Canucks prospect Brett McKenzie is off to a flaming hot start in the OHL. The 194th overall pick in the 2016 draft has scored seven goals in his first eight games. With five assists to go with them, he is producing at a clip of a point and a half per game.
While those numbers are quite impressive, they come with a caveat – no, not sample size, though of course that’s a relevant point. Instead, I’m referring to his age. Though the Canucks picked up McKenzie’s rights just a few months ago, he’s already playing in his draft-plus-two season, as he was taken in his second year of draft eligibility.
And so we have to ask the question: how good is good in a draft-plus-two season?
Warning! This article contains math! But also highlights.
The way that we’ve grown accustomed to determining what level of scoring is “good” relevant to age and league is by creating and comparing against historical cohorts. Following the 2015-16 season, McKenzie had a pGPS percentage of 15, with 34 of the 229 OHL forwards in his cohort playing at least 200 NHL games. In terms of expected value based on draft position, he was Vancouver’s best value selection in 2016, with a delta pGPS score of +11 percent.
As the 2016-17 junior seasons are just a few weeks old, I haven’t bothered to pull the new season’s data yet, as it’s preferable to build a larger sample of games played before running the prospect probabilities system – therefore, we won’t be using pGPS to answer this particular question.
That’s all well and good anyhow, as I’m more curious as to how the rate of success changes when production is increased and decreased, all across the spectrum of ages seen in a draft-plus-two season.
The Historical View
I always define ages based on September 15th, as that is a significant date in terms of draft eligibility. A player’s first draft eligible season is the one in which he turns 17 on or before September 15th prior to the start of the season, thus turning 18 on or before the September 15th following the draft. By this measure, a player is in his draft-plus-one season when he is 18 on or before the September 15th prior to the start of the season, and is in his draft-plus-two season when he is 19 on or before the September 15th prior to the start of the season, and so on and so on. Using these guidelines, any player in their draft-plus-two season will have an exact age beginning with 19.
Brett McKenzie was born March 12th, 1997, making him 19.511 years old on September 15th, 2016. He’s 188 centimeters tall (6-foot-2) and is currently scoring at a clip of 1.5 points per game. Those are the three factors used when comparing against historical records.
I split the age and production into bins of two months and 0.2 points per game respectively, limiting the results to players that were within 5 centimeters (approximately 2 inches) of McKenzie’s height. The sample goes from 1980-81 to 2015-16, however only players who were eligible to be drafted in 2009 or earlier are included – this gives players are least seven years to reach the 200 games mark.
These are the number of matches within the OHL:
OHL Matches, 188 +/- 5 cm
It’s no surprise that there are fewer results as the production increases – were starting with commonplace levels and heading towards the CHL elite.
Once we strip out those players that played fewer than 200 NHL games, we get the following data:
OHL Successful Matches, 188 +/- 5 cm
We see a bit of parabolic effect here, as the decreasing frequency of players is at odds with increasing likelihood of NHL success as production increases. This is easier to visualize in the following graph, which is simply the rate of success:
OHL Rate of Success, 188 +/- 5 cm
The rate of success clearly increases as point per game increases, which is no surprise – the correlation between rate of success and junior production is both well documented and intuitive – but is interesting nonetheless.
Using the data for those three charts, I created a graph that better visualizes the relationship between production and success.
Just as a note of comparison, here are the same chart for the other CHL leagues, the WHL (on the left) and the QMJHL (on the right), both again filtered by +/- 5 cm. Note the QMJHL’s lower levels of success (which is typical of the Quebec League), but still increasing with production as expected.
Also interesting is the apparent rareness of a 19-year old OHL player scoring at this rate. Within the sample, which spans approximately 40 years, only 69 forwards in McKenzie’s height range (6-foot to 6-foot-4) have scored at a rate above 1.4 era adjusted points per game in their draft-plus-two season. Nearly half of them (49 percent) went on to play at least 200 NHL games.
Back to the original question: how good is good at Brett McKenzie’s age? Well, 1.5 points per game certainly counts as good. If he were to continue to produce at that clip for the entirety of the season, you’re looking at a pretty high likelihood (somewhere in the 40-50 percent range) of getting a long term NHL player out of McKenzie, which is especially encouraging given where he was drafted.
