The case for the occurence of algorithmic vote flipping

[kathy88]

I bet that was one tasty morsel

Good work it's about damn time!

 

[TheGrinch]

Nice work!

Though the math is still beyond me, it was clear that you were never going to get past the demographic explanation, without demonstrating that it couldn't be the case.

Some rich irony there, that the demographics (that have been used as an explanation for the anamoly), actually flat-line like the election results are supposed to.

Not only does that clearly debunk demographic explantions, but wouldn't you say that this only makes the anamoly that much glaring?

 

[RonRules]

I have received skeptical e-mails about the sparseness of my demographic choices.
My response is, if you feel that the above charts are not representative demographic samples, have at it. Here's your choice in California:

Pick any of these for the years between 2002 and 2010 and get charting!
http://swdb.berkeley.edu/d00/index.html

DEM Democratic Party Registration
REP Republican Party Registration
AIP American Independent Party Registration
PAF Peace and Freedom Party Registration
MSC Miscellaneous Registration
LIB Libertarian Party Registration
NLP Natural Law Party Registration
GRN Green Party Registration
REF Reform Party Registration
DCL Declined to State Registration
MALE Male
FEMALE Female
HISPDEM Latino Democrats
HISPREP Latino Republicans
HISPDCL Latino Declined to State
HISPOTH Latino Other Party
JEWDEM Jewish Democrat
JEWREP Jewish Republican
JEWDCL Jewish Declined to State
JEWOTH Jewish Other Party
KORDEM Korean Democrat (All Asian surname matches unconditional)
KORREP Korean Republican
KORDCL Korean Declined to State
KOROTH Korean Other Party
JPNDEM Japanese Democrat
JPNREP Japanese Republican
JPNDCL Japanese Declined to State
JPNOTH Japanese Other Party
CHIDEM Chinese Democrat
CHIREP Chinese Republican
CHIDCL Chinese Declined to State
CHIOTH Chinese Other Party
INDDEM Asian Indian Democrat
INDREP Asian Indian Republican
INDDCL Asian Indian Declined to State
INDOTH Asian Indian Other Party
VIETDEM Vietnamese Democrat
VIETREP Vietnamese Republican
VIETDCL Vietnamese Declined to State
VIETOTH Vietnamese Other Party
FILDEM Filipino Democrat
FILREP Filipino Republican
FILDCL Filipino Declined to State
FILOTH Filipino Other Party
DEMMUNK Dem Male age unknown
DEMM1824 Dem Male age 18-24
DEMM2534 Dem Male age 25-34
DEMM3544 Dem Male age 35-44
DEMM4554 Dem Male age 45-54
DEMM5564 Dem Male age 55-64
DEMM65PL Dem Male age 65 Plus
DEMFUNK Dem female age unknown
DEMF1824 Dem female age 18-24
DEMF2534 Dem female age 25-34
DEMF3544 Dem female age 35-44
DEMF4554 Dem female age 45-54
DEMF5564 Dem female age 55-64
DEMF65PL Dem female age 65 Plus
REPMUNK Rep male age unknown
REPM1824 Rep male age 18-24
REPM2534 Rep male age 25-34
REPM3544 Rep male age 35-44
REPM4554 Rep male age 45-54
REPM5564 Rep male age 55-64
REPM65PL Rep male age 65 Plus
REPFUNK Rep female age unknown
REPF1824 Rep female age 18-24
REPF2534 Rep female age 25-34
REPF3544 Rep female age 35-44
REPF4554 Rep female age 45-54
REPF5564 Rep female age 55-64
REPF65PL Rep female age 65 Plus
OTHMUNK Other Party male age unknown
OTHM1824 Other Party male age 18-24
OTHM2534 Other Party male age 25-34
OTHM3544 Other Party male age 35-44
OTHM4554 Other Party male age 45-54
OTHM5564 Other Party male age 55-64
OTHM65PL Other Party male age 65 Plus
OTHFUNK Other Party female age unknown
OTHF1824 Other Party female age 18-24
OTHF2534 Other Party female age 25-34
OTHF3544 Other Party female age 35-44
OTHF4554 Other Party female age 45-54
OTHF5564 Other Party female age 55-64
OTHF65PL Other Party female age 65 Plus
DREG1G Dem registered 1 cycle (since g08)
DREG2G Dem registered 2 cycles (Since g06)
DREG3G Dem registered 3 cycles (since g04)
DREG4G Dem registered 4 cycles (Since g02)
DREG5G Dem registered 5 cycles (Since g00)
DREG6G Dem registered 6 cycles (Since g98)
DREG7G Dem registered 7 cycles (Since g96)
DREG8G Dem registered 8 cycles (Since g94)
DREG9G Dem no or earlier reg date than 94
RREG1G REP registered 1 cycle (since g08)
RREG2G REP registered 2 cycles (Since g06)
RREG3G REP registered 3 cycles (Since g04)
RREG4G REP registered 4 cycles (Since g02)
RREG5G REP registered 5 cycles (Since g00)
RREG6G REP registered 6 cycles (Since g98)
RREG7G REP registered 7 cycles (Since g96)
RREG8G REP registered 8 cycles (Since g94)
RREG9G REP no or earlier reg date than 94
OREG1G OTHER PARTY registered 1 cycle
OREG2G OTHER PARTY registered 2 cycles
OREG3G OTHER PARTY registered 3 cycles
OREG4G OTHER PARTY registered 4 cycles
OREG5G OTHER PARTY registered 5 cycles
OREG6G OTHER PARTY registered 6 cycles
OREG7G OTHER PARTY registered 7 cycles
OREG8G OTHER PARTY registered 8 cycles
OREG9G OTHER Party no or earlier reg date than 94

