Welcome to DU! The truly grassroots left-of-center political community where regular people, not algorithms, drive the discussions and set the standards. Join the community: Create a free account Support DU (and get rid of ads!): Become a Star Member Latest Breaking News General Discussion The DU Lounge All Forums Issue Forums Culture Forums Alliance Forums Region Forums Support Forums Help & Search
11 replies = new reply since forum marked as read
Highlight: NoneDon't highlight anything 5 newestHighlight 5 most recent replies
 

MrSlayer

(22,143 posts)
3. Most people simply do not pay attention.
Wed Aug 22, 2012, 09:29 PM
Aug 2012

If everyone objectively looked at what has been going in in Washington the past four years there is no way this would even be close. There is no way we would have gotten trounced in 2010.

But they don't and we're in serious trouble here. Serious trouble indeed.

 

Dkc05

(375 posts)
6. I think the poll is a fraud
Wed Aug 22, 2012, 09:32 PM
Aug 2012

Don't panic let's just work hard to get people out to vote on election dayI I think the poll is a fraud

samsingh

(17,598 posts)
7. if it predicted a bush win in 2000 and 2004, then it is clearly wrong
Wed Aug 22, 2012, 09:34 PM
Aug 2012

2000 was definitely a Gore win. 2004 was likely stolen in Ohio

CJCRANE

(18,184 posts)
8. It's a "postdiction" model, as mentioned in the other threads.
Wed Aug 22, 2012, 09:39 PM
Aug 2012

They looked back and analysed the data to find the commonalities. They didn't predict anything.

You could use different data points to get a different result.

longship

(40,416 posts)
9. Rubbish. Data mining and post hoc ergo propter hoc.
Wed Aug 22, 2012, 10:17 PM
Aug 2012

This is like taking stock market data, looking for correlations and then announcing that you can predict the future stock market by the correlates.

This is a monumental statistical fallacy.

Correlation does not imply causation.

Enough with these bullshit threads.

You might as well correlate the election to the zodiac signs. I am sure their are correlations with the numerological elements of the year number.

Latest Discussions»General Discussion»Poll predicted last elect...