The case for the occurence of algorithmic vote flipping

[kathy88]

Not realistic to get that amount of money....on the other hand you guys/girls/people (with all due respect) dont seem to move from the spot (are not going to deliver killing blow)....Maybe try to find newspaper to finance it instead with promise to get exclusive or something like that...

That's a great idea, Barrex. I would be willing to chip in on this, but corporate cash may be the way to go.... Brainstorm on possibilities?

 

[chri5opher]

//

 

[Bohner]

We need to contact more members of academia. These people are busy and it will take a while before we find someone that's willing to drop their current research and tackle this rather large problem. Most likely they will need funding.

The UCR Statistical Consulting Collaboratory (collaboratory.ucr.edu, see video on main page) would love to help but they also love to get paid. They proposed to do a large multi-variate study that will eclipse any doubt opponents would claim. If you have money and want to see a formal scientific paper written, please call their office at 951-827-7939. Their On-Campus Rate is $62/hour.

How Can You Help:
http://collaboratory.ucr.edu/gift.html

Gift Student stipend:
$30: 1-day of work in the Collaboratory
$150: 1-week of work in the Collaboratory
$1500: 1 full quarter of work in the Collaboratory
$4500: A full year of work in the Collaboratory
All gifts are greatly appreciated!

So as you can see, we are not just "talking" about it.

Hey RR...

Have you considered proposing this to one of Paul's SuperPACs?

I personally don't see the Paul campaign putting money into this as they need to use their funds wisely, and investing in this would sort of reinforce the "there goes crazy uncle Ron babbling about another conspiracy" stigma.

The SuperPACs on the other hand have cash on hand, and would allow Paul to personally distance himself from the whole flippergate scandal if they were the ones funding it.

Just a thought.

 

[drummergirl]

Hey RR...

Have you considered proposing this to one of Paul's SuperPACs?

I personally don't see the Paul campaign putting money into this as they need to use their funds wisely, and investing in this would sort of reinforce the "there goes crazy uncle Ron babbling about another conspiracy" stigma.

The SuperPACs on the other hand have cash on hand, and would allow Paul to personally distance himself from the whole flippergate scandal if they were the ones funding it.

Just a thought.

I suggest we try the Colbert Super PAC; they seem to have plenty of cash to throw around.

 

[Steve-in-NY]

Yknow, I stop in this thread pretty often, I find it very interesting. Im honestly not sure if I believe whats happening is what is being claimed, but I am also keeping an open mind. I think that just maybe you may be onto something... just not sure what. Yet.

That being said - I'll reach right to the top and say this:
If you guys are correct, and there is voter fraud going on, it seems to me that even $100,000 is nothing if it can definitively prove fraud taking place, considering the ramifications of such a scenario on the country.
If you guys are wrong and there's some previously overlooked data set that causes this phenomenon it would seem that it would be caught rather quickly, therefore ultimately costing perhaps a few thousand and put it to bed.

Therefore, I would start a chip-in to have an experienced, accredited third party have a look and if they smell a rat so to speak, THEN would be the time to push to drive the study forward.

</IMHO>

 

[drummergirl]

If you guys are correct, and there is voter fraud going on, it seems to me that even $100,000 is nothing if it can definitively prove fraud taking place, considering the ramifications of such a scenario on the country.

</IMHO>

I keep having nightmares about what happens the first week of November when massive voter fraud is uncovered the day after the election by the Democratic party in several key states. It would make Florida 2000 look like a kindergarden playground fight. So, compared to that scenario $100,000 is an incredible bargain.

 

[RonRules]

A few posts ago, I think I was doin't wrong.

Because there were so many delegates in RI I grouped them with one chart per candidate. Of course, there is no competition between delegates for the same candidate! That was the wrong way to chart delegates.

Instead you have to pit each candidate delegates against one-another, for each corresponding delegate slot.

Specifically: Romney's Delegate #1 vs. Paul's Delegate #1 vs Gingrich Delegate #1 vs. Santorum Delegate#1.

Here's the four RI delegates in position #1: Note how the evil is unveiled?!!

As it looks in RI, they were flipping the delegates! Good job Romney!

May I kindly ask if the volunteer analysts of the Alabama thread could chart each delegate position for each candidate? I just want to see if the same things happens in AL.

