As the votes were counted in the riding of Victoria during the byelection on Nov. 26, 2012, things were looking grim for the NDP candidate Murray Rankin.
As poll after poll reported, the Green Party’s Donald Galloway was in the lead and it seemed as though Rankin was on his way to losing the seat that New Democrat Denise Savoie had held since 2006.
But then, right at the end of the night, the last few polls reported their numbers and the NDP crushed in each one of them. Rankin would win, edging out the Green’s Galloway by 1,118 votes, a three per cent margin of victory, and go on to a Parliamentary career that will end with his retirement as an MP as of this 2019 general election.
What made the difference at the end of Rankin’s first narrow by-election victory? It was the advance polls, the last polls to be counted that byelection night. Galloway had won the most votes among those ballots dropped in the box on election day only, but the Greens could not overcome the “NDP machine” on southern Vancouver Island, and it was that NDP machine that pulled the vote with a better ground game than the Greens could manage.
The strength of that ground game was made manifest in the dominance by the NDP in the advance vote results.
It was one of the most striking examples of how a party’s focus on getting its partisans to a voting booth as soon as advance polls open can often make the difference between a win and a loss.
And with what appears to be a tight every-vote-counts and every-seat-counts general election right now, a party’s ability to put resources into getting out its vote over the weekend’s advance poll could mean the difference between government and opposition.
And if there is one federal party that has, for more than a decade, focused heavily on the value of an advance vote, it is the federal Conservatives.
From 2004 to 2011, the federal ridings with the highest advance vote turnouts tended to be ridings held by Conservatives. The priority on voter identification towards winning every day of an advance poll was a focus of Conservative war rooms led by the late Doug Finley and by his successor Jenni Byrne.
Some of the progeny from the Harper campaigns of 2004 and 2006 have implemented the same focus on ground game and advance polls. That would include Scheer’s war room general Hamish Marshall, as well as Nick Koolsbergen, campaign director for Jason Kenney’s successful United Conservative Party campaign in Alberta and Kory Teneycke, who ran Doug Ford’s successful Progressive Conservative campaign in Ontario.
Byrne, Koolsbergen and Teneycke all saw first-hand the kind of electoral success that came from Finley’s original playbook. Marshall’s success or failure on Oct. 21 could hinge on how well he and his campaign team absorbed that philosophy.
The Trudeau Liberals have not been unaware of this phenomenon. Indeed, one of the chief criticisms by the Trudeau insiders that won the 2015 election was that under leaders Paul Martin, Stephane Dion and Michael Ignatieff, too much emphasis was placed on the so-called “Air War” — advertising, speeches rallies, and so on — and not enough on the ground game.
But Trudeau’s 2015 campaign director Katie Telford and his current one, Jeremy Broadhurst, know they must match, if not beat, the Conservative ground game.
The Liberals found some success on that front the last time around. The five ridings in the country with the highest advance voter turnout in 2015 all went Liberal and, notably, were all in Atlantic Canada: the ridings of Fredericton, Beauséjour, Egmont, Acadie-Bathurst, and Madawaska-Restigouche. More importantly, four of those five ridings were Liberal pickups from other parties and only one, Beauséjour, was a Liberal hold.
Message: win the advance polls and you have a much better shot of winning after all the votes are counted on election day.
And while the New Democrats understand the value of driving voters to the polls as soon as possible — that 2012 byelection scare in Victoria emphasized that — the NDP in 2019 have far fewer resources than they’ve had in the last few federal elections. As a result, sources in the NDP war room acknowledge that winning advance polls this year will be a challenge for them, meaning their chances in tough races are further handicapped.
As for the Greens, organizing a sophisticated ground game, which can involve expensive voter contact databases and voter contact programs, has always been a problem. It’s likely one of the reasons that, very often, the Green’s popular vote, once all the votes are counted on election day, underperforms the level of support pollsters find for the Greens in the days leading up to the vote.
The Greens have simply been unable to capitalize on reservoirs of votes that might be out there for them.
Still in 2015, the NDP and the Conservatives significantly overperformed in their advance polls versus the Liberals and the Bloc Quebecois.
