I'm surprised that Smith pulled ahead late in Oregon. I guess the outstanding precincts in Portland was favorable to him. Coleman is ahead by less than 1000 votes, and it looks like Stevens won.
Actually, most of Portland hasn't reported yet. Look at the presidential race, and at previous Multnomah turnout. There's some sort of mistake there in the Senate race. I think Merkley will win in the end, making 58, which I guess isn't too bad.
The way Oregon tallies their mail in ballots thoroughly baffles me. I remember from the primary that a county's precincts reporting could go from 33% to 100% with little or no new results added.
For what it's worth, I wouldn't be so sure that Merkley wins. Smith's lead has been fairly stable throughout the night and into this morning. Then again, see my previous paragraph.
Still, how do you get to 58? Franken is losing with 100% in, pending a recount. Corrupt Alaskan Senator Ted "not-convicted yet" Stevens is up by 1.5 points with all but 3 (likely small) precincts in and 40,000 absentees (which usually break slightly Republican). And Chambliss will likely win outright or face a run off that he'd likely win anyway.
Even if Merkley wins, Democrats + the 2 independents should be at 57, not 58.