Strategic Voting – #hashtagfail

Strategic voting sites were all the rage this election. Websites like Project Democracy used advanced seat projection models and the latest polls to determine what ridings would be close and how progressive voters should ‘strategically vote’ to stop the Conservatives.

What an utter failure (or ‘hashtagfail’ as my buddy Jack would say). Pundits Guide warned us that Conservatives love strategic voting sites, but few people listened. The promise of having your vote count in a system where so many votes are wasted was too strong to resist.

If you look at the closest races in Canada involving the Conservatives, all of them decided by less 2% of the votes cast, the ‘strategic’ recommendations were horrible. In the 10 races where strategic voting could have been effective, Project Democracy got 5 of them wrong, and the Conservatives won all five of those ridings.

Riding Result PD Strategy Verdict
Montmagny – L’Islet – Kamouraska – Rivière-du-Loup Cons +1101 +10 over NDP Vote Bloc FAIL
Nipissing – Timiskaming Cons +14 over Libs Vote Liberal PASS
Etobicoke Centre Cons +25 over Libs Safe Liberal Seat – Vote Anyone FAIL
Yukon Cons +132 over Libs Vote Liberal PASS
Elmwood Transcona Cons +284 over NDP Safe NDP Seat – Vote Anyone FAIL
Esquimalt – Juan de Fuca NDP +406 over Cons Vote NDP PASS
Bramalea – Gore – Malton Cons +538 over NDP Vote Liberal FAIL
Don Valley West Cons +639 over Libs Vote Liberal PASS
Mississauga East – Cooksville Cons +661 over Libs Safe Liberal Seat – Vote Anyone FAIL
Lotbinière – Chutes-de-la-Chaudière Cons +777 over NDP Vote NDP PASS

1 – vote difference in Montmagny reduced to 10 after counting error detected.

Unfortunately, in a first-past-the-post voting system, only 2 parties can be viable alternatives. Strategic voting is a failed coping mechanism. We need to either change the voting system or get ready for polarized American-style politics with the Conservatives battling the NDP across Canada. Even as an NDP supporter, it’s not a prospect I’m looking forward to. I’d rather have more choices.

19 thoughts on “Strategic Voting – #hashtagfail”

  1. Try the same experiment on Catch22. It turns out they did even worse.

    I looked at the 14 seats where the Tories had the smallest margin of 14. Project Democracy got 6 of 14 wrong. Catch22: 10 of 14 wrong.

    Like

    1. Hi Mike Moffat, you seem to be presenting some misleading data. A fair analysis of a strategic vote site’s successes and failures would not include ridings where they chose not to take a stance. Check out http://dataj.posterous.com/a-conservative-majority-was-avoidable for a better analysis. Supporting the third place candidate against the Conservatives is an obvious failure. Not taking a stance or admitting ignorance on some ridings seems intellectually honest.

      One big challenge for strategic voting, of course, is getting good data. But that’s an empirical challenge.

      I hear a lot of right wing commentators complaining about strategic voting. I think people are only opposed to the principle of strategic voting for the following reasons:
      1) has an interest in splitting votes (e.g. right wing voter)
      2) rejects basic tenets of game theory

      Like

      1. Well, I’m definitely not right wing. I’ve just listened to the Liberals suggest an NDP or Green vote is ‘wasted’ for too long to think ‘strategic’ voting is a good idea, even if it did benefit the NDP this election. The non-Conservative parties are not interchangeable. There is a reason there is a Liberal, NDP, Bloc, and Green party.

        Instead of encouraging ‘strategic’ voting, parties should focus on convincing Conservative voters to switch parties. Every Conservative voter that flips gets you 2 votes closer to victory.

        Like

      2. I think it’s a good idea to flip Conservatives. Strategic voting does not happen instead of this. Strategic voting happens in addition to this. If you’re against strategic voting because you think there are always huge differences between the Greens, the Liberals and the NDP then just say so. You don’t need to dress up data to make this argument. In my view, some Liberals don’t have the right policy and values, so I would have a hard time voting strategically for them. But many Liberal MPs were, and are, sophisticated enough to warrant a strategic vote. Same argument applies to Greens and NDP, given the context.

        Like

  2. Ultimately all the energy used to convince supporters of party X to hold their nose to support Y to stop party Z is a flawed paradigm. The energy and effort needed to effectively strategically vote is so large as to be a waste of time and energy.

    Much better would be to put energy in locally to get people to vote for you preferred candidate. Given the low voter turnout, everyone knows at least a couple dozen people that are not voting. If you spend election day getting them to vote, and vote for your candidate, you will have a lot more success. A single person working hard on election day can bring in 20-40 non voting friends.

    Winning the election this time is also not the only thing, coming second matters because it is what everyone uses next time to decide who has a real chance to win or not. Many a VERY good person has not been elected because their party came third in the previous election and therefore they were written off as having no chance. Electoral politics are not a short game of a single election, you have to think at least two to four elections down the road and what needs to happen this time to change the future.

    Strategic voting is inherently short term thinking and even worse is backward looking. It also sends the wrong signal to the politicians. Many strategic voting sites were suggesting voting for the Bloc. The Bloc has no interest in good governance of Canada, good social services for the Canadian public, though good ones Quebec yes. The Bloc is also inherently tribalistic, it is an Us and Them La Pur Laine Contre Les Autres. Advocating a vote for them is not advocating a ‘progressive agenda’, what ever that means.

    Like

    1. I agree that it is good to bring in new voters. And I also agree that voting strategically is a short term decision. Short term thinking can still be good thinking. This is not an ‘exclusive or’ logic: this is an ‘and or’ situation: this is a let’s-use-every-tool-we-have situation.

