Player Discussion: Winnipeg Jets Defense

[MardyBum]

I think you're the closest on this, because honestly I'm not quite sure why Pionk is having quite a rough go at it over the last month.

Based on Corey's All Three Zones microstats tracking - sample size alert, only 11 games tracked with most of them in October, a few in December - he shows up well based on what people were impressed by in his Canadian Division season: retrieves the puck well and moves it up in the DZ, and didn't turnover the puck himself when carrying it out. That being said, boy, the Jets in this sample had a lot of entries against that led to scoring chances.

Maybe the repeated scenario is something as such: Pionk is too slow or doesn't step up at the blueline, Hellebuyck makes a save, maybe there's a rebound or two that Hellebuyck also saves, and Pionk finally goes in to retrieve the puck and move it out, somewhat successfully most of the time.

Watching the game yesterday and the previous one, there's no doubt that Pionk still does most of the things he's good at (passing out of the DZ). Even I recall a few plays at 5v5 where he stopped a cross-seam pass by VAN and quickly moved it out. But (IMO) these days, he hasn't been doing that as often as he's been making plays to avoid situations - or rather, not making plays - that lead to dangerous shots and chances against, and it just shows up everywhere:

• In terms of 5v5 rate stats (CA/60 and xGA/60):
  • Fine in November - 53.46 and 2.49
  • Horrid since December (cutting this to just Dec 15+) - 64.93 and 3.53
    • This is over 10 more shots and nearly 1.5 xGA relative to team average in this span.
• In terms of Evolving-Hockey's RAPMs, this recent stretch has dragged his overall numbers to be amongst the worst in the league, never mind the team:
  • xGA/60 RAPM: 0.395 (3rd worst amongst 249 defencemen with 100 minutes)
  • CA/60 RAPM: 5.82 (2nd worst, behind only Erik Gudbranson)
• Just a plain ol' WOWY showing how bad things are get in the DZ when he's on the ice versus off it.

Meanwhile, his Quality of Teammates is best on the team only being slightly better than Morrissey's, and Quality of Competition slightly easier than 44 (but is 2nd on the team) - but really the percentage points difference in the top 4 of Jets defencemen isn't even 1% in either. In other words: Pionk is getting the same teammates and same competition as the three other Jets top 4, yet is producing the worst results in the defensive zone by far (lately).

Something's off, and if it is injury as some local media has hinted at, then there's absolutely no doubt in my mind that he should be resting to reset and not trying to play through it, especially now that the team is healthier in its forwards and on the backend, and with Heinola putting together a strong stretch of games that is deserving of more TOI to see what's there - especially when the Jets are essentially playing every 2nd night up until the All-Star Break.

I'll add to this with regards to zone entries, as this has been an issue last year as well when he was obviously injured, but he's giving up skating backwards far too quickly. He has zero confidence in it. Even with clear position and space on a 3 on 2 he'll flip to forward skating to try and rub the player out on the boards instead of maintaining position and he opens himself up to easy zone passes in the space he vacates.

 

[BigZ65]

Pionk is the type of player the Canucks would pay 8...no 10 million bucks a year in free agency.

 

[voyageur]

I think you're the closest on this, because honestly I'm not quite sure why Pionk is having quite a rough go at it over the last month.

Based on Corey's All Three Zones microstats tracking - sample size alert, only 11 games tracked with most of them in October, a few in December - he shows up well based on what people were impressed by in his Canadian Division season: retrieves the puck well and moves it up in the DZ, and didn't turnover the puck himself when carrying it out. That being said, boy, the Jets in this sample had a lot of entries against that led to scoring chances.

Maybe the repeated scenario is something as such: Pionk is too slow or doesn't step up at the blueline, Hellebuyck makes a save, maybe there's a rebound or two that Hellebuyck also saves, and Pionk finally goes in to retrieve the puck and move it out, somewhat successfully most of the time.

