[speedgibson] Speed Gibson: Within the Margin of Error

[Email subscription to blog articles]

Posted by Speed Gibson:
Within the Margin of Error
http://speedgibson.powerblogs.com/posts/1231026418.shtml


   Yesterday, I posted that every election has a margin of error, if for
   no other reason that it is conducted by humans. I also contend that
   the final margin of victory in the 2008 Minnesota Senate race is
   likely to be well within that margin of error.
   Some are talking up a run-off system for Minnesota, like Georgia used,
   where the top two face off a month later if no one gets an outright
   majority, 50 percent plus 1. But if it's 49 to 44 with a smattering of
   third parties, this is pointless, maybe even wrong, just a protracted
   version of "instant runoff voting." The voters have made a clear
   choice.
   A run-off election does make sense if the margin of victory is
   comparable to the margin of error. The outcome truly is in doubt. So
   fine, let's say we say the threshold is 0.1% - one in a thousand for a
   state-wide race, which would be about 2,400 votes in the Franken -
   Coleman race. Rather than watch this endless, embarrassing
   hair-splitting, we would have voted again four weeks later in December
   and all but certainly had a winner by December 3rd.
   There are some details to work out, like financing the extra
   campaigning and avoiding frivolous demands for recounts or re-votes if
   the difference is 2,401 votes. Maybe the Governor should ultimately
   have the power in the gray areas.
   Right now, this looks like a much better solution than the mess we're
   in now, no matter who wins.