Jump to content
The Official Site of the Vancouver Canucks
Canucks Community

(Rumour) Virtanen on the trade block


Recommended Posts

5 hours ago, khay said:

.965 sv% is pretty impressive, considering only 36.2% ozone starts.

 

And no to Debrusk. Not the type of top 6 winger that we need.

 

That oZone start% is a meaningless stat.

 

Out of 417 5v5 shift starts for Jake, only 50 (11.9%) were in the dZone, 40 (9.6% in the oZone... the rest (about 80% were on the fly or in the neutral zone which are discarded by this stat ).  So 2.3% more of his shifts in the dZone than oZone.

 

So he had .41 shifts a game more in the dZone than the oZone.  That represents one shift difference every 2.5 games.... that is a meaningless distinction and in no way indicates he is being used in any sort of defensive role.

 

If anyone uses oZone or dZone start % to describe how defensively a player is used, they are misleading you.  It doesn’t represent what they purport it does.  It is a fraction of a shift per game difference... and the difference between players is even a smaller fraction.

 

For comparison, Hoglander has more dZone starts than Virtanen at 55 (versus 50) so is leaned on more defensively than Jake.  Yet his oZone start % is 55%.  As with every other player in the league, the vast bulk of his shifts are in the neutral zone or on the fly, which is completely ignored by that meaningless stat.

 

Edited by Provost
  • Haha 1
  • Vintage 1
Link to comment
Share on other sites

1 hour ago, Provost said:

That oZone start% is a meaningless stat.

 

is it tho? you can pick apart any stat in isolation, but I do think this one can tell you a lot about how a player is used if its not apparent by just watching the game or you don't see a player often. 

 

I think you can see it in how rookies are used, e.g. Sergachev's 1st year he was heavily used in ozone starts and you could see it at the end of the game when  he had no shifts and he wasn't trusted yet to close out games.  But you could just learn that by watching the game too. 

 

Its like a lot of stats, they only tell you something on the top or bottom 20% and then there's a big pack of players in the middle. Corsi is the same way, its "useless" for most players but the ones at the extreme ends usually tell a story. 

 

I don't think there is a great individual stat yet of any kind for defence in particular, so going to zone starts and entry/exit info is probably the best we have for d men.

 

Edited by Jimmy McGill
  • Thanks 1
Link to comment
Share on other sites

17 minutes ago, Jimmy McGill said:

is it tho? you can pick apart any stat in isolation, but I do think this one can tell you a lot about how a player is used if its not apparent by just watching the game or you don't see a player often. 

Not really at all.  Is one shift difference in deployment every 2.5 games going to tell you something more about a player than you can see?  Does it represent how they are used?

 

You don’t have to trust me on this one, if you want to get into the weeds about it, go read the paper by a phD mathematician who is highly regarded in the analytics community.... link below.

 

If you don’t want to get in the weeds, the summary is that even if you just take the players with the most extreme oZone/dZone start % difference and correct for where they start their shifts... less than 5% of those players see a difference of more than 1% on their Corsi.  That is a tiny difference for just a few players on the extremes.  If you take the whole population of NHL players and correct for zone starts, even that tiny difference disappears into random variation.  Zone starts don’t actually impact Corsi in any statistically meaningful way, they just don’t.

 

So anyone who puts Corsi and zone starts together to make any assessment of a player is just telling a tale, there is no correlation.  Again, don’t believe me... go look for yourself.

 

Zone starts is really only useful on a team level as it shows if a team overall has relatively good or bad possession numbers and if the ice is “tilted” in their favour.  The reason it exists is because it is easy to infer from other data that is captured... mainly faceoffs.  Just because a stat exists, doesn’t mean it tells you much and certainly doesn’t mean what people wrongly infer it does.

 

https://hockeyviz.com/static/pdf/rithac.pdf

Edited by Provost
  • Thanks 1
Link to comment
Share on other sites

5 minutes ago, Provost said:

Not really at all.  Is one shift difference in deployment every 2.5 games going to tell you something more about a player than you can see?  Does it represent how they are used?

