Canes Art provided by Chris Clark.
Matt’s Introduction: If you are catching up and want to see the other articles for Canes and Coffee’s reader/guest week, you can find links in the other Wednesday article HERE.
Wednesday’s guest article for reader/guest week at Canes and Coffee is another great one that lives at the cutting edge of today’s NHL and at the same time harkens back to Canes hockey coverage from years past. The author, Corey Sznajder, is busy working on an NHL microstats project to collect and publish a database of microstats for the 2016-17 season, but I figured it couldn’t hurt to ask if he would be interested in writing a guest post. Somewhat to my surprise, he agreed.
Corey’s formal bio is below, but please also allow me to introduce him from my Hurricanes hockey fan viewpoint.
Nowadays those of us who are passionate hockey fans see quite a bit of statistical information, analytics and visual representations for players. It has become common enough that it almost seems like this analysis has always been commonplace when in fact the reality is that the advent (or at least the surge) of fancy stats, analytics and similar in the public domain is really only about 4-6 years old. And back in the early pioneering days of this movement, Hurricanes fans were incredibly lucky to have a knowledgeable tour guide in Corey Sznajder. He started incorporating Corsi, other shot total and other analysis in his writing at his blog The Shutdown Line early on. I have always viewed Corey as the first leader and teacher on this front in the Hurricanes hockey community, and for Hurricanes fans who dug into advanced stats early, I cannot imagine there are many who did not utilize Corey’s Canes-centric articles to help climb the learning curve.
The Hurricanes community eventually lost Corey as a writer when he spent two years working for an NHL team, but with him back in the public domain, I consider Canes and Coffee lucky to include his insight today and at the same time take a jaunt down memory lane.
About the Author
Corey Sznajder (Twitter=@ShutdownLine) is an independent statistics and analytics expert for the NHL. He makes his online home at The Energy Line and is currently working on a project to track and publish a public database of microstats for the entire 2016-17 NHL season. Prior to his current project, Corey spent two years as a consultant for an NHL team. His tracking and analytical work also includes his All Three Zones project that tracked zone entries and exits for all 2013-14 NHL games, and he has ties to the Carolina Hurricanes community from his writing at The Shutdown Line from 2011 to 2014.
The Hurricanes entered this off-season facing a lot of questions, the biggest of which being which defenseman would they trade to acquire a top-six forward. This move never happened, but Carolina hasn’t had a quiet off-season. They managed to add Justin Williams to their top-six, helped their goaltending situation by trading for Scott Darling and added depth in the form of Marcus Kruger & Josh Jooris. The defense, however, was left mostly untouched. The only moves they made came on the edges by sending Ryan Murphy to Calgary while acquiring Trevor van Riemsdyk from the Vegas Golden Knights shortly after the expansion draft.
If you ask anyone who follows the NHL, most will rave about the Hurricanes defense. They’ve done well at building through the draft and we’ve seen the rewards of it over the last two seasons with Jaccob Slavin and Brett Pesce emerging as top-four defensemen. The argument now is that the Canes have a surplus of defensemen with mainstay Justin Faulk being pushed to the second pair and 2015 first-round draftee Noah Hanifin playing on the third pair for most of his NHL career. Carolina opted to keep these two along with their prospects and will enter the season with this as their top-four.
Most consider the Hurricanes defense to be in good hands and while that is true, it will probably have a different look. This will be the first time Hanifin will start the season in a top-four role and draw tougher assignments, likely taking over the role of departing Ron Hainsey. Say what you will about him, but Hainsey played a lot of minutes for this team over the years, especially on the penalty kill. The third pair will also be completely new, featuring the newcomer, van Riemsdyk and one of Klas Dahlbeck or Haydn Fleury, the latter being a bit of an unknown.
Evaluations for the 2016-17 season
With that in mind, let’s take a look at what makes the Hurricanes defense so unique and how they can optimize it next season.
Evaluating defensemen is a tough job because there’s a lot that plays into their role. It’s something that I’ve put a lot of work on during my time blogging on my old web site, Shutdown Line, where I tracked neutral zone events and zone exits to get a better idea of how each Hurricanes player was performing. I’ve continued that this year, looking at the entire league. Before we get into that, let’s look at some basic stats for the Hurricanes defense to see which pairs worked on the macro level.
|Slavin – Pesce||1030||56.6||47.7||54.3%||2.97||2.1||58.6%|
|Hainsey – Faulk||686||57||55.4||50.7%||1.75||3.15||35.7%|
|Hanifin – Tennyson||347||51.6||56.2||47.9%||1.73||3.28||34.5%|
|Slavin – Faulk||325||57.7||53.8||51.7%||2.13||1.36||61.0%|
|Hainfin – Faulk||293||61.5||47.6||56.4%||2.06||2.06||50.0%|
|Hanifin – Pesce||220||58.9||57.8||50.5%||3||3||50.0%|
|Dahlbeck – Murphy||178||48.1||55.7||46.3%||1.01||2.35||30.1%|
|Dahlbeck – Hanifin||150||52.5||48.9||51.8%||0.87||1.3||40.1%|
|Dahlbeck – Tennyson||133||47.8||65.2||42.3%||2.26||4.06||35.8%|
|Hainsey – Pesce||105||62.7||44.8||58.3%||0.7||2.1||25.0%|
This chart shows the shot attempt share of each Hurricanes defensemen during 5-on-5 play and the percentage of Carolina goals that they were on the ice for. It’s not the end-be-all, but it gives us a general picture of how the play went when each pair was on the ice. There are a few obvious things that stick out here. First, Jaccob Slavin is awesome. Second, almost every pairing that didn’t have him on it had some kind of problems. Justin Faulk and Ron Hainsey were above water in terms of shot attempts, but had a hell of a time keeping the puck out of their net. Some of that is out of their own control, as we all know the Hurricanes had some major goaltending problems last year (and years before that). Still, that’s opponents were creating a lot of shots with those two on the ice. They were probably getting some bad luck on the offensive end since the Canes were also generating a lot with them on the ice, but the end result was still negative.
