The case for the occurence of algorithmic vote flipping

[drummergirl]

In my mind for validation it would have to be taken to an outside source, re-run the work re-checked the numbers by someone without a horse in the race. This would ultimatley be a huge factor in proveing or disproving the results and your conclutions.

I don't really think there is such a thing as someone "without a horse in the race" when it comes to presidential politics. If you wanted someone with no opinion, you'd probably need extraterrestrial assistance And frankly, at this point in the campaign, Santorum, Gingrich, and Obama have as much or more to lose than Ron Paul. This only started here because Ron Paul was affected first.

Part of the point is the math is pretty easy; it's not rocket science. Anyone who's completed stats 101 can do the math. Most people hate math, so no matter how many times you say, "the math shouldn't do this", their emotions which scream, "this can't be happening because it would ruin my world" take over and they just simply refuse to see the truth.

This is one of those times when the truth is excrutiatingly painful. So, it's ok to have a cry, take some peptobismol, do some kickboxing, take a xanax, or whatever is legal that helps you to grieve - go do that. Then please consider what you can do with this information to make a difference.

This needs to go to election administrators where primaries have yet to be conducted. This needs to go to attorney generals where fraud has already happened. If you wait for the media to tell you what to do, may your chains rest lightly upon you.

[drummergirl]

I agree with jbauer, what is the next step?



I am currently emailing drummergirl's summary to the county clerks in Wisconsin, because it is voting next week. I am trying to make it clear in the email that I am contacting them as a concerned citizen who wants to ensure that the voting process is on the up and up, whether the claims are true or not. I am including some questions about how voting machine testing is done, based on some concerns that came out in the thread.

Thank you!

Really, all we know, is something is wrong. We don't know who is behind this or exactly how it's done (we have a couple ideas, but they are speculative). And just like you don't have to know every detail to report a theft (I don't know how they got in, but my stuff is gone), we don't need to have every piece of information that you'd put in a documentary to get this ball rolling.

[jbauer]

There has to be a firm out there that will do the independent analysis of your data without being given your conclutions. If they come to the same synopsis that you did independantly from your observations then you have validation. My guess is someone with the right connections (I don't have them) could get this infront of the right people.

Heck with our "younger" crowd there has to be someone that works or attends a university that would love to pour over the data. Heck give it to some flaming Democrat, they'd be all over it if it meant they could destroy the Republicans. My fear would be that if you just start emailing this report to every tom dick and harry its going to look exactly like it does as some sort of massive conspiracy theory. Until it can be confirmed its a theory.

I don't really think there is such a thing as someone "without a horse in the race" when it comes to presidential politics. If you wanted someone with no opinion, you'd probably need extraterrestrial assistance And frankly, at this point in the campaign, Santorum, Gingrich, and Obama have as much or more to lose than Ron Paul. This only started here because Ron Paul was affected first.

Part of the point is the math is pretty easy; it's not rocket science. Anyone who's completed stats 101 can do the math. Most people hate math, so no matter how many times you say, "the math shouldn't do this", their emotions which scream, "this can't be happening because it would ruin my world" take over and they just simply refuse to see the truth.

This is one of those times when the truth is excrutiatingly painful. So, it's ok to have a cry, take some peptobismol, do some kickboxing, take a xanax, or whatever is legal that helps you to grieve - go do that. Then please consider what you can do with this information to make a difference.

This needs to go to election administrators where primaries have yet to be conducted. This needs to go to attorney generals where fraud has already happened. If you wait for the media to tell you what to do, may your chains rest lightly upon you.

[RonRules]

I am currently emailing drummergirl's summary to the county clerks in Wisconsin, because it is voting next week. I am trying to make it clear in the email that I am contacting them as a concerned citizen who wants to ensure that the voting process is on the up and up, whether the claims are true or not. I am including some questions about how voting machine testing is done, based on some concerns that came out in the thread.

Very good. I also submitted drummergirl's summary to my Registrar of Voters. I did this in person and made a one hour presentation along with a knowledgeable friend. I urge everyone to do the same, even if you are unsure of the evidence thus far. Let them decide. The staff that I met immediately "got it", and we are continuing the dialog. I'll soon do the same with three neighboring counties: Los Angeles, Orange and San Bernardino.

