If you were a bookie... Polls and bets on the 2004 elections

The cell phones aren't the point, they are just a metaphor.

The point is that the polls are as much art as science. Pollsters do a lot of number crunching and estimating. The figures are always adjusted for this or that, and these adjustments are often little more than guess work.

One big adjustment (bigger than cell phones) is what constitutes a "likely" voter. Pollsters ask several questions about emotional state (how excited are you) and previous voting history and then make a calculation about the chances you will make it to the polls. If your score is too low, you are not a likely voter and your poll answers are discarded.

I am confident that the number of "likely" voters is being significanly underestimated. It is certain that many more people will vote this year that didn't vote in 2000. It is also clear that the greater the turnout in swing states, the better it is for Kerry.

I am quite confident of a Kerry victory for many reasons. The biggest of which is that the Democrats have done what everyone knew they needed to do. For the past few weeks (at least) the race has been about Bush, not Kerry.

This is the fact and Bush loses... probably convincingly..

Nimh, good evening. And good evening to yall.
My thinking is the time for polls is over. Tomorrow is the day. I am still optimistic but certainly not confident that my candidate will win. It should be an interesting, but perhaps very long, evening.
(I can't resist a final comment on the cell phone/polling issue which first came up many pages ago. I concede that it may be a non-issue, but, nimh, if you take the numbers that you spongily (ha) used in your post of 5 hours ago, massage them a bit, 20 to 30% of younger people having only cell-phones and so on and on instead of your numbers, you could come to a different conclusion. The post-mortem on polling should be interesting).

Johnboy will be one of the first people at the polls tomorrow, 6:30 am or so.
He won't be the first one there. Some guy, some older man, was always #1 until some other guy, equally as old, moved into the precinct and decided he wanted to be 1st to vote. Quite a battle, there. They are both Republicans, I think. -johnboy-

Hmm, try it, it's instructive (at least it was to me). I tell ya, you'd hafta really crank up those numbers to get anywhere near a statistically significant impact.

State polls today are not hopeful. Quinnipiac has Kerry leading by 5% in New Jersey, but the race tied in Pennsylvania and Bush ahead by 8% in Florida. SurveyUSA has Kerry up by 4% in Washington, just 1% in Pennsylvania, a surprising tie in Arkansas but Bush up by 1% in Florida, 2% in Ohio, 5% in Missouri and 8% in North Carolina. In fact, three polls came out today on Ohio, and they all have Bush ahead (SurveyUSA, Fox and Univ of Cincinnatti). Fox also has Bush up by 3% in Wisconsin and 4% in Iowa, though it has Kerry up 5% in Florida. Finally, a UNH poll has Kerry up by just 1% in New Hampshire.

Meanwhile, the missing WaPo tracking poll today has Bush gaining 1% to 49%, and Kerry steady at 48%, with Nader back up to 1%.

The 11:00 Line moved dramatically back towards Kerry again. I'll be real curious if they run another tonight.

Nimh's poll numbers on Florida:
(Quinnipak: Bush by 8)
(Survey USA: Bush by 1%)
(Fox: Kerry by 5%)
Enough already with the polls. johnboy's grandmother tried to learn to speak English but she had trouble sometimes with idioms. As she might say "It's time to put this puppy to sleep." -rjb-

Here's the updated graph on the weekly average of all national polls. (Sorry johnboy, but this is what I do in this thread - its what the thread is about, an' all ;-))

Last week in is the week that ended yesterday. Tomorrow's tracking polls (will there still be tracking polls tomorrow?) will still have to be calculated in as well, plus any other polls that might still appear tomorrow, but this average of B +1,1% is already based on the data of 16 pollsters.




Here's the graphs with the individual polls, first three-way then two-way races (the latter including much fewer polls). Note that the lines of some pollsters in the 3-way race graph are already drawn onwards into week 45. Thats the polls that have stated a projected outcome of the elections. Eg, Gallup's last poll had Bush at 49 and Kerry at 47, but they added a projection for the actual election outcome to be 49/49. Thats in the graph as their "week 45" data.

And the pollsters' data on Bush's job approval ...

Many pollsters polled his approval rating several times this past half-month, so the graphs are based on the average of all polls of the half-month for each pollster.

Tomorrow it'll be all about exit polls Nimh. So what's your final answer... who wins?

Never claimed the Zogby poll had Bush bias. Quite the opposite, I have repeatedly said that Zogby is the poll you should most pay attention to.

