It’s now been a couple of days since the New York Islanders last took the ice in Pittsburgh, and the excitement among the fanbase doesn’t seem to be quieting down. Which... is good, it shouldn’t!
In a pretty stunning turn of events, the Islanders swept the Pittsburgh Penguins out of the playoffs in four games over a span of seven days. The team now awaits the winner of the series between the Washington Capitals and Carolina Hurricanes.
In the meantime, let’s go back to our series preview. Within, we outlined 10 keys to the series for both teams (though it admittedly was written in more of an Islanders lens). Now that the dust has settled, let’s go through each one and take a look at how everything shook out.
1. The Goaltending Battle
When the series began, both Robin Lehner and Matt Murray were coming in playing strong hockey. However, as the games began to unfold, it was clear that one of the two was playing at an entirely different level.
NYI-PIT— Cole Anderson (@ice_cole_data) April 8, 2019
all 3 possible goalies have been pretty good recently pic.twitter.com/qXQ9LB3SIo
Let’s break this down further. During 5v5 play, Murray faced 89 shots and stopped 79 of them (.888 save percentage). Lehner faced 112 shots and stopped 108 (.964). This carried over to high danger shots, where Murray had a .824 save percentage on 34 shots and Lehner had a .962 save percentage on 26 shots.
Put shortly, Lehner allowed a single high danger goal against during the entire series, which is nothing short of remarkable.
In totality, goals saved above average is a metric that measures individual goalies against league average goalies. Lehner is currently second in the playoffs with a +3.83 GSAA at 5v5 play. Matt Murray is second to last with -3.77 GSAA.
Both goalies were good in penalty kill situations, but once again, Lehner shined saving 12 of 13 shots, including going 4-for-4 in high danger attempts by Pittsburgh. Murray was good here too, saving 21 of 23 shots and 7 of 9 high danger shots. In fact, both goalies rank in the top-5 in playoff GSAA on the penalty kill.
These are small samples and should not be used with any real predictive context, but that’s all we have right now. And in terms of describing how the series between New York and Pittsburgh unfolded, it is clear that Robin Lehner was a clear driver for why the series ended so quickly.
2. Don’t Take Penalties At All
One of the biggest concerns from the Islanders’ point of view heading into this series was the ability to shut down Pittsburgh’s power play. With an elite cast of Sidney Crosby, Evgeni Malkin, Phil Kessel, and Kris Letang combined with the Isles’ 30th ranked expected goals against per hour (via Evolving Hockey) on the penalty kill, this was an area that was pinpointed as a potential series “x-factor.”
Turns out that it was, but not in the way conventional wisdom would have led you to believe.
Pittsburgh did a good job getting shots from the slot during their powerplay chances. Their total attempts per hour (114.93) ranked third among all playoffs teams and their scoring chance rate per hour (65.23) is currently fourth.
However, the Islanders penalty kill basically swallowed everything up before it reached the net, as Pittsburgh’s 40.38 shots-on-net per hour ranks dead last among all playoff teams. While the Isles collapsed well and protected the front of their net with mastery, the Pens still got some opportunities as their 6.00 expected goals per hour was a slightly better 12th (of 16).
But between Lehner and the Isles’ defense keeping shots near the higher portion of the slot, the Pens only scored a single powerplay goal over the four games, and with three of the four games tied or within a goal late into the third period, the Isles’ shutdown of the Pens’ stars served as a major focal point for this series to end as fast as it did.
3. Find a Way to Play Even on Special Teams
The Isles did one better here as they outscored the Penguins on the power play over the course of the series. Granted, it was by a single goal (2-1 in total), but that’s more than the original point called for.
The Isles’ power play during this series was pretty good, despite the fact they whiffed on multiple two-man advantages and initially started 0-for-5 in Game 2. During the series, New York finished with 9.22 expected goals per hour on the power play, ranking third among playoff teams.
If we look at the 5v4 heatmap, we can see pretty clearly how that was the case. The Islanders took the vast majority of their attempts below the hashmarks. They also utilized Jordan Eberle’s spot on the half-wall and Devon Toews’ spot at the point effectively as well, creating some other looks that allowed them to further accentuate their strategic direction of getting “close-in” shots. The Penguins had little answer for this, as the Isles’ 66.4 shots-on-net per hour ranks a solid third in the league.
