Doing maths in Mexico

[cal04]

While Mexicans take to the streets over the presidential vote, democracy's fairweather friends are standing silent

(snip)
the stalwarts of democracy outside Mexico are silent. Bush has congratulated Calderón, not waiting for the court to rule. Reuters and Bloomberg echo the confidence of the elites that Calderón will win in court - never mind whether he won at the polls. When The New York Times is heard from, the headlines tell us of the "leftist claims" about the occurrence of fraud, while Calderón is described as "presidential." The Times never doubted that fraud did occur in Ukraine. In Mexico on the other hand, it seemingly renounces any duty to examine the facts on the ground.


(snip)
Mochán's work calls attention to at least four important anomalies in the count.

1. Calderón's percentage lead in the count started at around seven percentage points, and diminished steadily in percentage terms through the first part of the count. This corresponded to a remarkably constant absolute differential between Calderón and AMLO as the count progressed. Is this normal? The count depended on the arrival of the boxes; if this were absolutely random then the proportions should have held roughly constant while absolute differentials widened, as actually happened to the differential between Calderón and the third major candidate, Madrazo of the PRI, for most of the evening. Why did the Calderón-AMLO differential follow a different rule?

2. The PREP results went on view only after the first 10,000 boxes had been processed. If those first 10,000 boxes resembled what came later, then extrapolating backward should produce a line intersecting the origin - each candidate should have started with zero votes. For Calderón this is the case, but for AMLO it is not: the AMLO intercept is actually at minus 126,000 votes. Thus, the first 10,000 boxes were markedly different from those that followed. How?

3. There are gross anomalies in the number of votes counted per five-minute interval as the count finishes. Over the course of the evening, the pattern of vote counts set a normal range for this variable. As the last boxes came in, however, it was radically violated, with many more votes piled in, per interval, than was normal before. Moreover, toward the very end, PREP reset the box count, which regressed from 127,936 at 13.17 on July 3 to 127,713 at 13.50, meaning that records for 223 boxes disappeared. 33 minutes had by then passed with no updates. When they resumed, there were updates with absurd results: more than 6000 votes per box at 13:57, and then updates with large negative votes per box at 13:57 and 14:03.


more by James K Galbraith
physicist Luis Mochán of UNAM
http://commentisfree.guardian.co.uk/james_k_galbraith/2006/07/the_mexican_standoff.html

[larrysh]

1. Does he mean Obredor is fu-------????

[rman]

2. No, rather the contrary

[IndyOp]

4. Welcome to DU, larrysh! James K Galbraith is not saying

Lopez Obrador has no hope, he is saying - in no uncertain terms - that Lopez Obrador won. What the Mexican Election Tribunal will do, is not yet certain. I think 1,000,000+ people in the streets for the next 7 weeks might increase the likelihood of a fair ruling.

:hi:

[IndyOp]

10. Correction for my post just above:

Galbraith is NOT saying that Lopez Obrador won - he is saying that there appear to be some anomalies in the returns.

I would argue that the anomalies are more than sufficient for a full investigation - on the ground - with police detectives and lawyers and judges.

It is interesting to me to see so clearly how the candidate and citizen activists are expected to do the full investigation and prove without a shadow of a doubt that election fraud occurred before the corporate media and other citizens will listen. If this is the case, it indicates that election law is gravely inadequate. The law should clearly indicate what citizens need to do to activate a full investigation of a race and what that investigation should entail.

:kick:

[michaelvail]

3. Fighting for democracy: Nearly one million mexicans protest elections

Mexicans are fighting for their own democracy. Nearly one million Mexicans out on the street representing the other Presidential candidate. They are not sure if he would have been any better than the establishment lackey that has been placed in office but it is the principle which counts. Would Americans ever rise up and defend their democracy? The Mexicans are representing their country and their views, right or wrong I respect their integrity and their will to fight the aristocrats that aren’t even mexican that have destroyed their country.

http://www.thought-criminal.org/2006/07/17/fighting-for-democracy-nearly-one-million-mexicans-protest-elections/

[IndyOp]

5. Exclusive to DU !! Guanajuato 2006 is Mexico’s Florida 2000

Please come read & kick a new thread in GD...

http://www.democraticunderground.com/discuss/duboard.php?az=view_all&address=364x1654610

[EFerrari]

6. Done. n/t

[Kurovski]

7. K&R

I particularly like how the article points out the elite corporate press reaction to Mexico's election.

[ProgressiveEconomist]

14. K&r

[OnTheOtherHand]

8. a note of caution

Mochán's Spanish-language page is very interesting. As I've commented elsewhere, it is inherently risky to talk about "anomalies" when we don't have a solid empirical baseline to compare them to.

"1. Calderón's percentage lead in the count started at around seven percentage points, and diminished steadily in percentage terms through the first part of the count.... This corresponded to a remarkably constant absolute differential between Calderón and AMLO as the count progressed. Is this normal?"