Good luck.

 

[RonRules]

My cat just caught a large grasshopper. He has not killed it yet, but he's just playing with it. He's ripped off one of the legs so it can't jump too good.

He's now letting it try to fly away only to catch it spectacularly in mid-air.

Good kittie:

I feel the same way about this demographic argument.

I tried to find the most obscure demographics:

Filipino people that declined to state their party affiliation. I thought maybe, just maybe this demographics agrument would have a little bit of life left in it.

Oh well:

 

[drummergirl]

Not only does that clearly debunk demographic explantions, but wouldn't you say that this only makes the anamoly that much glaring?

yes

 

[RonRules]

Hi Liberty1789, I have a fairly urgent request from a VIP to do more charts like this for counties that exhibit flipping:

I don't think I personally can do them quickly because I have not set up the confidence bound stuff.

Ideally, if you could do these 4 charts that would be great: (we need examples of flipping and no flipping, both from the same state/county)

PS: The color for Romney in the above 2012 chart is non-standard (ugly green) and can be easily confused with Gingrich.

 

[drummergirl]

Hi Liberty1789, I have a fairly urgent request from a VIP to do more charts like this for counties that exhibit flipping:

I don't think I personally can do them quickly because I have not set up the confidence bound stuff.

Liberty just posted in the other thread that he's still out of town for a couple more days. He's a far better Excel master than me, but we can probably kick them out if we need to.

 

[Anti Federalist]

OK, I've avoided this along enough.

Can somebody give me a one paragraph summary of what the argument here is?

 

[drummergirl]

OK, I've avoided this along enough.

Can somebody give me a one paragraph summary of what the argument here is?

It's a page, but I don't think it can really be done in a paragraph... well, maybe Stephen Colbert could do it

https://docs.google.com/document/d/1nycaPPtpBCyc1DOt9ybWT1tnOXeczMHav3nfdkBzdBw/edit?pli=1

And here's the summary, though if you are not a mathophile, stick to the first 10 pages:

https://docs.google.com/document/d/1EokVx9tDsrjAJ-7H9XoPv3KmZYDvVjSFJ4cuxJTo1iE/edit

 

[drummergirl]

The best analogy I have is the one that helped my math-phobic kid to understand.