Both of the Isles’ power play goals were scored in the home plate area of the ice, driven by their 37.53 high danger chances per hour (ranking second in the league). Combine that with how strong their penalty kill was, special teams turned into an asset for New York during the series and was a key component in their victory.
4. Protect the Front of the Net
Truthfully, this is one area that could be worked on heading into the next round.
The above heatmap shows all of the 5v5 score-venue adjusted attempts for both clubs. And as we can see, there is quite the dark hue that starts at the Isles net and basically stretches out into the high slot. Without the magic of Robin Lehner and his .962 high danger save percentage, perhaps this series ends up a lot differently. Let’s quickly show Natural Stat Trick’s map of what constitutes a high danger attempt (in blue).
The chances came from pretty much everywhere. Patric Hornqvist, who did not have a productive series, led Pittsburgh with 7 individual high danger chances at even strength. Evgeni Malkin (6) and Sidney Crosby (5) were right behind him. After that, only Garrett Wilson and Jake Guentzel (who we highlighted in our preview piece) had four or more.
In total, the Islanders allowed the sixth most high danger chances per hour of all teams through the first four games. Granted, it is really hard to continually shut down players of Crosby’s ilk, but it’s something the Islanders will have to do - especially if Washington ends up being their next opponent.
Mark this one down as an area of opportunity.
5. Power Versus Power
In the above matchup charts via Micah Blake McCurdy’s HockeyViz.com, we can see that over the course of the series the Islanders had a fairly simple philosophy at home. Head Coach Barry Trotz opted to primarily play a Martin-Cizikas-Clutterbuck-Pelech-Pulock unit against the Crosby line, which worked pretty well.
Game 1 was a lot more matchup based than Game 2, with the Nelson line moving from the Malkin assignment in Game 1 to a mix of Crosby and Malkin in Game 2. The Mathew Barzal line was largely used against a few different looks in Game 1, but ultimately were mostly on the Simon-Bjugstad-Kessel trio in the second game to great success.
Mike Sullivan must have taken notice what he saw in Game 2 because the Penguins played an extreme matchup heavy game in Game 3. Sullivan moved the Cizikas line off Crosby and onto the Malkin line, keeping the Barzal line against Kessel, and largely aiming to get the Crosby line out against the Nelson line.
However like Trotz, there was a clear adjustment in Game 4 with matchups becoming a lot flatter - in their desperation, Pittsburgh aimed less for matchups and more for “keeping Sidney Crosby and their stars out on the ice as much as possible.”
In short, we saw some similar patterns by both coaches during their first home game relative to their second. It’s clear there was a lot of tinkering going on throughout the mini home series, but the impacts itself on the series are hard to directly and accurately gauge.
6. Pittsburgh’s Defense Might Be Underrated… Except Jack Johnson
There’s a lot of talk coming out of Pittsburgh right now regarding the effectiveness of their defense, and in fairness, this one is a bit more difficult to pinpoint. There’s no question there were some simple mistakes made by the Pens over the course of the series, but there also appears to be a bit of an overreaction to a small sample size as well.
Kris Letang is an interesting case. He made some pretty obvious mistakes, such as this giveaway:
However, overall, he was actually pretty good. He led the Penguins in expected goal share and high danger chance share, as seen above. And in fact, he was the only Penguins defenseman with a high danger chance share of over 50%. The downside for Letang has to be headlined by an on-ice shooting percentage of 1.94%. In other words, the Penguins - as a team - shot under 2% while Letang is on the ice. That’s going to make any player appear worse than they are actually playing, which combined with Letang’s visual mishaps was the case here.
On the other hand, Jack Johnson was indeed as bad as advertised. The Penguins’ expected goal share while he was on the ice was an abysmal 32.21%. Johnson’s on-ice high danger chance share was an even worse 28.45%, and in real goals, the Penguins were outscored 3-0 while he was on the ice.
In short, Mike Sullivan’s quick hook of Olli Maatta to insert Jack Johnson was the best possible outcome for the Islanders. It had a very real impact on the series, and was a self-inflicted wound by the Penguins which likely caused the series to end earlier than expected.
7. Maximizing The Lineup
There’s at least some crow for me to eat here, given the outcome of the series. The focus is going to be on the decision to play former Penguin and noted two-time Stanley Cup winner Tom Kuhnhackl over Michael Dal Colle, which is something that — if you’ve read my work — I would not have recommended.
