clock menu more-arrow no yes

Filed under:

Retiring The Blender: Finding the optimal lines for the Montreal Canadiens

New, comments

Using microstats to determine ideal configurations for the Canadiens’ current forward corps.

NHL: Montreal Canadiens at Florida Panthers Robert Mayer-USA TODAY Sports

As free agency winds down and rosters begin to take shape, discussion inevitably shifts from how a roster should be built to how it should be deployed. The Montreal Canadiens forward corps has certainly been in the news this off-season with the arrival of Jonathan Drouin, the departure of Alexander Radulov, and the continued perception that the team lacks a bona fide top-line centre.

Yet despite the turbulence, the vast majority of last year’s forwards return, including breakouts Artturi Lehkonen, Phillip Danault, and Paul Byron. These three in particular have demonstrated capabilities beyond expectations, and have complicated line assignments in the process.

Quantifying playstyle synergy

Prashanth Iyer and Dom Luszczyszyn, both of The Athletic, have taken an interesting approach to the age-old exercise of figuring out ideal lines by utilizing Ryan Stimson’s player categorization system. To briefly summarize, Stimson labels players as “balanced,” “dependent,” “playmaker,” and “shooter” by clustering them based on a combination of eight different offensive variables over the two most recent seasons.

Full methodology can be found here: https://hockey-graphs.com/2017/04/04/identifying-player-types-with-clustering/

Unsurprisingly, in terms of expected goals for per 60 minutes, it’s the specialist archetypes that generate the most offence. Stimson then took this one step further and examined which archetype combinations resulted in the most expected offence.

As Stimson himself cautions, remember to keep in mind that these numbers pertain to averaged archetypes: a third-line playmaker is not better than a first-line shooter. However, all things considered equal in terms of talent, a player with a specialized skill set will generate more offence than a jack-of-all-trades.

The Canadiens’ arsenal

What does this mean for the Montreal Canadiens? Based on Stimson’s criteria, here’s how the Habs forwards are categorized:

Detailed metrics for all players can be found here: https://public.tableau.com/profile/publish/Player_Types/Dashboard1#!/publish-confirm

With only two playmakers and (perhaps) no dependents in their top 12, the Canadiens have relatively limited permutations available if the goal of maximizing xG% is pursued. Stimson’s model dictates that playmakers should play together, and this makes sense if they can feed off each other — just because playmakers don’t shoot, doesn’t mean they can’t. However, playmakers tend to be the best players on a team, and a lot of people would prefer to “spread the wealth.”

To that end, the Habs have two configuration options for their top six: playmaker-playmaker-balanced/shooter-shooter-balanced (average xG% 55.95), and playmaker-shooter-balanced/playmaker-shooter-balanced (average xG% 55.00). Playmaker-shooter-shooter/playmaker-balanced-balanced is a relatively distant third at an average xG% of 53.65, and playmaker-playmaker-shooter/balanced-balanced-shooter comes in below that at 53.15.

The dynamic duo

Based on Stimson’s metrics, Alex Galchenyuk is the best playmaker-type player on the team, and slots in as first-line centre. With a playmaker-playmaker-balanced configuration, Jonathan Drouin slides in on Galchenyuk’s wing. It must be noted that although Drouin’s metrics are substantially lower than Galchenyuk’s, his numbers are weighed down by a relatively poor 2015-16 season. The prospect of Drouin becoming the player Galchenyuk already is should have Habs fans salivating.

The natural inclination now is to place the team’s best shooter, Max Pacioretty, on the line, and certainly there’s no statistical argument against it. While Stimson’s model says that a balanced player yields better results than a shooter, Pacioretty is no ordinary shooter. The Canadiens captain is extremely well-rounded and a top quartile player in just about every facet.

In times where the Habs need a goal, they could decide to load up with a first line of Pacioretty, Galchenyuk, and Drouin playing his off-side. However, under more normal circumstances, Pacioretty is needed to complement the more limited skillsets present on the second line. To that end, it may be best to place Artturi Lehkonen on the top line.

