Gallup Poll Results on Drudge

-- Ain't Gonna Happen (Not Here, October 27, 2000


Poor Al is dead

a candle lights his head

He's lookin' so peaceful and serene

He looks like he's asleep

It's a shame that he won't keep

but it's summer and we're runnin' outta ice

Poor Al

-- (, October 27, 2000.

Almost time to put a fork in him...

-- Ain't Gonna Happen (Not Here, October 27, 2000.

That Drudge sure is a dumb right-winger.

The Clinton News Network uses reverse psychology to get more people to come out and vote for Gore, guaranteeing an overwhelming victory, and what does the sludge-drudger do?... help out the lefties by spreading this propaganda.


-- (righties@not.2smart), October 27, 2000.

Drudge again gets some of the interesting news out first. The Gallop poll went from a +7% Bush lead (Oct.23-25) to a +13% lead (Oct. 24-26), a pretty significant increase, so Drudge's headlines appear appropriate. Interestingly though, is how the news outlets commsioning the poll chose to play it, First, CNN: Tracking poll: Bush holds on to advantage October 27, 2000

WASHINGTON (CNN) -- Today's CNN/USA Today/Gallup tracking poll continues to give George W. Bush an advantage over Vice President Al Gore. ---------- Hmmm...a five percent increase, and Bush "holds on" to an advantage? LOL....
Now lets hear from USA today: Presidential preference poll

Friday's results: Texas Gov. George W. Bush jumped to a thirteen percentage point lead over Vice President Al Gore Friday, up from a seven point lead Thursday. Bush leads Gore, 52% to 39%. This is Bush's largest margin over Gore since August. ---------- USA Today at least paid attention to the figures. Both downplayed it a great deal. Maybe they looked around at Zogby and other polls, and said, "somethings wrong here"...(during the same period, Zogby had Gore ahead, Now Bush back ahead). Most polls show Bush leading anywhere from 3 to 8 points, and I would guess the average at around a four point lead. What is really hard to predict is who will vote on election day....and for the Democratic supporters, how many ;)

-- David (, October 27, 2000.

Bold off, in theory...

-- David (, October 27, 2000.

Of course, theories are often proven incorrect,,/b> as I have just demonstrated.

-- David (, October 27, 2000.

Zogby is the most accurate, showing Bush with a 1 point lead, 44 to 43.

Friday October 27 6:34 PM ET Bush Edges Ahead Again in Reuters/MSNBC Poll

By Alan Elsner, Political Correspondent

WASHINGTON (Reuters) - Republican George W. Bush (news - web sites) reclaimed a one-point lead over Democrat Al Gore (news - web sites) in Friday's Reuters/MSNBC daily tracking poll, as the U.S. presidential race remained too close to call.

The survey of 1,206 likely voters in the Nov. 7 election, conducted Oct. 24-27 by pollster John Zogby, found the Texas governor with 44 percent and the vice president with 43 percent. That represented a two-point decline for Gore and a one-point advance for Bush over the past 24 hours. Eleven days remain until the Nov. 7 election.

Green Party nominee Ralph Nader (news - web sites) polled 5 percent; Reform Party candidate Pat Buchanan (news - web sites) stayed at 1 percent and the rest remained undecided. Gore had spent three days in the lead before Friday's survey.

The race remained well within the statistical margin of error of plus or minus three percentage points. A candidate would have to be more than six points in the lead to be outside that zone of uncertainty -- something neither man has achieved since the poll began on Sept. 29.

``Bush had a solid day and has now reclaimed his lead over Gore among men and among parents with children at home,'' said Zogby. ``Gore has been diligent lately in his attacks against Nader. But with Nader still receiving 13 percent from independents, 13 percent from progressives and 8 percent from 18-24 year-olds, it appears these attacks, at least in their initial phase, perhaps are not working.''

In a race this close, national polls cannot predict a winner because the election is likely to be decided in key swing states. Reuters and MSNBC will begin publishing daily tracking polls in nine such battleground states on Saturday while continuing the national surveys.

In the equally tight race for the House of Representatives, voters prefer the Democrats by four points. The Democrats need a net gain of seven seats to regain control from the Republicans.

When undecided voters were asked which way they leaned in the presidential race, they broke more than 2-to-1 for Gore.

Eighty percent of respondents said they had definitely made up their minds. Forty percent have ruled out voting for Bush and 44 percent for Gore.

Gore leads Bush by double digits in the East while Bush is equally dominant in the South. The two were tied in the West and in the all-important Midwest region where many experts believe the election will be decided.

The vice president leads overwhelmingly among voters aged 18 to 24, who in past elections have been the least likely to vote. Other age groups are closer.

While Bush is winning the support of more than 84 percent of Republicans, Gore has in recent days increased his backing among Democrats to 80 percent. But Nader is still winning 7 percent of the Democratic vote, hurting Gore.

Bush leads among white voters by 50 percent to 37 percent. But Gore is winning substantially among Hispanics and is taking more than 82 percent of the black vote.

Reuters and MSNBC will release a new poll every day until the election at 6:30 p.m.

-- (zogby@best.poll), October 27, 2000.