Suppose the entire neighborhood is away from home one Friday night at the local high school football game. While we're all away, thieves secretly enter our homes, being careful to leave no trace of their activities. They are after silverware. But, rather than take all the silverware from one house (ballot box stuffing) they want to be more subtle and not get caught. So, they take a fork and a spoon from 1 house, 2 knives and a server from the next, a spoon and knife from the third, etc. How long would it take before anyone noticed they were missing anything? Even when they did notice, would it occur to them that it was a theft or when/how it happened? Would they ever compare notes with the neighbors and realize what actually happened?

This is what vote flipping is like; the vote totals remain the same, so the procedures for canvassing miss them. And where recounts are done on a handful of precincts, they usually pick low vote totals if they can, so the flippers systematically avoid low vote total precincts. Then they flip a percentage of votes (5% we see a lot) from other candidates to Mitt Romney. 5% may not sound like a lot, but it can make the difference between first and second place. And where delegates are distributed proportionally, it can make a very big difference.

 

[RonRules]

OK, I've avoided this along enough.

Can somebody give me a one paragraph summary of what the argument here is?

This is the largest election fraud ever. I could stop here, but let me continue.

In ALL the States that have been voted thus far, except Puerto Rico, a large number of votes are transferred to Romney (and only Romney) from various candidates. The victims so far are: Ron Paul, Gingrich, Santorum, Huntsman, Rick Perry and Bachmann.

This table will give you an approximate proportion of votes that were stolen from the various candidates. The values highlighted in red are the more egregious ones:

There are different ways to see this happening but you need charts to see it. Some charts are easier to understand than others but you DON'T NEED MATH to understand them.

Let me bring back the old style chart to help make the point. Lets say you want to see if a particular candidate is up to no good and you want to know if he/she does better in bigger precincts.

You may decide to do a chart like this one below, which is simply the proportion of votes each candidate got as a function of precinct size:

(it's hard to see anything, but your suspicions may be enhanced that Romney is creeping up for no justifiable reason)

Then you have a stroke of genius and think it's a good idea to make the X-Axis cumulative in order to eliminate demographics from the charts. (I'll explain why that is in a minute).

You get THIS chart and things become very clear:

Romney is indeed up to no good, but how is he doing that?

We believe (us flippers) that no vote flipping occurs in small precincts (<~250 votes). This is done to prevent easy detection.

We believe (us fundamentalist flippers like myself) that no fraud occurs at the individual voting machine, whether or not the machine is of the paper ballot scanner type or Direct Entry (DRE) video screen type. Even hand counted ballots are not immune to this problem as was proven by Liberty1789 in NH.

That's because the fraud is most likely happening at the central tabulator. At the county level, the tabulator starts flipping votes once a precinct exceeds 250 votes (approximately). This can happen in miliseconds, as soon as the central vote tally is requested by the operator.

The Central Tabulator flips votes in an exponential fashion as a function of precinct size. This results in a near perfect straight line on the cumulative charts, generally with a .99 (99%) correlation.

We don't know exactly how the central tabulator machine does that. The central tabulator is usually a software program (like the Deibold GEMS) that simply runs on a PC. The software itself could have nefarious code causing the vote flipping or it could be infected by a virus once it has been installed.

If the PC containing the Central Tabulator software is ever allowed to connect to a network (LAN or directly to the internet through a modem and ISP), the software has a high risk of infection. The software could also be infected as soon as someone inserts a non-virgin USB key or DVD disk in the Central Tabulator PC.

I'll next explain why demographics do NOT explain Romney's rise on the cumulative charts. (I'm looking for Iowa demographics data)

Meanwhile have a look at this. People say Romney does better in bigger precincts because more old people vote for Romney and there's more of them in bigger precincts. If you chart age demographics as a function of precinct size, you get a nice flat line.

Demographics -- ANY demographics -- will not explain the steady upward trend that Romney makes on Cumulative Precinct Vote Tally charts.

That's why the Cumulative charts are so beneficial and instructive to this analysis. The demographics results will be the same in Iowa (Flat-Line). I'll update the post when I find the data. I may only get to that tomorrow.


What can you do about it?

ALERT ALL COUNTY ELECTION CLERKS.

