John Carlson has been on the shelf since December 23rd after taking a Brenden Dillon shot to the head. This is the second injury-induced absence of the season for Carlson, who also left in the first period of the October 29th tilt with the Nashville Predators and didn’t return to the ice until November 11th.

While Carlson’s offensive output since the 2017-18 season is among the best in the league for defensemen, his defensive play has resulted in rather polarizing outlooks on his overall performance and value.

In this post, we’re going to examine the impact of Carlson’s absence on the Capitals’ overall performance. The statistics used in this post are courtesy of Natural Stat Trick, Evolving Hockey, HockeyViz, and Hockey Reference. If you’d like to learn more about the statistical terms used in this post, please check out our NHL Analytics Glossary.

Carlson’s on-ice statistics

Before we get into the impact on the Capitals’ play as a result of Carlson’s absence, we’re going to dive into Carlson’s body of work during five-on-five play and his possession metrics. Here’s how he fared:

There’s not much to dislike, outside of Goals For percentage (GF%) and high-danger chances for percentage (HDCF%). Carlson’s shot attempt metrics in Corsi For percentage (CF%) and Fenwick For percentage (FF%) are rather solid. That positive ratio of shot attempts results in more shots for (SF%).

The influx of shot attempts and overall shots on goal drive Carlson’s expected goals for percentage (xGF%) up to the 53.3% rate we see above. This is counteracting his sub-50% marks in HDCF%, which typically have the highest impact on expected goals.

Carlson is also getting rather sheltered zone starts, starting 66.2% of his shifts in the offensive zone. Although that’s the case, zone starts aren’t the end-all-be-all for driving possession and scoring chance metrics. In fact, Peter Laviolette’s deployment of better offensive players in the offensive zone more frequently a better strategy. That’s where Carlson’s value lies, and at this point in his career, can’t be relied upon to be a stalwart, defensively.

The interesting takeaway here is that while most of the metrics show a positive performance, Carlson is on the ice for more goals against than goals for, resulting in that 49.98 GF%. One explanation is the on-ice save percentage of 89.84%, which is the lowest Carlson’s experienced in his 14 seasons with the Caps. On the other hand, Kuemper and Lindgren have a combined 91.73 save percentage during five-on-five play this season. This can mean either the Capitals and Carlson are rather unlucky while he’s on the ice, or there’s more high quality chances against, resulting in more difficult saves for the goaltenders to make. The likely scenario is some percentage mix of both.

Carlson’s player value in Goals Above Replacement (GAR)

A good way of measuring the total value a player brings to the ice is measuring his performance in relation to a replacement level player. Goals Above Replacement (GAR) measures a player’s value in all aspects of the game, including offensive and defensive value in all game situations. Here’s how Carlson has fared so far this season:

Realistically, this basically tells the story of what we’ve seen on the ice from Carlson this season. He’s solid offensively during even strength scenarios, but also struggles equally in even strength defense. This is a huge driver into why he gets such sheltered zone starts.

Carlson is still leaned upon to skate over 23 minutes a game on average, when he’s healthy, likely since Laviolette and the front office still view Carlson as an all situations number one defenseman. It’s probably not the best utilization for him, although he’s still rather elite offensively in terms of output since the 2017-18 season, scoring 349 points in 391 games.

The interesting story is that Carlson has been struggling in terms of power play offensive GAR. Here’s an explanation, via a chart from Evolving Hockey:

Although Carlson is above the replacement watermark by a standard deviation in xGF/60 and CF/60, he’s sitting nearly a whole standard deviation below replacement level in GF/60.

With his xGF/60 nearly the polar opposite of his GF/60, it’s hard to say that it’s due to bad performance on the power play, but likely due to his GF% while on the power play dropping to 87.5% due to being on the ice for two shorthanded goals against. On top of that, the power play has a shooting percentage of 10.77%, which is the lowest Carlson has been on the ice for in his career as a Capital. For context’s sake, the power play shooting percentage while Carlson was on the ice was 13.22% last season and 16.36% in 2020-21.

Impact on the Capitals

Now that we’ve set the context for how Carlson himself has performed this season, let’s take a look at how the Capitals have fared without him. Here’s a chart showing the team’s rolling xGF% throughout the season:

You can see a rather steady positive trend in xGF% from the doldrums in late October until the subsequent flat line after Carlson’s injury.

There’s not just one thing we can point at and declare the reason the Capitals haven’t continued that upward trend. For one, as the season goes on, you get a larger sample size of data, and numbers tend to stabilize a bit. Early in the season, very positive or very negative results could really impact the overall xGF% for the Capitals. Now that there’s more expected goals for and expected goals against racked up through 53 games played, we can likely see these stabilize and we have.

On top of that, the Capitals were playing their best hockey in the month of December, and their body of work in terms of results on ice really tapered off in January leading up to the All Star break. Part of that is likely relying on Matt Irwin to play a significant role on the ice instead of Carlson. Another could be the fact that the Capitals got Nicklas Backstrom and Tom Wilson back to the active roster and the subsequent line chemistry turmoil we saw as a result.

To show how the Capitals fare with Carlson on the ice, here’s a graphic from HockeyViz:

Overall, the Capitals have a +11% xGF/60 with Carlson on the ice during five-on-five play. This lines up with what we covered earlier. It also just so happens that one of the areas of excess shot attempts on the ice comes from the right point (where Carlson is posted). We can also see an ever-so-slight bump in xGF/60 on the power play, but it’s only a fraction of a percentage difference there. We know that Ovechkin is the fuel that powers the power play, so that’s not exactly out of sorts.

Defensively, it’s exactly what we can expect based off the eye-test and the advanced analytics. He’s not driving value defensively in five-on-five situations or on the penalty kill. The Capitals are exactly the same during five-on-five defensive situations with or without Carlson on the ice.

Conclusion

The Capitals are certainly missing Carlson’s offensive production from the blue line, and it’s evident in the value he drives during five-on-five offensive play. The Capitals are better with him offensively, and he drives a higher rate of xGF for the Capitals when he’s on the ice instead of off it.

There’s a caveat to Carlson, though. The years of Carlson being a plus defender during five-on-five play has passed him by. The Capitals and Peter Laviolette should probably be utilizing him a bit differently. Perhaps a reduction of ice time and penalty kill deployments could make him even more effective. The fact of the matter is, Carlson is still a top four caliber defenseman in the league just as a result of his offensive prowess. And one thing’s for certain, he’s still one of the six best defensemen on the Caps. The team will be better with him on the ice, especially if he’s used optimally.

By Justin Trudel