 

You don’t have to trust me on this one, if you want to get into the weeds about it, go read the paper by a phD mathematician who is highly regarded in the analytics community.... link below.

 

If you don’t want to get in the weeds, the summary is that even if you just take the players with the most extreme oZone/dZone start % difference and correct for where they start their shifts... less than 5% of those players see a difference of more than 1% on their Corsi.  That is a tiny difference for just a few players on the extremes.  If you take the whole population of NHL players and correct for zone starts, even that differ from disappears into random variation.

 

So anyone who puts Corsi and zone starts together to make any assessment of a player is just telling a tale, there is no correlation.  Again, don’t believe me... go look for yourself.

 

Zone starts is really only useful on a team level as it shows if a team overall has relatively good or bad possession numbers and if the ice is “tilted” in their favour.  The reason it exists is because it is easy to infer from other data that is captured... mainly faceoffs.  Just because a star exists, doesn’t mean it tells you much and certainly doesn’t mean what people wrongly infer it does.

 

https://hockeyviz.com/static/pdf/rithac.pdf

thanks I will dig into that.

 

The way I've been looking at it is e.g., on this site: https://www.naturalstattrick.com/playerteams.php?stdoi=oi

 

E.g, when there's a player I don't know much about, to me it tells me something about his usage when its highly skewed one way or another. But for sure most players have something roughly equivalent in all zones. Motte. e.g. doesn't, he's heavily skewed to neutral and defensive side draws. So if I wasn't familiar with Motte that would tell me something. 

 

Edited by Jimmy McGill
Link to comment
Share on other sites

9 minutes ago, Jimmy McGill said:

thanks I will dig into that.

 

The way I've been looking at it is e.g., on this site: https://www.naturalstattrick.com/playerteams.php?stdoi=oi

 

E.g, when there's a player I don't know much about, to me it tells me something about his usage when its highly skewed one way or another. But for sure most players have something roughly equivalent in all zones. Motte. e.g. doesn't, he's heavily skewed to neutral and defensive side draws. So if I wasn't familiar with Motte that would tell me something. 

 

Natural stat trick is excellent because you can see the “counts” instead of the zone starts %.  That gives you a much clearer picture of how a player is used.  You can see exactly how many shifts in all zones and on the fly... and then roughly divide them by the number of games played in your head to see how they are deployed without throwing out 80% of all their shifts and then correcting for ice time like the stat does.

 

Horvat had almost twice as many dZone starts last year as Beagle.  That shows that Horvat is leaned on more defensively.  If you look at the oZone/dZone start % stat... it gives the exact opposite impression.

Edited by Provost
Link to comment
Share on other sites

1 minute ago, Provost said:

Natural stat trick is excellent because you can see the “counts” instead of the zone starts %.  That gives you a much clearer picture of how a player is used.

 

Horvat had almost twice as many dZone starts last year as Beagle.  That shows that Horvat is leaned on more defensively.  If you look at the oZone/dZone start % stat... it gives the exact opposite impression.

yup thats a very good example. I really like this site for the line tool where you can compare how a player does with different partners, gives a bit of a window into coaching decisions too.

  • Cheers 1
Link to comment
Share on other sites

8 hours ago, khay said:

.965 sv% is pretty impressive, considering only 36.2% ozone starts.

 

And no to Debrusk. Not the type of top 6 winger that we need.

 

It is impressive.  And both those metrics give useful indicators, with limits.

It's exceptionally high - unsustainably high - but it does to a certain extent indicate how low the quality of on ice chances against have been this season.

It also needs to be kept in context - of the players he plays with.  All of Sutter, Beagle and Horvat - are very good shutdown/matchup centers - all of them handle hard minutes on a regular basis - all of them are very good faceoff guys (and in the case of Sutter and Beagle, they are also principal penalty killers).  And in each of those cases, there also tends to be a good complementary defensive winger (ie Motte, Pearson, Roussel on that line as well).

So Virtanen's defensive numbers reflect his units - moreso than an indvidual effect - but at the same time, they also reflect how well he complements those units.

 

It's also not really an outlier outcome - while it is exceptionally high in this season's small sample - he was also 3rd among forwards in on ice sv% last season (Sutter and Beagle lead the team).

