Palm Beach Post
Chad theories continue to pile up
By Joel Engelhardt
What if they held the most talked-about election in American history and a year later, when no one seemed to care anymore, you thought you had found the smoking gun?
You came upon something so intuitively simple you have to wonder why no one mentioned it when every less-than-scintillating detail surfaced during those dreadful days of count and recount, dimpled chad and political spin?
Well, you'd put it on a Web site.
That's just what University of Iowa Associate Professor Douglas Jones did after he began tinkering with the famous Votomatic machines used in Florida and throughout the country to cast votes on Election Day.
And he's not alone. The dimpled chad of Palm Beach County election lore lives on interminably on the World Wide Web.
You can find microscopic chad closeups, Dimpled Chad coffee mugs and even the view of dimpled Chad from space. Yes, the African nation of Chad has a crater, the Aorounga impact crater, visible to astronauts.
Jones' eureka moment came when he took the back off a Votomatic punch-card voting machine and saw two braces -- two strips of glistening metal blocking the area where the chad are supposed to fall.
What if, he thought, chad got jammed behind those braces? Would that be enough to explain the rash of dimpled chad, those combustible bits of evidence that set Democrat against Republican for 36 days of presidential posturing after the 2000 election?
Chad in class
Jones, 50, a computer science professor who is chairman of the Iowa Board of Examiners for Voting Machines and Electronic Voting Systems, is teaching a course this year on computers in elections. Students -- and anyone else -- can read his definitions and pictures of hanging chad on his Web site, and view his attempts to dimple a chad and ferret out the importance of the braces.
Chad designers challenged skeptics to pack the machine with the tiny bits of perforated cardboard and vote. Many said their inability to produce a dimple proved that it couldn't be done by a serious voter.
A Votomatic is designed to hold 1 million chad bits, more than enough to handle even the most crowded precinct on Election Day, pioneer Votomatic engineer John Ahmann declared when he testified for George W. Bush in Leon County Circuit Court last December.
But thousands of voters left dimples anyway. Barring the tampering claims of conspiracy theorists, something must have gone wrong. But what?
Braces found, mapped
Jones thinks he found the answer when a Votomatic arrived in the mail unsolicited. As the veteran academic is wont to do, he tinkered. And he saw the braces.
He wondered if those braces could have blocked the free flow of chad. He mapped them, to determine which chads they would block. And he started voting. And counting.
Hours followed of punching, ballot-removing, ballot-replacing and punching anew -- "a couple of hours of real boredom," Jones said.
The punching paid off.
Jones found he could create an impenetrable jam with just 317 punches on a Votomatic. It took him longer, 668 punches, to create a logjam on the Data Punch machine, the king of under-votes in Palm Beach County.
That doesn't surprise Ahmann, who said the Votomatics are designed for just 100 to 125 votes per election. After repeated use, they must be cleaned -- although folding up and moving the machines between elections is enough to shake loose some chads, he said.
But Jones said he couldn't dislodge the chad jam, except by banging the machine repeatedly on his desk. Anyone voting directly above those jams would have been stymied. The stylus would press with enough force to remove the chad but it wouldn't go anywhere. It would hit the chad jam, dimple and remain in place. Some chads ripped off and contributed to the jam when he pulled the card out.
Could that, rather than frail voters, have been the cause of those 5,000-plus dimples seen in the reviews of Palm Beach County's ballots after the election?
Tony Enos, who's in charge of cleaning Palm Beach County's voting machines between elections, doesn't think so.
He takes the back off the Votomatic and vacuums up the loose chads. The machines were cleaned before the September primaries. But even with a primary, a runoff and the presidential election, it's doubtful the same area of the ballot would receive enough votes to create a Jones jam.
Chad jam suspected
But if the chad jams remained despite frequent cleanings over the years, the number of punches on Election Day would be irrelevant, Jones conjectured. And if that happened, his mapping of the braces showed, punching chads 3, 4 or 5 -- the locations of George W. Bush, Pat Buchanan and Al Gore on the Palm Beach County ballot -- could have contributed to a chad jam.
To top it off, the cleaning of the slimmer Data Punch model is less thorough. Enos takes off the small door to the battery-sized compartment in the back, the one marked "Shake well to remove chad" and, well, shakes well.
However, Jones notes, you have to remove the rest of the backing and do some major digging to dislodge a chad jam in a Data Punch. If that doesn't happen with regular cleaning there's no telling how blocked the machine may be on Election Day.
So it seems reasonable, Jones said, to analyze the test ballots punched by election workers in the hour before the polls opened to confirm his theory.
Those ballots, reviewed by The Palm Beach Post, show no propensities for dimples on the chads Jones would have expected -- the ones at or near braces.
How about the results of actual voting? The Palm Beach Post review of Palm Beach County's ballots casts doubt there too.
One of the candidates closest to the brace, Monica Moorehead, had a high level of dimpling as a percentage of her total votes. But the other, John Hagelin, did not.
Al Gore and George W. Bush had the most dimples -- more than 5,000 between them -- but the lowest rate of dimpling.
Jones was nonplussed. The braces may not explain the dimples but they underline the importance of proper maintenance of the machines, an issue that's still important since more than 25 percent of the nation's voters cast ballots with punch-cards.
Other factors that could contribute: the size and type of stylus and the quality of the ballot card, Ahmann said.
That research vein remains mostly untapped. For now, the strange case of Florida's dimples remains unsolved.
"At this point, the pattern of dimples doesn't look random, it looks systematic, but aside from that, I haven't got a decent hypothesis to explain what's going on," Jones said. "It just doesn't add up."