Explain the problem. Urge them that if elections have not yet happened, to NOT USE the central tabulator. They DON'T NEED IT. It's a simple matter to add all the precicnt results in Excel and use that to post the results. They do it that way in 9 counites in Wisconsin and it turns that that there's exactly 9 counties that don't exhibit vote flipping. Outagamie county is a great example to show. (http://i269.photobucket.com/albums/...2012_WI_OUTAGAMIE_County_PresPrimariescsv.png)

If elections have occured, to re-canvas and check the individual candidate counts for each precinct. Many counties only check the totals and vote flipping will not be identified because the total votes remain the same.

One more thing, there should be ZERO tolerance of error when transferring numbers from the individual precincts to the central location. Error tolerance is only justified with ambiguous paper ballots that are scanned or hand counted.

 

[ratio]

The problem with conspiracy theories is that just like in this case, it replaces one question with another (usually bigger) one. Let's presume there is actually vote flipping. How would Romney go about organizing this? It would involve buying every person involved in the counting, across the entire nation.
From a probabilistic point of view: I think it's safe to assume that at least 1% of the human population have enough moral fiber to stand up against it and go to the press. Now let's say it would take 1,000 people in the scheme. What's the probably of nobody speaking up? 0.99^1000 = 0.004%. And 1,000 people is low balling it, there would have to be far more to pull off the scheme.

 

[RonRules]

Now let's say it would take 1,000 people in the scheme.

Answer: NO.

Hi ratio,

I note that you're a new poster. Rest assured that we'll keep a close eye on you. If you need to post links because you can't right now, send me a PM and I'll help you out. Maybe.

Vote flipping can be effected nationwide with two people: One with money and one with programming skills.

A programmer with enough free time can do this alone.

There is no need for a complex, unlikely conspiracy.

 

[The Man]

Explain the problem. Urge them that if elections have not yet happened, to NOT USE the central tabulator. They DON'T NEED IT. It's a simple matter to add all the precicnt results in Excel and use that to post the results. They do it that way in 9 counites in Wisconsin and it turns that that there's exactly 9 counties that don't exhibit vote flipping. Outagamie county is a great example to show. (http://i269.photobucket.com/albums/...2012_WI_OUTAGAMIE_County_PresPrimariescsv.png)

Hey RonRules can you steer me to see conclusively the correlation between the 9 locally-counted counties and flat-lining versus the centrally tabulated counties? Thanks.

 

[RonRules]

Hey RonRules can you steer me to see conclusively the correlation between the 9 locally-counted counties and flat-lining versus the centrally tabulated counties? Thanks.

You asked at the right time!



Right now, I'm making all the charts more easily accessible. There were hard to find in the hundreds of pages posted on the internet.

You can get all the charts by a series of 50 links for each state:
www.photobucket.com/flipping_WI
www.photobucket.com/flipping_NH
www.photobucket.com/flipping_SC
www.photobucket.com/flipping_VA
....
Many states have yet to be voted on, so you won't find any charts.

I also have www.photobucket.com/flipping_US (for upcoming nationwide charts.

The Wisconsin counties that exhibit no flipping (using the eyeball analysis method) are:

Marinette, Oconto, Oneinda, Polk, Richland, Wahsburn, Waukesha, Outagamie ...

I have called them all and they confirmed that they don't use central tabulators.

Dunn and Burnett counties may also be "no flipper" candidates, but have not called them yet.

 

[The Man]

You asked at the right time!



Right now, I'm making all the charts more easily accessible. There were hard to find in the hundreds of pages posted on the internet.

You can get all the charts by a series of 50 links for each state:
www.photobucket.com/flipping_WI
www.photobucket.com/flipping_NH
www.photobucket.com/flipping_SC
www.photobucket.com/flipping_VA
....
Many states have yet to be voted on, so you won't find any charts.

I also have www.photobucket.com/flipping_US (for upcoming nationwide charts.

The Wisconsin counties that exhibit no flipping (using the eyeball analysis method) are:

Marinette, Oconto, Oneinda, Polk, Richland, Wahsburn, Outagamie ...

I have called them all and they confirmed that they don't use central tabulators.

Thanks! I just found your earlier post on the 9 Wisconsin unflipped counties. Truly a great find RonRules! I wonder if someone could do the legwork and determine in other states which counties are LOCALLY or CENTRALLY tabulated. If this holds at a high correlation, this is HUGE.

