Fairvote.org is currently undergoing an upgrade, and some features may not be working as usual. We apologize for any inconvenience, and expect to be back at full capacity soon.

Instant Runoff Tallies Start in Court of Appeals & Superior Court Races

Released November 29, 2010

Media Advisory

November 29, 2010

Contacts 
Rob Richie  (301) 270-4616  [email protected] 
Toby Rowe  (301) 270-4616  [email protected]

Instant Runoff Tallies Start in Court of Appeals & Superior Court Race

On November 2, North Carolina voters recorded their preferences in four elections with instant runoff voting (IRV) to fill judicial vacancies in a single trip to the polls. Every voter had an opportunity to cast a ranked choice ballot in the statewide race to fill Judge James Wynn’s former seat on the state Court of Appeals, while three counties held instant runoff elections with three candidates seeking to fill Superior Court vacancies. Voters handled their new ballot well, as voter turnout was higher in the Wynn vacancy election than in most of the other Court of Appeals races.

On November 23rd, election officials completed the statewide canvass of ballots. Today, November 29th, they are turning to the process of tabulating votes and determining the winner of the instant runoff elections. “The instant runoff this year is not being tallied instantly, of course, because North Carolina does not yet have the software necessary to administer the tally automatically, unlike some other jurisdictions,” commented FairVote’s executive director Rob Richie. “But from a voter’s perspective, the runoff was indeed ‘instant’ – they were able to do their job in one trip to the polls.”

The Wynn vacancy election will be determined by an instant runoff, as will the Superior Court race in Cumberland County. Election officials will use two IRV tabulation processes depending on the voting technology used by different jurisdictions. As a reminder, here are the principles governing the count:

  • The instant runoff was triggered only if no candidate won an initial majority (more than 50%) of the vote.
  • To win, a candidate must have finished in the top two in his or her race. That means 11 candidates have been eliminated in the Court of Appeals race and one candidate in Cumberland County’s Superior Court race.
  • No ballot ever counts for more than one candidate at a time. That means that if a voter indicated one of the two frontrunners as a 1st choice, that ballot will count for that 1st choice only.
  • A voter’s rankings are used as “backups.” If a voter’s 1st choice has been defeated, then that ballot will count for the voter’s 2nd choice candidate if the 2nd choice is one of the two frontrunners. If that 2nd choice also has been eliminated, then that ballot will count for the voter’s 3rd choice if the 3rd choice is one of the two frontrunners. After the instant runoff, the candidate with the most votes is declared the winner.

In counties using optical scanning machines (in which votes are registered by scanning a paper ballot), IRV ballots will be sorted into four groups before they are tallied:

 

  1. The first group will include ballots ranking one of the top two candidates as a 1st choice. As indicated by the 1st choice ranking totals published after the first round of counting, the top two candidates in the Court of Appeals race are Cressie Thigpen (with 395,220 ballots) and Doug McCullough (with 295,619). These two candidates will advance to the instant runoff, while the others will be eliminated.
  2. The second group will consist of ballots that ranked an eliminated candidate as a 1st choice and either Thigpen or McCullough as a 2nd choice. These ballots will be run through the machines, and those votes will be added to the totals of Thigpen and McCullough, according to which candidate is listed as a 2nd choice. (Note: these ballots do not count “twice” or count as “two votes.” Once the candidate that they rank 1st is eliminated, they are counted instead for the candidate ranked 2nd. Never do they count for more than one candidate at a time.)
  3. The third group is composed of ballots that ranked eliminated candidates as 1st and 2nd, but ranked either Thigpen or McCullough as 3rd. These ballots will also be counted by machine and those votes added to the totals of the two frontrunners. (Again, these ballots are not counted as multiple votes.)
  4. The final group is composed of ballots that do not include either Thigpen or McCullough among their three rankings. Since none of the candidates selected on these ballots remains in the race, the ballots will not play a role in the instant runoff. Those voters at least had three chances to vote for one of the two frontrunners rather than only one chance, as was true of North Carolina’s prior system for these judicial vacancy elections.

 

Candidates’ totals from the counties that use optical scanning will be added to the totals from counties that use touch screen voting in order to determine the revised statewide totals and the final winner.

In counties with touch screen systems, the count will be completed electronically, although verified on paper. Data from the touch screen machines will be transferred to computer spreadsheets. Once the data has been transferred to a spreadsheet, the logic governing the rankings will be the same as that of the tabulation process outlined above: each voter’s ballot will count for whichever frontrunner is ranked higher on that voter’s ballot. 

This procedure was first prepared for use in previous IRV elections (in Hendersonville in 2007 and 2009) and approved by the State Board of Elections for use in this year’s Court of Appeals election. Making use of the voter-verified paper trail that comes with these systems, a random audit will confirm that voters’ rankings have been accurately converted into data for being tallied in a spreadsheet.

Once the totals from all jurisdictions have been added together, the candidate with the most votes will be declared the winner – a majority of those with a preference between the two frontrunners. The delay between the casting of votes on Nov. 2 and the determination of a winner is attributable to the fact that IRV-friendly voting technology has not yet been fully implemented in North Carolina. In the future, the tabulation process could be completed at the same time as the tally of first choices.

More information on instant runoff voting in North Carolina – including an explanation of the counting process and background material on each of this year’s IRV races – can be found at www.ncvotes123.com.