Black Box: Special Teams



Every Monday we’ll be producing a statistical summary of the Flames, starting with an OZQoC chart to describe each player’s role and put their stats into context, followed by their even-strength data, and finally the piece being described today – their performance on the special teams.

The purpose of the weekly Black Box is to provide a running statistical snapshots of the Flames as the season progresses, in the clearest and least confusing way possible for everyone.  The tables will be accompanied by some interpretive analysis, but in most cases you can’t read too much into the data without spinning off into a separate and more detailed examination.  Nowhere is that more true than with special teams data, owing to its unusual and infrequent nature.  In fact, we’ll limit ourselves to 5-on-4 play only, to avoid skewing the numbers with even more unusual and infrequent situations.

Power Play

We’ll focus on the three most important power play statistics: who is playing, how much they’re scoring, and how well the team is doing when they’re on the ice.

There are many reasons why a player may get more power play ice-time than he deserves.  Perhaps better players have injuries, are being saved for even-strength or penalty killing duty, or simply have their skills overrated by the coaching staff.  That being said, the better players generally get more playing time.

Despite the importance of playing time, in the spirit of even-strength data we’ll rank players based on how well the team does when they’re on the ice.  To determine the team’s performance, we’ll remain consistent with the even-strength tables and use Corsi-based data – all attempted shots (goals, saves, missed nets and blocked shots). 

It’s possible a player’s CE/60 (Corsi Events per 60 minutes) is being skewed by playing with particularly good or bad linemates, or against particularly good or bad penalty-killing opponents, and there’s a long-standing debate about whether players consistently generate on-ice shots of different qualities, but over-all this has proven the most reliable way of determining whether a unit is effective at generating offense with the man advantage.

While it may be the most interesting part of the table, a player’s scoring level is actually the least important part, because it is so highly dependent on luck and bounces over the small sample sizes we’ll be studying.  Generally PTS/60 will go hand-in-hand with CE/60, but a player has far greater control in his ability to generate shots than the ability for them to actually go in.

We won’t split up the forwards and the defensemen, since the roles sometimes get blurry when playing with the man advantage, with key forwards occasionally working the point, and large defensemen occasionally playing in the Holmstrom position – directly in front of the opposing goalie. Hopefully not facing the goalie, though (that would be the Avery position).

Player              TOI/GP PTS/60 CE/60
Brendan Morrison     2.3    4.7    95.3
Tim Jackman          0.8    2.9    94.7
Olli Jokinen         3.0    5.2    93.4
Clay Wilson          2.7    1.5    92.6
Chris Butler         1.2    1.0    90.9
Rene Bourque         2.9    3.7    86.4
Mark Giordano        3.4    4.1    85.6
Jarome Iginla        3.8    4.6    84.9
Jay Bouwmeester      2.7    2.1    84.6
Niklas Hagman        1.7    4.0    84.6
David Moss           1.5    4.2    83.9
Anton Babchuk        2.5    4.2    83.8
Alex Tanguay         3.5    3.3    83.8
Matt Stajan          1.1    0.8    78.1
Curtis Glencross     1.2    3.3    76.3
Lee Stempniak        1.8    2.4    74.1
Mikael Backlund      0.6    5.4    65.9

Those at the top of the list are those who are most effective at generating offense with the man advantage, like Brendan Morrison and the much-maligned Olli Jokinen, whereas those at the bottom are the ones that normally cause the crowd to yell SHOOOOOT.

Always keep an eye on playing time so you can see how the numbers might be getting skewed by a few lucky bounces over a small sample, and when you notice a big difference between a player’s scoring and the team’s Corsi rate, you can generally expect their fortune to change.

Penalty Killing

While it’s nice to have pressure on an opponent’s power play squad, it’s not as important to keep track of scoring when studying short-handed data. In the on-going effort to promote clarity and reduce confusion by presenting as little data as possible to paint the picture, we’ll include just the time on the ice and how well the team does at preventing shots with each player on the ice.

This time we’ll rank it by playing time, because those getting more playing time are typically also the ones getting the toughest assignments, which generally inflates the number of attempted shots the opponents are generating. 

There can be a big gap between each team’s top power play lines and the secondary units, making it somewhat unfair to compare the likes of Jay Bouwmeester and Scott Hannan to Mark Giordano and Cory Sarich (though you do have to admire how great their numbers are).

Bear in mind that the players at the bottom are usually ones that come in at the end of a penalty when all the real penalty killers are exhausted.

Player           TOI/GP CE/60
Jay Bouwmeester   3.2   101.1
Scott Hannan      2.6    90.4
Curtis Glencross  2.3    85.0
Mark Giordano     2.1    71.6
Brendan Morrison  1.9    86.7
Cory Sarich       1.8    66.0
Tom Kostopoulos   1.7   100.3
David  Moss       1.6    87.4
Alex Tanguay      1.2   110.9
Brendan Mikkelson 1.2   103.8
Rene Bourque      1.1    80.4
Chris Butler      1.0   111.3
Mikael Backlund   0.9    77.5
Matt Stajan       0.7    92.1
Brett Carson      0.6    93.6
Steve Staios      0.6    80.2
Stefan Meyer      0.5    93.0
Anton Babchuk     0.5    76.8

When you’re watching a game, remember that the best penalty killers can keep the opponents to three attempted shots or less – so count along.


Nothing is written in stone, either now or as the season progresses.  If you have suggestions on how to improve the clarity of these efforts, either by using different statistics or presenting them in a different format, please be generous with them.  On the other hand, if you have suggestions on my personal grooming, feel free to be a little more stingy with them.

Next week we’ll take a quick look at how the goaltenders will be covered, and then we’ll get started!

  • ChinookArchYYC

    Robert, doesn’t TOI track the trust level a coach has in his players on the PK, more than his value on the PK? While Bouwmeester is likely still our best PK player, I think you should put more emphasis on his ability to prevent CE in the first place. My thinking is that the best PK, is the one with no shots on net.

    Clay Wilson’s numbers on the PP are intriguing, he received some good minutes on the PP and managed a decent level of CE/60. Does he have to clear waivers for the Flames to keep him?

    • Penalty-killing ice-time is a reflection of a coach’s confidence in a player’s skill, relative to two things:
      – His other choices
      – That player’s other skills (he may be saved for other duties, no matter how good he is at killing penalties)

      Of course, a coach’s confidence in a player is often also a reflection of a player’s actual abilities.

      If you have reason to believe that the coaching staff’s appraisal of Bouwmeester is woefully inaccurate then there ought to be ample statistical evidence.

      As it is, there’s a big difference between most team’s top power play unit and their secondary squad, so Bouwmeester is playing far, far tougher minutes than Giordano.

      I’m not saying Bouwmeester is better than Giordano at killing penalties, and I’m not saying that you can’t compare the two players. I’m saying that it’s important that you keep the difficulty of the role in mind when comparing two players, which is why I recommend ranking the list by playing time.

      If there’s a strong push to rank by effectiveness at preventing shots, we can do that. This is for all of you, after all.

      As for Clay Wilson, remember that’s an awfully small sample size but yes, the results are intriguing.

  • Giordano’s short-handed rates are really good. Butler’s are flat out awful (although maybe the whole Sabres team was bad at giving up shots). Just looking here I’d suggest keeping Butler off the PK as much as possible though. Ditto Tanguay.