Electoral Fraud in Maricopa By Charles Taylor (Florida)
Here is an update on the 2020 US electoral fraud issue, as Dr Shiva uncovers 200,000 ballots with mismatched signatures. It shows how deeply flawed the US electoral system actually is, even apart from deliberate corruption and manipulation.
“In this Discussion, Dr. Shiva shares the results of his groundbreaking study revealing over 200,000 mail ballots with mismatched signatures were counted without being reviewed (“cured”) in Maricopa. This is the first study to calculate signature matching rates to provide a quantitative framework for assessing signature verification of mail ballots.
Key Points:
- First study to calculate Signature Matching Rates and to provide a quantitative framework for assessing Signature Verification of early voting mail ballots (EVBs).
- In Maricopa County, 1,911,918 EVBs were received and counted.
- The County reported no more than 25,000 of these ballots (1.3%) had signature mismatches and required review (“curing”); and of the 25,000, 2.3% in post-curing – 587 – were confirmed signature mismatches.
- A Pilot Study recruited three novices and three experts (forensic document examiners) to calculate signature matching rates on the same sample of 499 EVB envelopes. The purpose of this Pilot Study is to determine if results warrant any further investigation.
- All six reviewers who were presented images of EVB envelopes to evaluate if the signatures on those envelopes matched with genuine signatures on file concurred 60 of the 499 (12%) EVBs as signature mismatches.
- Based on this Study, over 204,430 early EVBs should have been cured vs. the 25,000 that the County actually cured; and, using the County’s 2.3% post-curing rate, 5,277 EVBs should have been disallowed.
- Though this Pilot Study is compelling on its own, an expanded study is warranted.
Rough Transcript (Auto-Generated)
SUMMARY KEYWORDS
signatures, ballot, people, mismatch, cured, maricopa, study, signature verification, match, brandon, envelopes, novices, images, mail ballots, agreed, experts, election, pooled, called, reviewers
Hello, everyone, this is Dr. Shiva Ayyadurai. I hope you’re well.
I’m going to be sharing with you a study that I did a pilot study, which I shared some of it yesterday, but I thought it would be good to redo it again today, because I think sometimes these kinds of things when they’re rushed, it’s not as good. So I wanted to go into a very detailed on the pilot study, revealing that over 200,000 mail ballots in Maricopa, which had mismatched signatures, that’s what revealed in the study, were counted and weren’t fully reviewed. Full Review is also noted as something called curing.
Okay, so that’s what we’re going to go over today. So we’ll wait for people to join. And that’s what we’re going to discuss today.
So again, let me bring up the title of the talk today. As you can see, it’s called irreconcilable differences. Over 200,000, mail ballots, with mismatched signatures counted without being reviewed, which means cured, and you’re going to understand what that terminology means in Maricopa, and this is a first study, to the best of our knowledge, to calculate signature matching rates, to provide a quantitative framework for assessing signature verification of mail ballots, the whole area of mail in ballots obviously has become a very controversial area, it’s split up the country, there’s a lot of division.
But one of the aspects of mail ballots that you’re going to learn shortly, is the aspect of verifying signatures. And we’re going to discuss that. So today we’re going to you’re going to learn about signature verification, we’re going to go over the results that took place in the 2020 election in Maricopa with, how they verified signatures, what results they had, then we’re going to share with you two important experiments that we did, where we got novices, to do the same exercise of having side by side signatures, and then telling us whether the signatures matched or not from the same person.
And similarly, we did the same thing with what we call experts foreign, sorry, forensic document examiners, F. D. S, they did the same thing.
And in the last part of it, I’m going to talk about how we combined both results, contrast them, but also pull them together to come up with the most conservative signature mismatch rate. And we’re going to go over that and that mismatch rate came to 12%. And when you multiply the tour 12% by all the signatures, all the early voting ballots that came in our analysis shows that over 200,000 ballots had signature mismatches.
And they should have gone through this full review process called curing. in Maricopa however, only 250,000 ballots were cured. Okay.
And our intention in doing this is to provide a scientific study so we could educate people. And we hope that the election officials in America will engage in a discourse and discussion. But this study is being delivered to the Arizona Senate as well as the attorney general.
