Slashdot Mirror
The South Carolina Primary and Voting Machine Fraud
cSeattleGameboy writes "South Carolina sure knows how to pick 'em. Alvin Greene is a broke, unemployed guy who is facing a felony obscenity charge. He made no campaign appearances and raised no money, but he is the brand new Democratic Senate nominee from South Carolina. Tom Schaller at FiveThirtyEight.com does a detailed analysis of how a guy like this wins a primary race, and many of the signs point to voting machine fraud. There seem to have been irregularities on all sides. 'Dr. Mebane performed second-digit Benford's law tests on the precinct returns from the Senate race. ... If votes are added or subtracted from a candidate's total, possibly due to error or fraud, Mebane's test will detect a deviation from this distribution. Results... showed that Rawl's Election Day vote totals depart from the expected distribution at 90% confidence. In other words, the observed vote pattern for Rawl could be expected to occur only about 10% of the time by chance. ... An unusual, non-random pattern in the precinct-level results suggests tampering, or at least machine malfunction, perhaps at the highest level. And Mebane is perhaps the leading expert on this very subject. Along with the anomalies between absentee ballot v. election day ballots..., something smells here.' Techdirt.com points out that South Carolina uses ES&S voting machines, which have had strings of problems before; and they have no audit trail."
The problem I see with this being some kind of fraud - is what kind of idiot would choose, as their puppet, this person. There must be hundreds of people who, in return for a hefty sum, would do your bidding, all while looking a whole lot more respectable. This looks to me more like a case of people voting for the 'other guy' without actually knowing who the other guy is.
Control is an illusion, order our comforting lie. From chaos, through chaos, into chaos we fly
In other words, the observed vote pattern is something you will expect to see a lot when checking various machines and various elections over time.
A 10% chance of a pattern in no way suggests any tampering. Perhaps together with other evidence it is a tiny indicator. It's hard to take any article seriously that doesn't examine the facts properly. Now if the chance was one in a million it might suggest tampering, but one in 10? I'll put it bluntly: Give me a fooking break
The P value of this test is 0.1, pretty much all research I read demands a P value of 0.05 to justify a hypothesis. How many elections are there in the USA every year? By this standard even if all of them were not tampered with and totally legitimate 1/10th of them would be found to have been tampered with. That's a large percentage of false positives for such a serious accusation.
Basically, bullshit, either do better research to get a lower P value or stop drawing such spurious conclusions.
The kind of "idiot" who wants a Democratic candidate that's sure to lose. The people who are alleging fraud are claiming that this is a scheme to ensure that the Republican incumbent is re-elected.
People do weird things some times. Why did Nixon commit felonies in the 1972 race against McGovern (and thereby destroy his Presidency) when it was obvious to almost everyone that McGovern had no chance of winning anyways?
The people who are alleging fraud are claiming that this is a scheme to ensure that the Republican incumbent is re-elected.
It would be a silly scheme though considering that this is a safe Republican seat anyway. Ok if we are going to be throwing conspiracy theories around, how do you know that this is not a scheme by the Democrats to create a scandal that they could blame on the Republicans?
Negative moral value of force outweighs the positive value of good intentions.
What are you talking about? Nixon committed no crimes at all related to the election. Some of his subordinates who were so used to committing felonies on a daily basis did it without his knowledge. What he did that was a felony was to hear their confession and then, rather than turning them in for their felonies, violated a large number of laws to cover up the stupid acts. I would bet that if G. Gordon Liddy had asked first, Nixon would have told him to not do it. It was senseless, but Liddy is a "patriot" in that anyone that he thinks will do something that is bad for the US deserves death, or at least a listening device so he can spy on them.
Learn to love Alaska
It's worth noting that in some precincts, Mr. Greene received more votes than were cast. As in, he got 115% of the votes. In others, he won the election day votes by 20 points but lost the absentee votes by 60. There are major, major discrepancies in vote tallies in this election. You can quibble about confidence intervals and statistics all you want, but it won't change the fact that *something* went wrong here. While it's probably not malicious, it absolutely should be investigated.
My other sig is clever.
Would you say to meteorologist that 9 out of 10 of hurricanes like this one were destructive, "That's meaningless unless it's 19 out of 20"?
The threshold for statistical significance is an arbitrary convention, not some ironclad law that lets you ignore evidence. As a guideline it is more appropriate in some circumstances than in others. Something does not stop being evidence simply because it does not reach that threshold. I read scholarly papers all the time that say "while X does not achieve the threshold of significance, it is suggestive and worthy of more research." When there is other evidence to support it, such a result can be valuable. And there is such evidence: this calculation was done precisely because the election looks fishy.
