The analytical Era has started: Are the Canucks ahead or Behind the curve?

[Reverend Mayhem]

are you implying that i dont have a basis for calling THoR a pile of trash because i haven't worked on something better, or just musing?

Not bashing you, that's all I'll say.

 

[RobertKron]

Meat&Potatoes/60, GRIT Rel, etc.

until you can quantify courage and character

Holy ****.

 

[Yossarian54]

dellow was probably the best overall analytics guy online and he was a lawyer

in fact, every time there's truly academic work on hockey, it's all garbage. remember THoR?

Yeah the thing about Dellow's and Tulsky's work (maybe Charron's a bit, I haven't read much of his lately) in particular was that they were trying to relate analytics data to specific on-ice events and explain or develop tactics that could alter possession numbers. Not the stereotypical 'spreadsheet watchers' that some seem to characterise analytics proponents as. In fact, the opposite, as they were watching probably more video than everybody else. This is an area where someone with a stats degree and no familiarity with hockey is probably going to be less useful.

Anyway, on topic, apart from Gilman and previously Gillis being into it, the Canucks definitely track scoring chances in some manner. Not corsi/fenwick, but AV has mentioned scoring chances for a particular player in interviews over the years. I am pretty sure Thomas Drance rounded up all the examples and found they seem to correlate reasonably well (but not exactly with) fenwick?

 

[Reign Nateo]

I'm pretty sure every team in professional sports is using advanced stats and analytics at this stage at least in some capacity.

 

[carolinacanuck]

jonathan wall

and how the canucks use statistical analysis is unique to them. i'd venture to say how each team uses stats is unique to their team. take what you want from them.

 

[Iridescent*]

Didn't analytics bring up Keith Ballard, David Booth, Marco Sturm, etc?

Sturm was a desperation depth signing. Ballard was brought in to be the primary transition defender our team lacked. Booth was supposed to be the second line winger for Kesler we lacked, and was addition by subtraction by dumping Samuelsson. I don't think any of them were analytic signings.

Malhotra to me is and will always be the poster child for our top analytic signing. He was a stat sheet darling, and it translated extremely well into on ice production.

As for the question, I think the Gillis regime was obsessed with high end analytics. I think they even had guys modifying Corsi/Fenwick to apply variables to where shots were taken from, passes through the box, etc.

 

[mrmyheadhurts]

Canucks were already ahead of the curve, what these recent hirings tell me is that there were some teams that had next to nothing in terms of analytics and are now playing catch-up. As Linden said, he believes the Canucks have one of the most cutting edge analytic departments in the league and will continue to do so under Gilman's direction. We know the Leafs were way behind in this area, all you have to do is look at their recent signings and roster decisions.

And I completely agree with Benning when he says it's much more useful at the NHL level than at the JR level. There are so many more variables like physical maturity, quality of competition, etc. That is to say, a bad scout will always be a bad scout, regardless of how much analytic info he has because the numbers mean nothing if you don't have a deep understanding of hockey and what you're actually watching on the ice.

 

[Verviticus]

analytics are a thing that are useful everywhere. they're useful in life, they're useful in posting, they're useful in driving or sports or flying planes or logistics for large companies. they are just as useful at the pro level as they are at the jr level. a dropoff in application is entirely because people like benning can't understand that to get a good result from something like analysis, you have to actually do things instead of just hoping numbers fall out of the sky in an ordered fashion

 

[Nona Di Giuseppe]

Didn't analytics bring up Keith Ballard, David Booth, Marco Sturm, etc?

Jesus... and Santorelli, Erhoff, Tanev...

... Booth still has good analytics and I think he's going to be great on a north/south team.

 

[Nona Di Giuseppe]

Canucks were already ahead of the curve, what these recent hirings tell me is that there were some teams that had next to nothing in terms of analytics and are now playing catch-up. As Linden said, he believes the Canucks have one of the most cutting edge analytic departments in the league and will continue to do so under Gilman's direction. We know the Leafs were way behind in this area, all you have to do is look at their recent signings and roster decisions.

