Looking into Edmonton's defensive pairs for 21-22
How should they look with the departures of Adam Larsson and Caleb Jones, and the addition of Duncan Keith?
Ranging from Adam Larsson signing in Seattle, to Caleb Jones being dealt for 38-year-old Duncan Keith, Edmonton’s defence has changed quite a bit so far, and it’s not even Free Agency Day yet.
Edmonton only has 2 right defencemen under contract (Ethan Bear and Evan Bouchard) due to Adam Larsson leaving, so it’s likely that there are still going to be more changes.
It was reported that the Oilers may be a dark horse candidate for Dougie Hamilton, a pending UFA and one of the best right-defenceman in the league.
Unfortunately, Holland pretty much ruled out that possibility in his most recent press conference, saying that the Oilers “won’t be big game hunting on the blueline” this offseason.
So, how much does Edmonton need to improve their blueline? What should the pairs look like on the opening night against Vancouver if the Oilers hope to be a playoff contender this season?
Top Pair: Darnell Nurse - Ethan Bear
For me, the top pair is pretty obvious. Darnell Nurse is pretty much locked in as the top left-defenceman, and the best option for Nurse’s partner should be Ethan Bear. Last year, per EvolvingHockey, Ethan Bear’s impact on scoring chances at even strength ranked 17th in the league for defenceman with at least 200 minutes of even-strength ice-time. Furthermore, Edmonton’s entire top 6 plays better with Bear on the ice.
The same is true for Nurse, as both his on-ice shot and scoring chance differential improves with Bear on the ice.
Bear played with Nurse on the top pair for almost the entire year in 2019-20, a year where Bear averaged more TOI per game than any other rookie, which included Cale Makar and Quinn Hughes. He excelled in this role, as Edmonton’s goal and expected goal differential rose with him on the ice.
He began 20-21 with Nurse on the top pair, but after missing almost a month due to a concussion, Barrie took that role and has played there ever since. Some still advocate for Tyson Barrie to continue to play there, but the main issue with Barrie is how he gives up more than he creates. Edmonton’s forward group in general improves defensively by a very large margin when Barrie is off the ice.
Barrie’s isolated impact on GA (goals against) was 214th in the league out of 231 defencemen with at least 200 minutes at even strength, while his impact on xGA (expected goals against) was 225th. To add on, he brings down Connor McDavid’s offensive game as well; McDavid has a 3.35 5v5 Points/60 with him on the ice, and a 3.84 5v5 Points/60 without him.
Barrie isn’t a terrible option to play on the 3rd pairing in a sheltered role, as he actually helps Edmonton’s bottom 6 improve in terms of possession/scoring chances by a small margin. His defensive numbers are also superior when he plays against 3rd and 4th lines per PuckIQ. The issue is that after finishing 1st in the league in points for all defenceman, he’s most likely looking for at least $4-5M, and you shouldn’t spend that much money on a 3RD.
Unless the Oilers manage to somehow sign Dougie Hamilton, the clear option for the top pair in 21-22 should be Nurse - Bear.
2nd Pairing: Jake McCabe/Mike Reilly - Nick Jensen/Dylan Demelo
The 2nd pair is an interesting topic. There’s a very valid argument that playing on the 2nd pair might be a more difficult role than the top pair in Edmonton.
The 1st pairing may face the hardest competition and play the most minutes, but at the same time, they spend the majority of their ice-time with McDavid and/or Draisaitl, while having favourable zone start usage as well. Meanwhile, the 2nd pair still plays tough opposition, while spending a lot less time with Edmonton’s stars and having difficult zone start deployment too.
Adam Larsson, who played on the 2nd pair in Edmonton for all of 2021, ended the season with the lowest QOT% (quality of teammates) per EvolvingHockey, and the 4th highest QOC% (quality of competition) per PuckIQ for Oilers defenceman. He also had 146 of his shifts start in the defensive zone, while only 67 of his shifts started in the offensive zone.
It can be pretty difficult to find players who can thrive in this type of role, but there are some good options available via trade or free agency. They may not replicate Larsson’s exact results in 20-21, but they have a very high chance of at least playing solid on the 2nd pair.
For the left side, Jake McCabe and Mike Reilly make a lot of sense to me.
The first three bars are offence, the last two bars are defence. Blue means the player is good in that category, while red means they struggle in that category. These stats already account for a large variety of factors that may influence their on-ice results, such as teammates, competition, zone starts, score and venue, etc.
Mike Reilly is a very strong possession player with great offensive metrics and above-average defensive stats, while Jake McCabe has a very encouraging track record of being a solid defensive defenceman. He’s been one of the few bright spots in Buffalo. Both are pending UFAs as of now, and both have had experience and success (positive relative scoring chance metrics) playing against top competition per PuckIQ. They’re among the best realistic options for the 2LD spot.
As for the right side, Dylan Demelo and Nick Jensen would be good choices.
Jensen is pretty much a slightly worse version of Adam Larsson, which isn’t a knock against him at all, and Jensen might be the best replacement for Larsson in terms of stylistic fit. Demelo is very similar to Ethan Bear in a way, as both are strong two-way defenders. The issue is that neither are free agents, but both don’t seem to hold too much trade value, as both were left exposed for the Seattle Expansion Draft by their respective team, and carry cheap cap hits (Demelo has 3 more years left on his current deal with an AAV of $3M, Jensen has 2 more years left with an AAV of $2.5M).
Niklas Hjalmarsson is also a nice idea and is currently a UFA. He’s 34, but it still looks like he can defend very well, as his shot suppression numbers have always ranked very high in the league.