You have it exactly wrong when you say "that's a large percentage of false positives for such a serious accusation." The election process is not innocent until proven guilty. We apply the presumption of innocence to human beings. An election is treated in the opposite way. It is not enough for it to be fair: it must be seen to be fair. It must be must be demonstrably legitimate. We do not let suspicious elections slide simply because the accusation is "serious." On the contrary, that is why we investigate them. This needs to be investigated precisely because of its seriousness.
Voters in S.C. are allowed to vote for the Democratic or Republican primary regardless of party affiliation. One of the theories was that Republicans crossed and voted in the Democrat primary to try to shaft them with a bad candidate. But if you look at the election results, you'll see that 424,893 people voted for the Republican primary while 197,380 voted for the Democrat primary. The electorate there is so strongly Republican that if 30k Republicans crossed over to give Greene his minimum 100k vs 70k margin of victory, the Democrats are looking at having to overcome a 2.7:1 margin of voter registrations against them to win, instead of "merely" 2.1:1. If you assume Greene is a nobody and should've gotten 10k votes max, then that means over half the people who voted in the Democrat primary were Republicans, and so the Democrats would need to overcome a 6.4:1 margin to win.
All in all, none of this makes any sense. There's no motive on either side. Why would Republicans poison a Democrat primary for a safely Republican seat? Why would Democrats not want to put forth the best candidate? Something does smell, but the most plausible explanation is simple voting machine tallying error with no nefarious purpose behind it.
I had an in-depth discussion with several people years ago about doing electronic voting. That was before the whole electronic voting fiasco started.
On the site that I was the Sr. SysAdmin for, and I did a good bit of programming for, it had a voting system. The original programmer couldn't handle the number of votes coming in, so he randomly took 1 in 10 votes and counted it. Sampling is fine and dandy, but in my world I like completely accurate numbers. The final system stayed in place for years. It very typically maintained millions of votes for thousands of items. It had some primitive components, but that was by design. The votes were stored in flat files, as it would bog down the database server trying to insert the votes in real time. The end user submitted their vote, and it was counted immediately (like milliseconds). The entire vote database was retabulated every 15 minutes. Two people had root access to the server, and it required root access to be able to view the voting information.
In that system, it wasn't a simple "pick a candidate". It was a scoring system (1 to 5) for the item being voted on. For years, one lonely dual 400Mhz machine with 512Mb RAM handled the tabulation and reporting. We did on occasion have someone question the results. It was usually on something that they were responsible for. "Why did my score drop from 4.5 to 3 in a hour?" It was simply that as the voting numbers rolled in, it adjusted their score. The preliminary numbers were favorable, but subsequent votes weren't so favorable. I could generate reports off of it for that specific item (it took about 10 seconds), where you could see the votes, and how it adjusted the score.
After a while, we had more robust equipment, and I began storing the voting information in a database. A replica of the database was used for tabulation, so the tabulation machine didn't slow down the vote recording process. That, and a better tabulation machine, brought processing tens of millions of votes down from 5 minutes to less than 1 minute.
So we talked about what else we could do with such a system. Real political voting could be managed in such a way. We ran into the same problems that are being questioned with the voting machines in use. Only two people with no interest in the outcome of the voting had access to the system. To manipulate the votes would be a very cumbersome task (by design). What if we did the voting for real politics.
Problem 1) How would we prove to the voting public that the people running the servers had absolutely no interest in manipulating the votes. There's no way to prove that.
Problem 2) How could we provide for anonymity of the voters. We stored the IP and identifying information with the votes, so we could eliminate voting fraud. Those who voted multiple times on the same item were categorically eliminated from all voting. Their records were stored, but ignored for tabulation. Real political voting requires anonymity. We could provide pseudo-anonymity by storing an ID number with the vote, that would associate with the voters registration. It would then be traceable back to the voter, which is illegal/immoral/just bad. For our application, no one cared.
Problem 3) How would the general public know that our tabulation program gave an honest result. When the votes don't go your way, people assume there had been some tampering with the results. Really, it would have been easy to lower votes ($vote = $vote -1), and make someone score poorly. Who would you trust more, a couple computer experts, or the government. I know I don't trust the later, but the general voting public wouldn't know if we were trustworthy. If presented with $100 million in cash, who's to say we wouldn't subtly adjust the results in favor of the group who paid us. Again, I believe in honesty in voting, but the general public doesn't know I won't accep
Serious? Seriousness is well above my pay grade.
There were threads about him with 1000+ Diggs *** AFTER *** the election, due to an interview with Keith Olberman (AFTER the election) where he appears to be several bricks short of a load. What does his becoming known after the election have to do with Alvin Green being unknown prior to voting in SC? Illogical argument.
