Aug 02, 2024 07:00:00

Suspicious data patterns found in Venezuela's presidential election, whose results have been questioned around the world

In the Venezuelan presidential election held on July 28, 2024, the local electoral commission announced that President Nicolas Maduro won the election with over 51% of the votes. However, the election monitoring division of the Organization of American States , which includes the United States and Latin American countries,announced that the voting results announced by the Venezuelan electoral commission cannot be recognized . A blog summarizing statistics and social sciences from Columbia University has pointed out that suspicious data patterns have been found in the results of the Venezuelan presidential election, which has been questioned around the world.

Suspicious data pattern in recent Venezuelan election | Statistical Modeling, Causal Inference, and Social Science
https://statmodeling.stat.columbia.edu/2024/07/31/suspicious-data-pattern-in-recent-venezuelan-election/

More than six hours after polls closed in Venezuela's presidential election on Sunday, July 28, 2024, the chairman of the National Electoral Commission announced President Maduro's victory. The National Electoral Commission stated, '80% of the votes have been counted and strong trends continue. The official results of today's vote will be announced,' and announced the following voting results.

Candidate Name	Number of votes	Vote share
Nicolas Maduro	5,150,920 votes	51.2%
Edmundo Gonzalez	4,445,978 votes	44.2%
others	462,704 votes	4.6%
total	10,058,774 votes	100.0%

If we were to write the vote share data in detail to seven decimal places, it would look like this:

Candidate Name	Number of votes	Vote share
Nicolas Maduro	5,150,920 votes	51.1999971%
Edmundo Gonzalez	4,445,978 votes	44.1999989%
others	462,704 votes	4.6000039%
total	10,058,774 votes	100.0000000%

The blog points out that this result is 'as if the vote percentage was not calculated from the number of votes, but rather the vote percentage was determined and then the number of votes was calculated.' In other words, it is pointed out that the election results were created based on the assumption that President Maduro would win.

For example, if the total number of votes was decided first, and President Maduro received 51.2% of the votes, his opponent Gonzalez 44.2%, and the other candidates 4.6%, the number of votes for each candidate would be as follows: However, if the data were to be made public as is, it would be obvious that the data had been tampered with.

Candidate Name	Number of votes	Vote share
Nicolas Maduro	5,150,922,880 votes	51.2000000%
Edmundo Gonzalez	4,445,978,108 votes	44.2000000%
others	462,703,604 votes	4.6000000%
total	10,058,774 votes	100.0000000%

So when they rounded down the results, they got the results announced by Venezuela's National Electoral Commission.

Candidate Name	Number of votes	Vote share
Nicolas Maduro	5,150,920 votes	51.2%
Edmundo Gonzalez	4,445,978 votes	44.2%
others	462,704 votes	4.6%
total	10,058,774 votes	100.0%

In order to verify whether the above figures could be obtained in real-world voting, the blog ran a simulation in which a randomly distributed vote share was assigned to each candidate, the vote share was rounded up or down to the nearest whole number, and then multiplied by the total number of votes, 10,058,774, and rounded up or down to the nearest whole number. However, even after repeating the simulation 1 million times, the rounding never worked.

The blog also points out that it seems a bit odd that the electoral commission announced President Maduro's victory with 20% of the votes still remaining.