Lehkonen’s metrics are generally not first-line worthy, but what stands out is how the young Finn has shown above-average shot production, especially from dangerous areas. This plays right into the hands of the playmaking abilities of Galchenyuk and especially Drouin, who will draw attention and open up space for Lehkonen to exploit. Finally, it has to be noted that the Frölunda alumnus offers a defensive counterbalance allowing his linemates to focus on offensive production.

With Lehkonen-Galchenyuk-Drouin penciled in, the second line will contain two shooters and one balanced player. Pacioretty has already been assigned here, and Phillip Danault neatly fits into the 2C position. That leaves a shooter on the right wing, and Brendan Gallagher beats out Andrew Shaw and Ales Hemsky for this role.

I touched briefly on Pacioretty’s brilliance above, but I think fans generally don’t do justice to just how well-rounded the Canadiens’ captain is. Ostensibly a shooter, Pacioretty nonetheless is top quartile in every one of Stimson’s metrics, including the playmaking ones.

While this line is clearly designed to have Danault feeding Pacioretty’s shot and Gallagher looking for net-front rebounds, Pacioretty certainly has the ability to fake a shot, draw the opposition towards him, and find a wide-open teammate (paging #11) for a better scoring chance.

Speaking of the diminutive dynamo, despite his relatively underwhelming 2016-17 campaign, Gallagher still ranks quite highly on many metrics that one wouldn’t normally associate with his playstyle (passing), and somewhat low on one metric that you would (dangerous shots).

Gallagher needs to play with a little more offensive diversity. He’s made his mark by going to the net with or without the puck, but there should be more to his game than just whacking at a goalie’s pads. Hopefully Claude Julien’s system will take a little more advantage of the Edmonton native’s shot and hockey sense, and rely less exclusively on his fearlessness and grit.

The traditionalist approach

If the Canadiens are uncomfortable with the prospect of Drouin and Galchenyuk on the same line, the alternative playmaker-shooter-balanced/playmaker-shooter-balanced configuration simply sees the playmaker Drouin exchanged for the shooter Pacioretty.

Pacioretty-Galchenyuk-Lehkonen would centre around Pacioretty’s shot with Lehkonen offering defensive security and an offensive wildcard with a nose for danger spots. However, without Drouin’s creativity complementing Galchenyuk’s, this alignment creates a more static offensive system, which in turn forces the young centre to risk more as a puck-carrier in the offensive zone.

On the other hand, Drouin-Danault-Gallagher makes for an interesting combination. Here, Gallagher becomes the primary shooter with Drouin serving as the offensive playmaker. Danault acts as the safety blanket, giving Drouin the freedom to try to break defensive structures.

For this line to succeed, Gallagher will need a different tactic than just driving the goalie. Instead, he should hang around the slot area, looking for breaks in the defence and creating passing options for Drouin.

Based on his strong metrics, Gallagher clearly has the ability to play this more all-around style of game. Whether he can break the mental or coaching blocks preventing him from being multi-dimensional is something that time will tell.

Dominating through depth

For the bottom six, the Habs have many options. If Michael McCarron is in the lineup (which will be unlikely given his waiver-exempt status), then they have three shooters and three balanced players. If Peter Holland makes the cut, those numbers become two and four, respectively. If Andreas Martinsen, then it’s two, three, and one dependent. From those choices, the optimal configuration is two shooters and four balanced players arrayed as shooter-shooter-balanced/balanced-balanced-balanced lines.

From here, the exercise is fairly straightforward. Andrew Shaw and Ales Hemsky are your two best remaining shooters and Tomas Plekanec is your best remaining centre. Shaw-Plekanec-Hemsky gives you a line that can take the tough defensive assignments and not just withstand the barrage, but also has enough ability to flip the ice.

As maligned as Plekanec has become, the Czech veteran remains a very serviceable player. While his overall shot production has slipped, his ability to initiate a play, his passing, and his dangerous shot generation is still borderline first-line level. The caveat is that these numbers are produced from a two-year sample size, but even with last year’s epic slump, Plekanec’s overall numbers are still second-line level. If he shows any signs of rebounding at all, it will mean that the Habs will have a top-tier second- or even borderline first-liner playing on their third trio.

