples. It also should be used with caution in polls, since it really applies to individual measurements (like batting averages or exam scores), and is less applicable for a measurement which is itself a statistical sample of measurements.

[TheGlobalizer]

I generally agree, but I don't see much evidence that it's anything beyond regression toward the mean, which is rarely linear.  And even if it were, McCain would be tied at best Election Day -- and would therefore almost certainly lose.  McCain needs something to happen.

Agree, I just pile on top my conjecture that the party ID weighting is a bit to the left of where it should be.

I think it's a 2-4 point Obama advantage today, without a foreseeable prospective bump for Obama (barring grandma's passing (hope not) creating a sympathy vote).  It's a horserace, and I think McCain can even pull out an EV win on a -1 PV margin, given the states that are in play.

Not saying it's likely today, but it's not much of a stretch, either.

[© tweed]

10/21/08

RV

Obama 52 (nc)
McCain 41 (nc)

LV expanded

Obama 52 (nc)
McCain 42 (-1)

LV traditional

Obama 51 (+1)
McCain 44 (-1)

[ChrisFromNJ]

10/21/08

RV

Obama 52 (nc)
McCain 41 (nc)

LV expanded

Obama 52 (nc)
McCain 42 (-1)

LV traditional

Obama 51 (+1)
McCain 44 (-1)

GALLUP SHOCK: OBAMA UP 7

[Sbane]

Hmm wonder why Rasmussen and Gallup disagree so much. Change in party ID?

[Eraserhead]

Awesome.

With numbers like this, he almost has to take a fall tomorrow though.

[Wiz in Wis]

Awesome.

With numbers like this, he almost has to take a fall tomorrow though.

agreed, though I think its just as likely that he picks up a point or two in Rass, so it will probably be a wash. Interesting that all the trackers post between a 1-11 point Obama lead. Some spread there.

[Lief 🗽]

God, those numbers are delicious.

[Ronnie]

How much I'd love for us to lose the election by just a couple points.  Doubt it's gonna happen, though.

[freedomburns]

You must be logged in to read this quote.

Could you be bothered to refresh me on what regression to the mean means?  I seem to remember some of this from General Ed statistics at UCSC twenty years ago...

Regression toward the mean is the effect of looking at outliers of a sample and finding them to be closer in the next measurement. This effect is strongest when the samples and the method to collect them are identical and dominated by statistical error. In that case one expects that an outlier is really just a statistical fluke and the next measurement will have high probability of landing closer to the mean.

This does not apply if there are systematic differences in the samples. It also should be used with caution in polls, since it really applies to individual measurements (like batting averages or exam scores), and is less applicable for a measurement which is itself a statistical sample of measurements.

Thanks, muon2!

[Sam Spade]

10/22/08

RV

Obama 51 (-1)
McCain 42 (+1)

LV expanded

Obama 52 (nc)
McCain 44 (+2)

LV traditional

Obama 50 (-1)
McCain 45 (+1)

[Eraserhead]

I called it.

[Reds4]

I wonder if this means McCain had a good night last night or just that a massive Obama night fell off.

[J. J.]

Probably weekend numbers dropping off.  Unless there are solid sustained movement across polls, Obama has it.

[Sam Spade]

Gallup doing its usual midweek movement.

10/23/08

RV

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

LV expanded

Obama 51 (-1)
McCain 45 (+1)

LV traditional

Obama 50 (nc)
McCain 46 (+1)

[Eraserhead]

lame

[J. J.]

Interesting, and not repeated in Rasmussen.

[Sam Spade]

lame

To beat out Hawk, this is an article from Gallup that I find most fascinating.

http://www.gallup.com/poll/111331/No-Increase-Proportion-First-Time-Voters.aspx

(conducted October 17-19 2008 among 2700 RV)

[J. J.]

lame

To beat out Hawk, this is an article from Gallup that I find most fascinating.

http://www.gallup.com/poll/111331/No-Increase-Proportion-First-Time-Voters.aspx

(conducted October 17-19 2008 among 2700 RV)

J. J.'s Second Rule of Elections?