Website Search

Sunday, December 26, 2010

N.Carolina's Statewide Instant Runoff Voting Contest - the facts, the regrets

48 days after the election, instant runoff voting produced a "winner" for the NC Court of Appeals, for the  "Wynn"seat. Thanks to IRV, an experimental tallying method was used, the election was almost a tie, a recount was called for, and we had a plurality result, not a majority win.

After counting the IRV votes,  the highest vote getter lost his 100,000 vote lead, and the contest became so close that a recount was requested. The declared winner won with 28% of votes from all ballots cast on election day. We do not know how many 2nd and 3rd choice votes each candidate got, as this data was not released to the public.

(Click on the above picture to enlarge.)

Footnotes to spreadsheet/chart:
[1] State BoE First Round results Nov 2, 2010 (page 20)
[2] State BoE IRV results Dec 13 2010  (screen shot)
[3] State BoE IRV results after recount

The reason for using IRV in this election:
"NC Court of Appeals Judge Wynn was elevated to the 4th U.S. Circuit Court o Appeals. Under the NC Constitution, the vacant office must be filled by an election this November. Thirteen candidates are running for the office. Too late to hold a primary, under NC law, this election will use an Instant Runoff Voting method. Voters indicate who they prefer among the candidates at the "instant" they vote, by ranking as many as three choices 1, 2, and 3. If your 1st choice loses, your 2nd or 3rd choice could determine who wins. Bobby Coggins.(blog)

There were 13 candidates in this non partisan contest and there was no primary. Candidates had access to public funding for their campaigns, but the limited funds had to be shared amongst a large field.

In the first round, Cressie Thigpen (D) was leading Doug McCullough (R) by 100,000 votes.

The top 2 candidates advanced to a runoff. The ballots were counted based on whether they ranked Thigpen or McCullough higher. 45% of those who voted in the Court of Appeals races voted for neither. 2nd and 3rd choice votes for Thigpen & McCullough were added, erasing Thigpen's 100,000 vote lead and putting McCullough about 6,500 votes ahead.

With IRV votes tallied, McCullough won with 28% of all voters' ballots cast on election day.The election was so close that Thigpen called for a recount. This was done by machine so it was no surprise that the results changed only slightly. Thigpen went down 120 votes, from 537,445 to 537,325. McCullough went down 43 votes, from 544,023 to 543,980. McCullough won by 6,655 votes.

McCullough got 19.7% of transfers, Thigpen 11.2%, and nobody 69.2%

The Democrat-registered incumbent appointed by a Democratic governor got a large share of the vote. But there were so many other Democratic candidates, that they weren't likely to result in many successful transfers, especially since the ballot instructions said simply to vote for the candidates in order of your preference - and not vote strategically so your ballot would not be stuck in the trash dumpster. 2.7 million voters went to the polls in November. Only 1.1 million had their vote count in December.

There were fewer Republican candidates, so there was a much greater chance of the vote counting, if you could just get voters to fill out the ballot. And there were more Republican voters in the first place.

News and Opinion:

Raleigh News & Observer, December 24, 2010. Instant analysis

Gary Robertson.AP. December 20, 2010. Concession after N.C. Court of Appeals recount widens lead

Doug Clark. Greensboro News & Record. December 16, 2010. Weak defense of IRV

Reid Overcash. News&Press.December 16,2010 “Instant Runoff Vote” loses its appeal when put to test

Gary Pearce.Talking About Politics. December 16, 2010 Instant Runoff, Eternal Confusion

Rocky Mount Telegram December 13, 2010 Instant runoff voting runs into issues, too.
That was one ugly way to elect a judge.

Charlotte Observer Dec. 13, 2010 Not best way to boost confidence in elections
Instant Runoff Voting was alot faster than the counting.

AP. Raleigh News & Observer December 9, 2010 Judge Thigpen wants Appeals race recount

Doug Clark. Greensboro News & Record. December 9, 2010 Editorial: Instant runoff failure

Gary Robertson. AP News. December 6, 2010 McCullough overtakes Thigpen in NC court race

Tom Sullivan. Scrutiny Hooligans. November 9, 2010. Short Cuts. 

Andrea Pacetti. News 14. October 28, 2010. Board of Elections decides how to count instant runoff votes (changes tallying method)

Bobby Coggins. October 19, 2010. 2010 Election and this IRV ballot for NC Appeals Court.

Sara Burrows. Carolina Journal. September 29, 2010 How the Instant Runoff for State Court of Appeals Will Work

North Carolina State Board of Elections. September 29, 2010 Meeting on Tallying IRV

Bobby Coggins. September 20, 2010. Instant Runoff Voting Strategy for NC Court of Appeals.

Joyce McCloy. IRVbad4NC blog. September 19, 2010.  2 Simple Rules for Voters in Instant Runoff Voting for NC Court of Appeals

Charlotte Observer September 9, 2010. Potential mess looms in judicial election

BY ROBERT ORR. Raleigh News & Observer September 5, 2010. Order in the court election!

Doug Clark. Greensboro News-Record September 3, 2010. Editorial: On the wrong track

Raleigh News & Observer. September 3, 2010. Test vote

Doug Clark. Greensboro News and Record. August 31, 2010. Lucky 13

GARY D. ROBERTSON AP. August 31, 2010. 13 candidates file for open NC appeals court job

ES&S and PrintElect voting machine co. August 31, 2010. Vendor letter on legal issues with NC Court of Appeals Instant Runoff Voting

§ 163-329.  Elections to fill vacancy in office created after primary filing period opens.

