Let's begin with the obvious. The Dallas Stars penalty kill stinks. They kill off penalties at a rate of 79 percent, which is good for 24th in the league. And the worst part is, the penalty kill has been trending down for awhile.
Special teams have, I think, a very Dostoyevskyian link to a team's identity. They hold up a mirror towards specific elements within that club. And they constitute close to 20 percent of any one game. Hence why people like Arik Parnass are doing their best to understand just how beneficial/detrimental that 20 percent is to any particular team.
So just how bad is the defensive half of that 20 percent for the Dallas Stars? First let's talk about what Dallas' strategy is.
The Pledge to Wedge
There are a number of different strategies available to a team on the PK. You've got the diamond formation, with its "Blueline? What blueline?" method of collapsing in and around the slot. You've got the box, with its "limit shots to the perimeter" philosophy. And then you have what you see here, called the wedge: two defensemen protecting the low slot (in this case Johnny Oduya and Jason Demers), one forward in the high slot, with the extra forward trying to 'eat chicken past midnight' (Patrick Sharp), if you will.
Fundamentally, the wedge is a good strategy for a team like Dallas that loves puck pressure to begin with. With their speed, and abundance of shooters, Lindy Ruff and James Patrick aren't wrong about philosophy. In addition, it's a strategy that takes advantage of the four-forward movement among power play strategies. To note:
When we look at the defensive numbers, it becomes quite clear that there are risks to the 4 forward approach. While the increase in shot attempts against is modest, the decrease in save percentage is actually fairly substantial. Goaltenders playing behind 4 forwards on the powerplay stop shots on goal at about the same rate as goaltenders killing a penalty, which is obviously not a good thing. Teams that play with 4 forwards are definitely exposing themselves to higher percentage opportunities defensively, whether they be odd-man chances or offensive rushes against a winger not used to playing the point, and these plays result in more shorthanded goals than we'd otherwise expect a team to see.
That's Matt Cane over at Hockey-Graphs responding to how a goaltender's save percentage dips from 92.5 percent in a 3 Forward/ 2 Defensemen setup to 89.5 percent in a four forward/one defensemen setup. Why the sacrifice? Because the goal differential per 60 is 6.1 to 4.9 in favor of the four-forward setup.
How this happens is because the four forward power play strategy can take its form in the Adam Oates' designed 1-3-1 formation (which Anaheim is using in the above pic); one of the more common PP formations because it maximizes a team's shooting talent. Only drawback is that it's a bit dangerous if the team on the PK can counterattack. There's no blueline basically, which helps explain why Dallas themselves have taken full advantage of this scheme, as they'rthird in the league in shorthanded goals with seven (only Ottawa and Montreal lead them).
The PK according to Boudreau
So if Ruff and Patrick are right on the philosophy, what's the problem? I suspect not every player can speak Socrates. First let's look at players who can. Right now the best PK unit in the league belongs to the Bruce Boudreau led Anaheim Ducks. Here you see their box formation trying to defend against whatever Dallas' macaroni and cheese science fair power play is supposed to be.
Jamie Benn feints the pass to Jason Spezza, who you can't see in the first pic but will in the second. At this point, all Anaheim wants to do is cut passing lanes and take away high danger scoring area chances. The difference between Anaheim and Dallas is movement. Anaheim's players constantly have their head on a swivel, and they quickly switch formations while maintaining just enough space between one another, but not too much. At this point Benn doesn't have a great pass, but as soon as he looks for that lane...
[Madden]Boom!/Madden] Anaheim shifts into the diamond formation, which is different from 'the box' because it can switch into a counterattack strategy with the forward posted higher (and another just high enough for support). Sure enough, Andrew Cogliano (the lone forward protecting Demers) reads this, and heads straight to the point man.
Unfortunately there's more. Because Cogliano so quickly closes the distance, he ends up in a puck battle. Ryan Kesler (the forward on the dot) being the premier PK'er that he is, sees the play develop, and quickly supports Cogliano. Demers can't beat Cogliano and Kesler at the same time because he's not Thanos, and yes, you guessed it, this leads to a shorthanded shot on goal for Anaheim. Spatial awareness, and adjustment; the two keys to a good penalty kill.
