clock menu more-arrow no yes

Filed under:

Yes, Auston Matthews Really Is That Good

New, comments

Matthews scores goals like no one else in the NHL.

Buffalo Sabres v Toronto Maple Leafs
We’re used to seeing this
Photo by Claus Andersen/Getty Images

Auston Matthews is the best even-strength goalscorer in the league. Frankly, it’s not close. In a market that has overhyped its share of players, we can say without exaggeration that in this attribute, he is without peer.

The raw numbers pop out. In his rookie year, he scored 32 even strength goals, leading the league. Sidney Crosby, Jeff Skinner, and Vladimir Tarasenko tied for second with 30. Last season, Matthews scored 29 even strength goals in 62 games. Connor McDavid led the league with 35. It’s safe to say that with a full bill of health, Matthews would likely have challenged for the title again.

The numbers only get more impressive when you add further context. When you normalize for ice time, Matthews goal scoring at even strength remains at the top of the league over the past two seasons, with a staggering 1.58 goals per 60 minutes (G/60). The next highest over the same time span was Sonny Milano, who ‘only’ scored 1.36 G/60, and he just barely makes my cutoff of 500 5v5 minutes over the last two years. The difference between Matthews and Brad Marchand, who ranks 14th with 1.12 G/60 is about the same as the difference between Marchand and Nick Bonino, who scores 0.66 G/60. That puts Bonino at 148th among forwards, making him a good third liner in terms of goal scoring.

It illustrates just how much of an impact Matthews makes relative to even the best players in the league with his goal scoring. This generally holds true across all stats in hockey. The super-elite players provide more surplus value compared to good players than those good players provide when compared to depth players. Matthews’ incredible performance is made more impressive by the fact that he’s done this as a centre and at the ages of 19 and 20.

What’s incredibly unique about Matthews is exactly how he does it. Broadly speaking, there are three ways to excel as a scorer.

  1. Shoot more than average (shot generation)
  2. Shoot proportionally more from great locations / positions (shot selection)
  3. Convert more shots to goals, independent of location / positions (shot efficiency)

Matthews does all of them.

Shot Generation

By whatever metric you choose, Matthews gets off more shots than almost anyone in the league. He records 17 shot attempts per 60 minutes, a pace which puts him 21st leaguewide (league leader is Brendan Gallagher with about 21 shot attempts per 60). Stratify further to unblocked shots and he climbs up to 15th. Look at shots on goal and now he’s 14th, less than one shot on goal per 60 minutes behind the league leaders. So he gets a lot of shots away, and gets a high proportion of them past defenders, and on target. Below is a box plot of unblocked shot rates among NHL forwards over the last two years. The long horizontal line indicates Matthews’ results, as it will in further box plots in this piece.

Shot Selection

Not only does Matthews shoot a lot, but he also shoots from prime positions of the ice.

He’s certainly no perimeter player. He gets right to the middle of the ice, and shoots from where it is likely to become a goal. Corsica’s expected goals model rates his expected conversion percentage from unblocked shots to goals at 8.19%, which ranks 37th in the league. It’s somewhat unusual for a volume shooter to be near the top of the league in this statistic. The tops in this category tend to be low-shot players like Paul Byron (11.08%) and Mark Scheifele (9.57%). Guys like Tyler Bozak also succeed here, often because they’re so pass-first that only shoot when it is by far the best option to score. See the box plot below for a sense of the distribution of this statistic.

Logically, there’s a natural tradeoff between shooting a lot and shooting from great locations, analogous to the tradeoff between volume scoring and efficient scoring in basketball. It’s incredibly hard to do both. If we plot unblocked shot rates and expected shot conversions we can see that.

There is a weak negative relationship here which indicates the tradeoff discussed above. Looking at Matthews’ dot, we see there is no one who is a better shot producer than him who also has a higher expected conversion rate. Similarly, no one with a higher expected conversion rate also produces more shots than him. He and Cam Atkinson are the only players in the top 50 of both categories, and Atkinson just barely squeaks into the top 50 of expected shot conversion. Matthews is like James Harden in the NBA. He shoots a lot, and he takes efficient shots (3s and layups in the NBA, net front and slot shots in the NHL) almost exclusively.

Based on that, it’s not surprising Matthews leads the league in expected goals per 60 minutes (xG/60), as it is just the product of expected conversion rate and shot rate.

However, it’s worth noting that Matthews’ lead here is small. He records 1.15 xG/60. Brendan Gallagher is right behind him at 1.12 xG/60. If you’ll recall, I stated at the start of this piece that Matthews has a huge lead in G/60... well, that’s because of point 3 on that list above.

Shot Efficiency

Not only does Matthews shoot a lot, and not only does he primarily shoot from dangerous locations, he also converts from those locations at a rate that is far above NHL average. If we divide a player’s shooting percentage by their expected shooting percentage, we can obtain a measure for how efficiently they convert shots to goals, accounting for the types of shots they are taking.

If we do this for Matthews, we see that a shot taken by him is 37% more likely to become a goal, accounting for shot location and type. As you’d expect, this ranks quite highly around the league.

He’s no longer at the very top of the league, but he’s still above the 75th percentile, and many of those above him just barely cross the minutes played threshold. Also, it’s worth noting this stat is incredibly volatile, and I would describe it as entirely descriptive, as opposed to predictive. We could see Matthews regress in his ability to convert, although, qualitatively, he certainly has the shot and skills to maintain it. For what it’s worth, Patrik Laine, Brock Boeser, T.J. Oshie, and Nikita Kucherov are the other players at the very top of the list.

All told, this combination of high level shot generation, shot selection, and shot efficiency is essentially unmatched. From our list of players, we can select everyone who was above the 75th percentile (first line level) in unblocked shot rate, expected conversion on unblocked shots, and shot efficiency (as defined above). This takes over 400 players, and whittles it down to 2. One is Auston Matthews. The other is Michael Grabner. So I guess I was wrong... Matthews isn’t a total unicorn in terms of how well-rounded he is at all aspects of goalscoring. But I’ll take our unicorn over Arizona’s this time.

Acknowledgements

All stats are 5v5 and via Corsica. In the article, shooting percentage is always in reference to unblocked shots. In the interest of open-facing work, I am more than willing to share the dataset I used, if anyone is interested. Throughout, I am looking at forwards over the last two seasons who have played more than 500 5v5 minutes total.