SvoNotes: Surviving in the Time of the Statistical Revolution

By Jeff Svoboda on November 29, 2017 at 8:40 am
Even John Tortorella likes numbers ... or so he says
Aaron Doster-USA TODAY Sports

Last Wednesday night, as I trekked home on the Ohio Turnpike to spend Thanksgiving with my family, I found something akin to a unicorn: a northern Ohio radio station broadcasting the Blue Jackets' game vs. Calgary.

(I only slightly kid; one battle the Jackets really need to fight is finding a way to better expose the team's game broadcasts throughout the state.)

Soon enough, I heard I was in for another treat: My good friend Alison Lukan would be joining the broadcast during the first intermission. (Another aside: If you're not following Alison, you should be, and not only for the pupper pics.)

Little did I know I was in for a show. One of Lukan's niches when it comes to covering the Jackets is how she has embraced analytics, and one of her weekly pieces at The Athletic is a deep dive into one advanced stats topic or another.

Very quickly, Blue Jackets radio voice Bob McElligott made his thoughts on advanced stats clear.

"There's a place for those numbers," McElligott said, though I must admit I'm paraphrasing based on memory. "Where is that garbage can again?"

The discussion continued into Lukan's appearance on the air, leading to a lively debate over the merits of stats-based analysis (Lukan's side) vs. just watching the dang game (McElligott's take, again paraphrased). It truly was great radio, and you can listen to it here.

So, where do we go from here?

Well, I don't have to say anything, I guess. But that wouldn't make a great column.

I'll start here and admit I'm generally a believer in numbers. As a baseball fan, I've followed the sabermetrics revolution in that sport with great interest and come to accept that there are simply better ways to evaluate players than just home runs, pitcher wins and RBI.

The whole field grew largely out of the writing of a number of rather smart people, perhaps most famously Bill James, who didn't really set out to revolutionize the game of baseball. He just thought some of what people had come to accept as gospel was generally balderdash. If you're going to say or believe something, essentially you ought to be able to show your work, James thought, and went about doing so. 

Baseball is a sport that particularly lends itself to statistical analysis, I think, because the game at its heart is a battle between the batter and pitcher. It's a two-man fight with a largely defined number of outcomes, and that particular competition repeats itself thousands of times throughout a season. Do it enough times and you get some data ripe for studying.

The analytics revolution has come in different ways in different sports, and it seems to be in its infancy (or perhaps young adult years) in hockey. As a free-flowing sport, hockey has an almost infinite number of variables, so it can be quite tough to study, and many of the people doing cutting-edge statistical work are trying to distill what exactly means what. 

At the same time, let's be honest here – it's kind of silly to think we can just stop at goals, assists, penalty minutes and save percentage and think we've figured out all there is in this great sport (don't even get stat people started on plus-minus). Sure, those numbers are useful, and a 30-goal scorer had a good year and a guy with 10 straight 30-goal seasons is probably pretty darn good.

But take assists, which aren't really created equal. A second assist and a first assist can be quite different; so can a second assist and a second assist. Did the pass come 10 seconds before a goal, or was it the first of a tic-tac-toe series? What about penalty minutes? Does that stat mean someone's a fighter or a flat-footed defenseman who takes a bunch of hooking penalties? And save percentage, well, I don't think I have to explain all saves have differing degrees of difficulty.

In other words, what people are looking for when they look at stats is looking for knowledge. Of course it's good to know how many goals and assists a player has each season. But adding context to those stats can be just as important as the stats themselves.

Some might prefer to stick with the traditional numbers, and that, of course, is fine. And I've made it clear a few times in this space – if you're looking for deep-end statistical analysis, this is not the spot for it. While I enjoy looking at some new ways to evaluate players and using that information to predict what might happen in the future, I'm far from an Alison Lukan or a Dom Luszczyszyn. That hasn't stopped me from finding things in their writing that help me look at teams, players and the sport in different ways. And it's also clear organizations like the Blue Jackets are using data in a multitude of ways to help them build a winner.

But as someone who watched the information revolution hit baseball, I think cultivating a divide between the two sides is folly. Everyone seems to have either read "Moneyball," saw the movie, or at least has used the term to talk about stats. One of the central conflicts of the book is how the statistics community and the scouting community were often at odds with one another, and the book is not shy about taking the side of the former.

The truth is there's no real reason for conflict. There's a place for the Bob McElligotts of the world, who prefer to use their eyes to tell them the story. You can learn a lot by watching. But there's also a place for the Alison Lukans, who are constantly pushing for different ways to learn about the game and its players.


View 2 Comments