I wrote a piece on Hockey-Graphs this week about how we should and should not draft goalies in the NHL and whether things will change. I wanted to apply this to the Islanders’ drafts from the Garth Snow era. For those who are unfamiliar with how analytics treats goalie drafting, and I went over how we deal with goalies in general in my last livecast:
But in sum, there are a few points to remember about goalies:
1. Goalies are HIGHLY variable over single seasons due to small sample sizes and the small differences between NHL goalies. A goalie can be great one year and then be lousy the next year -> in fact when a goalie does have a great year, we do expect him to take a clear step back the next year due to regression. With relatively few exceptions (Rask), you should never sign goalies to long term or expensive contracts, because the odds of failure are really high.
2. Goalies peak REALLY early (age 24 or thereabouts) and then decline throughout the rest of their career. This is on average, of course – some individual goalies may peak later and some will not suffer the effects of aging for a while longer than we’d expect, but for the most part, goalies decline every year and expecting a goalie over 24 to get better is making a sucker’s bet.
The above two points would seem to emphasize the importance of drafting goalie talent – after all, drafted players are the cheapest (not expensive) and will be on your team at the youngest age, near a goalie’s peak, so as to avoid the issues of goalie decline you get on the free agent market (most goalies don’t hit UFA till age 28 or even mid 30s). But this brings us to point 3:
3. Goalie prospects are far more variable than skater prospects and have a much higher rate of failure, due to the difficulty of scouting goalies and the extreme small sample size of pre-draft goalie prospects. As such, you generally should NOT draft goalies early, and probably never draft a goalie before the third round.
So yeah, this doesn’t make it easy – to a large extent, you simply have to get LUCKY to draft a good goalie. But it helps to have a good process, such that bad goalie drafting luck doesn’t result in you missing out on better skating prospects than you have to. So let’s look at Isles drafted goalies under Garth Snow:
2008 – Round 5 – 126 Overall – Kevin Poulin
–.893 SV% on 2135 shots (QMJHL)
Round 5 is basically the crapshoot area of the NHL draft. You could argue
that area starts as early as round 2 or 3, but by round 5 you’re in real
crapshoot territory. Teams treat 2nd and 3rd round picks as valuable (probably
more so than they actually are), but 5th rounders are basically treated like junk
(for rightful reasons). So using a 5th rounder on a goalie is not a bad idea.
Poulin wasn’t a great goalie prospect – not huge (6’2″), and one without a
great QMJHL track record at the time of the draft (his #s would improve the next
two years) – .893 isn’t good. But if a scout saw Poulin and thought he was
projectable, the fifth round is not a bad place to take a shot on Poulin, despite
So this is an example of the Isles drafting a goalie properly. It didn’t work
out as you’d dream, but for a fifth rounder, Poulin actually probably was an
above average return on the investment, even as bad as he was.
2009 – Round 2 – 31 Overall – Mikko Koskinen*
–.932 SV% (SM-Liiga) on 895 shots.
2009 was the year the Isles really made an attempt to find their goalie of the
future, picking two goalies early and picking them both very high. The Isles had
the #1 pick in each round, which meant a bunch of potentially impressive talent
was still available at each selection. We see that starting here with Koskinen
(picked two picks before Ryan O’Reilly, sob)
The top of the second round is WAY WAY too high for a pick on a goalie. Hell
usually the second round is, but this is the FIRST pick in the second round.
Eek. For what it’s worth, Koskinen was a high end goalie prospect to pick. The
statistics for his one season in a men’s league at the time (SM Liiga is the top
finnish league) were excellent, and he was ranked the #2 international goalie
(http://www.nhl.com/ice/page.htm?id=31414) coming in to the draft. Koskinen was
also overage (he was 20 when drafted) and the team probably expected him to
Koskinen is an example of a paradox of drafting goalies, really. On one hand,
you shouldn’t pick goalies early, because even the best goalie prospects have a
higher chance of going bust than the top skaters, or really even the skaters just
below the top level. On the other hand, the best goalie prospects (the ones who
have the least chances of going bust), are often picked as early as the second
round, meaning if you wait till the risk/reward for goalies makes sense, you wind
up with lesser prospects. For example, Koskinen was the #2 international goalie
prospect, and the #1 went 14 picks later in the second round (Robin Lehner –
sob). The Isles may not have gotten a decent goalie prospect if they waited till
later – this draft is a perfect example: the only goalie prospect who turned out
decent (so far) after round 2 was Darcy Kuemper.
This doesn’t mean that this justifies the pick. Your odds of busting are high,
even amongst guys rated that high by CSS – from 2006-2010, your top 2 euro
goalies featured two hits (Varlamov, Lehner) and 8 busts (although Enroth might
still have time). And these are the top rated guys!
2009 – Round 3 – 62 Overall – Anders Nilsson
–2.05 GAA (SV% not available) in 37 Games in Swedish Jrs. in 08-09
Taking two goalies in a draft is by itself risky – you’d LIKE to get at least 2
actual NHL players out of a draft if possible (your first rounder, and then at
least one guy from rounds 2-7), and by taking two goalies, you’re reducing your
odds at that. In this case, taking two so early REALLY reduces your odds,
especially given that the Isles had the first pick in each round.
