A Deep Dive into RPI's Puck Possession Problem
A stat-heavy breakdown of why the Engineers couldn't keep the puck
With the season all wrapped up, we thought it would be a good time for a little reflection on the past five months of hockey. A reflection on what has gone wrong, on what we can chalk up to bad luck, and on where the team can go from here.
Obviously, this season was pretty bleak from a fan perspective. Did we see the team come back to life a little bit during the playoff run? For sure. Was it fun to knock out Clarkson? Absolutely.
But giving up over 4 goals a game was not fun. Getting regularly outshot was not fun. Having a penalty kill success percentage below 70% was not fun.
So, what went wrong?
RPI Struggled with Puck Possession and Failed to Consistently Convert Possession into Chances
In retrospect, the Maine series probably should have been more of a red flag that the team was going to have problems keeping the puck for sustained periods of time. Yes, Maine was and still is a great hockey team, but they truly made RPI look like a Division III team that weekend.
The underlying numbers were ugly:
Now, if we look at the season as a whole, it got better, but never near good enough:
Here’s the thing: you could have the best penalty kill and goaltending in the league, but if your 5v5 numbers look like this, you just aren’t going to win a lot of games.
The penalty kill and goaltending (earlier in the season especially) rightly got brought up as big weaknesses for this team, but the reality is that they weren’t the difference between winning and losing. They were the difference between losing by 1 and losing by 3. The core problem was the play at even strength.
For a quick NHL comparison, RPI basically had the same underlying metrics (and win percentage) as the Blackhawks and Sharks. The two worst teams in the league.
The natural follow-up question is: “how did we end up like that?” This is a much harder question to answer, but we’ll give it a shot here.
1. Too Many Entries Into the Offensive Zone That Didn’t Result in a Shot
First, a quick primer on “OZ Possessions” and what that is referring to. Instat (our data provider) defines it as a period when the team is in the offensive zone with the puck for at least 5 seconds and they complete at least two actions (shot or pass, typically). For reference, Quinnipiac leads the country in Even Strength OZ Possessions Per 60 with 25.2. The worst teams in the league only had about 9.27 Per 60.
Compared to the rest of the country, RPI did not get into the offensive zone enough with good possession, and when they did get set up, they failed to get a shot off too often. RPI was just about dead-last in the country, only taking a shot on 73% percent of their offensive zone possessions. Again, for some reference, Clarkson led the country in this category with shots on 85% of their OZ possessions.
This matches up with the eye-test – a lot of the team had a pass-first mindset. And that’s all well and good if the passing is good enough to result in a high-quality scoring chance. But for RPI, a lot of the time it wasn’t.
For context, the best teams are making over 300 passes per 60 at even strength and completing about 87% of them. So, this difference isn’t huge, but it isn’t insignificant either.
It’s just another part of the game where if you start connecting a few more passes a game, then you can stay in the offensive zone for longer and be the team taking the shots instead of blocking them. You start driving play more, and don’t have tired legs from spending most of the game in the defensive zone. It all adds up.
2. RPI Didn’t Win the Puck Back Enough, Especially in the Offensive Zone
So, we know that RPI wasn’t taking enough advantage of their time in the offensive zone. But, what about how they faired with regaining possession?
Well, that wasn’t great either. Especially in the offensive zone.
Two metrics that give some insight here are takeaways (how often a team forces turnovers) and retrievals (how often a team wins possession back after a shot).
The number that stands out most here is takeaways in the offensive zone. This is a decent metric for the strength of a forecheck – you’d expect for teams with a dominant forecheck to be near the top of this category. Teams like Quinnipiac, RIT, and Alaska lead the category this season, averaging closer to 30 OZ takeaways per 60. These teams will lose the puck in the offensive zone but use a well-organized forecheck to win it back before the it can even get to their own defensive zone.
Another key metric here is the ability to retrieve pucks after shots. That’s what OZ and DZ Retrievals measure. How often do you win back the puck in the offensive zone after taking a shot and likewise, how often do you regain possession after the opponent takes a shot in your defensive zone?
*OZ and DZ Retrievals are normalized by the amount of time the team spent in the offensive/defensive zone. The retrievals are given in terms of “per 60 min of OZ time/DZ time.” Otherwise, teams that give up a lot of shots will naturally tend to have more DZ retrievals, and vis versa.
