by Bob Mckenzie
The fight is over. In truth, it’s no contest and probably never was.
It never is. Not in a society like ours, where there’s such a premium placed on more information, more science, more technology or anything that helps to promote greater understanding and/or knowledge.
This same so-called advanced-statistics fight took place in baseball and basketball. The outcome was predictable there; it is in hockey, too.
Corsi, Fenwick, PDO—they’re here to stay. Unless, of course, they’re replaced by some other new innovative number or metric that is deemed to be better or more useful. Which is always a possibility, but it’s always going to be more, never less.
I know the argument against that. Hockey is different. It’s not baseball. It doesn’t lend itself to numerical evaluations because there’s a flow and unpredictability to it that doesn’t exist in any other sport. There’s some truth to that.
On that note, actually, the traditionalists and new-wave number guys do agree, in a manner of speaking. As analytics expert Gabe Desjardins freely admitted, “There is no Holy Grail number in hockey,” no one metric that singularly defines success, whether individually or as a team. For mathematicians, scientists, deep thinkers or whatever these #fancystats guys ultimately are, it was surprising to learn how much they do ascribe to, or write off as, old-fashioned puck luck. So much of what they try to quantify is set against the backdrop of hockey’s intrinsic randomness, which makes their pursuits all the more challenging.
Now, no one is forcing anyone to subscribe to their work. If you’re a fan, a member of the mainstream media, a blogger, an agent, a player, a coach, a manager, an owner or whatever, you’re free to go about your hockey business or pleasure as you see fit. Watch, play or coach the games through whatever lens you choose.
Numbers aren’t for everyone. I get that people may not want to embrace Corsi or Fenwick or PDO, but for the life of me, I don’t understand mocking or dismissing them, either.
Part of the problem, I think, is that critics of #fancystats may have as much or more animosity towards the individuals putting out the numbers as the numbers themselves. It’s a fascinating dynamic, as are all conflicts.
It’s fashionable to portray the advanced-stats crowd as geeky or nerdish, especially in a sport that is as macho as hockey, and quite a few them actually seem to embrace that in a humorous yet prideful sort of way. I mean, the #fancystats label is almost an embodiment of that. Ferrari joked often about being a “hockey nerd.” But you have to understand that these guys also see themselves as pioneers in a new field, that they’ve come up with some information that, in their minds, is incredibly revealing and relevant—groundbreaking, if you will. And for a time, maybe even still, no one wanted to listen or recognize their work. Worse still, they were dismissed out of hand; they were mocked.
As they made some inroads—information is power—they did become more empowered and confident. Where once they might have had a raging sense of insecurity about any role in the hockey world, in some instances that insecurity has morphed into an air of superiority. Their numbers add up, they will tell you, and if you’re going to be disrespectful to them—well, then, they might just call you on it. The ridiculees have become the ridiculers; they’re the smartest guys in the room, and if you don’t believe them, just ask (some of) them. That said, though, it’s been my experience that so many of those who are “into the numbers” are welcoming and receptive to sharing or explaining their craft to anyone who is interested or doesn’t understand it.
I shake my head sometimes at the trading of barbs between the #fancystats believers and non-believers. At times, it reminds me of the dynamic in society between creationists and the scientific community, though obviously on a much lower, more secular level. I understand the backlash coming from the stats guys, but there are times when I think they’d be far better off just letting their work speak for itself than antagonizing those who choose not to see it as they do. But I also understand their frustration at those within the hockey establishment who are not only wilfully ignorant but derisive of them and mocking of their work.
As usual, I always see the world in shades of grey when, it seems, most everyone else has it as black and white.
“Corsi is the best.”
“Corsi sucks.”
I can’t believe some colleagues in the media who are totally dismissive of advanced stats, who treat them with contempt, finding zero practical application for them. I’m equally miffed at how some in the advanced-stats community feel like the numbers are all that matters and how little time they have for the so-called intangibles.
