Eight games into the pandemic-shortened 2021 NHL regular season, the Capitals sit atop the thunder-dome that is the MassMutual East Division at 5-0-3, even with major contributors Alex Ovechkin, Evgeny Kuznetsov, Dmitry Orlov, Ilya Samsonov, Tom Wilson, and Lars Eller missing from recent games. Overall, we’ve seen some analytical carry over from previous years, but the Capitals and new head coach Peter Laviolette have had success so far.
In this piece, we’ll take a look at the performance of each forward line and defensive pairing and how they’ve have faired thus far this season, as well as review overall goaltending performance. Statistics used in this piece are courtesy of Charting Hockey, Evolving Hockey, Natural Stat Trick, and Hockey Reference.
Line and Pairing Expected Goals Rates
There’s a delicate balance between generating and allowing scoring chances. Coaches will often construct lines or defensive pairings to serve a particular role, like a shutdown line or pairing to match up against opponents’ top offensive talent, or offense first lines that typically get offensive zone starts to capitalize on scoring chances.
The difference, for successful teams, is shutdown lines or pairings aren’t black holes offensively, or offense-focused lines routinely give up high-quality scoring chances against. The graphic from Charting Hockey below shows how each forward line and defensive pairing fares in terms of expected goals for versus expected goals against: (Click to enlarge)
Of course, we’ll have to take sample size into account here. Any of the lines or pairings with Ovechkin, Kuznetsov, or Orlov have had significantly less ice time than the other lines pictured above. But, directionally, we can see what lines and pairings have had success so far this season, both generating scoring chances (boosting xGF) and stifling opponents’ scoring chances (boosting xGA).
The third line that started the season, consisting of Conor Sheary, Lars Eller, and Richard Panik were extremely effective at limiting expected goals against at rates that were among the top in the NHL.
On one side of the coin, they’re placed firmly in the “dull” category because they were not generating as many high quality scoring chances. On the flip side, you have the defensive pairing of Zdeno Chara and Justin Schultz that are in the “fun” category, because they generated over three expected goals for per 60 minutes of play, but allowed just under three expected goals against per 60 minutes of play.
Ultimately, the ideal situation is to have the majority of your lines in the “good” quadrant, or at least in the top half of the y-axis. The only Capitals’ line that is in the “good” quadrant consists of Jakub Vrana, Nick Backstrom, and Tom Wilson.
The Vrana, Backstrom, Wilson line is strong because you have strong 200-foot players in Backstrom and Wilson, and Vrana who has been much improved so far in limiting expected goals against.
Expectation Versus Results
Now that we’ve covered expected results for goals for and goals against, let’s take a look at what has actually resulted on the ice. Below is another graphic from Charting Hockey that illustrates expectations versus results: (Click to enlarge)
Since the lines and pairings aren’t labeled in the graphic, here’s how each pairing and line that had at least 20 minutes of time on ice together performed:
Good
- Vrana – Backstrom – Wilson
- Hagelin – Dowd – Hathaway
- Dillon – Schultz
- Chara – Schultz
Lucky
- Vrana – Kuznetsov – Wilson
- Ovechkin – Backstrom – Wilson
Unlucky
- Sheary – Eller – Panik
- Dillon – Carlson
- Siegenthaler – van Riemsdyk
Bad
- Orlov – Carlson
- Chara – Jensen
- Ovechkin – Backstrom – Oshie
- Vrana – Backstrom – Oshie
- Panik – Eller – Oshie
Given that the Capitals have yet to lose in regulation, it’s interesting to see only six pairings and line combinations that are above the average goal differential rate of 1.33. Realistically, there are some common denominators on those six lines or pairing combinations that lead to better results. Those are Tom Wilson, Nick Backstrom, and Justin Schultz.
Backstrom is currently leading the team in points with 10 in eight games played. Wilson is scoring above a point per game currently with seven points in six games. Schultz has recently had three straight multi-point efforts, scoring all six of his points in the last three games.
