pith. sign in

arxiv: 2509.00118 · v2 · submitted 2025-08-28 · ⚛️ physics.soc-ph

Statistical study of the transcript of vote counts in multi-member constituencies: Identifying electoral fraud and reconstructing voting returns

Andrey V. Podlazov , Vadim Makarov This is my paper

Pith reviewed 2026-05-18 19:56 UTC · model grok-4.3

classification ⚛️ physics.soc-ph
keywords electoral statisticsballot stuffingstatistical testsvote reconstructionmulti-member constituencieselectoral fraudvoter behavior modeltranscript credibility
0
0 comments X

The pith

Statistical tests on vote transcripts can detect ballot stuffing and reconstruct multi-member election results.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper proposes new methods of electoral statistics for studying transcripts of vote counting in municipal elections. It constructs and applies statistical tests to detect ballot stuffing at the level of individual candidates and for ballots as a whole. The authors also examine the difference between falsifications for the administration and against the opposition, and construct indicators for assessing the credibility of transcripts as a whole. They propose a model of voter behavior to reconstruct the results of multi-member voting at the level of an individual precinct and carry out an independent verification of the reconstruction. A sympathetic reader would care because these tools could help identify fraud and recover accurate vote distributions from public data in elections where direct observation is limited.

Core claim

The central claim is that new statistical methods applied to vote counting transcripts can effectively detect ballot stuffing for individual candidates and overall ballots, distinguish types of falsifications, assess transcript credibility, and through a voter behavior model, reconstruct multi-member voting results at the precinct level with independent verification.

What carries the argument

The statistical tests for detecting ballot stuffing and the voter behavior model for reconstructing precinct-level results, which together identify anomalies in vote distributions and enable verification against observed counts.

If this is right

  • Ballot stuffing can be detected at the level of individual candidates.
  • Falsifications can be distinguished between those benefiting the administration and those against the opposition.
  • Indicators can be used to assess the credibility of entire vote count transcripts.
  • A voter behavior model can reconstruct multi-member voting results at individual precincts.
  • Independent verification can confirm the accuracy of the reconstructions.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • The methods could be used to analyze elections in other countries or types beyond municipal ones.
  • Reconstructions might provide evidence in legal challenges to election results.
  • The statistical distributions assumed could be tested against data from transparent elections to refine the model.
  • This work connects to broader efforts in using data science for election monitoring.

Load-bearing premise

The assumption that observed deviations from the proposed statistical distributions or voter behavior model primarily reflect ballot stuffing or falsification rather than unmodeled natural variations in turnout, preference clustering, or administrative errors.

What would settle it

If the statistical tests are applied to a set of elections known to have no fraud and they do not detect any ballot stuffing, or if the reconstruction model is applied to precincts where the actual individual votes are known and it accurately reproduces the results.

read the original abstract

We propose new methods of electoral statistics. With their help, we study transcripts of vote counting in municipal elections. We construct and apply effective statistical tests to detect the ballot stuffing at the level of individual candidates and ballots as a whole. We study the difference between falsifications for the administration and against the opposition. We construct indicators assessing the credibility of transcripts as a whole. We propose a model of voter behavior to reconstruct the results of multi-member voting at the level of an individual precinct. We carry out an independent verification of the reconstruction.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 2 minor

Summary. The paper claims to introduce new statistical methods for analyzing vote count transcripts in municipal multi-member elections. It develops and applies tests to detect ballot stuffing at the level of individual candidates and overall ballots, examines differences in falsification favoring the administration versus harming the opposition, constructs indicators for transcript credibility, proposes a voter behavior model to reconstruct precinct-level results, and performs an independent verification of the reconstruction.

