Well, we definitely can’t blame this on an August snowstorm. But we had better come up with some excuse because voter turnout in last week’s primary election was at its lowest level since 2013, when a snowstorm hit the city on primary election day.

No, city and school board primary elections weren’t held in August back then. They were held in late February. It has only been in recent years that a state law change moved city and school board races to be held in August for primary elections and November for general elections. In fact, this past week’s election was the first August primary for city and school races because we didn’t have enough candidates in 2019 to have a primary.

So, the sample size is small, but thus far the result is that an August primary isn’t creating a lot of excitement. Preliminary numbers show that Tuesday’s primary produced a voter turnout of 10.1%, according to numbers from the Douglas County Clerk’s office. To be clear, that’s not just voters who turned out on Election Day, but a combination of voters who turned out on Election Day plus everyone who voted in advance.

That 10.1% is the lowest turnout since an 8.6% turnout in February of 2013. I went back to read the Journal-World article from that day, and it noted turnout likely was impacted by a snowstorm that hit the city as polls opened.

The only snowmen I saw this past week were on my golf scorecard. So, what’s the excuse for the low turnout in this election? I suppose you could argue that it is pandemic-related. But such arguments seemingly would be muted by the fact that it is very easy to vote by mail in these elections. But the bigger reason I don’t think the low number had much to do with the pandemic is because voter turnout numbers for the 2017 primary — long before the pandemic — stunk too. They were just 10.3%.

In general, voter turnout numbers for city and school primary elections have long stunk. But now we are talking about a different degree of ripeness. Here’s a look at turnout numbers dating back to 2003:

• 2021: 10.1%

• 2019: No primary

• 2017: 10.3%

• 2015: 13.8%

• 2013: 8.6%

• 2011: No primary

• 2009: No primary

• 2007: 15%

• 2005: 16.3%

• 2003: 20% Estimated.

The last number is estimated off of figures from an old Journal-World article. For whatever reason, the summary sheet listed on the Douglas County Clerk’s webpage doesn’t list a voter turnout number for that election. I’m sure they have a precise number somewhere, but the clerk’s staff is in the middle of certifying this current election, so I didn’t want to bother them with finding that old number.

But as you will see in a moment, there are all types of signs the 2003 election was a busy one by primary standards. It is the one that involved a coalition of candidates who united under the banner Progressive Lawrence and swept the top three spots in the election. (Perhaps you can win a bar room trivia contest in Lawrence by naming those three candidates: Mike Rundle, Boog Highberger and David Schauner.)

By most standards, even 20% is a rotten number when it comes to voter turnout, but it would be quite a success story in the last couple of primary races in Douglas County.

The percentages, though, only tell part of the story. The number of ballots cast in each election is also interesting. Here’s a look:

• 2021 primary: 6,523 voters (subject to change as provisional ballots are counted.)

• 2017 primary: 6,081 voters

• 2015 primary: 8,818 voters

• 2013 primary: no data. The county’s summary sheet didn’t list a total.

• 2007 primary: 9,382 voters

• 2005 primary: 8,934 voters

• 2003 primary: 11,019 voters.

It is noteworthy that this week’s primary produced about 400 more voters than the 2017 primary, but the number is still depressing. Think about these numbers: The number of voters in 2021 was about 40% less than the number of voters in 2003. That is despite the fact that the city’s population is much larger in 2021 than it was in 2003.

Based off of population estimates produced by the city, Lawrence’s population in 2021 is somewhere north of 103,000. In 2003, the city’s estimated population was just under 85,000 people. Another way to look at that is that in 2003, about 13% of Lawrence’s entire population participated in the primary election. Last week, about 6% of Lawrence’s population participated in a primary election.

Yes, I know not everybody in Lawrence’s population is eligible to vote. Don’t turn me in for voter fraud. Rather, I’m just using the numbers as a metric to show how a smaller and smaller percentage of the community is determining who is in power.

But the numbers are imperfect because sometimes a primary only has Lawrence City Commission candidates on the ballot and other times it has school board candidates or candidates in other races from places like Eudora or Baldwin City. In other words, the total votes cast don’t all come from within Lawrence.

So, another set of numbers that might be useful is to look at is how many votes the top vote-winner in the Lawrence City Commission primary received each year

• 2021: Lisa Larsen, 3,518 (subject to change)

• 2017: Lisa Larsen, 3,743

• 2015: Leslie Soden, 3,618

• 2013: Mike Amyx, 2,989

• 2007: Mike Dever, 4,977

• 2005: Mike Amyx, 4,234

• 2003: Mike Rundle, 6,129

Another way to look at those numbers — Larsen’s vote total in this year’s election would have been good enough for fifth place in the 2003 election. That’s not an indictment on Larsen. It is an indictment on the many voters who stay home.

It is an indictment leaders may want to spend a little more time thinking about. The idea of creating districts for Lawrence city commissioners — currently all five seats are at large — is still floating around. Does anyone really think that is going to produce better turnout for a primary election?

It is going to create confusion because a good number of voters are going to be uncertain whether they even have a primary to vote in. You’ll have years where you’ll have some parts of town voting in a primary election and other parts of town not voting in one. Obviously it is doable. It happens with legislative races all the time. But it doesn’t seem realistic to think the complications of it are going to lead to better turnouts in primaries.

And, if turnouts decline as a result, there won’t be very many voters deciding who is in power. Is it possible that a winner of such a primary might only need to get a couple of hundred votes? Is that too few for a town Lawrence’s size? Would it produce better government?

It is something to think about while we root for an August snowstorm.