We have enough documentation. We have a good summary document. It's time to alert the people in charge. Forget the media for now, they'll spin this immediately. Check with people that know statistics.

Where vote has taken place, urge them to NOT certify the results.
Where voting has not taken place, alert them to the problem, explain the possibility of vote flipping and if you are knowledgeable suggest additional checks.
1) For the states that have voted: Have them re-check the numbers between precincts and central tabulation. NOT JUST THE TOTALS, but each individual candidate. That's extremely important because currently this is not done in the places I have checked.
2) Tell them to load up the central tabulator software on a virgin computer or disk drive. Don't ever let that computer hooked up to the internet.
3) Check the machines with at least vote counts > 300. Thus far, in the procedures I have reviewed the vote count in test never exceeds 50.
4) Set the machine date to the election date and time before testing. This is not currently done! It's an easy way for a virus to hide and "wake up" on election day.
5) Think of more checks yourself and report here.

[drummergirl]

The statistical argument assumes that the precinct size and vote percentage are independent variables. If there's a correlation between precinct size and vote percentage, then the statistical argument is incorrect.

The central problem is that a correlation exists between precinct size and vote percentage. The fact that this correlation exists is bizarre. Historically, votes per precinct (size) and a candidate's percentage of the vote are independent variables. This is why pollsters can conduct polls. If it were normal for these to be dependent variables, polling would have to account for this or be meaningless. Pollsters do try to account for known demographic variables - sex, age, race, religion. The "number of votes in your precinct" is not a known demographic variable.

And, if you think a bit about what constitutes a small precinct or a large precinct, it shouldn't be. A small precinct might be rural or it might be predominantly democrat or it may have had low turnout for an odd reason (weather, local problem, etc.). A large precinct might be suburban or urban or highly republican or had a high turn out for an odd reason. And in some cases here, the difference between "small" and "large" is about 20 votes.

I'm truly not sure from your posts if you don't understand the math (if so, please read the summary) or if you do understand the math but you just don't want to see what it is saying. Please keep in mind, the reason that the lines should go flat is that the total number of votes is constantly increasing, so slight variations in vote totals from precinct to precinct will be less and less "noisy" as you approach 100% counted. (remember the first graph in the summary?)

[Nirvikalpa]

If you have nothing positive to add to this discussion, do not post. Stop derailing the thread.

[jbauer]

What you say might be absolutley correct. If the creators of this want to validate it, it needs to be tested by an outside unbiased source. They'll either find out that its completley incorrect or that it has some legs. If its as easy to disclaim as you say it is, then it should be a pretty short interview and thus quite cheap.

I don't think they've proven anything other than Romney won. But what I can say is that they've put so much time and effort into this that they probably know the numbers better than anyone else. If they take it to the PHD and he/she discredits it within minutes it isn't going to cost much and it can be put to rest. However, if its taken to the PHD and there's something there then its could be earth shattering.

I'm a risk vs reward guy, the risk ($) of having it evaluated is very small. The reward, even if its not what they want to hear finally puts this to bed....or they find out there's something to it and they all get to write books and movies and do leno and letterman etal.

Seems like a good investment to me.

It doesn't need an outside party to check it. Anyone that is a little bit math inclined and taken a few upper level university courses can tell you that they haven't produced one iota of proof about anything. Nothing they are doing is even nearly complicated or advanced enough to require an expert to look at it. That's like saying you need to get a Ph.D mathematician to check your kids arithmetic homework.

[arsenius]

Very good. I also submitted drummergirl's summary to my Registrar of Voters. I did this in person and made a one hour presentation along with a knowledgeable friend. I urge everyone to do the same, even if you are unsure of the evidence thus far. Let them decide. The staff that I met immediately "got it", and we are continuing the dialog. I'll soon do the same with three neighboring counties: Los Angeles, Orange and San Bernardino.

1) For the states that have voted: Have them re-check the numbers between precincts and central tabulation. NOT JUST THE TOTALS, but each individual candidate. That's extremely important because currently this is not done in the places I have checked.
2) Tell them to load up the central tabulator software on a virgin computer or disk drive. Don't ever let that computer hooked up to the internet.
3) Check the machines with at least vote counts > 300. Thus far, in the procedures I have reviewed the vote count in test never exceeds 50.
4) Set the machine date to the election date and time before testing. This is not currently done! It's an easy way for a virus to hide and "wake up" on election day.
5) Think of more checks yourself and report here.