And I completely agree with Benning when he says it's much more useful at the NHL level than at the JR level. There are so many more variables like physical maturity, quality of competition, etc. That is to say, a bad scout will always be a bad scout, regardless of how much analytic info he has because the numbers mean nothing if you don't have a deep understanding of hockey and what you're actually watching on the ice.

indeed. these numbers are supplemental and need to be used in context, but certainly can clarify and enable a team to refine things

 

[BeardyCanuck03]

Gillis and AV were ahead of the curve. Case and point, Malholtra-Sedin/Sedin zone starts. And how they "pumped up" Hodgson's trade value.

Torts wasn't as much into analytics.

I think Benning and Desjardins will use them, but not sure how much and where they would sit on the curve at this point.

 

[The Stig]

analytics are a thing that are useful everywhere. they're useful in life, they're useful in posting, they're useful in driving or sports or flying planes or logistics for large companies. they are just as useful at the pro level as they are at the jr level. a dropoff in application is entirely because people like benning can't understand that to get a good result from something like analysis, you have to actually do things instead of just hoping numbers fall out of the sky in an ordered fashion

Do you have any proof or basis that Benning doesn't understand that or are you just saying it?

 

[Ryp37]

I get the same feeling about Benning.

I personally feel like we're behind the curve when it comes to advanced analytics. Sure, we don't completely ignore them, but we're definitely under-utilizing them.

I think Toronto has taken big steps in the right direction and once it starts paying dividends for them, other teams will take notice.

I feel like Gillis started out with this progressive thinking mindset but it all fell apart after the 2011 Finals loss.

It's too bad really, Gillis had some really forward thinking ideas (sleep doctors, specialty trainers, etc) but he completely abandoned everything that made him and the Canucks a success and overreacted to the loss in 2011, ultimately making blunder after blunder.

There was a time when the Canucks were the class of the league, (progressive thinking, great reputation - we had free agents picking US as their top destination - remember Sundin?) and now I feel like we're just another member of the pack.

The Canucks have been using advanced stats for years and it shows; you have no clue if they're behind the curve or not. Teams don't care if Toronto hired a blogger, most teams already have their own guys so can't really see anyone taking note other than the bloggers fans on hf

Also I'm sure that 2 years 20 million had something to do with Sundins choice

 

[John Bender*]

Jesus... and Santorelli, Erhoff, Tanev...

... Booth still has good analytics and I think he's going to be great on a north/south team.

Jesus??? There is there solid example of horrid additions. You provided three. Jesus.

 

[Orca Smash]

Benning on adv. stats:


Sounds like he's a fan of using them for pro level scouting, but prefers more of a gut feeling for amateur scouting.

He got this/shared this with peter chiarelli, peter was using them but was skeptical how much they helped at the amatuer side and made the same comment at that analytic confrence in 2012 he attended with burke. Peter basically said they use them but to what degree of importance he placed on them seemed to be touch and go. He did comment and ask how you fairly use advanced stats for amateur scouting when the kids are playing in various leagues of varying difficulty.


However this summer (a few weeks ago) chiarelli commented that this is the way the nhl is moving and im pretty sure the bruins gave the analytic guy (Ryan Nadeau) a big promotion in the organisation and he also does scouting for the bruins.

I will say i dont disagree with chiarelli to some degree on the jr side, i would use them to supplement identifying good players later on in the draft, but using them to determine a choice between a nylander and ehlers just as an example when one was playing in the Q and one was in modo playing men makes it much more difficult.

 

[Canadian Canuck]

Didn't analytics bring up Keith Ballard, David Booth, Marco Sturm, etc?

And Ehrhoff, Higgins, Torres, Lapierre, Samuelsson, Malholtra etc.

 

[Jay Cee]

analytics are a thing that are useful everywhere. they're useful in life, they're useful in posting, they're useful in driving or sports or flying planes or logistics for large companies. they are just as useful at the pro level as they are at the jr level. a dropoff in application is entirely because people like benning can't understand that to get a good result from something like analysis, you have to actually do things instead of just hoping numbers fall out of the sky in an ordered fashion

But that's not entirely true.