It was commissioned by the Arizona Senate for me to do so let me just jump right into it. And I want to welcome everyone for joining us. So let me go right into it.
By the way, this is Dr. Shiva DeRay. We completed this study on January 14, we presented it to the Arizona Senate liaison on January 16.
We did the final report a few days ago. And we did a public presentation recently. But I thought it’d be better to do it again.
I did it from home at the time, and I thought it’d be better to do it a little more formally because of the importance of this. Okay, so let me go right into this. By the way, notice, it’s a pilot study, a pilot study is typically done, where we have an idea an inkling in science, something’s happening, so we want to explore it.
But we go into it, obviously, with a hypothesis. Okay. So by the way, if you want to know more about me, it’s all in there.
My background is in engineering systems, over 40 years of work, and 40 plus years in pattern analysis. And all of that stuff is in here. If you want to go to VA shiva.
com. If you want to know more about my background, please do that. But this publication is a publication of the elect election systems integrity Institute.
It’s a newly launched into Institute ESI and we’re dedicated to providing independent research and infrastructure For those of you who are interested in doing audits or want to learn, please go check out the institute. And it’s really to support election, election systems integrity. And this publication documents the work completed by echo mail, which is commissioned by the Arizona State Senate to perform the work in this study.
So we have various items we’re going to cover today, we’re going to go through a quick executive summary, a background, the methodology that was used experiment one, experiment two, that combine analysis of novices and experts. We’re going to do that and then we’re gonna have a discussion, and then we’re going to go to conclusions. Okay.
So here is the executive summary. First was this was a first study to calculate signature Miss signature matching rates and to provide a quantitative framework for assessing signature verification of early voting male ballots. EVPs I’m going to use that acronym.
in Maricopa there were 1,911,918 early voting, mail ballots that were received and counted, and the county reported no more than 25 of these ballots 25,000 of these ballots 1.31% at Signature mismatches, and required review and curing to resolve those signatures to see if there were indeed mismatches. Now our pilot study recruited three novices and three experts forensic document examiners to calculate signature matching rates on the same sample of 499 AVB envelopes.
So we took a sample, I will explain that and the purpose of this pilot study is to determine if results weren’t any further investigation. And all six reviewers who were presented images of EVP envelopes evaluate if the signature on those envelopes match with the genuine signatures on file, concurred 60 of those 499. So all six, agreed that they were all mismatches.
And they did it all independently. So that would be a 12%. Mismatch rate.
Maricopa says 1.3%. Now based on the study, over 200,000, early voting ballots have should have been cured, versus at most 25,000, that the county actually cured.
And though this pilot study is compelling on its own, we believe an expanded study is warranted. Okay. So let’s go to the background.
What is signature verification, again, to quickly review this for those of you who are new, and by the way, I didn’t know what a lot of this was until I got involved in this. So we’re on the same boat, I’m just about a year and a half ahead of you. But signature verification is that multi step process aimed to verify signature based on review of two signatures side by side, one being genuine, the other being questionable.
And what happens is early mail, ballot envelopes come into a central facility they’re scanned to create these envelopes images, okay. And those are the early voting ballot envelopes are scanned to produce images. Then what happens in step one, an initial review is done.
It’s about 40 to 30 seconds using the county’s procedures, reviewer compare signature on envelopes with genuine signatures on file to determine match or mismatch. So they say, Oh, does this match this, okay? And by the way that people are trained in this in a very short period of time to do this, it’s novices who do this. And then if it is a no match, it goes to curing.
So all these ballots come in the 1.9 million plus. And the ones that are if the if the people here say they don’t match, it goes into curing, which means more detail review, if they match, the envelope is open, and it’s sent for tabulation when it’s cured.
Curing means it takes more than just four seconds to 30 seconds. By the way, initial reviews only four to 30 seconds, curing is three plus minutes. And the investigation includes attempts to contact voters to determine if the signature is indeed a match, or no match.
Okay. So and after hearing if it’s confirmed as a no match and the ballot not counted. If it’s confirmed as a match, then it’s sent for tabulation.
Okay, so basically the voters got two shots, one at the initial reviewing, and one during query hearing was set up really for fairness. So now in Maricopa County, let’s look at what they actually did. There were 1 million let me just edit this right here.
This is an error. There were 1,000,911. Let’s go back to this.
