Now we turn to the 1,346 absentee ballots that were identified by election officials as improperly rejected. The two campaigns must agree that a ballot was wrongfully discarded before it can be counted — and GOP Sen. Norm Coleman’s campaign is not making that easy.

The Franken team has agreed that all 1,346 ballots should be counted. But the Coleman camp is recommending the inclusion of just 778 of these ballots, as well as 67 ballots that were not flagged by election officials as improperly rejected. And guess what? The ballots are mostly from Republican-leaning areas!

From the Star Tribune:

Coleman’s proposed additions skew heavily toward suburban and rural counties where he did best in the election.

I explained in my last post why Coleman wouldn’t want all 1,346 ballots to be counted, but he seems to be going a bit overboard here. His campaign has given no indication that there’s an overarching logic to its choices, other than cherry-picking for Coleman’s own benefit.

The Coleman team will be taking a closer look at these ballots today, tomorrow and Friday.

UPDATE: The Star Tribune has analyzed 93 percent of the absentee ballots in question and determined that Franken is likely to gain from them. More motivation for Coleman’s selective approval of the ballots.

The Coleman team argued that these ballots, whose originals and duplicates could not be properly matched, had been counted twice, and that the canvassing board should not certify the election results until this issue was resolved. Most of these ballots came from Minneapolis, which voted overwhelmingly (78 percent) for Franken, and their exclusion could have helped Coleman overcome his current deficit of 46 votes.

But the court ruled unanimously that the Coleman campaign lacked sufficient evidence of double counting, making it likely that Franken will be certified as the winner sometime in January. Coleman’s lawyers have said that they will contest the certified result.

The court also granted the two campaigns more time to identify absentee ballots that were wrongly rejected.   The deadline for the allocation of the ballots was extended from Dec. 31 to Jan. 4. The extra time was alloted to allow for more fair and more effective counting. The decision also makes it “highly unlikely” that Minnesota will be able to seat its second senator on Jan. 6, according to a Coleman lawyer.

The 1,500 or so absentee ballots deemed wrongly rejected during the recount are expected to add slightly to Franken’s lead. For this reason, the Franken campaign has been pushing to include them, while the Coleman team has resisted.

So Franken inches closer to the finish line, but we won’t know for sure which campaign will reach it until next month. Stay tuned — or better yet, enjoy the holidays.

The Coleman campaign, getting a bit desperate as Franken emerges as the likely winner, this morning began challenging the challenges, arguing that 16 challenged ballots were not properly ruled on. But the board would have none of it, and it stuck with its initial rulings. After reviewing these 16 ballots, it expressed even “more confidence” in its decision-making. Board member Eric Magnuson, who also serves as chief justice of the Minnesota Supreme Court, warned against further challenges, saying, “I think you have a sense of how we’re disposed to rule on these things.”

Last night, Franken was projected to lead by 48 votes at the conclusion of the process, consistent with his campaign’s earlier prediction of a 35- to 50-vote lead. That projection held as the board completed its review.

Still on the table:

1. Today’s Minnesota Supreme Court ruling on double-counted ballots, as alleged by the Coleman campaign.

2. Around 1,500 improperly rejected absentee ballots that should be counted, though the implementation of this process will be quite contentious. Both campaigns must agree that a ballot was improperly rejected for it to be counted — and given their track record, agreement between the two campaigns is hard to come by.

3. Lawsuits, lawsuits and more lawsuits.

Of course, it’s nearly inconceivable that this lead will hold. Two factors could easily shift the balance: the Coleman campaign’s allegation of duplicate ballots, and the Franken campaign’s court-backed push to count around 1,500 improperly rejected absentee ballots.

As I argued earlier today, if Coleman is successful in excluding the alleged duplicate votes, he is likely to end up with a slim lead (around 26 votes, by my calculation). But that’s a big “if” — and it would probably be more than offset if all of the absentee ballots in question are counted, as Franken is expected to gain around 100 votes from this process.

