Informed folks tell me that today’s count was disproportionally city 43% of the votes counted rather than 33%ish. They say its more like 3350 city votes left to count…......... Hell I’ll tale 50% + 1. We’ll see!

Doug - and if those who study this see fewer mayoral votes left to count than I do, then they are probably right.  They have the software tools and the money to buy the data from the Auditor.  Readers should know that these results are normally closely held by political organizations and the last thing they want is for any media to print or post them.  We are left with my poor calculations.  But it will be fun to see how many more mayoral votes will be counted. (error corrected)  4,300 or somewhere towards 3,350.  Big difference.

This years election is unusual in the fragility of incumbents’ seats:  Pike, Crawford, Buchanan and Larson would normally be considered to have an edge by the mere fact of their incumbency.  Those with better memories than mine might be able to remember an election with more incumbents turned out or imperiled, but I think this year’s election is remarkable.

Floyd McKay in Thursday’s Crosscut and today’s Herald have articles on the role of the coal train controversy on the elections.  Supposedly, the proximity to the coastal rail line equates to voter support for opponents of the Gateway terminal.  If this is the case, a similar geographic pattern should show in the county races with increased support for Maginnis, Kremen and Louws increasing near the rail line.  It would be interesting to see if this is so.

The numbers coming in since the first count on election night don’t show much of a pattern like the shift John keeps mentioning.  In my earlier post, I described the older voter profile that applied to absentee ballots back when we voted at polling places.  I gave some reasons why the early returns now might show a slight bias towards voters with more rigid views and why the later returns would look much more like the final results.  I haven’t seen an explanation for a systematic shift in late returns towards more conservative candidates.

My guess is the early votes look different than the final results, rather than the later results are much different from the overall results.  If this is so, then the percentage change will be less with each later count, barring the randomization that occurs inside the postal system.  In any random system, there will be statistical lumps that look like signal but really are just noise.  Maginnis’ surge in the Saturday count looks like one of these noise bursts because it affected her race far more than the other county races.

One has to expect a considerable amount of correlation between candidates, so if there is a big jump in one race and not in others, it’s more likely to be randomness than a systematic effect.

I agree with the above comments by Paul, this election was very unusual. To answer his question about defeated incumbents. Four incumbents were also defeated in the 1993 elections. Dennis VanderYacht and Marge Laidlaw on the County Council. Foster Rose (he defeated Paul for the Ward 2 position in the previous election) and Mark Asmundson on the City Council. Mark was defeated in the primary. On January 2, 1985, Mark was appointed to replace David Wolf. He ran unopposed in the 1985 and 1989 elections. He was defeated the first time he had opposition but came back two years later to be elected mayor of Bellingham.

While the 1993 general election ballot had a controversial referendum of the Critical Areas Ordinance the turnout was lower than this year. The general election turnout was 56.5 percent and 79.3 percent were poll voters.