Systems Analysis: Different Looks at the Swarm

Game one between Winnipeg and Edmonton offers us an edu-taining systems battle. Without the puck, both teams use what is broadly called a ‘swarm’ system, but each team employs it very differently. Inside we’ll talk about what a swarm is and why it’s used, and then examine some practice video and still images to see if we can spot the differences in their systems play.

To begin, let’s be clear that we’re talking about a system without the puck. You’ll hear people call it a ‘swarm defence’ for that reason, but the defence/offence binary is less meaningful in hockey than in games with less flow to them. Personally, I think about the game through the lens of possession, and so a more useful binary for me is with/without the puck. Moreover, this system isn’t necessarily limited to the defensive zone. No, in fact, the Oilers use it all over the ice as part of what new coach Dallas Eakins is calling "a quick and immediate plan to get [the puck] back."

We have a lot to cover in this post, so we’ll focus on the defensive zone and come back to other areas of the ice in the future.

A swarm is an aggressive variation on the overload. Where the overload is designed to force players into a single, less desirable option, the swarm is designed to take away even that choice. Typically, a player attacks the puck carrier, while one or both vertical (or board) lanes are closed and a 3rd or 4th player (we’ll see how aggressive it can get below) plays inside the puck carrier’s nearest puck support. Did I just hear you say blerg? Here’s a picture!

 

This picture is meant to show the principle of the swarm more than any nuance that the Oilers put on it. That will come later. For now, notice that the Oilers defenders play according to where the puck is (all 12 eyes are on that corner), and every player has positioned himself ‘inside’ the Rangers puck supporters (that is, between the puck and the bad guys). In this case, Arcobello (centre, 62) is going to support below the goal line to retrieve the puck that the Ranger is trying to reverse behind the net. In some systems, that would be Belov’s (defence, 77) job. 

Just to give a contrast, here is a picture of the Rangers defending a similar play in the same period:

The Rangers play a very conservative man-to-man collapsing system. We won’t get into detail about the Rangers and our analysis of this photo will focus on better understanding the swarm.

The differences will at first seem like one of those picture puzzles on a cereal box, but they’re important even if small.  The defenders play between their check and the net here, rather than between their check and puck as in the swarm above. The ‘collapse’ aspect actually means that the defenders work inside-out. So if Vigneault coached the Oilers, he would give hell to both Oiler wingers in the 1st picture for playing too close to the perimeter of the play. See how the Oilers’ defender up the boards stands between the puck and the open player. The Rangers’ defender stands between the net and the attacker. Staal (18) stands, again, between the net and his check, where as Perron (57) is between the puck and Girardi (5), leaving the lane to the net uncontested.

In this example, Moore (centre, 28) looks to be lost, but he’s actually reading the Oilers’ defenceman before circling to cover Eberle (14) as he floats down the seam. In fact, it’s Dorsett’s (15) check (Belov, 77) that Moore is looking at (out of screen at the right point). Belov is above his defence partner Schultz (19) who is on the sideboards supporting the puck. In other words, the Oilers have committed 5 guys to support this puck and are showing a very dense offensive grouping. The Oilers are exploiting a tension in playing man and collapse to create a man-advantage on the puck. The coach has to choose whether Dorsett (or Moore, or someone) goes over there to stay close to a guy, or whether they simply hold formation inside the attacker’s perimeter. You see Vigneault’s choice above and why it’s not an overload system. 

For some reason, systems are often talked about as simply extensions of coaching personalities, or worse, as some eternal code of ‘this then that.’ Torts is mean so he makes people hit. If a team plays ‘through opponents’ then they win. The reality is that systems represent a theoretical approach to hockey grounded in ideas of which structures, areas, and plays are most critical in reaching the same stated goal.

For Eakins, we can hear him saying that he sees the swarm as part of a possession system. To take a step further back, Eakins believes the game is about the puck. Control where the puck goes, and you control the game. That might sound tautological until you consider the theory behind other systems. The collapse system, for example, is focused on opportunity through control over one particular area of the ice. Limit the number of opportunities at or near your net, and you might win the scoring chances battle.

Still, possession is not the only reason a team might employ a swarm. We know that Claude Noel evaluates players by scoring chances, and as we’ll discuss below, when a swarm breaks down it gives up 5-alarm chances. In some cases, it’s a theory of offence that brings a coach to the swarm system. The Jets under Noel believe in a vertical, transition attack, and it starts with having puck movers on the back-end. 

Non-swarm systems often rely on big, strong defenders to box out the front of the net, win one-on-one battles in the corner, and break up cycles by taking up a big patch of ice. Both the Jets and the Oilers employ more mobile defenders than trench-dwellers, and aside from Smid after a whistle, neither group has much malice or might.

The swarm can be used to support defenders in battles by avoiding one-on-one situations. Both teams want a fast transition game, and are using a system that gives guys like Tobias Enstrom and Justin Schultz a fighting chance against NHL sized forwards. 