Tomorrow is the big day when the canvassing board is expected to complete its review of the challenged ballots and the Minnesota Supreme Court is set to rule on the duplicates. Stay tuned.

The Franken team’s prediction, announced on Saturday, is actually a bit more modest than The Star Tribune’s projection of a 78-vote Franken victory. From The Hill:

“We now have enough data that I can say with a very high degree of confidence that Al Franken will win this count and be seated as the next senator from Minnesota,” Franken attorney Marc Elias told reporters Saturday.

However, the Coleman campaign has argued that a number of ballots were counted twice, due to an improper correlation of original and duplicate ballots. According to the St. Paul Pioneer Press, Coleman seeks to “have about 130 ballots, mostly from largely Democratic Minneapolis, tossed out.”

I ran some calculations based on these numbers, and here’s what I found:

Minneapolis gave Franken 141,675 votes on Election Day, compared to 39,309 for Coleman. That’s 78.3 percent for Franken, and 21.7 percent for Coleman.

Let’s assume, for simplicity’s sake, that there are exactly 130 ballots in question, that they’re all from Minneapolis and that these voters cast their ballots along the same lines as the city as a whole — big assumptions, I know, but bear with me. In this case, Franken would lose 102 votes if these alleged duplicates are discarded and Coleman would lose 28 — a net gain of 74 votes for Coleman.

This would leave Coleman with a lead of 24 to 49 votes, based on the Franken campaign’s prediction. Using The Star Tribune’s model, Franken would retain a lead of just four votes (coincidentally, this is the exact margin that the Franken team predicted more than two weeks ago.)

However, this is far from a done deal, because Franken’s lawyers don’t plan to go down without a fight. From the Pioneer Press:

Elias said the Coleman campaign’s duplicate-original contention is sheer fiction.

It also is unfair, he said, because while the Coleman campaign was able to challenge and separate out originals for which no duplicates could be found in Minneapolis, the Franken campaign was not allowed to do so in several more Republican areas.

The only resolution would be to reopen the recount in all 87 Minnesota counties, Elias suggested.

So there’s still a lot up in the air. Even if everything goes Coleman’s way on this issue and he takes a lead, Franken could still win easily if all of the improperly rejected absentee ballots are included, as they are likely to be.

The bottom line: by the end of the day tomorrow, when the canvassing board completes its review of the challenges and the Supreme Court rules on the duplicates, we should have a much clearer picture of the outcome of this election.

Withdrawn challenges have not yet been processed. Since Franken withdrew about 400 more challenges than Norm Coleman, Coleman will make up significant ground once these are taken into account. Exactly how much ground depends on the nature of the challenges — a withdrawn challenge to an initial non-vote (when the election official decided that voter intent could not be discerned, or that the vote went to a third-party candidate) will result in no change, whereas a withdrawn challenge to a vote for a certain candidate will result in a gain of one vote for that candidate.

For example, let’s assume that half of each campaign’s withdrawn challenges were initially aimed at turning votes from the other candidate into non-votes, and the other half were aimed at turning non-votes into votes for the campaign’s candidate. In this case, if Franken withdrew 400 more challenges than Coleman, then Coleman would gain 200 net votes when the withdrawn challenges are processed. If, on the other hand, three-quarters of the withdrawn challenges were to initial votes for the opposing candidate, then Coleman could expect to net 300 votes.

With Franken’s lead currently at 266, the makeup of these withdrawn challenges could make all the difference in determining which candidate will come out ahead at the end of this challenged ballot review.

Then, of course, we get to the rejected absentee ballots, which are expected to help Franken, and the inevitable litigation, which will more likely come from the Coleman camp.

This election is by no means decided, although I still feel good about my bet on Franken, even with the 2-to-1 odds I gave our CEO.

Side note: InTrade’s been monitoring the race, but their odds have lagged a couple days behind mine. On the 15th, they put Franken’s chances of winning at just under 55 percent. It took more than two days, and a lot of developments in Franken’s favor, for them to catch up. Now they give him close to 80 percent.

