2004 User Predictions - Discussion

[Gustaf]

why have both of you given Kerry Ohio and Bush Pennsylvania? Surely if Bush wins PA he will win OH.

Not necessarily...depends how the philly latte-drinking liberals vote.  If they vote for Nader Kerry might not win PA.

Boss Tweed,
   Hi, Mr. Devils fan it's earth here.  Most Philly Democrats are not latte drinking liberals.  In fact a lot are pro-life and pro-Iraq war.  Ther reason Philly votes Dem is because of Jesse Jackson and the big unions.  The suburbs have a mix of union workers and bleeding hearts though.  I don't think those voters will be going Nader.  Maybe someone livingin Society HIll though.  

He didn't say that most were...if one in 10 Gore+Nader votes vote for Nader again, that could be enough.

The electorate is aware of the very close 2000 election. There will be very-very-very little votes for Nader this time around.

Oh, I agree. Note the 'if' in MiamiU's statement.

Still, if the race is close, like in Florida-2000-close, every single vote could matter. If the margin of victory is around 0.01% like it was then, Nader could still give the race to Bush. In fact, latte-drinking liberals probably make up about 0.01% of the PA electorate...

[Gustaf]

If that happens I fully hope that Gustaf has a room set up for me in Sweden. I hear the Swedish meatballs there are excellent. lol

Sure.

[Gustaf]

http://www.uselectionatlas.org/USPRESIDENT/GENERAL/CAMPAIGN/2004/pred04.php?action=indpred&id=3267

Kerry does well enough to take Alabama, Georgia, Arkansas and Arizona but Bush manages to take a slight lead in the EC by taking New York.

The guy is from Alabama.  That might have something to do with it.

He's a Democrat, but despite having a seriously odd map, he still has Bush winning.

[Gustaf]

UPDATED Kerry v. Bush:


Bush/Cheney 394
Kerry/Edwards 144

Switched IL, ME, and DE to Bush.

Wow. And how comes CA is still Dem and IL isn't?

And I think you're going over-baord now. It isn't gonna be THAT bad.

[Gustaf]

Wow. And how comes CA is still Dem and IL isn't?

And I think you're going over-baord now. It isn't gonna be THAT bad.

1. IT would take a 650K vote swing in CA, only a 240K or so swing in IL.

2. It will be that bad

%-swing is more imprtant than in absolute terms. It would take a 45K vote swing in Wyoming, but that doesn't mean it's gonna happen.

[Gustaf]

I think that DC is the only place that will break 60%. So, I am using the colors to represent the range within the 60% max range. The colors from lightest to darkest represent: <50%, 50%-53.3%, 53.4%-56.6%, and 56.7%-59.9%.

Here's my election night prediction:

And then that one CD in NE goes to Kerry, right?

[Gustaf]

Not at all unreasonable Kelpie. In fact, pretty likely.

[Gustaf]

StatesRights, though asking 500 000 people is techically feasible, think about the costs...this is all statistical, an MoE of 3% in 95% of the cases is sufficient in most peoples minds.

The state-owned polling firm in Sweden, SCB, make 2 big polls a year, with a sample of about 9 000.

[Gustaf]

StatesRights, though asking 500 000 people is techically feasible, think about the costs...this is all statistical, an MoE of 3% in 95% of the cases is sufficient in most peoples minds.

The state-owned polling firm in Sweden, SCB, make 2 big polls a year, with a sample of about 9 000.

If telemarketers can use automated machines and call thousands of people at a clip then what would stop a rich person from funding such a poll?

It's not worth it. The increase in accuracy becomes less and less cost-effective, simply. So you'd pay a lot of money and get very little for it. You would always have the MoEs anyway.  

[Gustaf]

John Engle has a prediction entry

Yeah...WA is lean Bush, but GA is strong Kerry...

[Gustaf]

The 2nd one. That someone would get 80% in WY but below 40% in ID makes very little sense...

[Gustaf]

Also, we have to remember that even though a lot of Perot's voters were conservatives they supported him over Bush BECAUSE they were unhappy with Bush and his broken tax pledge. They might well have refused to vote for him anyway, even if Perot wouldn't have been around.

[Gustaf]

Also, we have to remember that even though a lot of Perot's voters were conservatives they supported him over Bush BECAUSE they were unhappy with Bush and his broken tax pledge. They might well have refused to vote for him anyway, even if Perot wouldn't have been around.

I think Dazzleman said it best:  The very presence of a Perot-type candidate illustrated Bush's weaknesses.

Yep, that's it.

[Gustaf]

40%-20%=20%

20%*19%=4%

So, Clinton still wins, though by a much smaller margin.

[Gustaf]

Shouldn't Clinton, if he campaigns a lot, be able to win Arkansas for Kerry?

In 1992 Clinton did abot 10% better in Arkansas than he did nationally. In 1996 the difference was about 4.5% (Arkansas stayed the same, whereas Clinton's national vote went up from 43% to 49%).

Now, I know it isn't the same when he runs as compared to when he endorses, etc. But still, he's a fairly recent ex-president and favourite son. If he could knock up Arkansas by a few % that could well be enough.