I was recently asked by an experienced coach to explain the differences between statistics and analytics. It was a fair question to ask from his point of view, since he’s been keeping track of numbers on shots, faceoff wins, hits and turnovers for his team for years. My answer to him was the following:
“Statistics like goals, assists and wins are really good at telling you what happened in the recent past. Analytics is about finding numbers that help you predict the future, which not all stats are very good at.”
Casting a net
In any case, I figured it’d be fun to put theory into action and try to figure out who the NHL’s next 30-goal scorer(s) could be. “The NHL’s Next 50-Goal Scorer” has a better ring to it, but with offense down so much across the league, we might be waiting for a while.
Searching for the NHL’s top goal scorers among forwards aged 25 or younger provides us with a good starting point. We are looking for players who are already starting to make their bones in the toughest league in the world, and who are young enough to still be on the front half of their career arcs. However, we also need to control for number of years in the NHL since our initial list is dominated by guys like Stamkos, Seguin, Tavares, Nugent-Hopkins; former top picks who have been in the Show since their late teens; players who everyone (and their great-aunt) know to be superstars of the game.
Once we control for that factor and restrict the list to players with three years or less of NHL experience, we get a list which looks like this (on April 1, 2015) :
(source: Hockey-Reference, data retrieved on April 1, 2015)
Comparison shopping
The top-20 of our list features some pretty recognizable names, including Vladimir Tarasenko, who is already a 30-goal man. Last year’s Calder Trophy winner Nathan MacKinnon, suffering through the dreaded “sophomore slump,” just misses the cut at #21.
As things stand, these players’ statistics in the Goal column suggest that most, if not all, of them will indeed score 30 at some point in their careers. But of course, these are stats. Good for telling us what happened, not so good at telling us what’s going to happen. In order to invoke the powers of analytics, we’ll have to go through a few more steps.
1) Look at individual shooting percentages
At the simplest level, a scorer can create goals in one of two ways: either put a ton of shots on net, and hope that some will go in, or take just a few shots, and hope that most of them will go in. Historically, in pro hockey, the first approach is a far better recipe for enduring success.
If we assume that the top young snipers in the world can convert their scoring chances at a 10% clip (as opposed to the league-average 8%), then we still notice a few abnormally lucky shooters this season. We needn’t concern ourselves with Tarasenko anymore since he’s already scored 36 and is therefore out of the running, but then we see Sean Monahan (who has since reached the 30-goal plateau) and Nikita Kucherov shooting at over 15% and wonder whether that’s really sustainable (most likely no). On the other side of the spectrum, we’ve got MacKinnon (7.3% this year after shooting 10.0% last season) and Jonathan Huberdeau (8.2% over 2 seasons after winning the Calder in 2013)
2) Look at powerplay goals
It’s easier to score at five-on-four than at even strength at any level of hockey, especially in the NHL. While some players’ skillsets are more conducive to putting up numbers on the powerplay, opportunity is the biggest factor to consider when examining a young scorer’s raw numbers on the man advantage.
Sean Monahan has 10 powerplay goals to Kucherov’s 2, but then again, he isn’t playing for a coach who has the option to roll Steven Stamkos, Valtteri Filppula or Ryan Callahan over the boards on the first PP unit.
3) Look at advanced metrics
After looking through “traditional statistics” while being mindful of context, we can dive into advanced stats in the form of usage-adjusted Corsi and per-60 measures in order to get to know these young players even better. Going through their HERO charts, three distinct “player personas” emerge:
A) Young Veterans
These guys are good and their coaches know it. They play top-6 minutes, don’t hesitate to shoot the puck, drive possession when they’re on the ice, and score at a high clip even when accounting for their ice time. They probably get lots of looks on the powerplay and are generally trusted by their teams to be difference-makers despite their young age.
Prime Example: Filip Forsberg, Brendan Gallagher, Nathan MacKinnon
B) Sheltered Kids
Whether it is because of perceived defensive flaws, or because they play on a deep team with a veteran-friendly coach, these players just aren’t getting as much ice time as their offensive efficiency suggests they should. Good news is, the only thing holding these guys back is a lack of opportunity. Chance are, your next fantasy hockey stud is currently a Sheltered Kid.
Prime Examples: Anders Lee, Mark Stone, Tyler Toffoli
C) At-risk Youth
While is it a bit early in the game to proclaim these players as flashes in the pan, there are definitely warning signs associated with young scorers who are putting up numbers right now, but without generating strong underlying numbers in terms of puck possession, individual shot attempts and even-strength scoring.
When their high shooting percentage deserts them (and it eventually will), they may not create enough positive things on the ice in order to keep earning their coaches’ trust. The inevitable slide down the depth chart may also lead others (media, fans, execs) to unfairly brand the player as “lazy” or “streaky.” These are your future Petr Pruchas, Wojtek Wolskis and Chris Stewarts – guys who put up 20-plus goals at a young age, and then fall off the face of the Earth due to a series of unfortunate events.
NHL organizations would be well-served to find out which of their players fit under their archetype and find ways to improve their all-around play before confidence, patience and trade value run low. As for fans and fantasy hockey owners, best to thread carefully here.
Prime Examples: Nick Bjugstad (especially considering his recent back surgery), Sean Monahan, Nail Yakupov
What it means for the Habs
With a career-high 23 goals and 3.01 shots per game, Brendan Gallagher should be well positioned to become his team’s, and the NHL’s, next 30-goal scorer. This is especially true considering Gallagher has been one of the Canadiens’ best puck-possession players since breaking into the team. Even when he’s not putting the puck in the net, he brings enough to the party to merit a spot in the top-six.
The main factor holding back his offensive production happens to be a lack of powerplay production, as he only has three PPGs on the year and just 14 during his entire NHL career. It’s interesting to note that Brian Gionta, a winger similar in style and stature to Gallagher, had 33 even-strength goals and just 2 powerplay goals in his first three seasons in the NHL before exploding for 48 total goals in 2005-06, the post-lockout year during which there was a record number of penalties called. That year, the 27 year-old Gionta lit the lamp 23 times at even strength and added a career-high 24 goals on the man advantage. He would never score more than 11 times on the PP in any other season, nor would he ever surpass the 30-goal mark again.
