One day and two years ago, Rhys Jessop (with some input from Josh Weissbock and Cam Lawrence) wrote one of the best prospect articles I have ever read: Jake Virtanen is Good, So Don’t Draft Him. In the article Jessop discussed how fans (and ultimately the organization) were making a mistake with their willingness to pass on the better statistical player when trying to use the draft to fill team need. He also also showed the importance of statistical analysis in drafting.
There’s a lot of talk out there about the Jets needing to go after a left hand defender, and there has been some I’ve seen pushing for the big 6’7 defender Logan Stanley.
The Jets, however, would be wise to avoid Stanley at 22 and here is why.
Before I continue, I would suggest reading the Jessop piece linked above, as there is no point in me rehashing everything that’s already been said. I will note some key points though and how they apply here.
This is nothing against Stanley, despite what some of this article may make it look like. The truth is that Stanley is still a legitimate NHL prospect that would be a welcomed addition to the Jets’ prospect cupboard. The issue is the opportunity cost in there likely still being better options available at pick number 22 and Stanley’s numbers need to be considered when taking player’s ceiling, floor, and likelihood of making the NHL as a regular contributor.
I get why there is the appeal. Stanley’s size balances out Jets’ top left-hand prospect Josh Morrissey. Even without knowing anything else, the thought of a 6’7 left hand defender playing in a top-four that includes Morrissey and Jacob Trouba is quite alluring. If Morrissey, Trouba, and Stanley all hit their potential ceilings, the Jets would be a hard team to play against for years.
The biggest issue with all this is that there is the need for a conditional qualifier with “if” which means there is a chance that it could not come to pass. Now, there is always chance with any prospect that they could become great or totally bust, but the chance is not equal in each case.
…and there are many signs that Stanley won’t be the best chance available.
“What’s the problem?”
Logan Stanley does not score points.
Okay, it’s not like Stanley is completely void of offense but 17 points in 64 games, 18 in 69 when including playoffs, is not exactly something that stirs confidence in a player.
Just how low is Stanley’s scoring? Here is the 10 closest SEAL-Adjusted Scoring defenders first-time eligible for the 2016 NHL Draft:
You might notice a pattern above; almost every defender above 6′ tall is ranked, while every defender below 6′ is not. There is also a huge (punny) outlier, and that is in Logan Stanley. Apparently three inches of height is worth around 100 draft positions.
Stanley stands out as an outlier where he is ranked. Ryan Pike recently created a consensus list combining multiple public lists to show which players projected as most worth a first round selection, while also pointing out a few honourable mentions. We can isolate the defenders from this list in order and show their rankings in SEAL-Adjusted Scoring for first-time draft eligible defenders:
Again we see a pretty significant outlier with Stanley not fitting with the rest.
“The position is DEFENSEman, not OFENSEman”
Of all people, I get that there is more to the game than scoring points, even though hockey at its core is a goal scoring contest where the most goals wins.
As regular readers know, I’m a fairly strong proponent of shot metrics, like Corsi. Corsi has a relationship with outscoring but outshooters and scorers are not one-and-the-same (although there is a significant relationship). So, I’m well aware of scoring not being a “be all measure” in determining who is the best player, but scoring still matters.
Long ago Jessop dissected draft success for defensemen. He split draftees into two buckets by an arbitrary line in points per game, and then split a further three times by where players were drafted. Here is a visual displaying what Jessop found:
This graph is immensely interesting, as it tells you two things:
- Scouts can legitimately evaluate which players have the non-scoring factors that garners success.
- Scouts are overvaluing those factors relative to scoring performance.
A top-selected (picks 1-25) who was below Jessop’s arbitrary scoring line of 0.6 points per game were successful picks just as often as the above mid-selected (picks 26-50) 0.6 points per game scoring players. That’s a steep over estimation of talent.
This is why I truly believe analytics and traditional scouting processes would best be served working together, while understanding their relative strengths and weaknesses.
“This one time… This one draft pick… Chara!!”
There is always the chance that any draft pick can be exceptional, but chasing the exceptions will most likely end up with you being that person desperately praying to every deity they’ve ever heard of while they chase the big win and throw their last quarter into the slot machine… All because someone has won before.
The truth is these exceptions are exceptions because they are so rare.
Whether by Canucks Army’s pGPS model, Hayden’s DEV tool, the Projection Project, or the remnants of information I have from when I worked with Cam Lawrence and Josh Weissbock on PCS, they all paint a similar picture.
