Myers The 6million dollar man 2.0

[Provost]

2 minutes ago, aGENT said:

You've literally already posted a bunch of them. They're all North of $7m.

 

It's really not that complicated.

Ummm, I can only assume you are just trolling at this point?  The ones I listed aren’t all north of $7 million... half

of them are under that, and that is just the list of the very highest contracts given out to UFA D in that time period for higher end top pairing guys.

 

There are a bunch lower than Myers $6 million that were signed.  Some of them are top pairing guys, and some of them 2nd pairing in the $4-6 million range.  Heck, Tanev just signed for $4.5 million and is getting the 2nd most ice time of Calgary D.  That is 2nd pairing contract value... not $6 million.

[CptCanuck16]

On 3/26/2021 at 6:48 PM, knucklehead91 said:

Tyler Myers is in a 3-way tie for 3rd last in points, however he has technically 9 ESP which puts him at 10th on this list.... BUT that ESP does NOT include his 2 shorthanded points, which if they added that because it is a handicap, he would be tied with Doughty at  6th in

I dunno. I think comparing him to others at his pay grade is a flawed comparison.  There's market forces at play here. How desperate were the Canucks for a big D man? How desperate was Myers to get out of Winnipeg?  How good is his agent at getting deals done? All these factors (and many more) come into play here. Myers received what the market would bare at the time based on supply and demand. He's a  okay D man. Not the best, not the worst. Is he probably a little over paid? Yes. He is probably more of a $4m-5m player but he filled a need we had at the time and at that time the asking price seemed reasonable to management.  

 

Myers is the least of our worries. If management really wants to change this team around they need to start looking at the guys behind the bench, namely Green, Baumer, and Brown. 

[aGENT]

51 minutes ago, Provost said:

Ummm, I can only assume you are just trolling at this point?  The ones I listed aren’t all north of $7 million... half

of them are under that, and that is just the list of the very highest contracts given out to UFA D in that time period for higher end top pairing guys.

 

There are a bunch lower than Myers $6 million that were signed.  Some of them are top pairing guys, and some of them 2nd pairing in the $4-6 million range.  Heck, Tanev just signed for $4.5 million and is getting the 2nd most ice time of Calgary D.  That is 2nd pairing contract value... not $6 million.

Tanev is a second pairing guy and, as much as I love the guy's defensive play, puts up very little offense... while also being injured a lot. Offense pays, always has (as does staying healthy).

 

Some of that 'half' aren't first pair either or at best 2-3's. Some are guys that signed longer ago at lower cap ceilings. A few are guys that signed sweetheart deals for contenders and/or are more # 3's. 

 

Sorry you're having so much trouble with this 

 

And again, I'm not saying Myers deal is a 'bargain' AT ALL. As I said, he's paid on the high end for a second pair D. Myers on a a 'good' deal would probably be $5.5m... Anything approaching $5m would probably be pushing it into an outright bargain/major discount territory. I don't think I've seen anyone declare his deal anything close to that. It's hardly the albatross some make it out to be though. The sentiment that Seattle wouldn't take him because of his contract, for instance, is laughable.

 

It's neither a 'good' or 'bad' deal. It's a slightly high end, market value deal for a second pair D. He probably gets a bit of a bump for his size and skating ability at that size too.

 

It's not UFA, first pair money. 

[CptCanuck16]

6 hours ago, knucklehead91 said:

If zone starts and oiSV% were such terrible stats, they wouldnt bother tracking it

While I agree that they aren't terrible stats, they track all sorts of mean-nothing stats. Who has the most goals on a Tuesday? Which team has the most 4pm starts? Its the information age. Compiling meaningless stats has never been easier.

[canuck73_3]

7 hours ago, Timråfan said:

 Nope, it’s been told that Loui is a type of player that no coach in the Canucks used him for. 
so if Green doesn’t use him as he was used before the blame is either on Green or wich coach he had here or Benning that got the wrong player to his coach. 
As Xerau often point out, you don’t try to force a cube into a round hole. 

Your need to blame coaching and GMs for awful players being awful is showing.

 

Eriksson sucked here from day one and my initial thought was it was 1mil too much 1 year too long. Hindsight being 20/20 it turns out it was 6mil too much and 6 years too long. 

[knucklehead91]

9 hours ago, SID.IS.SID.ME.IS.ME said:

I have to wonder about the accuracy of the expected goals stats from hockey-reference.com (I’m assuming those are the ones being used)?

 

According to what’s been posted here:

 

Myers: 13.4 xGF, 14.3 xGA, -0.9 E+/-

 

Hedman: 11.1 xGF, 14.9 xGA, -3.8 E+/-

 

But when I look at other stats websites:

 

naturalstattrick.com (5v5):

 

Myers: 23.38 xGF, 30.38 xGA, 43.49 xGF%

Hedman: 22.29 xGF, 20.5 xGA, 52.08 xGF%

 

naturalstattrick.com (5v5 score and venue adjusted):

 

Myers: 22.98 xGF, 30.97 xGA, 42.60 xGF%

Hedman: 22.7 xGF, 20 xGA, 53.16 xGF%

 

evolving-hockey.com (5v5)

 

Myers: 23.66 xGF, 32.71 xGA, 41.97 xGF%

Hedman: 24.06 xGF, 21.19 xGA, 53.17 xGF%

 

evolving-hockey.com (5v5 score and venue adjusted):

Myers: 23.46 xGF, 33.23 xGA, 41.38 xGF%
Hedman: 24.41 xGF, 20.87 xGA, 53.91 xGF%

 

And even the raw counts (GF/GA) from NHL.com:

 

Myers: 26 On-ice EV GF, 34 On-ice EV GA

Hedman: 37 On-ice EV GF, 23 On-ice EV GA

 

(Worth noting that EV stats include empty net situations, and TBL leads the league by scoring 11 EN goals, of which I’m sure Hedman was on-ice for quite a few. Could probably also say the reverse, when it comes to Myers and his EV GA.)

 

Those xG numbers from hockey-reference.com just look so wonky, compared to everything else I’ve looked at.

 

That is interesting for sure. Perhaps there is an issue with some of hockey references calculations.

