We have finally linked an increase in 45 votes to a specific voting machine

[LaydeeBug]

Link to tweet

[malaise]

1. Off to the greatest page for you

[MaeScott]

3. Thorough analysis, yes. Data/othergeeks, what say you?

[dalton99a]

2. Kick.

[wildeyed]

4. Did you verify the original analysis?

https://sean-case.github.io/avcedgetrump.html

It's very dense, more than it needs to be, IMO. I'll look more carefully later. FYI, that Farb Twitter account also posted some info earlier about Wisconsin voting that was either misleading on purpose or just very wrong. They didn't seem to understand that Wisconsin is a same-day registration state and that is why the tallies changed a few weeks after the election. SDR votes are counted as provisional until the identity and residence of the individual is verified. Once that happens, they are added to totals. They were acting like it meant the change in tallies was a sign of fraud. Very strange. Just based on that, I would be skeptical about any suspicious 'proof' they offer.

I know I'm not changing the minds of the true believers, but seems to me that if you are going to spend days analyzing data and then tweet inflammatory speculation about your findings, you would know a very basic fact like how the state's voter registration system works.

Trump DID steal Wisconsin, but with voter ID and fake news propaganda.

[Bernardo de La Paz]

6. Thanks for providing a non-tweet link to read. . . . nt

[wildeyed]

11. Lemme know if you can make head or tail of it.

They are choosing to say a very simple thing in a very complicated which is suspect on its own. Knowing that they also did zero research or ignored basic info about Wisconsin's elections already, makes it hard to take their 'linear regressions' seriously.

[Bernardo de La Paz]

12. Thorax

The proportion of votes coming from use of the Sequoia AVC Edge electronic voting system in Wisconsin was significantly correlated with increases in vote % for Trump in the 2016 Republican primary and the 2016 Presidential vote. This was not the case for the 2012 Presidential contest.

Meaning: The Sequoia AVC Edge machine rigged the 2016 vote but not the 2012 vote.

This finding held true even when the use of AVC Edge voting machines was considered alongside a range of municipality-level demographic and voting method variables.

Meaning: The finding above is robust and is not affected by other variables.

Results suggest that if manipulation occurred, it did not artificially increase the number of voters but instead modified the votes of genuine voters.

Meaning: Votes were changed, not added.

An investigation of the physical voting machines used in Wisconsin in 2016 is needed to determine whether they have been tampered with before or during the 2016 elections to favour Trump.

Meaning: Need to examine the machines themselves.

That seems reasonably clear. The rest of the paper seems to elaborate on those points. It has the flavor of a scientific paper (defining terms, explaining data organization, etc.) which is detailed and careful and tries to be explicit. This makes them verbose.

[wildeyed]

13. Yeah, I get that, but my eyes cross when I try to go back and check his math

and sources.


It would be very easy to say it this way:

In counties with X% of this voting machine, we saw a Y% increase in votes for R prez candidate in 2016 compared to 2012

In counties with X% + 20% MORE of this voting machine, we saw a startling rise of greater than 20% increase for Rs in 2016 compared to 2012, indicating that there MIGHT be a problem with the machines.

At this point, you have enough to go to a legit voter protection group who can do more research and file a complaint or lawsuit. But they don't do this. Why?

One paragraph, a table or graph and links to data and sources could show the correlation. I would also want to see basic demographic info like age, race, education and income levels. Population of area too. I could make the calculations myself IF I KNEW WHICH DISTRICTS. But that is conveniently left out.

The thing is, I caught this same account making wild claims about how a county in NC had OBVIOUSLY been stolen by GOP, but since I am from NC and know how to look up the info easily, I found they were completely misleading their followers. There was 5% swing, D to R presidential votes, when you compared 2012 to 2016.

[dragonlady]

18. On same-day registration...

Wisconsin voters do NOT have to cast a provisional ballot when they register at the polls on election day (unless they have a driver's license or state ID but don't have the number with them to put on the form). The usual situation for a provisional ballot is not being able to provide an acceptable photo ID, which can be cured by bringing it to the polls by closing time or to the municipal clerk's office by the next Friday. In my years of experience in observing elections, provisional ballots are quite rare.

