These are forecasts based on pictures of ballot tallies. And this website is made by the opposition. I'm not saying they're wrong, I just think it's silly to take their word for it instead of just waiting until we have more data and these results have been looked over by independent parties

> Many Venezuelan people have verified their records. Including the “witnesses” who have signed all the records. The signatures can be seen on the pictures

The thing is that these witness signatures could be forged (together with the tallies)

I assume that with "verified their records" you mean exactly what I'm asking for (but since I don't know any Venezuelan living in Venezuela, I haven't seen anyone verifying that stuff)

There is no need to do any difficult OCR, the QR code contains the tally in machine-readable format. You just need a phone and copy the result to a csv file

You could make a public spreadsheet so others can join in on the crowdsourced effort

Just make a list of every precinct and whether or not we have a screenshot for them. People can help by uploading more pictures and which precinct it corresponds to and also manually reading the results and updating the spreadsheet

Do you have any details about the kind of mechanism that they used?

If it's a private/public kind of mechanism, they should be able to disclose the public key for signature verification.

If it's not, and it's some kind of a HMAC, and the political parties have all access to the key... Then this doesn't protect at all against the threat implied (the different parties don't trust each other, and both claim that they are trying to "steal" the election), since these signatures could be forged by any of the political parties with access to the key

Even in the former case, it could be possible that a machine could be compromised, and could have emitted two tallies (one for the actual election, and another one with different numbers and forged signatures). In that case, we would still want to check that the local polling station can confirm that the Acta that we're seeing is congruent with what they have

>The signatures on the Actas are digital, not ink.

Yes, each acta has a digital signature, gathered ahead of time. It is there to compare against the inked signatures signed by the members of the mesa, after confirmation that the sampled ballots converge toward the computer's results. The ballots are the source of truth here, not what the computer receipt says. And the link between the ballots and the receipt are the inked signatures (or fingerprints) of the members of the mesa.

>What's more likely, that the opposition forged tens of thousands of receipts in less than a day, or a dictator reported fake results to remain in power?

The opposition need not have been the one to hack the machines. A third party could have done that. And again, the opposition haven't released "forged" receipts, merely receipts that have not actually been certified. How they have obtained those receipts is an open question at this point.

>Receipts, mind you, copies of which are given to each witness from the top-three political parties, at any point now could have been called into question but not a single counter example has been shown.

90% of their receipts lack any inked certification from the presidents, secretaries, members, witnesses, or operators of the mesas on the ground. That should be garnering an enormous amount of skepticism from a crowd that is normally adamant about not trusting computers during elections.

I can't see what this has to do with the analysis. Obviously, the counts themselves are subject to error, but it takes even more manipulation of vote counting to express counts that work out to round percentages, given error, not less.

Meanwhile: it is definitely not the case that the norm is to simply make up the vote counts, or to back them out from percentages.

You can certainly tell stories about how this publication isn't a smoking gun for the election, but they will all involve some deliberate manipulation of the numbers; for instance, you can try to rationalize a story where someone took the original vote counts, worked the percentages out, rounded them (so far so good), then discarded the original counts (uhhhhh) and synthesized new ones (triple yikes).

(I'm sorry to keep hedging this way, but I have to say again: cards on the table I think Maduro is awful, but my interest in this story is in the statistical anomaly, not the politics --- this for me is like that time HN got all wound up believing Google was dragnetting searches for pressure cookers after the Marathon bombing, which flunked plausibility due to a similar statistical argument; I like a pressure cooker, sure, but my real thing there was the base rate fallacy).

I don't really think it's even a plausible mistake. Remember, to make the mistake, you have to retain the original raw vote total, but discard the original raw per-candidate totals, then recompute them from the rounded percentages and the retained original total. That doesn't make sense, for reasons having nothing to do with my level of trust in the election authority. It's several steps of extra effort for a result, read live on television by the election authority, that instantly destroys the credibility of the election.