Watching the game yesterday and the previous one, there's no doubt that Pionk still does most of the things he's good at (passing out of the DZ). Even I recall a few plays at 5v5 where he stopped a cross-seam pass by VAN and quickly moved it out. But (IMO) these days, he hasn't been doing that as often as he's been making plays to avoid situations - or rather, not making plays - that lead to dangerous shots and chances against, and it just shows up everywhere:

• In terms of 5v5 rate stats (CA/60 and xGA/60):
  • Fine in November - 53.46 and 2.49
  • Horrid since December (cutting this to just Dec 15+) - 64.93 and 3.53
    • This is over 10 more shots and nearly 1.5 xGA relative to team average in this span.
• In terms of Evolving-Hockey's RAPMs, this recent stretch has dragged his overall numbers to be amongst the worst in the league, never mind the team:
  • xGA/60 RAPM: 0.395 (3rd worst amongst 249 defencemen with 100 minutes)
  • CA/60 RAPM: 5.82 (2nd worst, behind only Erik Gudbranson)
• Just a plain ol' WOWY showing how bad things are get in the DZ when he's on the ice versus off it.

Meanwhile, his Quality of Teammates is best on the team only being slightly better than Morrissey's, and Quality of Competition slightly easier than 44 (but is 2nd on the team) - but really the percentage points difference in the top 4 of Jets defencemen isn't even 1% in either. In other words: Pionk is getting the same teammates and same competition as the three other Jets top 4, yet is producing the worst results in the defensive zone by far (lately).

Something's off, and if it is injury as some local media has hinted at, then there's absolutely no doubt in my mind that he should be resting to reset and not trying to play through it, especially now that the team is healthier in its forwards and on the backend, and with Heinola putting together a strong stretch of games that is deserving of more TOI to see what's there - especially when the Jets are essentially playing every 2nd night up until the All-Star Break.

I don't worry about xGoals because they aren't really a factor. It's a fancy stat, that coaches don't really use, I'm sure there are more profound metrics than just shot based ones in their decision making. Does Pionk give up more shots than other defensemen? Probably, he does tend to give more space than other d-men, especially in protecting the zone. He often gets caught covering the middle of the ice, more than any other d-man I've seen, maybe to protect his partners, or just instinct I can't say why. Quality of teammates doesn't matter to me. Dillon and Pionk just aren't that good together, that's been pretty obvious since they were put together. And that's not all on Pionk either, Dillon can get beat when pucks get around him, he's definitely the weakest skater on the defense. It seems to work with Schmidt more than Pionk I just think it's the way the two read off each other. I've also noticed that Pionk has been beaten at least 3 times in the last stretch of games, where he is outnumbered on the puck, and that means a forward has cheated.

To say that Heinola has put together a strong stretch of games is kind of untrue too. He had one really good game, a couple of decent games, but those are all home games. His CF% which probably affects his xG is worse than Pionk's. Put him on the road against the league's best and then you can really say that he is what he is. But he hasn't been put in that position. And I don't think it's because our coaches are idiots...they have the same info as you, and more, and they still put Pionk out there in high pressure situations, and the team is still winning. So I think there's a narrative being written that is a common theme in these boards. A player who the fans don't like, because they see inconsistency, and mistakes, and so the next young player in the lineup is the answer. Except Logan Stanley. He just got the big slow useless stereotype from the time he was drafted, and a lot of people on here said should have take Cholowski (who isn't even in the NHL).

 

[voyageur]

Pionk is the type of player the Canucks would pay 8...no 10 million bucks a year in free agency.

I think we actually saw one of the worst defensive pairings in the league yesterday in Ekman-Larsson and Myers, for over $13 million combined.

 

[garret9]

I don't worry about xGoals because they aren't really a factor. It's a fancy stat, that coaches don't really use, I'm sure there are more profound metrics than just shot based ones in their decision making. Does Pionk give up more shots than other defensemen? Probably, he does tend to give more space than other d-men, especially in protecting the zone. He often gets caught covering the middle of the ice, more than any other d-man I've seen, maybe to protect his partners, or just instinct I can't say why. Quality of teammates doesn't matter to me. Dillon and Pionk just aren't that good together, that's been pretty obvious since they were put together. And that's not all on Pionk either, Dillon can get beat when pucks get around him, he's definitely the weakest skater on the defense. It seems to work with Schmidt more than Pionk I just think it's the way the two read off each other. I've also noticed that Pionk has been beaten at least 3 times in the last stretch of games, where he is outnumbered on the puck, and that means a forward has cheated.