[wildeyed]

20. In NC, all same-day registrations are counted as provisional until residence is verified,

regardless of the ID you present at the polls. It would be straight-up crazy to do it any other way. Wisconsin verifies EDRs with a postcard system. http://elections.wi.gov/node/3881

In my years of experience, provisional ballots are not rare at all. It depends on the demographics and the enthusiasm the election is generating. They are pretty common in poor areas with many transient voters. Provisional/EDR ballots generally favor Democrats, but in NC in 2016, they leaned Republican. Trump turned out a bunch of unlikely white voters here. I bet he did in Wisconsin too.

[dragonlady]

22. This shows how much election systems can vary between states

Wisconsin registration forms are signed under penalty of perjury and that is accepted as proof of residence for voting when accompanied by one of the acceptable documents such as driver's license, bank statement, utility bill, government-issued document, etc. "Provisional ballots are also NOT given when a person is attempting to register in-person at the polling place and does not provide the required proof of residence" (http://elections.wi.gov/clerks/provisional-ballots). A voter who has successfully registered at the polls puts the ballot into the machine and that is that. I have seen it happen hundreds of times.

The postcards you reference are sent out after the election to verify addresses but have no effect on the current election. I think this postcard procedure's main result is to hassle people who move around a lot and may be gone when the postcard arrives (think students and low-income folks).

[wildeyed]

24. Can you show me the link to how EDR are actually counted?

That site is a hot mess when it comes to that topic. http://elections.wi.gov/elections-voting/election-day-registration

What level of ID did Wisconsin require for EDR before strict photo ID laws were introduced? NC Allows Same Day Registration, but only during the early vote period. The tried to do away with it AND impose strict ID laws, but were blocked by the courts under Voting Rights Act. Their intent was obviously to block participation of AA voters, so the courts made them stop. I guess in WI it not considered a Voting Rights case

Super excited about the WI gerrymandering ruling! NC has a similar case wending its way through the courts. If we get a good ruling on those, can probably make many of the laws designed to artificially decrease voter participation for partisan ends go away too.

But back to this Farb guy.... Their account posted a bunch of misleading stats about a small county in NC. The 'suspicious' numbers were easily explained by any state political junkie. Even if you weren't from the state, if you look at past election results, the swing was totally inside the margin of possibility.

In addition, we have many independent election protection groups who go through all election numbers with a fine tooth comb. Part of the reason we do that is to provide information for SUPREME COURT cases on voter disenfranchisement. So why does some anonymous cyborg account on the interwebs think that their data analysis is more valid than groups looking to take the GOP down for voter suppression at the highest level? Why are the casting suspicion on statistical events that are easily explained by anyone with basic knowledge of voting patterns in this state? Why divert attention from the real problem?

The NC GOP made all sorts of noise about 'voter fraud' this last election when the odious Pat McCrory lost the governor's race. Wake county, home of the main campus of University of North Carolina, had a problem entering early vote results, making appear that McCrory was going to win early on election night. When the early vote totals hit (Democrats favor early voting over election day in this state), the outcome appeared to change radically. That is how GOP 'knows' that there was something fishy going on, that Dems cheated Oh, and they claim there was over 100% voter participation, so obvvious fraud!!11!!! But NC is an SDR state and Wake has a huge number of young, transient so it is NOT evidence of anything.

This Farb account makes similar accusations with similar flimsy evidence. This diverts from the very REAL voter suppression that occurs in both WI and NC. They have already decided that the voting machines are at fault and just go back and cherry-pick evidence that corroborates their opinion. We DO need to keep an eye on the voting machines and upgrade election cyber-security. Duh. But if their accusations are legit, they need to get some outside group with experience on the ground involved and stop hiding behind and anonymous Twitter handle.

[dragonlady]

25. I'll try to answer your question

First, registration and voter ID are two completely separate things.

