Algorithms in the news

I am very busy today, so I don’t want to say too much, but have had so many questions recently that a clarification is in order.

Today, the Gateway Pundit published a story about the NY voter roll algorithms. The article mentions me, my paper on the subject, and a reference to a recent substack post. Most of what they wrote is correct, but in a couple of places, I’d like to add clarity to avoid misunderstandings.

To be clear: the algorithms are very complicated. They are difficult to describe, and any writer who tries will have to cut out a lot of detail to finish their article in a reasonable word count. As far as I can tell, everyone who has written about the algorithms has compressed the message, from articles that came out a year ago, to today’s article. This is simply the nature of that type of writing.

Clarification #1: the SubStack article does not say that the NYSBOE changed “all” of their voter ID numbers. Rather, five counties changed their county ID numbers. This remains an interesting development, but the way it is written overstates the finding by implying all of the state ID numbers were changed.

Clarification #2: I have looked at about 7 states, not the 13 mentioned in the article. Out of that group, I am confident that algorithms exist in the voter rolls of many of these states based on scatterplots, but not in all. That doesn’t mean they aren’t there, but I haven’t seen the evidence for it yet. I am comfortable saying there are 4 states using unusual algorithms in their rolls, and one that has a suspicious tagging mechanism built into their rolls. The number 13 may be derived from a combination of what I’ve looked at and reports from other people.

I agree with the article’s assertion that, “While it is premature to assert that the algorithmic alteration of SBOE voter registration databases has been used to conduct election fraud, coding alteration of SBOE voter registration databases is inherently suspect of malevolence that needs to be investigated further before the presidential election on Tuesday, November 5, 2024.”

I also believe it is plausible that, as the article states, “The advantage of SBOE algorithmic election fraud is that a small group of National Security Agency professional cryptographers could conceivably carry off an election steal that would be difficult, if not impossible, to detect by honest SBOE officials as the election was being conducted.”

This point has been raised before and has led to questions sent to me by email. Those questions show that a little more information is needed to understand the idea. I am not saying, and do not think it is plausible, that a voter roll algorithm of the type I found in New York is sufficient on its own to commit voter fraud.

As I explained to one person, the voter rolls contain no information on who any person votes for. Therefore, it cannot control votes for a specific candidate. Also, apart from recording the voting method (early, by mail, in person, etc) it doesn’t have information on ballots either. On top of that, there is no way to link a voted ballot to any ID number because they are designed to be anonymous. Therefore, the algorithm cannot be used to control anything related to voted ballots.

I look at it this way: if someone wanted to commit voter fraud on a large scale, either by using fraudulent physical ballots or a fraudulent digital count, they would need a way to reconcile the fraudulently counted ballots with the number of voters who voted. The voter roll algorithm is designed in such a way that a person could covertly track false records. This would make it possible to safely alter them to reflect fraudulently inflated vote totals. It would also be nearly impossible to detect, as stated in the article.

This does not mean I have seen evidence that the algorithm is used this way. Rather, it has that functionality built-in. New York does have a large number of what appear to be fraudulent registrations. This would create a need to track them covertly if they are to be used as a group. As far as I can tell, the NY algorithm does only two things. 1) it maps CID to SBOEID numbers, and 2) it hides its presence extremely well. The first function is what can be used to interact with records of interest, and the second would make it very difficult for anyone else to identify which records were affected.

Clarification #3: The article states, “Paquette insists that the algorithmic alteration of SBOE voter registration rolls would permit rogue SBOE officials [to] steal elections either as voting were occuring, or after a pause in vote tabulation to permit the request, printing, and voting of “clone” or “ghost” fraudulent SBOE IDs that would be counted because they would be certifiable, even if fraudulently so.”

I don’t recall ever saying this and wasn’t asked a question about this by anyone at the Gateway Pundit. It does sound like a digest-sized version of something I believe is true, mixed with a theory I am undecided on.

The algorithm can be used to help certify elections by making it possible to reconcile votes counted with the digital record of voters who voted. Whether this happens at all or is managed by rogue SBOE officials is a question that I haven’t been able to answer with available data.

My global impression is that it is literally possible to use the algorithms as votes are counted, but want more data to say anything stronger than that. For all we know, foreign hackers or some other non-official entity is involved. Frankly, I find that more plausible than local or state officials. The officials I’ve met, for the most part, seem sincere and honest. The few that react in a defensive manner, leaving the impression they are insincere or dishonest, come across as genuinely in the dark regarding the algorithms.

As for when the algorithm would be used, I have heard other people talk about this. I don’t know enough about that part of the process to comment, which is why I’m content to let other people talk about it. Jeff O’Donnell has done some excellent research on this subject that appears to show algorithmic control of elections in real time. Not, by the way, the algorithm I found in the voter rolls, but a different one for a different purpose.

I think the TGP article conflates my research with extrapolations based on my research. That said, any real time covert access to records of interest would allow what the TGP article is alleging, so I don’t totally disagree, I just didn’t “insist” this was happening, nor do I recall saying anything like it publicly or privately. At most, I’ve agreed with someone else that it is possible.

Normally, I don’t like issuing corrections. This is because once an article is published, there is no controlling where it goes after that. In this case, I’ve spent enough time answering questions about this and related questions from Dr. Jerome Corsi’s recent War Room appearance, that I decided to write this out so I can point people to it in the future. The questions in both cases are very similar, so I’m only responding to the TGP article.

Bottom line is that they’re pretty much correct, but some missing details have led to misunderstandings. The most important issue is that the voter roll algorithm, if it is used nefariously, is one of several necessary components to any election fraud scheme. It is important, but would require help to do everything suggested by some of my correspondents.

It’s like saying the steering wheel of a car can be used to evacuate bank robbers from the scene of a crime. It is important, but the rest of the getaway car must be there as well, not to mention a driver.