The case for the occurence of algorithmic vote flipping

[dr.k.research]

Oh man, I can hardly wait to see the precinct level data

New York was one of the first states to impliment the Dominion Voting Systems machines. These machines are functioning in 52 counties.

The machines are also found in Upstate Michigan. Dominion is a relatively new player in this business.

http://www.dominionvoting.com/field/new_york

Now, here's the big one. Dr. Paul did BEST in Schuyler County, also where the Super Brochures where most extensively used. Schuyler apparently was counted by Dominion machines:

http://en.wikipedia.org/wiki/List_of_electronic_voting_machines_in_New_York_state

I believe this is the data you need for further analysis.

 

[RonRules]

Montana results, nothing new, Romney's a flipper.

Here are the three most populous counties in MT:

Data source:
http://electionresults.sos.mt.gov/resultsCTY.aspx?type=FED&rid=450000711&pty=REP&osn=100&map=CTY

Clearly the problem is pervasive and they (whoever "they" are) can't stop it. There's was no need at all for Romney to flip the vote in MT. The fix was in, before the Jan 3 Iowa primaries.

 

[drummergirl]

thank you Dr K

 

[RonRules]

The machines are also found in Upstate Michigan. Dominion is a relatively new player in this business.

Dominion is an old Canadian company. What's important to know is that they purchased the Deibold equipment. (I believe that ES&S was prevented from buying Deibold as that would have caused a near monopoly in the voting equipment business)

Now we have Syctl, a Spanish company who bought Clarity elections and now running elections in 26 states (a majority) and nobody cares.

 

[drummergirl]

And it's yet another piece of information for the puzzle:

No candidate getting over, say, 55% of the vote would need to flip. Yet there's Romney, over 60% and still flipping. It's not a sophisticated program/virus/algorithm.

Montana results, nothing new, Romney's a flipper.



Clearly the problem is pervasive and they (whoever "they" are) can't stop it. There's was no need at all for Romney to flip the vote in MT. The fix was in, before the Jan 3 Iowa primaries.

 

[RonRules]

Now, here's the big one. Dr. Paul did BEST in Schuyler County, also where the Super Brochures where most extensively used. Schuyler apparently was counted by Dominion machines:
I believe this is the data you need for further analysis.

PRESIDENT OF THE UNITED STATES for Schuyler County (Republican)

CANDIDATE VOTES RECEIVED PERCENTAGE
Ron Paul 167 29.715%
Newt Gingrich 77 13.701%
Mitt Romney 253 45.018%
Rick Santorum 65 11.566%

Indeed Ron Paul did quite well there. Unfortunately, there's no precicnt-level data from that county on their website. If you can make a request for it, I'll analyze it.

 

[RonRules]

No candidate getting over, say, 55% of the vote would need to flip. Yet there's Romney, over 60% and still flipping. It's not a sophisticated program/virus/algorithm.

Certainly true. Dumb criminals!

We'll see how dumb these criminals are when I find Utah to be a flipper.

 

[RonRules]

South Dakota has a great election website and they have precinct-level results for each county.

They also have a downloadable spreadsheet with all the low-level results, but the link is bad.
http://electionresults.sd.gov/resultsCTY.aspx?type=SWR&rid=3934&pty=REP&osn=105

The resource cannot be found.

Description: HTTP 404. The resource you are looking for (or one of its dependencies) could have been removed, had its name changed, or is temporarily unavailable. Please review the following URL and make sure that it is spelled correctly.

Requested URL: /MasterPages/ResultsExport.aspx


Can someone call them and bitch about it for me. I've got a lot to do today, besides charts.

 

[RonRules]

Come on girls, join in!

 

[RonRules]

What else is new? Here's New Jersey:

 

[RonRules]

Anybody care about Los Angeles?

LA County has one of the suckiest data format. It's the stuff that came out of computers about 30 years ago. It's also categorized by Congressional districts, not individual precincts. If you can find precinct-level data, please tell me.

