Skip to content

OUR VIEW: What strategic voters need to know

So you want to vote strategically? Let’s make one thing clear. We’re not fans of voting for someone you don’t really like to keep someone you really don’t like out of office, meanwhile forgoing a vote for the person you actually like.

So you want to vote strategically?

Let’s make one thing clear. We’re not fans of voting for someone you don’t really like to keep someone you really don’t like out of office, meanwhile forgoing a vote for the person you actually like. But with calls for strategic voting in tight races, including Burnaby North-Seymour, we’re going to give you a few words of warning. Make sure you know what you’re doing.

In order to vote strategically, you need to determine the candidates’ standings. And to do that, you need to look at the right polls, and there’s not a whole lot of riding-specific data available at this point.

Our story on page 1 features the latest poll from Insights West, commissioned by the Dogwood Initiative, one of the few polls that actually uses riding-specific data. But another Burnaby North-Seymour poll released shortly before our press time put the Tories in the lead with the Liberals and NDP tied for second. Polls are incredibly problematic because a flaw in the methodology can skew results, or even worse, crooked pollsters can “cook” the poll to get the results their clients are looking for. Here’s what to watch for when assessing polls: Look at the sample size. Is it a large enough randomized sample? Ideally, something like 1,000 people would be best, but 400 isn’t bad either. The smaller the sample size, the larger the margin of error. What’s the margin of error you say? It reflects how much the poll results could be off by. So if the margin of error is higher than the lead the first place party has, as is the case in Burnaby North-Seymour, the results are statistically insignificant. Look for undecided voters. Were they included in the poll? If a pollster has a sample size of say 400, but removes a large number of undecided voters from the final tally, that sample size has shrunk, which will affect the results. If possible, find out the order of questions used. Were there suggestive or damaging questions leading up to the final query? That can affect how people answer. Also, what methodology was used? Anything using self-selecting respondents or robocalls is suspect. Ideally, you want live calls with a randomized sample.

If you’re looking at sites like ThreeHundredEight.com and Signal.TheStar.com, remember these more closely reflect national trends than on-the-ground reality for local ridings. 

We’ve said this before and we’ll say it again: Take all polls with a giant grain of salt. Remember the last provincial election? Pollsters were wildly incorrect. Basing your vote on what the polls are showing is a risky game, and if you’re not careful, you could end up with the opposite of what you wanted. Proceed with caution.