Virtanen also lead all forwards in the playoffs last year, at .963.

The previous season he was second among all forwards, at .924.

It is a consistent, large sample indicator - that he complements those units defensively quite well, consistently (whereas his production on the other hand, is obviously far more streaky/confidence related).

Link to comment
Share on other sites

2 hours ago, Jimmy McGill said:

yup thats a very good example. I really like this site for the line tool where you can compare how a player does with different partners, gives a bit of a window into coaching decisions too.

you can view the counts every game on hockey reference as well.  

of course a guy like Horvat who plays twice the minutes of a Beagle is going to have twice the dzone starts - to go with twice the ozone starts - and a whack of powerplay ice time.

 

Horvat kills no penalties - at all.

4:02 a game of powerplay ice time.

0:02 on the pk.

 

Beagle is obviously the inverse.

3:10 a game of penalty killing - and obviously no pp ice time.

And 27.5% ozone starts at es.

 

So that needs to be factored in - importantly - because ozone starts are an even strength metric - (as are corsi/fenwick) - and penalty killing involves almost exclusively dzone starts, which are not reflected in those 5 on 5 metrics.  This in part feeds into the misunderstanding of the roles of players like Sutter and Beagle - who are mistakenly conceived as secondary matchup/hard minutes centers - while it is not that simple.

 

Beagle is, needless to say, leaned up far more heavily in shorthanded situations - obviously - as principally a hard minutes specialist (as is Sutter) - whereas Horvat has divided minutes - a dual, two way role (and plays far more 5 on 5  minutes than Beagle).  And, again, needless to say, no NHL team has simply one top 6 line - they all have top 6 / multiple lines that need to be shutdown - and some, particularly the more veteran and contender teams, are built with secondary scoring 3rd lines - so the idea that Horvat's line is 'the matchup line' is very oversimplifed - he's one of three lines that handle that role, and the other two handle the even more difficult defensive task of penalty killing.

 

When you adjust the reality to include the reality of special teams - and how heavily Beagle and Sutter are leaned upon, there is literally nothing misleading about looking at their 5 on 5 ozone starts - if anything those numbers actually under-represent their overall role.

Edited by oldnews
  • Cheers 3
Link to comment
Share on other sites

5 minutes ago, oldnews said:

you can view the counts every game on hockey reference as well.  

of course a guy like Horvat who plays twice the minutes of a Beagle is going to have twice the dzone starts - to go with twice the ozone starts - and a whack of powerplay ice time.

 

Horvat kills no penalties - at all.

4:02 a game of powerplay ice time.

0:02 on the pk.

 

Beagle is obviously the inverse.

3:10 a game of penalty killing - and obviously no pp ice time.

And 27.5% ozone starts at es.

 

So that needs to be factored in - importantly - because ozone starts are an even strength metric - (as are corsi/fenwick) - and penalty killing involves almost exclusively dzone starts, which are not reflected in those 5 on 5 metrics.  This in large part feeds into the misunderstanding of the roles of players like Sutter and Beagle - who are mistakenly conceived as secondary matchup/hard minutes centers - while it is not that simple.

 

Beagle is, needless to say, leaned up far more heavily in shorthanded situations - obviously - as principally a hard minutes specialist - whereas Horvat has divided minutes - a dual, two way role (and plays far more 5 on 5  minutes than Beagle).  And, again, needless to say, no NHL team has simply one top 6 line - they all have top 6 - and some, particularly the more veteran and contender teams, are built with secondary scoring 3rd lines - so the idea that Horvat's line is 'the matchup line' is very oversimplifed - he's one of three lines that handle that role, and the other two handle the even more difficult defensive task of penalty killing.

 

When you adjust the reality to include the reality of special teams - and how heavily Beagle and Sutter are leaned upon, there is literally nothing misleading about looking at their 5 on 5 ozone starts - if anything those numbers actually under-represent their overall role.

I didn't realize that - thanks. I thought that was an all situations number. 

 

Link to comment
Share on other sites

13 minutes ago, Jimmy McGill said:

I didn't realize that - thanks. I thought that was an all situations number. 