To say that Heinola has put together a strong stretch of games is kind of untrue too. He had one really good game, a couple of decent games, but those are all home games. His CF% which probably affects his xG is worse than Pionk's. Put him on the road against the league's best and then you can really say that he is what he is. But he hasn't been put in that position. And I don't think it's because our coaches are idiots...they have the same info as you, and more, and they still put Pionk out there in high pressure situations, and the team is still winning. So I think there's a narrative being written that is a common theme in these boards. A player who the fans don't like, because they see inconsistency, and mistakes, and so the next young player in the lineup is the answer. Except Logan Stanley. He just got the big slow useless stereotype from the time he was drafted, and a lot of people on here said should have take Cholowski (who isn't even in the NHL).

I'm just going to breakdown all the errors in here, but I will point out that your second sentence is flat out false:
I have not talked to all 32 coaches, but I know that more than 50% use xGoals and 100% of teams do in some capacity their opperations.
They may use their own xGoal model based on private data, but that's going to refine and adjust relative to public, not flip results upside down.

FYI: The guy that created xGoals got a day with the cup and a ring this summer.

 

[voyageur]

I'm just going to breakdown all the errors in here, but I will point out that your second sentence is flat out false:
I have not talked to all 32 coaches, but I know that more than 50% use xGoals and 100% of teams do in some capacity their opperations.
They may use their own xGoal model based on private data, but that's going to refine and adjust relative to public, not flip results upside down.

FYI: The guy that created xGoals got a day with the cup and a ring this summer.

Ok let's be realistic here. You notice that 3 of the most used defensemen (Morrissey, De Melo, Pionk) in late game situations, where the team leading tends to give up more shots, are among the 3 lowest on the team in CF%. Which I understand impacts xGoal scenarios. Now when you rate those stats, how do they compare vs. things like zone possession time, which is as important if not more important than shots. A guy like Pionk in the offensive zone will often shoot wide of the net to keep possession, is that a bad play? How do they compare with passing efficiency, and zone transitions. Zone clears? I've never heard any analytic people on here use those stats, but you can be sure that coaches rely on those, yes?

 

[surixon]

Yeah to me Pionk is just not turning into the type of player we need in that spot at that cap hit. Hopefully he pulls it together as imo he's the weak link in our top 4 atm.

 

[garret9]

Ok let's be realistic here. You notice that 3 of the most used defensemen (Morrissey, De Melo, Pionk) in late game situations, where the team leading tends to give up more shots, are among the 3 lowest on the team in CF%. Which I understand impacts xGoal scenarios. Now when you rate those stats, how do they compare vs. things like zone possession time, which is as important if not more important than shots. A guy like Pionk in the offensive zone will often shoot wide of the net to keep possession, is that a bad play? How do they compare with passing efficiency, and zone transitions. Zone clears? I've never heard any analytic people on here use those stats, but you can be sure that coaches rely on those, yes?

1) Morrissey-DeMelo are the most common used defense pair in late minutes. They have a similar Corsi and higher xG than Dillon-Pionk pair, who are used in slightly softer minutes. They are lower than Samberg-Schmidt, but those see much softer minutes. So matches Corsi and xGoal when you use nuance.
From a general point: Corsi% or xGoal% (especially if you do relative to teammate) correlates with average TOI, while +/- does not. So if that's your type of analysis, coaches use those numbers but don't use +/-.

2) Zone time performs worse than xG or Corsi, and coaches/management use it less. Think of it this way, you want possession to create scoring chances, not the other way around.