Significance. If the tests and model hold after addressing the noted concerns, the work would provide a useful framework for statistical detection of electoral anomalies in multi-member systems, extending techniques from statistical physics to electoral forensics. The independent verification of the reconstruction is a positive element that enhances the potential applicability of the results for practical analysis of election data.

major comments (2)
  1. [§4] §4 (Statistical tests): The null distributions for honest voting in the ballot-stuffing detection tests do not include explicit modeling or robustness checks for natural correlations such as preference clustering or turnout heterogeneity common in multi-member constituencies. This assumption is load-bearing for the central claim that observed deviations primarily indicate fraud, as unmodeled variations could produce similar patterns without falsification.
  2. [§6] §6 (Reconstruction model): The voter behavior model used for precinct-level reconstruction relies on specific behavioral assumptions whose validity is not demonstrated against alternative explanations like administrative rounding or preference dependencies; without sensitivity analysis or bounds on reconstruction error under violated assumptions, the independent verification does not fully establish reliability of the reconstructed returns.
minor comments (2)
  1. [Abstract] The abstract could briefly specify the form of the proposed distributions or null models to improve immediate clarity for readers.
  2. [Results] Figure captions in the results sections would benefit from explicit comparison of observed data to the expected null distributions to aid interpretation.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive review of our manuscript on statistical detection of electoral anomalies in multi-member constituencies. We address each major comment below with point-by-point responses and indicate planned revisions to enhance the robustness of the null models and reconstruction approach.

read point-by-point responses

  1. Referee: [§4] §4 (Statistical tests): The null distributions for honest voting in the ballot-stuffing detection tests do not include explicit modeling or robustness checks for natural correlations such as preference clustering or turnout heterogeneity common in multi-member constituencies. This assumption is load-bearing for the central claim that observed deviations primarily indicate fraud, as unmodeled variations could produce similar patterns without falsification.

    Authors: We appreciate this observation on the null model assumptions. The current null distributions are constructed from a baseline of independent candidate selections within the fixed ballot structure of multi-member elections, which aligns with standard approaches in electoral statistics. To strengthen the analysis, we will incorporate explicit robustness checks in the revised §4. These will include Monte Carlo simulations introducing moderate preference clustering (via correlated Dirichlet distributions) and turnout heterogeneity (via precinct-specific participation rates drawn from empirical distributions). Our preliminary checks indicate that while such factors generate some variance, the magnitude and specificity of deviations in the studied transcripts (particularly candidate-level overperformance patterns) remain inconsistent with these natural variations and continue to support the fraud detection claims. The revised text will present these results and update the interpretation accordingly. revision: partial

  2. Referee: [§6] §6 (Reconstruction model): The voter behavior model used for precinct-level reconstruction relies on specific behavioral assumptions whose validity is not demonstrated against alternative explanations like administrative rounding or preference dependencies; without sensitivity analysis or bounds on reconstruction error under violated assumptions, the independent verification does not fully establish reliability of the reconstructed returns.

    Authors: We agree that additional validation of the behavioral assumptions would improve the reconstruction section. The model employs a probabilistic framework for voter allocation across candidates, calibrated directly to the observed transcript aggregates. In the revision, we will expand §6 with a dedicated sensitivity analysis that perturbs key parameters to simulate administrative rounding effects and introduces controlled preference dependencies. We will also compute and report explicit bounds on reconstruction error under these alternative scenarios. These additions will complement the existing independent verification and provide quantitative support for the reliability of the precinct-level results. revision: yes

Circularity Check

0 steps flagged

No significant circularity; derivation relies on external statistical assumptions and independent verification.

full rationale

The paper introduces statistical tests for detecting ballot stuffing based on deviations from proposed null distributions for honest multi-member voting and a separate voter behavior model for precinct-level reconstruction, followed by an independent verification step. No quoted equations or sections show a prediction or result reducing by construction to a parameter fitted from the same data, nor does any central claim depend on a self-citation chain that itself assumes the target result. The approach is self-contained against external benchmarks, with the weakest assumption being the interpretation of deviations as fraud rather than unmodeled natural variation, which is a modeling choice rather than a definitional loop.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Abstract-only view yields no explicit free parameters, axioms, or invented entities; any statistical distributions or voter behavior assumptions remain unspecified.

pith-pipeline@v0.9.0 · 5619 in / 1110 out tokens · 31732 ms · 2026-05-18T19:56:47.631777+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

