Claims of vote flipping in Republican primaries.

[davidm]

Here’s someone else making similar claims.
http://www.ukprogressive.co.uk/breaking-retired-nsa-analyst-proves-gop-is-stealing-elections/article20598.html

A retired NSA analyst has spent several sleepless nights applying a simple formula to past election results across Arizona. His results showed across-the-board systemic election fraud on a coordinated and massive scale. But the analysis indicated that this only happens in larger precincts because anomalies in small precincts can be more easily detected.
…
When Duniho applied a mathematical model to actual voting results in the largest voting precincts, he saw that only the large precincts suddenly trended towards Mitt Romney in the Arizona primary – and indeed all Republicans in every election since 2008 – by a factor of 8%-10%. The Republican candidate in every race saw an 8-10%. gain in his totals whilst the Democrat lost 8-10%. This is a swing of up to 20 point, enough to win an election unless a candidate was losing very badly.

I don’t want to believe this. It sounds crazy; but it’s not impossible and it fits in perfectly with attempts to trim the Democratic vote by a few percentage points with things like voter purges and ID laws. It has to be close for something like this to work believably.

Help me out here. Convince me this can’t be happening on this scale.

[BrainGlutton]

@davidm

I’m not trying to dismiss this out of hand. It’s extremely troubling. I work in IT and I DO NOT think that we should be voting in this manner.

What system would you recommend?

[davidm]

Hand counted paper ballots.

[bup]

Well, the first step might be to try to reproduce Choquette and Johnson’s results, as they invite the reader to do at the end of their pdf.

They show their data sources.

Another step would be to try to get Nate Silver interested.

[davidm]

@bup

Well, the first step might be to try to reproduce Choquette and Johnson’s results, as they invite the reader to do at the end of their pdf.

They show their data sources.

Another step would be to try to get Nate Silver interested.

I was considering trying to do that. Perhaps for my own state. I’m not sure if it’s within my capabilities. Maybe I’ll look at it this weekend.

[Chronos]

OK, if you don’t like the alternate explanations posted so far, here’s another possibility: Romney concentrated his ad spending in places where it’d do more good. Even if this paper is completely correct, all it shows is that there’s a strong correlation between precinct size and Romney’s vote share. It does not show any evidence of wrongdoing, unless you can first explain why wrongdoing would follow this pattern, and second show that it fits the pattern much better than any alternate explanation.

[davidm]

It’s been explained in this thread why wrongdoing would follow this pattern. Flipping votes is more dangerous in smaller precincts because it’s more likely to be noticed.

As far as concentrating resources in larger precincts, wouldn’t the other candidates do that also?

[bup]

Here’s another amateur analysis of primary vote behavior, from a Ron Paul site, that not only looks at the primaries this year, but also 2008, where it’s claimed Romney also benefited from larger precincts, and from other elections, where no trend is seen (except consistency) ordering precincts from smallest to largest.Summary of Vote Flipping Analysis

Very similar style of writing and graphs and data source listing.

[Chronos]

As far as concentrating resources in larger precincts, wouldn’t the other candidates do that also?

Presumably, but Romney had more resources. And then again, the other candidates were also massively incompetent, as evidenced by the Virginia primary where none of them save Ron Paul even qualified to be on the ballot.

[bup]

@davidm

So, what’s the flaw in the analysis? Why would Romney’s percentage increase with the size of the precincts, except for those with paper ballots? Is it even true that that’s the case? Is this just a case of being selective about the data?

One thing that just occurred to me, on my drive to work -

Could the likelihood of having paper ballots versus machine voting also be highly correlated with rural versus urban precincts?

I don’t know if it’s true, but if it is, it could explain a lot of the patterns claimed - choosing Romney, larger precinct, and machine voting would all be a result of the urban-ness of the precinct.

Still, the *directness *of the correlation they claim is troubling.

[bup]

Here it is again, in Arizona.The analyst this time is a ‘retired NSA analyst.’ The NSA, of course, is decent math/stats cred right there, but otherwise, I can’t find much out about the guy - Ph.D? Masters? BS? What major?

[Enkel]

In 2006, a group of computer hackers reprogrammed a Dutch electronic voting machine to play chess. The Dutch government subsequently imposed a moratorium on the use of electronic voting machines. In 2009, German Federal Constitutional Court ruled the country’s electronic voting systems unreliable to the point that their use was deemed unconstitutional. In June, Ireland paid a metal recycling company €70,267 to dismantle and recycle the country’s €55 million worth of electronic voting machines.

In the United States, as many as 35% of this year’s general election votes will be cast on electronic voting machines that, “in terms of effort required to compromise the systems, are in the same ballpark as the Irish voting machines,” according to Dan Wallach, a professor in Rice University’s Department of Computer Science and a computer systems security specialist. “The current voting machines were designed with insufficient attention to computer security.”

[…]

Still, there may be reason for concern about the systems in general. In 2007, Ohio convened a team of 23 computer scientists, security specialists and other election equipment experts to review the state’s electronic voting systems. According to their report, “Evaluation and Validation of Election-Related Equipment, Standards and Testing” (EVEREST), “The systems uniformly failed to adequately address important threats against election data and processes. Central among these is a failure to adequately defend an election from insiders, to prevent virally infected software from compromising entire precincts and counties, and to ensure cast votes are appropriately protected and accurately counted.”

Could Romney-Linked Electronic Voting Machines Jeopardize Ohio's Vote Accuracy?

The Ohio report is linked above.