Did voting machines trump exit polls?

* They have used uncertified code in prior elections and covered it up.
* They told one of their developers to "Print 'System tests passed'" on bootup in lieu of actually performing any tests.
* One of their main developers has a prior felony conviction.
* Their database contains two sets of voting books. A secret key combination enables the hidden book and the machine will report on it.
* etc, etc.

Democratic Underground is flooded right now and won't let unregistered users read or even register, but here's a quote from the main page:

This post and this post suggest that exit polls closely match reported results, except in those states that had electronic voting machines with no paper trail. This definitely deserves more scrutiny.

http://www.democraticunderground.com/

SoCalDemocrat has done a statistical analysis...

...on several swing states, and EVERY STATE that has EVoting but no paper trails has an unexplained advantage for Bush of around +5% when comparing exit polls to actual results.

In EVERY STATE that has paper audit trails on their EVoting, the exit poll results match the actual results reported within the margin of error.

So we have MATCHING RESULTS for exit polls vs. voting with audits

vs.

A 5% unexplained advantage for Bush without audits.

Technology is the answer to all our problems!

How bad is this going to be for the States? In the debates, you saw the man unexpurgiated; he may be a strong campaigner, but he is not a strong president. It seems that all he did - or at least most - in his first term was geared towards getting re-elected, and pleasing the base.

Hopefully, he will be able to think to the bigger picture at last... but I somehow doubt it; the neo-conservatives will see this slender win as a vindication. It is hard however to see things being *quite* as bad... but, well, depends on how you look at it. First genuine signs will be the administration he appoints...

And then, magically ... though they were neck and neck in the polls to the last day ... bush pulls a huge win out of thin air.

Hum.

All the provisional and absentee votes get counted regardless of the concession, which is entirely symbolic and not at all legally binding.

"Even when complete, an exit poll still has the same random variation as any other survey. NEP says typical state exit polls will have a sampling error when complete of +/- 4% at a 95% confidence level, and +/- 3% for the national exit poll. Even if comparable to the final numbers – which they are decidedly not – the mid-day leaked numbers would have much greater error, perhaps +/- 7% or more."

In other words, it's entirely possible that the 51-49 spreads in the exit polls that people so excited for Kerry's chances yesterday were just noise.

If this really happened, Kerry's concession will be seen in history as the most stupidly premature concession since Carter.

But I give Kerry a little more credit than that, and that he probably knew something that we don't.

Actually, according to this Slate article, Ohio's votes will still all be counted. But if Kerry has misgivings about (1) pursuing an Electoral College victory in the face of a popular vote defeat, or (2) presiding over so deeply divided a country, I don't blame him.

Lefty Web sites are buzzing with a supposed "statistical analysis" that they say proves that Republicans stole the election by rigging paperless touch screen machines. According to the analysis -- the link is here, but the site's down at the moment due to high traffic -- performed by a Democratic Underground regular who goes by the handle SoCalDemocrat, states that use electronic machines were all showing strong Kerry support in exit polls, but when the results came in, the states went to Bush. The polls (which are based on interviews with voters as they leave the polls) indicate that voters in the state really voted for Kerry, the lefties say; the machines distorted or changed their votes. "Maybe Dubayah believes God will see him through this, but it's going to take more than blind faith to pull the wool over the data and the facts," SoCalDemocrat writes.

But as we see it, SoCalDemocrat's evidence is quite thin. For one thing, he appears to be wrong on the facts. While he's correct that exit polls showed a Kerry victory in many states that actually went to Bush, this didn't only happen in states that use paperless touch screen machines.

The most obvious example here is Nevada, the only state in the nation to use what many computer scientists consider to be the safest touch screen machines -- machines that print a paper ballot that is reviewed by the voter as each vote is cast, a so-called voter-verified paper trail. In Nevada, the last exit polls showed Kerry leading Bush by 49 to 48 percent, with 1 percent for Ralph Nader. The actual result was a win for Bush by 51 to 48 percent.

And even in states that do use paperless touch-screen machines, it's not clear that Bush made his gains in touch-screen areas of the states, rather than regions that use other machines. For instance, in Florida, it's the state's large South Florida counties that use paperless touch screens. But Bush did worse in these regions in 2004 than in 2000. In the 2000 race in Miami-Dade, Bush got about 47 percent of the two-party share of the vote, while Al Gore received 53 percent; this year, Bush only got 46 percent of the two-party vote there, while Kerry got 54 percent. What this means is that in the move from punch-card machines (which, as everyone remembers, Miami-Dade used in 2000) to paperless touch screens, Bush actually did worse, not better. At the same time, the president gained in Orange County, Florida. In 2000, Gore beat Bush in Orange County, whose largest city is Orlando; this year, Kerry lost to Bush there. And Bush didn't need rigged machines to do it -- Orlando uses paper-based optical scan voting machines, which computer scientists consider more reliable than the touch screen systems.

Even if unfounded, the Democratic Underground set's suspicion is understandable. The exit polls were odd. And who can forget the infamous pledge of Wally O'Dell, CEO of touch-screen machine vendor Diebold, to deliver Ohio's electoral votes to Bush this year? (Ohio, though, used no paperless touch screens.) One way to make sure future presidential races are above such suspicion would be to add paper trails to all the paperless machines.

http://www.salon.com/politics/war_room/archive.html?blog=/politics/war_room/2004/11/03/machine_fraud/index.html

ibuprofen time