NHL Draft: Drafting Older Players

Mike Gillis shakes hands with goaltending prospect Joe Cannata, a 6th round pick in 2009.
Cannata is one of several "older players" selected by the Canucks over the past four seasons.

We’ve spent the past few days ruminating on the Canucks, and General Manager Mike Gillis’ "draft strategy." In particular we’ve looked at which leagues the Canucks tend to draft a disproportionate number of their players from, and whether or not "player size" is a big factor governing who the team selects. 

Today we’re going to briefly discuss player age – because the Canucks are a significant outlier among NHL teams when it comes to their willingness to select older players on draft day. This is a trend that preceded Mike Gillis, but, it’s one that has continued in earnest during his tenure.

Advertisement - Continue Reading Below

Read past the jump for more.

As Cam Charron wrote last year around this time, "there are teams that make decisions, particularly in the late rounds, based on player size and strength rather than observable skill." Obviously players who are less highly touted are scouted less rigorously and logically, this means that less is known about their skill-set or their potential development. As a result, teams often find it instructive (and necessary) to rely more on other metrics: be it age, size, or stats when making late-round picks.

An example would be that, based on their drafting record, the Canucks organization seems to spend more time and effort on scouting the QMJHL and the Scandanavian leagues (Allsvenkan, SEL, SM-Liga, J20 etc.) than they do scouting the other CHL leagues, or the US National Development Team. This "bias" or "preference" is evident throughout their drafting over the past decade, but it especially comes to the fore in later rounds. 61.5% of the time during his tenure, or eight out of thirteen times, Gillis has selected players in the fifth, sixth or seventh rounds who were playing in the team’s two most fruitful "scouting regions" (the Q or one of the Scandanavian leagues).

This is partly why, rather than looking into prospects that might be available for the Canucks with the 26th pick of the first round, we’re approaching next week’s entry draft by trying to identify what paradigms seem to influence, or even dictate the Canucks’ decision making. One of the most unique, discernible trends in Vancouver’s drafting strategy under Mike Gillis has been their willingness to draft "older" players (19 and 20 year olds). 

Advertisement - Continue Reading Below

Player age and hockey recently became a hot mainstream topic as a result of Malcolm Gladwell’s book "Outliers." In his book, Gladwell discussed the "relative-age effect," which dictates that players born in the early parts of the year tend to accrue a compounding developmental advantage over younger players born that same calender year.

Basically, when you’re six years old, being 6.8 years old and being 6.2 years old can make a pretty big difference in terms of your physical development, reaction time etc.. As a result the more physically developed kids, who are naturally seen as "the best" – get the most on-ice opportunities and the most attention from their coaches. Hypothetically, these players will continue to more highly regarded as they move up into more difficult levels of hockey. Gladwell theorized that this is partly why so many "elite" hockey players are born in January and February.

Malcolm Gladwell’s theory is a reasonably good one – in that it’s backed up by the data. As Rob Pettapiece showed last summer, players born in the first quarter of the year are over-represented at the NHL draft. This effect is especially pronounced at the OHL draft, where 44% of all players selected are born in the first quarter of any given year. While the effect is somewhat lessened for 18 year olds about to be drafted by NHL teams – the fact remains that a disproportionate number of players selected at the NHL draft (38% of all players) are still born in the first three months of the year.

What Rob’s data shows is extremely interesting for our purposes. First off, his findings indicate that older 18 year olds are overvalued at the draft. Whenever a definable group appears to be overvalued, that means that there’s probably a market inefficiency that an aware General Manager can exploit, in this case by drafting the youngest group of players on the board in later rounds.

The Canucks under Mike Gillis, however, have generally shirked this approach and while they’ve drafted two 17 year olds in the past four years (Sawyer Hannay, who the team recently allowed to re-enter the draft, and last years third rounder David Honzik) their average draft age is exceedingly high relative to other NHL clubs. That average age gets even higher in the later rounds (6th and 7th) where the average Canucks pick is well over 19 years of age. Last years seventh rounder – Frolunda defenseman Henrik Tommernes – was literally days away from turning 21 when they selected him. When I first saw the pick, and saw that he was a 1990 birthday (in contrast to most Canucks picks that year, who were born in 1993) I had to double check to make sure Tommernes was even draft eligible.

Statistically speaking, Mike Gillis’ habit of drafting older players in later rounds puts the Canucks at a slight disadvantage. Again according to Rob’s data, as a group, the youngest players drafted in later rounds outperform their draft slot by the most, and individually, produce the highest rate of over-performers. The oldest players (20+) rarely outperform their draft slot, though, they do produce individual over-performers at a comparable rate as the youngest group – and at a higher rate than any of the middle groupings.

Advertisement - Continue Reading Below

If the goals and logic behind Vancouver’s "drafting older players" strategy are that

a) older players are more ready to step into professional hockey, and

b) that you’re just as likely to hit a homerun late in a draft with a player well over 19-years old as you are with a player who just recently turned 18…

then it makes some sense. If the team, however, just doesn’t pay attention to birthdays – well, that would be wasteful.

Advertisement - Continue Reading Below

In the meantime, the Canucks’ late-round strategy of selecting late-bloomers and older players seems to have paid some dividends, so I don’t expect it to change. Consider the 2010 draft, for example: the Canucks selected two extremely young players in Jonathan Iilahti and Sawyer Hannay in the sixth and seventh rounds, and picked a 19-year old with their other sixth round pick that year: Alex Friesen. Iilahti and Hannay are going back into the draft this June, while Alex Friesen has a shiny new ELC, and will compete for a spot on the Wolves next season.

Beyond that anecdote, several of the "older players" drafted by the Canucks have come along well in their development. Certainly the team should be pleased with the progress made by the likes of overage picks Kevin Connauton, Joe Cannata, and Alexandre Grenier over the past couple of seasons.

While I somewhat question the logic of regularly drafting older players, beyond the stats it seems like all of those players have a reasonable shot at significantly outperforming their draft slots. 

Recent posts about the NHL Draft:

  • Spiel

    Very appropriate article given the way the 2012 draft unfolded for the Canucks.
    There is a post draft interview video with Gillis on the canucks website that basically says the Canucks feel that older players are more of a known commodity and finished product. The philosophy of drafting older, more finished products is right out of the Moneyball playbook of favoring college players over high schoolers.

    Is there any analysis out there that answers the question about whether older players taken in the NHL draft arrive earlier in the NHL or fair better than younger players taken in the same draft position? I suspect the Canucks have data on this.