What Can the Recount Tally Changes Tell Us?

There's one critical piece of information that I'm paying careful attention to in the recount for the Wisconsin Supreme Court election, the updated vote tallies. The changes to the election results from the recount are not only important because they tell us who presumably is the winner, but they *should* also confirm the winner.

There appears to be at least 8 issues with the chain of custody of ballots in Waukesha County such as torn ballot bags and missing seals/labels. So there's definitely ways in which those ballots could have been modified. Of course, the media reported the issues with the ballot bags in the first few days of the recount, but there's been no mention of the more recent issues or how any of the issues were resolved. The GAB also hasn't explained the resolution of these ballot bag issues or even acknowledged many others. This is very concerning, because the whole idea behind the recount was to get to the bottom of issues, not to continue hiding them.

So, how can we tell if there are unresolved issues with the election results? Theoretically, and with a large enough amount of votes, the distribution of changes in results should match the final distribution of results. That is, if Prosser has a majority of the overall votes, he should also receive the majority of newly-counted votes. So if, say, Prosser retains the lead in the overall results when the recount is complete, but Kloppenburg received a majority of the newly-counted votes, then there is reasonable concern to doubt the vote tallies from one or more precincts.

This is particularly true if there is an unexplained wide gap between the two distributions and is a result of the mathematics of probability. If Prosser has a majority of the overall results when the recount is complete, but didn't receive a majority of the newly-counted votes, we could have a problem. It wouldn't be proof that an unresolved issue exists, but it would be a bright red flag.

So where do things stand now? Well, Kloppenburg is leading 437 to 140 in the newly-counted votes. That's 50% of the newly-counted votes going to Kloppenburg which is also 75% of the newly-counted votes cast between Kloppenburg and Prosser. If the recount were complete, these numbers would be cause for concern. However, I must stress that the latest numbers from the GAB are unofficial, and we can't make any judgements until almost all counties complete the recount (if we can at all before the full recount is complete). Also, this technique won't always uncover issues and it may imply issues when none exist, so we must carefully make judgements when using it.

Still, since we're getting very little information from the GAB and the news media, I thought I'd share this bit of helpful information now rather than later. If the distribution of results continue the way they are currently, we're going to want a strong argument for further investigation.