Monday, April 13, 2009

Number crunching

According to the latest state board of elections numbers, Democrat Scott Murphy is leading the special Congressional election vote count by 25 votes over Republican Jim Tedisco out of over 155,000 cast. The numbers may have changed yet again by the time you've read this.

Apparently, the Tedisco people are upset that ballots from soldiers in war zones might not arrive in time to be counted. Tedisco insists idea that American citizens might be disenfranchised on a technicality is 'shameful' and an 'embarrassment.' His hypocrisy is exactly what you'd expect from someone with Tedisco's used car salesman-like reputation.

But there's something that's bothered me about this whole recount thing. We've seen probably over a dozen different vote counts publicized. And this is just from the relatively non-controversial voting machines. There is human error involved and that election night voting totals processed by tired poll workers might change as their work is re-checked is understandable. I get this.

But common sense suggests that AT SOME POINT, officials should be able to check the voting machines twice in a row and get the same totals. If you tried balancing your checkbook 10 different times and got 10 different results, wouldn't you keep doing it until you were able to replicate one of them? If the totals keep changing from static machines, then how can anyone be sure that the final total is the right one?

In a close election, the count from the voting machines is supposed to be the easy part. If "results" change this often from unchanging voting machines, what's going to happen during the counting of absentee ballots, when the two campaigns are going to use the flimsiest excuses to try to get opposition votes disqualified?

2 comments:

Mark Wilson said...

Excellent point re: machine tallies.

As for absentee results, you don't have to get far in school before you are introduced to the concept of scientific error, or the probable certainty of a subjectively attained result. It's a concept which does not sit well with our culture's binary mindset, and we continue to believe--despite all reason to the contrary--in the infallibility of numbers.

The news media is not much help in this regard, ignoring margins of error when reporting polling data (ahead=ahead. period.).

If the 20th CD does come down to Tedisco's ineligibility to cast a vote in the race, then the race is most certainly within the margin of error (i.e. greater than 1/155,000). and ought to be re-run.

Mark Wilson said...

Re: Tedisco's request

What I find peculiar about this request is that the March 30 deadline for postmarking the absentee ballot would not be changed. By implication Tedisco doesn't trust the US military to get its mail back home within a two-week span. Now perhaps in a situation of open conflict where deliveries are delayed, suspended and lost would this be an issue. This combat zone by comparison is far more static, with supply and communication lines wide open. How many votes could we possibly be talking about?