For the record, I told @habermetrics that Tom Kuhnhackl would a) have a two-goal game in this series and b) score the OT winner.— Arthur Staple (@StapeAthletic) April 11, 2019
He did both last night, tbh
Still, Kuhnhackl responded in kind playing a set of noticeable games for the Islanders while on the second line alongside Brock Nelson and Josh Bailey. Finishing with two assists in the four games, Kuhnhackl’s series was highlighted by 8 individual scoring chances, which ranked fifth among all Isles forwards. He also, less than a minute into the first game, added a goal that never actually happened due to an offsides play.
That said, it was not all roses for Kuhnhackl. As he normally does, Kuhnhackl really struggled with the possession side of the game. He was a -12.03% in shot attempts for relative, essentially meaning the Islanders were 12% worse when it comes to shot attempts while Kuhnhackl was on the ice. This stretched across multiple metrics as Kuhnhackl was also -23.17% expected goals relative, -16.42% scoring chances relative, and a -34.35% high danger chance relative.
This is almost certainly a case where the eye test does not mesh with the data. There is no question that Kuhnhackl was working hard and trying to create as much as he could, but against tougher competition, the puck was in the Isles’ end more often than not. And if not for Robin Lehner’s heroics, perhaps the scoresheet looks a lot worse than it does now.
While there is no reason to expect changes, this is still an area that the Isles may want to look to improve if the start of the next round turns in the opposite direction.
8. The Nassau Coliseum Factor
The ever popular Josh Bailey song. #Isles lead 3-1 with 8:15 left in the game pic.twitter.com/h1qWpV1dHJ— Christian Arnold (@C_Arnold01) April 13, 2019
There isn’t really a whole lot to add here. The Nassau Coliseum was loud, electric, and generally a great place to be for both games during the first round. It’s a shame the team won’t be playing there anymore this year, but it is what it is.
It’s a tough place to play in the playoffs, as many of the Pens stated, and that’s due to the amount of fan engagement and intensity that exists. The Isles obviously went 2-0 here during the series, which gave them the early series edge they needed to finish Pittsburgh off.
This is a totally intangible element, but it was palpable throughout both home games.
9. Play Physical, Not Reckless
There was a very clear motivation early on to frustrate Evgeni Malkin. With Johnny Boychuk as the primary matchup in the first game, the Islanders took a noticeable attention to play physical against Malkin, and over time it certainly worked.
They appear to be getting under Malkin’s skin which appears like an intentional strategy.— Carey Haber (@habermetrics) April 11, 2019
Malkin, to his credit, did have a strong first game on Long Island but struggled for the rest of the series. Following a game ending melee in Game 2, Malkin was held pointless at home taking a total of three shots in the two games played in Pittsburgh. Without the effectiveness of one of the Penguins’ prominent weapons, the Penguins were held to just two goals during their home games.
10. “Get The Bounces”
At the end of the day, teams don’t sweep other teams at this level without getting bounces. The Islanders finished the first round with an almost incomprehensible 107.7 PDO at 5v5 play, which is second behind the Vegas Golden Knights as highest in the league. Between timely goals and incredible goaltending (plus a few open nets by Phil Kessel and Sidney Crosby), the Islanders got every bounce they needed to dispose of one of their division rivals as quickly as they possibly could have.
Final Round 1 5v5 SVA Metrics vs. Pittsburgh (via @NatStatTrick)— Carey Haber (@habermetrics) April 17, 2019
Pens 53.89% Attempts
Isles 51.39% Scoring Chances
Isles 53.37% High Danger Chances
Isles 50.73% Expected Goals
Isles 56.43% All Situations Expected Goals
All in all, there is little question who the better team was in this series. While much of the narrative has been about “how Pittsburgh lost the series,” the Islanders objectively outplayed Pittsburgh in three of the four games (Games 2-4), and were clearly deserving of this series win. Perhaps not as quickly as it unfolded, but remove the names from the teams and it is clear which team should have the opportunity to play into early May.
That, of course, would be the New York Islanders.
All data for this piece is from NaturalStatTrick.com, unless specified that it is from HockeyViz.com or from Evolving-Hockey.com.
All team metrics are as of games played through April 17.