Lastly, the third pair stunk. They went with the combo of Noah Hanifin and Matt Tennyson from roughly November until the trade deadline and the two got beaten to a pulp in both shots and goals. It was arguably the biggest problem with the defense last year and why van Riemsdyk was brought in. That should add some stability to this pair if things go to plan. Although, the greater concern here is Hanifin’s numbers. Carolina was on the wrong end of the shot battle and goals with him on the ice and a lot of it was coming while he was playing on the third pair. His numbers improve with Faulk & Pesce, which is predictable when you consider that they’re top-four defensemen and Tennyson was in his first full NHL season. It’s somewhat encouraging but it’s not the greatest sample to work with.
With the past done, let’s look at what the Hurricanes could do to optimize their defense. The first decision is an easy one, keep Slavin and Pesce together to start the year. They were outstanding last season, and the team shouldn’t tinker with something that isn’t broken for now. It gets trickier after that. One would think that Hanifin is a natural partner for Justin Faulk and it’s not the worst idea on paper. You have Slavin and Pesce handling the tougher match ups, as they did for most of last season, and that should free up Faulk and Hanifin to play in some more offensive situations. I mentioned earlier that there are new ways to evaluate defensemen and one of them is looking at their performance at the micro-level by tracking certain events like zone entries, zone exits and passes.
The latter is the most interesting thanks to the great work of Ryan Stimson. He and his volunteers have been tracking passes that lead to shot attempts for every game of the last two seasons and while the dataset isn’t complete yet, we have a decent enough sample to start looking at different ways to evaluate players. One way we can do that is using the passing data to see which type of playing styles each defenseman falls under and which types work the best together. Here, we can look at some potential defense pairs for next season. As Stimson says in his article, the idea is to find the best types of players that work together based on expected goals. For instance, defensemen who shoot the puck a lot (volume shooters) have a better chance of working well together than when they’re paired with someone who is more of a safe, defense-oriented player. Puck-moving defensemen also do well.
Let’s look at what Carolina has to work with:
|Shots||Shot Assists||Build-up||Passing||Transition||Danger Shots||Total Influence||Type|
|Brett Pesce||58.10%||59.94%||50.76%||55.96%||35.47%||45.87%||58.10%||Volume Shooter|
|Trevor van Riemsdyk||26.61%||54.74%||80.12%||70.34%||62.08%||53.82%||50.15%||Defensive-Oriented|
Note: Percentage = Percentile ranking
A small caveat here is that we’re limited to offensive data through this method, but this does give us a good idea of how good each player is with the puck and generating shots for their respective teams. The good news is that the Hurricanes have a lot of puck-movers to work with. I mentioned earlier that a shooter and a puck-mover is a good combo, and that’s what they had last year in Pesce and Slavin. They’re not offensive dynamos, but they hold their own and are excellent at keeping the puck in the right place, making their forwards jobs a lot easier.
As for finding a partner for Justin Faulk, the idea of putting him with Hanifin doesn’t seem to bad after reviewing this. Hanifin has his defensive struggles, but this highlights his offensive upside, and maybe he and Faulk can crush second-pair minutes next season. He’s also a more appealing option than the other two on the roster in Dahlbeck and van Riemsdyk, both of whom fall in the Defensive-Oriented category, which generally drags down your team’s Expected Goal Percentage unless you’re pairing them with an All-Around type. You could also see the Canes tinker with the idea of putting Faulk with Slavin again while putting Pesce with Hanifin. The potential upside of a Hanifin-Faulk pair is fun to think about, though.
Then there’s the third pair again, which looks like it could be a problem, but there are a few unknowns heading into the year. First off, it’s worth noting that van Riemsdyk grades out well in most categories except shooting and is pretty close to being in the puck-moving category. He also posted good numbers in a sheltered, third-pair role with the Blackhawks last season, so TVR has an okay track record and is probably an upgrade over Tennyson. That said, this year could be more of a challenge for him, as he will have either Klas Dahlbeck or rookie Haydn Fleury playing alongside of him, which is a downgrade from Brian Campbell and Duncan Keith. Still, he’s a capable defensemen and that’s all you’re looking for out of someone who plays 13-15 minutes a night.
The big question mark in all of this is whichever rookie earns that last third pair spot out of camp and how much of an impact will he have. The educated guess right now says it will be Haydn Fleury, but we all know that can change. After all, who saw Slavin and Pesce making the jump to the NHL so early? The Hurricanes have built arguably their strongest looking defense corps in franchise history, but there are still a few questions to be answered.