Sri Lanka Contingent Vote (method used in North Carolina)

Wikipedia Article on this contest,_2010#Court_of_Appeals_.28Wynn_seat.29

If you have other analysis on this contest, we'd love to hear from you. Comments on this blog are blocked but you can email me at joyce (at) ncvoter (dot) net

To receive updates by email visit this link

Monday, December 20, 2010

After 48 days the instant runoff voting contest for NC Court of Appeals is over

Just in time for Christmas. 48 days after the election, instant runoff voting produces a "winner" for the NC Court of Appeals contest for Wynn's seat. Initially, Cressie Thigpen was leading by 100,000 votes but he only had 20% of the votes cast amongst the 13 candidates. Next, 2nd and 3rd choice votes for Thigpen & McCullough were added, erasing Thigpen's lead and putting McCullough about 6,500 votes ahead.  With IRV votes, Doug McCullough had the lead. Even after all the adding of IRV votes, McCullough won with 28% of all voters' ballots cast on election day.The election was so close that Thigpen called for a recount. This was done by machine. The results changed slightly. Thigpen went down 120 votes, from 537,445 to 537,325. McCullough went down 43 votes, from 544,023 to 543,980. McCullough won by 6,655 votes (including IRV). I'm basing this on numbers from data at NCSBE site on Dec 13 - see screen shot  and results now up at the State Board of Elections website on Dec 20 

Concession after N.C. Court of Appeals recount widens lead
The Associated Press © December 20, 2010 By Gary D. Robertson 
Doug McCullough is returning to North Carolina's Court of Appeals after a completed statewide recount Monday showed McCullough still in the lead and incumbent Judge Cressie Thigpen conceded the election.
Thigpen conceded defeat after the complete recount in all 100 counties had McCullough, a former member of the state's intermediate appeals court, leading by 6,655 votes. Nearly 1.1 million votes were recorded between the two candidates in the instant runoff race.
Thigpen sought the recount when unofficial results showed him trailing by about 6,000 votes, saying he wanted to make sure election officials counted properly ballots cast through the unusual voting method. The margin widened to more than 6,500 votes before the recount began.
The recount began as early as Dec. 10 in some counties and wrapped up early Monday. Election reform advocates have been keeping an eye on the race because the ranking concept hadn't been used in a statewide race anywhere in at least 70 years.
"I would like to thank the voters who showed faith in me during this process and to the N.C. State Board of Elections for their diligence to insure votes were properly counted," Thipgen said in a prepared statement. "The voters have spoken and I congratulate Judge McCullough for his victory and will do all I can to make any transition seamless."
McCullough is a former federal prosecutor from Atlantic Beach who served on the Court of Appeals from 2000 through 2008. He didn't immediately respond to a request for comment.
State elections director Gary Bartlett said the next step for the board is to certify the outcome. It was the board's final unresolved race in North Carolina from the 2010 elections. No date had been set Monday afternoon.
Gov. Beverly Perdue had appointed Thigpen in August to serve through the rest of the year on the court and fill the seat vacated when Jim Wynn was confirmed to the 4th U.S. Circuit Court of Appeals. A 2006 state law required the instant runoff method to fill the seat because Thigpen and 12 other candidates ran for an eight-year term. Voters were asked to rank three candidates from the 13 on the Election Day ballot.
Thigpen and McCullough advanced to the second round of voting because they received the most first place votes. Election officials then separated the ballots of voters whose first-choice candidate was eliminated and counted how many of them made Thigpen and McCullough their next highest choice. Those choices were added to the original counts of the two. The candidate with the most combined votes was the winner.
McCullough trailed Thigpen by 100,000 votes in the first round.
Rob Richie, executive director the election reform group FairVote, said the statewide use of instant runoff voting was successful although it took 48 days after Election Day to finalize the winner. He said the method saved millions of dollars it would have cost to organize a separate runoff election. The state elections board and other groups provided voters information on how the ranking worked.
"It's a pretty fair reflection of a fairly close race," Richie said in a phone interview. "It seems like the voter education that was done was effective."
But critics said the process was still hard to understand for many voters, and the permutations of voting outcomes among 13 candidates hard for election officials to ensure the outcomes were accurate. Counties that used electronic voting machines also counted second-round totals with untested tabulating software, according to Joyce McCloy with the N.C. Coalition for Verified Voting.
"Instant runoff voting was not instant, nor was it like a runoff election," she wrote in an e-mail.
Bartlett said there were no major problems with the count and voters got to participate in picking the winner.
"Whether you like it or not, it worked," he said.

To receive updates by email visit this link

Sunday, December 19, 2010

Weak defense of IRV - Instant Runoff Voting in NC Court of Appeals oddities

Some say instant runoff voting did what it was supposed to do in the NC Court of Appeals Contest. For sure the IRV algorithm that sorts, allocates and eliminates votes is capricious in nature. IRV election is like a crap shoot. The top 2 candidates had 20% and 15% of first round votes. Thigpen led McCullough by 100,000 votes. After sorting, allocating, eliminating and reshuffling the IRV votes, McCullough and Thigpen had a near tie. The top two, Thigpen had 27.62% of the vote from all ballots cast, and McCullough had 27.96% of all votes. But with IRV, majority is redefined to 50% of "remaining" votes, and excludes about 800K voters. Thigpen lost his lead and now trails McCullough by around 6,700 votes. Close enough that Thigpen called for a recount. This was conducted by machine, not hand - last week. Results aren't in yet. 