      Tactical voting might have won the day for Randall Garrison in Esquimalt Juan de Fuca where he won by a very small margin over the Conservative.

      Like

  3. Cool analysis. The 2% marker might be arbitrary and skew the results though. Why not 3% or 1% or 5.5%? Your suggestion is important though! Are there studies that include more comprehensive data sets?

    Like

  4. Also, I just realized, if your findings do bear the weight of further scrutiny, might we alternatively conclude that strategic voting sites need better data on which to base their projection?

    Like

      1. Hi CanadianVeggie, I did read the article by Funke, but found it a little lacking. There are some good points especially about the relative weaknesses of strategic voting. But I can’t see how to conclude from the premise, that strategic voting has flaws, to the conclusion, that we should not vote strategically. Strategic voting, for example, might have carried the day in Esquimalt Juan de Fuca.

        Like

  5. If you look at some other areas where the vote wasn’t as close, the strategic voting sites also got it wrong. Their suggestions across New Brunswick were terrible, as they were wrong in quite a few ridings like Fredericton and Miramichi (and so ridiculously wrong in Saint John that it was hillarious).

    This whole scheme is flawed.

    Like

  6. counting ridings where the sites did not give advice is a bit statistically dishonest if your comparing it against voting for the party that best represents them.

    The proper way to phrase it would be Project Democracy was correct 5 times, wrong twice, and did not give advice 3 times.

    Like

    1. But they did give advice. They advised people that voting strategically wasn’t necessary. “Jim Maloway now looks safely on his way to defeat his Conservative rival. Vote your preference.” That sounds like advice to me.

      I would agree with you if in Montmagny they would have said it was too close to call between the NDP and Bloc for who had the best chance, but they didn’t. They explicitly made predictions in all 10 ridings.

      Like

  7. So your point then, is that voters were not strategic enough? You seem now to be arguing that voters should have voted tactically more often? If you are saying this, then I agree with you.

    Like

    1. What? No.
      Since you seem to be obsessed with strategic voting, can I ask you what you think the end purpose is? I realize the short term benefits (assuming you can accurately predict the outcomes ahead of time – which, as I tried to point out in this post, is very difficult), but what is your long-term plan for encouraging strategic voting? Is it a Liberal government? An NDP government? A Green-NDP-Liberal coalition?

      The only thing strategic voting seems to encourage is a 2-party system where the smaller parties supporters are constantly “strategically voting” for the largest party with views somewhat close to their own. Which is great if your goal is a 2-party system. Not so great if you think democracy is better served with more voting options.

      Like

  8. Full disclosure: I voted NDP in my riding and Denise Savoie who is an excellent candidate, won again: I asked Liberals to vote NDP in Esquimalt Juan de Fuca to vote NDP as well: I also asked NDPers and Liberals to vote for Elizabeth May in Sidney Saanich Gulf Islands.

    I’m not obsessed with strategic voting. I think there are many challenges to strategic voting. I think some of the problems and concerns raised by you and Funke and James McKinney and others are relevant and interesting.

    I just haven’t been convinced by any of these articles that we shouldn’t be interested in tactical voting. Of course it can be done badly. So what? It can be done well, as well. Of course it shouldn’t be done at the expense of other organizing. In my view, we should think and act and vote strategically all of the time.

    You know, Mike Moffat said yesterday that in his view, the strategic voting sites did more harm than good in this election. I think that is interesting. I’m not sure about that though and I would like to see the argument. But even if it’s true, that is not, by itself, an argument against strategic voting. It could mean, for example, that we would have to do it better next time.

    PS I think our FPTP system is broken by the way, and I’m totally in favour of electoral reform.

    Like

    1. Would you have encouraged your friends to vote for Elizabeth May in the absence of polls showing her ahead of the Liberals and NDP? I would have voted for her. That’s not ‘strategic voting’, that’s voting for the best candidate. I have no problem endorsing progressive candidates in certain ridings (like the Georgia Straight Slate), but it shouldn’t be based solely on polling data. Elizabeth May wouldn’t have stood a chance if people didn’t support her regardless of what the polls said.

      Like

  9. These are good questions. Actually the polls in May’s riding were done by her people. So we didn’t exactly have great data. And the NDP candidate was very good. And in the last election, the Liberal was awesome. Brionny Penn (sp?) lost because thousands of NDPers voted for a non-existent candidate. Fathom that.

    The thing is, no one is saying that a strategic vote should be based solely on polling data. I think. I don’t think they should say that. 🙂

    When I ask people to vote strategically, I can’t expect people to vote for a candidate they absolutely hate. They have to make an assessment. I think what is interesting is that your use of the phrase “best candidate” probably imports, on the sly, some probability. Probably part of your complicated matrix of ideas of “best” includes electability. You won’t throw your vote away if you think a truly terrible candidate x will win over a mediocre candidate y. This is the sense in which most people are already strategic voters.

    Perhaps what we’re really arguing about here is the degree to which people should accept a mediocre candidate. But I don’t want to argue about that.

    The Conservatives are truly terrible. They have a majority of the seats with a minority of votes. Many good MPs and many good candidates were dispatched because progressive voters couldn’t get their act together. I don’t want a two party system. Electoral reform might not be here for the next Federal election. In the mean time, we need to think and communicate very carefully about strategy.

    In my humble opinion, parties need to think hard about the possibility of withdrawing candidates in key ridings.

    Like

Leave a Reply

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

WordPress.com Logo

You are commenting using your WordPress.com 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