clock menu more-arrow no yes

Filed under:

Building A Case for Matt Duchene: A deeper look into the numbers

New, comments

Evaluating Matt Duchene's performance with quantifiable data

NHL: Ottawa Senators at Colorado Avalanche Isaiah J. Downing-USA TODAY Sports

Introduction

I find myself on the edge of a divide amongst Colorado Avalanche fans. Matt Duchene is being shopped around the league and most fans have taken a hard stance in the ‘trade him’ or ‘keep him’ camps respectfully. Initially, I fell into the ‘keep him’ camp and wouldn’t have suggested trading Duchene myself. However with such a large proportion of the fan base of the belief that moving on from Matt was the smarter move for the franchise I decided that to truly base my opinion of Duchene in fact and not just my intuition I would have to analyse the quantifiable data we have available about his career.

Duchene knows that Nathan MacKinnon is this team’s first line centre. However, the question remains of how quickly Tyson Jost will be able to take over the second line centre position. There is time for Duchene to play this role and teach younger players how to be successful at the NHL level. Teams that win the Stanley Cup have a strong first and second line centerman. Keeping Duchene would only work if he were comfortable playing the second line role or even left wing should Jost develop as we hope.

Throughout his career Duchene is a reliable second line centre if not a solid first line centre on some teams. Having played 77 games last season he has shown over his career to be a dependable player that doesn't miss much time. This stability is important for a young team learning the game at this level.

Age / Aging Curves / Comparable Players

Duchene is 26 years old and entering his eighth year in the league. He’s still a relatively young player who brings some experience to a locker room that trends younger. The team as a whole is coming off of a horrendous season and like many players, Duchene had one of the worst seasons of his career. This context is important when evaluating Duchene.

Based on player ageing curves the first significant drop in production for forwards is during a player’s 25/26-year-old season, for Duchene, this was his previous season. The next significant drop in production isn't until a forwards 28/29-year-old season. This isn't saying that Matt isn't a good player or will become significantly worse as a 28/29-year-old but that if we assume Duchene follows the average ageing trends of NHL forwards he has a couple of seasons to improve his trade value if we look past his contract status. Coming off of one of his worst seasons, there isn't much to lose by letting Duchene play and increase his trade value. Worst case scenario he could be moved to a team as a rental player at the deadline. Best case scenario he signs a bridge ‘prove it’ contract which allows performance at a discounted rate and the ability to move him for a greater return on assets. Assuming of course that Duchene would agree to those terms.

This graph by @DTMAboutHeart and @EvolvingWild shows the yearly change in wins above replacement by age from 2008-2016.

When looking at players with comparable career trajectories over the same amount of league experience Matt Duchene can be considered comparable to Paul Stastny, Mats Naslund, Vincent Damphousse, and Rod Brind’Armour.

Statistics

Looking at Duchene’s performance when compared to all other centermen from last season makes one appreciate his performance and production on a team that could not win. Duchene scored 18G, 23A for a total of 41 points. The league average for these categories amongst centermen was 9.93G, 14.96A for a total of 24.89 points. From a production standpoint, Matt is an above average centerman and a good contributor to the team’s offence. Of these 18 goals, 3 were scored on the powerplay and 1 was scored short handed showing that his goals are not being propped up by powerplay tallies, but are being driven by even strength production. Last season he also scored 3 game winning goals.

Last season he registered 160 shots on net compared to the league average for centerman of 91.86. This means he is finding the net more often than a majority of centermen and while shooting at 11.3% compared to the league average for centermen of 9.2%. We can take this to mean that Duchene is shooting more frequently and more precisely than the average centerman in the league. He took 275 shot attempts, of which 160 found the net. This means that Duchene was hitting the net on 58.2% of all shots attempted.

Matt spent 1409 minutes on the ice compared to centermen on average skating 810.73 minutes. The Avalanche lack forward depth and the first two lines were overplayed last season while the bottom six forwards were protected from quality competition by the coaching staff. Playing a smaller role may allow Duchene to focus more on his own game and flourish instead of worrying about making up for his teammate's mistakes.