3) If you keep possession and it works out, they will lead to more shots for and fewer against in the long run. For Pionk, it doesn't. Whether that's by goal differential, xGoal, or Corsi.

4) Pionk actually performs poorly in transitional numbers. He has the highest defensive zone failure rate (aka turnovers per breakout attempt) after Stanley and tied with Samberg for the lowest offensive zone transition with puck control per attempt.

5) Show me one player who over multiple seasons combined (3+) has a good goal differential but a bad xGoal differential, try also with Corsi.

6) You haven't heard any analytic people here use transition stats??? You literally liked one of my posts here when I responded to you about Harkins. I used transitional stats and trend in Corsi/xGoal to show he's not driving the bus on his xGoals/Corsi.
Transitional stats have been tracked by the analytical community for years (including by me) and show the hows/whys a player is good or bad, where Corsi and xGoal are part (you keep confusing that it's part not entire) of how good or bad that player is.

Yeah to me Pionk is just not turning into the type of player we need in that spot at that cap hit. Hopefully he pulls it together as imo he's the weak link in our top 4 atm.

He's weak link in the top 6 by my analysis.

 

[JetsFan815]

Dynamite game from Pionk tonight other than the bouncing puck at the blue line that resulted in a 2 on 1 and goal against (which was just bad luck, that bounce could have happened to anyone). So many times he generated a scoring chance or got himself open for a chance or kept the play alive in the o-zone with his iq. Must have read the latest posts in this thread.

This game shows his value to the team and what he brings to the table. He can contribute positively to the Jets, he is not a Myers/Sbisa/Beau type "worst D on the Jets" who basically rarely did anything good. If they can get him to improve his d-zone play a bit... he could be a key contributor in the playoffs if we make it there. He might benefit from simplifying his game in the d-zone and just chipping or rimming the puck from time to time, I notice he is always trying to make a play with the puck in his stick in the d-zone and sometimes skates himself into trouble.

 

[voyageur]

1) Morrissey-DeMelo are the most common used defense pair in late minutes. They have a similar Corsi and higher xG than Dillon-Pionk pair, who are used in slightly softer minutes. They are lower than Samberg-Schmidt, but those see much softer minutes. So matches Corsi and xGoal when you use nuance.
From a general point: Corsi% or xGoal% (especially if you do relative to teammate) correlates with average TOI, while +/- does not. So if that's your type of analysis, coaches use those numbers but don't use +/-.

2) Zone time performs worse than xG or Corsi, and coaches/management use it less. Think of it this way, you want possession to create scoring chances, not the other way around.

3) If you keep possession and it works out, they will lead to more shots for and fewer against in the long run. For Pionk, it doesn't. Whether that's by goal differential, xGoal, or Corsi.

4) Pionk actually performs poorly in transitional numbers. He has the highest defensive zone failure rate (aka turnovers per breakout attempt) after Stanley and tied with Samberg for the lowest offensive zone transition with puck control per attempt.

5) Show me one player who over multiple seasons combined (3+) has a good goal differential but a bad xGoal differential, try also with Corsi.

6) You haven't heard any analytic people here use transition stats??? You literally liked one of my posts here when I responded to you about Harkins. I used transitional stats and trend in Corsi/xGoal to show he's not driving the bus on his xGoals/Corsi.
Transitional stats have been tracked by the analytical community for years (including by me) and show the hows/whys a player is good or bad, where Corsi and xGoal are part (you keep confusing that it's part not entire) of how good or bad that player is.


He's weak link in the top 6 by my analysis.

Respectfully Garret I'll try to summarize with this. I'd definitely not arguing for Pionk on a +/- basis, since he is the worst on the team in that respect, it wouldn't help my argument. But I see an intelligent player with a lot of attacking zone chances coming from his passes, and I'm sure there must be a metric in there somewhere that has him as good transition player, and at least the 2nd best puck mover on the team.