Franken will hold a solid lead, likely in the low triple digits, by the time the canvassing board concludes its review of the challenged ballots. About 300 ballots remain to be reviewed, and the board hopes to finish the process tonight.

Also, last night the Minnesota Supreme Court ruled that absentee ballots that were improperly rejected must be counted in the final tally, reports The Minnesota Independent. Over a strong dissent, a three-justice majority decided that these ballots should be included, but that both campaigns must agree that a ballot was rejected without cause for it to be counted. This represents a big victory for the Franken campaign, which is expected to gain about 100 net votes from these ballots.

It now appears that unless Coleman can win a major lawsuit to disqualify a high number of Franken ballots, Al Franken will be the next U.S. senator from Minnesota.

UPDATE 12:12 PM: Franken’s lead has now surpassed 150 votes, with around 100 challenged ballots left to review. My prediction of “low triple digits” is looking pretty good.

—–

It appears that Al Franken is on his way to taking a slim lead over Sen. Norm Coleman by the end of the day or early tomorrow, as the third day of the Minnesota State Canvassing Board’s review of challenged ballots draws to a close.

Coleman’s official lead now stands at 38 votes, and it continues to shrink as more challenges are reviewed.

The board finished reviewing Franken’s challenges this morning, and Coleman’s official lead grew as most of them were rejected. Now it has begun going through Coleman’s challenges, and his lead has diminished at a faster rate than his team had hoped. About 400 Coleman challenges remain.

If Franken continues to net votes at the current rate, my very unofficial prediction is that he will end up with a lead in the high double digits.

And I’ve got some very unofficial confirmation from some very unofficial Star Tribune projections. Given the success rate of the two campaigns’ challenges, the Strib projects that Franken will come out of this process ahead by 90 votes. Readers of the Strib also judged the challenged ballots for themselves, and their average projection put Franken up by 40 votes.

Here’s a chart from the Strib showing where things stand as of 5:25 PM. As you can see, Franken’s challenges have had a much higher success rate (not a huge surprise, since he withdrew many more of his frivolous challenges):

Now, if Franken falls short of the lead, he still has a very good shot at winning this election if the improperly rejected absentee ballots are counted. (The Minnesota Supreme Court is currently deciding on their fate.) These ballots are expected to benefit Franken, perhaps by about 100 votes.

Of course, no matter what the result of the canvass and the Court’s deliberation, we can expect lawsuits galore — no one’s going down without a fight.

As the canvassing board plows through its second day of challenged ballot reviews (livestream here, resuming shortly after 3pm ET), Coleman’s official lead has grown to 328 votes. But that’s because only Franken’s challenges have been reviewed so far.

I’ll explain this as succinctly as I can. When Franken challenges a ballot, it means one of two things: the vote was initially given to Coleman and Franken thinks it should be discarded; or the vote was initially discarded and Franken thinks it should be his. (I have yet to see a case where a Coleman vote could be reasonably reassigned to Franken.)

Ballots are immediately removed from the official count when they are challenged, and they’re re-included when the challenge is resolved. Since the canvassing board usually agrees with the recount judges’ initial rulings, most of the challenges fail. So a Franken challenge will usually result in a vote for Coleman or a non-vote.

If Franken successfully challenges a ballot and an initial Coleman vote becomes a non-vote, the official count does not change (the ballot had already been yanked when it was first challenged). But this is still a net gain for Franken because his opponent has lost a vote.  The only way that Franken can gain in the official count is if he wins a challenge that results in a vote being assigned to him.

So where do we stand? So far, 339 Franken challenges have been ruled on with about 100 left to go.

Here’s how they’ve been allocated:

Coleman: 188
Franken: 48
Allocated for neither/other: 103

That means the net change in the official count is 140 votes for Coleman, bringing his lead from 188 at the end of the hand recount to 328 now. But actually it means a gain of somewhere between 48 and 151 votes for Franken, depending on how many of the “allocated for neither/other” votes initially went for Coleman, which we do not know.

