Mark Barbour (@18sktrs, patreon.com/18skaters)
The start of the NHL season has arrived and that means it’s time for me to start my weekly articles here at Apples & Ginos. In my first few articles I’m going to look at how statistics that are relevant to fantasy hockey are distributed among NHL skaters based on age. This type of data can help us set reasonable expectations about skater performance (temper those expectations about rookies!) while also helping us identify which skaters could be about to hit their prime.
First up: the distribution of goals.
How Were Goals Distributed By Age In 2021-22?
The goals data from last season are plotted below. Each square in the plot represents a skater, and its location shows the skater’s age and the number of goals he scored. The squares will overlap when skaters who are the same age score the same number of goals. When that happens the squares get darker, and the effect is essentially a “heatmap” where the dark areas represent the highest density of skaters.
Comments on this plot:
- There were a lot of 50-goal scorers last year! This is a departure from the recent historical data (below) so it may not be reasonable to expect a repeat this season.
- Auston Matthews scored 60 goals in his age-24 season. Assuming he stays healthy, I think it’s very reasonable to expect him to score a lot of goals again this season.
- That square in the top right is Alex Ovechkin. Ovechkin is an outlier – he scored 50 goals in his age-36 season. You can see that no other skater in his mid-30s was was remotely close to Ovechkin. Is it reasonable to expect another 50-goal season from Ovechkin this year?
- More generally, you can see the faint outline of a trend where skaters in their mid-20s score the most goals. This trend will be more apparent when we look at a larger sample of historical data.
How Were Goals Distributed By Age From 2014-19?
The time period of 2014-19 represents the five most recent full-length seasons prior to COVID-19 affecting the NHL schedule. What does this historical data look like?
The heatmap effect is more prominent in this plot and shows that skaters in their mid-20s generally score the most goals.
Where did all the 50-goal scorers go? This is a five-year sample and there are only four 50-goal scorers. Actually, that’s misleading. Alex Ovechkin alone accounts for three of those 50-goal seasons (plus a 49-goal season). Ovechkin’s presence in the data creates the false impression that the best goal scorers are age 30+. Here’s what the data look like if you exclude Alex Ovechkin.
Comments on this plot:
- The drop-off in scoring as skaters age into their 30s is pronounced (notwithstanding what Alex Ovechkin is doing). The goal totals decrease, and there are simply fewer skaters over the age of 30 who are still scoring goals in the NHL.
- There will be outliers, both young and old. Do you want to bet on an outlier though? Especially a skater who is 34 or older? That’s your call (obviously) but skaters with growth potential are more likely to be 24, not 34.
- If you’re curious about the 40-goal scorers who were under 20 in this final plot, they were Patrick Laine and Auston Matthews. Patrik Laine scored 44 goals in his age-19 season (2017-18), and Auston Matthews scored 40 goals in his age-19 season (2016-17). Matthews looks like he’ll be an absolutely dominant scorer in the NHL for years to come. It will be very interesting to see what Laine can do this year while playing with Johnny Gaudreau. He’s entering his age-24 season this year, so set your expectations accordingly.
- An amusing observation about the data: Jaromir Jagr is in there! He’s another outlier, scoring 27 goals in his age-43 season (2015-16). Incredible. Can Ovechkin do something similar?
Just Show Me Some Numbers!
Maybe scatter plots that look like heatmaps aren’t your thing. Perhaps you want some simple numbers that tell you what’s going on here. OK, here’s a count of the number of skaters who scored at least 30 goals in a season, separated by age.
AGE DISTRIBUTION OF 30+ GOALS
| AGE | COUNT |
|---|---|
| 18 | 1 |
| 19 | 2 |
| 20 | 5 |
| 21 | 10 |
| 22 | 12 |
| 23 | 12 |
| 24 | 13 |
| 25 | 17 |
| 26 | 13 |
| 27 | 12 |
| 28 | 11 |
| 29 | 6 |
| 30 | 15 |
| 31 | 5 |
| 32 | 4 |
| 33 | 4 |
| 34 | 1 |
The End Of The Article
That’s all for now. I’ll be back soon with another article examining historical data.
Apples and Ginos 