Unless we are comparing apples and apples then analytics are certainly not the only god. I don't think with his comments Benning meant that he was dismissing them entirely but that it was imperfect at best to use a mostly stats based analysis across leagues in North America and Europe in evaluating talent. If you have a good staff of scouts it could well muddy the waters trying to rely too much on advanced stats.

 

[me2]

Linden specifically said on team 1040 that Gilman is big on analytics and involving "cutting-edge" tactics that he won't discuss on the team 1040. Good teams do not divulge their analytic tactics, unlike Toronto who announces these "sexy" hires.

Toronto is announcing these hires because the represent a significant change of direction that fans should be aware of given Toronto's very public problems with understanding and processing stats in the past.

 

[me2]

Jesus... and Santorelli, Erhoff, Tanev...

... Booth still has good analytics and I think he's going to be great on a north/south team.

Booth has good analytics - relatively speaking he is a poster boy but he fails to turn the numbers into results.

Ballard though I really have wonder what was going on there. His corsi numbers have never been great and are often weak. The year before we traded for him he posted a 42.3% corsi at ES. That is about as anti-advanced stats as you can get. Poster boy for someone you don't get based on advanced stats.

 

[Wilch]

Booth has good analytics.

Ballard though I really have wonder what was going on there. His corsi numbers have never been great and are often weak. The year before we traded for him he posted a 42.3% corsi at ES. That is about as anti-advanced stats as you can get.

Probably has more to do with the fact that there weren't many defensemen available at the time and the team was desperately in need of blueline depth, considering Shane O'Brien was a top 4 regular and scoring was the least of the Canucks concern back then.

 

[Yossarian54]

Booth has good analytics - relatively speaking he is a poster boy but he fails to turn the numbers into results.

Ballard though I really have wonder what was going on there. His corsi numbers have never been great and are often weak. The year before we traded for him he posted a 42.3% corsi at ES. That is about as anti-advanced stats as you can get. Poster boy for someone you don't get based on advanced stats.

If you're talking about Corsi, absolutely, he comes out ahead. He falls down in on-ice sh%, and is kind of a poster boy in a way that he can drive play towards the offensive zone and prevent shots against, but actively hinders the whole putting it in the net thing. Yet we still did better in terms of +/- per 60 with Booth than without Booth last year. Weird.

 

[biturbo19]

The real question here is...

Does being "ahead of the curve" in embracing "advanced analytics" put you "ahead of the curve" on the ice?

 

[Lonny Bohonos]

The real question here is...

Does being "ahead of the curve" in embracing "advanced analytics" put you "ahead of the curve" on the ice?

Someone needs to make a stat for that....

 

[Wilch]

Someone needs to make a stat for that....

They should invent something called the win/loss record and a point system to track that, I heard it works pretty well for other competitive sports.

 

[LeftCoast]

Benning on adv. stats:


Sounds like he's a fan of using them for pro level scouting, but prefers more of a gut feeling for amateur scouting.

There are a lot of good reasons to down play analytics for amateur scouting.

• Fewer stats are collected - even basic stuff like TOI, EV vs PP/PK, FO%, shots, etc.
• Different leagues - statistical comparisons across the CHL, Tier II Junior, USHL, and European Junior and Senior leagues are meaningless. Maybe useful to compare players within a particular league - but there are also ...
• Age disparities - in the CHL you have 16 and 20 year old players in the same team/league, like wise in the Europe where a 17 year old could be promoted to the SEL, SMLiga or KHL senior leagues and play against adults. Ice time, roles and match ups are going to vary wildly.
• Different levels of competition - some leagues are very balanced, others not so much.

I don't think you totally ignore stats for amateurs, but they are certainly weighted far lower than for pros and need to be considered in the context of the league, age and competitive context.

For Pros, as Benning says - everyone plays in the same league and the NHL has upped its game in the last 10 years in terms of stats collected. Teams also employ their own staff to fill the holes and collect their own metrics.