 

[affa]

The problem with conspiracy theories is that just like in this case, it replaces one question with another (usually bigger) one. Let's presume there is actually vote flipping. How would Romney go about organizing this? It would involve buying every person involved in the counting, across the entire nation.
From a probabilistic point of view: I think it's safe to assume that at least 1% of the human population have enough moral fiber to stand up against it and go to the press. Now let's say it would take 1,000 people in the scheme. What's the probably of nobody speaking up? 0.99^1000 = 0.004%. And 1,000 people is low balling it, there would have to be far more to pull off the scheme.

I believe it is fraud. I also believe Romney probably doesn't even know about it.

 

[bbwarfield]

The problem with conspiracy theories is that just like in this case, it replaces one question with another (usually bigger) one. Let's presume there is actually vote flipping. How would Romney go about organizing this? It would involve buying every person involved in the counting, across the entire nation.
From a probabilistic point of view: I think it's safe to assume that at least 1% of the human population have enough moral fiber to stand up against it and go to the press. Now let's say it would take 1,000 people in the scheme. What's the probably of nobody speaking up? 0.99^1000 = 0.004%. And 1,000 people is low balling it, there would have to be far more to pull off the scheme.

it actually takes 1,000 of people not noticing.... and really maybe 2 or 3 people setting it all up... Romney probably would have no knowledge of it... one person puts the line of code into the machines.... 1,000 of people ignoring it.... thats all it takes

 

[affa]

Thanks! I just found your earlier post on the 9 Wisconsin unflipped counties. Truly a great find RonRules!

link?

 

[affa]

Ok kiddies, this is another big piece to fit the puzzle:

As I have been saying for a while, I suspect the vote flipping takes place at the Central Tabulator equipment. I don't know if the software from the manufacturer itself is corrupt or if it's a virus that affects it.

Central Tabulators are a piece of software with names like Diebold "GEMS" (now Dominion), Sequoia's "WinEDS", Scytl/SOE "Results Consolidation Platform (RCP)", Election Systems & Software (ES&S) also with "GEMS" and "Unity Election Reporting Manager".

Here are 8 counties from Wisconsin below. What's common with all these counties? Among other things, note they don't have computer generated reports output like in Dane county (see links below each county name to get each county's data). Their results are provided as a spreadsheet output saved as PDF's or HTML. One county (Richland) types directly into their Web-Based program to display the results on the county website

Here are the counties:

Marinette County, raw results and processed chart:
http://www.marinettecounty.com/i_marinette/d/tally_-_county_board_sg.pdf

Oneida County, raw results and processed chart:
http://www.co.oneida.wi.gov/section.asp?linkid=1162&locid=135

Oconto County, raw results and processed chart:
http://www.co.oconto.wi.us/i_oconto/np/official_4-3-12_29_pres._sum.pdf

Outagamie County, raw results and processed chart:
http://www.outagamie.org/index.aspx?page=837

Polk County, raw results and processed chart:
http://www.co.polk.wi.us/coclerk/election-results.asp

Richland County, raw results and processed chart:
http://co.richland.wi.us/election/april_3_2012/presidentialRepublican.html

Washburn County, raw results and processed chart:
http://www.co.washburn.wi.us/news/county-clerk/362

Waukesha County, raw results and processed chart:
http://www.waukeshacounty.gov/uploa..._Results/Election Results_Pres and County.pdf

Notice there NO FLIPPING?

Here's the big news: I've called everyone of these counties and NONE
use any kind of Central Tabulator software.

They perform the final tabulation by hand, enter it in Excel and upload it to their website. They do however use voting machines at the individual precincts, mostly Sequoia AVC, but no central tabulator at the county level. That's the key.

I paraphrase a famous communist: "It's not who votes or how they vote, it's who (in this case what) counts the votes!"

You can see what individual voting machine equipment they use with this map of election equipment in Wisconsin:

It's a little hard to read, so use this to first find where the county is:
http://en.wikipedia.org/wiki/List_of_counties_in_Wisconsin

found it.