Wisconsin Supreme Court candidate JoAnne Kloppenburg declared victory on election night when preliminary, unofficial returns put her 204 votes ahead of Justice David Prosser.
Fifteen days later, she sought a statewide recount after the official county-by-county returns showed her trailing by more than 7,300 votes out of 1.5 million cast.
In explaining her decision, Kloppenburg told reporters that a recount may not get her over the top, but would shine a light on "an election that right now seems to so many people to be suspect."
She also went on the offensive, raising questions about the legitimacy of the vote count around the state.
"There are legitimate and widespread anomalies," said Kloppenburg, a state Justice Department attorney, "and widespread questions about the conduct of this election, most visibly in Waukesha County, but also in counties around the state."
The Waukesha County problems are well known: A vote-tallying glitch belatedly boosted Prosser’s total in the official count by almost 7,600. County Clerk Kathy Nickolaus, a Republican in a GOP stronghold, didn’t report her error for more than a day after the election.
Given the outrage over the Waukesha incident, Kloppenburg seems on solid ground suggesting that there are "widespread questions" about the election.
But are there "widespread anomalies" in the statewide count?
That’s a much different claim and one we can check.
In order to justify her claim, we think Kloppenburg has to show that numerous questionable situations remain that are "irregular" or "abnormal" -- a dictionary definition of "anomalous."
In addition to the Waukesha situation, Kloppenburg at the news conference cited an "undervote" in Milwaukee and Racine; ballot shortages and long lines in Fond du Lac and several other municipalities; and changes in the unofficial vote totals in Winnebago County.
Let’s take a look:
"Undervotes": This is when a voter chooses not to vote in all the races listed on a ballot -- creating an "undervote" in one race compared with another. For instance, this year in the city of Milwaukee, the Milwaukee County executive’s race drew about 2,000 more votes than the Kloppenburg-Prosser race at the top of the ticket.
Kloppenburg campaign manager Melissa Mulliken said Racine and Milwaukee had an unusually high undervote rate in the statewide court race compared with local races there.
"You wonder why," Mulliken said. "Did voting machines misread ballots?"
But in the city of Milwaukee, the 2011 court contest was not the largest court-race undervote in even the last three years. In 2008, when Michael Gableman defeated incumbent Louis Butler for the high court, the Scott Walker-Lena Taylor race for county executive drew 3,500 more votes.
A review going back further shows the court races sometimes draw more and sometimes less than other contests, depending on the competitiveness of the races involved.
In Racine, the Kloppenburg-Prosser race drew fewer votes than the Racine mayor’s race. It was about 150 votes short, or 1 percent.
That, too, has happened before: In 2009, Justice Shirley Abrahamson’s breeze to re-election drew substantially fewer votes than a special primary election for Racine mayor. But in 2008, the Gableman-Butler race drew just slightly more than a statewide referendum in Racine.
Not much of a pattern there.
Neil Albrecht, deputy director of the Milwaukee Election Commission, said undervoting is not really an anomaly.
"It’s more a reflection with people's comfort and familiarity with certain contests and the candidates," he said.
UW-Madison political scientist Barry Burden agreed.
"‘Dropoff’ is the term we use," said Burden, who studies election administration. "It doesn’t surprise me at all. That’s reasonable to expect when you have a local race that is grabbing a lot of attention."
(Local conditions played out the other way in another Democratic hotbed, Dane County. The Milwaukee Journal Sentinel’s Craig Gilbert reported that 10,000 more people voted in the court race than in a county exec’s race that was not even close.)
Ballot shortages: The unusually high voter turnout in the high court race -- at least 14 points higher statewide than predicted -- fueled ballot shortages in various communities. Kloppenburg cited Fond du Lac County. State election officials ordered the clerk in the city of Fond du Lac to issue substitute ballots because of shortages, and the resulting lines at touchscreen machines and delays in voting.
Our research found other shortages in the counties of Eau Claire, Sheboygan and Winnebago. Together, they involved more than 2,000 votes on photocopied ballots that had to be hand counted.
How unusual is this? There are mixed views.
Burden said the use of photocopied ballots is fairly common but rarely noticed unless a race is close.
It’s common enough for the state to have a law that sets out the procedure for issuing substitute ballots when shortages occur, said Reid Magney, spokesman for the state’s Governmental Accountability Board.
One clerk, Lisa Freiberg of Fond du Lac County, said it was "very unusual" in her county to resort to copying paper ballots. Fond du Lac City Clerk Sue Strands, who orders ballots, agreed.
Strands said she considers her vote count solid. She said if there was an "anomaly" it was limited to four instances in which voters tried to put their copied ballot into voting machines. They jammed.
Kloppenburg’s campaign says the question is how such ballots were counted.
There is some concern about reliable counts of photocopied ballots, because they can’t be run through voting machines, election clerks say.
That necessitated some hand-counting on a busy election night.
Discrepancies in unofficial results: It’s clear the big Waukesha County vote tally error on election night was unusual. PolitiFact Wisconsin earlier examined a Nickolaus claim that errors in that part of the process are "common." We found that her error was extreme but not unprecedented, and we rated her claim Barely True.
It’s important to note that errors such as the Waukesha County tally and the 1,100-vote miscommunication in the unofficial tally from Winnebago County are reviewed -- and typically corrected -- in the official vote canvasses done by the counties.
That process -- plus a special state probe of Waukesha County -- is now complete and state officials say the Waukesha and Winnebago discrepancies have been cleared up.
So if that verification process is already done, does that leave Kloppenburg with much justification for including Waukesha on her problem list? She points out the state will do a deeper investigation of Nickolaus’ past work -- and that the preliminary state probe did find a handful of discrepancies in the county’s canvass of write-in votes.
So, has Kloppenburg supported her claim?
Her "undervote" claims do not hold up; they are not anomalous based on historical patterns. Errors in the unofficial tally were mostly routine and by now have likely been corrected by the standard post-election canvasses. Waukesha County’s situation was rare, but it is only one county, and the big discrepancy there has already been cleared up, according to state elections officials.
In Kloppenburg’s favor: Ballot shortages -- though not rare and not by themselves evidence of miscounts -- did happen in multiple areas and are by no means the norm. And the Waukesha County situation, while already investigated at length, still has a few loose ends, and questions remain about Nickolaus’ methods.
So, Kloppenburg’s claim had an element of truth but skipped critical facts that would give a different impression. Some might call it mostly false. But on the Truth-O-Meter, that’s the definition of Barely True.
Editor's note: This statement was rated Barely True when it was published. On July 27, 2011, we changed the name for the rating to Mostly False.