Despite trailing his opponent by slightly more than two hundred votes, Democratic challenger Al Franken stands a strong chance of passing Sen. Norm Coleman during the upcoming recount, according to at least one prominent political scientist. Professor Michael C. Herron of Dartmouth College, has put together a new study of the voting patterns in Minnesota, in the process determining that the majority of voters who cast unrecorded ballots in the Senate race were likely Franken supporters. "If someone put a gun to my head and said, 'You have to bet,' I would bet Franken," Herron said, when reached by phone. "It won't be a wipe-out. Two hundred votes is effectively tied. We just know that, in this case, Democrats tend to [screw up their ballots] more often [than Republicans]." In Minnesota, the "intent" of the voter is considered during recounts. According to Herron's analysis, of the 2.9 million people who went to the polls in Minnesota, there were approximately 34,000 residual voters in the Senate race. In other words, there were 34,000 more ballots cast than total number of recorded votes for all the Senate candidates. Why the difference? A good portion of voters, Herron concludes, voted in the presidential election but deliberately did not vote for a Senate candidate. These people won't matter when it comes to a recount. There is, however, a portion of the 34,000 who intended to vote for one of the Senate candidates but messed up. Voters were supposed to fill in the circle next to the name of the candidate they supported. Some, however, marked X's. Others circled the name itself or crossed out the names of candidates they didn't like. This group is key to determining the Minnesota Senate victor. In his study, Herron looked at the figures from the 2006 congressional election and the 2008 presidential election to determine which areas of the state have the most residual voters. By isolating these areas, Herron could determine which group was most likely to have wanted to vote one way but failed to cast their ballots properly. He found that the majority of residual voters came from two, not necessarily distinct places: African American communities and traditionally Democratic communities. With the former, he theorized, there was likely a "turnout surge" -- many people went to the polls to support Barack Obama and no one else (or at least not Franken). The latter, however, contained voters who "almost certainly intended to cast a vote in the Senate race [and likely for Franken] but for some reason did not do so." Story continues below How big that group is, is crucial. And a way to figure it out is to first look at how Barack Obama and John McCain fared in the state. According to Herron there was an approximately 0.34 percent residual vote rate in presidential race voting among Minnesotans. This means that of the 2.9 million votes cast for a presidential candidate, nearly 10,000 individuals wanted to vote but screwed up. There may have been people who wanted to vote in local and congressional contests, but not the presidential race. But this group is likely quite small. In other words, there are probably somewhere between 8,000 and 10,000 Minnesota voters who had clear problems filling out a ballot when voting for all contests (Senate and presidential). Many of these individuals, moreover, hailed from Democratic communities. "Ultimately, the anticipated recount may clarify the relative proportions of intentional versus unintentional residual voters," writes Herron. "At present, though, the data available suggest that the recount will uncover many of the former and that, of the latter, a majority will likely prove to be supportive of Franken." All Franken needs is to win more than 207 votes from this group than Coleman, and he will take over the Senate seat.