Electoral Vote Simulation Models Point to John Kerry Win Tuesday

[Nicholas_J]

Washington, D.C. -- The Washington Redskins poll may be the most accurate electoral vote simulation model with every incumbent political party loosing the White House if the Redskins loose their home game before the Tuesday election.

The points have been counted: Packers 28, Redskins 14. Touchdown Kerry!?

Under the Redskins rule, that means John F. Kerry will win the Tuesday national election, with George W. Bush becoming a one-term president like his father.

Professional pollsters may scoff, but the rule has been correct in every election since Franklin Roosevelt won in 1932. The probability of 18 elections in a row being decided this way is estimated to be one in 260 million.

Professional pollsters are NOT scoffing, however, at the most recent electoral college simulation models. Various academic models are now predicting a narrow win of both the popular vote and the electoral college vote by Kerry over Bush.

http://www.mywisecounty.com/news/110104-2.htm

[kranich]

1. They are trying to

say that since the Redskins no longer play in DC, but in Maryland, all bets are off.

[Goldmund]

3. One in 260 million?

Hardly. The probability of it happening each time is 1 in 2.

So, 18 times in a row, it's 1 in 2^18, which is 1 in 262 thousand.

OnEdit: oops, meant to reply to the original post.

[salib]

4. Funny that...

The odds that some sports statistic would be correlated with the election in the last 18 cycles are about 1. I.e., it is virtually guaranteed that some team would accomplish some feat related to the election consistently. Sorry to be such a skeptic.

[Goldmund]

5. Exactly

I totally agree with you. I'm sure you can find a town in the US in which weather on the day before the election has been an accurate predictor of the election result since 1864. If it rains, the incumbent wins... otherwise, the challenger.

And today, it's sunny!

[BootinUp]

6. Wouldn't it be higher odds

that it correlated to incumbent party wins/losses?

[Goldmund]

7. No

each time it's in in 2 -- there's 1 incumbent party and one challenger party.

[BootinUp]

8. I guess I was thinking

there are always 4 possible outcomes. But its not important enough to waste any more time on. :)

[mconvente]

15. actually...

thats kind of faulty logic - that assumes 50/50 probability for each iteration. But Im sure there have been instances where Washington had been a great time that year and then crappy the other years (like this year) - plus it depends on the teams they play. If the Skins were playing AZ they would be favored but since they played GB they weren't. So it's sorta accurate but be weary on that probability.

[jackson]

9. And they're saying Green Bay won on a technicality.

So the election might come down to a technicality. If Washington had won, repugs would be using this trend to forecast a win for their candidate.

[pelagius]

10. Didn't see the game, but...

...up by two touchdowns seems like a little more than a win on a "technicality". Of course, in Repug land, the "technicality" was probably something like "crossing the goal line with the ball."

[MaineYooper]

11. GB vs Washington

the "technicality" was a penalty that nullified a Redskin touchdown when it was still 20-14 GB.

Unfortunately for * (and the Redskins), all that counts in football is final score, not the manner of reaching it.

[newyawker99]

14. Hi MaineYooper!!

Welcome to DU!! :toast:

[pelagius]

12. So if the "technicality" had gone the other way...

...the 'Skins would have still lost? Sheeesh...

Typical Republican bullshit. They can't just laugh off the whole thing as silly. No, they have to be weasals on everything.

You know what? -- I do hope the door bangs these guys hard on the way out.

[Psst_Im_Not_Here]

2. Booya!

Good thing I wore that Packers green nail polish on Sunday!

[NewYorkerfromMass]

13. Kerry will get at least 275 votes

depending on Florida and the west....

[TruthIsAll]

16. He quotes numbers from the Election Model in the article

http://www.geocities.com/electionmodel/