The risk and reward with Erik Karlsson
By NHL_Sid
1 year agoAbout a week ago, Chris Johnston reported that the Edmonton Oilers have explored a potential fit for San Jose defenceman Erik Karlsson.
Erik Karlsson leads the entire league in even-strength points and is on pace for over 100 points, a feat not achieved by a defenceman since 1992.
With a massive $11.5M contract, it will be difficult to trade for him, but the possibility of Karlsson to Edmonton is legitimate. Thinking about a team containing Connor McDavid, Leon Draisaitl, and Karlsson is incredibly fun as an Oilers fan, and is likely a nightmare for other teams to think about defending them.
However, is he the right fit on Edmonton’s roster? They would likely need to pay a hefty price for Karlsson, who’s currently at age 32. Do the pros outweigh the cons?
Let’s take a look at the potential upside and the risks of trading for Erik Karlsson.
*All stats via EvolvingHockey and Natural Stat Trick, all microstats via AllThreeZones unless stated otherwise
Just how good is Erik Karlsson exactly? What are the advantages of trading for him?
As mentioned earlier, Karlsson’s production this season has been phenomenal.
In the past two seasons, Erik Karlsson’s impact on generating EV scoring chances (RAPM xGF) ranks first among all defencemen. Furthermore, here’s a look at his offensive microstats:
Look at all that deep blue! Karlsson is just flat-out dominant, as he ranks at or around the top of the league in about every offensive area you could possibly think of. It’s difficult to find a metric that doesn’t say Karlsson isn’t a top-five offensive defenceman (at worst).
There are several ways Karlsson could improve this team.
Edmonton has struggled with making efficient breakouts this season. Karlsson’s high volume of possession exits could be a huge help in that area.
For once, the Oilers are having a bottom six that’s helping the team win games. With McDavid and Draisaitl off-ice, the Oilers have out-scored the opposition 52 – 39 at 5v5. Their room for 5v5 improvement actually comes with McDavid/Draisaitl on-ice. McDavid and Draisaitl together are at a solid 54 percent goal differential together, but McDavid is dead even with a 50 percent goal differential without Draisaitl, while Draisaitl is at a poor 42 percent without McDavid.
This is where Karlsson could be a massive help. Here’s a look at his impact on goal and expected goal differential throughout his career with San Jose:
Karlsson had a very rough 20-21 season, ranking below average in both goal and expected goal share. But in the other four seasons, the Sharks have out-scored and out-chanced the opposition at a higher rate with Karlsson as opposed to without. This year, Karlsson is at an excellent 56 percent goal share.
McDavid is at a 45 GF% and 53 xG% with Edmonton’s top pairing of Nurse-Ceci pair on-ice. Without that pairing, he’s at 54 GF% and 59 xGF%. The pattern is similar with Draisaitl, who has an awful 40 GF% and 46 xGF% with Nurse-Ceci, and a 54 GF% and 54 xG% without.
Put in simpler terms, Edmonton’s stars out-score and out-chance the opposition at a significantly higher rate away from the pairing of Nurse and Ceci. Part of that is influenced by the quality of competition, but nonetheless, the top pairing needs to improve.
With Erik Karlsson on the roster, that changes everything, as he could largely help Edmonton’s top-six at 5v5. Either Karlsson could be on the top pairing with Nurse, as Nurse could largely benefit from a superb breakout passer, or Nurse and Ceci can be a bit more sheltered, with more TOI against top opposition allotted to Karlsson.
With Karlsson, the Oilers would be a team with a top-six that’s unstoppable offensively, and a bottom-six that can out-score the opposition at high rates.
In the 2022 playoffs, the Oilers already ranked first among all teams with 4.03 goals per game. Now imagine that team, but with RNH and Hyman producing at a 100-point pace, alongside the addition of Erik Karlsson. Scoring at 5 goals per 60 is a feat never achieved by a team in the playoffs in the 21st century, but if one team could do it, I’d bet it’s an Oilers team with Karlsson.
And even then, the Oilers rank 10th in the league in 5v5 goals per 60 this regular-season. There’s certainly room to improve offensively and to maximize McDavid and Draisaitl.
Karlsson is the best player available on the market, and with Draisaitl’s contracts expiring in two more years after this one, this seems like a marvelous opportunity to trade for a superstar, and go all-in.
But what about the risks of trading for Karlsson?
One thing that’s quite established is that Erik Karlsson is brilliant offensively. But what about the other side of his play?