Registration: The voter who needs to register must have proof of residing in the ward where he or she wants to vote. There is a list in the voting law of the only acceptable items. When registering in person the voter must show one of those items (on paper or brought up on phone or computer). Most people can come up with something, but it can become impossible if the voter is appearing near when the polls close and has no time to go home and get it. There is no option to cast a provisional ballot and bring in something later. Of course some people have none of the possible documents because of their life conditions, and one person is too many, but that's where we are now. (Before the Republicans captured the entire Wisconsin government, voters could show anything that the poll worker would accept or bring along someone who would vouch for their residence. No more.)

Since the 2016 election the state has started a new system for those who hold a current and valid WI driver's license or state ID. They can register to vote through the DMV online, and if they have a new address since the ID was issued, they can change the address online without any proof. This is somewhat helpful, but not for those whose license is suspended for various reasons, and there are many in this situation. They have to register the way described above.

Voter ID: Before the Republican takeover, there was no photo or other ID requirement for voting, and no poll book to sign. Walk in, give your name to the poll worker who would check you off on the list of registered voters for the ward, and receive your ballot. Same-day registrants would be entered on a separate list of voters and receive a ballot. As I said in the last post, every voter filled out the ballot and put it into the machine or ballot box (some small towns still use hand-counted paper). There was not, and still is not, a provisional ballot associated with same-day registration. This may seem unbelievable to you, but I assure you it is true. I have seen hundreds of freshly registered voters stick their ballots into the machine, where they become anonymous. The rationale was that no one would risk becoming a felon, by impersonating another person or lying on the registration form, just to provide one more vote to the total. The only real voter fraud was from a few felons who didn't understand being disqualified to vote or were carried away by enthusiasm. (One Republican was found to have voted absentee for two relatives, for himself from two different addresses, and in another state where he owned property. He pled insanity but that didn't fly.) Given the tiny amount of actual fraud compared with the huge numbers of disenfranchised voters, I'd like to see the old system return.

On the gerrymandering case: I too hope for the best. The people who put this case together have done a great job. I'm worried about timing, though. As far as I can see, it is not scheduled for argument yet, and the Court's term is only a few more weeks. Justice Kennedy, the one they crafted the argument around because of his comments in a prior case, is supposed to retire this summer. Unless a replacement can be delayed long enough, this may not work out so well.

[wildeyed]

27. Voter fraud is incredibly rare.

And it usually takes 1000s to influence an election.

In NC, you can register with just a SS # or DL #, but they send a card to validate before the registration is accepted. If the card is returned, they try to follow up and ascertain what the problem was. If you do not include a SS # or DL # on your original form, then you can still register, but you bring a bill or something with the same address where you registered to verify. Otherwise you just show up, get your name checked of the list and vote.

Why would they not hear this case ASAP, considering the importance? We have elections coming soon. The country is in turmoil with widespread civil unrest as a result of their poor rulings on Voting Rights and Citizens United, and now they sit on their hands? Bizarre.

[dragonlady]

30. I think the Supreme Court is pretty much on schedule

Unless there is some very urgent matter, they have procedural steps that do take a while--motions, time for amicus briefs to be filed, etc. The main problem with this is that Kennedy was an important part of the case's strategy. They may still hear it this term--the next argument calendar should be posted quite soon. This case directly affects only the State Assembly districts (called the House in most states) and therefore the State Senate as well, because each senate district is composed of three assembly districts. While candidates don't have to file nomination petitions until late spring, they would like to know what district they live in well before that and start campaigning. I'm sure the justices are smart enough to realize that.

[mythology]

26. The "analysis" is crap

He doesn't present the math, he only says if it's a positive or negative amount, doesn't account for voter ID laws, doesn't seem to have properly analyzed voting patterns in other states to understand the demographic differences between 2012 and 2016.

It also ignores that counties in Wisconsin that were hand recounted didn't differ from machine recounted counties in any unexpected ways. The differences were largely due to incorrectly filled out ballots.

Merely looking at vote total differences isn't actually analysis.

[wildeyed]

28. Do you have insight into why they would be doing this?

It's clearly BS, but it's time consuming, high level BS. Why is this group putting so much effort into the project?

[FakeNoose]

5. It's a list of tweets

I can't find any actual report on this link, it's only a list of tweets.