Alex Edler, also a UFA, provides a veteran presence that Holland seems to like, and Edler has put up solid two-way stats in the past few seasons playing in a tough role. However, it is worth mentioning that foot speed is an issue with him and that he could start to decline soon.
Dmitry Kulikov put up very strong play-driving numbers this year, playing primarily in the top 4 for both New Jersey and Edmonton, but his stats prior to this year were pretty average. I would love to have him return as the 3rd pairing LD, but considering how this year may have been an outlier, I’m not extremely confident about playing him on the 2nd pair unless he manages to replicate his recent results.
Ryan Murray is an OK option. He struggled in 20-21, but he had fine defensive results before this year. It’s worth keeping in mind that he doesn’t have a very pretty injury history though.
David Savard is more of a poor man’s Adam Larsson in my opinion. They’re similar in terms of how they’re both low-event, defensive defencemen, but Larsson is better offensively and defensively than Savard. Furthermore, Savard is a net negative in terms of overall impact despite his solid defensive ability. He’s more suited to play in a 3RD role, rather than a shutdown 2RD role.
Josh Manson was also talked about a bit by Bob Stauffer, but I’m not a big fan of him. After having a very good season in 17-18, he’s declined quite a lot, as Anaheim sees an increase in goal and expected goal differential with Manson on the bench. To be fair, he has struggled with concussions, and there is a chance that he could bounce back, but it is somewhat risky.
I think McCabe and Reilly are the best 2LD options, while Demelo and Jensen are the best 2RD options, but in my mind, it should be pretty clear that the Oilers need to add 1-2 more top 4 defencemen if they plan to contend this season.
3rd Pairing: Duncan Keith - Evan Bouchard
Duncan Keith was acquired by Holland to play on the 2nd pair in a shutdown role, and it looks like he will begin the season in that role. In my opinion, this is a recipe for failure.
Keith has declined quite a bit from the dominant Norris-winning defenceman he once was. His offensive stats are still a bit above average, but his defensive play has collapsed. In 20-21, his isolated impact on GA (goals against) was 229th out of 231 defencemen with at least 200 even-strength minutes played. His isolated impact on xGA (expected goals against) was 231st, or in other words, dead last.
A lot of people were quick to point out that Keith’s metrics suffered because he was paired with young, inexperienced players like Adam Boqvist, with one certain article going as far as saying Keith’s time spent alongside Boqvist was “adventures in babysitting.”
Many also shifted the blame onto the defensive system of Chicago’s coach, Jeremy Colliton, saying that no defenceman did well under his system. The flaw with this argument is that Boqvist’s metrics were better away from Keith. In fact, all of the players that Keith primarily played with (Mitchell, Murphy, etc) saw a rise in their on-ice shot and scoring chance shares with Keith off the ice.
The defensive system in Chicago is probably not helping Keith at all, as the team is bad with and without Keith. But if you don’t believe that Keith has struggled, what’s the reason for Chicago as a whole getting worse the moment Keith touches the ice, for the past 3+ years?
As a result, the best role for him might be to shelter him and play him to his strengths. As stated before, Keith’s offensive stats are above average, so the chances of him bouncing back in an offensive role are higher than if he played in a shutdown role on the 2nd pair. Furthermore, Keith was a positive in relative expected goal share and high danger chance share with Toews, Kane and Debrincat off of the ice at 5v5, so essentially with Chicago's bottom 6.
The problem is that Keith is an anchor to his team’s top 6 players and can’t play well against tough competition, but perhaps an offensive, sheltered 3rd pairing role may not be a bad idea?
Don’t give Keith a lot of ice-time with the top 6, as he brought down Chicago’s best players (Kane, Debrincat, Toews) as stated before, and the same could happen with McDavid and Draisaitl if he played a significant amount of time with them. But he could help provide some offence to the bottom 6 in Edmonton if he primarily played with them, perhaps similar to how he helped improve Chicago’s depth. Additionally, aim to give Keith favourable zone start usage as well.
As for Keith’s partner, start the season with Evan Bouchard on his right side, who could help maximize the offensive potential of the 3rd pair. As the year goes on, maybe promote Bouchard to the 2nd pairing. I think Bouchard has a very high ceiling, and there’s a very strong chance that he could take over the 1RD spot from Bear in 22-23. But for now, I think it’s a safe bet to start him on the 3rd pair as long as Edmonton can get a strong 2nd pairing.
Playing an offensively-oriented / sheltered 3rd pair of Keith - Bouchard might be the best chance for Keith to succeed in Edmonton to some degree.
Conclusion
The pairings discussed in this article may not be entirely realistic, and they’re primarily based on what I personally think should occur. I highly doubt Keith will start on the 3rd pair, and you also can’t rule out the possibility of Edmonton bringing back Tyson Barrie to play on the top pair.
If Barrie does re-sign, Edmonton could still try to bring in a player who’s similar to Larsson in terms of defensive ability, and considering how Holland has said that Bouchard is going to be playing every night, it means Bear is the odd man out. In other words, Barrie re-signing may mean Bear might be traded.
It’s an interesting situation, as Holland still has a decent amount of cap space left, and he can save more if Neal is bought out and if he can find a way to trade Kassian.
Either way, in my opinion, it’s pretty clear that the current defence core does not look very promising at all. If the Oilers want to contend next year (and they obviously should), they should be looking for a strong 5v5 replacement for Larsson and preferably a 2LD (not named Duncan Keith) as well.
Find me on Twitter (@NHL_Sid)