Maybe they were careless in AL and rushed to fix it in RI. Inquiring minds want to know!

 

[RonRules]

For your enjoyment:

The Rhode Island District 1 Delegate Battle Royale!

 

[The Man]

Hey RonRules- good work. In Bama Romney's delegate positions show a rise in ascending precinct total order, but remember that we found a hinge point where Romney's "votes minus delegates" takes off. This is probably as strong evidence as there is to date independent of simple comparison of low and high vote total precincts. It would be interesting to create the same plot "votes minus delegates" for the RI Primary.

 

[The Man]

The UCR Statistical Consulting Collaboratory (collaboratory.ucr.edu, see video on main page) would love to help but they also love to get paid. They proposed to do a large multi-variate study that will eclipse any doubt opponents would claim. If you have money and want to see a formal scientific paper written, please call their office at 951-827-7939. Their On-Campus Rate is $62/hour.
How Can You Help:
http://collaboratory.ucr.edu/gift.html
Gift Student stipend:
$30: 1-day of work in the Collaboratory
$150: 1-week of work in the Collaboratory
$1500: 1 full quarter of work in the Collaboratory
$4500: A full year of work in the Collaboratory
All gifts are greatly appreciated!
So as you can see, we are not just "talking" about it.

Make sure Dr. K is aware if this opportunity. He may have the financial means to get started.

 

[RonRules]

It would be interesting to create the same plot "votes minus delegates" for the RI Primary.

Here it is:

It was worth the effort. Both Romney's line and Ron Paul should be completely flat, but as always, Romney's goes up, this time completely independent of demographics effects, just like in Alabaman. The slope is not as steep, but the effect is confirmed.

Grinch & Santo look like their voters were perfect! The candidate votes almost match the delegate votes perfectly. It's much more likely instead that because they were not subjected to vote flipping (see straight lines on candidate chart, post #713) the subtraction yields near zero.

 

[The Man]

Here it is:
It was worth the effort. Both Romney's line and Ron Paul should be completely flat, but as always, Romney's goes up, this time completely independent of demographics effects, just like in Alabaman. The slope is not as steep, but the effect is confirmed. Grinch & Santo look like their voters were perfect! The candidate votes almost match the delegate votes perfectly. It's much more likely instead that because they were not subjected to vote flipping (see straight lines on candidate chart) the subtraction yields near zero.

OK. You graphed the delegates cumulative graph. What happens when you create "votes minus delegates" for each candidate in each precinct and graph only the difference? This is the controversy in Alabama.

 

[RonRules]

OK. You graphed the delegates cumulative graph. What happens when you create "votes minus delegates" for each candidate in each precinct and graph only the difference? This is the controversy in Alabama.

Here's the simpler "votes minus delegates", but because it is raw data it looks nasty. I included a polynomial fit to express the trend.

Romney minus each of his delegates:

Santorum minus each of his delegates:

And clearly the victim of abuse, Ron Paul:

The difference is absolutely startling. One more reason to act.

 

[drummergirl]

Holy hand grenades, batman! The resemblance to Alabama is astonishing. Just. Absolutely. Mind boggling.

Here's the simpler "votes minus delegates", but because it is unfiltered it looks nasty. I included a polynomial fit to express the trend.

Romney minus each of his delegates:

Santorum minus each of his delegates:

And clearly the victim of abuse, Ron Paul:

The difference is absolutely startling. One more reason to act.

 

[hosef]

Hey all, I want to thank everyone who helped out on this.

I am a Ron Paul supporter and I was a delegate to the Nevada State Convention this last weekend. I am sure many of you know how much effort was put into Nevada in the weeks leading up to the caucus here. Unfortunately even after all the work put in by the supporters here, the final results were quite disappointing. However, after reading this and related posts I am no longer disappointed, I am rather upset. If I can help with this in any way, please let me know.

My background on the subject:
Over the last 2-3 days I have read several of the PDFs that have been made by you guys and I have read most of this thread. I have a thorough understanding of what we are looking for and what has already been used as arguments to debunk this.

My technical knowledge level:
I have a fairly advanced understanding of computers and am getting into programming. I have a fair understanding of mathematics, although I have never studied this advanced statistical stuff.

I don't want fraud to happen in TX or CA.

 

[RonRules]

My technical knowledge level:
I have a fairly advanced understanding of computers and am getting into programming. I have a fair understanding of mathematics, although I have never studied this advanced statistical stuff.