Speaking of maligned, Shaw has been an incredibly polarizing player during his first year in Montreal. Contract aside, Shaw has proven to be a decent addition, with surprising proficiency in terms of transition play (his 2015-16 season with Chicago affects that number, but since all metrics are calculated relative to team averages, the impact is mitigated). Shaw is not a passer by any means, but if he can get a play started, Plekanec can direct traffic for a Shaw or Hemsky finish.

As for the newest member of the Canadiens forward corps, Ales Hemsky has been an excellent, well-rounded, offensive hockey player over the last two years. If he can come anywhere near this level, Marc Bergevin will have struck a coup.

The fourth line can actually play

Last, but not least, we come to the fourth line. Paul Byron and Torrey Mitchell are set, while the third spot is more of a toss-up. It’s a bit of a pity that Byron, who surprisingly demonstrated first-line-level playmaking skills over the last two seasons, is on the fourth line, but it’s a testament to the depth available to Julien.

We mostly talk about Mitchell in a defensive context, but it has to be noted that his offensive metric profile is fairly similar to that of Byron. If Mitchell can elevate his game a little bit and get a little puck luck, we might be seeing another breakout season.

The candidates for the final forward spot, as touched upon earlier, include Charles Hudon, McCarron, Holland, and Martinsen. We don’t have enough NHL-level data for Hudon, but his demonstrated goal-scoring ability in the AHL would well complement the playmaking skills of both Byron and Mitchell.

McCarron is classified as a shooter by Stimson’s system, but I don’t think the hulking youngster has quite carved out an offensive niche for himself yet.

Going strictly by the data available to us, Holland represents the best option.

The question is whether the Habs are willing to sacrifice a little bit of fourth-line offence to give a youngster a proper shot to establish himself. Whether Holland, Hudon, or McCarron takes the last forward slot won’t hinder the Habs in a significant manner, but it may cost them unseen potential.

Oh, right... Martinsen.

Perhaps not really an option at all.

Play the kids together

Ultimately, by the numbers, the Canadiens have two available optimal lineup configurations.

Playing Drouin and Galchenyuk together is the more aggressive option, but it’s also the more natural option. It exploits a potential natural synergy between two creative offensive players with great instincts, and the Habs are replete with shooters and opportunists capable of taking advantage of the attention that these two dynamic youngsters will draw. It frees up Max Pacioretty from facing top defensive competition day in, day-out, and restores the Pacioretty-Danault combination that worked surprisingly well last season. With Lehkonen on one line and Danault on the other, the Habs won’t have to worry about overcommitting offensively either.

Playing the two apart is the “safe” approach, but does not represent a very significant downgrade in Stimson’s model. Pacioretty-Galchenyuk-Lehkonen is a more defensively responsible first line than the version with Drouin on it, but represents a downgrade in creativity. Drouin-Danault-Gallagher is a youthful line with lots of potential, but how the three will play together is relatively unclear.

Both of these options yield better xG% values than the Canadiens recorded in the 2016-17 season (52.9). Given that the Canadiens actually overperformed last year in terms of actual goals-for percentage (54.3), improving xG% is an important key to minimizing potential regression to the mean.

The Habs have a lot of varied pieces to play with this year. Note that none of the Habs’ top 12 forwards are dependents, whereas the much-vaunted Toronto Maple Leafs have at least two, and possibly as many as five such players.

Montreal also now has a coach that knows how to develop schemes and strategies to account for different player archetypes. Given that, Montreal definitely has the potential to generate more offence from the forwards than they did last year.

We’ll all be eagerly watching to see if they can.

(A special stick-tap to Prashanth Iyer and Dom Luszczyszyn for the inspiration, and Ryan Stimson for all of the data and graphics. 2016-17 team xGF% and GF% numbers courtesy of Corsica Hockey and visualized by Sean Tierney.)