Slate has a wonderful essay on why these polls produce such different results. Sorry I don't have a link. It explains statistical concepts so they're clear to non-mathematicians, and shows how it's impossible NOT to introduce bias into a poll, but possible to USE the inevitable bias for different (but valid) purposes. Gallup's goal is to identify changes and trends, and their polls are designed to identify these. Zogby's goal is to describe the electorate's underlying traits (as opposed to transitory preferences), which are necessarily less volatile.

Zogby's approach is probably better at predicting the winner, while Gallup's approach is probably better at determining who won which debate. So these different approaches each have their own strengths and weaknesses, and which is "better" in an absolute sense is a meaningless question. It all depends on what you are trying to find out; otherwise you are asking whether a telescope is inherently "better" than a microscope *in general*. Silly question, leading to a silly answer.

-- Flint (, October 27, 2000.

I agree that Zogby is one of the better polls, but no longer consider it the best. The best are the non-media polls, the ones done for the parties...the Republican party, the Democratic party, more scientific, weighted for the % of Rep/Dems responding, etc.

Zogby is a rolling (typically 3 day) average poll. Other good media polls are the Battleground poll, and the Potrait of America tracking poll (3 day average, +/- 1.8 percentage points with a 95% level of confidence). I have found the POA poll even more consistent than Zogby, and its accuracy is arguably better.

-- David (, October 27, 2000.

"which is "better" in an absolute sense is a meaningless question"

Bullshit. As usual, Flint is trying to impress us with his overinflated ego by being overly analytical. The better poll is the one which has been closest to the actual outcome more times than the others.

-- (just@common.sense), October 27, 2000.

One study indicated that the most accurate polls for the primaries was POA (Rasmeutten (sp?), followed closely by get this...Gallup, and then Zogby. I thought flint offered decent discussion of the of polling methodology, saw no BS there...

-- David (, October 27, 2000.

I stand by my conclusion, that the one with the highest success rate is the best.

-- (just@common.semse), October 27, 2000.

I agree with your conclusion, "just".

-- David (, October 28, 2000.

common sense:

Interestingly, the last paragraph of that Slate article spoke of predicting the 1988 election. The 3-day rolling average (which they'd been publishing all along) got the wrong guy, while the 4-day average (which they'd never published) had it right. So of course *after the election* they touted their *four* day rolling average, and claimed they nailed the final result, so their polling technique was validated!

NOW, here you are saying methodology doesn't matter, only the accuracy of the prediction. Un huh. Maybe if you understood what was going on under the hood, you'd have a more realistic viewpoint. Did they predict correctly or not? Well, the 4-day average got it right, the 3-day average got it wrong, the 2-day average got it right....which one do YOU pick ahead of time? Which one is "real"? YOU have the "common sense", so tell us. We're all ears.

-- Flint (, October 28, 2000.

Such a smartass Flint. You are the one who started blabbing on and on about methodology, not me.

You stated that it is "meaningless" to try to say that one type of poll is better than the other. I say, BULLSHIT! The poll which is the "best" (IMO, that would mean "most accurate") is simply the one which has been correct more times than the others!

Why is THAT so hard to understand? You have a serious problem with trying to over-intellectualize everything (just as you did with Y2K), because you think you can impress everyone with your meaningless details. I'm NOT impressed.

If I want to know which poll is best, I simply look at the track record. You are welcome to do some research and find out the truth about these different polls, but I have a feeling you will get hung up in the "methodology", because YOU think that makes you sound more intelligent. LOL

-- (just@common.sense), October 28, 2000.

Nailed him good, Flint. Trouble is, he don't know it.

-- Brian McLaughlin (, October 28, 2000.

Oh yeah, Brian. Go ahead and kiss Flint's ass because he is "so much smarter" than you. LOL! His strategy works with people like you, because you are too dumb to see what he does.

If the San Francisco 49'ers have the best record for the season, then go on to win the Superbowl, they are the best team. Flint will tell you, however, that that isn't true. He will tell you that it depends on what the meaning of the word "best" is. Is it the best running game, is it the best defense, is it the best offensive coaching strategy at utilizing their wide receivers? Nice way to befuddle and confuse dimwits (or as Flint would say, "kick the anthill"), but the best team would still be the 49'ers. Duuuuuh.

-- (brian@flint's.butt.smoocher), October 28, 2000.

Which poll produces the more accurate result?

Neither. It is only he poll held on the day which will produce an accurate result. All prior polls are only working on a sample of the population, and can be no more than an aproximation of what some part of the population are feeling at the time.

As I do not lve in USA I can view the coming election with impartiality, and no matter which candidate wins, my life style will not change. I know little about each candidate's policies or history other than the litle that is published in the media over here. However from the information that is made available to us it does appear that on a couple of issues I am familiar with, Gore has some firmly fixed ideas, while Bush has commented that more research needs to be carried out. I must admit to being impressed that a politician has the honesty to admit that there may be issues on which he is not alreay an expert.

-- Malcolm Taylor (, October 28, 2000.

How come Bush does not have the honesty to admit that he used cocaine, and that he has been involved in an abortion with at least one woman?

-- (shrubya@not.very.honest), October 28, 2000.

Who needs polls? When you have Larry King on CNN telling Dick Cheney he expects to see alot of him on TV after the election, the race is over.

-- Buster Collins (, October 28, 2000.

Moderation questions? read the FAQ