I don't want fraud to happen in TX or CA.

Welcome to the Ron Paul forums. This thread has brought in several new people to the forum and we're glad to see new people join.

What would be really useful is for someone to write a script in the statistical language "R" to produce the cumulative charts. I was going to do that, but every time I try to sit down to learn R, some new election has produced results that need to be analyzed quickly and I revert back to either Excel or Program4Liberty's Java program.

R is used widely in academia and has the richest set of charting options. If we code our analysis and make our charts with R, academia will much more quickly jump in and help. They use standard tools and right now, we're not. In my discussions with professors and graduate students, they see this effort as a really big project and would need considerable funding to get started, primarily to write the software needed. By having scripts that run off the tools they already use, they will be more likely to participate.

Ultimately, the R script needs to be set up to read the election data directly. It is likely that someone has already done that, so please scour the landscape thoroughly to see what has already been done in R with regads to election analysis.

For your sake, since you are getting into programming, that's a great opportunity to learn the most popular language that's used in statistics. I also need to mention that we don't use any "advanced" statistical stuff. We just happen to use a technique that is rarely used (Cumulative Vote Tally), but it's not complicated at all. It's just a matter of adding the votes from each precinct from smallest to largest and chart the corresponding candidate results.

So again welcome and don't hesitate to ask questions. Most likely, others that want to participate will have similar questions.

 

[hosef]

Alright, I am going to look into R after work today.

In the mean time could I have a more definitive feature list to work with. Eg. Do you want to do multiple states/counties on one chart? How many different ways do you want the data to be graphed? How will the input data be formatted? In some states the delegate votes have uncovered large issues with the election, how do you want the program to handle that? Also, how should I format the output to make it easier for a paid professional to work with, and therefore reduce the cost of a formal review?

 

[RonRules]

Alright, I am going to look into R after work today.

In the mean time could I have a more definitive feature list to work with. Eg. Do you want to do multiple states/counties on one chart? How many different ways do you want the data to be graphed? How will the input data be formatted? In some states the delegate votes have uncovered large issues with the election, how do you want the program to handle that? Also, how should I format the output to make it easier for a paid professional to work with, and therefore reduce the cost of a formal review?

It's best to start small, achieve one little thing and grow from there. You'll make small mistakes and that's fine, but you'll get quick feedback. (The internet is good about that). If you try to plan a whole large project, most likely you'll make large mistakes and maybe it won't get done.

I would recommend for now to just try to make a cumulative chart with "R". The data format we have been using is like this:

You save that data in the ".csv" format, a common format for exchanging between databases and spreadsheets.

The algorithm you use to produce the cumulative chart is this:

You can also learn how to produce the cumulative charts from the Java program that's been written by "program4liberty". That's what I use most of the time. The code in there should help you code the R script. http://sourceforge.net/projects/voteanalyze/ (Latest version is 1.4)

If you can manage to do only that, that would be huge. I now have over 2,100 ".csv" files that could readily be analyzed with "R".

"R" has a rich set of graphics and people share their R scripts in the public domain:

http://rgm2.lab.nig.ac.jp/RGM2/images.php?show=all&pageID=1042


It's quite likely that there's R scripts that read election data directly from the various state and county websites. (If anybody here could help him find that, that will help a lot). If not, that would be the next part of the project.

Presentation is a huge part of conveying information and getting accepted. People who won't understand the charts will not accept them.

Salesman dictum: "A confused mind always says no."

The fact that many people on this very RPF don't accept Vote Flipping is an indication that we have not presented the results correctly.

Good luck and give us an update regularly.

 

[RonRules]

Here's a ".csv" file reader written in R.

Load Data from csv File:
http://rgm2.lab.nig.ac.jp/RGM2/func.php?rd_id=quantmod:getSymbols.csv

The free programming environment for R that I use is this:
R for Windows, Version for R-2.15.0 (That's the latest)
Download here:
http://cran.r-project.org/bin/windows/base/

 

[hosef]

I am not that new to programming(by 'getting into it' I meant that I was trying to make money at it). I just wanted to get an idea of what you wanted it to do and what you might want it to do in the future. Also, From my work on an open source project called Drupal, I have had quite a lot of experience with people on the internet telling me about my own code.