 

https://www.hockey-reference.com/teams/VAN/

You'll see under skater's advanced statistics - that corsi, fenwick, zone starts are (EV).  Percentages are not 'meaningless' - they are a clear indication of the degree to which a coach makes use of a player in particular situations/stoppages - their situational deployment/role - regardless of the volume of minutes.  That information is only useless if people could not care less about the nuances of player deployment.  Like any metric, they have a degree of utility, particularly with added context, and at the same time, the have limits.

I like the natural stat trick site as a tool as well, but their format is cumbersome - particularly on anything without a giant screen/display.

They have the option of separating 5 on 5, pp, pk, or including all situations - but for the purposes of comparing players who do not play parallel roles - ie Horvat, Sutter, and Beagle can be 'compared' 5 on 5 -  but overall, when you throw all situations in the blender, the relative analytical value is muddied, particularly when one of those guys eats large powerplay minutes, whereas the other two eat penalty killing minutes.  Separating even strength minutes - for the purposes of comparing shot attempt differentials, or scoring metrics - gives a better 'possession' and goal metric take.

 

Principal penalty killing duties = matter.

And interestingly, in the case of Virtanen - if you look at his numbers - they are excellent defensive outcomes accross the board, regardless of who he's played with.

 

 

Edited by oldnews
  • Thanks 1
Link to comment
Share on other sites

33 minutes ago, oldnews said:

https://www.hockey-reference.com/teams/VAN/

You'll see under skater's advanced statistics - that corsi, fenwick, zone starts are (EV).  Percentages are not 'meaningless' - they are a clear indication of the degree to which a coach makes use of a player in particular situations/stoppages - their situational deployment/role - regardless of the volume of minutes.  That information is only useless if people could not care less about the nuances of player deployment.  Like any metric, they have a degree of utility, particularly with added context, and at the same time, the have limits.

I like the natural stat trick site as a tool as well, but their format is cumbersome - particularly on anything without a giant screen/display.

They have the option of separating 5 on 5, pp, pk, or including all situations - but for the purposes of comparing players who do not play parallel roles - ie Horvat, Sutter, and Beagle can be 'compared' 5 on 5 -  but overall, when you throw all situations in the blender, the relative analytical value is muddied, particularly when one of those guys eats large powerplay minutes, whereas the other two eat penalty killing minutes.  Separating even strength minutes - for the purposes of comparing shot attempt differentials, or scoring metrics - gives a better 'possession' and goal metric take.

 

Principal penalty killing duties = matter.

And interestingly, in the case of Virtanen - if you look at his numbers - they are excellent defensive outcomes accross the board, regardless of who he's played with.

 

 

that's clear from the natural stat trick line tool as well, he doesn't vary much in his numbers regardless of line mate in that regard. You're right its a bit funky but once you get used to it its OK.

 

Of the two signings, Roussel is the guy thats declining too fast, and you don't need a stat site to prove it. We may have paid "too much" for Beagle, but I don't think by much, he was a guy many teams were after. 

Link to comment
Share on other sites

3 hours ago, oldnews said:

It is impressive.  And both those metrics give useful indicators, with limits.

It's exceptionally high - unsustainably high - 

That was a lot of text to explain that the stat means virtually nothing and can’t be used in the way you always use it.

 

Virtanen playing just one more shift in the defensive zone every 2.5 games than he does in the offensive zone is not unsustainably high, nor does it indicate his relative merits defensively or even how the coaching staff want to deploy him.

 

The icing stats say Jake was on the ice for 43 icings against this season.  Meaning 43 times there was a face off in the defensive zone that Green wasn’t able to pull him off the ice.  So the dZone Facebook stat is terrible too.

 

 

 

Edited by Provost
  • Haha 1
  • Vintage 1
Link to comment
Share on other sites

8 hours ago, Provost said:

 

Virtanen playing just one more shift in the defensive zone every 2.5 games than he does in the offensive zone is not unsustainably high

Look who is engaging - yet again, ironically.

 

Woosh. 

The on ice save percentage is what is unsustainable (at .965).  Ozone starts are entirely under a coaches control - nothing 'sustainable' or 'unsustainable' about deployment - it's 100% about how a coach intends to utilize a line/unit.