Weak defense of IRV
Greensboro News and Record. Doug Clark. December 16, 2010.
It's disappointing that Duke professor Mike Munger calls our opposition to Instant Runoff Voting "absurd," and counters with arguments that are just plain wrong.
The IRV, which decides a winner in a multicandidate race by adding second- and third-choice votes when no one receives a majority, achieves a result "precisely the same as if we conducted two separate elections," says Munger.
Talk about an absurd statement. That would be true only if voter turnout wasprecisely the same in a second primary and no voter changed his or her mind about the candidates in the interim. But it's worse than that, because IRV effectively excludes a high proportion of voters from this runoff.
IRV was used in the special election for an appeals court seat, which drew 13 candidates. On Nov. 2, Cressie Thigpen led with 395,220 votes, or 20 percent. Doug McCullough was next with 295,619 votes, or 15 percent. As the top two, under the IRV rules their second- and third-place votes were added. Last week, five weeks after the election, the State Board of Elections announced that McCullough was now the winner with 544,023 votes to 537,445 for Thigpen.
This was, of course, not precisely like a separate runoff for another important reason: 1,943,771 people voted in this race on Nov. 2. But the votes for McCullough and Thigpen after the "instant" runoff totaled 1,081,468. What happened to the other 862,303? These were voters who did not mark McCullough or Thigpen as their first-, second- or third-choice candidates. So the "instant" runoff was like holding a second separate election in which 862,303 people who voted in the first election — 44 percent of the total — were excluded from participating. How does Munger think that yields precisely the same result?
Next, responding to our view that second- and third-choice votes should not count as much as first-choice votes, Munger says "the idea that the votes are second-place or third-place votes is balloon juice." I don't know what balloon juice is — hot air, maybe? — but I do know that the ballot instructed voters to mark their second and third choices as well as their first choice. So they indisputably are second- and third-choice votes. For many voters, they may have been an afterthought.
If these second and third votes are considered to count as much as first-choice votes, then they should have been counted for all 13 candidates. If they were, it's possible that a candidate other than McCullough or Thigpen could have received the most votes. We don't know because the State Board of Elections has not released how many 1, 2 and 3 votes each of the 13 candidates polled.
Finally, Munger employs some "balloon juice" of his own: "The editor seems to prefer plurality rule, but sophomore political science students learn that plurality rule rarely chooses the best candidate in simulations."
Simulations? Why don't we look at real elections? In Guilford County's Senate District 28 race, Gladys Robinson won a plurality election over Trudy Wade and Bruce Davis. There was no "instant runoff." On what basis would Munger contend that Robinson, a community leader for many years and a member of the UNC Board of Governors, was not "the best candidate"?
For that matter, how does Munger conclude that Thigpen was not the best candidate in the appeals court race and McCullough was? Thigpen led McCullough by 100,000 first-choice votes — meaning a lot more voters thought Thigpen was the best candidate. After adding 2 and 3 votes, McCullough led Thigpen by 7,000 votes (pending a recount). On that basis Munger says McCullough was the best candidate? That strikes me as more than a little absurd.
On top of its other faults, IRV is arbitrary. Why not just count first- and second-choice votes? If that had been done, Thigpen's initial lead over McCullough probably would have held up. Why add third-choice votes? That's sliding further from the voters' real preference. Or if the idea is just to count more votes, how about giving voters fourth and fifth choices? Or let them rank all candidates in order of preference? That would be worse, in my view, but it's no less logical than the IRV format applied in this race.
My question to IRV supporters is this:
Compared to a plurality election, does IRV produce an outcome that is faster, less expensive or more credible?
The first two answers are no, plain and simple. Those are facts. The third question requires an opinion. My opinion is no — and professor Munger sure didn't change it.

To receive updates by email visit this link

Tuesday, December 14, 2010

North Carolina instant runoff voting: "That was one ugly way to elect a judge"

The Rocky Mount Telegram has withdrawn support for instant runoff voting. The RMT editorial dept promoted IRV over and over in spite of our warnings that Instant Runoff Voting isn't instant, and isn't the same thing as a runoff. We also warned that IRV is akin to a crap shoot. Well, we told you so.

Instant runoff voting runs into issues, too
Monday, December 13, 20100 Rocky Mount Telegram

That was one ugly way to elect a judge.

More than a month after the Nov. 2 election, North Carolina finally has decided that Doug McCullough is its newest member of the N.C. Court of Appeals, thanks to a new procedure called “instant runoff voting.”

In the interest of full disclosure, it should be noted that we used this space several times before Nov. 2 to applaud the innovation. We had high hopes that instant runoff voting would make low-turnout, costly runoff elections a thing of the past. It might still do that, but it’s not without issues.

McCullough won his seat on the court as a result of counting second and third choices cast by voters in November. That’s the way the process is supposed to work, but consider the fact that Cressie Thigpen actually had about 100,000 more votes than McCullough in the count taken on Election Day. His total wasn’t a majority, however, in a field of 13 candidates.

Eighty percent of the voters cast first-choice ballots for someone other than Thigpen. Eighty-five percent of the voters cast first-choice ballots for someone other than McCullough.