 What is consistent is the xGF Myers is on par with Hedman 5v5 in all your samples from other sites 3/5 calculations have Myers with a higher xGF than Hedman.

 Which Myers is on a weaker team compared to Hedman who is on a powerhouse. So 5v5 Myers is contributing the same as Hedman 5v5.  
 

According to Hockey-reference:

Myers

740gp 

564 ES goals against

(786 goals against total)

Hedman

795gp 

648 ES goals against

(880 goals against total)

 

Hedman has been on the ice for alottttta goals.... on one of the best teams for the last decade..

 

Myers has maintained a career oiSV% of 91.8 over the course of his career, never going below 90.1 and has had 6/13 seasons of above 92.0 oiSV.
We all know Myers has played on some really sh*tty Sabres teams

 

Hedman has maintained a career oiSV% of 90.8 over the course of his career, 4 of which were below 90.0 2 of which were 88.7oiSV

Once again....Hedman has been on one of the best teams for the last decade...

Now some posters disagree with the oiSV% stat and say that Myers oiSV% is a result of good goaltending. Either Myers has a horseshoe stuck up is arse  and Hedman broke a mirror and has 7 years badluck..... or the consistent oiSV% is an indication of a players defensive ability. @Provost you may have scoffed at the idea of Myers being in the same stratosphere as Hedman.... but perhaps Hedman is a product of his environment? He plays on the best team in the league who has an unbelievably talented roster, yet when he is in his own zone, for some reason his goalies sV% dips. But when McDonagh is on the ice (and has a 57% dZs) his goalies sV% increases.

 

Sure you can talk about high danger chances, but thats 5 guys on the ice that let that chance happen.

 

Zone starts and oiSV% go hand in hand. EVEN IF ITS JUST 1 MORE FACEOFF IN THE DZONE PER GAME. 1 lost faceoff in your own zone, is a game tying goal or a game winning goal. Each faceoff is crucial and there are roughly 50-70 whistles in a game

 

Zone starts + oiSV are an indication of an individual's influence

Higher dZs = trust

High oiSV% = doing their job, making sure shots are blocked, passing lanes are taken away, or making sure the goalie can see the puck to make the stop.

 

Lower dZs = not trusted

Lower oiSV% = liability

 

Hedmans first 6 seasons on a pretty good Tampa team he averaged 30.33pts (22.5pts through his first 4 seasons) and an 89.0 oiSV%.

That pretty good tampa team has turned into an unbelievably good Tampa team and yes Hedmans points have increased tenfold, but he has also racked up a TON of PPP. But if he was as good as McDonagh in his own zone, his oiSV% should have increased dramatically, however it has barely improved to a career average of 90.8. Which his high oZs% indicates why Hedman is not deployed in his own zone.

Take tampa out of the equation, how much better would Hedman make Buffalo?

 

Lets say Hedman has 15 oZs and 12dZs 20% higher oZs deployment in a game. And Myers has 10 oZs and 14dZs for roughly 30% more dZs. (This is from a faceoff)

Without a faceoff, when either player is not on the ice and bottom pairing D are on the D-side and the play transitions up the ice and into the oZone, who do you think is going to be going onto the ice as the bottom dpair jumps off? 
Hedman for sure will be going on the fly into the oZone for Tampa. 
For Vancouver it will be Hughes. 
Hughes and Hedman are going to be more effective in the oZone than the dZone, so even though there may be only a couple different “zone starts” i.e 2-3, the % shows the reliability and where that reliability is. On the fly changes will increase the zone designation.

 

Also, out of curiousity... do high danger chances take into account PP expiration? Because if that isnt factored into the equation... Myers 12 point whatever to Hedmans 9point something high danger chances allowed isnt a very fair stat to focus on, because Myers averages 3:17 of PK whereas Hedman averages 30seconds less, and averages 2:43PP time more than Myers, which after an expired penalty and some tired PKers on the ice, Hedmans 5v5 stats are going to get boosted. 
 

Anyhoo...

 

 

Edited March 28, 2021 by knucklehead91

[SID.IS.SID.ME.IS.ME]

@oldnews

 

I’m not going to quote your whole post. It’s very late, and I’ve had a really long day, so I might do a full quote and response another day, but for now, I’m just going to jot down a few ideas, and save a little board space by avoiding the big quote and reply.

 

(Although, knowing me, this post will probably ramble on anyway, so I’m probably fooling myself if I think I’m going to save much space.)

 

You’re right that there are two types of “zone starts.”

 

I tend to think of the faceoff based ones as “traditional zone starts” because the stat initially just looked at faceoffs. The “shift start” type used to get referred to as “true zone starts” and more recently, it’s kind of become convention to refer to “shift starts” just as zone starts, and the older, “traditional zone starts” as “faceoff percentages” (as in DZF%, NZF%, OZF%).

 

But I agree that it can be confusing. And I probably should have been more clear about the stat I was using, in that earlier post.

 

I disagree about the older stat being more meaningful, when it come to analyzing players, and making adjustments to metrics, based on deployment.

 

Rather than write an essay here, I’ll just add a couple links that do a pretty good job explaining my own views on the two stats, and the overall impact of zone starts, when it comes to impacts/adjustments to Corsi, etc:

 

https://puckplusplus.com/2015/01/15/how-much-do-zone-starts-matter-i-maybe-not-as-much-as-we-thought/
 

https://puckplusplus.com/2015/01/20/how-much-do-zone-starts-matter-part-ii-a-lot-on-their-own-not-that-much-in-aggregate/
 

(These are blog posts from Matt Cane, B.Eng, M.Sc Applied Mathematics, who’s done some good work on this topic,)

 

I do think both stats have value, and help us to get a better picture of what type of minuets players play, and how coaches attempt to control deployment.
 

But when it comes to Myers’ faceoff based zone start counts, I’m not sure they are all that favourable, compared to his “shift starts.”

 

To quote Matt Cane, “The main issue with the current [faceoffs count based] approach to measuring zone starts is that the measurement is often skewed (and sometimes heavily) by the performance and talent of the player in question. Bad players tend to end up with more defensive zone faceoffs because their opponents tend to get more shot attempts against them, which leads to more opportunities for their goalie to freeze the puck and more defensive zone faceoffs.”

 

When Cane looked at how many “traditional zone starts each true zone start is worth,” he came up with the following:

 

 

So, we can attempt to plug in Myers’ “shift starts” and see how many DZ/NZ/OZ faceoffs he’d be expected to have, based on his shift deployment, and the averages for faceoffs per shift type.

 

Again, it’s late, and my eyes are having trouble staying open, so I’m not going to do any calcs tonight, but I suspect if you ran those numbers, the results would show that Myers is ending up with more defensive zone faceoffs than expected, based on averages for his “shift start” deployment.

 

And this might not reflect well on him, in terms of his effectiveness when starting shifts in the defensive zone.

 

(In particular, Myers has some poor icing numbers (62 icings for, 27 against, which is the worst differential on the Canucks’ D), so it’s possible that a far number of his defensive zone (shift) starts are resulting in icings, and creating additional, mid-shift DZFs, which are increasing his DZF%.)

 

To quote Cane again, “often a player’s zone start percentage is impacted by their own performance: bad players end up with more defensive zone faceoffs due to their inability to drive possession, which incorrectly inflates their defensive zone start percentages. This also helps to create a false link between zone start percentages and possession numbers, leading people to incorrectly infer that tough zone starts are a key driver behind a player’s results.”

 

(Just to be clear, even though the quote says “bad players,” I don’t think of Myers as a bad player. I like his PK value. I do have some questions about his EV defensive value, and see some concerning indicators in his overall stats profile. But he’s a legit NHLer. I also don’t think his full value can be measured with stats alone, which is true for most players.)

 

Sorry, this post is pretty rough and disjointed. I just wanted to give you a reply, and some of where I’m coming from, when it comes to the discussion on zone starts. I’d like to come back and really dig into this stuff, and also clean up the mess I’ve written here, but unfortunately, tomorrow is going to be another busy day, and next week’s not much better, so I’m not sure I’ll ever get to it.

 

Regardless, hopefully at least this gives you something to chew on, and maybe furthers the discussion (and I think the links are a good read for anyone interested in these kinds of stats.)

 

Cheers!  

 

 

 

 

 

 

[SID.IS.SID.ME.IS.ME]

4 hours ago, knucklehead91 said:

That is interesting for sure. Perhaps there is an issue with some of hockey references calculations.

 What is consistent is the xGF Myers is on par with Hedman 5v5 in all your samples from other sites 3/5 calculations have Myers with a higher xGF than Hedman.

 Which Myers is on a weaker team compared to Hedman who is on a powerhouse. So 5v5 Myers is contributing the same as Hedman 5v5.  
 

According to Hockey-reference:

Myers

740gp 

564 ES goals against

(786 goals against total)

Hedman

795gp 

648 ES goals against

(880 goals against total)

 

Hedman has been on the ice for alottttta goals.... on one of the best teams for the last decade..

 

Myers has maintained a career oiSV% of 91.8 over the course of his career, never going below 90.1 and has had 6/13 seasons of above 92.0 oiSV.
We all know Myers has played on some really sh*tty Sabres teams

 

Hedman has maintained a career oiSV% of 90.8 over the course of his career, 4 of which were below 90.0 2 of which were 88.7oiSV

Once again....Hedman has been on one of the best teams for the last decade...

Now some posters disagree with the oiSV% stat and say that Myers oiSV% is a result of good goaltending. Either Myers has a horseshoe stuck up is arse  and Hedman broke a mirror and has 7 years badluck..... or the consistent oiSV% is an indication of a players defensive ability. @Provost you may have scoffed at the idea of Myers being in the same stratosphere as Hedman.... but perhaps Hedman is a product of his environment? He plays on the best team in the league who has an unbelievably talented roster, yet when he is in his own zone, for some reason his goalies sV% dips. But when McDonagh is on the ice (and has a 57% dZs) his goalies sV% increases.

 

Sure you can talk about high danger chances, but thats 5 guys on the ice that let that chance happen.

 

Zone starts and oiSV% go hand in hand. EVEN IF ITS JUST 1 MORE FACEOFF IN THE DZONE PER GAME. 1 lost faceoff in your own zone, is a game tying goal or a game winning goal. Each faceoff is crucial and there are roughly 50-70 whistles in a game

 

Zone starts + oiSV are an indication of an individual's influence

Higher dZs = trust

High oiSV% = doing their job, making sure shots are blocked, passing lanes are taken away, or making sure the goalie can see the puck to make the stop.

 

Lower dZs = not trusted

Lower oiSV% = liability

 

Hedmans first 6 seasons on a pretty good Tampa team he averaged 30.33pts (22.5pts through his first 4 seasons) and an 89.0 oiSV%.

That pretty good tampa team has turned into an unbelievably good Tampa team and yes Hedmans points have increased tenfold, but he has also racked up a TON of PPP. But if he was as good as McDonagh in his own zone, his oiSV% should have increased dramatically, however it has barely improved to a career average of 90.8. Which his high oZs% indicates why Hedman is not deployed in his own zone.

Take tampa out of the equation, how much better would Hedman make Buffalo?

 

Lets say Hedman has 15 oZs and 12dZs 20% higher oZs deployment in a game. And Myers has 10 oZs and 14dZs for roughly 30% more dZs. (This is from a faceoff)

Without a faceoff, when either player is not on the ice and bottom pairing D are on the D-side and the play transitions up the ice and into the oZone, who do you think is going to be going onto the ice as the bottom dpair jumps off? 
Hedman for sure will be going on the fly into the oZone for Tampa. 
For Vancouver it will be Hughes. 
Hughes and Hedman are going to be more effective in the oZone than the dZone, so even though there may be only a couple different “zone starts” i.e 2-3, the % shows the reliability and where that reliability is. On the fly changes will increase the zone designation.

 

Also, out of curiousity... do high danger chances take into account PP expiration? Because if that isnt factored into the equation... Myers 12 point whatever to Hedmans 9point something high danger chances allowed isnt a very fair stat to focus on, because Myers averages 3:17 of PK whereas Hedman averages 30seconds less, and averages 2:43PP time more than Myers, which after an expired penalty and some tired PKers on the ice, Hedmans 5v5 stats are going to get boosted. 
 

Anyhoo...

 

 

Bah, I’m in bed, but regrettably, still awake at this ungodly hour (my kid just had a nightmare), so since my night’s sleep is ruined, I’ll try to reply to you too!  

 

I don’t think comparing raw xGF counts between players is offering much value, especially without also considering games played or TOI. You could look at xGF/60, I suppose, if you wanted to compare them, but probably you’d need to look at a lot more, before you’ll get a good picture of these players’ offensive on-ice impacts.

 

Also, why compare just the xGF I posted and not mention the xGA or the xGF%? If you’re comparing Myers and Hedman, why not consider the other stats too, especially since, when comparing between the two players, the xGA and xGF% numbers for Myers and Hedman (at least the ones I added... still not sure about those Hockey Reference stats) have some pretty significant differences?

 

Moving on to oiSV%.

 

There’s really very little evidence, if any, of skaters “driving” save percentage. There are only a handful of players who you might make a case for (and even then, it’d be a fairly dubious claim), and Myers certainly isn’t one of them. If I have time tomorrow, I’ll try to link an article or an analytics blog post that shows this clearly. For now, I hope you’ll take my word for it (it’s actually something I dug into quite a bit, several years back, so I’m pretty confident on this).

 

But for now, if you still think Myers is driving oiSv%, I’d suggest looking over his career, and see how Myers’ oiSv% ranks among his teammates, year for year. I think you’ll find he’s all over the place. Top of the team chart some years, other years middle of the pack, and sometimes near the very bottom. Hopefully this will convince you that Myers doesn’t drive save percentage. If he did, you’d expect to see a very clear pattern where his on-ice Sv% consistently outperformed his goalies’ Sv%, season after season.

 

Zone starts: see my post above, and maybe check out the links, if interested.

 

High danger chances: no, they don’t take into account things like PP expiration. Stats like HDC and HDCF% are just simple event counts, rates, and percentages. There’s really nothing more exotic to them than that.

The NHL tracks shot location coordinates, and any unblocked shot, inside the scoring chance area (also referred to as “home plate”—see image) and from either the slot or inside the crease, is a high danger chance.

 

Every unblocked shot inside “home plate” is a scoring chance.

 

[knucklehead91]

6 hours ago, SID.IS.SID.ME.IS.ME said:

@oldnews

 

I’m not going to quote your whole post. It’s very late, and I’ve had a really long day, so I might do a full quote and response another day, but for now, I’m just going to jot down a few ideas, and save a little board space by avoiding the big quote and reply.

 

(Although, knowing me, this post will probably ramble on anyway, so I’m probably fooling myself if I think I’m going to save much space.)

 

You’re right that there are two types of “zone starts.”

 

I tend to think of the faceoff based ones as “traditional zone starts” because the stat initially just looked at faceoffs. The “shift start” type used to get referred to as “true zone starts” and more recently, it’s kind of become convention to refer to “shift starts” just as zone starts, and the older, “traditional zone starts” as “faceoff percentages” (as in DZF%, NZF%, OZF%).

 

But I agree that it can be confusing. And I probably should have been more clear about the stat I was using, in that earlier post.

 

I disagree about the older stat being more meaningful, when it come to analyzing players, and making adjustments to metrics, based on deployment.

 

Rather than write an essay here, I’ll just add a couple links that do a pretty good job explaining my own views on the two stats, and the overall impact of zone starts, when it comes to impacts/adjustments to Corsi, etc:

 

https://puckplusplus.com/2015/01/15/how-much-do-zone-starts-matter-i-maybe-not-as-much-as-we-thought/
 

https://puckplusplus.com/2015/01/20/how-much-do-zone-starts-matter-part-ii-a-lot-on-their-own-not-that-much-in-aggregate/
 

(These are blog posts from Matt Cane, B.Eng, M.Sc Applied Mathematics, who’s done some good work on this topic,)

 

I do think both stats have value, and help us to get a better picture of what type of minuets players play, and how coaches attempt to control deployment.
 

But when it comes to Myers’ faceoff based zone start counts, I’m not sure they are all that favourable, compared to his “shift starts.”

 

To quote Matt Cane, “The main issue with the current [faceoffs count based] approach to measuring zone starts is that the measurement is often skewed (and sometimes heavily) by the performance and talent of the player in question. Bad players tend to end up with more defensive zone faceoffs because their opponents tend to get more shot attempts against them, which leads to more opportunities for their goalie to freeze the puck and more defensive zone faceoffs.”

 

When Cane looked at how many “traditional zone starts each true zone start is worth,” he came up with the following:

 

 

So, we can attempt to plug in Myers’ “shift starts” and see how many DZ/NZ/OZ faceoffs he’d be expected to have, based on his shift deployment, and the averages for faceoffs per shift type.

 

Again, it’s late, and my eyes are having trouble staying open, so I’m not going to do any calcs tonight, but I suspect if you ran those numbers, the results would show that Myers is ending up with more defensive zone faceoffs than expected, based on averages for his “shift start” deployment.

 

And this might not reflect well on him, in terms of his effectiveness when starting shifts in the defensive zone.

 

(In particular, Myers has some poor icing numbers (62 icings for, 27 against, which is the worst differential on the Canucks’ D), so it’s possible that a far number of his defensive zone (shift) starts are resulting in icings, and creating additional, mid-shift DZFs, which are increasing his DZF%.)

 

To quote Cane again, “often a player’s zone start percentage is impacted by their own performance: bad players end up with more defensive zone faceoffs due to their inability to drive possession, which incorrectly inflates their defensive zone start percentages. This also helps to create a false link between zone start percentages and possession numbers, leading people to incorrectly infer that tough zone starts are a key driver behind a player’s results.”

 

(Just to be clear, even though the quote says “bad players,” I don’t think of Myers as a bad player. I like his PK value. I do have some questions about his EV defensive value, and see some concerning indicators in his overall stats profile. But he’s a legit NHLer. I also don’t think his full value can be measured with stats alone, which is true for most players.)

 

Sorry, this post is pretty rough and disjointed. I just wanted to give you a reply, and some of where I’m coming from, when it comes to the discussion on zone starts. I’d like to come back and really dig into this stuff, and also clean up the mess I’ve written here, but unfortunately, tomorrow is going to be another busy day, and next week’s not much better, so I’m not sure I’ll ever get to it.

 

Regardless, hopefully at least this gives you something to chew on, and maybe furthers the discussion (and I think the links are a good read for anyone interested in these kinds of stats.)

 

Cheers!  

 

 

 

 

 

 

Just hittin back on bolded and yea I totally was goin crosseyed last night and gettin elbows in the ribs from my partner telling my to gtfo my phone XD

anyhoo.

The statement by that Cane doesnt really jive with me. a players defensive zone start numbers being inflated due to being bad and having more shot attempts against that lead to whistles is kinda poorly assumed. After a whistle, if your worst players are on the ice, the coach has the ability to change the lines, UNLESS it is after an icing. So realistically the only way their dZS is being inflated is due to icings. Or the coach is an idiot and is leaving bad players on the ice and tanking the season. An icing could be due to forwards A) not gaining center and taking an icing call b) too much pressure and panic in the dzone and they take an icing to settle it down c) missed passes leading to icings. Which would result in a faceoff in their own zone where the coach cant change the lines. BUT when there is a chance, Guys like Hedman are not goin over the boards, its McDonagh and Sergachev etc. For us it is Truly Edler. But im just here to defend Myers and what hes bringing to the table, compared to guys around the league with the same or substantially higher cap hits.

5 hours ago, SID.IS.SID.ME.IS.ME said:

Bah, I’m in bed, but regrettably, still awake at this ungodly hour (my kid just had a nightmare), so since my night’s sleep is ruined, I’ll try to reply to you too!  

 

I don’t think comparing raw xGF counts between players is offering much value, especially without also considering games played or TOI. You could look at xGF/60, I suppose, if you wanted to compare them, but probably you’d need to look at a lot more, before you’ll get a good picture of these players’ offensive on-ice impacts.

 

Also, why compare just the xGF I posted and not mention the xGA or the xGF%? If you’re comparing Myers and Hedman, why not consider the other stats too, especially since, when comparing between the two players, the xGA and xGF% numbers for Myers and Hedman (at least the ones I added... still not sure about those Hockey Reference stats) have some pretty significant differences?

 

Moving on to oiSV%.

 

There’s really very little evidence, if any, of skaters “driving” save percentage. There are only a handful of players who you might make a case for (and even then, it’d be a fairly dubious claim), and Myers certainly isn’t one of them. If I have time tomorrow, I’ll try to link an article or an analytics blog post that shows this clearly. For now, I hope you’ll take my word for it (it’s actually something I dug into quite a bit, several years back, so I’m pretty confident on this).

 

But for now, if you still think Myers is driving oiSv%, I’d suggest looking over his career, and see how Myers’ oiSv% ranks among his teammates, year for year. I think you’ll find he’s all over the place. Top of the team chart some years, other years middle of the pack, and sometimes near the very bottom. Hopefully this will convince you that Myers doesn’t drive save percentage. If he did, you’d expect to see a very clear pattern where his on-ice Sv% consistently outperformed his goalies’ Sv%, season after season.

 

Zone starts: see my post above, and maybe check out the links, if interested.

 

High danger chances: no, they don’t take into account things like PP expiration. Stats like HDC and HDCF% are just simple event counts, rates, and percentages. There’s really nothing more exotic to them than that.

The NHL tracks shot location coordinates, and any unblocked shot, inside the scoring chance area (also referred to as “home plate”—see image) and from either the slot or inside the crease, is a high danger chance.

 

Every unblocked shot inside “home plate” is a scoring chance.

 

Just like you i was getting pretty tired and waay to wrapped up in this debate last night and I cut a few corners such as just comparing xGF and the reason being, is that there was a few posters here saying Myers does not produce and he sucks blah blah blah... Well I chose the top dman in the league and Myers 5v5 production is on par with Hedman who is on the #1 team in the league.... Myers is not on a #1 team, nor a team that is producing. Myers has 9ESP+2SHP = 11pts when the opposition has 5 players on the ice... Hedman has 14 of his 33pts at 5v5. Tampa has 78GA to our 120GA..... Theres a landslide of a difference in goals...  Its a team game and its not all on Myers. His contract, for $6,000,000 is not bad.

 

Yes the amount of games played does affect these comparisons and isnt entirely fair to the argument. But you would have to expect once Tampa catches up in GP, Hedman will be on the ice for a few more GF and GA.

 

I mean come on man, Im talkin Vancouver vs Tampa bay. 6mil Myers vs Conn Smythe Norris winning Victor muthaf***ing Hedman its not like im choosing Erik Karlsson or PK Subban.

 

Heres a fun fact.... Tyler Myers is 2nd in the league for short handed goals over the last 10 years. He has had 1 SHG per season the last 4 seasons. He's in a 7-way tie for 3rd for short handed points the last 4 years for dmen with 6 SHP. 6-way tie for 6th over the last 10 years with 10SHP (Hedman has 11)

Giordano has 10SHG and 22SHP... No one is even close. 

 

Anyways back on topic. Looking at his career and other teammates for Myers, his numbers are on par and better than most of his teammates and sometimes its the bottom, But he has never dipped below a .900oiSV%. And right now this whole debate was about Myers playing to his contract. , etc etc etc So Ive gather stats and crap like like xGF xGA zone deployment, PP and PK time, E+/- points, 5v5 pts etc etc...from all over, with top teams and top players salary ranges from 4.5-11+mil 

$6,000,000 is pretty fair

 

Just this year alone, Victor Hedman has the lowest oiSV% on his team out of D. Most of them are in the .930 range. Other years, He's still one of the lower oiSV%

 

I will take a look at those articles today at some point, I am interested in them.

 

As for the high danger chance stat.... Thats a 5 guy failure, not a 1 man let down. Its about as fair as comparing +/- 

What led to the +/-? Stepped onto the ice from a bad line change, got a dash 1. Stepped on and got +1? Its not a great stat, it does tell small truths, but not enough

just like how did that guy in the slot get open? how about broken sticks, forwards turning the puck over, hit reff and stayed in, was it on Myers side of the home plate? etc etc etc.... 

 

oiSV% gives an overall picture of his entire  time spent in his own zone.... the goalie has a .918SV% when he's on the ice, which is .001 higher than Demko's current SV%... then theres Holtby's oiSV%.........

 

Vasilevskiy has a .930SV%,

Hedmans oiSV% is .916% - His goalies SV% is takin a bit of a dip. 

Cal Foote .932oiSV%

Cernak .929oiSV%

Schenn .913%oiSV%

Sergachev .933oiSV%

 

For some reason my computer will not screen shot. Take this link, scroll down and go over to E+/-  The entire TBL team is minus except 2 players who have played a combined 6 games. Hedman is the worst....

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

 

Also Myers is our 2nd best PKer on D (technically hes 5th best on the team for regular PKers) with a 4.9 GA p60 (Jordie Benn has a 3.7 GA p60 with an average of 1:35PK time.... so Myers still is better with double the time on the PK of 3:17)

(Hedman has a 5.9 GA p60 on the PK with less PK time 2:41)

 

Myers is 2nd on the team, not just D... but on the entire team with an 8.4GF p60 on the PP.(JT Miller is 1st with 8.5GF p60) 

Hedman has a 11.4GF p60 on the best team in the league. (Hedman has 4x more PP time)

 

idk man, I just dont see the terrible influence through stats or game play that Myers brings... I dont find myself screaming at him. He's been good for us.

[Guest]

You guys don't want to see what the Canucks' defence would look like without Myers....  trust me.  We would be the next Buffalo Sabres.  

 

Myers has probably been our best d-man this year, and I've watched 36 out of the 37 games this year.  You could argue that Schmidt has been better, but I'd rank Hughes and Edler bellow them 

[DownUndaCanuck]

People can try and analyse the points and "improved" stats as much as they want and you get all sorts with Myers - some decent stats but then some bad ones. Watching him play though, he's one of the most complete defencemen we had. I was hoping Schmidt would come in and be a better version of Myers but to be honest it looks like it's the other way around. Myers has been safe on the ice, hardly given up bad chances unless it was due to a team-error, has been dominant defensively along the boards, in front of the net and with his long stick and then up the other end of the ice he's jumped up in the rush perfectly, scored some nice goals and racking up assists.

 

6M gets you a top-4 defenceman in today's day and age and 6M for a guy who kills penalties, is scoring at a 30-40 point pace and defends well is a bargain. Myers is in his prime right now and without him this team would be even worse.

 

I really feel for our defencemen, they're all decent players but they've been thrust into a terrible system that doesn't collapse down well, work as a 5-man unit and is always looking to make risky offensive plays that leads to odd man rushes against. That, combined with how badly both our goalies were at the start of the season, has really led to all sorts of statistic variations.

 

On the whole Myers has been our most consistent all-around defenceman by a long shot. Schmidt has looked average compared to previous years, Benn has looked a bit better but still Benn, Edler has regressed slightly and is starting to look slow, Hamonic has been an under-rated blessing but Hughes has been one of the worst defensive defencemen in the NHL this season, so who knows how bad we'd be without Myers.

[oldnews]

On 3/28/2021 at 2:19 AM, SID.IS.SID.ME.IS.ME said:

 

To quote Matt Cane, “The main issue with the current [faceoffs count based] approach to measuring zone starts is that the measurement is often skewed (and sometimes heavily) by the performance and talent of the player in question. Bad players tend to end up with more defensive zone faceoffs because their opponents tend to get more shot attempts against them, which leads to more opportunities for their goalie to freeze the puck and more defensive zone faceoffs.”

 

The main issue with 'bad' analytics - is that 'bad' analysts - substitute 'bad' assumptions - subjective assumptions - for analysis.

Analytics are only as good as the analyst.

The above slanted categorizaion - of "bad players" (entirely subjective/anti-analytical) - leads me to see extremely litte analytical value in a really poor, hopeless oversimplification like Cane's. 

The problem with the relatively undeveloped 'science' of analytics in hockey - at least at the amateur, public/blogger/media type level - is the obvious what I call corsi-gazing that takes place, with a not even subtle or implicit bias in favour of 'offensive' players.  Cane't take conveniently serves a confirmation bias against defensive players.  The problem with this discussion in a Myers thread - is that he is a more complicated player - who plays a dual role, in various situations, with various partners (of different kinds) - with quite a bit of variance in his minutes and in the end, not much margin between o/dzone, etc - that make 'generalizations' about the player even less meaningful. 

There are a number of contestable points that can easily be brought to bear on that assumption above - 1) that any coach can take said "bad player" off the ice if trapped in the defensive zone because they are a 'bad player'.   The irony is that players that typically have lower ozone starts - regardless of how you calculate the metric - are coaches' choices for the 'harder minutes' - the players they rely upon to handle those dzone starts/terrritorial disadvantages, where the situation is tilted toward the necessity of defending first, and thinking about 'producing' only if you can reverse the tilted ice.  Another obvious factor - is the quality of competition that involved - almost any NHL coach elects to 'shelter' their younger players, and insulate their less defensively capable players - meaning those hard minutes guys tend to face opponent top units in those situations - at least those where coaches have the last change.  Shutdown depth forwards - who are also typically your penalty killers - are not there to shut down the opponent's fourth line - they face top powerplay units, they handle top6/9 forward groups on a regular basis 5on5 - that is the point of their 'specialization'.   Which means that they are not necessarily going to have the most flattering corsi - and that anyone with an analytical mind - should not be dismissing context as if it bears no meaning - and should realize not only the significance of the kind of minutes they play, but also 'adjust' their 'possession' expectations accordingly.  There are some replacement level 'bottom six' forwards - that are not particularly good at (the NHL level) at either end of the ice - but when the term "bad" is applied to guys like Sutter, Beage - it only underlines the absurdity of the 'analytics' being relied upon.  It makes more sense to think like professionals in those instances - why championship level teams would bother with a Jay Beagle - why a young team with a young core would want a player like that to help insulate/enable their youth - as opposed to overcooked simplifications that attempt to explain away what many NHL coaches do (not talking simply Green here, or the Canucks - but a team tendency from at least the AV era to present - and one that is not entirely limited to the alleged lack of wisdom of the present management group).   I have preferred to try to understand what they're actuall doing as opposed to dismiss significant indicators and one up them no unanalytical assumptions/reductions.

If a person tracks enough players, over enough time, on various teams - and further, those players that move to other teams - the significance should not be thrown out with the "bad" bathwater that Cane washes his numbers in.  If you take, for example - the primary 'defensive' / hard minutes forwards of the Canucks - and then try to explain away low ozone starts overlapping with team best goals against metrics and on ice save percentages, for example - the idea that they are simply "bad" players is appropriately dismissed - and replaced with a recognition of what actual NHL teams and coaches do - why they bother with these players - and what the actual situational utility of 'defense-first' players are.  Every sport in the end is a two-way sport - and the science of 'defensive analytics' is pretty poorly developed where hockey is concerned / subject to a underlying/latent offensive bias.   The truism that 'defense wins championships' is not a whole truth - it has it's limits - is only a partial truth - but it certainly should not be dismissed out of hand - as if defensive players are "bad players".   That is so new age 'analytics' wadr.

Edited March 31, 2021 by oldnews

[DefCon1]

On 3/28/2021 at 11:46 PM, VancouverHabitant said:

You guys don't want to see what the Canucks' defence would look like without Myers....  trust me.  We would be the next Buffalo Sabres.  

 

Myers has probably been our best d-man this year, and I've watched 36 out of the 37 games this year.  You could argue that Schmidt has been better, but I'd rank Hughes and Edler bellow them 

Why are you comparing our defense without Myers to the worst team in the league? Myers was like the 6th Dman for the Jets and Jets fans did not like Myers. If you want to win the cup, you need to look at teams that are cup contenders and compare our defense to their team instead of comparing us to the worst team in the league to justify why Myers should stay. We could get a lot better D man than Myers at 6 million. This offseason there was Tanev at 4.5 million, TJ Brodie at 5 million or so and some other great defensemen that other teams got through trades like Devon Toews. All those D men are a lot better than Myers both in mobility and defense and they are all playing in the top 4 or top pairing for their respective team. We could either improve the team by moving these big contracts or stay the course and remain mediocre, wasting ELC years and bridge deals that Hughes, Pettersson and Demko will eventually sign. 

[DefCon1]

On 3/29/2021 at 1:38 PM, DownUndaCanuck said:

People can try and analyse the points and "improved" stats as much as they want and you get all sorts with Myers - some decent stats but then some bad ones. Watching him play though, he's one of the most complete defencemen we had. I was hoping Schmidt would come in and be a better version of Myers but to be honest it looks like it's the other way around. Myers has been safe on the ice, hardly given up bad chances unless it was due to a team-error, has been dominant defensively along the boards, in front of the net and with his long stick and then up the other end of the ice he's jumped up in the rush perfectly, scored some nice goals and racking up assists.

 

6M gets you a top-4 defenceman in today's day and age and 6M for a guy who kills penalties, is scoring at a 30-40 point pace and defends well is a bargain. Myers is in his prime right now and without him this team would be even worse.

 

I really feel for our defencemen, they're all decent players but they've been thrust into a terrible system that doesn't collapse down well, work as a 5-man unit and is always looking to make risky offensive plays that leads to odd man rushes against. That, combined with how badly both our goalies were at the start of the season, has really led to all sorts of statistic variations.

 

On the whole Myers has been our most consistent all-around defenceman by a long shot. Schmidt has looked average compared to previous years, Benn has looked a bit better but still Benn, Edler has regressed slightly and is starting to look slow, Hamonic has been an under-rated blessing but Hughes has been one of the worst defensive defencemen in the NHL this season, so who knows how bad we'd be without Myers.

The problem here is the fact Myers is being described as our best defenseman when couple of years ago when he was on Jets, he was on their 3rd pairing. This is the main reason why Canucks are terrible defensively, because we rely on Edler and Myers to be our best defenseman. If Myers was paid 4 million and he was on our 3rd pairing, I would not be concerned. What Benning needs to do is to trade for a couple of top 4 dmen better than both Edler and Myers. Not sure how he would find those, whether through trades of FA but thats one way to improve our defense. Another step in order to improve our defense is to fire Baumgartner and hire an actual NHL assistant coach that knows defense.

[SID.IS.SID.ME.IS.ME]

4 hours ago, oldnews said:

The main issue with 'bad' analytics - is that 'bad' analysts - substitute 'bad' assumptions - subjective assumptions - for analysis.

Analytics are only as good as the analyst.

The above slanted categorizaion - of "bad players" (entirely subjective/anti-analytical) - leads me to see extremely litte analytical value in a really poor, hopeless oversimplification like Cane's. 

The problem with the relatively undeveloped 'science' of analytics in hockey - at least at the amateur, public/blogger/media type level - is the obvious what I call corsi-gazing that takes place, with a not even subtle or implicit bias in favour of 'offensive' players.  Cane't take conveniently serves a confirmation bias against defensive players.  The problem with this discussion in a Myers thread - is that he is a more complicated player - who plays a dual role, in various situations, with various partners (of different kinds) - with quite a bit of variance in his minutes and in the end, not much margin between o/dzone, etc - that make 'generalizations' about the player even less meaningful. 

