Thursday, July 2, 2009

Chuck Marohn on The Coleman-Franken Recount

On the air we’ve discussed the Coleman/Franken recount process on many occasions. Those conversations include frequent suggestions that we develop a different process for elections, adopting a new approach like an “instant runoff” or “top-two runoff” to determine the victor. My contention has been that neither of those approaches would have helped us avoid the current dilemma, which is that the election was too close for us to determine a winner.

Before Franken supporters start hurling curses my way, understand that I am not questioning the legitimacy of his victory. This is our system and, going through the process we have established, he won. The same goes for George Bush in Florida in 2000. In both instances, the elections were too close to call and recognizing that would have allowed us to pursue a clearer result.

What I am talking about is math. This may matter to me more because of my days in engineering school, where they drilled the concept of “significant digits” into my head. The idea of “significant digits” goes something like this:

A road lane is 12 feet wide. If we have two lanes, that makes the surface 24 feet wide. It is not, as our calculators like to tell us, 24.000 feet wide. In other words, our ability to measure in that instance is in feet, not the thousandths of a foot.

If we wanted to measure the distance between my home north of Brainerd and the KAXE studio in Grand Rapids, we might get out a map and count the miles as listed in red. That would be around 80 miles. We could also go to Google Maps, which says it is 81.6 miles. We could get a surveyor to run a loop between our locations, and if they were extremely meticulous they might be able to say that it is 81.57 miles. We could get a highly accurate laser measuring device and that might be able to tell us the distance is 81.57234 miles.

Our method of counting ballots is somewhere between Google Maps and our meticulous surveyor in terms of precision. That is actually quite impressive for a human endeavor involving literally hundreds of unpaid volunteers. And it is a good system. In nearly every election held, this method works very well. I would not change it today, largely because the extremely high cost to achieve laser-measuring precision would be unjustified in nearly every instance. Why would we spend billions on a laser measuring the distance to Grand Rapids when Google Maps will get us the precision we need at no cost?

Consider that the final official count was 1,212,206 for Coleman and 1,212,431 for Franken, a difference of 225 votes. That is a difference of 0.0093%. We all saw how people marked their ballots in crazy ways, ballots were (I believe innocently) misplaced, recounts varied from original counts, etc…. Understand that if we did a third count, we would get a different number. That number would be 1,212,xxx, with the xxx being where our precision fails us.

So what do we do about elections like Coleman/Franken 2008, where the result is too close to accurately measure? We simply rerun them. We agree ahead of time that our system has limitations, we have a reasonable degree of accuracy that we can have confidence in, we establish that degree of accuracy in a bipartisan way prior to the election and, if the election results wind up to be too close to call, the remedy will be a do-over.

With that approach, we would have saved ourselves seven months, millions of dollars and we would not have the acrimony we have today, where partisans bristle over the legitimacy of one outcome or another. To me, that is not only Minnesota Nice, it is Minnesota Smart.

Chuck Marohn is the President of The Community Growth Institute in Brainerd a regular commentator on KAXE's Morning Show

1 comment:

Anonymous said...

Chuck - I appreciate your optimistic, non-cynical viewpoint. Thanks for writing this!