Tonight's game is a good summary of Pionk's season. He was on the ice for 2 goals against, one was the most egregious, I'd say that both Detroit goals that were scored with Pionk on the ice could be attributed to a lack of physicality at times. The 1st goal he scored was a direct result of zone possession time, with a shift length that had the opponents on for a full minute defending in their zone, which contributed to a full forward line change, while the defense held the line. I think any coach would have interest in shifts that directly put the opponent in a situation where they give up more shots or a penalty. I'm surprised that Pionk doesn't contribute to this positively in some way. 2nd goal Pionk was on for was quick pass, that gave Connor enough room to manoeuvre for offense. In terms of scoring chances he ran one of the post that would have ended his night as a +1 player, but his mistakes were costly.

I appreciated your information on Harkins. It is helpful analysis. A lot of that information isn't publicly available, or maybe requires a subscription. Some of the information you provide privately would certainly impact roster choices. Mark Stuart is a notable one where you definitely proved me wrong, and perhaps you are right again.

I see the Jets lacking a true #1 RD at this time, and part of that is the physicality they lost in both Byfuglien and Trouba. I'd be interested to know where Morrissey and De Melo rank among the NHL's top pairing. I'm not surprised if Dillon and Pionk underperform as a pairing, and there's probably a lot of metrics that support replacing Pionk with Schmidt.

If Pionk is on your 3rd pairing he's not your #2 d-man anymore, and maybe that's a good thing, but I still see him as the 2nd best offensive player on the team by a wide margin. I'd be interested to know your thoughts on a Samberg-Pionk pairing, and where Samberg's PKing metrics rank among the NHL's best, seems to me like's he in there for one of the best.

I don't think the Jets can consider De Melo as a true #2 d-man, and Schmidt probably isn't either. Do you think it would be a good time to integrate Ville Heinola into the mix, and throw 7 d-men into the blender see if reduced minutes for some key players increases production. I honestly don't see Pionk dragging this team down, but someone who might need to be put in a different role.

 

[Jet]

Based on the amount of concern shown about Pionk's game I tried to look closer at his play.
Don't get me wrong I'm not trying to say that what people are saying isn't true... I'm just not seeing as much bad play as is being reported here. I'll have to keep watching to see more examples.

I agree. I think people are hyper focused on it right now as our other D have upped their games.

I feel like Pionk knows his game isn't where it needs to be and is 'over trying'. He does some good things and then some really dumb things all in one game.

I'm hoping the coaching staff can work to get it sorted out.

I don't think he's been bad enough to get benched though.

 

[DRW204]

prior to the year, i wondered/questioned how much goal-scoring our D group could provide. bowness was adamant on getting more scoring out the back-end, after years of being found in the doldrums of the league.

for reference
season: goals/goals rank
19-20: 21/29th
CDN DIV: 10/31st
21-22: 24/31st

this year
already at 23 goals! ranked 7th in the NHL. top team has 25 (FLA/NYI/WSH/SEA), worst is VAN with 10.

 

[DRW204]

These are the D pairs sorted by highest rate of chances given up when on ice.

A bit of a similar names at the top to last years group

 

[ps241]

I agree. I think people are hyper focused on it right now as our other D have upped their games.

I feel like Pionk knows his game isn't where it needs to be and is 'over trying'. He does some good things and then some really dumb things all in one game.

I'm hoping the coaching staff can work to get it sorted out.

I don't think he's been bad enough to get benched though.

To me Neil looks like a guy who has been struggling with his confidence. When players are at their best they play instinctual and quick, mentally and physically. They are on the front of their skis metaphorically. Neil looks like he has been playing on the back of his skis for a while.

 

[ps241]

I'll add to this with regards to zone entries, as this has been an issue last year as well when he was obviously injured, but he's giving up skating backwards far too quickly. He has zero confidence in it. Even with clear position and space on a 3 on 2 he'll flip to forward skating to try and rub the player out on the boards instead of maintaining position and he opens himself up to easy zone passes in the space he vacates.

You make a great point about Neil transitioning out of backwards skating quickly. Karlsson use to have this tendency early in his career in Ottawa.