Now, the mere mention of “intangibles” is a sore spot for many numbers guys. Rightly, they cite a lot of really bad decisions (that don’t add up statistically) being made because a certain player “is good in the room.” Fair enough on that. But there are still elements of hockey, which is an incredibly visceral team game, that can’t always be quantified. A player may have a fantastic set of advanced metrics, but he may also have personality traits that, in the context of a team sport, work against him or serve to discount his numbers in some way. Maybe he’s a bad teammate. Maybe he’s creating negative energy on the bench. Maybe he’s a coach killer. Or maybe he’s needy, in spite of his great numbers. Teams, like individual players, are living, breathing things. There’s no equation to measure that positive or negative energy.
Again, it seems like—some of the time, anyway—there are too many extreme viewpoints. Some guys believe in nothing but the magic and wonder of the game; some guys believe in nothing but what the numbers tell them. Each individual instinctively knows which side of the intangible/tangible side of the coin is more appealing to them, but I’ve always thought they shouldn’t be mutually exclusive.
At the end of the day, though, we all have to do what works best for each of us. As someone who’s not predisposed to enjoying working with numbers, that means challenging myself to keep an open mind on this relatively new information, embracing the parts of it I think make sense in terms of giving me a better understanding of the game and how I do my job. What metrics or numbers I can’t understand or choose not to use, I just set aside.
I look at so-called traditional hockey stats now, like plus-minus and conventional shots on goal, and I see virtually no value in them. From my perspective, keeping in mind my sterling math background, they seem antiquated.
Quite frankly, I’m incredulous that anyone could find more value in those numbers than Corsi or Fenwick or PDO. I don’t need to equate Corsi rankings to who will win or lose a game, or the Stanley Cup, but riddle me this: How could anyone think simple SOG is better than knowing how many shots were attempted (Corsi)?
Some of the advanced stats give me a better, clearer picture of what’s happening. It’s a big league. You can’t watch every game. I don’t need Corsi or Fenwick or PDO to be anything more than one tool in the belt. It’s just information, subjectively better in my mind than what’s currently being released as official by the NHL in their real-time stats.
One of the reasons advanced hockey stats are going to become more mainstream is better packaging. The traditional websites—Desjardins’s Behind the Net, for example, but there are countless others—have been a boon to the cause since their inception, but many of them were or still are decidedly user unfriendly, unless you’re a real numbers guy with a natural inclination for this stuff.
As you know, I’m not.
But along came, as an example, Extra Skater (www.extraskater.com), and suddenly #fancystats have a shiny, new user-friendly package that isn’t overly intimidating for a neophyte. When I’m watching an NHL game at work now, I’ve got the NHL website open to all the real-time stats for that game, and that’s great, but I’ll also have Extra Skater open as well, as a companion, for all that useful information I can’t get anywhere else.
When it came time for me to fill out my official NHL awards ballot in 2014, I got as
much or more relevant information from Extra Skater as the official NHL stats package. Between Corsi, Fenwick, PDO and a plethora of derivative metrics, it’s possible to get an incredible amount of detailed information on any player or any team. At a glance, you can see which players are being used in which situations, who’s matching up with whom, who’s playing with whom.
What’s not to like about having more information?
If there’s a day when I just want to put my feet up and watch a game, forget about the numbers, just soak up the wonder and magic of sport, that’s fine, too. There’s so much more to the game than math or science; there’s no quantifying emotion, drama, joy and heartbreak. The numbers are always there if I want or need them; they’re not going anywhere.
It’s funny, actually. While we think there’s this battle going on for mainstream acceptance of Corsi, Fenwick and PDO, and superficially I suppose there is, the reality is that next-generation numbers and principles are being worked on like never before.
Gabe Desjardins has, more or less, hung it up as a hockey analytics innovator. In 2014, he got an offer from an NHL club to jump on board and do some extensive research work. He thought about it, but ultimately took a pass, “because there are other things I want to do with my life now.”