To illustrate how a strong ground game that focuses on advance polls can make a difference, Global News ran the 2015 election counting only those votes cast in advance polls and none of the votes cast on election day. About 3.7 million Canadians voted in 2015 in advance polls, which were about 20 per cent of all votes cast.
If only those votes had been counted and no others, the result would have been a Liberal minority of 163 seats with Thomas Mulcair and the NDP holding the balance of power with 52 seats. The Conservatives would have won 118 seats and the BQ would have won just 5 seats. (For this analysis, Global News counted polls in each riding which are numbered in the “600” range, i.e. poll 600, 601, 602, etc. )
Think of how the political history of the country might have been different if the Trudeau Liberals had only won a minority in 2015. NDP Leader Thomas Mulcair, holding the balance of power, would have almost certainly survived a leadership challenge, likely by using that balance of power to force Trudeau to keep his election promise to make 2015 the last first-past-the-post election.
But those considerations will have to remain in an alternate reality. In the actual event, the result was a healthy Liberal majority: Liberals 184, Conservatives 99, NDP 44, BQ 10 and Green 1.
So comparing our advance-polls-only election to the real thing, the Liberals would have won 19 fewer seats, the BQ five fewer, while the Conservatives would have won 19 more and the NDP 8 more.
Now, let’s run the 2015 election again but this time count only those ballots cast on election day and ignore any of the ballots cast in advance polls. The end result is a much stronger Liberal majority of 194 seats (+10 versus the actual result) with a smaller Conservative caucus of 96 (-3 vs. the actual result) and 37 for the NDP (-7 vs. the actual result) while the BQ and Greens won 10 and 1 respectively.
That’s the sort of result one would expect in a wave election where many voters had not made up their minds who to vote for or maybe had not even decided if they would vote until the end of the campaign.
But this 2019 election has a much different dynamic to it than the 2015 election. Voter turnout and each party’s ability to get its partisans to the polls will be critical.
For one thing, getting more of your supporters to an advance poll means freeing up more campaign resources on election day to get the balance of your identified supporters to vote. Second, it can help innoculate a campaign against any last-minute events that might cause a partisan to re-consider their support or not vote at all and mitigate the effects, as happened in 2015, of a wave moving away from your party.
Finally, when people vote, they tend to tell their friends and families and, by voting early, there can be days when that word-of-mouth can be effective rather than the few hours on an election day.
Two good illustrations from 2015 of the value of strong advance vote work:
In the Ontario riding of Barrie-Springwater-Oro-Medonte, Conservative Alex Nuttall was in a tight race against Liberal Brian Tamblyn. Nuttall won the advance polls 5,443 to 5,075, an advantage of 368 votes. But counting only the votes cast on election day, Tamblyn rode the Trudeau wave to a win: 15,930 to Nuttall’s 15,648, an advantage of 282 votes.
But Nuttall’s stronger performance in the advance vote gave him the narrowest overall win in the country: 86 votes.
The Conservative advance poll team, in that instance, had done just enough to stave off the Trudeau wave that crested on election day.
Another example: In the riding of Edmonton Mill Woods, Liberal Amarjeet Sohi would win the 2015 election by beating incumbent Conservative Tim Uppal by 92 votes. But Sohi would have to survive a last-minute wave in Alberta back towards Stephen Harper’s Conservatives.
Sohi would have lost the election if only the votes on election day were counted, but because he beat Uppal on the advance polls by 95 votes — he could overcome the tiny three-vote margin Uppal had on the election day-only vote. That’s how close it was on election day — a three vote difference — in a riding where more than 49,000 people voted. And it was a great example of how important the advance vote turnout can be for any campaign.
Alternate realities
Here is the list of ridings that had different results in 2015 if the vote count is restricted to either election day only or advance polls only versus the overall actual result:
Newfoundland and Labrador
- St John’s East: LPC wins overall but NDP wins advance polls.
New Brunswick
- Fundy Royal: LPC wins overall but CPC wins advance polls.
Quebec
- Beloeil-Chambly: NDP wins overall but LPC wins e-day only.
