Diving into the defensive zone

Data either tracked manually or provided by Evolving Hockey

Introduction

Heading into the break, I thought it would be a good idea to dive into a mini project on the Bruins to try to learn more about the team. After the collapses in Philadelphia and Pittsburgh, Bruce Cassidy specifically called out the back end.

Bruce Cassidy seemed to focus his comments towards the breakouts, but with many of the goals against starting from behind the net, I was more curious about the in-zone defense more than anything.

I think it’s important to keep in mind that the Bruins are one of the best defensive teams in the league at 5v5. The Bruins rank 5th in shot attempts against per hour, 2nd in expected goals against per hour, and 3rd in goals against per hour. Therefore, any commentary on the Bruins flaws in coverage should be combined with this context.

---

The Process

So how are we going to look into their defensive coverage? To my knowledge, no one in the public realm has looked at tracking puck touches against in the defensive zone, so I set out to do that.

On Wednesday, I hopped on my computer with a game plan. I would go through Tuesday night’s game and track all of the possessions that the Golden Knights had in the Bruins end. For each individual possession a Golden Knight’s player had I tracked:

• How the possession started
• Where the possession started
• How the possession ended
• Where the possession ended.

A possession could start from a controlled zone entry, a pass reception, a rebound, or a loose puck recovery. A possession could end in a shot, pass, turnover, or a stoppage in play. For the sake of subjectivity, a loose puck recovery could be a takeaway, and a turnover can still be recovered by a Golden Knights player. It just means the puck was not clearly in possession of the Golden Knights proceeding or following the possession respectively. The raw data is presented below.

This is where and how each Golden Knights possession ended on Tuesday night. There are a few points off of the rink above, which is a discrepancy between the application I tracked the data on and the function I graphed on.

---

Analyzing the Data

The reason for tracking this data was to first look for trends that might exist before looking at the video. When you are able to visualize the things that take place, it’s almost like being able to view the game in its entirety all at once.

The first thing I did was visualize where possessions started and ended, and colored it by the result. This will give us an idea of how the puck is traveling in the zone. Are the Knights getting closer to the net consistently? This is the sign of a very problematic defense, so it’s not something I would expect to see out of the Bruins. Are the possessions short or long? Short possessions generally signal high pressure. If a player can travel with the puck for a long distance, there is not enough pressure from the defense.

If you looked at that chart and got a headache, so did I. Luckily for you, I starred at the graph for a few minutes in order to digest it. I think there were a few things that we can take from this. For one, outside of two or three outliers, the Golden Knights were not able to walk the puck into the center of the ice. Another one is that the possessions down low are rather short. This is a good sign of pressure, for the most part. And lastly, many of the long possessions are off of entries, which were mostly kept to the outside.

Moving a bit closer to goals, we should take a look at how the Golden Knights created their shot attempts. In the chart below, I plotted the last pass proceeding a shot (the solid lines), the where the shooter skated after receiving the pass (the dotted lines), and the shot locations (the dots).

5 of the 35 shot attempts came either from a pass from below the end line, or from a shooter coming up from below the end line. I’m going to want to take a specific look at these shots when I go back into the video.

One thing I though about doing, and I think is a good idea, is to split up the zone into segments and look at how efficiently the Bruins can create turnovers in each segment of the zone. I simply didn’t have enough data for this to come out well, but I think it could be useful for larger samples.

---

Video Analysis

Let’s take a look at where the breakdowns are coming from below the end line if that is possible. We will first look at the two shots where the shooter started his possession below the end line.

This shot shortly follows a defensive zone faceoff loss. I think we would all agree that this shot isn’t very dangerous, but why?

There are two things that take place that force the Golden Knights to take an odd angle shot. Carlo quickly activates to pressure the puck carrier in the first place. Granted we don’t know what his next play could’ve been, but he would have more options if Carlo had hesitated. When Carlo activates, he only has the option to rim the puck up high, or pass back to Marchessault as he did.

