[Image courtesy of maybeimdreaming]
Today's Albuquerque Journal has a story about Election Day problems in Sandoval County, NM. There, the county moved to vote centers for the first time for the 2012 election but encountered delays of up to five hours and problems at the polls on November 6.
Looking back, it now appears the cause was some bad math based on some rosy assumptions:
A perfect-case scenario.
That's what the Sandoval County's Bureau of Elections director relied on when he decided on the quantity of equipment needed to handle voter turnout in Rio Rancho on Election Day.
Eddie Gutierrez admits he didn't allow for non-routine situations when doing the voting math, like the folks who needed provisional ballots because they had applied for but lost absentee ballots, or voters who recently moved to Rio Rancho but were registered somewhere else.
Consequently, even though almost 400 fewer voters showed up than expected, processing times were twice as long, leading to the delays.
Not everyone agrees that the estimates were the culprit; one pollworker wrote the Journal and claimed that a shortage of equipment was to blame.
Either way, however, this story is a real-world reminder that in planning for Election Day, election officials need to be cognizant that averages for voter processing are just that - averages - and build in room for the inevitable unpredictability that arises as the day progresses.
This kind of analysis isn't simple; it requires a better understanding of when voters arrive, what services they need and how pollworker performance will change with both familiarity and fatigue as the day wears on.
But it's a lot better than simply expecting everything to go perfectly. Perfection is the goal of election planning - but it can't be the plan itself.