About that. We bring this back to everyone. There’s a slight error.
I notice I don’t like those errors and we bring it back. So there’s actually 1,911,918 Okay, early voting ballots, as we said mail ballots to the county set a maximum number 25,000 were cured and which is 1.31%.
The Confirm signature matches mismatches where EVPs were not counting According to accounting for 587, out of those 25,000. So if you compare the 587 to 25,000 will be 2.3%.
But if you compare the 587 net, which was not counted, that would be point oh, three 1%, three, one hundreds of 1%. All right. So what did we do? So what we did our team at ESI.
In our research institute, we put forward a methodology, which I’m going to walk you through, and we did two experiments. Let me go right to that. So the methodology we did here is the following.
First, we selected a representative statistical sample from population 1,000,009, or 11,009, or 19. Early voting, mail ballots, when we have those in here, at our echo mail had to be specific. And the confidence level of these were 95%, we want to do we wanted to select a sample size, we could have 95% confidence with a plus or minus 4.
4% margin of error. So out of those, this all these millions of ballots, we selected for 99 randomly. And to give you an idea in statistics, you don’t have to select all of them, you just have to do a random sampling.
This is how pollsters work. So again, this is a pilot study, we could do a much larger study, but even at this level of selection, we’re at 99 95% confidence with a plus or minus 4.4% error.
Okay, so we selected a random sample of 499. That was the first part. And then we organized a data set of 499 envelopes signatures, by randomly sampling that 1.
9 plus million AVB on both images repository we had Okay, so these are for example, the signatures from those envelopes. Alright, step three was we created a data set of the foreigner 99 genuine signatures that match names and addresses of the foreigner 99 envelopes signatures, okay. Now, how do we do that? Well, we source we source from the Maricopa publicly accessible deeds repository, we extracted 499 deed signatures.
Now, it should be noted that the source of the genuine signatures used in this study are likely different from this source of genuine signatures used by the county. They typically use voter registration signatures are from the DMV. However, experts and forensic document examination share that signatures from a deeds repository, which is what we did, may likely be more valid given such signatures are notarized.
Okay, so the county, by the way, if they’re listening wish to provide their genuine signatures, for these foreigner 99 samples, the study can be updated. So we’re willing to do that. Now, the Step Four was we create pairwise data sets of the 499 envelopes signatures, and the 499 genuine signatures here the signature from the deeds repository, and here’s a signature right off the envelope.
And what we asked the reviewers to do was they’re supposed to look at these and decide if this is a match, or no match. That’s simple, okay, and they can only spend maximum of 30 seconds four to 30 seconds and doing this. And that’s typically what’s allowed in the initial review process.
Now, let’s go to experiment one, we use novices here, we use three novices. So the novices were asked, instructed to follow the county’s guide. So they went through it reviewed it.
In fact, I was one of them. Okay, I was one among the three, and the three non FDA selected and instructed to follow the county’s guide. We presented for 99 pairwise signatures, no more than 30 seconds, and everyone was recorded for their match or non match selections.
And each of the non FTS mismatch rate was calculated, as well as what we call the pooled consensus mismatch rate. So what is each FTS mismatch, right? But then we looked at all these three people, we said, how many signatures are they all agree were mismatches. And that’s called pool consensus.
All right. So on that y, x axes are the signatures, because they’re going from signature zero all the way to signature foreigner 99. And you can see over time, how their signature mismatch rates are convergingdown to here.
And you can see we also averaged all them and this average came to 20.1%. Okay, so 20.
1% was the average of the mismatch rates, when you look at all the three, non FTS and novices and this is another way of looking at it. So you notice 86 out of foreigner 13 were no match for this f a non FD and so on. These two guys are very close, but the total signature mismatch rate by the novices average is 20.
1%. Remember, the counties was only 1.3%.
Okay, so just think about the counties was only 1.3%. All right, then We did a pool consensus.
This means how many times in all three non FTS novices for a pair of signatures concluded was a match, a no match or did not have agreement. And that’s what this looks like. So you can see the saffron, here’s non match the blue is how many times they matched.