Again, Nilsson wasn’t a bad goalie prospect – he had good size, wasn’t old (so
had time to improve) and CSS had him ranked #5 amongst European goalies. And
third round normally is about where goalie picks start being justifiable. But
again, if you’re picking so early a goalie, he better be a top tier goalie – and
those guys aren’t really available in the third most years (and weren’t this
year, either). And this is the FIRST PICK in the 3rd round – it’s just too high.
2013 – Round 3 – 70 Overall – Eamon McAdam
–.891 USHL SV% on 1666 SA in 3 years (.896 in his Draft year)
The Isles took a break from goalie selection for a few years (Cody Rosen
notwithstanding) and didn’t draft a goalie until 2013, when they again drafted
two. 70th overall is still relatively high for a goalie pick – again you can
justify it for a great goalie prospect, but they’ll rarely be available at this
point – in this draft for example, 4 of the top 5 NA Goalie prospects per CSS
were drafted prior to this point.
McAdam was ranked #6 – so he was a known goalie prospect, but not exactly super
highly rated. He also had LOUSY #s – .891 is not good in the USHL. This never
looked like a good 3rd round pick.
Particularly as the #4 NA Goalie, Calvin Petersen, was still available.
Petersen was the same height and weight roughly as McAdam and had better
#s….and PLAYED FOR THE SAME TEAM. The Isles drafted two guys with ties to
Waterloo, and didn’t draft the guy with the best #s who CSS thought was a better
goalie prospect. That’s….questionable, particularly so early. The Isles were
again drafting solely for projection here, like they did with Poulin, except they
were spending way too high of a pick to do it.
McAdam has time – and his PSU stats may be misleading as that’s a new College
hockey team which probably affected his awful #s from his draft+1 year. But
again, not the greatest process in drafting him.
2013 – Round 4 – 106 Overall – Stephon Williams*
.907 on 2009 USHL shots over 3 years
.923 on 894 NCAA shots in draft year
Williams is a more interesting pick here than McAdam. Williams was an overager –
he’d played also at Waterloo like McAdam, but also played a year of college at
Minnesota State in the WCHA. Williams’ #s are very solid – the NCAA ones are
very good and the USHL ones are okay. Now Williams wasn’t a CSS ranked prospect
– he’s not in their top 30. So one would suspect there was a higher risk than
other goalies with his #s of him collapsing. But this is the type of goalie that
a pick in the 4th round actually can make sense, particularly mid 4th round.
Now, as it turned out, Williams seemed to have flopped in his sophomore year, and
lost his starting job. That said, it was only on 232 shots, so if he gets
another shot, maybe he can turn it around. But here’s an example of not terrible
2014 – Round 3 – 78 Overall – Ilya Sorokin
.912 on 815 KHL shots in two KHL seasons.
Sorokin was AGAIN a third round draft pick – so again, we’re dealing with a black
mark in terms of process here. He also wasn’t as highly ranked a prospect as
Koskinen, but he was listed #5 amongst European goalies coming in to the draft by
CSS. More importantly, he had a .912 SV% in the KHL – the second best league in
the world, giving him some decent numbers to back up the scouts’ evaluations.
Mind you, the KHL had a freaky high average sv% last year, so .912 was still a
bit below average, but he was a 17-18 year old in a man’s league (a legit man’s
league) when he pulled that off.
Now again, this doesn’t make the pick a result of good process. You have the
Russian factor to consider – maybe he never comes over even if he IS good – and
the #1 Euro goalie was still available and was considered a legit good goalie
prospect (Ville Husso, who posted a .923 in SM Liiga at age 18). This doesn’t
mean he won’t be a legit player or better than that #1 ranked goalie – the css
scouts and other public ones aren’t always more accurate than a teams’ private
scouts! But you’re still drafting a goalie relatively early here, so it’s a bit
questionable unless it’s a goalie who might be a clear slam dunk.
2014 – Round 4 – 95 Overall – Linus Soderstrom
Again this was a double goalie draft (please stop doing that, Isles). Soderstrom
was actually ranked ahead of Sorokin by CSS (#3) and interestingly, he’s REALLY
REALLY young – he just turned 18 last month. On the other hand, we don’t really
have any reliable statistics for him – he had a .915 in under 20 Swedish play with an unknown sample size and god knows how that compares – so he’s a clear product of scouting, not stats.
And that’s okay for a 4th round pick, although a high one. I’m not opposed from using a 4th rounder every year on a guy like a Soderstrom or Williams (or Poulin), one SHOULD pan out every so often, which is all you need from goalies. It’s when you go lower that there are clear opportunity costs.
The Isles have not seemingly had good draft luck since Snow’s tenure, although the verdicts are still out on a few of these guys. That said, the process has been deficient as well – Multiple goalies per draft (3 times!), goalies drafted way too early….only three of these goalies were seemingly drafted at a reasonable position, and there’s at least one questionable pick based on the stats and scouting (McAdam).
The Isles shouldn’t stop drafting goalies – their system is barren – but should start picking em later (4th round or later) and if they want to go earlier, make damn sure their picks are slam dunks – and even then should refrain from going better than third round. Because even a seemingly good potential pick in Mikko Koskinen has been a bust so far, and analytics tells us that really isn’t that surprising.