I’m so confused . . . Everybody talks about Bush being ahead in the polls and the only issue is by how much. But this story (http://www.inthesetimes.com/site/main/article/on_the_rebounce/) from the October 11 issue of In These Times says Kerry has had a wavering but solid lead in the electoral vote count since July. Who’s right? Different poll-trackng websites give different answers. http://www.electoral-vote.com/ gives the tally as of today (9/29/04) as 241 electoral votes for Kerry, 273 for Bush, with only one state (Minnesota – 10 electoral votes) being “exactly tied.” http://www.race2004.net/, on the other hand, gives 165 votes for Kerry, 264 for Bush, with 109 “undecided” – presumably, because the latter site has wider criteria for classifying as state is “undecided.”
electoral-vote.com, BTW, says nothing about the popular vote. race2004.net says that if the election were held today, Kerry would win the popular vote by 49.21% to 48.71%; but Bush would win the electoral vote.
First of all, national polls are kind of meaningless. What matters is winning states.
Second, the polls I’ve seen on a state-by-state level include a lot of states where 1) It’s a tie, or 2) The lead one of them holds over the other is by only a couple of points.
Quite frankly, I don’t know who’s ahead, and most who’ll say otherwise are partisan to one of the two.
[minor hijack]–it seems to me that this is the first election where there have been so many totally contradictory polls. When polls vary from 13% lead for Bush to a 1% lead for Kerry–thats proof that the whole “science” of polling is as much guesswork as it is science.
Personally, I’m hoping that the pollsters will have to admit it finally,and in the future we’ll see less reporting of polls and more about the issues. (Unlikely, I know…but it’s a nice fantasy)
Unless the spread is well outside the margin for error, I don’t think polls mean too much. They are great ‘nyahnyah!’ ammo, but this +/- 2-10% or so we are getting doesn’t mean much, if anything.
[tangent]What’s going on in Wisconsin? electoral-vote.com has it as ‘strong Bush’ based on a 9/21 poll. I thought Wisconsin had been leaning (slightly) Kerry, but it just jumped to strong Bush.
Is it real? Is it because Kerry said ‘Lombard field’ or whatever he said instead of Lambeau? Or is it a bad poll?
If you click on any state in the map of the welcome page in race2004.net, it will take you to a page showing all the polls done this year, including the date, results, and polling agency. If you click on any state in the map of the welcome page of electoral-vote.com, it will show the same info in graphic form over time, with a red line representing support for Kerry and a blue line for Bush. From searching the electoral-vote.com site, I note that the graphs for all the “battleground states” or “swing states,” including Wisconsin (and Florida, Pennsylvania, Michigan, Ohio, Missouri, New Mexico, Arizona) show wild variations – the lines touch and cross and zig and zag and the results seem to depend as much on who did the poll as when it was done. If you click on Texas, you’ll see a more-or-less straght line for Bush holding steady above the 50% line, and a more-or-less straight line for Kerry holding steady below the 40% line; you’ll also see far fewer polls than are represented on the swing-state pages – which is understandable; Texas is such a solid lock for Bush, why should they bother to poll it?
Most of the polls I’ve seen for the individual states (the only important polls IMHO…national polls are worthless as we don’t have popular elections but state EC elections) seem to be pretty much within the margin of error. Translation: no one knows who’s really winning at this point, nor will we really know until election day is over and we count the ACTUAL votes for each candidate and tally up the EC votes and declare a winner…its too close to judge right now.
My gut feeling is Bush is slightly ahead atm, simply because I’ve been seeing more and more stuff about Republicans ‘stealing’ the election from the left (as well as other things in a similar vein)…and it seems more and more strident, like they are trying to set up excuses in case they lose. If Democrats were really confident right now I doubt they would be playing that card as hard…the Republicans seem pretty confident and aren’t doing similar things.
Thats just MY gut feeling though, which is essentially as meaningless as most of the poll data. Basically I think the debates are going to be the deciding factor…they will make or break whichever candidate comes out APPEARING to have done better.