 

Your ridiculous take on one dzone starts -  "1 more shift every 2.5 games" - is pulled right out of the same place you got the "Jake cost the team 30 goals" fantasy metric.

 

https://www.hockey-reference.com/teams/VAN/

Schedule / results.

 

Last game vs Montreal

7 dzone starts

3 ozone

30%

 

Two games ago - vs Toronto

12 dzone starts

5 ozone

29.4%

 

Three games ago - vs Toronto

7 dzone starts

3 ozone 

30%

 

3 games

26 dzone starts

11 ozone.

 

15 more dzone starts than ozone - in the past 3 games.   Your story is 1 per 2.5.   

36.2% ozone starts on the season....

 

This is why I consider 'discussion' with you to be a waste of time. 

Aside from the walls of meaningless side chatter - there is the simply oblivion/refusal/inability to deal with actual statistics.   

Instead you shoot from the hip and make numbers up.

 

The irony is that the actual outcomes are there for anyone to look at.

 

Your misrepresentation of them - is purely about an irrational axe you have to grind with the player, the need to be 'right' about your favorite whipping boy - as opposed to reality, in spite of the objective outcomes / actual real world metrics.

 

Cue the patented wall of smoke/text, complete with;

1) 'I've been called names!'  Ad homineminemineminum! 

2) 'Why is oldnews responding'?....  to my responses.  oldnews is a liar - I never agreed to mutual ignore....but oldnews was supposed to ignore me!

3) Next comes a pedantic rant about flat earth, vaccinations, whatever random straw of the day...

4) Followed by - 'I'm right, always right,  I'm Provost alright.  I'm always right!!
"5) And the favorite go to fantasy =  "All the 'experts' agree with me!"

 

w.o.t.

 

 

 

 

 

 

Edited by oldnews
Link to comment
Share on other sites

Based on Anaheim's trade tiers Jim Benning targeting a player like Heinen who's assigned to the lowest bracket

 

Trade tiers for the Ducks

UntouchableTrevor Zegras, and Jamie Drysdale.

Pretty untouchableHampus Lindholm.

Only if he wants to leaveRyan Getzlaf, John Gibson and Ryan Miller.

UntradeableCam Fowler, Jakob Silfverberg, Adam Henrique, Kevin Shattenkirk and Derek Grant.

Time to rebuild and need max value in returnRickard Rakell and Josh Manson. Not trading Rakell for just draft picks. Manson has a year left on his deal at $4.1 million but has dealt with injuries the past few years.

Useful and might be good to hang ontoJani Hakanpaa and Nicolas Deslauriers.

There is promise and what to see where it goes – Maxime Comtois, Isac Lundestrom and Max Jones.

Don’t give up just yet unless can get a proven talentTroy Terry, Sam Steel and Josh Mahura.

Only consider if getting young proven talent – Jacob Perreault and Lukas Dostal.

See if can get anything backDanton Heinen, Ben Hutton, Jacob Larsson, Carter Rowney, Sonny Milano and Brendan Guhle.

 

Oh yes Ducks you get anything back for Heinen. This anything is a former first rounder ....

Edited by Wolfgang Durst
Link to comment
Share on other sites

1 minute ago, Wolfgang Durst said:

Based on Anaheim's trade tiers Jim Benning targeting a player like Heinen who's assigned to the lowest bracket

 

Trade tiers for the Ducks

UntouchableTrevor Zegras, and Jamie Drysdale.

Pretty untouchableHampus Lindholm.

Only if he wants to leaveRyan Getzlaf, John Gibson and Ryan Miller.

UntradeableCam Fowler, Jakob Silfverberg, Adam Henrique, Kevin Shattenkirk and Derek Grant.

Time to rebuild and need max value in returnRickard Rakell and Josh Manson. Not trading Rakell for just draft picks. Manson has a year left on his deal at $4.1 million but has dealt with injuries the past few years.

Useful and might be good to hang ontoJani Hakanpaa and Nicolas Deslauriers.

There is promise and what to see where it goes – Maxime Comtois, Isac Lundestrom and Max Jones.