Unfortunately I'm overseas, so I'm limited to email. But, here is the list of things I mentioned in the email I sent out. Mostly the same, but number 6 is based on something I saw in that Bev Harris movie (name escapes me at the moment, obviously...).

1. Are voting machines and tabulators tested by setting the machines' dates to the date of the actual election?
2. Are voting machines and tabulators tested using a range of votes? (For example, under 50, more than 200, more than 1000?)
3. Are the voting machines and tabulators tested using actual candidate names?
4. Are recorded vote totals for each individual candidate reported back to precincts for checking, or only total votes recorded?
5. Are the voting machines and tabulators connected to the internet,
or other network?
6. Are voting machine vendors allowed to interact with machines or software at any point between testing and vote certification?

And this explanation:

Items 1 through 3 are critical in case any virus is programmed to only operate when those parameters are true. In particular, item 2 is
critical to the phenomenon observed in the document. It is my understanding that item 4 is a common practice, but it allows the
alleged flipping to go undetected. Also, it is critical to note that the alleged flipping can be performed using a relatively low number of
flipped votes per precinct, as low as 3-4 votes per precinct can add up to the effects shown.

[dsw]

I'm truly not sure from your posts if you don't understand the math (if so, please read the summary) or if you do understand the math but you just don't want to see what it is saying. Please keep in mind, the reason that the lines should go flat is that the total number of votes is constantly increasing, so slight variations in vote totals from precinct to precinct will be less and less "noisy" as you approach 100% counted. (remember the first graph in the summary?)

I think I understand the math. And we seemed to agree at one point that the argument that the line should go flat is only valid if there's no correlation between precinct size and vote percentage. Right?

But there clearly is, in a lot of cases, exactly that correlation. Given that, it's not surprising that the line doesn't go flat.

That's where the rigorous mathematical part ends. The graph isn't flat exactly because this correlation exists, in some races.

The non-rigorous argument is that the correlation shouldn't exist. That's based on "if you think about it" suggestions that are definitely in the "things that make you go hmmmm" category. But that's not math. (And your "if you think about it" here turns out not to be the case in the Va Beach data, where there's some very strong clustering of similar-sized precincts, so that the sorted precinct order is to a significant degree an ordering of different geographical neighborhoods.)

I'm not disagreeing with the math part. The summary doesn't, however, keep the non-math part distinct from the math part. It talks about the non-math part as if it were math, when it's more of a "something's not right here."

You write: Please keep in mind, the reason that the lines should go flat is that the total number of votes is constantly increasing, so slight variations in vote totals from precinct to precinct will be less and less "noisy" as you approach 100% counted.

And my point in a nutshell is that this argument ONLY makes sense if there's no correlation between precinct size and vote percent. So for this mathematical argument to apply you would have to first show that there's no such correlation, because otherwise the math doesn't work.

But in fact there IS such a correlation. So the math part of the argument doesn't apply. And the real point of contention is the question of why the correlation exists. That part of the argument has not been reduced to math. Which is not to say that there aren't things going on there that make one go hmmm.

[drummergirl]

My fear would be that if you just start emailing this report to every tom dick and harry its going to look exactly like it does as some sort of massive conspiracy theory. Until it can be confirmed its a theory.

A valid concern. As a matter of fact when this was first brought to my attention via a friend's facebook post, my first thought was literally, " Oh my God, more tinfoil hat stuff? I know my statistics so this will take about 5 minutes to dismiss." Then I started looking at the data. Then I got sick to my stomach. Now I'm here.

Honestly, if you know of a better way to disseminate this information please do it.

So far the media hasn't picked it up, not even the alternative press. I doubt it will get much press unless/until indictments start coming in. Too much math.

Election administrators know about the math and they generally care about the integrity of the electoral process (their reputations and careers depend on fair elections). So be a concerned citizen and politely ask questions. Be appreciative of their time and volunteer to help out. Hopefully, they will have some ideas about what is going on and how to prevent it in the future.