Well, yes and no. Only the state-by-state results will determine who is president. But if one candidate wins the electoral vote and the other wins the popular vote, that will have political consequences – it will diminish the winner’s perceived mandate, and inspire more demands for abolishing the EC.
Perhaps…but it won’t effect who is the next president. I also doubt there will be much of a movement to abolish the EC even IF it happens again that the person who wins the EC vote doesn’t win the popular vote…I just don’t see it happening.
So, in a very real sense, the state by state polls would be the only important ones…IF they were reliable enough to trust. However, if the polls are close enough to be within the margin of error (which most states are if memory serves), then its really too close to call at this point, and probably will remain so until they tally things up on election day and announce a winner. My only caviot to that is…what is the PERCEPTION of the polls in most American’s minds. If the PERCEPTION is that GW is winning, or clearly in the lead anyway, this could make it the REALITY that GW is winning or clearly in the lead. Its sad but true.
I expect things to tighten up even in the polls after the debates, as I don’t think there will be a clear ‘winner’ there…and this thing will essentially come down to the wire. I just wish they were letting some of the third parties participate in at least one of the debates.
The big headline on McPaper (USA Today) shows a deadheat in the latest USA Today/Gallup Poll. Of course it’s sort of meaningless in re the Electoral College, but, IMObservation, there seems to be a bit of a sociological version of the Van Heisenburg (badly spelled?) Uncertainty Principle, wherein the polls affect the voters. Sadly there are some people out there who will always go with a winner/whoever everyone else seems to vote for. The good news in a dead heat is that it might actually require some of these people to think for themselves. In that case I’d still give an advantage to the sitting president since he’s the devil that is known. Right now the EC vote seems locked for Bush, but expect that to change slightly over the next 28 days.
One note - although the site says “all info updated 10/3/04,” that doesn’t mean that all the polls are current through that date. Several states don’t have post-first debate polls. So it’s relying on a mix of polling data, some of which isn’t as current as it could be, and projecting from that to a conclusion that “if the race were held today, Bush would win.”
Good point. Furthermore, a lot seems to depend, for some reason, on who’s doing the polling. To iron out these differences, http://www.electoral-vote.com/ is changing its methodology:
But that hasn’t been done quite yet. The map shows 243 votes for Kerry, 295 for Bush, but none of the state polls date from after Thursday’s debate.
I’m not buying it. I say wait a few days, or even until mid-week next week, then look again. I think its tightened up and is basically a dead-heat atm…and will remain so until the election. Then flip a coin…its going to be THAT close.
There was an election by EC on the SDMB recently btw…care to guess who won EVERY EC vote?
Re: methodology - from what I’ve heard there’s a whole lot of wiggle room.
For instance, the polls of debate watchers drew from an audience 35% Democrat, 35% Republican, 30% Independent – but that’s not the makeup of the population.
Isn’t it Gallup that bases its Democrat/Republican blend on previous voter turnout? And haven’t there been a lot of reports of new Democrat registrations outpacing Republican? I know that’s the case in Ohio.
I don’t think they’re polling 1,000 people randomly, and I don’t think they really know who is going to vote this year.
I take polls with a lot of salt. For someone to give projected vote totals down to the units digit is preposterous and that pollster must have been asleep in statistics class when “significant figures” was discussed. The polls use scientific samples of the voter rolls, and different groups are represented according to their tendency to come out and vote. With the huge influx of new voters this year as witnessed by many media stories of the interest in this election, these new voters may not vote as you might demographically expect. Also, you can count on the right coming out in droves every year. The pro-life crowd, the anti-gun control crowd, and the anti-gay acceptance crowd all vote and all vote Republican. But now the rest of the voters are energized and those that typically don’t vote are coming out. My bet is that they all break Democratic and Kerry wins in a walk.
What odds are you giving? Not that Kerry wins, but that he ‘wins in a walk’? I think whoever wins its going to be as razor close as the 2000 election was…unfortunately. Unless something major breaks in the next few weeks of course, or there is some major screw up in the debates.
Cite? I mean, state and national polls are done by the same kinds of organizations, aren’t they?
Basically I think the debates are going to be the deciding factor…they will make or break whichever candidate comes out APPEARING to have done better.