For the Jets, Adam Pardy and Mark Stuart stand out as very poor puck movers in this system, but might look a little more steady in a system where they just chip it up the boards after some face washing in the corners. (Might.) The Oilers have a long history of Theo Peckams and Mark Fistrics, but it’s their first year using the swarm to my knowledge, and I don’t know much about the Marlies. It’s possible Eakins is adapting his system to his personnel, or that he convinced MacTavish of its promise in the interview that landed him a job before there was an opening. Maybe some Oiler people can help me out in the comments.

So far all we’ve done is differentiate the swarm from a man-collapse system. Next we’ll look at the Oilers and Jets for two examples of different types of swarms. In particular, we’ll see different roles for the centres and the Oilers using a double-pin in the corners to win puck battles and break the cycle. The result is also a different break out, and we’ll touch on that very briefly. 

The Jets system in pre-season is more conservative than the one we saw used by the team last year, but it still qualifies as a swarm. On the Bruins first goal we saw Olli Jokinen play a really soft overload zone. It’s a terrible play; it serves us well as a learning example.

The Jets’ set here is a simple overload. It gives us a chance to see the difference. You see three Jets lined up like a picket fence between the puck carrier and his most dangerous passing options. Trouba (3) has inside position on his man in front and should win any race to the end-boards should Marchand (63) put it low, and the Bruin’s best option is to go to the point. The Jets have chosen for him. Jokinen is the active player in this space for the Jets swarm. It’s his job to force Marchand’s hand here, and we’ll see later an example from the second period where Cormier does that.

But not right now. What Marchand does with those four seconds between pictures is skate up the boards and then circle back down. Krug (47) goes all the way to the net through the same lane Bergeron (37) is in now, and then circles out to his position at the bottom of the image. Instead of pressuring the play on the way up the boards, Jokinen waits and then makes a soft approach on Marchand’s way down the boards when he’s on his forehand, and he waits long enough that there is a gap behind him. Bergeron fills that gap because he’s smarter than Jokinen. Pavelec somehow reads the play going to Krug (47) and slides away from the shot and gets beat short side. Good gravy.

That’s not a defensive structure we saw last year in Winnipeg, and it’s something to keep an eye on through the season.

It doesn’t answer how the Jets use the swarm. With the Swarm, we start in the corner.

 

One of the struggles with seeing the Jets’ systems is that Dustin Byfuglien (33) is an absolute wild card. Here he’s standing below the goal line but between his check and the net. His partner, Clitsome (24) is attempting a single pin in the corner, and the centre (28, Cormier) is attacking the puck. As we’ll see below, this system can get a lot more aggressive, but for now, the Jets have four players more or less committed to this corner – roughly in an ‘L’ shape from Peluso (14) to Big Buff (33). 

It might not surprise you that the Jets were dominated on this play and in this period. The Bruins have two players with inside positioning, and the Jets are already chasing with the puck still in the corner. A better centre than Cormier might read that he should pin 32 white (or chase him away from the play) instead of making this leaning hack at the puck. 

Regardless, the team adjusts in the intermission, and we see the system tweaked toward being more aggressive. Below is the same line in another attempt to prevent the Bruin cycle, this time on the half-wall (like the Jokinen play above). The Jets have two players on the puck, and play man defence outside that scrum. That’s much closer to what we saw last year. Again, Dustin Byfuglien is beginning a pivot to go… god knows where. Clitsome might cheat inside that Bruin forward, or Peluso might adjust his space toward the net except that Buff is in no-man’s land without a check or an obvious decision here.

Let’s look at another line. During this game, Mike Johnson of TSN pronounced that Tangradi-Slater-Thorburn had played well enough to cement themselves as the Jets fourth line. Let’s have a look at them.

In this play, Scheifele and Wright both went for the change on the back-check, so Thorburn (22) took over the centre job below the goal line. Slater (19) is playing RW in the switch. Here we see a very classic swarm formation, as we saw in the example above. Both boards (up to the blue line and down behind the net) are covered (or closed), there is pressure on the puck, and a 4th skater (Tangradi, 27) is between the puck and the next opposing player. Everyone is between the puck and their check, and focus is on the puck and its carrier.

A lot happened in that 20 seconds between slides, but here we are back in the same corner with Slater (19) playing his natural centre and Thorburn (22) back on the right wing. This picture is extremely important for our comparison to the Oiler system below. Postma is attacking from an angle to try to close off one board play, Thorburn is closing the other, Slater is at the goal line but not attacking the puck. Three players surround the puck, but the Jets’ weak-side defender stays in front of the net even without an immediate threat and Tangradi stays in the middle slot. The Jets have isolated Thornton (22) so that Postma can attempt to pin him, but until that pin is made, his team mates keep their distance.  

I’ve already hinted at how aggressive the swarm can get, and the Oilers are the best example I’ve seen of it. Let’s look at a couple still images that give us some insight into what Eakins is starting to implement, and re-visit a video from their training camp practices that was captured by Blacque Jacque and posted in the comments over at Lowetide’s blog a few weeks ago.