It isn’t reasonable to argue that McCullough is the candidate that North Carolina voters prefer, based on those numbers. But let’s not forget that a runoff election — even one that’s statewide — traditionally draws a miserable turnout, as well. In the Democratic primary runoff for U.S. Senate last summer, for example, winner Elaine Marshall received fewer votes than her opponent, Cal Cunningham, received in the first primary.

There’s no clear answer to the runoff dilemma. Hold a runoff on a separate date and watch turnout plummet. Try something like instant runoff voting, and the results are awkward, to say the least.

We appreciate the innovative thinking behind instant runoff voting, but officials need to keep looking.

To receive updates by email visit this link

N Carolina Instant Runoff Voting 'Not best way to boost confidence in elections'

Here's op/ed in Charlotte Observer (one of NC's top newspapers) about Instant runoff voting. And how it wasn't everything it was cracked up to be.. My comments below the article.

Not best way to boost confidence in elections
Instant Runoff Voting was alot faster than the counting.

Monday, Dec. 13, 2010 Charlotte Observer

Perhaps the best that can be said of North Carolina's new electoral experiment is that Instant Runoff Voting (IRV) accomplished two of its goals: It produced a winner, though the loser has asked for a recount. And it saved N.C. taxpayers upwards of $5 million because it wasn't necessary to hold a statewide runoff election.

But the IRV process was anything but instant. Voters cast ballots fairly quickly for their first, second and third choices among 13 candidates running for an N.C. Court of Appeals seat, and first-round ballots were tallied fast. But computing the impact of those second and third choices took more than another month. A glaring error, once discovered, resulted in a dramatic change in the apparent winner. And the vote was so close that one candidate is exercising his right to a recount just to make sure.

Out of all this, are voters reassured about the electoral process? Maybe so. The election did result in the selection of an experienced, solid judge in New Bern lawyer Doug McCullough, who served previously on the Court of Appeals.

And the apparent loser in this election, Raleigh lawyer Cressie Thigpen, may continue to serve on the court even if he falls short in the recount. Gov. Bev Perdue had appointed Thigpen to the seat late last summer to fill a vacancy, and she could appoint him to another vacancy caused by the election of Court of Appeals Judge Barbara Jackson to the N.C. Supreme Court.

But the most troubling thing about Instant Runoff Voting was not that it did not produce an instant winner, nor that it has taken so long to persuasively produce a winner. The most troubling point is that Thigpen led the Election Day voting with more than 395,000 votes - 20 percent of the 1.9 million votes cast, but lost to McCullough, who came in second with more than 295,000 votes, about 15 percent of the total.

In other words, Thigpen was not the first choice of 80 percent of the voters, while McCullough was not the first choice of 85 percent of voters. Yet when all the second- and third-choice votes were added, McCullough had a lead of more than 6,000. It will strike some voters as curious that second- and third-choice votes count as much as first-choice votes - with the result that the candidate who led the election in the first round loses.

That's the way Instant Runoff Voting works - and it may be the reason that a number of jurisdictions have stopped using it to determine victors in multi-candidate races. N.C. legislators should re-examine the law to determine whether it's the best way to maintain voter confidence in elections. We believe that case has yet to be made.

We think the better course is to avoid using the IRV process in statewide elections, because it takes too long to compute a winner. Lawmakers would be wiser to allow judicial appointees named to court seats after statewide primaries are over to serve until the next regularly scheduled statewide primaries. That would give the nominee time to concentrate on the job - and give voters a clearer result on Election Day.

Here's my comment posted online to this article:

There was no majority winner here. Worse, the tallying process is error prone, as stated in the OP, although the exact type of error is not stated. Why not? Voters weren't prepared to rank choices and the only voter education in our state, a voting rights act state, was a voter guide sent to households. Around 5 cents was spent educating voters about this voting method formally known as "Sri Lanken Contingency Voting".

IRV is very complex to tally - even if NC had federally tested software to tally it, which we don't! IRV is not additive. Thats right, you can't just add up the votes at the polling places. A complex algorithm is used to weigh each 2nd and 3rd choice - sorting, allocating, eliminating and reallocating votes until finally the deck is shuffled all into one pile and a winner announced. The complexity of counting IRV ballots leads to great logistical problems and time-delays. It is so complex that the validity of the results is not verifiable. Who really won?

I wager that if you recounted this contest multiple times, you would get different results each time.

What about the voters? Voters had to rank their "runoff" choices without knowing which of the 13 candidates would be the "top two" in the runoff. So if a voter didn't rank either Thigpen(D) or McCullough(R) for any of the 3 choices, then that voter was shut out of the runoff.

The state's IRV procedures violate the intent and statutes of the Public Confidence in Elections Act, a law passed in 2005 to protect North Carolina's voters and votes from voting machine software flaws and miscounts. This law was passed to prevent headlines such as we saw in 2004:

"A Florida-style nightmare has unfolded in North Carolina in the days since Election Day, with thousands of votes missing and the outcome of two statewide races still up in the air."AP Newswire, Nov 13

Our situation was dire indeed: "NC has the worst election problem in the country right now." -- Computer scientist Dr. David L. Dill of Stanford University, about the votes lost in Carteret County. November 11, 2004.

We cannot implement IRV in North Carolina without exposing our future elections to fiascos as occurred in 2004, where over 50,000 votes were mis tallied - some missing, some double counted, some lost forever.