Don’t give up just yet unless can get a proven talentTroy Terry, Sam Steel and Josh Mahura.

Only consider if getting young proven talent – Jacob Perreault and Lukas Dostal.

See if can get anything backDanton Heinen, Ben Hutton, Jacob Larsson, Carter Rowney, Sonny Milano and Brendan Guhle.

 

Oh yes Ducks you get anything back for Heinen. This anything is a former first rounder ....

 

do we know that it was Benning targeting Heinen, or was it ANA trying to get Jake for Heinen? the Friedman rumour wasn't actually clear on who was trying to drive the deal. 

  • Cheers 2
Link to comment
Share on other sites

11 minutes ago, Jimmy McGill said:

 

do we know that it was Benning targeting Heinen, or was it ANA trying to get Jake for Heinen? the Friedman rumour wasn't actually clear on who was trying to drive the deal. 

Who needs the details when you're Benning bashing?:o

Edited by higgyfan
  • Haha 1
Link to comment
Share on other sites

2 hours ago, Wolfgang Durst said:

Based on Anaheim's trade tiers Jim Benning targeting a player like Heinen who's assigned to the lowest bracket

 

Trade tiers for the Ducks

UntouchableTrevor Zegras, and Jamie Drysdale.

Pretty untouchableHampus Lindholm.

Only if he wants to leaveRyan Getzlaf, John Gibson and Ryan Miller.

UntradeableCam Fowler, Jakob Silfverberg, Adam Henrique, Kevin Shattenkirk and Derek Grant.

Time to rebuild and need max value in returnRickard Rakell and Josh Manson. Not trading Rakell for just draft picks. Manson has a year left on his deal at $4.1 million but has dealt with injuries the past few years.

Useful and might be good to hang ontoJani Hakanpaa and Nicolas Deslauriers.

There is promise and what to see where it goes – Maxime Comtois, Isac Lundestrom and Max Jones.

Don’t give up just yet unless can get a proven talentTroy Terry, Sam Steel and Josh Mahura.

Only consider if getting young proven talent – Jacob Perreault and Lukas Dostal.

See if can get anything backDanton Heinen, Ben Hutton, Jacob Larsson, Carter Rowney, Sonny Milano and Brendan Guhle.

 

Oh yes Ducks you get anything back for Heinen. This anything is a former first rounder ....

based on whose trade tiers?

Where did you get that list?

Fowler is "untradeable"?

Grant?

  • Haha 1
Link to comment
Share on other sites

Motte gets a fair bit of love for his hustle, energy, and defensive game, but will get get the same love once he's making more money? He's due for a raise after next season and it's unlikely he'll continue making close to the 1.225M he currently is. If he steps back in this season and plays well, and then continues to do so next season I wouldn't be surprised to see him creep up closer towards what Jake is currently making. They're different players who fill different roles, but would Motte get the same love at say.. somewhere between 1.8-2.5m per?

 

Jake played some solid minutes alongside Bo and Pearson and matched up a lot against McDavid and Draisaitl last game, clearly he's not useless. He appears to be sound defensively and his speed allows him to match up against faster players. Perhaps the issue is that many Canucks fans fixate on where he was drafted and their desire to see him fill a top six role while he appears to be a solid middle six tweener? 2.5M for a tweener who's struggling offensively but who's historically shown he's capable of providing supplementary offense isn't a tough pill to swallow. 

 

Is it frustrating because he's got the physical tools to be more than a tweener? Sure, but that doesn't mean he's worthless. It's just a matter of whether he's got a fit on this team going forward as a tweener, and I think he very well could given how his ice time is distributed. Hoglander probably slots into our top six next season if we don't retain Pearson, which given cap realities is very possible. Podz is a wildcard, but I envision management and coaching giving him every opportunity to succeed. Where Gaudette ends up fitting is anyone's guess. I'll be surprises if MacEwan goes on to be anything more than a regular fourth liner. 

Edited by Coconuts
  • Thanks 1
  • Cheers 1
Link to comment
Share on other sites

Please sign in to comment

You will be able to leave a comment after signing in



Sign In Now
  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...