Published Voting Results and Their Accuracy

There was a lot of talk yesterday about how CNN and MSNBC were getting results faster than the Secretary of State website. What hasn’t been covered is that some of these numbers have been continually corrected and revised. With 100% precincts reporting on all sources, the numbers still conflict.

At 14:17, Nov 5:

CNN: Coleman 1,211,584 - Franken 1,210,894 (690 vote difference)
MSNBC: Coleman 1,211,629 - Franken 1,210,921 (708 vote difference)
SOS: Coleman 1,211,629 - Franken 1,211,167 (462 vote difference)

In addition to these differences, these numbers are all different from themselves where they were at when 100% of the precincts were originally reported early this morning. In fact, the SOS number had recently changed this afternoon.

For another example, CNN at 11:10 had these numbers: Coleman with 1,211,403 and Franken at 1,211,089, showing only a 314 vote difference. Again, that was with 100% reporting.

I wonder what the numbers will say this evening?

10 Responses to “Published Voting Results and Their Accuracy”


  • Does it matter? The recount will happen and there will eventually be a winner named. There will likely be court challenges either along the way or after the fact. None of it will change the fact that the victory will be within the margin or error.

  • IN EVERY EXAMPLE FRANKEN LOSES. WHATS ALL THE FUSS ABOUT? FRANKEN SHOULD MOVE BACK TO WHEREVER HE CAME FROM AND CONTINUE TO WRITE HIS PORN.

  • CNN just updated there numbers to reflect the 462 vote difference.

    I distinctly recall the 2002 Senate race where the vote was stopped around 3am and the vote would start again in the early morning. Mondale was far behind but he made a huge gain near the end closing in tens of thousands of votes. I wonder if there aren’t still absentee ballots to be counted, as I believe that was what also contributed to Mondale’s surge.

  • KH, sometimes your comments are just plain obnoxious. Just because something “doesn’t matter” in the sense that it won’t make an impact on the final result, doesn’t mean that it isn’t an interesting observation. For all of us that have been closely watching this race, DFLers and Republicans alike, it definitely is an interesting observation. The difference between 708 votes and 462 is an appreciable one in a race where every vote will matter immensely… But, yes, thank you for stating the obvious.

  • Aaron’s post, in the context of this site, seems more likely meant to give the impression that Franken could very well be leading by this evening and therefore the legitimate “winner” in place of Coleman. The fact is neither candidate will ever be able to claim with significantly more than a 50% certainty that he received more votes than the other. So really, neither 708 or 462 is an interesting or meaningful number in any way. Unless you don’t understand the math and variables involved.

    Quit spinning, the campaign is over. The only people worried about 708 or 462 are those looking to give false legitimacy to their candidates possible “win”.

  • The SOS numbers have been changing all day and will continue to change as they correct incorrectly entered data.
    Franken is currently behind 475 votes.
    CNN and MPR are getting their numbers from the SOS, but they are delayed.
    You have to reload the CNN page to see the updates.
    The SOS web site says all precincts have reported but some are showing 0 votes which is hard to believe.
    Also, some numbers have been put in the wrong columns.

    There are currently 4500 ballots with no Presidential vote and 30000 ballots without a Senate vote.
    These would be the ballots that would affect the certified results if people had marked them in a way that was not caught by the optical scanner.
    Franken should gain among these.

  • Since DFL leads among the no college cohort (never took an SAT) and among first time voters, Franken should pick up votes on incorrectly marked ballots.
    Assuming the certified results have Franken down by 500 with 2500 marked ballots that haven’t been counted, Franken would have to win by 20%, I’d call that unlikely but possible.
    A more likely path to victory is a cleanup of the data entry errors prior to certification.

  • tkg, thanks for all your comments at the various posts today. I don’t know how accurate it all is but you seem pretty sure. Thanks for keeping the dream alive!

  • Franken leads in the “can’t color inside the lines” constituency!!!! j/k

  • Does anyone have any information on overseas and absentee ballots (i.e. whether they were likely to have been 100% included in the current numbers)?

Leave a Reply