The second thing is Wagner’s positioning. Any sort of aggressive behavior in the defensive zone requires a chain reaction of support. When Carlo activates, Chara cannot cheat to Marchessault because the front of the net would be wide open. He places himself closer to the net to be able to cover for Carlo while also being able to defend Marchessault. However, this opens up the option to the third forward for the Golden Knights. Therefore, Wagner has to place himself in a position to support Chara while also defending the passing lane. When you look back at the clip, you see the slight hesitation by Wagner in order to stay in that lane. I wouldn’t say this was played perfectly, but things could’ve been more dangerous had he cheated.

Ironically, the same Golden Knights players are involved here as well. Again, I don’t think this is a very dangerous shot, but I think the pro’s and con’s of the Bruins coverage come out in this clip.

Before the pass, we see your typical zone defense. Krug is on the puck carrier, Bergeron is “shadowing” Krug, Pastrnak is covering the slot, Marchand is covering the point, and Carlo is covering the front of the net. However, there is a decision to made when the puck goes down low here. Option one is what we saw above. Bergeron can come back to the new puck carrier while Krug stays on the previous puck carrier and Carlo and Pastrnak make sure the third forward won’t be a threat in the slot.

The other option is to aggressively pressure the new puck carrier by activating Carlo and sending Bergeron down there as well. This pucks a lot of responsibility on Pastrnak, and the Bruins ultimately don’t want to do that, but the alternative sets up a 1-on-1 situation that Bergeron was never going to win.

The benefit of the conservative option that the Bruins chose is that all of the Golden Knights forwards are boxed out when the shot is taken. This prevents chaos in front of the net. We also can’t ignore that the wingers weren’t much help here either. This is just something to keep in mind, especially as the game of hockey keeps advancing.

Now that we have looked at the shots where the shooter originated below the end line, let’s take a look at the shots where a pass came from below the end line.

Oh boy, the same suspects again. This time Carlo bobbles the puck as he is looking to break out. Both Kuraly and Chara pressure the puck quickly, forcing the blind pass. Unfortunately, the wingers were already committed to the breakout it seems. Neither Wagner nor Blidh were able to get back and cover the slot. This could have been dangerous if this was a left-handed shooter, but the extra time to take the shot allowed Halak to get square. I think this is fluke more than anything.

This low coverage 101. We see the same concepts shown as the second clip. Carlo is on the puck carrier, Krug is defending the front of the net, and Bergeron is shadowing. When Tuch receives the pass, Bergeron is in good position, limiting Tuch’s options. Brad Marchand’s positioning in the slot also deters any ambition to drive Bergeron 1-on-1. He can either go back to the point, or put the puck on net and see what happens. He chooses the latter.

This was definitely the best chance that came from behind the net, and it has two root causes. For one, the effort level was quite low on this play. It’s hard to tell how hard the shove was from Stone, but McAvoy is completely taken out of the play because of it. Furthermore, Krejci is thinking more about the breakout than anything. He only has his stick in the battle, allowing Stone to come in and win it with ease.

Secondly, Heinen is caught napping. By the time he realizes that he needs to be helping out in the slot, it’s too late. Both wingers were out very wide anticipating a breakout, but there needs to be a commitment to defense first.

---

Conclusion

I wouldn’t be too worried about the Bruins defensive play. They are one of the best teams in the league in terms of their defensive results, and I would be careful not to put too much weight into the recent results. I think that it’s possible we are just seeing opponents taking advantage of the few mistakes the Bruins make each game, and that will balance out over time.

I think there is room for this team to grow defensively, believe it our not. They are successful at being a disciplined, conservative team, but they could build on becoming more aggressive over time.

I think the more they are willing to take the battle to the opponents, like Lauzon did above, the less zone time they will sacrifice. These changes aren’t necessary at the moment, but it will be interesting to see how the coaching staff evolves the system over the next couple of years.