And the grades when, you know one guy match, but to others did, okay? Because there are three people here. And you can notice here 63 out of 499 were when all three people, all three non FTS, agreed that those signatures on those ballots were mismatches. So that would mean if you use this number, that the non FD pool consensus signature mismatch rate is 12.
6%. Again, that’s close to almost 10 times higher than the Maricopa County signature mismatch rate. Okay.
So again, these are the novices. So in conclusion, from the first experiment, what we see is nearly 240,900 to AVB should have been cured, okay. 240,902.
So remember, this was the original the county had 1.3% 25,000. If we use the average mismatch rate for the novices, this study reveals that 384,295 should have been cured.
But if you take even the more conservative one where we only assume the pool of consensus, that would be 240,902. Okay. The pooled consensus mismatch, right? Okay.
Now, let’s go to experiment two. So let me just jump in here. So the first experiment, first experiment, is the experiment where we did novices, non forensic document examiners, now we’re going to look at how do the experts do it? And these are experts who are hired by courts to look at signatures to determine if there are they do forensics on them.
So we hired three of those. And they held this. Okay, so let’s go here.
So this is experiment two, we brought in experts. Again, the experts were given three of them for 299 signature, pairwise images, they had to say was a match or no match. Again, we calculated their mismatch rates and their average mismatch rates among all three of them.
And then we also looked at their pooled consensus mismatch rates. So again, x axis number of signatures going from left to right, they’re going and they’re processing them. And you can see that these two FTEs conversion, this FTEs, a little bit off, but meaning away from the three, but they’re far higher than the novices.
So they were when you average them at 60.3%. And then, when you again, here the numbers that you can see, they’re pretty close one of them, the first and the third are above 60.
The middle one is that, you know, 40 plus. But regardless, among the forensic document examiners, they said that 60.3% of the signatures mismatched, and so that in America said only 1.
3%. Obviously, the forensic document examiners have a much higher standard than the than the criteria that you know, the non FTS, were asked to follow. Again, just remind you now we did the pooled consensus, which means we looked at the FTEs and said, How many of the signature pairs that they both conclude were match all three? Are they all concluded was no match? Or were they all have disagreements? And this what you say? So you can see obviously, they had far more mismatches 195 out of nearly the foreigner 99 They said we’re mismatches.
They had 97 only where they matched Okay, and the rest they disagreed. So that but regardless, what you would find is that this means the pool consensus signature mismatch rate is 39.1%.
Again, far less than the percentage of Maricopa signature mismatch rates again, this is the pool consensus of the three FTEs. So if you use that pool consensus number, you find out that 747,560 EDB should have been cured. Okay.
And what you see here is, again, comparing it to Maricopa 1.31% If you look at the high rate, if you look at the average rate 60.3% or 1,152,887 ballots should have been cured if you use the average, but if you use a more conservative number, a 39.
1% where they all agreed on common signature mismatches. That means 747,560 should have been cured. Okay.
So, there you go, you have two very different numbers. So if you look at the novices, they’re seeing around two and 40,000. If you’re looking at the experts minimum they’re seeing around 740,000.
But this is far more than the 25,000, which Maricopa Curie. So that’s the main takeaway from this part so far. Now, what we’re going to do, is we’re going to now do an exercise where we look at bringing together the novices and the experts.
Okay, and what would we get there? So, let’s go there. So combine analysis of novices and experts, novices meaning the non FDS experts, meaning the forensic document examiners. Okay.
So again, this is all six. So the bottom here, right, here are the three novices. So where our mismatch rates are very low, the experts are here, they’re much higher, as you can see.
And then we did the average of all the experts, which is the saffron nerd 60.3%, the average of the non experts or the novice is 20.1.
But then if you average everyone together, that’s 40.2%. Which means the novices and the experts are saying, if you pull them all together, that 40.
2%, you can look at each individual FTS and the average we have here. But again, 40.2% is also still higher than the Maricopa signature mismatch rate.
My intention in doing this was to look at this problem in different ways. One is just the novices but novices average and offices pooled, then just the experts, their average, and their pooled, and then put them all together. And that’s what we have here, the average of everyone put together, you can see is 40.
2%. Still far higher than the 1.3%.
The next thing that I wanted to do was to say, if I brought all of the experts and novices together, and we looked at which ones they how many of the ballots, signatures did they have commonality of agreement, okay. And that led to this bar chart. So the x axis here is how many times how many reviewers all agreed to a number of EVPs being mismatched common mismatches so may explain.
