« What Is Bill Barr Going to Do? by Larry C Johnson | Main | Fox News–A 21st Century MEDIA Titanic? by Larry C Johnson »

12 November 2020


Feed You can follow this conversation by subscribing to the comment feed for this post.


Will wonders never cease. It seems Arizone never certified their election machines according to Arizona law.
Title 16 - Elections and Electors § 16-449 Required test of equipment and programs; notice; procedures manual The test shall be observed by at least two election inspectors, who shall not be of the same political party, and shall be open to representatives of the political parties, candidates, the press and the public.

Neither a Libertarian Party agency Representative NOR the Republican GOP Chair Rae Chornenky was preset at this REQUIRED test and by statute there SHALL be at least 2 official observers from differing parties present to certify the machines..."

Of course this came from the Dark Lord or the Evil one or some such place, so Trust, but unlike with AP stories, verify.

"... the Legislature that crafted the election laws, including the security features. And it was never thought necessary to require the kind of outside audit that she now wants."
Audits, why on earth would a state every think something should be audited......

Eric Newhill

Barbara Ann,
I'm spending more time on this than I should, but this comment should be definitive. I went back and watched the Dr. Shiva presentation again. He is definitely talking some smack that doesn't add up at certain points and now I see more clearly what the critics are talking about, though they are full of it too. really, this is just more of what I've been saying, but hopefully more clearly and this can be put to rest.

It is indeed a mathematical reality (maybe artifact, if one prefers) that as the % straight R ballots increases, unless the % individual ballots going for Trump increases proportionally, you will get a downward sloping line. As I mentioned, my own modeling shows that. That is because the equation being used for the Y-axis points is % individual ballots going to Trump - % straight R ballots. As straight R % gets bigger then the Y-axis points get more negative if their % stays the same or gets smaller. That is what the critics are reacting to.

To illustrate: If you have 40% of ballots straight R and 25% of individual ballots going to Trump, then, using Dr Shiva's methodology, you will get a Y-axis data point of 25% - 40% = -15%. If we then move along the X-axis to a district with 60% straight R ballots, we would need 45% of the individual ballots to go to Trump to get a Y-axis point of -15%. If we get less than 45% we get a downward slope from the first Y-axis point (15%) to this new one. And so on and so forth. To be clear, if we only had 30% of individual ballots going to Trump in this second district, then 30% - 60% = -30%, which is less than the -15% we got for the first district and we have a downward sloping line.

However, Dr. Shiva does mention that he recognizes that mathematical fact; sort of. Near the beginning of the presentation he quickly makes, then glides right past, the assumption that as a district becomes more "republican" (based on % straight R ballots) the % of individual ballots going to Trump should increase. Later he frequently returns to that assumption. Actually, it is the crux of his argument. The critics are missing that, on purpose?

Is it safe to assume, as Dr. Shiva does, that as % straight republican ballots increases that the % of individual ballots going to Trump should at least increase proportionally (and thus not forming a downward line)? I don't think so. I have already said that. It is suspicious that they don't, but I'd hardly call it definitive proof that fraud occurred. Should we expect that % straight R ballots increases we would see % individual ballots going for Trump being all over the board and thus forming a zig zagging pattern? I have no idea. What is the theory of human psychology behind that data?

Are the critics correct that the downward slope is a mere mathematical artifact? Not in the least. It is only an artifact - though that is not really the correct term - if % individual ballots going to Trump does not increase proportionally, or at a greater rate than, % straight R ballots.

When you're analyzing data to understand reality, you can never lose sight of the real world by getting lost in statistical minutia. Nor can you get lost in fantasy explanations about the real world without some data or other hard evidence to support your explanations. I see Dr. Shiva and his critics committing both of these sins.


Very good research by Shiva. Flipping votes in the most hardcore Republican precincts is the best way to hide it.

The issue with Georgia is that they are doing a recount not an audit. Quite why a Republican run state participated in the vote freezing makes me suspicious of the SoS and Governor there. A recount, without an audit, in a county where there weren't issues with votes being shifted supervised by a Republican SoS is how I would attempt to stymie the fraud claims.



If you recount the same illegal ballots that you counted the first time the numbers should be equal, but it is still fraud.


Observer residing in Switzerland tracking voting reports in Swiss real time during the wee small hours in America after Election Day, reports on the 3:42 AM (Zulu) instantaneous dump of over 100K votes in Wisconsin at that witching hour.