Here are the districts covered in the chart:
090CNDN 091CNDN 092CNDN 093CNDN 094CNDN 095CNDN 096CNDN 097CNDN 098CNDN 099CNDN 100CNDN 101CNDN 102CNDN 103CNDN 104CNDN 105CNDN 106CNDN 107CNDN

Data source:
http://rrcc.co.la.ca.us/elect/downrslt.html-ssi

 

[RonRules]

Hey anybody wondered about Scott Walker's recall election?

Duane is one of the few counties that have the low level data available. Please help me find more.

Here's the list of counties. Sort by size and call up the county administrators and get them to post the precinct-level data:
http://en.wikipedia.org/wiki/List_of_counties_in_Wisconsin

 

[RonRules]

Here's the Wisconsin Lt. Gov chart. It's understandably similar, but not identical to the Gov. chart.

I suspect flipping BTW.

Gov. totals:
Scott_Walker Tom_Barrett Hari_Trivedi
77449 175934 1236


Lt. Gov totals:
Rebecca_Kleefisch Mitchell_Mahlon
75141 174827



Data source:
http://countyofdane.com/clerk/elect2012r2.html

 

[RonRules]

Orange county California: 2044 data points on the chart, representing a total of 15948 votes.

Note the clean upward slope for Romney:

Data source:
http://ocvote.com/fileadmin/live/pri2012/run10/districts.pdf

Romney is still batting a 1,000.

 

[RonRules]

San Francisco:

Data source:
http://www.sfelections.org/results/20120605/detail.php

In the results thread yesterday, I saw this commen:
"No Fricken way one candidate could possibly get that many votes...I'm calling election fraud on Romney...82%....come on."

The above chart largely explains it.

 

[RonRules]

California 2012, Republicans, Entire State based on Congressional Districts:

This chart represents 53 data points (congressional districts) and 1,446,140 votes

Today I will be at the Riverside Registrar of voters for the 100% recount of the DRE machines as well as the 1% mandatory recount of the entire county. The Registrar promised me that if no randomly selected precincts have more than 250 votes, they will recount an extra one that has 250 votes.

 

[RonRules]

Libertarian Party, all of California:

Why is Gary Johnson not perfectly flat-lining like the libertarians did in 2008?

Two reasons:
1) There's only 58 data points in this chart. When I get the full state database on a precinct basis, I expect it to flat line perfectly.

2) In the above charts with the 58 data points, it's because he got most of his votes in Orange and San Diego counties and it skewered the results a bit. You can see that little percentage differences on the straight precinct vote chart below.

Verdict: The libertarians flat lined again.

Data source:
http://vote.sos.ca.gov/returns/president/party/libertarian/county/all/

Here are the 2008 charts.

Libertarians:

Republicans:

 

[RonRules]

Riverside County, CA is currently doing the 100% DRE Machine (touch screen, direct entry) manual count. I am an observer again today.

The touch screen voting machine puts out a "Poll Tape", on which every single vote is printed. There is a report at the end of the tape and this report is also displayed on the screen as I showed several posts back.

The election workers are comparing the Poll tape with the output of the central tabulator for each machine. DRE machines record about 5% of the votes in Riverside county. 100% of those votes are checked.

It's still not perfect, because if the machine itself is doing the flipping nobody will ever know. They also spend a small fortune for those machines, the maintenance, the management and all those election workers having to check 100% of the votes. With all that trouble the process is not at all guaranteed. Votes could easily be flipped though the machine's software.

A straight manual count public would be so much easier and virtually fraud proof.

Here are pictures of the election workers doing the 100% DRE manual check:

 

[RonRules]

This afternoon they will do the 1% county wide check of the paper ballots that were scanned though the Optech 400C scanners.

There are approximately 820 precincts in Riverside. The assistant registrar randomly picks 1% of those, rounded up, which is 9 precincts. If those chosen precincts don't cover some of the local elections, additional precincts will be added.
This year, the voter turnout is very low. It is quite likely that out of those 9 precincts, none will have 250 votes.

Yours truly has gotten the Registrar to agree that one more precinct with more than 250 votes will be manually counted.

Here the assistant registrar is rolling the dice to randomly pick the first 9 precincts:

That's how the sausage is made in Riverside.

 

[RonRules]

Here's a quickie. Green party, California. Flat-lining as expected:

58 data points, hence a little roughness.