We warned you *before* 2004 that computer vote counting could result in lost or inaccurate vote tallies, and pushed for paper ballots and standards for voting vendors and systems. No one listened to us until the Nov 2004 fiasco.

We warned you ahead of time that this statewide IRV contest 1) was error prone and violated election transparency laws; and 2) we told you that IRV voting for this contest was akin to participating in a crap shoot.

IRV is very bad for democracy, very bad for voters, and sets a terrible precedent of non transparent unverifiable voting. If the objective of an election process is to discern the will of the voters, then that process must be the simplest, most transparent and most enfranchising method for all voters.

More more information about the problems with instant runoff voting see our website at

To receive updates by email visit this link

Thursday, December 9, 2010

N. Carolina's instant runoff voting chickens come home to roost. Recount sought

The instant runoff voting contest for NC Court of Appeals is headed to a recount. Here we are, 38 days after the election, we get the results to the "instant runoff" and learn that we're going in for another round, due to lack of confidence in the tallying procedures. No surprise, since Larry Leake, the Chair of the NC State Board of Elections expressed concern about the tallying procedures when approved on Oct 28, 2010 in the middle of the election.

So, the instant runoff voting chickens have come home to roost. Proponents often have compared IRV voting to ranking different flavors of ice cream. They called it "progressive". Well, the "progressive" IRV chickens have come home to roost in NC, and those flavors taste like sawdust.

Judge Thigpen wants Appeals race recount
December 9, 2010 Raleigh News & Observer ...Spokesman Reid Overcash says Thigpen wants to make sure the count was performed properly because instructions may have been confusing to county officials.

Editorial: Instant runoff failure
THURSDAY, DECEMBER 9, 2010 (Updated 3:02 am)
North Carolina’s first try at statewide Instant Runoff Voting didn’t work out very well, except for Doug McCullough.

The second-place finisher in last month’s special election for a Court of Appeals seat apparently ended up winning. Although McCullough trailed Judge Cressie Thigpen by 100,000 votes Nov. 2, he gained enough second- and third-ballot votes to overtake Thigpen’s lead, pending a recount.

The outcome should convince legislators that this experiment failed and should be abandoned.

The goal, ironically, was to produce a more credible result in this sort of situation. The special election was called because a vacancy occurred on the Court of Appeals in August, far too late to schedule a primary. The last time a similar situation came up was in 2004 when a seat opened on the Supreme Court. Eight candidates filed to run, and Paul Newby won with only 23 percent of the vote. The legislature drafted a new law seeking to produce a larger mandate.

This time, the special election drew 13 candidates. Voters were asked to indicate a first, second and third choice — an exercise that surely taxed their familiarity with the relatively little-known contenders.

Only the first-choice votes were tallied Nov. 2: Thigpen received 20 percent and McCullough 15 percent. They advanced to a second round of counting.

Election officials added the second- and third-choice votes county by county. A mere five weeks after the election, a tentative result was reached. To help promote transparency, the State Board of Elections should release all vote totals so the public can see exactly how it came about.

Proponents of this experimental system say that 20 percent should not elect someone for such an important office. The additional second- and third-choice votes also indicate public support, they say. As things stand now, however, a candidate with just 15 percent of the vote has won. Only by giving second- and third-choice votes as much weight as first-choice votes can McCullough claim more public support than Thigpen.It’s simply not a democratic principle that a second or third preference counts as much as a first.

Fortunately, McCullough and Thigpen both are well-qualified for the position. Both have experience on the court. And Gov. Bev Perdue, who gave Thigpen a temporary appointment in August, can appoint him to fill another vacancy on the court — and should.

The best way to handle such vacancies would be for the governor to appoint a replacement who would serve until the next regular election two years later.

Short of that option, a special election ought to be decided more directly: Voters should cast a single vote for the candidate of their choice, and the candidate with the most votes wins. Having a candidate win with only 20 percent of the vote is a more acceptable outcome than having him lose to a candidate with 15 percent

Will a recount change the outcome? It could if some counties made errors in sorting, allocating, eliminating and re-shuffling the 2nd or 3rd choice votes. We saw how in Cary, NC in 2007 that officials had trouble tallying just 3,000 ballots correctly. How hard would it be to tally 2.4 million ballots cast in 100 different counties with different voting systems, correctly? With The saving grace is that the "winner" of the election is very qualified. The bad news is that this election method does not inspire confidence in the outcome and does not provide a majority winner, i.e 50% + 1 of the votes.

NC was fooled into using this convoluted chaotic non transparent boutique style election method. Lets not be fools again.

To receive updates by email visit this link

Monday, December 6, 2010

Instant runoff voting shocker-Thigpen loses 100,000 lead, is 6,700 behind

Incumbent Judge Cressie Thigpen had a 100,000 vote lead in the statewide IRV contest for NC Court of Appeals. He's lost that lead thanks to IRV votes and now is 6,700 votes behind. Although this was non partisan contest, voter education fell to the political parties mostly, because IRV was touted as a cost saving measure.

Monday Dec 6 2010 McCullough overtakes Thigpen in NC court race
AP News. RALEIGH, N.C. -- Nearly complete results from the instant runoff race for the North Carolina Court of Appeals show the second-place candidate overtaking the leading candidate from the first round of voting.
State elections director Gary Bartlett said Doug McCullough had a roughly 6,700-vote lead over incumbent Cressie Thigpen with counting complete in 99 of the 100 counties. The only one left - Warren County - didn't have enough votes cast to turn the race back to Thigpen.
If the race stays close, Thigpen would have until Thursday to ask for a recount.