Dallas' PK Problems from A to Z
Do the Stars have those keys? Not quite. Here's Cody Eakin, Jamie Benn, Alex Goligoski, and Jordie Benn all getting wedgied.
There are a ton of things wrong here, so let us count the ways. First and foremost, Cody Eakin is not "eating chicken past midnight". Instead he's the tail between the legs. The purpose of the extra forward here is to create pressure. Not to back up into the mangled triangle behind you. No spatial awareness? Check.
Eakin's not the only problem here. Not by a long shot. Goligoski and Benn end up far too close to each other, leaving Jordie Benn outmanned. But Jordie's job here, as it is for any weakside player (which is to say the player on the opposite side of the puck), is to protect against the backdoor play. Instead he accelerates away from the crease, functionally putting Antti Niemi in a 2-0 situation. No adjustment? Check.
Any good PK needs to work like one of those sticky stretch hands your fourth grade math teacher took away from you. Elastic, but retractable. Here's Dallas getting the elastic part right.
This is what the wedge would look if you gave it beer goggles. Mattias Janmark (the +1 in the "triangle") is pressuring the point, doing his job. Recall that Cody Eakin is 'the other guy' to see that this is what's supposed to happen. However, Antoine Roussel has backed into Goligoski, and is basically just screening his own man. He's no longer the high forward, which is a problem because...well, just look.
LA has enough space and time to assemble their own Ponzi scheme here. Believe it or not, Alec Martinez (point man) passes to Jamie McBain (left wall) who the dishes to Vinny Lecavalier (right side half wall). Vinny gets a shot off. Kari Lehtonen stops the puck, and then LA resets in the exact same formation with the exact same space (!) from the exact same pass (McBain to Vinny) to score on the second shot, by the exact same player (obligatory Scanners reference).
The rebound is especially problematic because there's zero rotation on the PK from Dallas. If you go to the 0:49 second mark on the PP when LA scores, you wouldn't notice a single difference in positioning. Yea it'd be nice to see Kari Lehtonen trust that he's 6-foot-4 and doesn't need a trampoline to get to the Hamburglar playground slide, but that's the least of Dallas' worries here.
In addition to the problems listed, Dallas' faceoff percentage in 2016 on the PK is 39 percent. While a lot of work has been done on the importance of faceoffs, and why they're overstated, I can't imagine that this helps them in any way. Especially in a vacuum where possession of the puck can quickly turn into shot quality given the odd man out.
Part of this is on personnel. Jason Demers, and Johnny Oduya log the most PK minutes on the backend. Vernon Fiddler and Eakin log the most minutes among forwards. Goligoski, and Benn are right there with them as rotating PK stalwarts. Between them, out of Dallas 22 skaters, only one of those listed (Goligoski) breaks the top 10 in shot attempt differential at 5 on 5, which is a good predictor of PK success (Jamie Benn is second on the team in CF Percentage at even strength, and you guessed it, is by far Dallas' best shorthanded man).
Erin Bolen has broken down a lot of the numbers already. With the PK in such disarray, you have to wonder why they don't try different four-man units. Size is likely an issue here, as the game's best shorthanded forwards are guys like David Backes, Kesler, and Sean Couturier. These are players with reach, and just enough speed to go along with their defensive acumen. Unfortunately Dallas doesn't haven't much of that.
Radek Faksa projects to be a good PK forward for this reason. Some of the more hot-blooded fans might lament Ruff for this, but Ruff actually has Faksa 11th among Dallas' 19 skaters who have registered TOI on the PK (despite only 18 games), and even had him killing a 5-on-3 situation once.
If the philosophy isn't working, at what point do you find players who speak the Socratic method? Or do you change your philosophy altogether? I do believe that the PK speaks to the flawed aspects of Dallas' defense, and to their bottom six. While not an indictment by any means, a lot of things will need to change if Dallas wants to avoid having the PK cripple them when it matters most. Lucky for Dallas, they have time and homegrown answers to address this.