When one candidate loses a close election, it’s pretty common for his or her supporters to come up with some reason why the election was “stolen.” It happened for Democrats after John Kerry lost the 2004 presidential race, it happened for Republicans after Norm Coleman narrowly lost a 2008 Senate race in Minnesota, and it’s happening again for Republicans now that Pat McCrory appears to have lost this year’s North Carolina governor’s race. These claims are generally judged to be baseless.
So when you read Gabe Sherman’s New York magazine report that “a group of prominent computer scientists and election lawyers” is urging Hillary Clinton’s campaign to request recounts because they believe “they’ve found persuasive evidence that results in Wisconsin, Michigan, and Pennsylvania may have been manipulated or hacked” in Donald Trump’s favor … well, then you should take a deep breath and look at the evidence that’s actually presented.
And there’s not much. In fact, there’s a grand total of one specific claim from this analysis that’s reported in Sherman’s piece, sourced to an anonymous person briefed on a recent call in which this group made its case to Clinton campaign bigwigs:
- “The academics presented findings showing that in Wisconsin, Clinton received 7 percent fewer votes in counties that relied on electronic-voting machines compared with counties that used optical scanners and paper ballots. Based on this statistical analysis, Clinton may have been denied as many as 30,000 votes; she lost Wisconsin by 27,000.”
But this alone proves nothing. For one, it’s quite possible that electronic-voting machines were more often used in counties that were already more likely to support Trump — for instance, counties with lots of low-education or rural white voters who strongly supported Trump across the whole country. Indeed, back-of-the-envelope insta-analyses by Nate Silver and Nate Cohn suggest this is the case, though it’s hard to rebut claims that haven’t been made publicly.
Further reasons for skepticism:
- The swing toward Trump in Wisconsin was part of a swing toward him across the whole Midwest — including Iowa and Minnesota, two states next to Wisconsin that are demographically similar to it, and which both use only paper ballots counted by machines that are not connected to the internet.
- Michigan also uses paper ballots, so there would have to be manipulation of a different sort going on there. The article doesn’t shed light on what evidence there might be regarding Michigan (and Pennsylvania — to change the Electoral College outcome, major fraud would’ve had to have happened in all three states).
- According to Sherman, Clinton campaign chair John Podesta and the campaign’s top lawyer Marc Elias were given all this data five days ago. They have apparently done nothing with it, even though the deadlines to request recounts in states are approaching quickly. That seems to suggest they did not view the analysis as credible.
So … be skeptical. Maybe this group of “prominent computer scientists and election lawyers” is sitting on more persuasive evidence than this. If so, they should post it publicly and let their claims be analyzed, rather than letting vague rumors swirl. But you definitely shouldn’t believe a vague, fantastic-sounding claim about a stolen election unless serious, solid evidence emerges to back it up, and independent experts validate how that evidence is being analyzed.
Update: One of the computer scientists mentioned, J. Alex Halderman, has now posted his views on Medium. They are much less dramatic than the secondhand account in New York magazine. “Were this year’s deviations from pre-election polls the results of a cyberattack? Probably not. I believe the most likely explanation is that the polls were systematically wrong,” he writes. But he argues that since hacking is possible, it’s simply prudent to conduct a recount in close states where there’s a paper trail, which is reasonable enough.