An analysis of the recounted ballots also shows a number of anomalies. For example:
* Most of the difference between the recounted and the original totals is due to 116 ballot boxes that lost an average of 63 percent of their votes during the recount. These were largely ballot boxes that contained more than the proscribed limit of 750 votes.
* Not all of the ballot boxes that had more than 750 votes were re-opened. The ones that were opened had a significantly higher percentage of votes for Lopez Obrador than the ones that were not opened. This raises the possibility that the recount gave Lopez Obrador a net loss of votes because of the way in which these “over voted” ballot boxes, which lost most of their votes during the recount, were selected to be opened.
* The majority of the null votes (17,129 or about 2 percent of the total votes) in the recounted ballot boxes were removed in the recount. The IFE did not explain whether any of these null votes became valid votes in the recount. If so, this is potentially important because the total number of null votes in the presidential race is more than three times the margin of difference between the two top candidates.
The authors note that it is possible that these and other anomalies found in the recounted data, described in the paper, have reasonable explanations. However, what is most difficult to explain is the lack of transparency in the process and the inordinate amount of time that the IFE has taken to publicize information – still very incomplete – on the recount that has taken place.
“It is unfortunate that the Federal Electoral Tribunal made a decision about which ballot boxes to recount before the results of this first partial recount were explained to the public,” said Weisbrot. “Furthermore, if this new recount is not conducted very differently than the last one, it is difficult to see how it will be of much use in obtaining a credible result.”