Desjardins knows, though, there is still going to be a surge in the movement. He said chip technology exists right now (in 2014) that would allow sophisticated tracking of the detailed movements and actions of NHL players while on the ice, that it’s only a matter of time until that’s how new groundbreaking data will be collected.
For every pioneer like Gabe Desjardins or Vic Ferrari, who take a step back or drop out of sight entirely, there are more bright, young guys committed to the cause, always moving it forward because there’s no going back. Tulsky, the San Francisco–based Philadelphia Flyer fan (his Outnumbered blog can be found at www.sbnation.com/outnumbered), is known for doing groundbreaking work on zone entries and countless other intriguing concepts; Dellow is churning out novel looks at what happens from offensive-zone faceoffs and myriad other subjects.
“No one is writing better stuff in hockey right now than Tyler Dellow,” Ferrari said, a sentiment echoed by Desjardins. “Eric Tulsky’s work is fantastic, too. Those two guys are really good. There are some others too. My time is past. I’m happy to have played some small part in it. I take pride in that. I hope people listen now to guys like Tyler and Eric.”
Who knows what other nameless, faceless individuals are out there, who may one day find themselves working for an NHL team on some concept or metric we can’t even begin to imagine today, a measurement that could change the way we look at the game tomorrow.
“People need to understand, it’s not about the math,” Ferrari said of next-generation analytics. “It’s about thinking. The math Fs people up. A lot of what’s written now is way over the top with numbers. It’s crazy—kids with graduate degrees in stats or math, and they’re always looking for raw numbers and they want to throw them in the regression hopper and say, ‘Your answer is this.’ Hockey fans, people in general, don’t know what this shit means. So they’re either impressed or they say, ‘This is shit. You’re a nerd.’ Neither is a good thing. The kid with the PhD is probably a good person but he’s f---ing up really bad. The conversation doesn’t move forward. At some point, you don’t want the conversation to move forward. You want to move forward ahead of the conversation.”
It’s endless, really. The possibilities are infinite. What’s that old saw about putting 100 monkeys in front of 100 typewriters? One of them is going to come up with a masterpiece?
Or maybe just explain PDO.
The #fancystats battles will, to varying degrees, always exist in hockey—conflict is the nature of our game—but there’s going to be more mainstream acceptance of them with each passing day.
Count on it.
This, of course, coming from a guy decidedly lacking in numerical acumen; smart enough, though, to talk his way from a grade of 53 to a 55 to get out of a final math exam and still know either one of those numbers would pass for a damn fine Fenwick.
Note from the author:
If the premise of this chapter was to prove the 2013–14 NHL season was the turning point for hockey analytics, the events of the summer of 2014 validated that supposition, over and over again.
In the first few months after this chapter was written and put to bed, in May 2014, the magnitude of #fancystats developments was mind boggling. It was so great, in fact, that it was difficult if not impossible to rewrite the chapter to reflect those changes.
Instead, here is a brief summary of those noteworthy developments that made 2014 the Summer of Hockey Analytics. It’s yet even more compelling evidence of the rapidly evolving world of hockey #fancystats.
• In mid-June, the New Jersey Devils hired their analytics expert. It was Sunny Mehta, the former pro poker player and Ferrari protégé from Irreverent Oilers fame.
• In July, renowned stats blogger Eric Tulsky was hired on a part-time basis by an NHL club, calling into question what, if any, blogging he would be able to do on his site in the future.
• On July 22, the Maple Leafs, of all NHL teams, fired two assistant GMs (Dave Poulin and Claude Loiselle) and hired 28-year-old Kyle Dubas, the Sault Ste. Marie Greyhound general manager who is a huge proponent of advanced statistics and was to oversee that department as assistant GM of the Leafs. He was hired by team president Brendan Shanahan to work in concert with Toronto GM Dave Nonis.