I’ve cued the video below to what I want to show, but if you’re a fan and haven’t seen it yet, there are all sorts of interesting tidbits to muse over. From where it starts (4:18), you’ll hear a whistle and the then see the defensive swarm with a double pin by two defencemen with the centre digging. I encourage you to pause again at 4:33. That’s all the video I want to talk about for now.

In that brief 15 seconds, we just saw a TON of what the Oilers are doing this year. As you’ve noticed by now, lots of teams use the swarm. I don’t know of another that uses both defenders in one corner, or allows both defenders pin on the cycle. As we see in the Jets swarm above, the centre supports below the goal line and is allowed to pin, but one defenceman stays near the net so a single-pin is used to limit options and the second play is generally a hit to separate the puck without sacrificing another man to the corner. Eakins is showing that he’s willing to put three people in one corner. 

To compensate, look how low the weak-side winger comes – all the way to the crease. And the strong side winger stays on the strong sideboards. That’s all five players below the hashmarks, committed to a single play. In the hockey world, this has a technical term: Big Brass Ones.

Okay, some game stills.

 

Look familiar? This actually came on a turn over by Larsen (36), who rounded the net and got caught by the Ranger F3 transitioning to F1. He got hit, smartly organized himself to make it a battle, and the Oilers re-grouped into this insanely tight formation.

And here we have the double-pin (look closely in the corner for two Oilers):

This is a good time to note that this system has its problems. Lowry (56) has come away with the puck, is about to beat Mike Brown (13) around the net, and there are three Jets available to fill the wide open spaces of the most dangerous part of the ice. (My favourite part is that Eager (55) is still looking at the corner. Could be a long year for that line.)

Here’s another issue Oilers fans are going to see a few times this year:

 

It’s a simple enough mistake. The Oiler defender lost containment on the cycle and the Ranger forward beat him to the puck. I’ve highlighted the puck so you can see what’s about to happen. (A goal. A goal is about to happen). When the system requires three or even four players be on the boards, small mistakes have very noticeable consequences. The Rangers’ skaters were seen using the reverse along the end-boards to try to beat the Oilers’ swarm from the puck drop, but it wasn’t until the 2nd period that the coach adjusted to have support for that reverse. The Rangers had 3 shots in the 1st and 24 the rest of the game (though still lost). They got the Oilers to start chasing back and forth, and if you’re late in this system, you get beaten off the boards. This is why Eakins wants that weak-side winger so incredibly low.

One final problem this system has is the chance for over-pursuit. The shot below is actually a mistake by Evander Kane (9) on the defensive side of the puck, missing a check and then just floating away from the play. But it leads to a two-zone break away for the Jets’ most dangerous skater. Such aggressive pursuit meant the Jets only had to slap the puck into the soft zone behind the Oiler attackers and watch Kane take off. It’s high-risk hockey.

One of the most interesting parts of Eakins swarm is that it comes with a unique breakout. On the breakout in the video above, we saw the Oilers reverse to a winger who made the first vertical pass to the other winger swinging across the zone. My first thought when I saw that video a few weeks ago was that the system would fail if that were the intended breakout play. The carrier had no other options except to eat it and start another battle. A reverse is back toward opponents, and everyone else is occupied. We also saw one of the defencemen jump up on the side closest to the camera (see that?). It would be two games before a team played a weak-side lock on them and scored 8 goals.

Thankfully for Oilers fans, that part of the drill was more about where the weak-side winger goes than what he does with the puck. I suspect we won’t see a lot of that single-option breakout from the Oilers this year. In pre-season, we saw the Oilers carry around the back of the net when they won that puck battle. It looked a little more like this:

Now, it looks a little different because the centre didn’t start low. But otherwise it’s the same. Joensuu (wing, 6) is swinging, Pitlick (weak-side wing, 68) swoops low through the corner, and options are developing for Larsen (36) with the puck.  

Still, I’m not sure what the Oilers plan is for a strong-side breakout, and if the Oilers have any success, we’ll see teams start to use F3 or the defender on that side to cut off that breakout very early (instead of chasing the puck around the net as Thorburn (22) does). The best breakout has multiple options, and so far we’re not seeing a lot of different looks coming out of that aggressive corner swarm.

Okay, it’s been a long one. Pop-pop gets a treat. Most of our future articles will deal with just one team and one aspect of their system at a time. 

In this case, it’s a useful case study to see both teams together, using the same system in different ways. The Oilers are going to be an exciting team to watch this year again. Jets hockey is high-event, and the Oilers’ new system makes them look conservative. The Oilers went 5-2-1 while learning this new system and the Jets were a terrible 1-4-3 while running out some odd line combos. It all gets real tonight.

 

*** Have a systems question that’s been bugging you? Please suggest it as a topic below or track me down on twitter @kevinmccart