So this means none of the reviewers agreed that 97 of the ballots were mismatches, which means all of these were matches, which means all agreed these signatures match. And so that was six agreeing all 97 signatures matches. And just to show you that when they do match, you’re the examples of some of those signature matches.
So again, all six people, novices and experts agree this signature in this signature match. And if you know you don’t, they look pretty similar. And again, here’s another one, three knobs and three experts agree that they both match.
Three novices and three experts agree these two match. Same here that three novices experts these to match and also with this fifth example here, so those are examples of where novices and experts have both agreed these signatures match. And there were 97 of those, okay.
But we’re interested in the mismatch, right? Okay. So now we want to look at this analysis here, which is it right bar here? We’re where all six reviewers, how many times have they agreed to valid signatures together? pairwise did not match and those were 60 signatures. So that’s what the 60 is where all free non FTEs and FTEs agreed that they did not match okay? So signature mismatches, okay? All free novice and free expert.
So what do we have here? We have the examples of this, okay, we’re three novice experts agree they do not match. So this and this do not match. Okay? This, and this, do not match all three novice and expert degreed.
And you can see they’re pretty clear they don’t match. Okay, same here. Very different slant, very different writing.
And same here. So those are examples of where now we have 60 examples of where the novices and the experts agree that out of the 499, they don’t match. So what does so that’s called the pooled consensus, signature matching rate among the common mismatch of knowledge and expert.
So what is that? Well, if you look at this, it’s this. Again, the saffron sliver here, 12.0%.
So 60 of the 499. Some of the examples we just shared with you, or both parties agree they don’t match. Okay? So that’s 12% pulled signature mismatch rate.
And look at that, that’s still much higher than the 1.3% Alright, but if you use a 12% number, you find out That 229,433 board and 30 EVPs should have been cured. Okay.
So again, comparing to Maricopa just 25,000 to 1.31%, if you use the pool mismatch rate of 12%, which would be the most conservative, you get 229,430. If you use the little more nonconservative, if you look at the average mismatch rate among all six, you would still get 768,591.
Okay? So that’s what you would get with that number. Okay? It’s a it’s a, it’s significantly higher. Okay, when you pull everyone together, alright.
And so let’s go here. All right, let’s have a discussion now. So, in conclusion, as we’re wrapping this up, the intention of this pilot study was exploratory.
We’re not here accusing anyone, we literally wanted to recreate how signature verification is going on. And by the way, the study has not been done. It’s quite amazing.
That has not been done. We’re the first to do this study at the election systems integrity Institute, which has published the work that echo mal did. But this these are the conclusions, what we find is that of all of these early, early voting mail in ballots, again, 1.
3% 25,000 What was cured. And out of all of these only point oh, three 1% were confirmed to be mismatches post hearing that was only 587 signatures. So just to let you know, only 587 signatures out of that 1.
9 1 million are finally confirmed the mismatch own and 25,000 initially, before we went through curing, what’s considered to be mismatches. So what we wanted to do was now look at different scenarios, because we presented a lot of scenarios here. So, if you look at this, this is all the possible scenarios.
So the first scenario is imagine if we use the novices average rate mismatch, right, that would have been 21 20.1%. And this says 384,295 should have been cured.
And if you apply the 2.3% of this, that means 1839 1839 ballots should not have been allowed after curing. If you use the non FDA novices pool consensus would be 12.
6%. And that would come down to 240,902. In which case 5541 ballots should have been disallowed.
If you look at the average of the experts, that a much higher mismatch rate 60.3%, in that case, 1.1 million 52,884 should have been cured, which means 26,516 would have been disallowed.
If you used the pool consensus rate, which means where all the FTE isagreed to the same common set of signatures not matching, that would have been seven and 47,560, which would have meant 17,194 ballots should have been disallowed. Now, the last piece is we’re averaging we’re combining novices and experts if you can, if you use the average rate, that would have been 40.2%.
As we mentioned, that would have been 768,591 valid should have been cured, which means 17,678 should have been disallowed. If we use the most conservative number here, which is all three people, I mean, all six people agreeing that a certain set of signatures did not match, which was 60 would have been 12%, which would still have been 229,000 out of 229,430 ballots should have been cured, which meant 5277 ballots should have been not allowed about 10 times more than what we’re allowed. Okay.
