This is how the statistics reveal the fraud in the Russian elections

After every election in Russia, the Russian-language blogger is eagerly waiting for Sergei Shpilkin to publish his plans. What it examines is how votes are distributed in relation to turnout.

Every election is the same: the normal distribution curve is broken by Putin’s party, which suddenly gets an uncontrollable number of votes.

Sjpilkin was widely praised and quoted by both the opposition and independent Russian media. The accounts he said Online publishing He even pushed the Russian Central Election Commission to invite him to the committee’s scientific board (which was canceled in 2020).

Unfortunately, this did not lead to anything. Sergei Sibelkin says it felt like a bureaucratic exercise.

himself A humble person calls accounts “my hobby”. When we met in a cafe near Kitaj Gorod in central Moscow, he indicated that he was neither a mathematician nor a statistician. He is a Soviet-trained physicist who enjoys making statistical calculations.

– There are no difficult things to do. Anyone can do it. I collect numbers in a diagram and interpret them according to accepted methods.

The graphs clearly show that the higher the turnout, the more votes for Putin’s party, United Russia. This turnout affecting party support is not unusual in itself – for example in northern democracies, high turnout is known to often benefit left-wing parties. But this does not mean that only left-wing parties get more votes. Usually, the sounds should be distributed more evenly.

In Russia you can instead Let’s witness a statistical anomaly: When turnout goes up, only United Russia gets the vote.

Could an increase in turnout only benefit Putin’s party? Statistically speaking, the answer is no. A normal distribution curve means that most variables of a given quantity are relatively close to the mean, making their distribution hill-like.

But in the Russian elections there will not even be a hill. The reason is simple: manipulation. In simple language, the pre-filled ballot papers that are filled into the ballot boxes.

Sgiblkins Point Diagram Describes the power party figures Something he calls a comet’s tail.

– We have about a hundred thousand polling stations in Russia. In the scatter plot, we see that the result from more than half of them describes a compressed cloud. They are all close to each other, in the sense that this is the real outcome of the election. Here, election officials were honestly calculating. But in unusually high turnout polls, United Russia numbers are growing rapidly and forming a guilty tail, says Sjpilkin.

A comet’s tail begins to form as the share rises. It depends on the completed ballot papers.

We also see that an astonishing number of polling stations got exactly 75 percent for United Russia. They are so many that they describe a fine line. It’s statistically impossible for so many polling stations to get exactly 75 percent, Sjpilkin says, people come to polling stations at random to vote.

The reason is that some regional electoral commissions decided to present a “beautiful” result: a high and equal figure for Putin’s party. Like 75 percent. This is achieved by placing pre-filled votes into the ballot boxes and then arranging them by modifying the final result in the election protocol.

Sergey Shpelkin points out that That there are no polling stations close to the official result in terms of the number of votes and turnout. The official participation rate in the elections to the State Duma this year was 51.72 percent. United Russia won the elections with 49.82 percent.

Usually, this should mean that polling stations form a block around that specific intersection. But they didn’t – the bloc was formed instead around the intersection where turnout is 38 percent and United Russia gets 32 percent. Then the total numbers are subtracted by the comet’s tail.

– Election officials who have been commissioned to cheat believe you can put a lot of pre-filled ballot papers into the ballot box without them being visible. Then they reported a turnout of 75 percent and United Russia got 75 percent. Sjebelkin says they don’t think these things are apparent in the statistics.

Although the Russian electoral authorities still penalize cheating, giving them a surprisingly large amount of access to election data. It’s impossible, for example, from the Syrian presidential election, to get evenly detailed numbers, says Sjebelkin.

– In Syria, you can only know the total number of votes and the number of votes the candidates got. Nothing about how the votes are distributed. The numbers show that they calculated by a calculator how many votes Assad should get, how many other candidates he should get, and that this should coincide with the turnout. They are similar in Belarus – Belarusian electoral authorities do not publish numbers by polling stations, but only by electoral districts.

How, then, is it possible for the Russian electoral authorities to continue to publish such detailed figures that, for example, Jebelkin could compile his schemes?

The system has been in place since the beginning of the first decade of the twenty-first century. Removing an existing system is difficult. It is a clear step backwards. But it is clear that my investigations and many other investigations bother them. Now they are trying instead to make it more difficult for us to gather information, says Sjpilkin.

In the last Duma elections On September 19, electoral authorities attempted to put the sticks in the wheel by encoding the numbers in a way that made them impossible to copy.

But the electoral authorities did not expect Russia to be the country of programmers.

– We’ve analyzed it. It’s not just me who deals with this, skilled programmers quickly fixed the problem, says Sergey Shpelkin and smiles ignorantly.

He maintains that his graph can’t really be seen as evidence of cheating. What they show is the extent of cheating. According to Sjpilkin’s calculations, United Russia received fourteen million false votes in the elections to the State Duma this year.

The distribution curve is not valid as evidence in court. You must have concrete evidence. We did so thanks to the election observers. But it is impossible to send election observers to all 96,325 polling stations. Statistical methods are a cheap way to get an idea of ​​the extent of fraud.

Sjpilkin confirms That there are many local electoral commissions that count the votes honestly.

– For example, I found polling stations in the Krasnodar region where everything went honestly, although they are in an area where cheating is common. Election fraud is more common in rural areas than in cities, but these polling stations prove that fraud cannot be blamed on any kind of “mentality”. Our election officials don’t cheat if left to themselves by those in power. They want to count the votes honestly.