This graph from hockeyviz.com shows the Avalanche’s distribution of ice time with the red line indicating league averages. Clearly the team’s top six forwards were overplayed while the bottom six did not see much ice. Defensively, only the sixth defence played noticeably less minutes than the league average. This can be understood by injury troubles on the blue line.

Point share refers to a number of points in the standings a player’s performance is worth to a team. This is close to wins above replacement in baseball but refers to total points in the standings earned and not just total wins. Using point share we can see that Duchene is worth 3.1 points in the standings to the Avalanche out of their 41 total points last season. This means that 13.23% of points the Avalanche earned in the standings last season were directly correlated to Duchene’s performance. If Duchene is moved, a return package or player that is worth a higher point share then Duchene theoretically means that the team would be better. The league average point share for centerman is 2.47 points. Obviously, an above average player would be needed to fill Duchene’s role on the team.

He blocks shots at an above average rate. Although this isn’t immensely important for a centre, it is a valued skill nonetheless. These shot blocking rates help make up for the fact that Matt isn't the best forward when it comes to shot suppression. He blocked 41 shots when the average centre only blocked 26.57. However, Duchene plays the body less often than the average centerman. This can be attributed to Duchene’s game being based on speed and playing dynamically, not being physical. Duchene throws 42 hits to a league average of 55.72 for centermen. Duchene only had 12 penalty minutes last season and has consistently drawn penalties at a greater rate than he has taken them. This is due to his overwhelming speed forcing players to hook and hold him to keep up through the neutral zone.

Faceoffs are important because puck possession is important, especially in high leverage situations. Duchene’s faceoff numbers were fantastic last season. He won 62.6% of all draws when the average centerman only wins 44.66%. Duchene is an important piece to have on the ice either as the primary or secondary faceoff man in a clutch situation late in games or on special teams. Sure this doesn’t mean much when the team isn't competing in necessarily important hockey games, but it will be should the Avalanche be successful and make a run at the playoffs.

Duchene had a CF%=51% corsi percentage in his previous season. Corsi is metric to understand puck possession, which results in shots, which results in goals. This means that during 51% of the time Duchene was on the ice the Avalanche had possession of the puck. The league average for centermen is 48.56% and most players fall between 45%-55%. Players with a possession stat above 55% are considered Elite. It would be advantageous for the Avalanche if Duchene were to improve his possession game and thereby increase Colorado’s scoring chances while decreasing their opponents’ scoring chances. Relative to other players Duchene provides the club with net 3 corsi events while the average centerman is worth -1.32 corsi events to their team. Corsi events include all shot attempts except blocked shots when the net is empty. Maximizing CF% while minimizing CA% correlates in winning hockey games. It is encouraging to note that Matt has improved his possession game in recent seasons.

Both the on ice shooting percentage and on ice save percentage are worse for the Avalanche when Duchene is on the ice. Shooting percentage falls from 7.2% to 7% when Duchene is on the ice compared to the league average of 9%. Save percentage drops from .899 to .877 when Duchene is on the ice compared to a league average of .913. This could be caused by multiple factors, many of which are out of Duchene’s control including ice time, line combinations, and line deployment. However, this can also be attributed to Duchene’s inability to suppress shots. Duchene is weak at suppressing opposing team’s shooters as we will take a look at later in this article. Another important number to look at for Duchene is PDO. A player or team’s PDO accounts for the amount of luck a player or team has had during a given time period. With a larger sample, this number will regress to the mean of 100 and is calculated using shooting and save percentages. Duchene’s PDO last season was 94.7 which makes for a very unlucky season. In hockey, luck is fairly important because puck bounces make the game rely less on skill than some other sports. The following video from Vox explains this luck vs skill paradox in greater detail.

Duchy was used in the defensive zone more frequently than the offensive zone this past season. 52.9% of his zone starts were in the defensive zone while 47.1% of his zone starts were in the defensive zone. This tells us that Duchene could not optimize his time in the offensive zone because pressure could not be sustained by all four lines in order to maintain control in the offensive zone and force faceoffs in the opponent's end. This could also be because Duchene’s faceoff statistics are so good that the coaching staff feels having him take draws in the defensive zone is to the team’s best interest defensively.

