Reviewing the Flames’ penalty kill, eight games in
By Ari Yanover
6 years ago[Editor’s note: Remember Mike Fail? He’s back! In pog form. Or, more accurately, to share in depth knowledge about his favourite topic: the Flames’ penalty kill. The contents below are all his, via guest post. Enjoy!]
It almost feels like there are two constants in life this season for the Calgary Flames. The first: the Flames are going to take a penalty. The second being the massive uproar of anger that yet again, the Flames will be shorthanded. Like the Lion King has taught us, it is the circle of life.
Recently I looked into the game versus the Ottawa Senators which drew attention to a handful of things, but three stood out as extremely important points:
- Zone entries against, specifically controlled/uncontested entries against which lead to shot attempts or ZEFR events.
- Formation issues that extend beyond 4v5, well into 5v5, and that are likely contributing to the results so far.
- A reluctance to apply pressure and look to suppress more aggressively.
There isn’t anything to scoff about when you’re in the top five in the league in terms of PK% (86.8%) or being tied for eighth with only five PPG against. Though I would be remiss to not mention that overall PK% – which blends 4v5 and 3v5 results – doesn’t tell the whole story. On the surface you’re almost certain to believe things are going well, but like many things in life, there are opportunities.
Digging in a little further we see the Flames’ CA60 while 4v5 sits 14th at 101.25. Their FA60 (79.2) puts them 20th while their SA60 (54.14) has them 15th in the league. All this in just under 60 minutes of 4v5 TOI this season.
So goals aren’t occurring which is great, right? Well, let’s dig a little deeper using my systems tracking through the season so far to see what is going on at a more granular level. You can view all data here and read about my project work on my personal site.
Neutral zone formation and forechecking
The first thing we’ll examine is the formation and forechecking schema through eight games. The big thing to keep in mind with all systems tracking is:
- Hockey is chaos and often hinges on the cusp of pure entropy.
- Most things are fluid in hockey and ever changing. When we get to video you’ll see what I mean. In short: hockey is a fluid.
- Forecheck formation tracking can be a pain if a forward gets too aggressive and F2 has to move out of formation to compensate. All tracking was done with keeping the most clear representation of what the formation should be.
- The ZEFR against data points to using Arik Parnass’ ZEFR statistic to measure zone entry to formation or dangerous rush attempts. The best example of this is an opponent’s power play attacking on the rush or they enter the zone on the rush, get set up, and shoot towards the net.
| Forecheck type | Times used | Break up percentage | Dump-in percentage | Carry-in percentage | ZEFR against percentage |
| Passive 1-3 | 57 | 15.79 | 12.28 | 71.93 | 42.11 |
| Retreating box | 29 | 13.79 | 13.79 | 72.41 | 37.93 |
| Tandem pressure | 3 | 33.00 | 0.00 | 33.00 | 66.00 |
| Same-side press | 4 | 25.00 | 50.00 | 25.00 | 25.00 |
The Flames so far have relied heavily on two common formations: the passive 1-3 and the retreating box. The most glaring and concerning number in raw counts and in the percentages is just how often zone entries against are transpiring.
Finding ways to stifle the breakout will not only be beneficial in time management, but it will afford you the luxury of limiting your in-zone time as much as possible. For the Flames, in a league where teams are getting faster and more skilled – and that look to maximize their 5v4 outputs – this will be a critical pillar in their 4v5 success this season.
Zone entries against
When looking at zone entry data I tracked all zone entries against, including some that happened when the team was in no clear formation or structure. I tracked faceoffs in the defensive zone as an entry against for future work on deployments on units and how the dichotomy of starts versus on-the-fly usage is extremely important as to penalty kill units.
| Entry type | Total count | Percentages against (C/D/B) | Percentages against (all) |
| Controlled entry | 79 | 73.83% | 50.00% |
| Dump-in against | 13 | 12.15% | 8.23% |
| Entry broken up | 15 | 14.02% | 9.50% |
| Defensive zone faceoff | 51 | null | 32.29% |
It goes without saying the Flames have struggled with controlled entries against. Part of it comes from being on the PK so often; the other aspect comes from neutral zone breakdowns like line changes, exploits of over-aggressiveness, passiveness or misreads on the opposing team’s breakout; or a lack of challenging the zone entry:
Failed to load video.
