Posted by: sweeneyblog | November 10, 2011

Voting Trends with Mail-in Ballots

Hello Loyal Readers,

First of all, let me make everyone here feel a little old. With one exception, I have voted by mail-in ballot my entire voting life. I was able to vote at a polling station once for the 2004 primary right as I turned 18. Since then, every place I’ve lived has had universal mail-in balloting. I am a big fan. I feel like it allows more people the flexibility to find time to vote, and to make informed decisions about who they are voting for. It also provides us a whole addition set of data as to WHEN people make their electoral decisions.

Behold the chart.

Number of Ballots turned in

Feel free to click on it to make it larger. This chart I made shows how many ballots arrived at the Whatcom County auditor’s office each day starting on the Monday after ballots arrived.  Each line represents a different demographic of voters (can’t share what they represent, sorry) but this is everyone who voted, I promise.

The thing that jumps out at me is the pretty flat procession until that last weekend. I always thought that the first weekend would offer a big bump of ballots. Voters who are energized about their candidates would quickly fill out and turn in their ballots, but that didn’t seem to happen this year. There was a surge that struck in the second weekend which makes me envision lots of parents hunkered down at the kitchen table with their voters guides while their kids whirled around them in a Halloween-candy sugar tornado. But the main conclusion is while we have the option to vote on any day of that three week window, most people choose to wait to the last minute. Just like if we had a polling location.



  1. This strikes me as an outcome somewhat peculiar to this election, and I’d need to see data from past elections to better understand how it fits into election history.

    Without intending to minimalize the very important issues in this election, stakes were low. Either candidate could plausibly be Mayor of Bellingham, and not much of an ideology swing to energize voters. Same can be said for County Executive. No “necessary enemy” existed to alarm voters and get them riled early. If you were not “of the tribe” of the candidate, little enthusiasm to vote; if you were “of the tribe” of the candidate, the challenger could plausibly serve in place.

    The county council races bucked this, and you see some the results of late campaign energy manifesting themselves in the returns.

    What you see in typical years is a brush fire of early ballots from people whose minds were made up early, fixed by ideology or partisan swing. This election did not have that feature, and so voters apparently held out to the last moment, perhaps collecting data or mulling opinion until the final hours. There was ultimate urgency to vote, yes, hence turnout levels well above the state average.

    One thing missing from elections FOR YEARS is the absence of “trending” Servais was predicting. Poll voters liked the social aspects of voting and were rooted in their communities, which yielded a conservative trend to late ballots. All bets are off on that with vote-by-mail, have been for several elections now—it appears conservatives are no more or less likely to hold back their ballots than anyone else. A new trend may build, but right now there seems to be only churn, not trend.

    My guess is “trending” may build among younger voters and, for example, around WWU campus, as these new voters feel their way around and develop preferences for how and when they vote.

    • I agree about trending, and will probably do a follow-up post to help put this in context.

    • My, lets have a cat fight. Amazing Tim Johnson acknowledged I exist as he has tried his best for years to pretend I and the NWCitizen website do not exist. When he takes issue with me, he usually refers vaguely to blogs or some such. So, now he specifically goes out of his way to say no county late vote right shift has existed and names me as holding this fictional belief.

      “One thing missing from elections FOR YEARS is the absence of “trending” Servais was predicting.”

      I enjoy different perspectives any of us bring to trying to make sense of election results. Some see trends or causes that others scoff at. For example, Tim in his Gristle this week covers why Dan Pike lost the election — and Tim totally ignores the red-light camera issue. He mistakenly referred to them as “surveillance” – which of course, state law does not allow them to be used for. But he only mentions this once and as a cause of a higher voter turnout, not for votes against Dan Pike.

      I wrote the red-light camera issue was the deciding factor in the mayor’s race. Regardless, I think even Dan would agree that it was a big issue. The vote against red-light cameras is 65% – overwhelming. And the target of the anger by the anti red-light camera folks has been Dan Pike. For Tim, it seems that things he does not like then simply do not exist. Nice to know I finally do for Tim.

      • Play nice.

    • John’s ugly nonsense is the reason I avoid him at all costs.

      • Play nice boys.

      • Tim, you’ve pretty much been ignoring the red light cam issue as if it did not exist – except to tell your readers how obnoxious Eyman was when he showed up in Bellingham (not that I disagree, but it happened to be one of two times that he visited Bham in the span of a year). I’m happy, though, that you had the good judgement in the end to tell your readers to vote yes on the initiative (in what I thought was a very schizophrenic voting guide, I should add).

