Every election the editorial boards demand that we change the ballot deadline to received by election day, and every election I have to explain—using data and math and stuff—that they don’t know what the fuck they are talking about. It is not the ballot deadline that is the bottleneck. It is the ballot processing:
Take, for example, King County Elections (KCE), which counted 556,083 ballots on election night. That was only marginally more than the 521,786 it had received by the Friday before the election. In fact, it took three more days of counting just for KCE to catch up with the 749,097 ballots it already had on hand election night, let alone begin to work through the 147,744 ballots that arrived the next day. Statewide, Washington tallied on election night only about two-thirds of the ballots it had on hand.
So how could moving the ballot deadline speed up the results? It can’t.
Sure, the piece quoted above is a couple years old, but I’ve been tracking ballot statistics since 2009, and it has proven true year after year: the only way to speed up ballot counting is to spend a lot more money counting them. That’s what Oregon does—we stop on Tuesday at 8pm, while they count ballots 24 hours. It is simply a fact. And one the editorial boards have never bothered to refute. Because they don’t know fuck about ballot processing. Whereas I know this elections stuff inside out. Seriously.
Furthermore, even if moving the ballot deadline would speed up counting by a day or two—and it won’t—to what end? The overwhelming majority of races can be called on election night. I fail to see the harm in being asked to wait a day or two longer to learn the final outcome of a handful of others.
What the editorial boards are pitching is a solution in search of problem—and a solution that simply won’t work.