The canvassing board expects to finish ruling on Franken’s challenges today. Then we’ll see the Coleman challenges, which will likely number around 1,000, although the canvassing board has urged Coleman to withdraw a portion of these. The effect of these challenges will be the inverse of Franken’s: the official count will register gains for Franken, while the actual count will move in Coleman’s favor.

It seems likely that Franken will come up a bit shy of overtaking Coleman at the end of the challenged ballot resolution. But then we move into the improperly rejected absentee ballots, on which the state supreme court is ruling right now (livestream available at The Minnesota Independent). If the court upholds the canvassing board’s decision to count these votes, it seems likely that Franken will take the lead.

I’ll keep you posted, if you can bear to read any more on this….

UPDATE: Yesterday, Secretary of State Mark Ritchie was careful to refer to Coleman as “Senator Coleman” and Franken as “candidate Franken,” correcting himself the (many) times he used the wrong appellation. Today he’s gotten tired of that game; it’s a simple “Coleman” and “Franken.”

The AP’s examination of the remaining challenges found:

- Fewer than half of the challenges left-about 1,640-are in genuine doubt. Still, that’s eight times more than the current margin between the two men.

- In ballots that could easily be assigned, Franken netted 200 more votes than Coleman. But that number was essentially meaningless because Coleman has withdrawn significantly fewer challenges than Franken-that is, the pool of challenges that can be awarded to Franken at this stage is notably larger.

- Nearly 300 challenges wouldn’t benefit either man because the voter clearly favored a third-party candidate or skipped the race.

- Of the challenges that can’t be reliably awarded to either candidate now, more than 400 possible Franken votes are being held up because on grounds that those voters identified their ballots through write-ins, initials, signatures, phone numbers or some other distinctive marking. At least 300 possible Coleman votes are in limbo for the same reasons.

- The next biggest class of ballot that can’t easily be awarded falls in the category of unclear voter intent. Nearly 600 involve cases where a voter filled in two ovals but crossed out one, put an X above or below their darkened oval or put differently sized partial marks in more than one. There are slightly more potential Franken ballots in that pile as well.

Now, as the AP explained, these numbers are subject to significant change and mean little on their own. This analysis included over 5,000 challenged ballots, but the Franken and Norm Coleman campaigns have withdrawn thousands of frivolous challenges, and the number of challenged ballots now stands at 3,497, down from its peak of 6,655. Over the weekend, the Franken campaign announced that it would reduce its challenges to fewer than 500 by Tuesday, and the Coleman camp followed suit and pledged to bring its number under 1,000. The state will begin reviewing these ballots on Tuesday.

So what’s the significance here? On December 3, the Franken campaign claimed the lead (by 22 votes, later revised to four) in the recount, while the official tally favored Coleman by a margin of 295 votes. The difference was in the methodology: while the official number excluded all challenged ballots, the Franken team’s number included the predicted outcome of the challenges, based on the election officials’ initial judgment.

As far as I’ve seen, the AP report is the first extensive third-party study of the challenged ballots, and even if the numbers change, it lends credence to the Franken camp’s expectation of significant gain when the challenges are resolved.

The Minnesota Secretary of State website now shows Coleman with a 188-vote lead. If the AP’s 200-vote-swing prediction is at all accurate, the challenged ballots could easily be the difference-maker in the race. And then of course there are all those absentee ballots that were improperly rejected but now seem likely to be counted. The Franken campaign is confident that it stands to gain from these ballots; that’s why it pushed so hard to have them included. In other words, everything that happens between now and the end of this process is likely to chip away at — and quite possibly reverse — Coleman’s small lead.

Let me put this whole jumble in more concrete terms: I just bet on Franken with the CEO of TWI’s publisher and gave him 2-1 odds. Donuts are on the line.