By Joe Uchill - 11-23-16 12:57 PM EST
Liberals searching for answers to Donald Trump's surprise victory in the presidential election have latched on to a new explanation: Hillary Clinton didn't really lose.
A growing "audit the vote" movement, which has garnered high-profile backers like "Avengers" director Joss Whedon and actress Debra Messing, argue election tampering or incompetence may have led to Trump being declared the winner.
They are demanding that the Department of Justice and election officials check voting machines and vote totals to make sure the results in the states of Wisconsin, Michigan and Pennsylvania are accurate.
Trump has been declared the winner in Wisconsin and Pennsylvania and has a lead in Michigan, where some media groups have declared him the winner.
If the results in all three states were for the Democrat, Clinton, who has a lead in the popular vote of nearly 2 million over Trump, would also have the Electoral College lead.
"Demand an audit. Make the call," Whedon wrote in a tweet to his 107,000 followers on Tuesday that included a photo of Clinton overlaid with the message: "SHE WON."
The idea that widespread election tampering decided the election, with evidence of enough uncounted votes that the results could be overturned, would appear to be far-fetched.
Trump's transition is in full effect, and Clinton conceded the results in the early hours of Nov. 9. No major Democratic officeholders appear to be backing the social media calls for an audit.
Clinton is behind in the state of Pennsylvania by 68,000 votes and in Wisconsin by about 27,000 votes.
Many of those calling for an audit have focused on the idea that the Russian government may have manipulated the outcome. Russia was implicated in data breaches of a number of Democratic institutions and hacking the email account of Clinton's campaign manager.
New York Magazine reportedTuesday that a team of researchers including voting-rights attorney John Bonifaz and J. Alex Halderman, the director of the University's Center for Computer Security has even spoken with the Clinton campaign last Thursday about contesting the results in Michigan, Wisconsin and Pennsylvania.
According to the New York Magazine report, the researchers argued that Clinton performed 7 percent better in counties of Wisconsin with easily auditable optical scan or paper ballots than she did in counties with electronic machines that leave no paper trail.
They estimate Clinton might have been robbed of 30,000 votes through some form of vote tampering, enough to swing Trump's current 27,000 vote lead, though they have not released their research.
Without the research, it is impossible to know what variables the researchers did or did not account for, and that team's argument has not been released for Pennsylvania.
Both the voting results and the technical peculiarities of hacking election machines cast some doubt on whether the theory that Wisconsin was hacked holds water.
Clinton's final voting percentages were similar across the state, county for county to losing 2014 Democratic gubernatorial candidate Mary Burke. No one has accused that election of being rigged or unjust.
In nine of the 13 counties Clinton won, she was within five percentage points of Burke's performance. Trump ultimately only flipped three counties Burke won, none of which Burke won by more than 5 percentage points.
Nate Silver of FiveThirtyEight speculated in a recent posting about what might account for the differences. He wrote that one reason parts of Wisconsin with optical scan and paper ballots performed better for Clinton might be their relative wealth. Rural counties tend to use the unauditable machines.
In Ashland, a rural county Clinton won that does not use election systems that leave a paper trail, Burke outperformed Clinton by more than 10 percentage points. Ashland is the second least affluent county in the state.
The number of missing ballots is suspiciously low for what an election hacking effort would need to be.
Until the election, polling indicated Clinton would win Wisconsin by nearly 200,000 votes. An outside attack would have to likely targeted well more than the researcher's 30,000 vote estimate.
But election hacking on the scale that could swing a normal election is a gargantuan feat. Though experts agree that hacking an individual voting machine is easy, hacking them at a large scale is far more difficult. Voting machines are not networked to each other or connected to the internet. Nearly all machines need to be hacked in person, one at a time.
Wisconsin, for example, has 1,800 voting municipalities, which often have a pair or more of machines.
The difficulty behind an election hacking effort would not rule out other forms of tampering, such as an insider threat.
The Wisconsin Election Commission did not answer phone calls Wednesday, the day before Thanksgiving.