The Capitals are in the midst of a rare two-day break before they begin a big two-game set against the Bruins in Bean Town on Wednesday. The break in play has provided the team a day of rest on Monday and a day for a rare practice session on Tuesday.
The break in play also provides us a good opportunity to take a look at how certain aspects of the team and players are performing. In this piece we will take a quick look at the forwards and defensemen’s performance through the scope of a few advanced metrics.
With 21 games now in the books, the Washington Capitals have completed 37.5% of their abbreviated 56-game regular season schedule. In this piece will present baseline shot and possession metrics, as well as develop a more indicative index value for each player, which will provide a more concise rating for each player’s performance through the first 21 games.
BASELINE SHOT AND POSSESSION STATS
The graph below provides the baseline shot and possession metrics after 21 games for all Capitals skaters with more than 50 minutes of time on ice so far this season. The chart includes shot attempts percentage (CF% – dark blue), ‘scoring chances for’ percentage (SCF% – light green) and ‘expected goals for’ percentage (xGF% – light blue). The red line represents the 50% demarcation point. (Click to enlarge).
Top performers for these statistics include Conor Sheary, Evgeny Kuznetsov, Jakub Vrána, Justin Schultz, Lars Eller, Nick Jensen and Richard Panik.
The previous statistic and associated graph provide a good jumping-off point for assessing player performance, but additional context is required to get a truer sense of each player’s performance.
The graph below adds the percentage of offensive zone shift starts (green) for each player. (Click to enlarge).
As we mentioned in several of last week’s posts, offensive zone start percentages (OZS%) tell a lot about a players general purpose, coach Laviolette’s intent, and in some cases, his confidence-level in a players offensive and defensive capabilities. As a result, the percentage also gives certain players a handicap or advantage when it comes to calculating offensive statistics.
For example, Evgeny Kuznetsov and Alex Ovechkin’s shot and possession metrics are drastically inflated because they essentially only get offensive zone starts (green bar), starting in the offensive zone 88.14 and 81.71 percent of the time, respectively.
ADJUSTMENT FOR OFFENSIVE ZONE STARTS
In order to address this statistical imbalance, we need to create an adjustment that accounts for players afforded more (or less) opportunities in the offensive zone. This can be ascertained by simply subtracting OZS%:
I = (CF% + xGF% + SCF%) – (OZS%)
Where ‘I’ is a new index value that accounts for the advantage/disadvantage of offensive zone starts.
If we apply the above formula to the aforementioned statistics, we can generate new values for each player:
The devolved index provides a more accurate individual rating for each player, but additional context can still be applied.
You will notice that the bottom line/checking line of Carl Hagelin, Nic Dowd and Garnet Hathaway come out on top. This is primarily due to their excellent play in the first 21 games this season, however it also indicates a very slight formulaic advantage over top-line skaters, because the reduction of offensive zone start percentages has less of an impact on their overall index rating initially developed.
What’s missing for a performance analysis? Goals and assists are not included in the index calculations derived above. We can simply expand the original formula to include two times the goals plus one times assists for each player. Let’s add those values for each player.
I + 2xGoals + 1xAssists
Advanced shot stats, subtracting any offensive zone starts advantage, and adding goals and assists will give us a better indication of how each player has performed through the first 21 games of the season.
The following graph includes the original index initially derived (orange), as well as incorporates the new values with goals and assists included (blue). (Click to enlarge).
Simplifying/clarifying the graph above, the graph below removes the original index values initially calculated and just presents the new values that include goals and assists.
The new values provide a better understanding of each players performance based on the shot and possession metrics originally presented. However, a more accurate reflection of the scoring values for each player may be obtained by incorporating the points per game the player is generating.
To refine the metric even further we can multiply the initial index by points per game (PPG) and divide by two.
(Index * (1+PPG))/2
The Capitals bottom six continues to perform well, although the top-six is starting to come alive, with the exception of Nicklas Backstrom, who has had his foot mashed on the gas pedal since the start of the season. Richard Panik is a brand new player under Laviolette’s new regime, generating offense and showing he is now well worth his salary.
The bottom line of Hathaway, Dowd and Hagelin have been solid in defending against opponents top lines, all while generating positive possession and scoring opportunities. You can’t ask for more from a fourth line.
The additions of former Pittsburgh Penguins players Conor Sheary and Justin Schultz have been a solid success to this point in the season, and are two of the team’s top performers in the first 21 games.
On the blueline, the addition of Zdeno Chara has been a significant improvement and likely assisted in the emergence of Nick Jensen, who statistically has been one of the Capitals top performers since he was scratched for three games in late January.
The first 21 games saw significant inconsistencies in the lines and defensive pairs deployed for each game, and that needs to be considered as well. Absences by key players not only hindered their start to the season, but also affected the overall lineup.
The next six games will tell us a lot about the status of this Capitals team, as they have two games with the Bruins and three games with the Flyers all in the next 11 days. Following those games will provide an ideal time to revisit player performances and how each player has evolved.
By Jon Sorensen