Regions
NewsOpinionsAnalysisServicesTrainingsAbout usRu
AnalysisFalsificationsKursk OblastOrenburg OblastSt. Petersburg23 September 2024, 10:56
Ivan Shukshin
Programmer, election researcher. Krasnodar region
Collage: Kseniia Telmanova

A lot can be said about the local elections that took place on September 8. Official data showed that in Kursk and Orenburg regions falsifications were used to avoid a second round of voting. In Saint Petersburg, the authorities didn’t dare to give the incumbent governor Alexandr Beglov more than 60% of the vote (knowing the citizens wouldn’t believe it). As a result, they were forced to falsify the results while also restraining the extent of the falsifications, and all over the country the voting of employees of public institutions was monitored and managed by a system called "Geo-SMS".  Now, let me explain it more in detail.

It seems that the Russian reality drives us all a little crazy with time.  We say “single voting day” although we know, that in most places it is actually three days. Elections in the regions were held on one day, two days, three days - differently everywhere, but three days were the most common. There were 4,129 election campaigns of various levels.

Voter turnout control

Some practices were consistent throughout the country . In many regions, including St. Petersburg, Tatarstan, Krasnodar Krai, Ryazan, and Penza, United Russia used its Geo-SMS service to monitor and manage voter turnout.How does Geo-SMS work? It is a printed QR code placed at the PEC (Precinct Election Commission). Sometimes it's on a table, sometimes on a wall; in Gelendzhik, it was even attached to the presiding officers and the volunteers designated to assist people with disabilities, Let me make a brief side note.. According to Marina Antipina, a candidate of the Communist Party of the Russian Federation (CPRF), the volunteers in Gelendzhik were 15-year-old children. It is important to note that instead of helping people with disabilities, as required by the decision of the Territorial Election Commission, the underage volunteers were seen holding a poster with a link to the United Russia party service, which is used to illegally pressure people to vote! This is political exploitation of minors in a nutshell.

CPRF candidate discovers a QR code for United Russia's election monitoring service on the desk of the chairman of PEC No. 0939, Gelendzhik, Krasnodar region.

Here's how the system works: Voters, urged to vote by their supervisors, scan the QR code with their smartphones. The United Russia website https://edg-qr.er.ru/visits-qr/10 opens, and after entering their phone number, they receive a text message with a link to confirm their geolocation. The United Russia party service then registers that the person is indeed at the PEC.

If the phone number entered is in the database, the message "You have been sent a message with a link to verify your presence" is displayed. If the number is not in the database, the message "This phone is not registered" is displayed.

What the website looks like from the link in the QR code and the two ways to respond

When criticized, United Russia has commented: "The service is used exclusively for party supporters who have previously agreed to the processing of personal data. Participation in the project is voluntary and does not violate the right of citizens to express their will".

So who are those supporters, that has agreed to be in the register?  I decided to make a little test. 

Since I was a member of the Gelendzhik Territorial Election Committee and had the phone numbers of all the commission chairmen, I checked the numbers of about 50 people using this service. Only four phone numbers showed that the person was not in the database, probably because they had changed their number.That’ strange because it not like all of them were nominated by United Russia - there are also members of commissions from other parties: the Party of Growth, meetings of voters at places of study or work, the Cossack Party, the Rodina Party, Civic Force, and Just Russia. In other words, regardless of which party nominated them, almost the entire leadership of the PECs in Gelendzhik is in the database of United Russia supporters.

Table of my "check for a unanimous candidate" of the Gelendzhik PEC leadership

I would like to ask ironically: could the QR code for the CPRF application also be distributed to volunteers and commission chairmen across the country? Or perhaps a link to the Smart Voting recommendations?

Voluntary registration of United Russia in the election monitoring system

I believe that the phone numbers in the United Russia database did not come from "those who gave their consent to the processing of personal data", but rather through the illegal actions of government organizations acting in the interests of the party.

While checking the phone numbers through the service, I also checked the phone number of a person that I know personally, who works in a state-funded organization but is an opponent of United Russia. This phone number was in the database. It seems like they’re adding everyone they can reach through administrative channels.

Now let us analyze the situation region by region.

Election of the governor of the Kursk region

The election was held in three days with the use of  Remote Electronic Voting and nine days of early voting. A month before the elections, the CEC website removed the addresses of commissions and their composition.

Shpilkin chart for the Kursk region

Anomalous votes make up a quarter of the total, but it seems that we are entering a period where honest votes are so few that algorithms now consider them to be the anomalies. Therefore, in this Single Voting Day it is not worth paying attention to the quantitative characteristics.

The distribution of blue dots for Alexei Smirnov, the deputy governor, on the right side of the chart shows a sloping trend starting from a turnout of 25% and a result of 45%. The tail extending upward from there is a comet of ballot stuffing. Obviously, the 12 days of voting and the secrecy were used exclusively for fraud. Turnout at the end of the first day of voting was 47%, rising to 61% by the end of the election. That means that two quarters of the voters voted on Friday, a working day, and only one quarter voted on the weekend. 

