Brad Wardell's views about technology, politics, religion, world affairs, and all sorts of politically incorrect topics.
Statistical analysis
Published on October 15, 2004 By Draginol In Politics

Polls are interesting things but statistical analysis is my thing.  And based on our models in The Political Machine, Kerry is going to squeak out a victory in November barring a sudden change in events.

The Florida of 2004 may be where you least expect it - Iowa. Ohio and Iowa are going to be the key states this time. But it is Iowa that people should be keeping an eye on even more so than Ohio or Florida. It has a relatively low population which creates all kinds of opportunity for messing with the final results.

Moreover, Bush can potentially win the election and still lose Ohio as long as he gets Iowa.  That's because Wisconsin is likely to go Bush this time.  So you could have a bizarre scenario where Bush wins the election despite losing Ohio.  In that event, it would be Bush 271 to Kerry's 267.

Obviously when it's as close as it is, it's really hard to make a definitive prediction. Our statistical model is not poll driven but is instead issue driven. The war in Iraq is no longer as strongly favored by independents and the jobs issue in Ohio has increased.

So our prediction right now is that Kerry has an advantage even though national polls currently favor Bush. And that the two states to watch out for are Ohio and Iowa. The computer controlled versions of Bush and Kerry in our simulations now focus almost exclusively on those two states.  New Mexico could also go either way but it isn't statistically relevant.

This article does not indicate a preference by Stardock or Ubi Soft. It is just a report from our statistical model and from waching the computer controlled versions of Bush and Kerry.  If you're an American, regardless of where you live, please make sure you are registered to vote and vote this November.


Comments (Page 2)
3 Pages1 2 3 
on Oct 15, 2004

The title is reassuring, but who knows what will happen. 

Anyone who protested about Gore's non-election on the basis of winning the popular vote is wrong.  The relevent reason for discord would be the fuzzy results in Florida.  I think that would be worthy of protest.

If irregularities in any state lead to Bush's re-election or Kerry's election this time out, it would make sense for the "losing" side to protest.

on Oct 15, 2004
Gore would have won in 2000 if Palm Beach voters could figure out how to use a slightly more complicated ballot.

Kind of gives lie to the myth that Democrats are smarter than Republicans. The only reason Gore isn't President right now is because Democrats in Palm Beech were dumb.

Keep a close eye on the candidates. I think you'll see them in Iowa and Ohio a lot in the coming weeks.
on Oct 15, 2004
Well - there are all kinds of irregularities in a couple of states with the 2000 election. Focusing on FL and palm beach in particular ignores other things.
on Oct 15, 2004
MUHAHAHAHA, look at all the red land mass we conservatives control, victory is sure to be our's!!!!!!!! sigh........stinking electoral college........if only size truly mattered...........in politics.
on Oct 15, 2004
We'd own almost all the natural resources, too and have all their bagel-eatin behinds bottled up in cities, ripe for siege. Where's my damn catapult...
on Oct 15, 2004
Thank you so much for doing the statistical analysis. I was very curious about who would win be it Kerry or Bush if it was ran in the Political Machine and I can see how the country shifted towards Kerry. It's really interesting to see how the process folds out as the swing states are in full swing. This is going to be an even tighter election than 2000.
on Oct 16, 2004
Draginol:

Not that I don't appreciate the model but I would be remiss if I didn't say that the election within 30 electoral votes or 4% (the usual margin of error for election polls) is "too close to call." I think we are still too far away to even make a choice of candidates. Even this morning, new information on Bush's Guard Service came out.

I would caution people not to rush to either candidate until we get to the last weekend. I know that many feel strongly about "their candidate" but this election is far from over.
on Oct 16, 2004

CrispE: I know, it's too close to call. That's the point of making a preidction because if it wasn't so close, any prediction would be "Well DUH."

Iowa and Ohio, if Bush wins either one, he wins. But I think Kerry is likely to win both.

on Oct 16, 2004
You're a smart guy, Drag. Just hope you're smart and wrong.

Cheers,
Daiwa
on Oct 16, 2004
Can't forget Colorado's voting initiative #36. This would split up Colorado's electoral votes according to popular vote. If it passes, Kerry will pick up four (or if he wins Colorado five) electoral votes. Final score: Kerry 270, Bush 268
on Oct 16, 2004
I'm calling Ohio for Kerry. I feel MO is gone to Bush as almost all the paid staff in the Kerry camp have gone to Iowa to slug things out there. Penn. will go to Kerry, but just barely...(got some friends up there...)
on Oct 16, 2004

Reply #26 By: Deference - 10/16/2004 3:53:56 PM
I'm calling Ohio for Kerry. I feel MO is gone to Bush as almost all the paid staff in the Kerry camp have gone to Iowa to slug things out there. Penn. will go to Kerry, but just barely...(got some friends up there...)


I also live in PA. And the state polls show it barely at 46 Kerry 45 Bush. Don't be to sure about it going to Kerry. A 1 point difference is way to close to call.
on Oct 16, 2004

Reply #25 By: Politically Active - 10/16/2004 2:06:59 PM
Can't forget Colorado's voting initiative #36. This would split up Colorado's electoral votes according to popular vote. If it passes, Kerry will pick up four (or if he wins Colorado five) electoral votes. Final score: Kerry 270, Bush 268


The current state polls don't back your play. Currently it's Bush @ 257 Kerry @ 243. Even giving Kerry the 4 points your talking about Bush is still up. If you would like to keep track go here:
Link

on Oct 16, 2004
well, the game is unproven so who can say. I hope it is wrong, wrong, wrong.
on Oct 16, 2004
well, the game is unproven so who can say. I hope it is wrong, wrong, wrong.
3 Pages1 2 3