Hey there, time traveller!
This article was published 21/9/2018 (1098 days ago), so information in it may no longer be current.
Hello, Winnipeg. I’m Andrew Berkshire -- and I’m going to be writing about the Jets this NHL season, as they attempt to build on the breakout campaign we had all been waiting for.
My niche in sports writing is analyzing the game using statistics different than what most people use (I have access to data provided by SPORTLOGiQ, a Montreal-based company that tracks every bit of minutia in the sport).
I will use this information to go beyond what we hear in the post-practice news briefings and after-the-game scrums -- providing insight into what is happening on the ice, how the team is trending in various areas of its game, and what we might expect to happen if those trends continue.
To kick things off, I thought it would be a good idea to introduce some of the concepts I work with regularly and what they’re useful for.
Scoring chances: One of the advantages of tracking every event on the ice is compiling accurate locations for shots, so we can evaluate how dangerous each shot attempt is.
Shot attempts from the slot area are identified as scoring chances, with shots on net from this area carrying an average expected shooting percentage of 17 per cent, while the average shot on net from the perimeter carries just a four per cent chance of beating a goaltender.
We count shot attempts from this area because total attempts are more predictive of future goals than just shots on net (though I often list both attempts from this area and shots on net).
High-danger scoring chances: When referring to high-danger chances, I’m talking about the inner slot area, which is where a whopping 48 per cent of NHL goals are scored.
Shooting from the inner slot gives an even more extreme bump to expected shooting percentage than getting into the slot overall, with shots on net from this area beating goaltenders 22 per cent of the time and shots from outside it averaging just a 5.6 per cent success rate.
Passes to the slot: Exactly what it sounds like. Getting into the slot area is important, and you can get the puck there by skating it in, recovering a loose puck, or receiving a pass.
The best option for scoring is receiving a pass, as it forces the goalie to move, creating more space for the shooter, and a more difficult save for the goaltender.
It’s no surprise star centres Sidney Crosby and Connor McDavid lead the NHL in such passes the last two seasons.
Possession driving plays/transition plays: Moving the puck with control is important; we know this from public research on controlled zone exits and controlled zone entries.
Combining all statistics that move the puck forward -- whether it’s skating the puck out of the defensive zone or over the red-line or passing the puck up the ice in the defensive or neutral zone -- reveals an all-encompassing transition statistic, allowing us to identify which players are responsible for moving the puck up the ice.
Scoring chance-generating plays: Counting scoring chances is great for goals, but what about quantifying playmaking?
Counting the plays that contribute to the creation of scoring chances gives us a ballpark figure of how many scoring chances a player creates for teammates overall.
Loose-puck recoveries: Exactly what it sounds like. For a huge portion of every hockey game, the puck is not under the control of either team. Loose-puck recoveries can be as simple as retrieving a dump in or picking up a loose rebound or a heated battle along the boards.
Loose-puck recoveries have a huge impact on shot-attempt metrics such as Corsi, as they determine puck possession more often than any other type of play.
Defensive plays: One of the main jobs players have while defending is to strip the puck from the opposing team. Defensive plays measure how often a player is able to accomplish that goal.
Combining blocked passes, stick checks, bodychecks, and shot blocks that result in a change of possession gives us a grouped stat that quantifies the area of defensive play.
Not surprisingly, Boston's Patrice Bergeron consistently leads the league in this area among forwards.
Turnover rate: The NHL lists giveaways on its real-time stats page, but they’re recorded very differently from arena to arena, and stripped of context. By contrasting how often a player turns over the puck with how many total plays with the puck they make, we get accurate turnover rates we can separate by zone.
With this information, it is possible to measure players relative to their own teammates because team structures vary in the NHL, and some teams are more prone to turnovers by design, as more dangerous plays are more difficult to complete.
Many of these statistics have predictive value, but for the most part, when I’m using them, it’s in a descriptive context. I like to answer the questions of how and why things are happening, and sometimes that leads into what comes next.
Sometimes the more interesting story is why a player is on a hot streak we know won’t continue, not that it will eventually end.
Next week, I will have a couple of stories: one on how I feel the Jets are inoculated from the type of collapse the Edmonton Oilers experienced last season; and second, a look at a few things the Jets will need to improve on if they hope to go deeper into the playoffs in 2018-19.
It should be an exciting season ahead for the Jets, and I look forward to being part of the Free Press coverage.
Andrew Berkshire is a hockey writer specializing in data-driven analysis of the game.