Proofiness - Charles Seife [57]
The press, always fond of calling a horse race, were happy to report whatever Potemkin numbers fell into their hands. Unsophisticated reporters simply reported the official recount tallies without even trying to account for the challenged ballots. (“The Republican incumbent held a slight edge, with a 192-vote lead over Democrat Franken,” reported the Associated Press.) Everyone else whipped out their crystal balls or attempted to read turkey entrails or used whatever other methods they could to generate their own projections. Poll watcher and sports statistician Nate Silver used regression analyses to predict that Franken would win by twenty-seven votes. Two weeks later, he reversed himself, projecting that Coleman would win the recount.
Of course, all of these numbers were completely worthless. The frivolous challenges ensured that the official tallies were devoid of meaning. There was no way anyone could know whether Franken or Coleman was ahead until the canvassing board finally started wading through the challenged ballots and discarding all the frivolous challenges. Yet nobody was willing to wait. Pundits and journalists created Potemkin numbers to titillate their audiences—those phony figures allowed them to give an answer to the unanswerable question of who was winning the race. More disturbingly, the campaigns used Potemkin numbers to foster a belief that their candidate was winning—and that their opponent was trying to steal the election. It was naked proofiness, manufactured to sway public opinion.
Not every challenge had a cynical motive. While the vast majority were frivolous, plenty of them were perfectly reasonable. Scores of ballots were challenged for valid reasons: double votes, illegible votes, bubbling-in errors, identifying marks that revealed who cast a particular ballot (one voter even got a ballot notarized).57 Every possible ballot pathology—every single way a voter can conceivably mess up a ballot—was on display.
At times the fallout was stupendously absurd, as in the lizard people fiasco. The race was so close that the judges fought a hard battle for a full five minutes before coming to their decision. Mr. Lizard People might indeed be a real person, therefore the ballot—which had been counted for the Democrat—had to be discarded. Lizard People had eaten a vote for Franken.58
The lizard people ballot was just one of the hundreds of perversities that Minnesota voters created for the amusement and confusion of the canvassing board. There were flying spaghetti monsters and Mickey Mice, ballots that had been initialed, stamped, and scrawled on in every conceivable way. There were dark circles that were nowhere near any blank ovals intended to receive the votes. A handful of voting marks had been placed with almost laserlike precision directly midway between the Coleman and Franken ovals. Mansky was right: most Minnesotans do know how to fill out their ballots properly. However, the ones who didn’t sure failed spectacularly.
In a population of a few million voters, it’s almost guaranteed that a few hundred will cast pathological ballots—ones where the voter’s intent is obscure or even impossible to determine. (Badly designed ballots can make this number even worse.) This is yet another source of error, no matter how foolproof your ballots are and how strict and clear your election laws and regulations