 

[DRW204]

prior to the year, i wondered/questioned how much goal-scoring our D group could provide. bowness was adamant on getting more scoring out the back-end, after years of being found in the doldrums of the league.

for reference
season: goals/goals rank
19-20: 21/29th
CDN DIV: 10/31st
21-22: 24/31st

this year
already at 23 goals! ranked 7th in the NHL. top team has 25 (FLA/NYI/WSH/SEA), worst is VAN with 10.

quoting above as a reference point

haven't looked at this in a while.... league leaders in team-D scoring this season:

 

[DRW204]

for reference the 3 previous years:

season: Jets D goals/goals rank
19-20: 21/29th
CDN DIV: 10/31st
21-22: 24/31st

what an uptick this year from the group:

sorted by limiting the opposition in chances allowed (xGA/60)

of Dmen with similar minutes played to dillon-schmidt (went with 270 mins, or 10% less overall as a benchmark) the Jets ranked 9th in least xGA/60 and best in GA/60 with those 2 on the ice over the year.

they didn't get scoring though, but imo id attribute that more to a fwd problem than a D problem. but those two seem to have rated well together on the defensive side in a top-4 role.

Also, Dylan Demelo showed some offensive iq/prowess that we really haven't ever gotten from him id say. i think it was a game where morrissey was out and demelo and samberg were on the top pair, and demelo really stood out in more of an offensive role. he also whiffed on a couple open nets too this year

he was 2nd best on the team in scoring efficiency at 5v5 (pts/60) and had his best overall output of his career.

 

[DRW204]

Let’s start with the Jets. Winnipeg’s been a game team across the board this year, but the results at the top of the lineup are truly impressive. The Morrissey and DeMelo pairing – they play primarily with the Mark Scheifele line – have cooked teams at even-strength, out-scoring their opponents by 19-goals while carrying about 53 per cent of the expected goal share. That’s playoff-calibre production from your top-pairing, period. Behind them, Winnipeg’s a much weaker team territorially, but even their depth group has out-scored the competition (+7 goals) on the year:

Travis Yost: Do Canadian teams have the two best pairings in hockey? | TSN

If we look at how the Jets and Canucks have played with these top-pairs on the ice versus without, it's a true night-and-day scenario. Travis Yost has more on the dominance of Quinn Hughes and Filip Hronek in Vancouver, as well as Josh Morrissey and Dylan DeMelo in Winnipeg.

www.tsn.ca

 

[trebendan]

Let’s start with the Jets. Winnipeg’s been a game team across the board this year, but the results at the top of the lineup are truly impressive. The Morrissey and DeMelo pairing – they play primarily with the Mark Scheifele line – have cooked teams at even-strength, out-scoring their opponents by 19-goals while carrying about 53 per cent of the expected goal share. That’s playoff-calibre production from your top-pairing, period. Behind them, Winnipeg’s a much weaker team territorially, but even their depth group has out-scored the competition (+7 goals) on the year:

Travis Yost: Do Canadian teams have the two best pairings in hockey? | TSN

If we look at how the Jets and Canucks have played with these top-pairs on the ice versus without, it's a true night-and-day scenario. Travis Yost has more on the dominance of Quinn Hughes and Filip Hronek in Vancouver, as well as Josh Morrissey and Dylan DeMelo in Winnipeg.

www.tsn.ca

View attachment 793891

We need to re-sign DeMelo!

 

[larmex99]

 

[Mortimer Snerd]

We need to re-sign DeMelo!

Dillon too.

 

[Adam da bomb]

Dillon too.

I wonder what cap looks like if they resign Dillion.

 

[Gabe Kupari]

Dillon too.

He needs to want to be here. I’m not sure he does.

 

[WolfHouse]

He needs to want to be here. I’m not sure he does.

Based on exactly nothing haha....

View attachment 793916

The two oilers duos makes me suspect this is a flawed stat

 

[AtomicJets]

Dillon will be 34 next year. He's still a good player but I'm not sure how much demand there will be for him in free agency. Teams would likely be reluctant to go beyond a one year deal.