Voters initially ranked up to three candidates among 13 who ran on Election Day. Thigpen was the top recipient of first-place votes - 100,000 more than McCullough. But McCullough caught up with second- and third-place votes.

DEMS did a poor job on voter ed, GOP did much better job. DEMS were told to bullet vote, to rank Thigpen 1st and 3rd, and other crazy stuff. Some voters mistakenly ranked 2 or more candidates in the first column. State funded voter ed was meager SBE flier to households. (Since IRV is easy as 1-2-3).

The winner will serve an 8 year term. The tallying method was error prone and risky. We have no way to know if vote tallies were accurate.

The irony is that IRV was pushed by Democrats and thanks to their reform, the democrat lost. The North Carolina Coalition for Verified Voting can now say - be careful what you ask for and "We told you so".

To receive updates by email visit this link

Sunday, November 21, 2010

For Polk County Democrat website, Ignorance is bliss - about instant runoff voting

Ignorance is bliss if you read what website owner of Polkdemocrats writes. They don't know or care how the statewide IRV contest will be tallied, in fact they don't even know that we have top 2 IRV and NOT San Francisco style IRV. Polk Dem's webmaster says no one cares about this contest so its ok if voters are confused, spoil their ballots, or that results will be hinky. They aren't worried about the non transparent tallying process, either. Guess it's be ok if someone accidentally mistallied the statewide contest and the wrong person won? Since the tallying method erases audit data, who'll know. What we don't know won't hurt us, right? Maybe it would save even more money if we had the State Board of Elections just pick out the winner for us.

Instant Runoff Vote

In the 2010 midterm election, North Carolina was faced with an unusual situation in a judicial election, where 13 candidates would be on the general election ballot for a single office. Situations like this happen all the time in primaries, and in fact the purpose of the primaries is to winnow the field down to two candidates for the general election.

But sometimes the primaries don't accomplish that (because no candidate gets over the threshold percentage of votes), and, rarely, as in 2010, offices become vacant after the primaries. In those cases the law requires a runoff primary, which is a huge expense for the state because a second election is just as expensive as the first, even if it is for a single office. In the 2010 primary season, there had to be a separate runoff primary for a single Democratic office -- Senator -- so the election machinery had to be ramped up in full: the Republican judges and workers had to show up even though none of their candidates were involved.

Cooler heads have prevailed, and for the 2010 general election that multi-candidate judicial race was decided at a single election. Here's how the system works:

The ballot lists all of the names of the candidates three times: first in a column for the voters to select their favorite candidate, then in a second identical column for the one they would prefer if their first choice wasn't selected, then a third column for their third choice, given that neither their first nor second choice made it. This is basically what you would do in a runoff primary, you're just doing it all at once.

Come election day, the Board of Elections in Raleigh tallies all the votes in Column 1 -- the voters' first choice -- and ranks them top to bottem in number of votes. If the top vote-getter gets more than 50% of the votes, the election is over. That's very unlikely in a 13-candidate race, and in that case the BOE begins calculating the instant runoff. What they do is take the top two vote-getters into the instant runoff, just as would happen in a primary, but carry out the voting using the votes already cast in columns 2 and 3. doesn't know (or much care) the exact mechanism the BOE uses to calculate the runoff, but in other jurisdictions using the system they start with the low vote-getter, checking how many people voting that person had column 2 votes for either of the top two, and adding those to the column 1 totals. Then they go to the next lowest vote-getter, add those votes, then see if there are enough votes for either of the runoff candidates to get over 50%. And so on until each of the losing candidates in column 1 has their column 2 votes tallied. If after that neither of the runoff candidates still has 50%, then the BOE goes to column 3 and repeats the whole procedure. That should produce a winner.

We presume that the BOE in its august attention to duty had a platoon of statisticians working out the probabilities and know that three times around the track is a great plenty.

Assuming that we ultimately have a winner in this 2010 race (we don't know yet), is the instant runoff a good idea?

Yes. Anything that saves the taxpayers $3 million to avoid a runoff primary for a race nobody cares about or will vote in is a good thing. It introduces a lot of complexity in the last thing anybody wants complexity introduced to, and surely mistakes at the polls were made aplenty. For example, people voting for their candidate in all three columns, thinking -- wrongly -- that doing so meant extra votes for their favorite. The only votes counted in columns 2 and 3 are those of voters whose favorite LOST in the first round. If your candidate makes it to the runoff, your votes in columns 2 and 3 aren't counted at all; if your candidate doesn't make it, your only votes in columns 2 and 3 that are counted are those for one of the two candidates that did.

We must take the mistakes and under-voting as a valid tradeoff for an actual runoff election.

Something else we must take is the philosophical one of the possibility that a decisive but not majority winner in the first round could lose in the runoff. That in fact happened this year in the City of Oakland California, which also uses this system. That could also happen in an actual runoff, of course, so count as a big supporter of the system.

To receive updates by email visit this link

Wednesday, November 3, 2010

Oakland CA voters confused by instant runoff voting on Nov 2 - video

How did Oakland voters feel about instant runoff voting, also called ranked-choice voting? Here is a news update and also a video interview by an observer.

Oakland spent $3.51 per voter in targeted areas to educate them on RCV. Many voters were still confused.

