Resisting the pull of cynicism since 1969.

Monday, December 05, 2005

An idealistic pragmatist's guide to strategic voting

The inevitable argument about strategic voting has started early this time. Buzz Hargrove is telling us to vote Liberal to keep the Conservatives out. Paul Wells is telling us that any NDP supporter who votes Liberal to keep the Conservatives out is shooting himself in the foot. Well, guess what? They're both wrong.

It's all well and good to say that this election should be about issues, and I don't disagree. In principle. But strategic voting is an inevitable by-product of our electoral system. There's a huge body of political science scholarship on it, and while the experts may not agree on everything, they certainly agree that it's here to stay. Simply saying "don't vote strategically! ever!!!" makes you look idealistic to the point of naiveté, and people will simply tune you out. Where the real problems occur is when people try to vote strategically and end up using faulty strategy. This is what happened all across Saskatchewan in 2004 when the NDP lost seat after seat to the Conservatives by only a handful of votes, with nary a Liberal to be found. In other words, I'm with the guy on the fence: if you must vote strategically, make sure you know what you're doing.

The key to making sure you know what you're doing, of course, is knowing a lot about the individual riding you live in. The biggest mistake people make in a strategic voting decision is thinking the big national polls about how each party is doing across the country are relevant to how they should vote. Until we have proportional representation, the national percentages have nothing at all to do with your vote, so you can ignore them completely. Concentrate instead on two things: 1) your riding's results in the last election, and 2) this year's campaigns in your riding. By looking at last time's results, you can figure out whether your riding is a safe seat, a seat contested between two parties, or a seat contested between three parties (rule of thumb: a "safe" riding is usually one that was won by at least 50% of the vote last time, a two-way race is one where two of the parties each got around 40% with everyone else lagging behind, and a three-way-race is one where three different parties each got at least 20%). Looking at this year's campaigns can tell you who's running serious candidates with well-funded and well-organized campaign teams behind them, and who's not really trying to win that riding (have the candidates been doorknocking like crazy? have they been sending out lots of campaign literature, do they have an informative campaign website? do they have lots of events planned, have you seen them in the media?). Once you have both of those pieces of information, then, you can move on to some actual informed strategy.

If you're living in a "safe" riding according to the 2004 results, double-check to make sure something hasn't changed drastically since then. If your current MP hasn't retired or died, if the opposition parties aren't running high-profile candidates with really well-organized campaigns, then you can safely assume you're going to end up with the same MP this time. This means that you can choose the candidate you most prefer without worrying about trying to vote strategically. Vote anyway, of course--each vote your party of choice receives means another $1.75 into their coffers, and this year's results will also help influence how each party decides to approach the next election. But you can safely vote your conscience.

If your riding is contested between two parties and the other parties are non-entities, then your strategy is simple. First, again, check to make sure something hasn't changed drastically since last time. If things look pretty much like they did in 2004, then look at the two candidates who have a chance at winning and vote for the one you prefer. This is a classic strategic vote, and it makes a great deal of sense.

If your riding is contested between three parties, then your race is wide open. I'd actually advise not trying to vote strategically in a case like this because it's too easy to get things wrong and adversely affect the party you really want to see win. But if you really feel strongly about trying to cast a strategic vote, then you've got to make that decision based on more than just numbers. Who's running--are all three candidates known quantities? How much visible support do they each have in terms of lawn signs or media attention? How well-organized are the three campaign teams--do they have good communications material, and are they taking part in events that make their candidates more visible? If you've met the three candidates, which ones seem more qualified than others? Other people are going to be facing the same decision, and they're going to have the same kinds of information you have, so there's a good chance they're going to be taking the same factors into account.

If you're not sure what to do after looking at all of the information at your disposal, then just vote for the candidate you'd most like to see in Ottawa. Voting strategies only make sense when they have an actual chance at affecting the outcome, and if you've taken a good look at the situation and see no obvious patterns, odds are it's a pretty close race. Vote your conscience, and may the best candidate win.

It's our first-past-the-post electoral system that makes this kind of thing necessary, and until we reform the system, we're stuck with it. Use it wisely, and if you must, use it strategically. But when you do, make sure you know your individual riding's situation well enough to make a truly informed decision.