• On August 5, stats blogger Tyler Dellow, who was interviewed extensively for this book’s #fancystats chapter, was hired by the Edmonton Oilers to work as a consultant to the hockey operations department, notably head coach Dallas Eakins. Upon his hiring, he took down his website.
• On August 6, the day after Dellow was hired in Edmonton, Hockey Night in Canada’s Elliotte Friedman wrote an excellent feature on www.cbc.ca that not only included an interview with the mythical Vic Ferrari, but also identified the mystery man as 47-year-old Tim Barnes, an engineer turned financial analyst from Alberta who had studied at the University of Calgary and had lived in England, Toronto and Edmonton before settling in Chicago. Friedman’s interview with Ferrari was the first ever published and his unmasking of Ferrari was groundbreaking within the hockey statistical community. Ferrari’s interview for this book was conducted three months before Friedman’s story was published.
• In mid-August, Dubas hired a “team” of analysts for the Maple Leafs. Two of them—Cam Charron and Rob Pettapiece—were #fancystats proponents from the blogging/writing community. The third was Darryl Metcalf, who created the Extra Skater website, which (sadly for those of us who used it) was taken down once he was hired by Toronto.
By any standard, the volume, rate and significance of 2014 off-season analytics-oriented developments were staggering, only reinforcing it as the year of #fancystats. But while two NHL teams gained analytics expertise in Tulsky and Dellow, the #fancystats blogging community also lost its two most noteworthy and original voices, to say nothing of the demise of Extra Skater, advanced hockey stats’ most user-friendly site.
CHAPTER 5
Grapes, Unplugged
Father and Son Cherish Precious Moments Watching
“The Best Hockey in the World”
“I hope we have a good referee for you. I don’t know this guy.
If the ref calls lots of penalties at the beginning, the game will be chaos.
If the ref is good, he lets the boys play.”—Don Cherry
* * *
It’s a blistering-cold Tuesday night in mid-December, and there’s pretty much just family and friends inside icy Victoria Village Arena. The utilitarian rink is tucked into an industrial section of the Don Mills area of Toronto, and the small group of fans sparsely dotting the stands are there for the 9 p.m. Greater Toronto Hockey League minor midget game be
tween the middle-of-the-pack Don Mills Flyers and the lower-end Toronto Young Nationals. It’s not exactly a marquee matchup. While some of the 15-year-old boys on both teams will get drafted into the Ontario Hockey League the following spring, and a couple may even play there as soon as the next season, there are no superstars on either team, no can’t-miss kids.
A veteran scout of more than 13 years, who works for the OHL Central Scouting Bureau, is in the house tonight. It’s his job. This is one of anywhere between two and four games he normally sees each week, and, as is their custom, his soon-to-be-80-year-old father has joined him on the drive from their Mississauga homes.
“I go with him all the time,” the father says. “If he goes, I go. He’ll phone me in the morning, and on the days I know we’re going to a game, that’s what I’ll look forward to all day. It’s joy—pure joy. I’d rather do this than anything. I’d rather be here than anywhere.”
The father and son walk to the far end of the rink, away from the arena entrance and past the few people who are there. They stand in the far corner, noses pressed right up to the glass, inches away from the action. There’s no thought given to sitting in the stands, even for the man whose 80th birthday (February 5) is less than two months away.
“Best place to watch a game,” the scout says. “You get a real feel for the speed and the size of the players. You can really see and hear everything here, too. It’s a little colder here, but it could be worse. In some rinks, when you stand in the corner, you’re actually standing on ice. Sometimes we’ll sit—maybe if it’s a doubleheader—but we like standing in the corner. That’s our spot.”
The father is wearing a black toque emblazoned with a red maple leaf, pulled down over his ears, tight to his eyes, to ward off the cold and dampness. The toque matches his all-black melton and leather jacket, which is also adorned with large red maple leaves, black sweat pants and black boots.