No one knows, as far as I know. I've certainly seen U.S. election returns in which one candidate jumped out to a lead, and then the race narrowed or even reversed. The vote counts don't arrive at random.

Note that the discussion in the OP refers to the PREP count; in the official count the following day, apparently AMLO took the lead and then Calderón overtook him. But as far as I can tell so far the biggest difference between the two counts is that the results seem to have arrived in different orders. Is there any sequence that we would feel confident wasn't "anomalous"?

(Point 2 basically restates point 1: obviously the first 10,000 boxes recorded in the PREP were markedly different.)

3. You can see this going-backwards in these two 'screen shots' courtesy of Mochán's website:
http://em.fis.unam.mx/public/mochan/elecciones/prep1/resultados172.html (13:17)
http://em.fis.unam.mx/public/mochan/elecciones/prep1/resultados174.html (13:50)
The index of all the screen shots is at
http://em.fis.unam.mx/public/mochan/elecciones/prep1/

But note that Calderón actually loses more votes (not many) than AMLO does. Did something sneaky happen? I don't know. Did it hurt AMLO? Well, conceivably, but not obviously.

Also, Galbraith's column asserts (in point 4) that "the differences in Calderón and AMLO's votes per box... ought to follow a normal curve...." Umm, maybe, maybe not. It isn't a law of nature. The normal curve isn't all that "normal"; all sorts of real-life distributions depart from it. Anyone who can look at that distro and tell you that AMLO did win -- or didn't win -- should, in a just world, forfeit credibility. Mochán did not do either of those things, and I have no gripe with him at all. I don't even have much gripe with Galbraith.

If form holds, someone is about to accuse me of defending Calderón, suppressing important debate, etc., etc. Nope. Calderón hasn't won yet, I don't know who received more votes, and there is a lot of work to be done. Even if the PREP count was clean, that wouldn't begin to prove that Calderón won. What I would really like to avoid, if possible, is latching on to factoids that don't mean what some people might want them to mean, and that are likely to be transformed by the folk process into unfactoids. I fear that very soon there will be people swearing that the results were obviously "statistically impossible," and then there will be people saying that with all due respect, they weren't impossible, and then there will be stuff about agents and right-wing foundations, and... well, I don't think the Mexican people will thank anyone for the ensuing spectacle. But I could be wrong.

[Adelante]

9. Thank you for that, OnTheOtherHand

Anybody could be wrong in a situation like this. I appreciate your presenting your insights. We should each keep an open mind until the evidence presented has been proven or disproven.

[IndyOp]

11. What are the strongest points Mochan & Galbraith make? (n/t)

[OnTheOtherHand]

12. the difference distribution may be most interesting

It's important to understand that Mochán's analysis basically is of the PREP, and most of it has to do with the order in which the PREP data arrived. Blatant discrepancies in the PREP data could indicate some form of fraud -- but if (I haven't seen any evidence of this) the PAN stole a bunch of votes, it wouldn't really care what order the votes were then recorded in.

I think the oddest thing about the PREP results (which does carry through to the full count) is the distribution of the difference between Calderón and AMLO votes -- because it appears almost "normal" (bell-shaped) except for a "dent": there are fewer precincts where AMLO wins by a bit than where he loses by a bit. That could indicate double-digit fraud in a lot of casillas, or it could be sheer coincidence. The distro will never tell us; at most it's a hint, maybe a false lead.

[cal04]

13. thank you for the explanations

I really appreciate your posts

[JamesGalbraith]

15. Cautions

I'm not opposed to cautionary warnings. The anomalies Mochan identified seem to us to warrant serious investigation. I said in the piece that they are consistent with the possibility of fraud. And I tried to clarify exactly what the indications detected by Prof. Mochan appear to show.

I agree that of all the anomalies, the departure of the difference (FCC-minus-AMLO) distributions from normal are the most interesting (and they do not depend on the order of count). They are consistent with the possibility that votes were shifted from AMLO to FCC in the close boxes, but not the ones that were heavily pro-AMLO or pro-Calderon. And this is consistent with a fair amount of visual evidence, from the handwritten tally sheets, that votes for AMLO were simply mistranscribed into the electronic count.

But the downward shift of the intercept for AMLO in the first 10,000 boxes is also interesting. If there were a similar shift, or an upshift, for Calderon, then it would be easier to dismiss this one as reflecting (say) a different part of the country or of the electorate. But how to explain an intercept of -126,000 for AMLO while the intercept for Calderon stays at zero?

It is, once again, *possible* that there are innocent explanations. It would, however, be a mistake to assume that because innocent explanations remain possible, that the anomalies should be brushed aside. If innocent explanations exist, then it's reasonable to ask those in the know to explain exactly what they are. The regression of the count in the late stages is something that could have a purely bureaucratic explanation. But it's clearly anomalous, it should not have happened even if it were not indicative of fraud, and therefore it especially warrants a documented explanation from the PREP authorities.

Having said that, as I follow the news I see many more bits of evidence, tending to support the view that the irregularities were very significant. I have not seen much that would point toward the contrary view.