clock menu more-arrow no yes mobile

Filed under:

Philipp Grubauer’s case for the starting role

He’s back to his old self, it’s just hard to see

Winnipeg Jets v Seattle Kraken Photo by Steph Chambers/Getty Images

If I’ve learned anything from the classic Kevin Costner movie Bull Durham, it’s that you “never fuck with a winning streak.”

The Kraken are winning games with Martin Jones in net, and losing them with Philipp Grubauer in net. That’s why it’s no surprise to see head coach Dave Hakstol giving the lion’s share of starts to Jones. Jones currently holds a 22-6-3 record (.783 win percentage) this year, which is the most wins he’s had in a season since 2018-19. Grubauer, meanwhile, is sitting with a 4-8-2 record (.333 win percentage), which actually puts him on pace for a worse season than last year in terms of wins and losses. There’s a distinct correlation with which goaltender is starting in net and Seattle’s chances of winning the hockey game.

Here’s the thing though: correlation does not equal causation. Jones’ save numbers aren’t actually any better than Grubauer’s. In fact, some of them are worse than Grubauer’s. Jones’ save percentage is currently .894 as of this writing, barely a hair better than Grubauer’s .893. In terms of quality starts (starts where the goalie had an above average save percentage, relative to the season’s league average) Grubauer has actually given the Kraken more per game than Jones — 60% of Grubauer’s starts have been considered “quality,” vs 53.3% for Jones.

Which then begs the question — why is Martin Jones winning so many more of his starts than Philipp Grubauer?

A tale of two defenses...

One aspect that is impacting the differing win rates between these two is the defense playing in front of them. Yes, it’s the same skaters, but the actual defensive performance the Kraken are putting in front of these goaltenders is quite different.

Overall, the Kraken have been fairly average defensively. Their 3.1 goals against per game ranks right in the middle at 17th in the NHL. However, a good chunk of that number comes from their poor penalty kill. At 5-on-5, Seattle actually sneaks inside the top-10 in goals against per 60 minutes of play. Based on raw shot totals, they look even better — only the Hurricanes and the Bruins allow fewer unblocked shots than the Kraken at 5-on-5, according to Evolving Hockey. The problem is that every single defensive number changes a significant amount depending on who is in net, and every one of them is bad news for Philipp Grubauer.

Kraken goalie stats table. First line: Martin Jones, 33 games, 1480 minutes, 37.8 fenwick against per 60, 2.19 goals against per 60, 2.09 expected goals against per 60, .942 save percentage, .945 expected save percentage. Second line: Philipp Grubauer, 16 games, 665 minutes, 39.4 fenwick against per 60, 2.25 goals against per 60, 2.26 expected goals against per 60, .943 save percentage, .932 expected save percentage.

One quick note about this table: the save percentages you see here are based on all unblocked shots, or Fenwick, rather than only shots on goal — that’s why they’ll look different than the save percentages seen in their official NHL stats.

There are two ways in which a defense can help out their goaltender: by limiting the quantity of shots their goaltender sees, or by limiting the quality of shots their goaltender sees. In Grubauer’s case, the team isn’t doing him any favors in either department. Seattle surrenders 1.6 more unblocked shots per 60 minutes of play with Grubauer in net than they do with Jones in net. The disparity in shot quality allowed is even more impressive — Seattle allows just over 2 expected goals per 60 with Jones vs 2.66 with Grubauer. That’s the difference between being the second best defense in the league (behind Carolina, at 2.01 xGA/60) and being the 20th best defense in the league (between the Flyers and the Oilers). The actual goals allowed and the save percentages are nearly identical between the two Kraken netminders, but the degree of difficulty in reaching those numbers is vastly different.

...and a tale of two offenses

These two are clearly getting significantly different support out front, but at the end of the day they’re allowing pucks in the net at nearly the same rate. That means that there’s an even bigger factor at fault for the wildly different win percentages between the two goalies — goal support.

Seattle’s 3.3 goals for per 60 minutes at 5-on-5 is the best in the NHL. Boston and New Jersey are the only other teams to crack 3 GF/60 so far this year. But there’s an extremely sharp difference in this offense depending on who is in net. With Martin Jones, it’s even more dominant — 3.8 GF/60. That’s the best 5-on-5 goal support for any goaltender this year with at least 400 minutes of 5-on-5 time on ice. With Philipp Grubauer in net, that number plummets to 2.1 GF/60. Of 70 qualified goaltenders, Jones ranks 1st in 5-on-5 goal support while Grubauer ranks 60th. It’s the largest difference in goal support between a team’s top goalies this season.

The largest goal support differences between a team’s top 2 goalies. Martin Jones gets 1.74 more goals per 60 than Philipp Grubauer. Buffalo’s Ukka-Pekka Luukkonen and Craig Anderson is second at 1.16.

It is worth noting that this difference doesn’t hold up on special teams. The Kraken have scored 9 goals in 60 minutes of power play time in front of Grubauer versus 20 goals in 180 minutes in front of Jones — a slight advantage for Grubauer, but not enough to offset the 5-on-5 differences.

Putting it all together, it seems that it’s time to start considering that Grubauer actually is back to his old form. Allowing one goal against the Avalanche in a losing effort this past weekend looks like a microcosm of his season as a whole — above average performances with below average goal support. An older Félix Hernández, if you will. Here’s a look at the above-average results for the two Kraken goalies at 5-on-5, where you can see just how stark the contrast is in their goal support.

Scatterplot showing NHL goalie goal support on the x-axis versus goals allowed on the y-axis. Jones has by far the best goal support while Grubauer is below average, and both are above average in goals allowed.

Adding more context to the raw numbers, we can see the same pattern holds true in terms of goal support, but Grubauer actually comes out ahead in terms of making saves. The Kraken are scoring more than a goal per hour above expectation with Jones in net when accounting for shot quality, and more than half a goal below expectation with Grubauer in net. On defense, Grubauer is saving more goals than expected, while Jones is actually a tick below expectation so far this season.

A scatter plot showing goals above expectation on the x-axis and goals saved above expectation on the y-axis. Jones again leads the NHL in goal support, but has worse goals against numbers relative to expectation than Grubauer.

Philipp Grubauer, despite his record, is actually putting up one of the best seasons of his career right now based on these numbers. According to Evolving Hockey, he’s never had a better save percentage relative to expectation at 5-on-5 than he has this season. Including special teams situations, only his last two years in Washington have looked better than this year. It’s hard to fault Hakstol for giving Martin Jones the starting role right now, but if the Kraken start to slip at all and their offense declines for a few games, he shouldn’t hesitate to move toward an even split between his two goaltenders.

The Kraken still aren’t winning games with Philipp Grubauer in net, and that’s a problem. It’s just not Philipp Grubauer’s problem.