The article's title is misleading, and it misrepresents the content and conclusions of some of its sources.<p>The 28 million ballots in question aren't "missing"; they're ballots that were mailed to voters and weren't used.<p>In some cases the voters in question likely didn't bother voting (common in local elections) or simply forgot (I've done this when I accidentally stacked other mail on top of a ballot and then didn't find it again until too late). In other cases, voters may have moved and failed to update their voter registration. This situation isn't unusual or alarming, and the opportunities for fraud aren't nearly as plentiful as the article tries to imply.<p>The article cites a 2008 report by the CalTech/MIT Voting Technology Project as recommending that states abolish absentee voting due to fraud concerns, but that report contains only a single paragraph proposing a theoretical _potential_ for fraud due to "unsecured" mail channels. It provides no evidence that this is common or has even occurred, and in fact devotes significantly more space to discussing the inconvenience of mail-in punch-card ballots with Styrofoam backings — something I had never heard of despite having voted by mail for almost two decades.<p>The article also cites a 2008 Reed College study, which again raises theoretical concerns about fraud but offers no substantive evidence that it's an issue, and in fact states that "A report commissioned by the federal Election Assistance Commission, authored by Tova Wang and Jed Seberov, concluded, on the basis of interviews with experts in the area, that there was little evidence of voter fraud."