Three predictions on July 2012 Federal Elections in Mexico based on past regularities

TL;DR

This paper analyzes past Mexican electoral data to identify regularities and predicts that similar statistical patterns will appear in the 2012 elections, based on sociophysics models and previous findings.

Contribution

It introduces three characteristic regularities observed in Mexican elections and applies them to predict the 2012 electoral outcomes.

Findings

Error distributions follow power laws with asymmetric lobes.

PRI shows a slope change after 70% of votes processed.

Vote distributions fit Daisy model distributions.

Abstract

Electoral systems are subject of study for physicist and mathematicians in last years given place to a new area: sociophysics. Based on previous works of the author on the Mexican electoral processes in the new millennium, he found three characteristics appearing along the 2000 and 2006 preliminary dataset offered by the electoral authorities, named PREP: I) Error distributions are not Gaussian or Lorentzian, they are characterized for power laws at the center and asymmetric lobes at each side. II) The Partido Revolucionario Institucional (PRI) presented a change in the slope of the percentage of votes obtained when it go beyond the 70% of processed certificates; hence it have an improvement at the end of the electoral computation. III) The distribution of votes for the PRI is a smooth function well described by Daisy model distributions of rank $r$ in all the analyzed cases, presidential and congressional elections in 2000, 2003 and 2006. If all these characteristics are proper of the Mexican reality they should appear in the July 2012 process. Here I discuss some arguments on why such a behaviors could appear in the present process