What do these tags mean?
matches
The paper's claim is directly supported by a theorem in the formal canon.
supports
The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends
The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses
The paper appears to rely on the theorem as machinery.
contradicts
The paper's claim conflicts with a theorem or certificate in the canon.
unclear
Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

11 extracted references · 11 canonical work pages

  1. [1]

    последовательные бюллетени независимы

    work page

  2. [2]

    В основе предположения 1 лежат особенности процедуры голосовани я

    вероятности голосования постоянны во времени. В основе предположения 1 лежат особенности процедуры голосовани я. Даже если избиратели схожих пол итических взглядов (родственники , соседи, сослуживцы) голосуют одновременно , их бюллетени случайным образ ом пада- ют в избирательную урну, где пер емешиваются, а при вываливании ее содер- жимого на стол и соби...

    work page

  3. [3]

    или сложные (участок 212). 23 Таким образом, наличие опорных да нных позволяет использовать ин фор- мационную энтропию как простой ка чественный индикатор простых ф альси- фикаций, нечувствительный к перемешиванию. Однако желательно им еть ана- логичный индикатор, способный хотя бы отчасти справляться и со сложными фальсификациями. В основу индик атора та...

    work page 2000

  4. [4]

    Ф о р м а л ь н ы е м е т о д ы в ы я в л е н и я м а с ш т а б н ы х э л е к т о р а л ь н ы х фальсификаций // Проектирование будущего

    Подлазов А.В. Ф о р м а л ь н ы е м е т о д ы в ы я в л е н и я м а с ш т а б н ы х э л е к т о р а л ь н ы х фальсификаций // Проектирование будущего. Проблемы цифровой реально- сти. 2019. Вып.2, 120-137

    work page 2019

  5. [5]

    Формальное выявление выдума нных результатов выборов // Проектирование будущего

    Подлазов А.В. Формальное выявление выдума нных результатов выборов // Проектирование будущего. Пробле мы цифровой реальности. 2020. Вы п.3, 176-190

    work page 2020

  6. [6]

    Совершенствование строгих м етодов выявления выдуманных итогов голосования: Практическая электоральная статистика // Проектирование будущего

    Подлазов А.В. Совершенствование строгих м етодов выявления выдуманных итогов голосования: Практическая электоральная статистика // Проектирование будущего. Проблемы цифровой реальности. 2024. Вып.7, 251-280

    work page 2024

  7. [7]

    Dual approach to proving e lectoral fraud via statistics and forensics

    Podlazov A., Makarov V. Dual approach to proving e lectoral fraud via statistics and forensics. arXiv:2412.04535

    work page arXiv

  8. [8]

    Реконструкция фальсифицированных результатов выборов с помощью интегрального метода Шпилькина // Проектирование будущего

    Подлазов А.В. Реконструкция фальсифицированных результатов выборов с помощью интегрального метода Шпилькина // Проектирование будущего. Проблемы цифровой реальности. 2021. Вып.4, 193-208

    work page 2021

  9. [9]

    Выделение основного кластера на диаграмме рассеяния элек- торальных данных // Проектирование будущего

    Подлазов А.В. Выделение основного кластера на диаграмме рассеяния элек- торальных данных // Проектирование будущего. Проблемы цифровой ре- альности. 2022. Вып.5, 193-204

    work page 2022

  10. [10]

    Фальсификация процедуры выборов без фальсификации итогов голосования: К десятилетию выборов мэра Москвы 2013 г

    Подлазов А.В. Фальсификация процедуры выборов без фальсификации итогов голосования: К десятилетию выборов мэра Москвы 2013 г. // Проектирование будущего. Проблемы цифровой реальности. 2023. Вып.6, 186-198

    work page 2013

  11. [11]

    Выборы депутатов Государственной Думы VII созыва: Выяв- ление фальсификации результа тов и их реконструкция // Социологические исследования

    Подлазов А.В. Выборы депутатов Государственной Думы VII созыва: Выяв- ление фальсификации результа тов и их реконструкция // Социологические исследования. 2018. 405(1), 59-72

    work page 2018