It looks like he's saying they proved that *some* Wisconsin voting machines were rigged for Trump.
I'm very interested in this because probably something similar happened in PA.
Where is the report published?

Thanks!

[Moostache]

7. Look, people will ALWAYS find a shred of doubt to hold onto their beliefs...

There will be those who find argument or disagreement with ANY statistic or mathematical or legal proof no matter what. In the end, much like a criminal case, the best we can hope for is proof beyond a reasonable doubt that something was amiss and then its up to devoted investigators to dig out the exact details and fill in the grey areas with full color and "who/what/where/when/why/how".

For me, there is already plenty of proof that Trump used Russian money and influence (Twitter bots, online trolls, Facebook lies and manipulations) to help sway the election his way. The central question I need to have answered is WHY?

If the Russians were in this only to extract a measure of revenge on Hillary for dressing down Vladimir Putin and generally being a pain in his ass, then the influence and money would be the end of it, they would have merely been trying to fuck with her and the surprising Trump "win" would merely be collateral damage as far as Russia is concerned...a welcome but unintentional outcome.

HOWEVER, if the Russians were involved in an effort to actually elect Trump, with expected payoffs in oil and gas money and investments and sanctions relief as well, now we are talking about complete subversion of the election and treason from anyone in Trump's campaign with ANY knowledge of these things, at all. If the Russians were involved to this later degree, then outright hacking of voting tabulations in Wisconsin, Michigan and Pennsylvania would seem extraordinarily likely, and in fact those states did prove to be the margin of "victory" for the Orangutan and his cabal.

There absolutely MUST be an independent prosecutor assigned to investigate this and ferret out all of the truth. This person needs to have complete authority to compel testimony and witnesses and documents and should be 100% free of the existing executive branch of the government - no FBI, no political appointees, no congressmen or senators or party affiliated hacks.

This is the darkest hour of the 21st century for American self-governance. We are failing right now and every minute that this matter is NOT under a special prosecutor.

[brooklynite]

21. WHY is simple: they didn't expect Trump to win.

They expected Clinton to win, and wanted her weakened by riling up the Republican base with all the "corrupt politician" stories they could muster, increasing the likelihood that the Republicans in Congress would be pressured not to work with her.

[hedda_foil]

8. He's gotten the name of the machine wrong. It's the Sequoia AVC Edge, Not EVC.

That automatically makes the discovery look less credible. It's a bad mistake, that will allow it be painted as CT even though it has nothing to do with the data.

[sarah FAILIN]

14. Could it be a simple mistake in hearing it wrong?

If the data is true, I hate to discount it over that.

[hedda_foil]

16. The computer scientists who have been in the trenches on this are examining the info.

Crossing fingers.

[sarah FAILIN]

17. +1

[sarah FAILIN]

15. The researcher had it right though

So problem appears to be the middle man that retweeted it.

[Achilleaze]

9. republicans cheated America in order to install Comrade Casino

as their freaking republican Draft-Dodger-in-Chief.

That is so freaking deplorable and disgusting.

republican TRAITORS against America will be brought to justice.

[byronius]

10. Well, of course. The Religion of Racism commands them to do such things.

Plenty of evidence for this. I wish us all luck in setting it right.

[aikoaiko]

19. Indirect evidence will never be enough to overturn an election.

But it should prompt people to make changes and require a paper trail.

[triron]

23. good find!

[aquamarina]

29. This election was a total sham

Between the media bias, Russian meddling, and suspect voting equipment, there was no way for HC to overcome the totality of it. And now look at the shitstorm we in as a result. It is an outrage.

[sarah FAILIN]

31. has anyone heard anything else on this?

Looking at the actual data makes it look interesting.

[LaydeeBug]

32. The guy on Twitter is Mike Farb

Dedicated and tenacious, he is indefatigable

[sarah FAILIN]

33. He is just reporting on this guy though

People were slamming him for using the wrong words and terms I think, but he is not the one gathering the info so he doesn't have the full knowledge of it.

[sarah FAILIN]

34. Seems very relevant now nt

[LaydeeBug]

35. I think so too. nt