Polls Fairly Accurate Predicting Election

Voting Box

Election Polls have always been used to gauge the the mind set of the electorate for decades.

It gives voters a chance to see who may be forming the next government well before voters head to the polls to mark their ballots.

For some, polls are just entertainment – for others it takes a factual snapshot of the day that looks at the standings of the candidates. You can take them or leave them.

SaultOnline.com used 338Canada.com for most of the election campaign as the site tends to be fairly accurate in their findings. 338Canada.com takes several different polling agencies into account and then uses an average based on those polls.

This poll from August 26 shows a virtual tie between the Liberals and Conservatives in the Sault Riding

For the entire campaign, the polls showed not only a close race nationally, but a very close race in Sault Ste. Marie.

The campaign started with local incumbent Terry Sheehan in a comfortable lead, days into the election however, Sheehan and the Liberals started to see that lead shrink.

Conservative candidate moved from third place to a tight second and for a few polls showed Sonny Spina in the lead. The polls with a margin of error at plus or minus 7 points showed only a few point difference between the Liberals and Conservatives.

News agencies such as SaultOnline.com is often criticized for using polls, many believe the polls and media are bias in their findings. Regardless, the polls painted an accurate picture of how the vote may go.

That’s not to say that polls can sometimes get it totally wrong. The 2016 U.S. Presidential election is a good example of the pollsters getting it wrong.

In Canada however, the polls remain consistent with the actual vote on Election night.

338Canada.com predicted a narrow win for the Liberals federally, and also predicted a neck in neck race locally.

On the eve before the election, Sheehan was predicted to take 35% of the vote and ended with (so far) 37.5%   Conservative Sonny Spina  was forecast to take 30% and achieved 37.4% on the final tally with just four polls yet to count as on Monday night.

This poll released on September 7 shows the race even tighter with Sheehan taking a slight lead – just like the results from election night

Days before the election both Sheehan and Spina were basically tied heading into the last weekend before the September 20th election.

The mail-in vote with be counted by late Wednesday afternoon to declare an official winner in the riding. Terry Sheehan continues to lead slightly before the mail-in vote is counted.

Interestingly enough, an online poll conducted by SaultOnline.com showed a much different story, with Spina taking the seat comfortably with 53%. Online polls can not be used officially because of who is voting. It’s not random, as they are in phone polls and don’t take into account, sex, age or political affiliation. Meaning an online poll can be bias in the final result.


  1. Reader polls reflect the editorial stance of the media conducting them. Most people tend to read media which reflect their views editorially.

    Perhaps you need to reconsider your right wing editorial leanings Dan. Clearly you’re leaving a lot of potential readers (and advertising dollars) on the table.

    • Can’t handle balanced journalism? Craig is hardly a righty. Other than policing and calling out stuff in city hall, doesn’t strike me as overly political in tone. Would a right slanting publication have Peter Chow’s articles?

      I applaud Craig for balancing out the site nicely.

          • The evidence you provided in your article. You provided 338Canada poll results based on properly randomized samples accurately reflected the election results. The election results themselves (provided in another article on the site) accurately portray the political leanings of the population. The results of the Saultonline poll were also presented in your article and were heavily skewed to the political right when compared with actual results. Studies (e.g. those done by Jesper Strömbäck of the University of Gothenburg) have shown that political media polls reflect the editorial stance of the media conducting the poll.

            Therefore, the evidence (results of two polls & election result) indicate that saultonline’s editorial stance is to the right of the average Sault resident’s political view.

          • Craig, I realize I should also have apologized to you. My “not balanced journalism” comment could be misconstrued as referring to your article. In no way was I claiming your article was biased. It was fact based and well written. My contention rather, is that the overall political tone of the website leans right.

  2. Too bad that so many had to waste their votes causing the Liberals to win a minority and Sonny to possibly lose when he should have won hands down. Fingers are still crossed that blood stained Sheehan loses like he deserves to.

    • Ontario voters also just voted a Liberal into a riding that had a sexual assault in his history.
      Canadian voters are uninformed and lack knowledge. If they had taken a look at the actual platforms, performance and voting record that would have been an eye opener and possibly had a different vote outcome.
      But let’s just continue to complain about crime, cost of living, inflation while we vote in the party that’s largely been the cause of these increases.
      We vote liberals in for a couple terms because they’re flashy and sexy with good PR and then when they’ve effectively caused enough damage to Canada we then vote in the conservatives to clean up the mess, but the conservatives are un-sexy, unliked by the media and are mostly business so the high school popularity contest that is Canadian politics will vote the liberals back in.
      Lather, rinse, repeat.

Comments are closed.