Of the 950+ precincts showing in the spreadsheet, there are 221 total ballots in dispute by one side or the other. I would be reluctant to say that Franken has netted another 43 votes until those 221 are dealt with.

The way I understand it, the 43 net votes for Franken is outside the 221 challenged votes. In other words those votes count and the challenged votes are put aside. Until the end. I am guessing if one of the candidates has a lead over and above the number of challenged votes, the loser will concede because even if all challenged votes went their way, the final result wouldn't change.

Basically then I disagree with John above and I think it is fair to say that Franken moved 43 votes close to Coleman.

One interesting question raised by the slideshow on the MPR site is the relative quality of the challenges by the respective sides' observers. I hope that one of the strategies is not to flood the canvassing board with lots of questionable challenges so as to facilitate the narrative of the "canvassing board is biased in their interpretation of the challenges." In any case, it will come down to the state canvassing board and adjudicating the challenged ballots - at least it appears that way.

Comments by local elections officials seem to suggest a belief that many, if not most of the challenges will be rejected. Adding those votes to each total would create an upper limit on the totals and may more closely resemble the final result.

All I'm saying is that the very narrow difference in votes (based on the initial election results) is close enough that each and every disputed ballot could become critical.

If the number of disputed ballots is less than the difference between the top two candidates at the end of the process, then it is a whole different situation. I have no issues with your disagreement, Dean but I'm going to take a "wait and see" approach to this.

Wait, why do you subtract out the canvass votes? I thought the recount votes are all that matters now.

There was a report in the Duluth paper today that older voting machines used in some Eveleth precincts resulted in substantial undercount of votes that were picked up in the hand recount. Both Franken and Coleman added votes as a result, but because these were DFL precincts Franken picked up more.

In theory, mistakes in counting ought to be roughly equal and resort in very little change. The only real hope Franken has is that there was a pattern of undercounting in DFL precincts, similar to what happened in Eveleth.

The other thing is that the recount in 1962-63 that ended with Karl Rolvaag defeating Elmer Anderson is a good model for what is happening here.

My father, who covered the capitol for the Saint Paul Dispatch (the evening version of the Pioneer Press) during the recount, always said that Rolvaag won because the DFL was much more aggressive during the challenged ballot process. If this comes down to the challenged ballots, I think it is going to get very nasty. And if Coleman loses, he is going to blame it on partisan politics.

I'm baffled by this math. I can certainly add and subtract, but how can Franken "net" 43 votes when the count for all candidates began yesterday with zero? The recount doesn't result in adjustments to the last pre-recount tally. It produces new numbers that are meaningless until all the votes have been counted.