Now, obviously, this was a very, very close race. We’re not saying that this would have affected the outcome. But the point is that the signature matching rates that we’re discovering are 10x Higher mismatching rates than what the county had.
Okay. And then final conclusion here. So it really is this takeaway, if you take the most conservative people can say I don’t agree with this.
I don’t agree with this. I don’t agree with this. I don’t agree with this.
We’re giving people literally, you know, six different choices. So if you take even the most one give the county the benefit of the doubt, it would be 12%. Okay.
So at the net of it as a result show minimum 229,500 for the 430 plus or minus 4% murmur that’s our margin of error should have been cured. Now if we subtract the 25,000 County did cure, you would get down to 204 1430 plus or minus 4%. That’s why we said more Then over 200 200,000.
Alright. So that’s what we have. Now the final conclusions of this study.
And by the way, I encourage everyone to go review this video. I know I’ve covered a lot here. But this video aims to be educational.
But it is a detailed study. Okay. That goes into the detailed analysis of this, okay.
And the reason we did this is that it’s important to recognize that the election integrity movement really needs to take a scientific systems approach. Unfortunately, there’s been a lot of Grifters, who jumped into this movement. And they’ve been taking advantage of people.
So we started the institute, because those in power have created institutes at Harvard, and MIT at Stanford. And these are really, really very smart academics who believe that there is no problem, okay? That there are no problems. Okay.
So what we are sharing here is we took a very, very detailed systems approach. And we’re not here to hype anything, we want to invite people to have a discourse, okay. And so that’s why we did this.
And we want to encourage everyone who’s interested in that approach to study system science, we offer a course, on system science, you can go to vashiva.com/join. But I encourage everyone to take a systems approach.
We live in a world where there’s the left approach or the right approach, or the approach or the anti-approach. But ultimately, what a systems approach reveals is that you need to apply the scientific method we go into like we did with this study. It’s a pilot study, okay.
We weren’t aware what was going to happen. We did the analysis, and we may do an extended study. So let me finish up with the conclusions.
This is the first study to quantify the signature mismatching rates during signature verification of EVPs. Again in Maricopa over 1.9 million voting ballots were received and counted.
The county reported no more than 25,000 of these ballots at Signature mismatches and required carrying a pilot study, as we said, recruited three novices and free experts to calculate the signature matching rates. Okay. On the sample of the 499 EVPs, all six reviewers were presented the AVB envelopes evaluate, and we found out was all six reviewers in this pilot study concurred 60 of the Forerunner 99 12% as signature mismatches.
Now based on the study over 200,000, early voting ballots should have been cured at most, versus at most 25,000. County did now. We want to be humble about this while the study is compelling, and expanded studies warranted confirm the findings of this study.
Okay. And that’s really the scientific approach. Again, the title of the studies, irreconcilable differences, over 200,000 Male ballots with mismatch signatures counted without being reviewed, cured America, first study to calculate signature matching rates to provide a quantitative framework for assessing signature verification of mail ballots.
Alright, I hope that helps people. I hope this was educational. Hey, John, I’m about to run out of my battery.
Can you plug me in? Thanks. Thank you, John. So let’s look at some of the comments.
Someone says Hi, Dr. Shiva, can I get in touch with you regarding antiviral removing heavy metals? Sure, you can send me an email at
Someone else sends another question that says I think the paper ballots should have been watermark, an official page. We’ve seen all the fraud. Yeah.
So look, the real issue here is that we did the valid image analysis and we showed at ESI. If we’d done that way before we could have done this audit for pennies. Okay.
The Grifters who did not want to give us a ballot images, which is really unfortunate. I’m not sure why they did that. But the only conclusion you can reach is they wanted to do this paper ballot for 910 months when if they use a ballot image, they could have discovered this stuff very quickly.
All right. What we discovered here is that you have to go way upstream in the process, even as the ballots are coming in and the signature verification has some weaknesses as we’re showing here. Okay.
All right. I hope this is valuable. We will be putting this up shortly on vashiva.com I wish you well be well. Thank you.”
Comments