That makes no sense. Even the left-leaning fivethirtyeight blog listed the South Carolina Senate seat as safely Republican back in late April, with a 95+% chance to be won by the Republican candidate.
According to TFA, these voting machines have a large number of problems and no audit trail. Who's to say this wasn't just a fuckup, rather than deliberate malice on anyone's part?
If this shows anything, though, it's the need for a non-electronic audit trail. I've often had people find it odd, given that I'm a programmer, that I'm so against purely electronic voting. I don't find it odd-I know exactly how easy it is to manipulate data on a large scale, even data that's supposedly secure and tamper-resistant. It's a whole lot harder to tamper with thousands or millions of paper ballots than it is to tamper with thousands or millions of electronic records.
That doesn't mean electronics have no place. An electronically generated human-readable ballot would be fine. In that case, the speed and reduced human error of electronic voting could be realized, but the voter would still have the ability to verify their choices after printing, and if wrong, go to an election judge, say "I didn't intend to vote this way", and have their ballot scrapped and recast. Backup paper systems should always be available at every precinct in case of a total failure of the machines, electrical failure, or just people who are not comfortable using them.
Having that type of mechanism in place would prevent exactly this type of scenario. It would allow for the result either to be overturned, or to say with certainty that, while unlikely, it is indeed the outcome.
To fight the war on terror, stop being afraid.
Stop the snow job. He's a military intelligence vet and a man with a Poly Sci degree. So what if he's unemployed after he leaves the service? It's tough out there. The ABC interview was a butchering.
Anyone seen my low uid? last seen 10 years ago while panning the #@$# out of Taco's 'web based discussion system'
What if it's actually the Republicans making a scheme to make us think that it's the Democrats trying to make us think it's the Republicans trying to make us think it's the Democrat candidate?
Safe Republican seat? Yes and no.
He won his seat in 2004 by around 9%.
Back in December of 2009, he had a 9% lead against a generic Democrat. That's not a huge lead against a completely unnamed opponent. AND there a lot of people here in South Carolina who really like this whole "anti-incumbent" trend. (Enough to make a difference? Probably not. But enough to scare DeMint a bit.)
Alvin Greene ran no advertisements. He didn't attend the Democratic Party Convention in South Carolina. He had practically no name recognition when compared to his opponent, Vic Rawl, who at least was a state legislator. He was able to pay the filing fee for running for the Democratic primary with a personal check (the filing fee is over 10 grand), but he's poor enough to qualify for a public defender for the felony obscenity charge against him. (Also, please note, that the law being used against him is one that is generally only used for people who show bestiality, extremely violent porn, etc., not the simple hetero porn that Greene allegedly showed someone. So that too comes across as a bit hinky.)
According to the FEC, at least through May 19, DeMint had around $3.5 million in cash on hand for this election cycle. Greene has $0.
Now, as to your last question, could this be the Democrats up to something instead of the Republicans up to something? I don't know. But the whole damn thing smells to high heaven.
Mr. Hu is not a ninja.
His name was the first one on the ballot. Many people just pick the top one. No scandal, human nature, get over it.
It was a silly scheme, but from what I have read this is business as usual in SC politics. Republican operatives sometimes pay entry fees for black candidates just to "stir the pot" of racial division among the Democrats during the primaries so that blacks will be less likely to vote in the general election.
I have also read that this is often not much more than a practical joke, especially in this case when the candidate did nothing but pay the entry fee and did not even have campaign signs up in his own yard. I think the Republicans really don't want these candidates to win because it would bring national attention to the way SC politics work, and they were probably just as shocked that Greene won as everybody else was.
Back in 1990, Rod Shealy used this exact same tactic in a Lt. Governor race in SC. He recruited a homeless black guy with a criminal conviction in an attempt to take out the Democratic frontrunner, so his sister (a Republican) could win. It was a crass attempt to play on the racial prejudices of SC (both for blacks in the Democratic Party and against blacks among the general populace) to get his sister elected. He almost succeeded to. And he is still working in SC Republican politics (most recently in the Bauer gubernatorial campaign).
All of you who are saying this is a preposterous idea have obviously never been involved in SC politics. This isn't even a particularly nasty tactic by SC standards.
SJW: Someone who has run out of real oppression, and has to fake it.
I've often had people find it odd, given that I'm a programmer, that I'm so against purely electronic voting.
It is amazing how often people find that odd. But don't just tell them you're against it...tell them pretty much the entire industry is against them, because computers do exactly what you tell them to do, including lie, and then they can lie about being told to lie.
People need to hear this more from people they regard as knowledgeable about computers. Over and over. Computers lie if told to do so. This is not detectable because they'll just lie about their lying to the people checking them.
If corporations are people, aren't stockholders guilty of slavery?