A Pioneer Press analysis of rejected absentee votes already counted in the U.S. Senate recount shows that you can't predict an outcome based on whether ballots come from a Republican or DFL hot spot.

That's interesting because part of the Coleman basis for continuing the effort to get 4,400 more rejected absentees added to the total is based on a belief that many (most?) of them come from Coleman-leaning precincts. In fact, Coleman legal spokesman Ben Ginsberg said after the recount: "If we were counting all the votes, we would feel very confident about picking up votes. If you look at counties and the precincts that these 400 came from, they were more from Franken precincts."

Coleman, himself, said recently: "I'm confident because the votes that are out there are votes that will turn a tide here. ... There's no question about that. They come from mostly Republican areas because the votes from the Democrat areas have already been counted."

But the Pioneer analysis shows that ain't necessarily so.

"Of the 1,284 absentee ballots that have been counted since Election Day, Coleman underperformed compared to the political tilt of the cities and counties from which the ballots came," the paper said in its Political Animal blog.

"During the recount, 933 absentee ballots were counted. On Tuesday, 351 absentee ballots were counted.

"Counties that Coleman won in November contributed about 46 percent of the ballots counted during the recount. Counties and cities that Coleman won in November contributed 61 percent of the ballots counted this week.

"But when the ballots were opened during the recount Coleman only won 33 percent, 13 percentage points less than the geography of the ballots would lead one to believe. When the ballots were opened Tuesday, he again won about 32 percent of the vote, underperforming by 29 percent."

Bottom line, the uncounted ballots -- which have been rejected for a reason -- may or may not make a difference.