Now, Timberland, let me tell you a thing or two about Rasmussen.

After hearing this Rassmussen poll being quoted all the time on talk radio-and almost nowhere else-I went to it's website. At that time it was called the Voice of America poll, or some such.

When I got there, I saw SEVERAL links to obviously right wing websites. For instance, one of these listed all the people Clinton supposedly murdered on his way to the top. That kind of thing.

The posting forum had ONE guy criticizing the poll, and he was surrounded by people saying such things as "Bush is for the Godly people in America!" and similar comments.

The one critic pointed out that Rasmussen had but two clients, the Washington Times and some other right wing site. That's it.

After looking at this mess, I found pollingreport.com doesn't even carry the Rasmussen poll, realizing the whole "poll" is a fake.

On top of this, in one of the closest elections in history, with Gore achieiving a mere 0.5% victory in the popular vote, Rasmussen shows a landslide margin of victory for BUSH.

I mean, can anyone SUCK any worse?

Now I ask you gentle folks, especially timberland: How can any individual possibly regard Rasmussen as anything other than as a fake poll for hire to the Republicans?

Honest-please give me a cogent, honest reason to take Rasmussen seriously.

Cant speak for Timber but I get your points Keltic.

Reason pollingreport.com doesnt carry Rasmussen however I think is because it was a daily tracking poll all along.

And though Rasmussen was way off in 2000, its data now are right in line with those of other pollsters.

I've gone for a boldly optimistic prognosis ...

Until, of course, Rasmussen comes out with an unscientific poll which shows the undecided vote going to the incumbent, Bush,this year instead of the other way around.

They're being more subtle this year. They are waiting to the end, then coming up with polls which supposedly show the undecideds going for the incumbent this year, when they traditionally go the other way.

Simply put, 2000 revealed Rasmussen for what he is, and I am amazed anyone pays attention to him at all. I frankly do not care if this year he finds it convenient to follow the crowd a little more in his estimates.

KW, lemme disabuse of of the notion I hold Rasnussen in any greater esteem than any other pollster. From the go, I've maintained no one poll, or pollster, in and of itself apart from its relationship to its contemporaries over time, is of much use or significance.

And whether you find it difficult to believe or not, Scott Rasmussen is a Democrat.

{And a special thanks to nimh} :wink:

Timberland, you have carefully avoided the points in my post.

How many times do you plan to do that?

Links to right wing websites listing the people Clinton was supposed to have murdered on his way up.

Only two right wing periodicals as his customers.

Predicting a landslide for Bush in an election when Gore eaks out a close victory.

And we are supposed to take this fellow seriuosly at all?

Don't give me any gobbledy-gook about you hold Rasmusssen in no higher esteem than any other pollster.

Rasmussen made it clear he wasn't even TRYING to be objective.

I repeat-in light of what I pointed out, how can you put Rasmussen in the mix with other pollsters, who at least try to be impartial?

Yes, and when Culture Club first became popular, Boy George was telling interviewers that in real life, he was "quite masculine" and the dress was just a stage act.

People tell interviewers whatever seems convenient at the time.

KW, I don't avoid your points, you apparently do not understand, or perhaps you reject, my point. I don't see Rasmussen as anything special one way opr the other. And while you may not take Rasmussen seriously, enough folks do look at his polls, whatever they might think of them, to make his one of the largest, and fastest growing, polling sites on the 'net, according to ALEXA. Credible or not, partisan or not, Rasmussen is a factor among those who factor such things as political factions.

Oh, and for those interested, the traditional first town in the nation to vote has done so. In Dixville Notch NH, its Bush 19, Kerry 7, according to the AP newswire feed.

(edited to correct a spectacularly screwed-up link)

Okay, I used my secret formula and arrived at the answer. It is:
Bush...... 296
Kerry..... 242
(Better hope Bush doesn't lose Florida )

Florida is going for Kerry. But I don't think he will even need it.

I predict Kerry with 301 evs.

Are you prepared for a tie?

http://slate.msn.com/id/2108751/

269 - 269, Bush wins since the selection of the President would go to the House where each state gets one vote. And there are more Bush states than Kerry states.

So Bush can win a second term with either an electoral vote win or a tie.

Kerry has to win the electoral vote to unseat Bush.

I just heard Zogby has Bush up by 6 in Florida!

Can this be true???

<yay>