In the last 20 years, scouting has come a long way and draft picks are made with more research behind them than ever before, which may have led to teams feeling more sure of themselves when stepping up to the podium. However, that self-assuredness seems to apply more to drafting forwards and defencemen as opposed to goalies.
Drafting goalies has always been a tough task for teams and it is a topic that’s often debated within the hockey community. When is the right time to take a goalie? Is it acceptable to take a goalie in the first round? Do goalies picked in the higher rounds turn out to be better more often?
We are coming off the heels of Spencer Knight going 13th overall, which some GMs may believe is quite high to pick a goalie. In this article we will question if that line of thinking is justified and what teams have seen the best results from their draft picks at the position.
Drafting goalies is certainly no easy feat, because you are drafting for a position that sees it’s players peak at a later age and there is much more variance in performance throughout the career of a goalie compared to one of a position player. When we take a look at this goalie aging curve chart from Garik16, we see that goalies typically peak between roughly 25 and 28 years old. Garik used a 20 year sample size dating back to the 1996-97 season to build this chart. Repeating the exercise today for the last 20 years, may result in different results.
With that in mind, it’s difficult for teams to know that a goalie will be an important part of their team when they are drafting a teenage netminder more often than not. The highest two points in the graph also speak to the unpredictability of a goalies progression.
Over the last 20 drafts, 474 goalies have been drafted with the following age breakdown.
Almost 70% of goalie picks made by teams are used on an 18 year old, who is most likely 8 to 10 years away from their peak and although scouting has certainly progressed, it remains the hardest position to get a read on at this age.
I pulled those 474 goalies drafted from 2000-2019 using the NHL Draft Finder tool on Hockey-Reference to sort through. Let’s start with the basics and take a look at the number of goalies drafted in each round.
As you may have guessed, not many goalies are picked in the first round, and the number of goaltenders selected increases the later we get into the draft. This chart reminds us that the NHL entry draft was 9 rounds long until 2004, while giving us an idea of the risk-averse behaviour of NHL GMs. Many are of the belief that using a first or even a second rounder on a goalie is a gamble they are only willing to take on three or four goalies in a draft, but are they right in not doing so?
To try and answer this, we’ll take the original 20 year sample size and shrink it down to 15 years by only looking at draftees from 2000-2014. We are doing this, because it isn’t entirely fair to say that a pick drafted within 5 years, who hasn’t played x number of games in the NHL, isn’t a successful pick. There is no one way to measure a successful draft pick, so we will simply do an analysis on games played versus draft position.
The scatter plot shows that there is no real correlation between draft position and games played. In fact, the r-value is -0.24, which is not indicative of any correlation.
In a publication by Michael Schuckers that analyzed the value of an NHL draft pick, we learned of the following probabilities for each position to play at least 200 games.
Goalies drafted in the first and second rounds are much more likely to reach the 200 game mark, but if you aren’t going to pick one in those rounds, you might as well wait until closer to the end of the draft since probabilities show that goalies drafted in the 7th round have the same chances of playing 200 games as those drafted in the third. We recognize that it’s tougher for goalies to play 200 games compared to a position player, but what we are more concerned with is the comparison of probabilities per round.
Michael Schuckers’ chart also gives us more insight as to why there is no correlation between draft position and games played. While the probabilities for the first two rounds may lead us to believe there is, the following rounds do not, as we see rounds four and five being more successful than three and round seven being more successful than round six.
This tells us that the three to four goalies in each draft that end up being drafted in the first two rounds are picked at justifiable spots, but every goalie drafted after that, which happens to be just over 85%, may as well be picked much closer towards the end of the draft.
Which teams have seen the best results?
With all of this data pulled, we can take a look at what teams have seen better results than others when drafting netminders and how the Canucks stack up. We are trying to determine how often successful goalie picks were made by each team, again, without a standard measure of success. Due to this, we will use a couple measures which are 1 game played in the NHL and 200 games played in the NHL, similar to Garik. Since we are using the 200 game mark, we will only use the 15-year sample size spanning 2000-2014 once more for the same reasons mentioned above. Additionally, the number of goalies drafted during that time and their average draft position will be included.
There is quite a range when it comes to the number of goalies drafted by each team with Philadelphia drafting 18 and Montreal only drafting 8, however, the average draft position for a goalie to be taken may be more similar across the board than one would have guessed. The differences in the number of goalies taken between teams could be due to draft philosophies or their comfort in their goaltending depth during this time.
What really matters is the last two columns when determining which teams have seen the best results when drafting goalies. Several teams see a rate over 50% of draft picks who start at least one game at the NHL level, but only a handful reach the 20% rate for picks who start at least 200 games, which is the column that should matter more. Anaheim, Ottawa and the Winnipeg/Atlanta franchise have done quite well, while Arizona, Boston, Buffalo, Minnesota and New Jersey couldn’t have done worse.
Of course, if you don’t draft goalies well it doesn’t mean your NHL team will have poor goaltending. Teams trade for goalies or sign and develop undrafted projects. Vancouver is an example of that with their mediocre drafting at the position for these 15 years, but still icing Luongo during that time. However, it is easier to keep assets when you are able to hit more often on these picks and not trade for a starter.
While the Canucks have seemed to be doing a better job of drafting for the position, their list of goalies drafted in 2000 and on who haven’t contributed to the big club in a meaningful way is a long one. That list includes David Honzik, Jonathan Iilahti, Joe Cannata, Morgan Clark, Alexandre Vincent, Julien Ellis, Lukas Mensator, Rob McVicar, and Matt Violin. On the surface, that seems like a lot of misses, but when we get some context for the rest of the league then it doesn’t look so bad.
Canucks goalies that were not included in the sample size for the above table make up a group that fans are rightfully quite hopeful for. Demko’s development has been steady and although he’s only 20, the team has to like what they’re seeing in the growth of Dipietro. The latest additions to that group are Matthew Thiessen, drafted in the 7th round of the 2018 draft, and Arturs Silovs drafted in the 6th round this summer.
It’s a group that the team can be hopeful for, but if the numbers tell us anything, it’s that drafting 18 and 19 year old goalies can be a shot in the dark that can be more justified to take in the first two or last two rounds.