Devils team discussion (news, notes and speculation) - offseason part III

[Camille the Eel]

In the Southern Hemisphere they spin the other way

also pigs tails curl in the other direction south of the line. No kidding.

 

[Better Call Sal]

Thing of beauty.

 

[Zajacs Bowl Cut]

I think the foremost shortcoming of analytics used in all sports is they tend to slant towards the previously held biases of the people coming up with the analytics.

Possession is certainly an aspect of hockey, and many analytic models are wisely doing their best to measure it. But to say that hits and blocked shots are irrelevant and then to omit them from the analytic model is equivalently unwise.

I don't understand this line of thinking. Analytics are all based on something. They are not just based on things pulled out of thin air. They are based on shot attempts, or zone entries/exits, or (insert something that happens in an NHL game here). They aren't slanted. They are just information.

Also, no one said hits and blocked shots are totally irrelevant. There are obviously times when you need some hits or you need some blocked shots. But similar to faceoff wins, they are almost irrelevant when judging one player vs another player that I really do not care about them at all.

 

[Patrik26]

Thing of beauty.

Now if we could get a Stanley Cup Playoffs (nice) or Stanley Cup Final (even better) logo there ...........

 

[njdevils1982]

about the left and right thing ....i played soccer for 35 years. ...from the age of 10 and on i was first a left half for about 3 or 4 years and then moved to my left d position which i played until 'retirement' in 2014

all being a natural right footed player...i managed fine enough (mainly due to learning to use the left from a young age) ...my brother has already taught his kid the same but since age 4 (hes 10 now and has already 'got it')


ill also add this for fun

Kawai announce new Left-Handed piano

You've heard of a left handed drum kit and a left handed guitar? Well now there's a left handed piano. Kawai pianos, the specialist piano manufacturer renowned for its innovation and creative thinking have announced the launch of a brand new concept - the left-handed piano. It's true that...

www.sheargoldmusic.co.uk

 

[Devils731]

about the left and right thing ....i played soccer for 35 years. ...from the age of 10 and on i was first a left half for about 3 or 4 years and then moved to my left d position which i played until 'retirement' in 2014

all being a natural right footed player...i managed fine enough (mainly due to learning to use the left from a young age) ...my brother has already taught his kid the same but since age 4 (hes 10 now and has already 'got it')


ill also add this for fun

Kawai announce new Left-Handed piano

You've heard of a left handed drum kit and a left handed guitar? Well now there's a left handed piano. Kawai pianos, the specialist piano manufacturer renowned for its innovation and creative thinking have announced the launch of a brand new concept - the left-handed piano. It's true that...

www.sheargoldmusic.co.uk

Soccer is a wholly different animal though.

You may have a preferred foot but they’re both your feet, you can mirror their usage when you switch sides.

For hockey, you can’t do that because your forehand and backhand have inherent strengths and weaknesses that will always exist.

 

[Guttersniped]

There are disadvantages for defensemen playing on their off-side defensively, or at least differences in approach.

Some people want to make it sound like there aren’t which is just weird.

There are advantages offensively for a defenseman playing on their off-side.

 

[Eggtimer]

There are disadvantages for defensemen playing on their off-side defensively, or at least differences in approach.

Some people want to make it sound like there aren’t which is just weird.

There are advantages offensively for a defenseman playing on their off-side.

oh for sure.More disadvantages than a person thinks. Simple little plays are made harder. I’d say one of the main things is receiving and trying to pick up the boards or a pass etc on your backhand instead of your forehand . Also dumping the puck on a PK . Takes precious time to switch to your forehand to dump the puck so you are forced trying to clear it on the backhand. Tons of times a year you see multiple plays where a D tried to clear on his backhand and whiffs on it and can’t elevate it out of the zone and ends up a goal against.

 

[njdevils1982]

Soccer is a wholly different animal though.

You may have a preferred foot but they’re both your feet, you can mirror their usage when you switch sides.

For hockey, you can’t do that because your forehand and backhand have inherent strengths and weaknesses that will always exist.

quality left D in soccer is like RHD in hockey in my opinion...scarce

a very important position due having to cover right footed wingers who are in abundance so teams have a larger pool to draw skilled players

not many people can pull off the mirror move....shit....after playing the left for so many years once in a while i did spells on the right and it was almost foreign to me. the steps i would use to turn to the outside defending, angles of attack, perception of space and the mental part. i obviously could do it but was more valuable on the left as there were never many (if any) substantial players that were naturally left footed on our club over the many years

 

[Goptor]

I don't understand this line of thinking. Analytics are all based on something. They are not just based on things pulled out of thin air. They are based on shot attempts, or zone entries/exits, or (insert something that happens in an NHL game here). They aren't slanted. They are just information.