      • Haven’t been ignoring the red light issue. Been following it. Went to the court hearing, went to the NWBC meeting on the topic.

        But, with all due respect, it was rendered a bit of a non-issue when the appeals court ruled the initiative was invalid and—at the same time—you had City Council members admitting, whatever the outcome, they probably would not renew the pilot project when the year contract elapsed. Whatever the vote, it was not going to end the contract; whatever the vote, it was not going to renew the contract.

        My surmise, based on opinion left and right alike, was the initiative would pass by a huge margin. Did it have coattails on other issues, like the mayor’s race? Sure, probably, but I didn’t see a way on Election Tuesday to tease out that information in a meaningful that was not purely anecdotal conjecture.

        I don’t see how you can reasonably claim, for example, a vote against red light cameras = a vote of against Pike when the one issue was decided hugely and the other is still in the noise. The single most powerful telltale in the mayoral race was the primary, against which the general appears to track closely.

  2. The Whatcom Watch has tracked for years, WHEN the mailed/dropped ballots arrived at the Elections Office:

    Here is the last Primary – this year:

    And previous years are also there…

    • That’s wonderful, thank you!

  3. I’ve had the chance to vote both ways and by far prefer going to a polling place, standing in line and poking the card with that special voting tool. Like Tim mentioned, there was a fun social aspect about it as well. Plus I still remember my father, and others, going to the Courthouse to hear the results updated.

    Now you just draw a line stuff it into an envelope and drop it off, hopefully combining that trip with another errand you had to do 🙂

    Election night is not very exciting after 8:00. There’s a single vote count, and you don’t have to listen to the radio or be at the courthouse…. yawn.

    I’d really like to see us go back to the old system, it just feels more authentic to me.

  4. I had a friend call me frantically on the day of to get some vote advice. I learned my lesson the hard way. After losing a ballot, I fill it out and mail it IMMEDIATELY!

  5. Voting is a sacred right – one that our forefathers (and foremothers) fought and died for. In many parts of the world people still risk their lives to cast a ballot. It should be a little inconvenient to vote. Some effort beyond licking the flap on the envelope should be expected even in our techno-heavy, 21st century culture to elect the people who make the laws we live under, levy (and spend) our hard earned tax dollars and decide whether or not a cop has to pull us over for running a read light or we just get the ticket in the mail. (This is not an anti-postal service rant – I’ll post that one for later.)

    I agree with Clayton, I wish we could go back to the old system. Besides, the little “I Voted” stickers made me feel self-righteous.

    • If you put your ballot in the auditor’s office drop box you can pick up a sticker.

      • Ah! But there is the rub. I work in Olympia during the week – only home on the weekends (been doing this since 2001 – before the all-mail system and never voted “absentee”). While I personally feel casting a ballot is worth taking a days annual leave, I don’t value the sticker quite that much. But thanks for the info.

  6. This chart is way cool.

  7. Yes, Riley, I did forget to mention this is an impressive bit of work. Wondering what software you used to build it?

    • I used a Google Motion chart. You make a google doc and then you can manipulate the data as moving bubbles, or lines. If you want, I can show you the set-up I have for precincts over the last five years.

  8. Okay, you guys with the nostalgic view on poll voting need a reality check. Is democracy really served when you have to wait 3 to 8 hours to vote as some areas of the country have experienced? And even in Washington, I had to wait in line (often in the rain) in the old days to vote. I became a fan of mail voting when I missed a school levy vote cause I flat out had no time during that particular Tuesday to go to the polls. Voting still requires some work and thought but as Riley (smart fellow that he is) alluded too, you know have an opportunity to review your ballot with some time to think through your decision. I can’t imagine making decisions on judges and some of those initiatives this year without having the time to reread the voters pamphlet, go on the internet and ask questions of friends. You can argue that we should all do our homework ahead of time but its far easier to have the ballot in hand when making those decisions–particularly since voting pamphlets often contain far more issues than are on any given ballot and its only when I get ballot that I’m actually sure which fire and mosquito district I should be paying attention to. Those of you who still pine for the social experience of going to the polls should find somebody under 30 and ask them to arrange a flash mob for you and others to gather and vote together.

