Four Goals Against Vancouver Don't Erase What Happened in March and April
The preseason scoreboard said 4–2 over Vancouver, and somewhere in Seattle a fan exhaled. That's understandable. After a season that disintegrated so thoroughly in its final weeks — two wins in the last seven games of April, a goal-per-game rate that cratered nearly a full goal below the team's own prior-stretch pace — any offensive life feels like oxygen. But a preseason result is a controlled environment. The real question isn't whether the Kraken can score in September. It's whether they can score in March, when it matters, with whoever is left standing.
That's where last season's numbers become genuinely uncomfortable. Seattle's late collapse wasn't a fluke of bad bounty — it was a structural failure. When the roster was intact and reasonably healthy, the team generated close to three goals a game over a 20-game stretch. When the injuries compounded and the depth was asked to carry weight, that number fell to 2.0. The difference between those two versions of this team is essentially the difference between a playoff contender and a lottery conversation. And the depth that failed them then has not been meaningfully reinforced now.
Look at where the production actually came from last season. Jordan Eberle and Matty Beniers anchored the top of the lineup with 33 and 32 points respectively, and Jared McCann was the team's most dangerous scorer per game when healthy — but McCann played only 23 games. Below that tier, the falloff was steep. Shane Wright and Kaapo Kakko both contributed, but their ice time tells you exactly where the coaching staff ranked them in the pecking order. The left wing depth chart entering this camp — Eduard Sale, Lleyton Roed, Carson Rehkopf at the top of the positional rankings — produced essentially nothing measurable last season. That's not a knock on young players; it's a structural observation about what happens when the top six goes down and those names are suddenly asked to fill meaningful minutes.
Eeli Tolvanen's departure sharpens this problem rather than resolving it. He finished fourth on the team in points last season with 29, logging nearly 50 games. He wasn't a star, but he was a reliable secondary contributor — exactly the kind of player whose absence gets felt in February rather than October. Mackie Samoskevich, acquired from Florida, is 23 and intriguing, but trading a draft pick for potential is a bet on upside, not a guarantee of replacement production. My read is that the Kraken are hoping Samoskevich grows into Tolvanen's role faster than the schedule demands. That's a reasonable hope. It is not a plan.
The goaltending situation adds another layer of genuine uncertainty that a 4–2 preseason win cannot paper over. Joey Daccord, the team's clear number one, has been sidelined since late June with no public timeline for return. Philipp Grubauer and Nikke Kokko split the backup load last season, and the depth chart shows all three goalies carrying meaningful portions of the workload — which is another way of saying no one ran away with the job when Daccord was healthy. If he isn't ready for October, the Kraken are opening the season with a goaltending question that they haven't answered since the spring. A team that went 10-20 over its final 30 games cannot afford to start the next season in a similar posture of uncertainty.
The front office clearly knows something needs to change — the Chuck Fletcher hire as senior advisor to Jason Botterill signals that the organization is seeking outside perspective on exactly these kinds of structural questions. Fletcher has built rosters in this league for a long time, and bringing that experience into the room suggests Botterill wants a second set of eyes on the depth problem, even if the official language around the hire stays vague. Whether that translates into action before October is the open question. Right now, the Kraken have a preseason win and a roster that looks very similar to the one that fell apart when the calendar turned to March. Four goals against Vancouver is a pleasant data point. It is not evidence that the underlying problem has been solved.