He recorded 60 takeaways and 46 giveaways. It’s good that this ratio is positive in favour of the Avalanche. This shows that he is playing two-way hockey and not forcing passes or shots, while using his speed to take the puck or passing lanes away from opponents.

The final two important numbers I’ve taken a look at are Duchene’s GF/60EV and his GA/60EV. These two numbers show the even strength goals for and against while Duchene is on the ice scaled to sixty minutes. When comparing him to the average centerman in the league Duchene is on the ice for 2.2 goals for to the league average of 2.17 for every sixty minutes of even strength hockey played. He is on the ice for 3.9 goals against compared to a league average of 2.48 goals against for every sixty minutes of even strength hockey played. Obviously, the Avalanche’s goaltending struggled last season and injuries did not help the cause. However, it remains a fact that while on the ice Duchene struggled to suppress shots and keep the puck out of the back of his net while only recording goals at a just above average pace.

Line Combinations

Here we have various line combinations that Duchene spent 50+ minutes with last season. His CF% playing with Nieto & Landeskog, MacKinnon & Rantanen, and Comeau & Rantanen were encouraging. If we remember, any CF% above 55% is considered elite so the combinations of Nieto & Landeskog, and MacKinnon & Rantanen should provide a good example of the production Duchene is capable of when surrounded by quality teammates.

When paired up with Nieto & Grigorenko he performed at the low end of average. Unfortunately, 55 minutes of Duchene’s season were spent in a pairing with Soderberg & Grigorenko which did not prove successful in the least. Any CF% below 45% is considered below average and they somehow managed a CF%=38.9%.

Hero Charts

Players scored and compared on a base 10 scale based on a three year moving average of all players and positions. From OwnThePuck

When comparing Duchene’s performance to the archetypes of first and second line centres over the previous three seasons it can be concluded that Duchene is not quite a first line centre. Although he produces goals at an above first line rate he lacks the minutes played, primary assists, shots generated, and shots suppressed to be considered an elite first line player.

When comparing Duchene to the archetype of the second line centerman we can see that he takes up second line ice time while producing goals at an above second line rate, and primary assists at a second line rate. He generates fewer shots than the average second line centerman and suppresses shots at a lesser rate than fourth line centermen whose archetype averages a score of ‘5’ for shot suppression.

Conclusion

There is a place on the Colorado Avalanche for Matt Duchene. However, it is not in his traditional role. If Duchene is to stay with the club he will have to assume the second line center role for the time being until Jost’s development can be properly evaluated. The time may come where Matt is asked to play left wing to make room up the middle for Jost. Duchene could be moved but he is coming off of his worst season in the league and the return for trading him right now will not be for his full value. Duchene could very well improve his level of play and therefore trade value as he comes towards contract negotiations and potential free agency in a couple of years. Once his value has risen he could be moved as a rental player at the trade deadline to a contender. This would likely see fair value returned for Duchene.

Otherwise he could be kept with the club provided that Matt, a lifelong Avalanche fan, would be willing to take a hometown discount on his next contract. The Avalanche shouldn’t be paying him more than his current cap hit of $6 million should he resign. Matt is at the same age that many players are when they decide to make a splash in the free agency market, and may be looking to cash out with his next contract. We can’t know for certain although he would be a valuable asset to keep around for the future of the franchise. If management does not foresee him resigning on a team friendly deal they should move him for assets that match/exceed Duchene’s offensive productivity while being more defensively sound. If management feels that Duchene is willing to return on a team friendly deal and play outside his traditional role he would be a welcome addition to the club.

Under what circumstances should Duchene stay with or leave the club? I’d love to read your opinions in the comments.

Data sources and additional reading:

http://ownthepuck.blogspot.ca/

https://hockey-graphs.com/2017/03/23/a-new-look-at-aging-curves-for-nhl-skaters-part-1/

http://hockeyviz.com/

http://datarink.com/#/home

https://www.hockey-reference.com/