Flipping the vote against Ron Paul in South Carolina?

[The Man]

Hey Liberty 1789, this exactly supports my observations. I would like to get your comment on this: Using the method of total votes vs. votes each, the precincts can be arranged in any random order and it will produce a "noisy", "jaggedy", "erratic", etc. looking line for Romney. But when the precincts are arranged in order from lowest vote total to highest, the curve is inexplicably "smooth." I concluded that this could not be natural, but rather had to be created by a computer algorithm. There's simply no reasonable alternative explanation.

This is critical and your intuiton is on the right track. I'll put it as simply as I can.

A visually straight line on the charts I publish is rarely perfectly straight. It joins points which are not perfectly aligned. The differences between the real points and the straight line that best fits them can absolutely be described a white noise. t-stat measures directly how stable is the slope of the line joining the points, how much if it "vibrates" along its general direction. F factor is even smarter in some ways: it measures if all those points which are not perfectly on the line are normally distributed away from it: more points closer than far way from the line, if you will.

On segment of lines where the vote flipper I suspect is active, something extraordinary happens: the nature of the white noise is totally transformed! T-stat is up 5/10 times, F factor sometimes 20-30 times. This all indicates that the slope is not natural. it has been "smoothed" by the vote flipping. It vibrates much less and points next to the line are much closer than before, with fewer outliers.

What happened is this: you're mixing the original, "natural" line with normal white noise to a perfect straight line with no noise. The combination has therefore massively less noise than the original. t-stats and F factors shoot up : the straighter the line, the higher they go. And they shoot up massively. The radical change in the noise is a proof of tampering! What one calls a mathematical fingerprint!

Another point: I have not seen yet a big F factor (>500) on its own: if the white noise is gone from 1 line, there is always another one where the noise is gone as well. Algorithmic vote flipping would do precisely that.

 

[kathy88]

I have no idea what you guys are talking about but I can't wait for it to come out I've said it before I'll say it again THANK YOU SO MUCH. I know you guys have been busting your asses on this for a while now.

 

[kathy88]

HAHAHAHA. If you hover over the paperclip next to the stars on the front of this thread I just noticed the file is named WTF. HAHAHAHAHA.

 

[affa]

bb/mark -- deep breath, guys.

liberty - once again, amazing work.

 

[cjm]

This is critical and your intuiton is on the right track. I'll put it as simply as I can....

Thanks for the explanation. My previous issues/challenges with regard to sampling and projection methodology are obviously not applicable here.

 

[affa]

ok, so i'm no graphic designer.

i took liberty's posts (starting around 737) in which he broke down SC by flipper on/off. Somewhere around there, he also pointed out that the flipper is not based upon total votes per county, but rather, total votes as percentage of... something. can't remember. just tried to find it and can't. liberty? anyway, knowing very little about SC geography, i wanted something a bit more visual.

gold star = Liberty detected NO vote flipping
BLUE star = Liberty detected theft from Gingrich to Romney
RED star = Liberty detected theft from Paul to Romney
NO star = Liberty didn't report on that county.

i put in approximate vote count for some of them - mostly where i found it interesting -- for example, the cluster of 8ks near each other that show vastly different results.

IMPORTANT: note that there are a couple 1:1 flips for Gingrich to Romney touching 1:1 flips for Paul to Romney. That's damning in my opinion. The string of 8k (Gingrich to no-flip to Paul) is also interesting.


image was too large to embed; resized:

this embed only seems to work sometimes. link is here:
http://image.bayimg.com/damcpaado.jpg

 

[Slutter McGee]

Seriously, I have no problem putting in the effort of reading an entire thread. But 85 pages is pushing it. Can anyone give me a bit of a summary?

Sincerely,

Slutter McGee

 

[affa]

Seriously, I have no problem putting in the effort of reading an entire thread. But 85 pages is pushing it. Can anyone give me a bit of a summary?

Sincerely,

Slutter McGee

it's worth reading. also, go to your settings and change number of posts per page. i only have 22 pages.

summary: it looks like we have established voter fraud in SC, and have evidence of it elsewhere.

important bits:

1) Romney always the benefactor
2) Paul, Gingrich, Huckabee (in 2008) have all been hit.
3) In most cases, it's 1:1 benefactor/victim, but in some cases Romney pulls from multiple others.
4) No one has offered a debunk that explains how this could only ever affect Romney the way it does (specifically, direct pull from one other candidate with no other candidates affected)
5) Statistical evidence that the slopes are algorithmic (smoothed) further indicating fraud (compared to fraudless areas where there is a lot of natural noise, which is how it should be)
6) it seems they only alter votes in larger precincts meeting certain other criteria. low vote areas are left alone.

our resident stat experts believe we have massive smoking guns.

 

[Slutter McGee]

Thank you so much. As much as it hurts me to say it. At least the voter fraud is benefiting Romney and not Santorum. I will try to read more.

Sincerely,

Slutter McGee

 

[Anna Karenina]

Just curious why they didn't use this vote flipping technique on Rand? How and why on earth did they let him win?

 

[Paul Fan]

Just curious why they didn't use this vote flipping technique on Rand? How and why on earth did they let him win?

Rand was too far ahead. It would have been too noticeable.

 

[Liberty1789]

A decluttered blog re-taking only the substantive analytical posts would be very useful actually, affa.

 

[S.Shorland]

This is the internet.Anybody could be anybody.I'd really like this to be in the hands of an indentifiable person with credentials and acknowledged to be so by a recognisable figure associated with Paul.That's my real beef.

