On paper, a 213-vote lead like Rochelle Swanson holds over John Munn does not seem that formidable. However, given the numbers we know from the election, the chances actually seem very remote that John Munn can overcome the lead.
We can calculate, based on reports from the County Elections office, that there are maybe 2000 to 3000 ballots that have still not been counted. We have no reason to suspect that there is a skewed percentage of the electorate still to be counted.
So, we really have two population samples to estimate from. The first is the group of voters who cast their absentee ballots prior to election day. The other is the group of voters who showed up on election day and cast their ballots at the polls.
The first group gave John Munn a 73-vote advantage. The second group overcame that advantage and voted for Rochelle Swanson by 286 votes.
It is our belief that the voters who turn in their ballot late are more like the voters on election day than the other group. But for purposes of this analysis, we will err on the side of John Munn to explore what happens if the remaining 2000 or 3000 votes are aligned with those who voted prior to election day.
If Mr. Munn received votes at the same rate he received prior to election day and there were 2000 uncounted ballots, John Munn would expect to receive 748 of those ballot while Rochelle Swanson would receive 724. That does not do much to cut into the lead.
To win, John Munn would have to triple his vote advantage over Rochelle Swanson in one-third of the votes, a tall task.
Instead, John Munn would have to increase his vote share to 843. That would mean he would have received votes on 42% of the ballots rather than the 37.4% he received prior to election day and 34.5% overall. To put that into perspective, Robb Davis received votes on 43.3% of the ballots, so John Munn would have to nearly match him in this sample to catch Rochelle Swanson.
Running the same numbers for 3000 unaccounted ballots, and using his pre-election rate, Mr. Munn would expect to receive 1122 additional votes to Rochelle Swanson’s expected 1086, a 36-vote difference.
Once again, John Munn would have to increase his vote share to 1211 in order to defeat Rochelle Swanson; while that number is a more palatable, he would still need 40.3% of all ballots to accomplish it.
Not only does John Munn have to increase his vote share from to 40.3% in the 3000 ballot scenario or 43.3% in the 2000 ballot scenario, Rochelle Swanson’s vote share would have to decrease from her 36.3 percentage to about 31.5% in the 2000 vote scenario and 33.3% in the 3000 ballot scenario.
The bottom line in this analysis is if there is a normal distribution of ballots that remain to be counted, John Munn would have to perform far better than he did on the pre-election absentee ballots to close the gap, and Rochelle Swanson, who was fairly consistent in terms of her vote share, would have to suddenly drop.
While these aren’t impossible numbers, the trend seemed to be going Rochelle Swanson’s way on election night, and it would not only have to reverse itself, but it would have to change drastically even from what happened before election day.
Possible? Yes. But does not seem very likely.
—David M. Greenwald reporting