There is still the caveat that the Capitals have been without top offensive talents in Ovechkin and Kuznetsov the past four games (and perhaps more).
Goaltending
After star veteran goaltender Henrik Lundqvist required a procedure on his heart prior to the start of the 2021 regular season, all signs pointed to former first round pick Ilya Samsonov taking the reins as the Caps’ starting goaltender. Then, Samsonov and his Russian comrades were required to miss at least four games due to breaking the NHL’s COVID-19 protocols, which gave way for Vitek Vanecek.
Vanecek has performed admirably in his first six NHL regular season starts. Here is his stat line, courtesy of Hockey Reference:
Vanecek is posting solid numbers for a goaltender making his first six starts in the NHL. In four of his six starts, he’s posted at least a .900 save percentage, including an extremely impressive NHL debut in Buffalo with a .968 save percentage and only one goal allowed. The strong piece here is his Goals Allowed percentage (GA%-). 100 would be exactly average, so he’s currently performing above average at 90. The lower your GA%-, the better (0 would be saving 100% of shots faced).
According to Natural Stat Trick, Vanecek currently has the sixth best save percentage during 5 on 5 play in the NHL and the sixth best goals against average amongst goalies who have played at least 180 minutes this season.
Here’s a graphic from Evolving Hockey that shows Vanecek’s even strength and shorthanded performance thus far: (Click to enlarge)
Glossary:
- xFSv%: Expected Fenwick Save Percentage; FSv%: Fenwick Save Percentage; dFSv%: Fenwick Save Percentage – xFSV%; xGAR/FA: Expected Goals Above Replacement over Fenwick Shots Against
This graphic is a way to illustrate expectations versus replacement level performance. At even strength, Vanecek is performing above replacement levels in FSv%, dFSv%, and xGAR/FA. During shorthanded situations, it’s an inverse result.
According to Natural Stat Trick, Vanecek currently has an .824 save percentage during shorthanded situations, which is 43rd in the NHL. The difference is, Vanecek has faced 10 high danger shots against during the penalty kill, tied for 8th highest in the league.
The Caps’ penalty kill is currently ranked 20th in the league, and are tied for 12th in overall penalties taken. The Capitals are also tied for second worst penalty drawn/taken differential with -11. Long story short, the Capitals need to stop taking as many penalties.
On the flip side, here’s Samsonov’s even strength and short handed performance thus far:
Samsonov has struggled so far at even strength. It’s an extremely small sample size, since he’s only had 101.73 minutes at even strength and 12.3 minutes shorthanded.
In two games this season, he’s 1-0-1, and is currently posting an .868 save percentage and a 3.36 goals against average. Here’s his stat line from this season and last season, courtesy of Hockey Reference:
Overall, he’s had the complete opposite start to the season as Vanecek. His GA%- is quite below average at 140, and based on his save percentage, is allowing two more goals above average.
Among goalies that have played at least 60 minutes of five on five play, his save percentage ranks 51st of 54. The Capitals will need much more out of Samsonov to be successful. The better teams in the league have two solid goaltenders, and with a compressed 56 game schedule this season, consistent goaltending from both goaltenders is more important than ever.
By Justin Trudel
The most interesting thing here is this.
Lucky
Ovechkin – Backstrom – Wilson
Unlucky
Sheary – Eller – Panik
Should probably change the labels to elite and pedestrian. If you do something consistently for 15 years, don’t think we can call it luck. On the other side, if stats consistently say you are very good (Eller, Panik) yet you don’t get the results one would expect, clearly something is missing. For the record, fan of both players.
Valid/fair points. Because the “expected goals” metrics utilizes historic shot success data, the ratings are more or a comparison of how they are doing in relation to their past performance. Agree, the terms need tweaking. “Lucky” and “unlucky” is a relation to how the players have performed in both ends, thus a “lucky” player or lines should have worse offensive statistics then they actually do. And on the flip side, a player or line rated “unlucky” has performed well so far, but the scoring has not materialized as of yet.