Here’s a look at where Karlsson’s isolated defensive impact on both goals and expected goals has ranked in every season he’s played in:
(I usually use percentiles for this stuff, but it’s a lot less time-consuming for me to use this method, inspired by JFresh)
To put it mildly, Karlsson’s defensive results have not been good. The only year he had a positive impact on suppressing both actual and expected goals was 2012-13, a season in which he played just 17 games due to suffering an Achilles injury. Compared to every other year, that sample isn’t exceptional.
His on-ice save percentage was at a career-high in 2019-20, which is the reason for his above-average GA results, and his xGA impact was pretty decent in his first year with San Jose.
But overall? The objective of skater defence is to prevent goals and chances against, and Karlsson has struggled at this for his whole career.
Of course, you could just chalk it up to the poor teams he’s played on, but RAPM is a model that already attempts to adjust for a player’s quality of teammates, quality of competition, zone starts, etc. Is it perfect? Certainly not, but typically, the larger the sample and the more linemates you’ve played with, the more accurate the RAPM output is. In this case, we have fourteen seasons’ worth of data.
Karlsson’s teams have consistently allowed more goals and quality scoring chances with Karlsson on-ice. Even if you believe these metrics are the product of taking lots of risks to generate offence on a poor team, it seems like this risk-taking has been sustained throughout his entire career. That’s simply a red flag that needs to be noted.
Karlsson’s defensive metrics could definitely improve on a better team, but that’s an uncertainty. I can’t think of many defencemen who’ve struggled defensively throughout their whole careers, but suddenly immensely improved at age 32, even if Karlsson will switch teams. Not impossible, but seems a bit unlikely to me.
This is further backed up by his microstats:
Karlsson has been below-average at defending entries in each of the past three seasons. He’s allowed controlled entries against at a decently high rate, and the amount of chances he’s allowed off those entries against is even worse. Furthermore, Karlsson is a very high-event breakout player.
As shown earlier, Karlsson ranks in the 99th percentile in possession zone exits per 60, meaning he’s superior to 99% of the league’s defencemen at exiting the zone with possession.
However, his failed zone exits per 60 ranks in the 15th percentile, meaning 85% of the league has been better than Karlsson at limiting turnovers/icings when attempting to start breakouts.
A similar story is present with his defensive-zone retrieval results. He can retrieve pucks at a high volume, but on the other hand, Karlsson botches a lot of retrievals, meaning he’s prone to turning the puck over once he retrieves the puck. In fairness, players who do have the puck more are going to make more turnovers, but his successful retrieval% is below the league average, and his exit success% is good, but not great.
Now, I did display the overall relative goal and expected goal share results earlier in the piece. Even on a poor team, his net impacts are positive. The good does considerably outweigh the bad!
I don’t think that’s even a question here, as Karlsson is probably the league’s best offensive defenceman. He would need to be really, really bad defensively to have a net negative impact.
However, I think it’s very safe to say that his defensive play is a concern, and it brings down the overall net value he could provide. The Oilers will likely be the best offensive team in the league with Karlsson. But on the defensive side, they won’t improve, and there’s a likely chance the exact opposite occurs.
Speaking of his goal differential impact, it was seen earlier that Karlsson’s impact on GF% was very, high, much higher than in recent seasons. Is this sustainable?
Here’s a look at Karlsson’s career-wide on-ice shooting percentage:
Here’s a look at Karlsson’s career-wide on-ice shooting percentage:
Now, I want to clear something up before people misinterpret my point; Karlsson’s offence is not the result of puck luck. As mentioned previously, his expected goal for impact ranks first among all defencemen. It would be a disservice to Karlsson by simply calling this season an outlier.
I think two things can be true at once; Karlsson is probably the best offensive defenceman in the league, but Karlsson is also facing some career-high puck luck which is unlikely to sustain.
Throughout his career, Karlsson has never exceeded an oiSH% (on-ice shooting percentage) of over 9%. This year? Karlsson is at 11.5%.
Some elite players can sustain an elite on-ice shooting percentage. Patrick Kane and Leon Draisaitl are prime examples of players who’ve always had a high oiSH% that regular players cannot sustain. It’s valid to suggest that Karlsson could be one of those players, but his career-wide results say otherwise.
Not even prime Erik Karlsson could hit a 9 oiSH%, and yet 32-year-old Karlsson is almost at 12%? Safe to say that many things point to his current 100-point pace being unsustainable.
Karlsson’s impact on expected goals is actually very similar to last season, and his impact last season was actually a bit higher. However, the large difference in goal differential/production can be explained by the massive increase in on-ice shooting percentage.
So yes, Karlsson is excellent offensively, but I think it’s important to recognize that he’s faced unsustainable puck luck. I’m not 100% confident he outscores his defensive shortcomings to the brilliant extent he has this season, meaning his goal differential this season could be an outlier.
