makes a great point
about Alaska's U.S. Senate vote count.
However many ballots Mr. Miller elects to challenge — and he could challenge every single ballot if he wanted to — only a certain fixed number of these challenges will have any realistic chance of being upheld. The fact that, on the first day of ballot counting, Mr. Miller’s campaign was already indulging in what seems to have been quite a number of frivolous challenges, and despite this, was still not able to challenge ballots at a rate consistent with what would eventually be required to deny Ms. Murkowski re-election, speaks quite poorly to his chances.
Miller's got an advantage in the PR war as long as people take the state's count of "challenged ballots" as gospel, and assume none will be counted. (It's very possible that the count ends with Miller up by a hundred or so votes with thousands of ballots technically challenged, giving Miller the chance to go on TV and argue that he won.) But lots of them will be counted. Look at this ballot , challenged by Miller and photographed by the Anchorage Daily News.
That's a vote for "Lisa Murkowski" by someone with sort of bubbly handwriting. There's no way that challenge survives. If even 10 percent of the challenged ballots have "problems" like this, Murkowski adds hundreds of votes back to her total.
UPDATE: More from the AP:
Shortly after the second day of write-in ballot counting began in the race, a Miller observer challenged a vote for Murkowski that appeared to have her name spelled and printed correctly, though the "L" in "Lisa" was in cursive handwriting.
Really, if only one in ten of the challenged ballots are over the clarity of correctly spelled names, Murkowski nets hundreds of votes. If it approaches 50 percent of ballots, she's winning easy.