As we close the chapter on a tough five-game road trip, and the team prepares for a six-game home stand starting Friday night, it seems like a good time for a first look at macro trends in the Seattle Kraken play data. We’ll start with team-level information before highlighting Kraken player-level results. The goal here is to trigger further thought and inquiry, rather than deep dive in any particular area. Any one of these data sets could be an entire article. Unless otherwise noted below, the data is from Natural Stat Trick or NHL Edge.

Before getting to it, it is important to note that the Kraken have played just 14 of 82 games. This is a solid sample (17 percent of the season) but still on the low end for drawing meaningful conclusions. For example, analytics outfits like Evolving Hockey tend to publish their models only after accruing 15 to 20 games worth of data. I think there is plenty we can learn here, but let’s keep the salt shaker on the table for now. Particularly with a new coaching staff, it is quite possible this picture shifts in important ways by the time we do another one of these check-ins around the 30-game mark. OK, let’s dive in.

Seattle Kraken lag in the standings and shot quality metrics

It may be “early” in the data world, but the first 14 games of the season have had very real consequences for Seattle’s playoff prospects. Sure, as of Nov. 7, Seattle is only “four points out,” but five teams separate the Kraken from that last wild card spot—several of which have games in hand on Seattle.

On a standings points percentage basis, Seattle is fifth-worst in the league and third-worst in the conference. And this position is fully “earned” because, as of now, no team has fared worse in generating overall shot quality than the Seattle Kraken at only 41.69 percent expected goals for (xGF).

For those requiring a bit more explanation, expected goals (xG) is a metric that estimates the likelihood of a shot attempt resulting in a goal based on shot location and type. Expected goals for percentage (xGF %) compares a team’s total “xG for” with “xG against.” For example, if a team has taken three shots worth .1 xG total and given up one shot worth .05 xG, the team’s xGF% is calculated .1/(.1+.05) = 67%.

Seattle’s poor xGF% is likely overstated somewhat by the amount of time Seattle has played with the goalie pulled in recent games. That said, even if the fairer figure is slightly more generous to the Kraken, the team’s expected goal share would still be near the bottom of the league.

Seattle’s struggle has been both on offense and defense. On a production basis, Seattle’s 2.79 goals scored per game ranks 21st in the league, and its 3.21 goals against per game also ranks 21st. Pivoting back to shot quality, Seattle has been well below average generating quality looks for itself and in suppressing opponent shot quality.

The chart above does not separate out performance on special teams, so it is worth a short detour to look for takeaways on those units. As shown below, Seattle has been slightly above average in drawing penalties, but slightly below average in converting those opportunities into goals.

As for the penalty kill, the Kraken are taking more penalties than average and suffering the commensurate consequences, conceding more goals with a player disadvantage than average.

These substandard real-world results may be a bit “lucky” and present an artificially rosy picture of the team’s performance, though. In terms of shot quality, we see that Seattle is below average in creating quality looks on the power play and perhaps the worst in the league in preventing opponent shot quality on the penalty kill. The goalies may be the team’s best penalty killers so far.

Kraken skater data reflects broader team struggles

Shifting focus to individual data, there are few true “positive” indicators in light of the team’s overall weakness. Nonetheless, there is plenty of interest when evaluating how the coaching staff has deployed the players and in comparing the players against each other. First, let’s take a look at overall deployment and individual offensive engagement (via shot attempts).

Brandon Montour has been a workhorse on the blue line—leaned on all the more heavily with Vince Dunn out for the last 10 games. Despite missing the Toronto game, he still leads the team in ice time by a wide margin. Chandler Stephenson stands out as the only forward deployed as much as any of the top-four defensemen. As we will see in a moment, the coaching staff certainly views him as an all situations stalwart.

The data shows Stephenson doesn’t shoot much—but we knew that already. It is more notable to see Shane Wright so low in total shot attempts too. The team should consider putting him back in a better line context to get more out of him and the offense overall. (Worth noting, Wright was back with Oliver Bjorkstrand and Eeli Tolvanen at practice on Thursday.)

What explains Josh Mahura and Tye Kartye’s light overall usage? As shown below, Mahura and Kartye have not found a home on special teams. (Given the penalty kill unit’s struggles to create pressure, I wonder whether there may be an opportunity to incorporate Kartye’s aggression there. He has practiced as a penalty kill forward.)

Speaking of deployment, NHL data reports provide information on a player’s “zone starts”—meaning the zone (defensive, neutral, or offensive) where the puck was when the player jumps on the ice. With enough sample size, this can provide insight on coaching preferences. Is the play pinned in the offensive zone? Send out Player X because he can keep the offensive pressure up. Need to win a key defensive-zone draw? Tap Player Y because he can win face-offs and defend.

In the chart below we see that no player is deployed in a more difficult ice position context than Adam Larsson (followed by Brandon Tanev, Yanni Gourde, and Jamie Oleksiak). On the other extreme, we see Josh Mahura (and would see Cale Fleury if he met the games played minimum) as players “sheltered” with many more offensive-zone starts.

