There’s now less than a month until NHL training camps open up and, and the first exhibition games will start getting played in late September. That means Auston Matthews’s debut with the Toronto Maple Leafs is getting pretty near. While Auston’s time spent playing in the World Cup may delay his first game with the Leafs by a week or so, it won’t be long until Leafs fans finally get to see him don the blue and white.
Expectations for the recent 1st overall pick are sky high, but Mike Babcock has tried to keep expectations in check by saying that Matthews will start the season on the third line. Any player’s production is going to be affected by their ice time, so just how many points the Leafs prized rookie can score is going to depend if (or, let’s be honest, when) Matthews starts moving up the lineup.
Most points projections are either based on a flat number (ex. 42 points this season) or points-per-game (0.5 PPG). But I’m not sure that’s the best approach, especially for a rookie, because ice time can be a major factor. In particular, how much of a player’s ice time is spent on the powerplay can have a big impact on their scoring totals. To that end, I’ve decided to provide a range of projections depending on how much ice time Auston gets, and what kind of split he might see between 5v5 and special teams.
First, I needed a baseline. To figure out what kind of scoring rate we might expect, I put together a list of all the centres taken either 1st or 2nd overall since 2009 who played in the NHL in the season immediately after they were drafted. I picked that year because league-wide scoring has been relatively consistent since then, so the projection will be more reliable than if I’d included data from earlier years when scoring was higher. I’ve put together an average points per 60 minutes of ice time at 5v5 and on the powerplay for this group, which you can see below.
|Player||Draft Year||Pick #||GP||P/60 (5v5)||P/60 (5v4)|
The 5v5 scoring rates are much more consistent among this group than those on the powerplay, which isn’t surprising since powerplay scoring has a lot more variation due to the much smaller amount of ice time.
The average scoring rate at 5v5 among this group is roughly equivalent to 1st/2nd line tweener. The powerplay rate is at a pretty solid 1st line level. That distinction makes sense to me. One of the things that players jumping to the NHL often say is that the biggest change is how fast the game is. On top of that, NHL players are much stronger than teenage opponents. That makes playing at even strength a major adjustment, even for most top draft picks. On the other hand, the game is slowed down considerably on the powerplay, so highly skilled young players can make the most of their abilities under less pressure and stress than even strength.
After grabbing scoring numbers, the next thing I needed was to come up with different ice time scenarios. I decided to grab the median ice time of an NHL forward playing on the 1st, 2nd, and 3rd lines at both even strength and the powerplay. Then I plugged in the scoring rates from up above, and that resulted in the following chart, which gives nine different projections for Matthews depending on the split between his 5v5 and PP ice time.
One thing that you can see is that powerplay ice time has a larger effect on the projection than even strength TOI does. While the focus on which “line” Auston plays on will be on how he’s deployed at even strength, it’s his powerplay minutes that will likely be the biggest factor in determining how much he’s able to score.
With all the numbers plugged in, I can provide a range of projections depending on how Auston Matthews’s ice time is split up. At the low end we get a projection of 36 points. That may seem low, but that projection assumes he gets very little PP time and gets pretty limited minutes at 5v5, both of which are not likely to be true over the course of 82 games. At the high end his scoring is predicted to match Jack Eichel’s 56 point point rookie season. That may sound low to some fans too, but it’s more than John Tavares had in his rookie year, and he’s gone on to become one of the league’s top scorers.
It’s also worth adding that the projections here don’t take into account the possibility of points scored at 4v4 or with the net empty, so the high end is actually probably closer to 60 points than 56. And even then, Matthews could certainly score more if everything goes right for him. My model can give a prediction, but it can’t read the future.
To close out this post, I thought it’d be interesting to compare to some other projections. Here are a few that I’ve found or have been shared with me:
NHLe is a formula used to project scoring based on the scoring of players who’ve entered the NHL from the same league as a given player. I think it’s a great tool for evaluating rookies coming from the CHL and NCAA, and less useful for players coming from Europe due to the small sample sizes, but I’m including it as a point of reference here anyway. NHLe has Matthews at 43 points.
At Sportsnet, Steve Burtch comes to the same conclusion. Using a modified form of NHLe, he has Matthews at 0.51 points per game, or 42 points over a full season.
Based on those examples, it seems like my model is on the high side (even my mid-range projection of 47 points is higher than any of the other projections), but I think the methodology holds up pretty well.
For a bit of fun, I also decided to ask my Twitter followers to vote in a poll on how many points they thought Matthews would score this season. As of the time I’m finishing up this post, 38% think he’ll score between 40 and 49 points, while 45% think he’ll score between 50 and 59. Only 17% have him outside that range. Based on my projections, that voting pattern seems pretty reasonable to me.