Oakland voters flock to polls, struggle with ranked-choice voting
By Scott Johnson and Janis Mara. Oakland Tribune Posted: 11/02/2010

OAKLAND -- As Oaklanders cast their ballots today, one of the hottest subjects is not the issues and the candidates, but the method of voting, a new ranked-choice system used in the mayoral race.
Diana Toutjian, 57, a native Oaklander who attended Piedmont Avenue as a kindergartner, was baffled by the system. She didn't understand how it worked, and voted three times for the same candidate. This means her second and third votes won't count. "They should have had a red box there to explain what we were supposed to do," Toutjian said...

To receive updates by email visit this link

Friday, October 1, 2010

What Carolina Journal wasn't told about Instant Runoff Voting for NC Court of Appeals

The article State : How the Instant Runoff for State Court of Appeals Will Work gives a misleading impression on the ease of continuing or expanding instant runoff voting. In fairness, the story IS based on an interview by Sara Burrows of the Carolina Journal with NC's top election official. Sara is only going by what she was told. When I read the article, I was stunned to see the claim that IRV could be automated for less than $5 million. That just isn't true. If you want to know the truth about IRV, you have to know which questions to ask, and you need to know when you aren't being told the whole story.

From the article, here's the misleading claim about cost to "automate" IRV, which I will debunk:
“It all comes down to whether the voters like it or they don’t like it,” Bartlett said. “If the voters like it and the candidates think it’s fair, then the General Assembly needs to look at expanding it and getting us automated software for it.”

Purchasing and installing the software, Bartlett added, would cost less than the $5 million it takes to run one statewide election.

IRV won't be automated so cheaply and it also won't be expanded without gutting several election transparency laws.
IRV already clashes with more than a few transparency laws in place, from violations of standards for voting systems and vendors to the law that mandates votes be counted where cast, and that all votes be reported on election night. This has to be corrected. Those standards were put into place toprevent a repeat of headlines like this: "A Florida-style nightmare has unfolded in North Carolina in the days since Election Day, with thousands of votes missing and the outcome of two statewide races still up in the air."AP Newswire, Nov 13

With IRV, not all voters get to participate in the "runoff". Only voters who had the luck to rank the top 2 candidates will have a say in the "runoff".

The real cost to "automate" IRV would be around $65 million dollars, the cost for all new voting machines for all counties. This is because the current voting system, including the "IRV" software would have to be certified together, and to different federal standards than our machines were certified to when purchased in 2006. It is very unlikely our current voting machines will meet the current federal standards, even if our current systems do operate very well. And these machines will last at least 10 years, if not 15 or more.

In an email dated Sept 30, 2010, NC State Director of Elections Gary Bartlett admitted to me that it was highly unlikely for our voting systems to be recertified with IRV software. Gary answers my questions
in blue/bold.

-------- Original Message --------
Subject: RE: request for documents, questions to SBE
Date: Thu, 30 Sep 2010 14:58:29 -0400
From: Bartlett, Gary
To: Joyce McCloy
CC: Wright, Don , McLean, Johnnie

*Wouldn't North Carolina's entire voting system -M100s, M650's, iVotronics, Unity + the "automated software" have to be federally tested and certified together in order to meet North Carolina's standards for voting systems and software? Yes. Due to the age of the equipment, I would be shocked if ES&S upgraded these machines because of cost.

*Isn't there a chance that the system you propose would not pass federal certification standards? I have not proposed software; I provided to the reporter of the amount ES&S stated it would cost to develop IRV software.

*Would this system you propose be certified to the old 2002 standards? If not, would our current voting machines + the "automated software" meet the higher standards? No; must be the current federal standards, thus, upgrades would have to be made.

From: Joyce McCloy []
Sent: Wednesday, September 29, 2010 10:52 AM
To: Bartlett, Gary
Cc: Wright, Don
Subject: request for documents, questions to SBE

"Automating" isn't automatic. To "upgrade" the system would mean physically changing out computer chips plus installing new software, but that would have to go through the federal testing and certification process which could take up to 2 years. Then, we could end up with voting systems that still coudn't tally IRV, as happened in Pierce County Washington when they bought special IRV software/systems. Their IRV precinct optical scanners couldn't tally IRV. See official recap of Pierce Co IRV election

Here's a letter to drive home the importance of using federally certified software and systems in North Carolina.
It is the law, and there are penalties to the voting vendor if they violate that law.
Hence, the letter from ES&S/PrintElect to Gary Bartlett, Director of the NC State Board of Elections

"IRV is not an approved function at the federal or state level of current ES&S software, firmware or hardware. Subsequently, we will work at the direction of the SBE and counties to assist but cannot be held responsible for issues as a result of IRV..".
~ letter from PrintElect to the North Carolina State Board of Elections, dated August 31, 2010.

Does IRV solve a problem or create one? It is highly likely that the winner of the Court of Appeals seat elected by IRV will be by much less than 50%, and as far as confusion, we won't know because in North Carolina, the NC State Board of Elections doesn't report all IRV votes, pretending that those aren't really valid votes unless needed to determine the "runoff". IRV is more complex than other voting methods, because you can't simply add up the vote totals, but have to sort, allocate, eliminate and reallocate, examining each individual ballot.

We could easily argue that this contest violates the voting rights act, because the lack of voter education and the demands put on the voters will disenfranchise many who normally would not be disenfranchised. Our state will only spend $500K in voter education, and already political groups are getting it wrong, telling voters to single shot or bullet vote this contest. *See attached document on voter education issues.

