On election night, Spitale was ahead of Hutnik by a scant two votes.
Then the canvassing began. That's a review, precinct by precinct, of the votes cast in each race. It involves checking the flash drive in each voting machine, a comparison of the votes ballots cast on the machine against who actually checked in on the electronic poll book, a review of the mail-in ballots cast, checking the provisional ballots and "remakes" of mail-in ballots rejected by the scanner.
As of Wednesday, Hutnik had picked up two votes and the race was in a dead heat. As of Thursday night, Hutnik is ahead by six. But it's still too soon to pop open the champagne.
One of my readers complains we all should have known the winners on election night. "[T]his extended counting is not what has been the history of voting in the USA," he observes. "This only leads to a further eroding of faith in the system."
This reader is wrong. The election results tabulated on election night or the following day are unofficial and stay that way until the votes are canvassed and results are certified by the Elections Comm'n. It always takes a week or a little more to conclude this canvass.
As I've mentioned in a previous post, 516 of the 16,010 mail-in ballots received (3.2%) were rejected by the scanner. Unlike a Presidential contest, the ballot in a municipal contest is much longer and this presents a problem when unfolded for scanning. A paper tear or poor unfolding sometimes renders the ballot unreadable, and that's when elections officials must complete "remakes."
You could reject all 516 of the ballots rejected by the scanner, but that would disenfranchise the voter.
In addition to the remakes, provisional ballots must be reviewed. Is the voter registered? Does he actually belong to the Democratic or Republican party? Should it count in the precinct where the provisional was cast?
These questions must be answered by canvassers before deciding to count or reject a provisional.
The canvass will also disclose pollworker error, especially now that the county is using electronic poll books.
When the polls close, the elections judge removes a flash drive from the voting machine. This is encrypted and is taken from the precinct and handed to county officials to tabulate the results. In the Forks and Palmer Tp race, the elections judge in one of the Forks Tp districts turned in two flash drives corresponding to two voting machines in that precinct. One of the flash drives showed that no votes had been cast at all. When contacted, the elections judge told officials no one had used that machine. This information was incorrect.
During the canvass, the electronic pollbook showed that 100 more voters had checked in to vote than appeared in the results. Officials then checked the machine that supposedly was never used. The missing votes were there and were tabulated. The elections judge removed the flash drive before closing the polls.
It's a long day.
This is why you do a canvass.
Without the electronic pollbooks advocated by Voter Registrar Amy Cozze and County Administrator Charles Dertinger, this error would never have been discovered.
Dertinger has overseen the selection of three separate voting systems during his time as Elections Commissioner, County Council member and Administrator. He has been through close races like this one.
In an interview yesterday, Dertinger said that the entire district will be audited today. Both campaigns have been informed. If the final tally is a tie, Hutnik and Spitale will draw straws.
Marc Crisafulli, a Wilson Borough police officer, has sewed up the Republican side.