- Chicoutimi-Le Fjord: LPC wins overall but NDP wins advance polls.
- Gaspésie-Les Îles-de-la-Madeleine: LPC wins overall but NDP wins advance polls.
- Hochelaga: NDP wins overall but LPC wins e-day only.
- Joliette: BQ wins overall but NDP wins in advance polls only.
- La Pointe-de-l’Île: BQ wins overall but finishes NDP wins advance polls.
- Longueuil-Saint-Hubert: NDP wins overall but LPC wins e-day only.
- Louis-Hebert: LPC wins overall but CPC wins advance polls.
- Mirabel: BQ wins overall but NDP wins in advance polls only.
- Montmagny-L’Islet-Kamouraska-Rivière-du-Loup: CPC wins overall but LPC wins if only e-day votes are counted.
- Pierre-Boucher-Les Patriotes-Verchères: BQ wins overall but LPC wins advance polls.
- Quebec: LPC wins overall but NDP wins advance polls.
- Rivière-des-Mille-Îles: LPC wins overall but NDP wins advance polls.
- Rivière-du-Nord: BQ wins overall but NDP wins in advance polls only.
- Saint-Hyacinthe-Bagot: NDP wins overall but LPC wins advance polls.
- Trois-Rivières: NDP wins overall but LPC wins e-day only.
Ontario
- Barrie-Springwater-Oro-Medonte: CPC wins overall but LPC wins if only e-day votes are counted.
- Davenport: LPC wins overall but NDP wins advance polls.
- Hamilton East-Stoney Creek: LPC wins overall but NDP wins advance polls.
- Hastings-Lennox and Addington: LPC wins overall but CPC wins advance polls.
- Huron Bruce: CPC wins overall but LPC wins on advance polls.
- Kenora: LPC wins overall but CPC wins on advance polls.
- King-Vaughan: LPC wins overall but CPC wins on advance polls.
- Kitchener-Conestoga: CPC wins overall but LPC wins e-day only.
- London-Fanshawe: NDP wins overall but LPC wins advance polls.
- London North Centre: LPC wins overall but CPC wins on advance polls.
- Niagara Centre: LPC wins overall but CPC wins on advance polls.
- Ottawa Centre: LPC wins overall but NDP wins advance polls.
- Parkdale-High Park: LPC wins overall but NDP wins advance polls.
- St. Catharines: LPC wins overall but CPC wins on advance polls.
- York Centre: LPC wins overall but CPC wins on advance polls.
Manitoba
- Elmwood-Transcona: NDP wins but CPC wins on advance polls.
- Kildonan-St. Paul: LPC wins overall but CPC wins on advance polls.
Saskatchewan
- Desnethé-Missinippi-Churchill River: NDP wins overall but CPC wins on advance polls and LPC wins on e-day.
- Regina-Lewvan: NDP wins but CPC wins on advance polls.
- Saskatoon West: NDP wins but CPC wins on advance polls.
Alberta
- Calgary Confederation: CPC wins overall but LPC wins advance polls.
- Edmonton Mill Woods: LPC wins overall but CPC wins on e-day only.
British Columbia
- Burnaby North-Seymour: LPC wins overall but NDP wins advance polls.
- Burnaby South: NDP wins overall but LPC wins e-day.
- Central Okanagan-Similkameen-Nicola: CPC wins overall but LPC won e-day only.
- Coquitlam-Port Coquitlam: LPC wins overall but CPC wins on advance polls.
- Kelowna-Lake Country: LPC wins overall but CPC wins advance polls.
- Kootenay-Columbia: NDP wins overall but CPC wins advance polls.
- Pitt Meadows-Maple Ridge: LPC wins overall but CPC wins advance polls.
- Steveston-Richmond East: LPC wins overall but CPC wins advance polls.
- Surrey Centre: LPC wins overall but NDP wins advance polls.
- Vancouver East: NDP wins overall but LPC wins advance polls.
- Vancouver Granville: LPC wins overall but CPC wins advance polls.
- North Island-Powell River: NDP wins overall but CPC wins advance polls
David Akin is Chief Political Correspondent for Global News.