The best case scenario comparison is Tyler Myers, but for the most part players that are similar to Stanley do not make the NHL. The ones that do are rarely that great either. For the most part you are dealing with Dylan McIlrath, Eric Mestery, and Eric Gryba.
“Safer pick as he can play bottom pair”
Players like Stanley are not even the safe bet.
Everyone wants the big player that can also score. Everyone wants a Anze Koptar or a Ryan Getzlaf, but those players are rare. The interesting debate comes when you have a smaller scorer being compared to a larger non-scorer. Often people will argue that the larger player is a “safer” selection despite a lower upside, and the smaller player is a “top six or bust” player.
The thinking goes that a larger player could still fill a defensive depth role if they do not develop into a top player. More potential places on the roster is equated to more likely a NHL roster player.
This actually turns out to be false for the most part. The players who are most likely to play on the team are the ones who are also most likely to play at the top of the roster. Upside and safety is related. The best depth players are failed top-picks and/or scorers.
Not long ago, I compared players who were statistically similar to Josh Morrissey versus those similar to Keaton Ellerby. Morrissey was a strong scorer in junior selected 13th overall, although is a bit “undersized” while Ellerby was not as strong of a scorer in junior, was selected 10th overall, and is quite a bit larger.
While the conventional wisdom that players like Ellerby are more likely to fit in on a third pairing or depth role than players like Morrissey was true, the higher scoring and higher upside players like Morrissey actually bust far less often.
“What about Stanley’s non-scoring factors”
I honestly question whether Stanley has the non-scoring factors that significant elevate his ability beyond his scoring performance. Looking at Myers being Stanley’s ceiling of statistical cohorts, we notice some differences right away when you look at traditional scouting notes.
Feel free to search on Youtube the two players, even when Myers was in junior. You will see a stark difference in their ability to skate. In the new NHL, even your worst defenders need to be able to skate. Quickness is so important defensively and in transition, and defensemen are being relied on to be able to both skate and pass the puck now more than ever before.
While skating can be taught and improved to some degree, there are limitations to growth. In addition, in my work with Hockey Data, I have noted that there is a strong relationship to those who pass successfully often and those who receive assists (duh).
Despite Myers not scoring that proficiently in his draft season, scouts did recognize offensive talents. Myers was touted as a potential offensive defenseman due to his ability to shoot, handle the puck, and skate, despite not demonstrating that performance yet in the boxscore numbers.
This is one of the areas where scouting can help analytics, in trying to determine which players who are not scoring now but will likely score in the future.
Even looking beyond scoring hurts the image for Stanley.
Stanley spent a good chunk of the season with an elite talent in Mikhail Sergachev. Sergachev lead the Windsor Spitfires’ defenders in even strength primary points per game, yet Stanley was fifth on the team and just over a third of the same production.
Stanley also sits fifth on the team of the seven defenders with a -5 relative goal share (relGoal%). Goal metrics do have their issues in sampling. I should note that in a random 10 game sample (note small sample), Stanley had a 43.7 Corsi% which was one of the lower ones of the 60 defenders I had tracked for a NHL team through Hockey Data.
“So Stanley is terrible and should not be drafted?”
No. That’s not true.
As I noted before, Stanley is still a player I would welcome into the Jets’ organization. After all the flaws I have pointed out, there are still aspects to his game that makes him better than other players who will be drafted, even to the Jets. There is also the chance that Stanley could be the exception and I understand there are things some professional scouts may have seen that I have not.
However, my issue is that Stanley is not a player that I would want my team to select at 22nd overall, as there are going to be players who have garnered results that indicate higher ceilings and lower floors. Even at the Jets 36th overall pick, I’d still be skeptical.
Where would I draft Stanley? I would use a third round pick on him, but he won’t be around then and the Jets’ do not have a third anyways. If the Jets do draft Stanley at 22nd or 36th overall, I hope they make up for it by garnering plus value later in the draft. Otherwise, the Jets could end up with another Lucas Sutter on their hands.
While there are differences that separate the Sutter pick and what would happen if the Jets took Stanley, there are comparisons as well. Sutter at the time was still a legitimate NHL prospect, had a projectable frame, scored but not well, had limited upside, and was taken earlier than he should have been.
Play the odds and remember that scoring matters.