(This is a letter that I’m sending to newspapers in Minnesota regarding the Senate contest there; it’s an adaptation of a letter that I wrote in 2000 regarding the Florida presidential recount. It’s kinda sad that the basic concept doesn’t seem to be catching hold: there should be a margin of error declared for any given election, and if the election result is smaller than that margin, it should be declared a tie and handled accordingly.)
“Precision” is the key word to understanding Minnesota’s post-election brouhaha. Precision is not the same as accuracy, but rather refers to the kind of measurement that one can expect from a given instrument. Your garden-variety twelve-inch ruler, for example, is not precise enough to measure the diameter of a blood cell, but an electron microscope could do the job easily.
Our existing election systems are very imprecise instruments for determining the will of the people. Machines malfunction, humans make mistakes, and parties can influence outcomes in many ways (some legal, some not). This imprecision is the subject of a certain amount of humor on the one hand and cynical resignation on the other, but is usually insignificant – most elections yield margins of victory that easily exceed the margins of error. (I’m told a three-percent margin of error is fairly standard for elections.)
In the Senate recount, the margin is orders of magnitude smaller than the level of precision that the election system can be expected to yield. Imagine trying to determine the winner in a footrace where yeah, sure, someone got across first, but the actual difference was a matter of millimeters and milliseconds and you’re eyeballing the result from 500 yards away.
In other words, it’s a tie. Rather than quibbling over recount procedures, the parties involved should simply settle on a tie-breaking mechanism, if the law does not provide for one already – whether it’s a complete redo or a mere coin toss – and let the Senator get to work.