In the Remote Electronic Voting, Smirnov's result was 50%. As a rule, the Single Voting Day is used to coerce public institutions workers, and since voters’ fear that their electronic vote may be verified, they vote with a corresponding bias in favor of the unanimous candidate. The last Single Voting Day showed that in most regions the results of the Remote Electronic Voting are higher than the results of the honest core.

The distribution and Remote Electronic Voting results suggest that Smirnov's true score is significantly less than 50%, starting where the sloping trend in the distribution begins. This means that Smirnov did not win this election in the first round, but instead received about 45-50%. 

Election of the Governor of the Orenburg Region

The election was held three days without the use of Remote Electronic Voting.

The distribution in Orenburg is very similar to the one in Primorye in 2023. It's unusually wide, almost as if they were trying to create a Gaussian distribution to make it look like everything was done fairly. This wide distribution even covers the "bell curve" that isn't distorted by ballot stuffing.

In the distribution, there's a noticeable dip just after 30% turnout, as well as a small cluster around 25% turnout and about 50% support for the candidate. These probably represent the commissions that counted the votes honestly, either because they had honest vote counting or were closely monitored by persistent observers

On the right side of the distribution, you can see a sloping trend typical of ballot stuffing, starting where the candidate received 50% of the vote.

Calculator spectrum for the Orenburg region

In the calculator spectrum, you can see some integer peaks, but they are not very pronounced: at 77%, 80%, and 90%. This suggests that the election went on for three days, relying mostly on ballot stuffing rather than fabricating numbers.

The real result for Governor Denis Pasler can be estimated to around 50% - nowhere near the 78% he gave himself.

St. Petersburg gubernatorial election

In Saint Petersburg, the election also lasted three days, and did not include Remote electronic voting. At first glance, the graphs of the Shpilkin method suggest that the election took place without significant ballot stuffing. The graphs show clusters at an unusually low turnout of 34%, with no "comet tails," double bell curves, or other irregularities. There is only a single bell curve, with few anomalous votes, almost as if the election day itself was conducted fairly.

But let's take a closer look. In the blue cluster for Beglov on the right side of the graph, you can see two strange densifications, which appear as horizontal lines.

Shpilkin chart for Saint Petersburg

Let's look at the "calculator spectrum" for these elections, which helps to detect when vote counts have been manipulated using a calculator. As expected, there are massive peaks indicating tampering.

Calculator Spectrum for St. Petersburg

The falsifications followed a very strange rule. If you look at the two largest formations, there's a peak at 54.5% and a steep slope just before 60% that drops off abruptly at 60%. These are the two formations visible in the previous distribution.

But why 54.5% instead of 55%? St. Petersburg has a quirk: sometimes the candidate's percentage is calculated incorrectly-they calculate the candidate's percentage as the number of votes divided by the number of valid ballots, not by the total number of ballots. If you create a Calculator Spectrum using only valid ballots, the largest peak shifts to a round number - 55%.

Calculator Spectrum for Saint Petersburg Based on Valid Ballots

The calculator spectrum for the turnout is unusual, showing peaks at integers: 30%, 35%, and 36%. Could they really have had to fabricate such low turnout numbers?

Calculator spectrum for voter turnout in St. Petersburg

From all this we can deduce what instructions were given to the falsification commissions: "For Beglov-NO MORE THAN 60%! Even better - 55%! Oh, and make the turnout 30-35%, not more than 36%." 

My first thought was that in these uncontested elections, with the other candidates clearly doing anything they could in order NOT to attract voters and with the presence of administrative coercion via "Geo-SMS," Beglov would still have received over 50%. However, the distribution pattern on Shpilkin's graph around 27% voter turnout leaves me uncertain. There are similar breaks in the curves to those seen in the presidential elections in March 2024, when Davankova was pushed further from second place. Therefore, my verdict is that it's still impossible to say for sure what the outcome in St. Petersburg would have been without falsifications. 

Conclusion

The elections in the Kursk, Orenburg, and St. Petersburg regions were rigged. To control the election results, even with strict measures to secure their place on the ballot, the authorities used both "turnout suppression" with administrative control of voting through "Geo-SMS" and multi-day voting, which facilitates falsification.

In addition, to illegally remove observers from commission documents or even from the counting process in combined elections, election commissions often used the excuse that the observer was only authorized to monitor their specific part of the count.

You can find more regional analysis on golosinfo.org. If you are interested in a specific region, do not hesitate to contact us.

Другие записи по теме «Falsifications»
InvestigationFalsifications5 years ago
The turnout was overestimated almost on 3 million people in 10 regions during 2018 presidential elections
Summary results of the sample check of video recordings from polling stations from the 2018 election of the President of the Russian Federation