Electoral College Meta-Analysis (election.princeton.edu)

[ParanoidPat]

http://election.princeton.edu/#final2004

From Prof. Sam Wang of Princeton University.

Interesting reading. :evilgrin:

Has anyone else looked at this analysis? :shrug:

Check out the charts at this link.

http://synapse.princeton.edu/~sam/pollcalc_validation.html

[dbonds]

1. Interesting

His comment: Wednesday, November 10, 8:00AM: Regarding the ongoing Florida voting fraud controversy, S. Doershuk, who has expertise in demography, makes a constructive suggestion: "It might be instructive to examine some similar counties in Georgia and Alabama, particularly those which border on north FL, to see if the same pattern can be found. Given the 'bright red' nature of both border states, I wouldn't expect anyone to have bothered to tamper with vote totals there, so a comparison could be instructive." This is excellent. If any of you has this information I would be very interested.

I think this is an excellent idea.

His graphs do show something fishy is up. Just another way to smell the fish.

[savetheuniverse]

2. good to see we've got some princeton conspiracy theorists;)

well for those of us non-scientific einstein fans, I thought this was interesting (albeit disastrously depressing) with re "public consicousness"

the site meter is devastating: peaks out on election day goes down to next to nothin' today. the opposite would have been more encouraging--an index that people are looking, but they aren't.

it's so maddening. we have the worst brains in the country in the white house and the best brains in the country trying to figure out how the hell they got there.

Guess that's the problem with trying to predict the past.

[noserotonin]

6. I don't think Prof. Wang thinks there is a conspiracy...

I used to go to Sam Wang's page quite frequently before the election since he provided an interesting, statistically-rigorous analysis of the pre-election polls. (His prediction of a Kerry win was a bonus!)

After the election, I expected him to analyze the exit poll results and say "what a minute something is really wrong", but he never did. I don't think he knows about CNN later changing exit poll results.

In fact, he attributed some of the discrepancy to the youth vote not materializing. (Okay, perhaps this was plausible explanation immediately after the election when all the pundits were busy blaming the youth vote. But later news stories have indicated the youth really did come out and vote in numbers not seen since 1992.)

I certainly appreciate the work Professor Wang did prior to the election, but on the whole he seems to have just "rolled over" and accepted the election results, e-voting glitches, provisional ballots, and other irregularities not withstanding.

I think he's lost interest in the whole project now. :(

[bleever]

3. The second table here caught my attention; it seems to show

pretty clearly that the greatest deviations were in the key states of Florida and Ohio (as well as New Hampshire, Pennsylvania, Missouri and Colorado). These deviations were more than sufficient to tip the state. And while six states showed deviations that favored B* by +1 or more, only one state deviated for Kerry by that magnitude.

To me this is graphical illustration of TruthIsAll's point in his 187 million to one analysis, and the 250 million to 1 analysis from the stats prof. :thumbsup:

[ParanoidPat]

4. I wonder if TIA has checked this site out?

:shrug: :kick:

[bleever]

5. I PMed him to draw it to his attention.