A Passion for Numbers: Associate Professor of Statistics Alexandre Leblanc’s First Love Remains Constant

When asked what initially drew him to study statistics, Associate Professor Alex Leblanc pauses for a moment, then leans back in his chair and smiles. “It goes way back [to] when I was about seven or eight. I was drawn to numbers.”

Growing up a baseball fanatic in Montreal during the heyday of Major League Baseball’s Expos, the young Leblanc’s favourite pastime became looking up statistics for the team. Ultimately, it was that love of numbers that drew him to major in Math and minor in Stats in university, which translated to a Master’s degree and ultimately led to a PhD in Statistics.

“The modelling and the computing were things that I was interested in”, Leblanc says, warming to his topic. “They kind of drew me more into Stats than actual pure math.”

Leblanc remembers his grad student days and the longstanding debate between two primary schools of thought within Statistics: Frequentist and Bayesian. (The essential difference is how each group uses probability. Frequentists use probability only to model certain processes broadly described as “sampling”. Bayesians use probability more widely to model both sampling and other kinds of uncertainty.)

“There is a certain type of formalism that is really elegant … that the Bayesians are using, but I wouldn’t dare say that’s the only way to make it work. What I’ve been doing since I’ve been here is … nonparametric statistics, which is model-free statistics. The goal is to stay away from modelling assumptions as much as possible … Bayesians are the total opposite of this: ‘Whatever is going on, just add an extra layer in your models and you can try and tackle it.’  So, they’re … modelling to the extreme, while I’m trying to avoid modelling as much as possible.”

“I started to work on the Bayesian side of things, mostly because of the computing. I was very much interested in the applied math/computing aspect of it … when I was a grad student. That was at the beginning of when people were starting to think of more efficient algorithms and [they thought] ‘… [N]ow we have computers we can actually do these things, so let’s get down to business and do them, implement them.’  Which is where I came in.”

As for the common perception that Statistics isn’t as exciting as the “bench sciences”, like chemistry or microbiology, where scientists are able to combine chemicals and run visually interesting experiments, Leblanc begs to differ.

“I don’t run wet lab experiments, mixing stuff up and so on, but on my computer, I run all kinds of experiments. We use random number generation to simulate potential outcomes from one thing.”

Leblanc also points to the influence of big data and analytics, which have caused a virtual explosion of interest in sports statistics. As professional teams invest more money in statistical analyses as part of their overall strategies, statisticians are in demand. Leblanc is aware of a number of academics who have been offered lucrative contracts by professional sports teams in the U.S.

Leblanc cautions against thinking that gathering data is all-important. When all is said and done, statisticians need to be able to look at the data and interpret it correctly. “The statistician needs to be thinking about the link between the number crunching and the guy who needs the number crunching.  That’s one of our specialties.”

Beyond interpretation, statisticians must be able to think. Leblanc stresses that relying on computers has its limits.

“Some stuff doesn’t make sense. You need to know that it doesn’t make sense. You need to know why it doesn’t make sense. How to fix it. Because the data is complex. The stuff you’re actually doing mathematically … the computing is complex. It’s not true that we can solely rely on [computers]… You need to use your brain. It’s not just numbers. It’s not just the calculator. It’s more than that.”

When asked what he finds inspiring about statistics, Leblanc is typically thoughtful. “I would be tempted to say because with statistics … you end up helping a lot of other people in other disciplines. You’re … involved with all kinds of other interesting things: biology, genetics, health sciences, finance. … [However], I’m really more on the math side. I just like it for what it is … I’m a purist in that sense. I came at this with just a love for numbers and tables … I’m just content with that.”

Leblanc’s response to the imminent fiftieth anniversary celebration in his department?

“Over the last fifty years, our department has grown significantly and established itself in the Canadian community. It has a long history of offering solid methodological training to its students. [It’s] quite exciting, as we’re looking into adding solid computing and data science skills to the set of tools [in which] our students are going to be trained….[T]he next twenty years are going to be offering new challenges. It will be quite exciting, and I’m thrilled to be part of that.”

(For an impressive example of the insights you can get from good data analytics skills, check out this BBC video, The Joy of Stats, featuring Hans Rosling.

The Department of Statistics hosts several important events this June. On Saturday, June 10, it celebrates its 50th Anniversary, as the oldest statistics department in the country. The following day it hosts the Statistical Society of Canada 2017 Annual Meeting, June 11 – 14, at the University of Manitoba, Fort Garry campus for the first time in 30 years, approximately 500 delegates are expected.

What: Dept. of Statistics 50th Anniversary Reception
When: Saturday, June 10, 2017, 6:00 p.m.
Where: University of Manitoba, Fort Garry Campus

&

What: The Statistical Society of Canada – 2017 Annual Meeting
When: Sunday, June 11, 2017 to Wednesday, June 14, 2017
Where: University of Manitoba, Fort Garry Campus

For more information please contact the Department of Statistics.

Jo Davies