Travis Yost has been involved in the world of hockey analytics for a decade and is currently part of TSN's Hockey Analytics team. Prior to joining TSN, Travis was a contributor at the Ottawa Citizen, the Sporting News and NHL Numbers, and has been a consultant for an NHL franchise. He will be contributing breakdowns on the Sabres for The Buffalo News this season. Follow Travis on Twitter: @travisyost.
The best penalty kill is the one that never takes the ice. But in the absence of incredible team discipline, having reliable penalty-killing units is essential for team success. And this season the Buffalo Sabres are seeing positive results.
Heading into the Christmas break, the Sabres have one of the more successful penalty-kill units in the league. Entering Monday's games, their 6.2 goals against per-60 minutes ranked sixth in the National Hockey League (their straight conversion rate of 83 percent was fifth). It’s a bit of found money for a Sabres team that hasn’t had results on the penalty kill in the last couple of years; in 2016-17, they finished at 7.5 goals against per-60 minutes (23rd in NHL), and in 2017-18, they finished at 7.6 goals against per-60 minutes (20th). Their current performance would make them about 10 goals better in the standings on their penalty kill alone over an 82-game season, which is surely material.
When team success starts to shift like this, I try to answer two questions. First, how real is the performance? Penalty-kill performance in smaller samples can be highly volatile and we need to know if the underlying numbers are indicative of quality play. Second, what is a team doing from a personnel standpoint, and how does the coaching staff think about deployment?
The first question is rather straightforward. We know that the Sabres are conceding very few goals on the penalty kill relative to where the rest of the league is right now. But how do they compare on measures that are much more telling in terms of volume, and generally correlated with long-term penalty-kill success? (Data via NST)
The good news is that Buffalo isn’t skating by on save percentage alone. The shot volume they have faced – whether you want to look at the entire sample, or only part of the sample that’s truly collinear with scoring chance generation – is around league average. That's not indicative of some oncoming massive regression, though it’s fair to say that Sabres goaltenders have certainly pulled their own when down a man. Linus Ullmark’s 91.8 save percentage and Carter Hutton’s 87.6 save percentage (over many, many more shots) were fifth and 22nd in the NHL, respectively. That’s going to get the job done.
So we know that Buffalo’s penalty kill has improved year over year. We might not be sure of the degree of improvement just yet, but it’s fair to say that all of the performance measures here are looking directionally better. The next question that comes into play: How does Phil Housley think about his penalty kill units, and what is he most inclined to do when a Sabres skater heads to the box?
The usage table below does a really strong job of capturing coaching deployment behavior. Here we have all 13 Sabres skaters who have been used (to varying degrees) on the penalty kill this season, and the minutes they have spent playing with one another as a percentage of total Sabres penalty kill ice-time. So when you see Zemgus Girgensons and Johan Larsson play 14.4 percent of the time together, that’s 14.4 percent of all available Sabres penalty kill minutes. (That’s a lot.)
Let’s take a look at all of these relationships:
I mentioned the Girgensons/Larsson duo – they have established themselves as the most frequent two forward combination for the Sabres on the kill this season. To a lesser extent, we also see Vladimir Sobotka and Evan Rodrigues spending some time together, as well as Sobokta paired with Zemgus Girgensons.
The forward group is particularly interesting because of how many minutes the team had to replace from last year’s roster. Buffalo’s second-, third-, fifth-, and sixth-most utilized penalty killing forwards – Ryan O’Reilly, Benoit Pouliot, Scott Wilson and Jacob Josefson – haven’t been available to the coaching staff. That’s hundreds of minutes the Sabres had to reallocate around their lineup, and sometimes that can open up quite the can of worms. To Buffalo’s credit, they’ve actually improved performance on this front.
Defensively, the story is much more straightforward. Rasmus Ristolainen has taken on the team’s biggest role on the penalty kill – a role probably more suited for him at this point in his career. He tends to play with Marco Scandella, though we also have seen some of those minutes allocated to Zach Bogosian. In front of him regularly are three names – Girgensons, Larsson and Sobotka, in that order. Jake McCabe, just like he was last year, tends to rotate in as a penalty killing option on their second unit.
That’s the recipe Buffalo has been cooking with this season, and so far, it appears to be working. The team is clearly getting some goaltending help, but not enough to wash away improvements made at the skater level. And subtly, it’s another area where Buffalo appears to be improved from where they were from just a year ago.