For a Pollster, the Provincial Election was a Decisive Win
Four years after an embarrassing debacle, Mario Canseco found redemption in predicting this month's too-close-to-call contest. But will anyone remember?
May 18, 2017
In May 2013, I spent a Tuesday night, on live television, figuring out what had gone wrong in the British Columbia electoral forecast, the 33rd voting process I had worked on as a public opinion pollster in North America and the first one where the results were incorrect. Massively incorrect. Not just for me, but for everyone who published their findings before election day.
The missed forecast was met with extreme derision. By Wednesday morning, my inbox was flooded with invitations to do something else with my life.
There was also empathy from unexpected places. There was the voter from Kerrisdale who apologized for changing her mind “at the polling station” after she took the survey and told me she was going to vote for a specific party. There were friends, former colleagues, reporters, even a wonderful lady in Saskatchewan who I once had a brief e-mail exchange with. When the pile-up was roughest, all these people took a minute to send their best. I will help them move a piano. Anytime, anywhere.
About 20 years earlier, the Philadelphia Phillies—my favourite Major League Baseball team—were defeated by the Toronto Blue Jays in the World Series. Phillies manager Jim Fregosi was questioned for his decision to stick with controversial closer Mitch Williams in spite of a few hiccups. Fregosi consistently answered: “Without Mitch, we would have never got to the point that we got.”
A minute after arriving at the office Wednesday morning, the day after the 2013 election, I realized I did not work for Jim Fregosi. I could not stay at a place where people had actively rooted for my failure.
Thursday was even weirder. After two days of televised apologies, a stranger said to me at the corner of Georgia and Cordova: “I know you! You’re that guy everybody hates.” We chatted briefly. He left. I was ready for more questions from reporters, but the phone stopped ringing. The first revelations of the Rob Ford crack video gave the media more entertainment than I could provide.
When You’re Right…
Since I was a kid, I had a particular obsession with looking at who else was born on the same day as me. My favourite birth-mate is Muhammad Ali, who once quipped: “When you’re right, no one remembers. But nobody forgets when you make a mistake.”
In November 2012, I spent a weekend at home, alone, drowning in data and figuring out 10 electoral forecasts for five American states and the nationwide presidential race while listening to a never-ending loop of U2’s The Joshua Tree. I was forecasting the election of God’s Country, summoning Lady Liberty to rescue me. She did. From Vancouver came the best prediction of (what was then) the most polled presidential race in history. But nobody remembers it. Because, as Ali would say, it was right.
By June 2013, I found myself at a superior company, with outstanding colleagues and a clear plan. I pleaded my boss to make the most of every opportunity for electoral forecasting redemption. He agreed and trusted me. We went to work on a wide range of issues that mattered to British Columbians and Albertans, later Canadians and, later still, Americans. We were mocked in the early going. Nobody could trust polls anymore, it seemed. Still, we endeavoured to go on.
We worked on 22 different elections in two countries at Insights West—all correct. In Alberta’s 2015 election, in which more than 40 years of Progressive Conservative rule ended in favour of the NDP, data-less pundits suggested we were going to be wrong. In the Metro Vancouver’s transit plebiscite, nobody else polled before the ballots arrived in the mail. In the United States, we went 14-for-14 and had the second best forecast in the country. None of it was as meaningful as the challenge I spent four years working towards: B.C.’s provincial ballot in May 2017.
I spent the weekend before the B.C. election, and most of Monday, looking at the data coming in and deciphering who was going to vote and how. On election day, trolls parroted the lines of pseudo-experts and backroom performers who can explain anything without numbers after the fact. It is a luxury that the professional pollster, the very few of us left, does not have.
Four years after that difficult night, I was back at the studio where I faced the cameras in 2013. “Are you going to be right this time, Mario?” a host joked during make-up. I struggled to find the right words, but offered reassurance.
The results started to come in. Our forecast would be close to perfect. My first thoughts drifted to my family, who guided me through these four years with nothing but encouragement and love. We were right. It felt great. But as Ali would say, nobody will remember it. On to the next one.
What do you think of the B.C. provincial election results? Let us know in the comments below!