Of course, that’s a major “if”. McKenzie’s sample is extremely small at this point, and he has a long way to go.
Expected Production for McKenzie in 2016-17
#Canucks prospect Brett McKenzie with 1G tonight – he now has 12 PTS (7-5-12) in 7 OHL games
— Ryan Biech (@ryanbiech) October 15, 2016
Last season, McKenzie put up 53 points in 62 games (somewhat of a breakout season for the Vars, Ontario native, with a 21 point increase from the season prior), which works out to 0.80 points per game played.
I created a sample of 220 OHL forwards that played at least 30 games in their draft-plus-one season between 2010-11 and 2014-15, and found their subsequent draft-plus-two seasons to compare against. Many did not return: some were promoted to the NHL, while others left for the American League – players born between September 16th and December 30th would be eligible to play in the AHL in their draft-plus-two season.
This left 163 players that played at least 30 games in their draft-plus-one and draft-plus-two season between 183 and 193 centimeters tall, which I used to generate the following graph:
The linear trend line indicates that we should expect Brett McKenzie to score about 0.94 points per game this season, his draft-plus-two year (0.9488 * 0.80 + 0.1776). Other players that scored in the vicinity of 0.80 P/GP in their draft+1 season produced at rates between 0.50 and 1.45 P/GP in their draft+2, but no player that scored less than 0.95 P/GP in their draft+1 season broke the 1.5 P/GP barrier the following year.
In fact, the only player who scored below 1.10 P/GP in their draft+1 season and broke 1.50 P/GP the season after was Dane Fox – and we all know how that turned out.
Of course, Fox had the benefit of finishing Connor McDavid’s plays, while McKenzie’s North Bay Battalion is missing a generational talent from its roster. In actuality, McKenzie is the only NHL drafted forward on his team, although there are three NHL drafted defencemen: a 68th overall pick in Cam Dineen and a pair of sixth and seventh rounders.
The Battalion are a pretty average team, sitting in third place out of five teams in their division, with a 4-5-0 record. They’ve scored 26 goals in the nine games they’ve played so far, which means McKenzie is accounting for 27 percent of North Bay’s goals so far this season. If that were to continue, it would be the best Team Goals Percentage in the OHL in at least the last ten seasons, so that’s quite lofty.
McKenzie has taken 37 shots on goal so far in the eight games he’s participated in, which gives him a shooting percentage of 16.3 percent. It’s high, but not excessively so for the OHL – certainly not high enough that we can write off his goal total to this point as pure lucky. What’s more impressive is that he’s average 5.38 shots on goal per game, which puts him seventh in the OHL in that regard. Many of the names ahead of him are much higher draft picks like Alex DeBrincat, Zachary Senyshyn and Mitchell Stephens, as well as top 2017 prospect Owen Tippett.
For those of you that are sick of reading about numbers, I put together a little highlight pack of all of McKenzie’s points to date. It includes several goals that come from banging away in front of the net, a few nice feeds for primary assists, and a couple of forced turnovers – all good signs for continuing to produce offence in the future.
If the question that we were trying to answer is, “is Brett McKenzie scoring at a good rate for a guy that’s almost 20?”, then the answer is certainly yes, especially when you take his team into consideration. If the question is, “Can he continue scoring at that pace?”, then things are a little less rosy.
Despite the evidence that he isn’t relying heavily on teammates or a sky high shooting percentage to rack up points, McKenzie’s current increase in rate of production from draft-plus-one to draft-plus-two seasons is unprecedented, especially for a seventh round pick. You’d have to think that his numbers are going to regress towards a point per game by the end of the season – heck, he’s been regressing already with just one goal and no assists in his last two games (when I conceived this article last weekend, he had been producing at a rate of 1.83 points per game).
If he can continue to produce at a level that’s even close to where he’s at now however, it’ll be nothing but good news for projections of his future as an NHLer, and good news for the Canucks. It’s very early still, but it seems as though Jim Benning and company might have found a gem in the seventh round for the second consecutive year.