Update: Our friend Max Hailperin, a math/computer
science professor at Gustavus, notes that ballots mistakenly cast in the 8B are likely just
to be pulled in three precincts.
They are:
Alexandria Ward 1- Precinct 1 where 26 voters who signed into Precinct 1 were given ballots for Precinct 2, thereby giving them an incorrect opportunity to vote in the Franson-Cunniff race.
Similarly, in Ward 5 (Alexandria City Hall), it looks like 6 voters in Precinct 1 were given the Precinct 2 ballot, allowing them to incorrectly vote in the Franson-Cunniff race. In W1P2, Cunniff is reported to have a 753-603-1 lead; in W5P2, it's 570-392-1.
Alexandria's Ward 3, which voted at New Life Christian Church. There, three more ballots were counted than voters signed in. The Cunniff-Franson-Write In counts there are 754-535-2.
A source well-informed in election procedure tell Bluestem that these are indeed where the ballots will be pulled according to the errors.
The odds are looking better for Franson in the luck of the draw, rather than from Cunniff, who did very well in Alexandria--possibly among voters who had no business voting for either candidate.
[end update]
Over at LeftMn, Tony Petrangelo reports in Mary Franson – Bob Cunniff outcome even less certain now that a report of obvious errors may put the One-vote Wonder of the House 8B race in question:
. .Anyway, the obvious errors in question appear to be voters in districts other than 8B, who were given an 8B ballot, 32 of them in all.
To correct this, Petrangelo reports, state law dictates:
. . . 32 ballots will be randomly pulled out of the pool of all the ballots prior to the results being certified. This means that there’s a pretty good chance the final margin will not be one vote after certification.
Going back to the link at the beginning of this post, the Douglas County Canvassing Board will meet again on Wednesday the 21st to do the after the fact reconciliation, that is unless one of the two candidates gets a court to step in.
Bluestem called the Douglas County Auditor's office to learn whether or not the 32 ballots would come from all of HD 8B or just Douglas County. We were told that the Douglas County Canvassing board would only be pulling from the ballots in their possession--not from the larger pool of the district.
Given that Cunniff outperformed Franson by over 1000 votes in Douglas County-- 6861-5854--there's very slight odds that more Cunniff votes than Franson votes will be pulled, although the randomness of the draw makes those slim odds. Maybe stats or math-inclined reader could figure the odds--AP reporting 35 not 32--of a Cunniff victory when that X ballots will be pulled in a pool of 12715 votes and one candidate has a 1007 vote advantage.
The auditor's office told Bluestem that a petition had been filed in court by one of the candidates.
A call to the Douglas County Court Administrator's office learned that Mary Franson had filed a petition. We are waiting for a scan of the court document.
Here's the petition--it concurs with the AP count, although onl 32 ballots are in non-8B precincts:
Petition of Mary Franson in the Matter of the Obvious Error --Case 21-CV-12-1973
The Associated Press is reporting that the error is 35 ballots, and there will be a hearing today.
Earlier this month, a tie for county commissioner in Meeker County was settled by a random ballot being pulled, the Litchfield Independent Review reported in Voting error goes in Fenrich's favor:
A voting error at a precinct in Litchfield Township broke a tie in the race for District 2 County Commissioner, resulting in the Meeker County Auditor’s Office Friday declaring Dale Fenrich the winner by one vote.
Unofficial election night results listed the two candidates in a dead heat, with 1,069 votes each for Fenrich and incumbent Dave Gabrielson. However, after reviewing results, staff from the auditor’s office identified a discrepancy between the number of voter signatures on the roster and the ballots counted for Litchfield Township, according to a news release from Meeker County Auditor Barb Loch on Friday morning. The discrepancy was also noted by the Litchfield Township election judges on their incident log.
On Friday, Loch said the log indicates a voter accidentally received and cast two ballots. She said the log states, "Voter received two ballots. Voted on front of one and back of the other. Deposited both into the scanner before election judge could intervene.”
Loch said voting errors are not unusual, though they rarely affect the outcome of a vote. "This one is notorious because it was dead-on even," she said.
Loch said her office followed state law requiring the number of voter signatures to match the number of ballots counted in a precinct. In the case of Litchfield Township, there was one extra ballot, Loch said. On Thursday, one ballot was randomly selected and removed from a box containing all ballots cast in Litchfield Township. Two election judges witnessed the process, Loch said.
In removing the extra ballot, the results for the County Commissioner District 2 race now show Fenrich with 1,069 votes and Gabrielson with 1,068 votes. The results were officially approved Friday at a County Canvas Board meeting.
This should be interesting.
Photo: Mary Franson who, with Bob Cunniff, totally defines randomness.
Max Hailperin can correct me if I'm wrong, but i calculated the odds of a Cunniff lead post-reduction using a hypergeometric distribution function. I calculated it in two cases: for 32 ballots and for 35 ballots. For 32 ballots we would need to pull at least 17 Franson ballots for Cunniff to take the lead. That gives a 26.455% chance of a Cunniff lead. For 35 ballots it is more interesting as it presents the possibility of a tie. A Cunniff win would occur with a maximum of 16 of his ballots being pulled, which occurs 31.8007% of the time. A tie would require exactly 17 Cunniff ballots and 18 Franson ballots to be pulled. The odds of this happening are 10.9358%.
Posted by: Archer Dem | Nov 19, 2012 at 11:19 AM