Fraude Mexico 2006 — part 5 of 10

I only half-watched  part five of Fraude Mexico 2006 — sorry — once in a while “real life” interferes with other plans.  Briefly, in this segment of the video, we see the fall-out from the initial count (the PREP) of the 2-July-2006 election vote.

The PREP is a mathematical model based on a vote sampling, which is supposed to accurately reflect the final vote.  Because Mexican ballots are hand counted, and because congressional representatives are based both on a winner take all for single districts and proportional representation for parties in multiple districts (its too complicated to go into right now) — and because in many places parties run joint tickets, and give those proportion seats based on how well the joint ticket did overall … and because things like last-minute polling and exit polling is illegal in Mexico (in this election, Fox and Sky News broadcasts were both blacked out because the two foreign cable networks refused to comply with Mexican election laws, and were reporting on the election on their English language broadcasts)  — and the media is waiting to report the results (and the candidates want to know who won)… PREP is pretty important, and watched closely.

This is where things went hay-wire in 2006.  The early returns showed a close election, but with Lopez Obrador winning with two or three percentage points above Calderon (and well ahead of the other three candidates).  Then, as the evening went on, Calderon’s percentage of the vote rose at an even pace, while Lopez Obrador’s fell.  The final PREP results led to duelling probability theories… and mathematical modelling suddenly taking on political coloration.

It was obvious early in the PREP count that something was wrong… Luis Carlos Ugalde is shown early in this segment saying getting the count out was going to take longer than expected.  Coupled with (what I still believe was an improbable) ratio of change in the sample count over the course of the evening, Calderon’s early claims of victory based on projections from software developed by his brother-in-law’s company are the best evidence that something was not right about the count.

The mathematician show in this segment’s explanation is that an unknown quantity of additional ballots were countedfrom districts that would be pro-Calderon, and an unknown quantity of ballots were subtracted from the sample in pro-AMLO districts, which would fudge the numbers. Given that the rate of change in the sample count was an even progression over time, it is difficult to believe there was not software manipulation.  The probability of this happening naturally is just too low.

At this point, Lopez Obrador’s call for counting vote by vote, ballot box by ballot box (“voto par voto, casilla par casilla”) was not that extreme — “I defended the principal of free elections”, he says.