    So Riley, here are my questions. Do late voters vote late because they can’t make up their mind or because they are procrastinators? Or put another way, how many who voted in the last week were ticket splitters? The other question I have is how does this fit with voting patterns in previous years and elsewhere and will this have implications on GOTV. In my neck of I-5, the political mail (except for the liquor initiative) stopped about two weeks before the election. Is that a smart political decision? Finally, how come you can’t tell us the cohorts represented by the lines? There are some differences between them.

    • Here’s my answers in order:
      1) Not really sure, probably some are procrastinators, some are indecisive, there is not an easy way to tell.
      2) With patterns from previous years . . . I did an analysis in 2010 and it showed a bulge at the first weekend and a bulge at the end. Perhaps because national races are better covered so people have their opinions set already? Or the partisan nature makes it easier for people to turn in their ballots.
      3) The political mail probably stopped because you turned in your ballot. Smart campaigns pay for “Matchbacks” which is a daily report from the auditor that tells them who has voted. They then clean these people out of their mailing list. So if you turn in your ballot early, they cross you off their list.
      4) The cohorts are different demographics that people have painstakingly gathered over the years and I don’t want to give away their data for free. The differences are not drastic.

    • Good answers, Riley. VBM has many advantages, but election drama is not one of them.

      In the first few years, when Whatcom was a leader in VBM, you saw big waves of voters turning in their ballots in the first week they received them. It certainly changed the way local media began covering the campaigns, as we all shifted the bulk of our coverage to land in and around the ballot mailing, rather than timing around the actual polls. Same thing with mailers and flyers. In some ways, the election was considered essentially over the day Shirley hauled her mailsack to the Post Office, weeks before ballots were counted.

      Now it appears all that is relaxing as the entire state went VBM. I suspect some of that has to do with changes in the way voters collect their information about candidates and issues, as old media fades and new media takes its place (but in weirdly partisan and—if I can use the term—epistemically closed ways). It’s hard to know exactly when voters are pumped and ready to receive the information about candidates and issues they are clearly ultimately interested to receive (but it does seem to be true that people who are not ready are also not interested).

      The Weekly held back its voter guide until the week after ballots were mailed in the belief (perhaps erroneous) that people who toss their ballots in the mail as soon as they receive them are already fixed and decided on how they intend to vote. Taking to the candidates (and we did ask), they also thought this was a logical way to proceed.

      • I thought the Cascadia Weekly’s voter’s guide was well-timed. Tip of the hat, good sir.

  9. I used to go to the polling booth with my Dad when I was too young to vote so count me in for that nostalgia nonsense 🙂 That said, before vote by mail was manditory, I always opted to vote that way precisely because n adult I disliked waiting in lines.

    I’m one of those that vote the day I get my ballot. It saves paper (less fliers get mailed to me) and cuts down on robo calls.

    • Dar! Sorry about typos. I’m on my phone. That’ll teach me!

  10. OK Sweeneys: here’s a reality check. I work for Planned Parenthood. We run phone banks. We have an in-joke about “controlling spouses,” those who will not allow their wives to come to the phone. The wives in question have often indicated that they were pro-choice through signing a petition or similar.

    Now what happens when two ballots get to that household? Think that husband is going to give his partner the same privacy and freedom that a voting booth affords?

    It’s more than nostalgia. That voting booth gave each and every individual the privacy and freedom to make choices on their issues and candidates without pressure.

    And need I remind you that, even if you hated to attend, share a doughnut with the sweet old folks volunteering, and smile at other voters regardless of their choices, the absentee ballot option existed–just for you.

    • Very true, which is one of the reasons I had hoped the Auditor would provide a safe voting location. It wouldn’t take much work, just set up one booth in the office for people to vote in privacy in those sorts of situations. People could opt-in for that opportunity, while the rest of the population could enjoy the higher turnout that mail-in ballots provides.

      • @ Stephanie and Riley – I’d never thought of that issue, or the fix – a safe voting place. They could even have the safe voting booth’s nullify the mail in vote for those spouses that’ll fill it in anyway. It’s a sick world that we need to address things like this…

  11. The difference between early and late mail ballots is slight compared to the conservative bent of absentee ballots when we were voting at the polls. I have a longer discussion at

    The later counts in mail ballots usually look more like the overall result, so what little difference exists is in the early count, as Tim describes. The difference is very small and in some elections has not been detectable.

    I enjoyed election night when we voted at the polls. With mail ballots, election night is not enjoyable anymore. It’s a matter of taste. There’s no sense of ceremony or closure anymore. The slight turnout increase due to mail ballots seems inconsequential.