Anonymous said...

Excellent post. On TV, the message should be "don't vote strategically". In print, it should be "go to Idealistic Pragmatist to see how to vote strategically".

Of course, you assume that only a small fraction of people vote this way. It's quite possible that 60% of the people in Westmount actually want the NDP to win, but vote Liberal to keep out the Bloc.

Idealistic Pragmatist said...


Aww. Thanks. :-) Although your theory about Westmount made my head hurt.

Anonymous said...

IP, this may be of interest to your viewers. CBC has now set up an online survey via which participants may determine which party best represents his/her ideologies. Completing the survey only reminded me that my allegiance remains unchanged!

Idealistic Pragmatist said...


Thanks! That looks great.

Q. Pheevr said...

This is one of the most sensible things I've ever read about strategic voting. In particular, I very much agree with your observations about the irrelevance of national percentages as opposed to how things look within one's own riding. However, I do think it's possible to vote strategically on a national level, provided one thinks in terms of seats rather than raw poll figures.

For example, suppose it's 1993 and I'm a left-leaning federalist voting in the riding of La Prairie. I've seen enough polls to be confident that the Tories, including the local incumbent, Fernand Jourdenais, are on their way out, and the Liberals are headed for a strong majority. My real sympathies are with the NDP, but in my riding, the NDP has about as much chance as the Natural Law Party. (Five years earlier, nearly fourteen hundred people in my riding voted for the Parti Rhinocéros, but Jourdenais was pretty much a shoo-in that time around.) So the real race here is between the Liberal candidate, Jacques Saada, and the B.Q. candidate, Richard Bélisle, and it looks pretty close. Naturally, I prefer the Liberals to the Bloc, but I'm confident that the Liberals are going to form the next government no matter what. The people who really scare me are the Reform Party--they're as much of a force for decentralization as the Bloc, and a whole heck of a lot farther to the right on social issues. And at the federal level, Reform and the Bloc are competing with each other for second place, even though they're not in direct competition in any one riding (the Reform Party have no candidates inside Québec; the Bloc have none outside). So I strategically cast my vote for Bélisle. And it works! Bélisle narrowly beats Saada, which does little to dampen the spirits of the Liberals, but which helps ensure that Lucien Bouchard becomes the next leader of Her Majesty's Loyal Opposition instead of Preston Manning. I'd rather it had been Audrey McLaughlin, of course, and I'm very sorry to see my own team lose official party status, but there really wasn't anything I could do about that.

In this hypothetical example, the strategic vote depends on both the percentages in the voter's own riding and the likely distribution of seats at the federal level.

Idealistic Pragmatist said...


Okay, I somehow missed your "suppose it's 1993" line the first time around, and I got very confused when you started in about the Reform Party and Lucien Bouchard. :-)

Anyway, I can certainly see the logic behind a strategic vote like that, but I don't think most voters think about it that hard. For all too many people, it's just: "Argh! the Conservatives are too high in the polls for my comfort! I'd better vote Liberal!"

Q. Pheevr said...

Oh, well, that's not strategic voting; that's just panic. It's not strategy unless you think about it. So I'm just observing that when one does think about it seriously, as you obviously have, there are sometimes occasions when it's possible to vote strategically at two different levels.

Patrick said...

Stephane Dion is reported today ( as saying it is preferable to vote Conservative or NDP than to vote Bloc. Making him your highest level endorsement.

I would add to your succinct and excellent analysis only this one point. You have to believe there are two federalist parties that deserve to be in a position to influence the decisions of the next government for strategic voting to make sense. In my case, I have long thought the NDP to be a party of principle, and so when faced in my own riding with a situation where the Conservative doesn't have a chance, I planned to vote NDP. I'm changing my mind rapidly every time I see Jack tear into the Conservatives, like a girl in the schoolyard trying to get the two biggest boys to fight. Jack should be aspiring to climb over the carcass of the Liberals, not to bring the frontrunner down despite the implications for Canada and his own party.

If the Conservatives and Liberals are close to a tie in the seat count, the NDP will lose the balance of power. Simple as that. Jack had better smarten up, or he'll tear down the only other party capable of providing an ethical alternative to the Liberals.