Charts & Graphs - Jan. 23, 2014
"Do you know what the perfect game is? The perfect game is no hits. Do you know why that is? It's because you have the puck. You don't have to hit anybody, you have the puck."
The Edmonton press wanted to grill Eakins on his team's lack of response against Zack Kassian, who has been a veritable thorn in the Oilers side all season, but all Eakins wanted to do was toss heuristics used for a year or so in hockey's online analytical world. "Hits", which are recorded by the National Hockey League and often used (incorrectly) as a barometer for physical play, also do a wonderful job measuring which players have the puck the most and the least.
To illustrate this point, did you know who the player with the lowest hit differential in the NHL is? Chris Tanev, according to ExtraSkater, has taken 137 hits and has thrown just 21 all season. His -116 hit differential is the lowest in the NHL. The second lowest? Jonathan Toews (22 hits thrown and 108 times hit). Third lowest? Alex Pietrangelo (13 hits thrown and 89 times hit).
It goes further than this. Things like blocked shots, giveaways, takeaways and hits are recorded inconsistently building to building in the NHL. What might count as a hit in Ottawa or Toronto certainly won't be recorded as a hit in New Jersey or Edmonton. The Canucks, by way of example, are 16th in the league in "hits" on the road but 29th at home. Almost every team in the NHL has a similar discrepancy. Does every team simply change its style from home to road, or is it more likely different buildings have different definitions?
I'd suggest the former.
BLOCKED SHOTS, AND SUCH
Blocked shots are another point of contention within the statistical world. A blocked shot is, ultimately, a good play because you're preventing a scoring chance, but you never want to be in a situation where all you're doing is blocking shots.
Put it this way:
Blocked Shots per 60 Minutes
Even Strength 13.9
Penalty Kill 24.6
Those are statistics from the 2011-2012 season, via ExtraSkater, encompassing all 30 teams in the National Hockey League. Teams blocked far fewer shots on the powerplay than they did at even strength, and far fewer at even strength than they did on the penalty kill.
Generally, blocking shots can be a good thing, but you'd rather not have to rack up a lot of totals. There are different ways that analysts have attempted to sort through the best shot blockers in the game without simply looking at the players that block the most shots. Typically, those players (New York's Andrew MacDonald or Calgary's Chris Butler) are simply players that spend a lot of time in their own end.
In a previous Charts & Graphs feature I noted the difference between two advanced stat terms you may have heard of, "Corsi" and "Fenwick". They are both puck possession measures, essentially measuring the same thing. Corsi counts up goals, shots, missed shots and blocked shots for both teams and looks at one team's share of the total. It is a very good measure of which team has the puck.
Fenwick ignores blocked shots entirely, both for and against. The general principle is that "Corsi lines up better with puck possession and Fenwick lines up better with scoring chances" (which would make sense—since a lot of blocked shots are perimeter efforts that probably wouldn't go in even if left unblocked). It depends, to your taste, which is more important: possession or scoring chances?
By looking at a comparison of the Corsi and Fenwick numbers for the Canucks and the Rangers over the last three seasons, we can get some indication of what John Tortorella prefers against his predecessor:
I find this table super interesting, but I wasn't able to find any evidence that there's any reason to prefer Fenwick to Corsi, or prefer Corsi to Fenwick. Both measure pretty similar things and over a small sample, most of the little differences are probably more due to noise than strategy. But Tortorella's shot blocking schemes gained him a lot of notoriety in New York.
Tortorella's shot blocking in 2012 was a bit more noticeable because the Rangers had a low Corsi % and thus spent a lot of time in their own end. The mistakes were corrected in the shortened season and New York spent far more time in the other team's end, judged by their increased Corsi%. Still, Tortorella's team blocked the most shots in the NHL and generated a very high Fenwick % for his squad, the 6th best in the National Hockey League.
His predecessor in Vancouver, Alain Vigneault, now with the Rangers, preferred to play a perimeter-style game that focused on point shots five-on-five and attempting to create rebounds and lucky bounces. No style is one way better or worse than the other, but the fact the Canucks went from one of the lower differences between Fenwick and Corsi to one of the highest just shows the difference a coach can make.
Two Canucks players have a negative Corsi % but a positive Fenwick %, meaning they're blocking enough shots to turn themselves from negative possession players into positive possession players: those are Chris Tanev (49.9% Corsi and 51.1% Fenwick) and Zack Kassian (49.0% Corsi and 50.5% Fenwick).
WHO HAS THE PUCK?
Possession means everything in hockey. Banking on a goaltender or a star forward isn't the model to sustainable long term success in the NHL, so it's best to look the stats that will show whether a player has the puck.
From a theoretical perspective:
Getting back to Dallas Eakins' point, I don't disagree with him about what makes a game a perfect game. From the Canucks perspective, in Tuesday's game they had 56 shot attempts to Edmonton's 53. They absorbed 28 hits to 13 for Edmonton. Penalties were even. The Canucks gave the puck less, but also had fewer takeaways.
And… Edmonton had 14 blocked shots to Vancouver's 13. Not surprisingly, most of the Oilers shots came in the third period as the team was pressing. These graphs show the total number of unblocked shots for a team during a game:
Vancouver was ahead for most of the game, and it was only late that the Oilers started trying to take more risks and tie the score.
The real problem for the Oilers wasn't the physicality. It was that they had less of the puck against Vancouver, and that's been a common problem for them over the last several seasons. The lack of physical play may be a component, but from listening to post-game interviews, you get the sense that the Edmonton media is a little more fired up over Kassian than the Oilers themselves are.
I wrote two weeks ago in this space that Kassian is a player who looks like he could be dominant with the puck. He may just need to focus on getting it a little more. Few guys his size are as skilled as he is, and if he can supplement that physical intimidation with a little more puck-possession ability, he becomes a very valuable commodity in the NHL.