Beyond this “zone start” stat, NHL Edge now gives us another layer of information by providing data on the time each player spends in each zone. This allows us to compare a player’s zone starts against his time spent in each zone to get a rough understanding of whether he is—on balance—maintaining/improving his team’s ice position or losing ice position. For instance if Player Z was deployed in the offensive zone 70 percent of the time but nonetheless spent 55 percent of his time in the defensive zone, we could conclude that he contributes to substandard play driving the puck toward the opponent’s goal (a negative “ice position” impact).

Setting aside neutral-zone starts and time, we can compare the ratio of offensive- to defensive-zone starts (horizontal axis on the chart below) against the ratio of offensive- to defensive-zone time on ice (vertical axis). Indicative of the team’s overall struggles, only two players have spent more time in the offensive zone than the defensive zone—Montour and Mahura. This is to be expected for Mahura since he’s deployed so much more in the offensive zone. This is not so for Montour, who sees only a 45.2 offensive-zone start percentage. Beyond Montour, the players indicated in blue also move play back toward offense, while the players indicated in red have had the opposite impact.

Turning from deployment to goal-scoring production and Natural Stat Trick‘s shot quality metric, we see that the two leading scorers on the team, Jordan Eberle and Jared McCann, are not there by accident; they also lead the team in expected goals. Matty Beniers is third on the team in shot quality, earning almost exactly his three goal total. The fourth-ranked skater on the team is a somewhat surprising one—Brandon Tanev. That said, Tanev has always been able to create transition looks. The challenge for him is handling and finishing those chances.

Next let’s take a look at the team’s overall scoring (via goals scored for percentage) and shot quality share (via expected goals for percentage) with each player on the ice. Across a big enough sample, if a team is not generating surplus goals or quality shots with a player on the ice, it tells you about that player’s offensive and defensive contributions beyond just the box score.

As of Nov. 6, the Seattle Kraken have only one player with a 5-on-5 xGF% above 50 percent—Shane Wright. Everyone else is underwater in that shot quality metric. That said, usual suspects Brandon Montour, Jordan Eberle, and Jared McCann have been able to drive positive real-world goal impacts. That group is joined by Ryker Evans, Oliver Bjorkstrand, and Matty Beniers as the “plus” players on the team.

At the other end of the spectrum, Stephenson and Andre Burakovsky have struggled to contribute to goals and shot quality. (That said, I should note that these two combined for perhaps their strongest effort in the team’s last game against Colorado. So there may be a glimmer of optimism there.) Eeli Tolvanen’s on-ice impacts have been weak too, but he does have four goals, so he’s contributing in that department.

Finally, NHL Edge data also allows us to supplement the “eye test” with a quantifiable measure of how “fast” a player is playing at even strength. In the early going, Shane Wright has topped Matty Beniers as the team’s player who skates the fastest average speed. He also has the fastest individual burst on the team this season at 22.54 miles per hour. The chart below pairs that speed data with shot attempts each player has taken off the rush. It’s no surprise that the four centers skate the fastest on average, since the center has to cover all 200 feet of the ice and often needs to push harder to get to the appropriate spot.

For the Kraken, rush offense and play speed have not been closely correlated. To the contrary, of the four players to take four or more rush shots, three are defensemen. This also implies that much of Seattle’s rush success to date has been due to the activation of defensemen (and, perhaps, a result of other teams sleeping in coverage).

The goalies perform differently but have been solid overall

Lastly, shot location data provides some interesting insight regarding the contrasting production of Seattle’s goaltenders.

Joey Daccord has drawn eight of 14 starts, posting a .915 save percentage and 2.75 goals-against average. According to Natural Stat Trick‘s metric, he has saved 4.8 goals more than expected in his starts, which is sixth-best in the league. He’s done it by being a brick wall on shots he “should” save. His save percentage on shots Natural Stat Trick classifies as “low danger” is .991, best in the NHL among all goalies with at least four starts. He has saved 4.16 more goals than an average goalie on these types of shots. By contrast, his high-danger save percentage is slightly below average at .767. He has allowed approximately two more high-danger goals than average on these shots. This confirms the “eye test” that high-danger cross-seam passes have been the best way to beat Joey early on.

Philipp Grubuaer has six starts, with a .876 save percentage and 3.27 goals against average. His -3.08 goals saved above expected is below average, but there may be reasons for optimism. His save percentage on high-danger shots is .795. This is actually better than Daccord’s mark and approximately league average. He has also been solid with his rebound control early in the season, allowing approximately two fewer rebound shot attempts per game than Daccord. In contrast, he has struggled on low-danger shots, posting a save percentage of only .909 on those shots. There may be some bad puck luck or luck with screens affecting his early season numbers. Only time will tell whether this weakness on perimeter shots persists.