There are several other risks with Karlsson as well. Karlsson is 32 years old, and has a concerning injury history. He makes $11.5M, and his trade package could be hefty.
It will take some cap juggling to fit him in, especially considering that San Jose doesn’t seem too fond of retaining a ton of salary. You could do a three-way trade and have the third team retain some money, but it’s doubtful if you’d be able to convince other teams to retain a considerable amount of salary for the next five years.
Karlsson’s exact asking price is a bit unclear, as various reporters have had conflicting reports this season. Some sources say San Jose wants three first-round picks, others disagree with that. Either way, it’s likely that multiple players, picks, and/or prospects will go the other way.
As mentioned earlier, Edmonton scored a league-leading 4.03 goals per hour in the 2022 playoffs. Let’s compare this to Tampa, the team that went to the SCF in each of the past three years, and won two cups.
From 2019-20 – 2021-22, Tampa scored 2.8, 2.9, and 3.2 goals per 60 respectively, considerably lower than last year’s Oilers team. Offensively, the Oilers are already at that cup-contending level.
The reason that Tampa achieved so much more success is due to their play at the other side of the rink; Edmonton allowed 3.7 goals per 60 last year in the playoffs, whereas Tampa was at a significantly superior 1.9, 2.0, and 2.6 GA/60 in each of those three years respectively. Defence matters, a lot.
I’m definitely in favor of going all-in. All of Edmonton’s first-round picks should be on the table, but the many risks of trading for Karlsson need to be noted.
Even with Karlsson, it’s certainly not guaranteed that Edmonton will win the cup, and they would still have many defensive flaws. Trading for Karlsson would mean that the Oilers would go the “all-offensive” route, trying to out-score their defensive deficiencies.
This strategy could definitely work; in recent history, we’ve never witnessed three players of McDavid, Draisaitl, and Karlsson’s caliber on the same team. Not to mention, the team also has Hyman and RNH, who currently produce at a 90-100 point pace.
Edmonton would be a top-two contender in the West with EK, and once you get to the Finals, anything can happen. There would be a lot of 7-6 or 8-7 games, but it could be a strategy that works and results in a Cup. But undoubtedly, there are risks with going this route.
Conclusion
Dec 22, 2022; San Jose, California, USA; San Jose Sharks defenseman Erik Karlsson (65) skates on the ice during warm-ups before the game against the Minnesota Wild at SAP Center at San Jose. Mandatory Credit: Robert Edwards-USA TODAY Sports
In a perfect world, you’d love to make multiple moves to improve this team.
A list of players the team could benefit from could include a defensively-inclined RD partner for Nurse, a top-four LD, a RW with finishing talent in the top-six, and perhaps even another defensive forward to help Draisaitl’s line. All of those moves would make the Oilers a very well-rounded team at both ends.
But with Edmonton’s cap situation? That’s a fantasy.
It’s difficult for Edmonton to even add one impact player. So is it worth it to try to make the team well-rounded when that’s likely difficult to do with just one move? Perhaps doubling down on “all-offence” is the move to make, and trying to maximize McDavid and Draisaitl’s offensive output. That’s one of the major arguments for acquiring Karlsson.
Out of all the individual players available, not one available player would make a larger overall impact than Karlsson. He’s the best player on the market and a true 1RD, even while considering his flaws.
That includes Jakob Chychrun. Even with Chychrun’s strong two-way results on an awful team, Karlsson’s brilliant offence exceeds that.
However, what makes Chychrun so appealing to me is his age and $4.6M cap hit. In today’s cap world, you need more value contracts like that. That same benefit isn’t present with Karlsson, who has a pricey contract, even with some retention, that ends when he’s 37.
If the Oilers don’t win the cup with Karlsson, then in that scenario, they’ve dealt valuable assets which would make it difficult to go all-in the following season. Elite players do tend to be exceptions to traditional age curves, but there is a chance of a decline to some degree after this season. Whereas, in comparison, if the Oilers dealt for Chychrun and didn’t win the cup this season, Chychrun is only going to get better at age 24, and will still make $4.6M until Draisaitl’s contract is finished.
Not to mention, Karlsson’s defensive metrics is an area for concern.
So all things considered, this is the textbook definition of a high-risk, high-reward situation, where both sides need to be noted before making a trade.
The reward and the best-case scenario of trading for Erik Karlsson is self-explanatory; we get to watch perhaps the best offensive team in the past decade, alongside a Stanley Cup victory, whereas the risk is overpaying for a pricey player that isn’t enough to put them over the edge.
The deadline is just a day under two weeks away. It will be very interesting to see what the Oilers will do.
Find me on Twitter (@NHL_Sid)
Recent articles from NHL_Sid