North Carolina's high standards passed in 2005 have improved our elections:
Under-vote rate for President is cut from 3% in previous elections to average of 1% statewide in 2008
Post election audits of 2008 Presidential election show voting machine results are accurate.
See "An Assessment of the Recount and the Certification of the Election Result for the November 2008 Election"


Joyce McCloy
NC Coalition for Verified Voting

To receive updates by email visit this link

Monday, September 20, 2010

Vendor letter on Legal Issues with NC Court of Appeals Instant Runoff Voting

We may have a voter guide, and we may have an Instant Runoff Voting task force, but sorry, no cigar! We still do not have IRV tallying procedures for touchscreen counties as of Sept 17, 2010. Additionally, in an August 31 letter, the voting vendor warns that our voting system cannot legally tally instant runoff voting, and the vendor will not accept responsibility for an IRV related election fiasco.

I contacted a member of the task force set up to devise procedures, he advised me on Sept 13 via email that:

"there are still several options on the table and I do not know at this point what method we will end up with. I think we need to spend some time testing and evaluating the options before a decision is made. Might as well do it right the first time."

Meanwhile, the voting vendor warns about the legality of using our voting machines for this process. Here's excerpt from Print Elect & ES&S letter to the NC State Board of Elections dated Aug 31 2010:

"IRV is not an approved function at the federal or state level of current ES&S software, firmware or hardware. Subsequently, we will work at the direction of the SBE and counties to assist but cannot be held responsible for issues as a result of IRV....
IRV Tabulation
*Printelect and ES&S will only take responsibility for and support tabulating the IRV contests individually. Methods for deciding a runoff winner by others will not be supported by Printelect or ES&S. This risk will be the sole responsibility of NCSBOE and the counties.

*IRV voting tabulation methods are not an EAC or state certified portion of our voting system and have not undergone the testing that would normally be required to receive these certifications."
read the full letter here

But pro IRV advocate Bob Phillips of Common Cause NC expresses confidence:

"It's going to be a reliable, accurate count," ... "I'm very comfortable with what I'm hearing so far." ~ Study calls NC best among 10 at protecting voters" By GARY D. ROBERTSON.

Why does the voting vendor worry when Common Cause NC does not? Because the vote vendor is legally held to high standards set in 2005 to prevent repeats of headlines like this:

"A Florida-style nightmare has unfolded in North Carolina in the days since Election Day, with thousands of votes missing and the outcome of two statewide races still up in the air."AP Newswire, Nov 13

Here are North Carolina's standards for voting systems:
§ 163-165.7. Voting systems: powers and duties of State Board of Elections.

(a) Only voting systems that have been certified by the State Board of Elections in accordance with the procedures and subject to the standards set forth in this section and that have not been subsequently decertified shall be permitted for use in elections in this State.

...The State Board may certify additional voting systems only if they meet the requirements of the request for proposal process set forth in this section...

(1) That the vendor post a bond or letter of credit to cover damages resulting from defects in the voting system. Damages shall include, among other items, any costs of
conducting a new election attributable to those defects.

(2) That the voting system comply with all federal requirements for voting systems.

§ 163-165.9A. Voting systems: requirements for voting systems vendors; penalties.

Sunday, September 19, 2010

2 Simple Rules for Voters in Instant Runoff Voting for NC Court of Appeals

1. Voters must not rank the same candidate twice, and
2. Use strategy, make or follow endorsements, rank potential winners only, vote for at least 2!

Read further to learn how to avoid having your vote be a total shot in the dark.


Do not rank the same candidate more than once!
You can rank up to 3 DIFFERENT candidates out of the 13
Ranking a candidate more than once means your 2nd and 3rd choices will not count! The voting machine is not set up for IRV so it will not warn you if you make a mistake in ranking.


Statewide endorsements and or sample ballots are a must.
This election will either be a crap shoot or it will be decided by the most effective endorsements. The endorsements should keep in mind that if there is a runoff, only the 2 candidates with the most votes will advance. Vote for 2 choices at least. If you fail to rank either of the 2 candidates, you wil not have a say in the "runoff".

Who should you rank 1st, 2nd or 3rd? With IRV, a candidate may win by getting the most 2nd choice votes. Keep that in mind as you head to the polls.

Stay with the pack to win. Basically there must be a statewide strategy where everyone stays in line behind one particular candidate and then rank an additional candidate who also is qualified or has a reasonable chance of getting votes.


a) Voters should vote for party favorite and or their own personal favorite.
Dale Sheldon-Hess, an expert on various election methods advises:
The voters, if they have a preferred party, should name their party's candidate first; or, if they're ABSOLUTELY certain there's no chance of them winning, name their personal-favorite candidate first, and the party favorite second.

b) Rank only candidates you think will get enough votes to make the runoff:
Professor Steven J Brams, a Professor of Politics and Election Method expert advises:
The only sensible strategy is not to rank high candidates one knows and prefers but only candidates one prefers who might make the runoff. If one votes for others, the probability that your vote will be transferred to your preferred of the two front runners is slim.

Steven J. Brams Phone: (212) 998-8510
Dept. of Politics FAX: (212) 995-4184
19 West 4th St., 2nd Floor E-mail:
New York University
New York, NY 10012

Voter education, endorsements, videos, and handing out palm cards at the polling places would be helpful.

To receive updates by email visit this link