People who are closely watching Whatcom County election returns noticed a distinct shift on Wednesday toward the conservative.
Big leads by progressive County Council candidates eroded by 2 or 3 percentage points. (The smallest lead is still by 5 1/2 points and appears safe.) More significantly, a close Port of Bellingham race swung away from Renata Kowalczyk (50.2 percent of the vote on Tuesday), supported by liberals, to Dan Robbins (51.1 percent of the vote on Wednesday), who is the choice of conservatives. The other port race followed the same trend: Mike McCauley's 1,892-vote lead from Tuesday shrunk to 898 votes on Wednesday.
Speculation abounded on social media as to what might have caused Wednesday's trend. Was a bulk of the ballots counted on Wednesday from outside Bellingham? (In this race and in general, Bellingham precincts voted liberal for the most part; county and small-city precincts voted conservative with a few exceptions.)
If that's the case, it wasn't the result of anything done consciously inside the county Auditor's Office, Auditor Debbie Adelstein said in an interview today.
She did say that what was counted on Wednesday were ballot-box ballots and only a few ballots received in the mail. But there are several steps in the Auditor's Office between the ballot box and the counting machine, and the ballots from yesterday's count should have been a thorough mix of all the boxes -- Bellingham, Western Washington University, Lynden, Everson, everyone.
You can do a quick "back-of-the-envelope" calculation, as the science professors like to say, to get a sense of whether a disproportionate number of ballots did in fact come from outside Bellingham, even if only by chance.
From the website with county results, I took the City Council race with the largest vote count from Tuesday -- Pinky Vargas vs. Clayton Petree (14,994 votes total) and used that as a proxy for the number of Bellingham voters counted on Tuesday, knowing that not everyone in Bellingham voted in that race. Divide that number by the total ballots counted on Tuesday, 41,318, and you get 36.3 percent. That is, as a quick-and-dirty estimate (underestimate) of how many ballots counted Tuesday were from Bellingham, you get 36.3 percent.
Doing the same calculation for Wednesday: The largest vote count, Vargas vs. Petree, was 16,944. Again this isn't the total number of Bellingham votes, but it's a valid enough proxy because we're using the same count for both calculations. The total count on Wednesday was 50,380. Divide, and you get 33.6 percent.
We can conclude with some confidence that a disproportionate number of yesterday's ballots were from out of Bellingham. This might be patently obvious just by looking at the results -- the trend to conservative. Still, there are two possible explanations: the ballots, despite mixing randomly in the Auditor's Office, were counted in such a way that a disproportionate number put through the machine happened to be from outside Bellingham. Or -- and this seems more likely -- county and small-city voters turned their ballots in late more so than city voters.
If the latter is true, that could mean the poll results will continue to trend conservative throughout the week.
Election observer Jim Fox says historically, about 60 percent of all the ballots are counted on election night. By Friday afternoon, Nov. 8, about 90 percent should be counted.