 

[cjm]

4) No one has offered a debunk that explains how this could only ever affect Romney the way it does (specifically, direct pull from one other candidate with no other candidates affected)

We have to be careful with this one. The "why Romney?" question is murky. The answer to the "Why?" question could vary from county to county and election year to election year based on demographics. Debunks have been offered that have not been statistically dis-proven. The relative Republican-ness can explain shifts between Paul and Romney in some counties. This doesn't explain Romney-Huckabee, but it doesn't need to. Different years, different demographics, etc. The burden is on the statistician to prove there's something there, the burden is not on the denier to prove there is nothing there.

The statistics focused on in this thread (algorithmic smoothing) cannot explain why, they can only demonstrate that it is Romney. I would just leave it at that.

 

[Danan]

I saw you guys speculating that Ron Paul could be better in heavily democratic areas and worse in republican as a possible explanation/debunk. Do you have any prove for this assumption? Or did I misunderstand you?

In the latest polls I saw Mitt Romney was ahead of Ron Paul with democrats (I even think by a margin that would extend Mitts overall lead). Ron only owns the voter segment of independents. This is not even astonishing if you look at their policies. So if anything Mitt should even increase his numbers in democratic areas. And even if Ron would do great with dems, there is no reason why guys like Gingrich and Santorum should stay equal and only Romney loses votes.

Also, more republican areas should - if anything - increase the numbers of Gingrich and Santorum against Ron Paul and maybe even against Mitt Romney.

 

[bbwarfield]

I saw you guys speculating that Ron Paul could be better in heavily democratic areas and worse in republican as a possible explanation/debunk. Do you have any prove for this assumption? Or did I misunderstand you?

It's a theory that was kicked around, but the original way of deciding "Republicanism" of a precinct never took into account the minority vote demographic. (Used McCain vs. Obama as a standard) So while that theory hasn't been dis proven, i don't see it as a likely explanation but there still crunching numbers on these types of things.

 

[affa]

We have to be careful with this one. The "why Romney?" question is murky. The answer to the "Why?" question could vary from county to county and election year to election year based on demographics. Debunks have been offered that have not been statistically dis-proven. The relative Republican-ness can explain shifts between Paul and Romney in some counties. This doesn't explain Romney-Huckabee, but it doesn't need to. Different years, different demographics, etc. The burden is on the statistician to prove there's something there, the burden is not on the denier to prove there is nothing there.

The statistics focused on in this thread (algorithmic smoothing) cannot explain why, they can only demonstrate that it is Romney. I would just leave it at that.

i disagree. every 20 or 30 posts or so we get someone who hasn't read the thread suggesting Romney might just be more popular in bigger threads because of demographics.

But this can NOT be explained by demographics or "relative Republican-ness" and there have been many posts on why. It could explain Romney winning, sure. It could explain a spike or a dip in a candidate, sure. But it can not and does not explain 1:1 flipping in a race like we're seeing.

I discussed this more here: http://www.ronpaulforums.com/showth...-Paul-in-South-Carolina&p=4212485#post4212485

To quote from that post: any precinct that is so incredibly Romney-centric that every last vote pulled from Ron Paul would unerringly go to Romney is a precinct in which Romney is so popular that he'd be pulling votes from all other candidates across the board, not just Ron Paul.

Just because 'demographics' is the most common 'debunk' attempt doesn't make it valid. Can it be disproven? Well... that's what liberty has basically done by showing that the odds of what we're seeing are staggering within a single county, let alone repeatedly. And to add to that, it's even sillier to think that the county next door, of a similar size, would show a similar 1:1 with Gingrich instead of Paul!

The assumptions in the 'debunk' - that Republicans reject the Constitutionalist Paul, and prefer the flipflopper Romney - are quite likely nothing more than media fabrications, especially based on what we're seeing in this thread.

 

[affa]

is my jpg in post 846 showing for anyone? it was a nightmare getting that to embed, and i'm still not sure if it's working correctly.

 

[Danan]

It's a theory that was kicked around, but the original way of deciding "Republicanism" of a precinct never took into account the minority vote demographic. (Used McCain vs. Obama as a standard) So while that theory hasn't been dis proven, i don't see it as a likely explanation but there still crunching numbers on these types of things.

Do these demographic explanations even matter at all if we can prove a statistically very unlikely (or impossible) voter shift kicking in and increasing at various precinct sizes?

In order to matter the demographic would have to change at the exact same rate as the voting results shift. And that all over SC (and probably even more states). That doesn't sound very reasonable. Also, I don't really think it's possible for anyone in the grassroots to check every possible demographical explanation - that sounds like a hell of work.


Do we see the same pattern of voter shift in various counties in every state or do different strange voting behaviors occur depending on the state? If the latter is the case, do these patterns correlate with the way the ballots were counted (do hand-counted ballots show a different pattern then machine-counted)?

If the same pattern emerged in Iowa as in counties that used electronic voting (even if it just happend in very few counties in Iowa) we do have a little problem. Either our Paul-campaign vote counters in Iowa didn't do a good job and were wrong (or fooled) or there is something very special in Mitt's personality that causes him to gain votes by precinct size with an astronomical precision that would force every statistician into insanity and this project is pointless. Because even if we could come up with a good story about voter fraud in SC, they could just debunk it and say that the same pattern exists in other precincts without electronic voting machines.

Again: Did the very same pattern that could have been caused by voting machines in SC occur at least once in another county where the ballots were hand-counted in public?

Sorry if these all concerns have been answered already. You can just ignore this post then and continue your work. =D

 

[Danan]

is my jpg in post 846 showing for anyone? it was a nightmare getting that to embed, and i'm still not sure if it's working correctly.

Well, there was a link that worked earlier this day. Now I don't see neither a embeded picture, nor a link.