Sign in with your NHL account:
  • Submit
  • Or
  • Sign in with Google
Lightning vs. Hurricanes
Tracking the Storm

Numbers Don't Lie

Tuesday, 06.10.2014 / 11:30 AM / Tracking the Storm
By Michael Smith
Share with your Friends

Numbers Don\'t Lie
The Carolina Hurricanes are one of a growing number of NHL teams that recognize the utility of analytics in player evaluation.
Analytics. Advanced stats. Fancy stats. Whatever your descriptor of choice, these metrics are seeing increased usage in both the general hockey fandom and front offices around the National Hockey League.

Michael Smith
Follow on Twitter

“It’s a hot topic these days, and it’s something that the Hurricanes use,” said Darren Yorke, the Canes’ video scout and hockey operations assistant.

But that’s about all he’ll tell you. The metrics, how they’re used, how they’re obtained and how they’re tracked over an 82-game season – all of that is preserved within the walls of the Canes’ front office, for their eyes only.

“There are a bunch of different balances and checks that we consider when we try to put things together to make sure we’re doing things right,” said Canes Executive Vice President and General Manager Ron Francis.

While terminology, metrics and application of those metrics vary from team to team, the usage of analytics aims toward the goal of comprehensive player evaluation.

“The general public is probably most familiar with possession analytics, which help paint a picture about a player in terms of how often they’re gaining the offensive zone and driving possession,” Yorke explained. “There are other metrics and analytics we use to help paint a total picture on a player.”

In beginning to understand the scope of analytics and their usefulness, it’s important to first consider the definition. A statistic, for instance, is goals scored. An analytic is goals scored per 60 minutes. Where statistics are the tip of the numerical iceberg, the analytics loom beneath the surface.

“A statistic is just one measure. With an analytic, you’re furthering a statistic,” Yorke said. “An analytic can help you further analyze one metric.”

Possession analytics such as Corsi and Fenwick have recently inched into the national hockey discussion. With names derived from the folks who propagated the metrics, Corsi measures shot attempts – including those on net, blocked or missed – for and against while Fenwick does the same minus blocked shots. Both analytics are derived from basic stats that the league tracks on a game-by-game basis: shot attempts. And, both help to quantify puck possession, as they operate on the idea that the more shots a team takes, the more time it spends in the offensive zone. Of course, these metrics can be skewed – for instance, when a team has a power play or there is a wide margin in score – and there are ways to account for these instances, but taken as a whole, these two analytics can provide insight beyond basic stats.

Take, for example, the performance of the Hurricanes and specifically Jeff Skinner in the team’s 4-2 victory over the Buffalo Sabres on March 13. Carolina registered 91 shot attempts (55 on goal, 20 blocked and 16 missed) to Buffalo’s 46 (23 on goal, 14 blocked and nine missed), and a vast majority of this domination in puck possession came when the score was even. Buffalo took a 2-1 lead in the third period, setting up what would have been a maddeningly frustrating loss, considering that the severe tilt in puck possession should have led to the Canes scoring more than just a goal. Carolina would eventually pull away and win 4-2.

Also in that game, Skinner alone had 13 shots on goal (plus a blocked shot and a missed shot), setting a single-game career-high and franchise-high. He had two takeaways, no giveaways and made an appearance on the score sheet with his first-period tally in what was a dominant game for the 22-year-old. Skinner’s shot numbers from that game are certainly on the high end, but over the course of an 82-game season, it’s easy to see why he’s a reliable producer and an important asset for the Canes.

“In my world and in passing this information along to Ronnie, we take a look at quantitative and qualitative data,” Yorke said. “You get an unbiased view, especially from a quantitative perspective. Numbers don’t lie.”

An analytic, though, is just one form of evaluation used in concert with more traditional scouting techniques. Hockey, after all, is about wins and losses and scoring more goals than the other team. Analytics can help achieve this goal when assembling a roster.

“The main purpose of using quantitative data is to try to get an unbiased view on some things that your eyes don’t necessarily catch,” Yorke said. “We want to use everything. Combined with traditional scouting and video scouting, analytics can give you as much information as possible to find the best players.”

Baseball, a notably individual sport, is fully ensconced in the sabermetric movement. Last September, the NBA and STATS LLC announced an expanded and rather unprecedented partnership to install STATS’ SportsVU player tracking technology – a system of six cameras and proprietary technology – in all NBA arenas (15 teams already had partnerships). Whether from STATS or other competing companies, player tracking technology will make its way to the NHL and could be implemented league-wide by the start of the 2015-16 season.

“I think now that you have major sports leagues and teams using this information, hockey will be going down the same roads,” Yorke said.

Both in terms of analytics and player tracking, hockey is a bit more dynamic than baseball or basketball. Hockey features on-the-fly line changes with players constantly entering or exiting the playing surface. Hockey features a puck that can bounce this way or that way off a rut in the ice or an end board and can skitter underneath or through a number of obstacles – specifically bodies and legs – in trajectories unlike that of a ball. Not to mention, hockey is much more of a team-focused sport than the individual match-ups of baseball. And, sometimes there’s just dumb luck or a red-hot goaltender that could defy the odds.

But in the long run, it’s hard to argue against math.

“There are lots of things that may not be considered an analytic,” Yorke explained, “but if you can quantify it, you can turn it into a valuable measure.”




1 TBL 48 30 14 4 156 127 64
2 NYI 46 31 14 1 151 129 63
3 DET 47 27 11 9 139 119 63
4 MTL 45 29 13 3 123 106 61
5 PIT 46 26 12 8 138 117 60
6 NYR 44 27 13 4 134 106 58
7 WSH 46 24 13 9 137 120 57
8 BOS 48 25 16 7 126 121 57
9 FLA 44 20 14 10 107 122 50
10 OTT 46 19 18 9 126 128 47
11 TOR 48 22 23 3 142 150 47
12 PHI 48 19 22 7 130 146 45
13 CBJ 45 20 22 3 113 142 43
14 NJD 47 17 22 8 107 134 42
15 CAR 46 16 25 5 98 120 37
16 BUF 47 14 30 3 89 167 31


E. Staal 41 15 13 -8 28
J. Faulk 46 8 18 -14 26
J. Skinner 41 10 9 -7 19
E. Lindholm 46 9 10 -10 19
R. Nash 46 7 12 -4 19
J. Tlusty 39 11 7 -12 18
N. Gerbe 42 4 13 -2 17
V. Rask 46 6 9 -9 15
A. Sekera 44 1 14 -8 15
C. Terry 33 6 3 -3 9
A. Khudobin 5 8 2 .916 2.32
C. Ward 11 17 3 .911 2.45