Evidence Against Fraudulent Votes Being Decisive in the Bolivia 2019 Election

In this document, Professor Walter R. Mebane, Jr, makes a case for the legitimacy of the general elections that took place in Bolivia in October 2019, despite the irregularities.


  • Even by discarding votes likely to be manufactured, and attributing those likely to be stolen to the second runner up, Evo Morales would still win by a more than 10% lead.
  • The number of "fraudulent" votes was determined through statistical analysis of the vote counts.

Electoral integrity analysis: General Elections in the Plurinational State of Bolivia

In this report the OAS state their concerns over the irregularities they perceived in the general elections that took place in Bolivia in October 2019.


  • The OAS can not certify the results of the TREP to be accurate because of the following reasons:
    • An unmonitored servers, BO1, was included in the infrastructure.
    • The data flow was redirected to an undocumented server, after the interruption 7:40 PM, despite the available contingency equipment.
    • The data flow subsequently reached the primary server in a way that eludes the audit company's oversight.
    • The used servers contain a different number of logs.
    • The interruption of the TREP was not properly documented.
    • Tally sheets were transmitted with dates outside of the TREP life cycle.
    • Metadata vital for transparency was discarded.
  • The OAS can not certify the results of the official count to be accurate because of the following reasons:
    • Best practices were not followed and the the software was not formally tested.
    • Authentication was weak.
    • Test data was mixed with election day tally sheets.
    • 1,575 TREP tally sheets were mixed in the official count.
    • During the process, the head of the software company recompiled the app, and was given direct unrestricted access to the databases.
  • Irregularities were found in 78 of 333 tally sheets analyzed by the OAS technical team.
  • Some tally sheets from Argentina (67 of 176 analyzed) showed more votes than voters on the registration lists.
  • The location of the official vote count was changed in some municipalities. Party delegates were not informed in time to be able to witness the count on the new premises.
  • Protocols for the handling of electoral material was not standardized. The handling of electoral material was not documented.
  • The increase in votes received by Morales in the last 5% is highly unlikely.


  • TREP: TransmisiĆ³n de Resultados Electorales Preliminares.
    A system used to quickly count votes and get a preliminary estimate of the results of the election. Those results are not legally binding, and are obtained with pictures of the tally sheets.
  • Official Count: Official counting of the election results. Those results determine the outcome, and are obtained by counting the votes on the tally sheets directly.
  • OAS: Organization of American States.