    Without knowing the demographic differences between the lines on Riley’s chart, all I can say is the only differences seem to be slightly different population sizes. Any other differences just look like statistical noise. People wait for deadlines, sales to end, etc. Procrastinators predominate.

  12. The curves seem to indicate there is still plenty of time to campaign after the ballots are mailed. I suppose that’s good for campaigns, but a bit painful for us slacker citizens who have to endure the doorknockers and phone calls.

    I don’t think trending in late returns is definitely dead, but I agree it is not what it used to be. I think it will be more in terms of data trends. For instance, if there had been significant developments in the coal port story, or some scandalous revelations regarding a candidate, later voters might herd up and show a trend. But, as a pundit pointed out over at NWCitizen, the media does of very good job of burying information on candidates and issues, so such trends will normally be small enough to go unnoticed.

    As far as polling goes, need a social voting experience? You can always throw your mail ballot away and go complete a provisional ballot at the Auditor’s desk. You could do it with your friends and cronies if you want to make a serious social occasion out of it. Do it late on election day and you might even get to wait in line a bit. Meanwhile, mail ballots have resulted in higher turnouts. Though I liked the poll experience and never had to wait three hours, better turnout is the best test!

    • Tip says, “…mail ballots have resulted in higher turnouts.” He doesn’t offer any statistics to support that conclusion.

      My judgement is completed different and it is based on the charts that have been printed in Whatcom Watch since the 2007 general election showing the number voting by mail and turnout. The number voting by mail increased from 21 percent in 1991 to 81 percent in 2002 and 100 percent in 2005. The 2011 primary election is an example of flat turnout. The 2011 primary turnout was 37 percent, that compares with a 34 percent turnout in the 1991 primary when 84 percent were poll voters. In previous odd and even year elections, both primary and general elections, the turnout results are the same. Voting by mail has not significantly increase turnout. The one thing voting by mail appears to have done is even out voting. In the past, there were large fluctuations in turnout depending on what appeared on the ballot. Now, people that only voted in peak years appear to vote in every election.

      The one exception to neutral turnout is presidential elections, the 2004 general election had the largest turnout in 66 years and 2008 general election had the largest turnout in 70 years.

  13. Mail balloting has changed the tempo so that about 2/3 of the ballots are cast before the last weekend. Look at the area under those curves, the final spurt isn’t even half of the votes cast.

    This has more or less wiped out volunteer political involvement and replaced with expensive mass-mailing and broadcast advertising. Campaign spending has increased a lot under mail balloting and participation has declined. That’s a large price to pay for a small increase in turnout.

  14. If you look at the area under the curves, the majority of votes were cast before the final weekend. That last spurt may be a bump in the curve, but it’s a small portion of the total.

    Spreading out the decision time vastly increased the amount of work for volunteer campaigns. Those are pretty much a thing of the past.

    Expenditures have skyrocketed and grassroots participation has been replaced by mass mailing and media buys. Under today’s rules, Tip Johnson, Louise Bjornsen and Barbara Brenner would never have had political careers.

    That’s a very high price to pay for a statistically questionable claim of slightly higher turnout.

  15. Look closely at the area under those curves: most of the votes, probably about 2/3 were cast before the last spurt at the final weekend. The graph is showing quantities, not rates. More were cast per day at the end, but less votes were cast in that spurt compared to the preceding two weeks. Riley has the numbers on a spreadsheet and can do the integration of the area under the curve. What percentage of votes were cast in the last four days?

    It’s an important point because it underscores how different the decision timetable is for poll versus mail campaigns. With poll voting, grassroots campaigns could concentrate on a last minute mobilization for voter contact and get out the vote efforts. With the decision smeared out over two weeks, that large amount of continuous labor is not available in a grassroots campaign.

    Mail balloting has undermined grass-roots campaigns. What used to be done by many small donations and volunteers is now mostly done by large donations, mass mailings and radio/tv ad buys. It has replaced citizen involvement with money – money controlled by a minority.

    Those of a mature frame of mind, as opposed to these feckless youngsters, should think back to the elections that put grass roots pols like Tip Johnson, Louise Bjornsen and Barbara Brenner in office. None of them would have had a political career under today’s mail ballot system. Instead it would be the likes of Jim Caldwell and Roger Almsgaar.

    The claim that mail voting has increased turnout is statistically weak. A questionable increase in turnout is a very large price to pay for sacrificing grass roots participation and jacking campaign expenditures into the stratosphere.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s


%d bloggers like this: