Answering the Critics

Why their case is weak

“I Will Set Aside the Claims About Sare’s Missing Votes”

We’ll start to examine a few critics and see why their arguments don’t stand up.

We’ll respond here to Charles Stewart III, an MIT academic who says in his June Substack:

"Anomalies" in Rockland County, NY Voting Patterns Are Not Anomalies

Let’s start with his first statement as he examines the evidence:

“I will set aside the claims about Sare’s allegedly missing Senate votes—the number is small, and such affidavits are inherently unverifiable. Instead, I focus on Harris’s underperformance relative to Gillibrand, as it involves more votes.”

What? You will “set aside” the claims about Sare’s allegedly missing Senate Votes? Sare’s missing votes are one of the two main pillars of the lawsuit. Read the complaint.

There are two main points of evidence in the petition.

1. Independent Senate candidate Diane Sare’s votes were counted incorrectly.

2. The drop-off rate in the county (the difference between the presidential vote and the downballot vote of the same party) is highly irregular.

In a debate, when the moderator asks you a question, if you say, “I’m going to set aside your question” you will lose the debate. Similarly with a lawsuit. If you are going to come out swinging against the lawsuit, “I’m setting aside half of your case, because I don’t feel inclined to discuss it” isn’t an impressive argument.

The Sare votes are critical to the case. In Ramapo District 39, eight voters signed sworn affidavits that they voted for Diane Sare and the Rockland County Board of Elections acknowledged another vote in Early Voting. Sare must have a minimum of 9 votes in Ramapo 39. But the Board gave her 5. That is almost a 50% reduction of her votes.

In Ramapo District 62, four voters signed sworn affidavits that they voted for Diane Sare and the Rockland County Board of Elections acknowledged another Mail-in vote. Sare must have a minimum of 5 votes in Ramapo 62. But the Board gave her 3. That is again, almost a 50% reduction of her votes.

Charles Stewart III is a professor of political science, not a mathematician. However, for him to “set aside” the reduction of almost 50% of a candidate’s votes as not worth discussing is highly problematic, and not scientific. If you extrapolate 50% of the candidate’s vote missing across the entire county it is a large number of votes, not a “small” number as he says.

We also have no way of knowing if it’s just Sare’s votes that have been counted incorrectly, or if this problem impacted other races.

Then he says “such affidavits are inherently unverifiable”.

That also is not true. If we count all the votes by hand as we have requested, we will have a better idea if the affidavits are correct or the Board’s count is correct. (Keeping in mind that we don’t know if the chain of custody is good enough to have confidence in the count of the paper ballots).

The judge seems to agrees with us and not Professor Stewart III, that voters have a constitutional right to have their votes counted accurately, not dismissed as a “small” number. That is why the request for the recount is moving forward.

Superficial Drive-By Critique

We are usually fans of Professor Stewart III’s work. He’s done an excellent analysis showing that voters of color often wait longer in line to vote than others. We quote his work in our legislative efforts to reduce wait times to vote.

However his Substack on Rockland County is really a drive-by critique without much thought. It’s unfortunate that something that seems to be done so lightly might influence public opinion about the case.

He points out that the same type of voting patterns that show up in the predominantly Hasidic communities in the town of Ramapo, exist in a Haredi (orthodox) community in neighboring Orange County, New York as well.

And?

That’s the critique?

The same pattern exists in another county?

Our concern is that the high drop-off patterns exists all over the country. It is more extreme in certain Hasidic districts, but it is also unusually high in the other four non-Hasidic towns in Ramapo.

The fact that there are 2 Haredi (or Hasidic) communities that voted largely for Trump as well as the Democratic Senate candidate does not in any way diminish the rest of the claims regarding Rockland County.

In the statistical analysis we presented in our initial press release, we made sure to demonstrate that the other four towns in Rockland County which have minimal, if any Hasidic presence, also have very unusual 2024 Democratic drop-off rates when compared to the 2020 Democratic drop-off rate. Those towns are:

• Clarkstown

• Haverstraw

• Orangetown

• Stony Point

The irony here is that Charles Stewart III himself participated in a study published in Scientific Data which stated that the typical drop-off rate between the presidential and downballot races in 2020 was 1-2%.

This is a quote from his own study: [In 2020] “Using this data, we show that in battleground states, 1.9 percent of solid Republicans (as defined by their congressional and state legislative voting) in our database split their ticket for Joe Biden, while 1.2 percent of solid Democrats split their ticket for Donald Trump.”

Here are the authors. They are from Harvard, Yale and MIT, among other institutions:

• Shiro Kuriwaki,

• Mason Reece,

• Samuel Baltz,

• Aleksandra Conevska,

• Joseph R. Loffredo,

• Can Mutlu,

• Taran Samarth,

• Kevin E. Acevedo Jetter,

• Zachary Djanogly Garai,

• Kate Murray,

• Shigeo Hirano,

• Jeffrey B. Lewis,

• James M. Snyder Jr. &

• Charles Stewart III

The drop-off rate in Rockland County is +23% for Trump and -9% for Harris. That is very different from the 2020 averages of 1-2% that Stewart III’s own study found to be typical.

p-value of zero

A well-respected statistician we are working with, who has published a text book on statistics, found that when comparing the 2020 Democratic drop-off to the 2024 Democratic drop-off in the non-Hasidic towns of Clarkstown, Haverstraw, Orangetown and Stony Point, the 2024 Democratic election results had a p-value of zero.

p-value is the probability value or in other words the likelihood that a result is happening in a random or natural way. The p-value (or likelihood) that the 2024 Democratic election results are not influenced by some unknown action or force is zero.

So, while Professor Stewart III can show that there is at least one other Haredi (orthodox) community that voted similarly to the predominantly Hasidic town of Ramapo, he has not even tried to address the irregularities in the non-Hasidic communities.

Hasidic Voters Are Not All Bloc Voters

From our latest press release:

Another oddity in the county's election results was not cited in the original petition, but was discovered after the lawsuit was filed. In multiple districts hundreds of voters chose the Democratic candidate Kirsten Gillibrand for Senate, but zero votes were recorded for the Democratic Presidential candidate Kamala Harris. Many scholars have pointed to "bloc" voting in Hasidic Communities as the reason for these Harris zero-vote districts. "Bloc" voting is where voters make candidate selections based on the recommendations of a leader. This is one possible explanation for the zero-vote districts, but it does not rule out other explanations.

Local Rockland County reporting indicates that there are many times when Hasidic communities do not vote as one monolithic block, "How those votes could go … is unpredictable, complicated by the fragmentation of votes among multiple Hasidic and Orthodox voting groups, particularly in Rockland, in spite of the popular misconception of a single voting bloc." Referring to statements from Yossi Gestetner, co-founder of the Orthodox Jewish Public Affairs Council and a Rockland resident, the article says, there are "thousands of other Hasidic and Orthodox voters in Rockland with varying degrees of fidelity to the political endorsements of community leaders."

Pro V&V - A Technology Company in Disarray

The Pro V&V voting machine software and firmware updates are extremely concerning, but Professor Stewart III completely discounts them.

A company whose website was publicly disintegrating from July 2024 through February of 2025 was in charge of the 2024 software and firmware updates to what appears to be most of the voting machines in the entire country, from both of the two major vendors: Dominion and ES&S. Why we would trust this company with our voting machine security, when they could not keep the front page of their website online is a mystery and one that deserves to be thoroughly investigated.

Professor Stewart’s argument here is again painfully thin.

Here is his argument,

“There’s one last issue to address in the Rockland County controversy. Some have linked the vote discrepancies to a claim that a de minimis pre-election update to ES&S voting machines contributed to tabulation errors. Rockland County uses ES&S equipment; Orange County doesn’t. The claim that this minor software patch to voting machines could have facilitated a targeted manipulation of votes in New York (and elsewhere?) is implausible on its face. The fact that Orange County uses a different voting equipment manufacturer should put to rest the idea that these discrepancies were due to software upgrades.

The problem with this argument is that Pro V&V put out numerous software and firmware updates throughout 2024. And they went to both major vendors: ES&S and Dominion. Rockland County uses ES&S, Orange County uses Dominion. We itemized all of the voting machine software and firmware update in our June 10th, 2025 Substack, which was published almost two weeks before Dr. Stewart’s Substack.

Throughout the year, Pro V&V provided updates to most of the ES&S versions in use such as this update:

ESS-1165- an update that went to all of these ES&S versions:

DS200/DS300, EVS 5.2.4.0, EVS 6.0.2.0, EVS 6.0.2.1, EVS 6.0.4.0, EVS 6.0.4.3, EVS 6.0.6.0, EVS 6.1.0.0, EVS 6.1.1.0, EVS 6.2.0.0, EVS 6.3.0.0, and EVS 6.4.0.0

And they provided updates to most of the Dominion versions in use such as this update:

DVS-100974 - an update that went to all of these Dominion systems: Democracy Suite 5.5-B, Democracy Suite 5.5-C, Democracy Suite 5.5-D, and Democracy Suite 5.17

Those are just two updates. Pro V&V provided 21 software and firmware updates to voting systems in 2024. Although the updates may seem minimal, if the company was compromised—or it’s management and security was lax, as the website errors indicate—malware could have been placed on voting systems through any, or many, of those updates.

From our first Substack on the Pro V&V updates:

Software updates are a classic mechanism to deliver malware to computer systems. For example approximately half a million Asus laptops were infected with malware through an official software update in 2018

“Researchers at cybersecurity firm Kaspersky Lab say that ASUS, one of the world’s largest computer makers, was used to unwittingly install a malicious backdoor on thousands of its customers’ computers last year after attackers compromised a server for the company’s live software update tool.”

I respect that Professor Stewart III as a political scientist, but he is not an election security expert, and his dismissal of this very serious threat is inappropriate at best and dangerous at worst.

He calls the Rockland County election results in this case a “nothingburger” The term is more aptly applied to his report.

Please Donate & Subscribe

We are making tremendous progress understanding what happened in the 2024 election, as we lay the groundwork for improved U.S. elections long term. If you find our work valuable, please subscribe and make a donation as well! All donations are tax-deductible. We are raising funds for our lawsuits as well as our program to protect the 2026 Midterms. Thank you for all that you do.

Think for yourself, or others will think for you without thinking of you.

~ Henry David Thoreau

---