From his own data analysis perspective, he observed this one second suddenly changed the leaning red Trump status of Wisconsin to leaning Biden blue. In one second. He concluded this abrupt switch in outcome was bull feathers. https://www.americanthinker.com/blog/2020/11/bidens_great_leap_forward_a_split_second_in_wisconsin.html

Eric Newhill

No. Shiva's research is only good *if* it is true that as "republicanism" increases (as defined by % straight R party votes in district), % individual candidate ballots going to Trump increases proportionally (or greater than proportionally). If % individual ballots going to Trump remains constant or decreases as republicanism increases, then you get the downward sloping line that Shiva sees. That is a mathematical reality. No two ways about it.

Random %s of individual candidate ballots going to Trump as republicanism increases would result in a zig zagging line.

So is it true that as % straight R party ballots increases, people using individual candidate ballots will vote for Trump in a proportional increase? Who knows? It's a mere assertion by Shiva and he offers nothing to support it. I call that shady research.

The idea is that there was an algorithm in the counting machines to switch Trump votes to Biden votes. Therefore a hand count would avoid that problem and get at the true results.

Barbara Ann

Eric, All

Given the gravity of Shiva's allegations I too have spent a lot of time on this, this should be my last word too:

I have repeated the work of Joe Bak-Coleman (Princeton PhD, though he doesn't use the title "Dr") shown in the link to his Twitter account I posted above - and here to save scrolling up to find it. To reiterate; this uses the publicly available data for Kent Co. MI, one of the counties Shiva uses to illustrate the supposed algorithmic fraud.

My scatter plots look identical to Bak-Coleman's, the important point being the Biden excess vs. Dem SPV plot has the same downward slope as the Trump excess vs. Rep SPV plot. The Trump one looks similar to Shiva's chart (from 37:12 in the video) but I note Shiva's plot has a steeper slope. The downward trend is evidenced in both.

Anyhow, this is enough for me, as Shiva's key argument is that the down slope in the Trump plot is evidence of votes being switched from Trump to Biden. As Bak-Coleman says, by this logic the down slope in the Biden plot is evidence of the imagined algorithm doing the converse and moving Biden votes to Trump in heavily Dem precincts. This stretches credulity to breaking point.

The fact that Shiva has not conducted such an obvious control (plotting Biden excess vs. Dem SPV) and the fact that this shows exactly the same 'suspicious' trend blows Shiva's credibility out of the water for me. I do not think it likely this is accidental and suspect him of opportunistic pseudo-scientific scaremongering for his own gain. If so, this is truly despicable. I would not touch this man's 'science' with a barge pole.

Eric Newhill

Barbara Ann,
Exactly. An MIT PhD X3 knows what is necessary to pass peer review, convince a court, etc. Why even no calculation of Rsquared? Why no comparison to previous elections? Why no literature review of the voting habits of non-straight party voters? And, as you note, why no comparison to straight D v individual Biden ballots in 2020?

I first encountered Dr Shiva when he was peddling some software smart gizmos that were supposed to lower healthcare costs. He either had no idea what what he was talking about or was trying to con people.

No integrity in that man.

That said, as you surely know by now, the results are not an "artifact" of the choice of the y-axis.


Barbara Ann,

"Shiva's plot has a steeper slope."

Which means just what? Your software isn't stealing votes as fast, or that the more Republican a district the more people voted against him individually?

Eric Newhill

Barbara Ann and all,
Dr Shiva's shady presentation and shaky conclusion should not discredit the apparent fact that one voting machine developed a glitch that transferred 3,000 votes from Trump to Biden in a district in MI. It appears that all involved admit that happened. That alone should merit a carefully monitored hand recount of *all* MI ballots, IMO. Software doesn't just glitch. It has to be programmed to glitch.


destroying Shiva's theory



When they recount Georgia, they need to look hard at DeKalb County and double check all signatures.

My family left DeKalb County Georgia about 2 years ago. DeKalb keeps everyone on its jury list (and I'm guessing voter registration list) forever. You are never removed - never.

Both my wife and I received a ridiculous number of emails asking us to vote in Georgia, including one claiming to be from DeKalb County government. They offered to help us register and vote, if we but asked. I suspect this is because DeKalb never removed us from the jury roles.

We now live in New Hampshire and voted in our new home. But I do wonder if (unbeknownst to us) we also "voted" in Georgia for Biden. I'd be very interested in checking if I had the power to do so.


Kurt -

You can check if a ballot was submitted in your name in Georgia by going to www.vote.org then click the Track your Ballot button. Or call the Georgia Hotline: 888-730-5816



That is correct on Ghoti = Fish

Barbara Ann,

I only have access to county level as well. I would love to know how to access the data set used by Dr. Shiva.


The comments to this entry are closed.

My Photo

February 2021

Sun Mon Tue Wed Thu Fri Sat
  1 2 3 4 5 6
7 8 9 10 11 12 13
14 15 16 17 18 19 20
21 22 23 24 25 26 27
Blog powered by Typepad