In Minnesota, Norm Coleman's lead has
grown back to 210 votes. (At the start of the recount
process, Coleman was up 215 votes, but during the recount Franken
had narrowed the gap to as low as 120.) The catch is that, with
77 percent of the recount in, the campaigns have challenged more
than 3,000 ballots, so it's hard to say for sure who is gaining
or losing votes.
Nate Silver has done some statistical work suggesting that
Franken's gains have been understated, and that actually he's the
favorite to win. His analysis is based on the idea that Franken
has gained in counties in which there are either no challenges or
few challenges, and that Coleman's gap widened as challenges
increased dramatically. But I'd take a bit different of an
approach. If Coleman's lead stays in the 200 vote range, it would
seem difficult for Franken to gain that many votes in a universe
of just a few thousand challenged ballots. Keep in mind that
these ballots have already been ruled on once by an elections
judge, so you'd have to assume that an overwhelming majority of
the challenges will fail, and right now the campaigns are
challenging roughly the same number of ballots. So, if Coleman
maintains his current margin as the Canvassing Board meets to
rule on the challenged ballots, I think he'd be an pretty good
position.
Meanwhile, in Georgia, Saxby Chambliss looks
like he has the edge in next week's run-off, with a new PPP
poll showing him crossing the magic 50-percent number with a
52-46 lead. An internal Democratic poll shows Chambliss below 50
percent, but still leading 48-46. I would add that the final PPP
poll prior to the election showed Chambliss with a 48-46 lead,
and he ended up winning the initial vote by a 49.8-46.8 margin.
So in other words, the poll didn't skew Republican. I continue to
believe that without Obama on the ballot, Martin won't benefit
from super-sized black turnout, and that the state's traditional
voting universe favors Chambliss.