Again, it’s offensive zone play that stands out here. Contrary to what you might think, RPI actually did well recovering pucks after shots from their opponents. What they didn’t do well was regain possession after taking shots on goal themselves. Their forecheck was not effective enough.
What I’m trying to get at here is that it’s easy (and natural!) to blame the defensemen when a team regularly gives up over 4 goals per game. But it isn’t easy when you’re always on the backfoot because the team can’t hold the puck for sustained periods of time.
3. RPI Didn’t Generate Enough in Transition Either
If a team can’t consistently keep puck possession in the offensive zone, then they must be able to generate quality chances off the rush and in transition to be successful. Teams like Harvard are typically good at this “counterattack” style of play (although not this season funny enough). They use speed and skill to quickly transition from the defensive zone to the offensive zone and create chances while the opponent is still trying to regroup.
Unfortunately, this was not something RPI had much success with either.
Counterattack here is synonymous with “a rush chance.” Basically, a transition from the defensive zone to the offensive zone that leads with an event (shot/pass/turnover/etc) and that all happens within 12 seconds. For context, Michigan State averages the most counterattacks per 60 (at even strength) with 19.6 and Arizona State converts at the best rate with 56.1% of their counterattacks leading to shots.
A team that doesn’t hold on to the puck well and that doesn’t consistently create quality chances in transition will typically find itself in the defensive zone for much of the game and on the losing end of the scoreline.
What Does it Look Like on an Individual Level?
The team clearly did not hold onto the puck enough, but which players were at least having a positive impact on possession? Well, we can answer that by digging into a few different individual metrics.
I’ll preface this by saying all of these are from even strength only and normalized by ice time (everything is given per 60 minutes of TOI). We’ll also present what percentile (nationally) each player is in for each for some reference.
First, takeaways. This is a straightforward one and paints a good picture of which players forced turnovers the most and helped RPI win back the puck. We have it split up by zone for both the forwards and defensemen.
We’ll focus in on “OZ Takeaways” for forwards and “DZ Takeaways” for the defensemen as these are naturally the two most-telling metrics. The former is a good measure of forechecking capability, and the latter shows who the strongest in-zone defensemen are.
The forwards at the top of the list aren’t a huge surprise. Guys like Payant, Hotson, and Brackett all clearly have a high work-rate, and the underlying numbers show that it leads to them being effective forecheckers. We alluded to this earlier, but there is a drop-off though after the first five or so names, and it’s a big reason why RPI is near the bottom of the country in OZ takeaways overall.
For the defensemen, the numbers look a little better. Smolinski leads that pack and is in the top 5% nationally of takeaways in the defensive zone. Matta and Agnew, both of whom should be back next season, strong here too.
What about the other side of the spectrum? Who’s losing possession of the puck the most. Here’s what the giveaways look like.
Side note: this metric is normalized a little different. It’s now tied to player puck control time instead of their total ice time. This is so players who just naturally have the puck more don’t get dinged more for having more turnovers.
Ryan Brushett was near the top of the country here, as was Jake Lee. This is one area where the defensemen don’t come out looking as well, suggesting that there were problems with making the first pass or two after winning back possession in the defensive zone.
The problem with the forward group wasn’t that they gave the puck away carelessly too much, it was that they were not good enough at retrieving it back after getting shooting.
Only three forwards were better than the national average at regaining possession after a shot. It The defensemen fair a little better in retrievals, both after a shot and after a dump-in:
Again, the point we’re trying to hammer home here is: yes, the defensemen could have been better, but if RPI wants to have success going forward, they need to become better at sustaining good possession.
So, What Now?
Emphasize adding players who help win back and keep the puck through recruiting/the transfer portal.
The team has good, skilled players who we expect to return. The problem is that if they rarely have the puck, those skills are much less useful.
Earlier this offseason, Stephen broke down what we think the to-do list for RPI should be this offseason. This included filling some key roles on the first and second forward lines as well as on the top two defensive pairings. Finding the right guys who can complement the likes of Hotson, Muzzatti, and Smolinski is crucial to turning RPI into a competitive team again.
Buckle up, we have a big offseason ahead.