There are a number of contestable points that can easily be brought to bear on that assumption above - 1) that any coach can take said "bad player" off the ice if trapped in the defensive zone because they are a 'bad player'.   The irony is that players that typically have lower ozone starts - regardless of how you calculate the metric - are coaches' choices for the 'harder minutes' - the players they rely upon to handle those dzone starts/terrritorial disadvantages, where the situation is tilted toward the necessity of defending first, and thinking about 'producing' only if you can reverse the tilted ice.  Another obvious factor - is the quality of competition that involved - almost any NHL coach elects to 'shelter' their younger players, and insulate their less defensively capable players - meaning those hard minutes guys tend to face opponent top units in those situations - at least those where coaches have the last change.  Shutdown depth forwards - who are also typically your penalty killers - are not there to shut down the opponent's fourth line - they face top powerplay units, they handle top6/9 forward groups on a regular basis 5on5 - that is the point of their 'specialization'.   Which means that they are not necessarily going to have the most flattering corsi - and that anyone with an analytical mind - should not be dismissing context as if it bears no meaning - and should realize not only the significance of the kind of minutes they play, but also 'adjust' their 'possession' expectations accordingly.  There are some replacement level 'bottom six' forwards - that are not particularly good at (the NHL level) at either end of the ice - but when the term "bad" is applied to guys like Sutter, Beage - it only underlines the absurdity of the 'analytics' being relied upon.  It makes more sense to think like professionals in those instances - why championship level teams would bother with a Jay Beagle - why a young team with a young core would want a player like that to help insulate/enable their youth - as opposed to overcooked simplifications that attempt to explain away what many NHL coaches do (not talking simply Green here, or the Canucks - but a team tendency from at least the AV era to present - and one that is not entirely limited to the alleged lack of wisdom of the present management group).   I have preferred to try to understand what they're actuall doing as opposed to dismiss significant indicators and one up them no unanalytical assumptions/reductions.

If a person tracks enough players, over enough time, on various teams - and further, those players that move to other teams - the significance should not be thrown out with the "bad" bathwater that Cane washes his numbers in.  If you take, for example - the primary 'defensive' / hard minutes forwards of the Canucks - and then try to explain away low ozone starts overlapping with team best goals against metrics and on ice save percentages, for example - the idea that they are simply "bad" players is appropriately dismissed - and replaced with a recognition of what actual NHL teams and coaches do - why they bother with these players - and what the actual situational utility of 'defense-first' players are.  Every sport in the end is a two-way sport - and the science of 'defensive analytics' is pretty poorly developed where hockey is concerned / subject to a underlying/latent offensive bias.   The truism that 'defense wins championships' is not a whole truth - it has it's limits - is only a partial truth - but it certainly should not be dismissed out of hand - as if defensive players are "bad players".   That is so new age 'analytics' wadr.

There’s some strawmanning going on here.

 

Firstly, Matt Cane has never, to my knowledge, claimed that Tyler Myers, Jay Beagle, or Brandon Sutter, are “bad players.”

 

I actually made a point, in my original post, of saying that I wasn’t suggesting Myers was a “bad player.”

 

The quote was never intended as a criticism of any particular NHL player, or anyone’s favourite Canuck.

 

I used it to help explain the reasons why traditional “zone starts” (ie: faceoffs based rather than shift based) are considered to be flawed and rather problematic as analytical tools.

 

Cane certainly never said that all defensive players, or even defensive players, in general, are “bad players.”

 

I’d suggest you read the links I originally posted (I’ll link them again at the bottom).

 

The issue with “bad players” is that they can skew the data for OZF%/NZF%/DZF%, which is why Cane argues that “true zone starts” (AKA “shift starts”) are more reliable (albeit still flawed—the work toward “game states” analysis may be more promising), when evaluating or comparing players, and considering things like the impact of deployment on events metrics (like Corsi).

 

Also, you should know that Matt Cane is currently employed as the Director of Hockey Analytics for the New Jersey Devils, and so he’s not just some “public/blogger/media type” who engages in “Corsi gazing.”

 

Again, I really recommend reading the links to his work on the topic:

 

https://puckplusplus.com/2015/01/15/how-much-do-zone-starts-matter-i-maybe-not-as-much-as-we-thought/

 

https://puckplusplus.com/2015/01/20/how-much-do-zone-starts-matter-part-ii-a-lot-on-their-own-not-that-much-in-aggregate/

 

Edited March 31, 2021 by SID.IS.SID.ME.IS.ME
Autocorrect

[IBatch]

On 3/27/2021 at 10:55 PM, aGENT said:

Tanev is a second pairing guy and, as much as I love the guy's defensive play, puts up very little offense... while also being injured a lot. Offense pays, always has (as does staying healthy).

 

Some of that 'half' aren't first pair either or at best 2-3's. Some are guys that signed longer ago at lower cap ceilings. A few are guys that signed sweetheart deals for contenders and/or are more # 3's. 

 

Sorry you're having so much trouble with this 

 

And again, I'm not saying Myers deal is a 'bargain' AT ALL. As I said, he's paid on the high end for a second pair D. Myers on a a 'good' deal would probably be $5.5m... Anything approaching $5m would probably be pushing it into an outright bargain/major discount territory. I don't think I've seen anyone declare his deal anything close to that. It's hardly the albatross some make it out to be though. The sentiment that Seattle wouldn't take him because of his contract, for instance, is laughable.

 

It's neither a 'good' or 'bad' deal. It's a slightly high end, market value deal for a second pair D. He probably gets a bit of a bump for his size and skating ability at that size too.

 

It's not UFA, first pair money. 

Exactly.   I'd also add RHD to that list, and maybe people don't consider this, but it was also pre-covid so by the time his five years were up at the rate the cap increases he'd be a 5-5.5 guy.   Have tried to explain this a few times compared to what MG was signing Ballard and co back when we were contenders.   6 today is what 4 was back then (Ballard got more) was back then.   Petry and Trouba are other recent top four RHDs that got their UFA pay-day, and Brodin did too...all 6.25 - 8 x 8.   Any RFA deal shouldn't even be used as a comp.   And on Tanev.   Seriously a double standard going on here, both him and Edler were constantly ripped on during the rebuild.   Both when they played and when they were out.   Even with Myers taking a big load off Tanevs shoulders, he did end up getting injured right at the end.   Covid gave him time to come back.   

 

Too early on that deal, but yes if we could afford him too that would have been nice.   Myers for sure isn't the worst of our worries, cap wise or otherwise. 

[TheNewGM]

I wouldnt hate Myers if his contract was shorter or less M.. 4 more years ? eh.. but thats Bennings fault not Myers

Edited April 2, 2021 by TheNewGM

[Junkyard Dog]

Too many long posts. You guys are quoting them too. 
 

It’s like you guys wanted to use 6M characters.