If we examine a game by game breakdown of entries against it becomes even more apparent:
| Game | Controlled entries against | Dump-ins against | Broken up |
| vs Edmonton | 5 | 0 | 3 |
| vs Winnipeg | 7 | 1 | 1 |
| vs Anaheim | 8 | 1 | 0 |
| vs Los Angeles | 8 | 2 | 2 |
| vs Ottawa | 13 | 2 | 2 |
| vs Vancouver | 18 | 3 | 3 |
| vs Carolina | 16 | 1 | 3 |
| vs Minnesota | 4 | 3 | 1 |
With the struggles of limiting controlled entries against there have been a handful of entries prevented that are worth drawing attention to. You can start to see the value of finding opportunities here:
Failed to load video.
Defensive zone formation
| Defensive zone strategy | Times used | Usage percentage |
| Passive triangle +1 | 47 | 65.28% |
| Pressure-based | 25 | 34.72% |
For the sake of simplicity I’ve blended the pressure-based strategies (Czech Press – 24 times and Low-High Press – once). Like Charlie O’Connor’s work on the Philadelphia Flyers’ penalty kill, I tracked when the opponent’s power play got into formation and the Flames’ penalty kill formation deployed to combat it.
So, what have we found out so far? The Flames are playing with a more passive pressed-based structure. It could be a contain/manage approach being utilized until an opportunity arises – which has been observed – or it could be a mandate to play safe, which is unfortunate since it often feels like safe is death in this league.
Failed to load video.
The other theory that is purely reasonable is a conservation of energy approach to avoid fatigue. Given the propensity of penalties against so far playing an aggressive approach while shorthanded might give way to errors – predominately forced – which could give way to more high-danger chances against or goals.
Now we can add another layer of data to this, from tracking with shot data per each tracked instance of their formation:
| Defensive zone strategy | Times used | Shots on goal per formation | Unblocked shots per formation (shots on goal + missed shots) | Total shot attempts per formation |
| Passive triangle +1 | 45 | 0.62 | 1.13 | 1.13 |
| Pressure-based | 27 | 0.30 | 0.85 | 0.85 |
During my tracking and checking the play-by-play data, I did find some discrepancies, so I ended up tracking shot data manually while in formation. In very, very, very limited usage the Flames did better when applying pressure. Part of the overarching list of problems impacting the team revolves around zone entries against and a passive pressure approach to defending in their own end.
If you’re attacking the power play formation with pressure – using angles, active stick in lanes, taking away time and space, breaking the PP formation, and forcing them to regroup/cycle the puck – in theory you’re going to limit shots or shot attempts against.
To contrast that, there’s always the likelihood you get too aggressive and open up seams for the opponent to capitalize on a high quality chance, or even a goal. Right now, though, it makes much more sense for the Flames to play more aggressively.
Has there been anything good?
In eight games, the Flames have generated 15 controlled zone entries for, which led to a Mark Giordano shorthanded goal. In those controlled entries for, most of the shots directed toward the net have been extremely low quality, with only a handful being high danger.
This is a good indication that they are capable of counter-punching, and should. It’s just a matter of time before they score another SHG.
The same relative sense of good can be said about some sequences which have resulted in better play both in the neutral and defensive zones. I’ve compiled a sped up brief snapshot of some sequences worth highlighting:
Failed to load video.
Recommendations and summing things up
The first recommendation to be made should be the Flames considering bringing in Paul Devorski, like the Winnipeg Jets did during training camp, as this might help with the penalty woes. You can’t play all season shorthanded, again.
Secondly, when it comes to their neutral zone forecheck, it might be time to take it up a notch. Use a formation like tandem pressure to eat more time off the clock while applying direct pressure to the opposing team’s breakout. Work done by Charlie O’Connor’s Flyers PK project eluded to the value of this configuration.
From there we can echo similar sentiments about defending the blueline to contest more entries against. Avoiding circumstances where seams open up due to gap control between a sagging defensive pairing and the forward tandem requires rectification. Again, using tracking data we can see the impact of this with ZEFR statistics.
Giveaways while down a man can be costly and the Flames have had their share early on. In many cases, giveaways have led to additional high event sequences:
Failed to load video.
Finally, when it comes to play in their own end, when set up the team should employ more pressure-based tactics. This might help alleviate the workload that Mike Smith has seen – and to his credit has been pretty good with a .923 SV% – while affording them opportunities up ice or killing time off the clock.
Recent articles from Ari Yanover
