Flipping the vote against Ron Paul in South Carolina?

[RPit]

I'm no statistical genius, but honestly I find this very compelling that at least 'something' is wrong. Whoever put this together deserves applause.

 

[affa]

Interesting speculation alone isn't worth broadcasting. If we had hard evidence here of a flipped vote with testimony to back it up, that would be worth bugging anchors and drudge about...

This is statistically damning. It needs to be peer reviewed, and number checked, but it seems accurate. Clear case of vote flipping. The statistical model is able to predict Santorum and Gingrich with, as the author notes, "deadly precision". This sort of precision in statistics is why they can often call elections early, or base polls on only a couple hundred respondents.

This is, assuming the numbers are correct, which they do appear to be... well, damning. Absolutely damning. There is no conceivable reason why in any county over a certain population every predicted Ron Paul vote, based on such a large sampling, would be perfectly flipped over to Romney.

This needs to get out. Again, it needs peer review by statisticians... but this is a smoking gun.

We need people, preferably trusted professionals, looking at this.

 

[goldpants]

I do have one question though. If we really did turn out the vote, were our votes spread out across age groups we normally fail to do well with or were the young voters out in multiple levels to the ones reported?

 

[BUSHLIED]

Nothing surprises me, I am convinvced there is fraud taking place. Just look at how the media treated Ron since '07....if they can get away with that out in the open then switching votes is so much easier...the media and the establishment don't care about Paul...

 

[RPit]

Plausible.

I want to see the same analysis run for different counties.
Then for different states.

I'd also like to see the number for Rick Perry included in there. I know he got some votes. Maybe not enough.

I'd also like to know more about that particular county and the precincts.

This could always be just corruption in one county or just one statistical fluke in one county.

Why would Ron Paul and Romney "flip" in voters minds? If you say Paul is strong in the small precincts you must say that Romney is strong in the big precincts. And Santorum and Gingrich have the same appeal to small and large.

But we don't know WHY some precincts are small and some large. Was it turnout? Was it rural/urban? Democrat/Republican? Black/White? Regardless it has to be something the uniquely impacts only Paul and Romney.

It would be nice to see this analysis for Iowa and New Hampshire where there were more candidates.

Rigging can be customized, so to assume the same analysis would 'transfer' isn't fair. Depends on the riggers.

 

[affa]

Can anyone dig up exit poll information pronto. Like, immediately, while it's still available. It would be interesting to know if exit poll results were in line with these predictions.

I'm off to go see Ron Paul in Vancouver or I'd do it myself.

 

[Mark37snj]

Does anybody know where to get all of these numbers by county, or possibly even already in excel format? If we could get that information on here, maybe somebody with the know-how would take that up.

*NOTE* I don't think that the statistical calculations used in the document for SC will be accurate for caucuses, only primaries. A small vocal minority can skew the numbers in a caucus much more.

I suspect the campaign is gona take a good look at this and probably launch an investigation like the one with the Huntsman video.

 

[FrancisMarion]

Did you do this for any of the other counties? Its well done, but in order to convince you have to include other data sets as a measure of control.

1. Do the same 20% as the basis for your straight line in other counties please. Charleston would be nice, that's mine.
2. For arguments sake, what are the results if you went back and used 10% for the straight line estimation? Is there a large deviation? Shouldn't be to much work if you are using excel....
3. Again, its eye-opening but it has to be compared to prove its not an anomaly.

Read the whole thing, its well presented.

 

[S.Shorland]

In the Maine Caucus it was 'don't count the votes aloud'.Evidence from Belfast of shenanigans due to one thoughtful chairman going against the state gop edict.Electronic and procedural rigging to solve the caucus/primary difference.

What a blessing that Maine is so small,actually.It forced them to deal with small precincts.

 

[low preference guy]

In the Maine Caucus it was 'don't count the votes aloud'.

I feel like there should be a national campaign FOR counting the vote aloud after this. There is video to make a cool ad.

Rev9

 

[flsurfer]

I agree, this needs to be done on more counties. If the graphs match up in a similar manner it should be a pretty solid evidence that is was rigged.

Those of you who haven't already looked at these graphs all the way please do: https://docs.google.com/viewer?a=v&pid=explorer&chrome=true&srcid=0B_wWkfsJPShUMWQxMTc2NzgtM2MzYy00ZGJhLWI1MmYtMWU2ZGU1OWZkZjhk

 

[bcreps85]

I suspect the campaign is gona take a good look at this and probably launch an investigation like the one with the Huntsman video.

It would be good, but we can't rely on that. Remember...they might not even know some of what is going on...they are running a campaign and likely not scouring the forums. If I could get the numbers, I'd try to work with this some and post what I come up with. My last statistics course was a while ago, but I could probably try to use excel to duplicate his methodology.

 

[roversaurus]

Along with doing the same thing for other counties and other states. We should look at exit polling for this specific county AND we should go back and look at 2008 voting results for this same county.

Perhaps we have discovered a "real" phenomena that people who vote in small vs large precincts have a Paul/Romney preference difference. We should see this in other locations then. We should probably see it in 2008.

I guess it's possible that Romney sent out campaign workers to large precincts in that county. But at the moment it seems best explained by statistical fluke or fraud.

 

[Mark37snj]

It would be good, but we can't rely on that. Remember...they might not even know some of what is going on...they are running a campaign and likely not scouring the forums. If I could get the numbers, I'd try to work with this some and post what I come up with. My last statistics course was a while ago, but I could probably try to use excel to duplicate his methodology.

I think affa is on the case.

Can anyone dig up exit poll information pronto. Like, immediately, while it's still available. It would be interesting to know if exit poll results were in line with these predictions.

I'm off to go see Ron Paul in Vancouver or I'd do it myself.

It's on Daily Paul, no way they are gona miss this. But I'm all for kicking around the numbers some more.

 

[hillertexas]

 

[EaSy]

I do have one question though. If we really did turn out the vote, were our votes spread out across age groups we normally fail to do well with or were the young voters out in multiple levels to the ones reported?

I think the voter info is comming from exit polls, not from actual voting.

 

[Mark37snj]

I think the voter info is comming from exit polls, not from actual voting.

Just above Figure 1 it says at the top of the paragraph that "A spreadsheet was created with vote data downloaded from the SC Election Commision's website for Anderson County. Not sure if thats what you are looking for.

The data for this analysis was actual votes.

 

[EaSy]

I wasn't talking about spreadsheet but about where more info about actual voters is comming from. Official results from SC are just precicts and numbers of votes.

 

[Mark37snj]

I wasn't talking about spreadsheet but about where more info about actual voters is comming from. Official results from SC are just precicts and numbers of votes.

Ah, sorry, CNN should be a good spot to look. They ran all kinds of polls on variables like that.

 

[S.Shorland]

Read point 8.Anderson apparently isn't a county that shows wide discrepancy between rural and urban.He used the votes actually cast and then extrapolated.My question was just about opinion polls on realclearpolitics running up to election day showing Romney well ahead but if his figures are correct,I can't see anything wrong with his logic.