Also, no one said hits and blocked shots are totally irrelevant. There are obviously times when you need some hits or you need some blocked shots. But similar to faceoff wins, they are almost irrelevant when judging one player vs another player that I really do not care about them at all.

Its a common problem when trying to do modeling.
The variables need to be weighted in order to get reasonable outputs but how should it be done and who should be doing it? Some sports stats even need to worry about who is collecting the data.

The entire sports analytics field is based upon which team can take advantage of others by being the least flawed.

 

[My3Sons]

quality left D in soccer is like RHD in hockey in my opinion...scarce

a very important position due having to cover right footed wingers who are in abundance so teams have a larger pool to draw skilled players

not many people can pull off the mirror move....shit....after playing the left for so many years once in a while i did spells on the right and it was almost foreign to me. the steps i would use to turn to the outside defending, angles of attack, perception of space and the mental part. i obviously could do it but was more valuable on the left as there were never many (if any) substantial players that were naturally left footed on our club over the many years

My middle son played defense in lacrosse. He’s a natural lefty and the advantage he had over kids that could only shoot right handed was dramatic. I imagine that is similar to how it would play out in soccer. Your ability to adjust and use your stick to alter or take away a shot is so much fatster with your natural handedness in play.

 

[Guadana]

I don't understand this line of thinking. Analytics are all based on something. They are not just based on things pulled out of thin air. They are based on shot attempts, or zone entries/exits, or (insert something that happens in an NHL game here). They aren't slanted. They are just information.

Also, no one said hits and blocked shots are totally irrelevant. There are obviously times when you need some hits or you need some blocked shots. But similar to faceoff wins, they are almost irrelevant when judging one player vs another player that I really do not care about them at all.

I'm ok with analytics, I love advanced stats, I just don't trust them overall, without eye test. A lot of players makin impact without good possession metrics, someone have them but it doesn't work. Some could be meh in 60% of time on the ice, but their decision making in offensive zone help them to create really dangerous chances.
There are too many factors in hockey, you can't count or/and balanced them. And different players are playing different roles. Someone could be dominant driving force but can't find or create open space in cycling, other one could move his body well and create space for partners, but useless in transition. What's about defensive forwards, who is very good against the cycling, close gaps and corridors, but isn't good with the puck or can play on the slot only. Or could play rebounder role and score 10+ goals on rebounds but isn't fast enough to make the rushes and bee your transition guy in your bottom. Metrics of that kind of guy could be bad, but "this" guy could make very important job.

Advanced metrics shows only what counted person wants to show you. May be he doesn't understand different things, but it is what it is. Ofcourse it helps, and we should use it. But without eye test and understanding of different aspects of the game, all this stories about how good Tatar and look at his metrics, cost nothing. Because when you watch him, you can clearly understand why stat people love him, but coaches sit him on the bench, and gm's traded him.

 

[Triumph]

I don't understand this line of thinking. Analytics are all based on something. They are not just based on things pulled out of thin air. They are based on shot attempts, or zone entries/exits, or (insert something that happens in an NHL game here). They aren't slanted. They are just information.

Also, no one said hits and blocked shots are totally irrelevant. There are obviously times when you need some hits or you need some blocked shots. But similar to faceoff wins, they are almost irrelevant when judging one player vs another player that I really do not care about them at all.

There's a moral dimension to opposing stats that don't take into account hits and blocked shots because these are the most sacrificial plays a player makes. He's giving up his body to make a play, or at least disregarding his own pain. What kind of a person would say these sorts of plays are 'worth nothing?' Well, they're not worth nothing, but we just can't really quantify the value of having more or less of them. What we do know is that players who generally hit a lot or block a lot of shots aren't all that good, but that's not very useful either.

The Devils who were best at being without the puck in the offensive zone, in my opinion, were Zach Parise and Jaromir Jagr, both guys were just great at forechecking and recovering the puck or forcing the opposition into a turnover. Parise hit a little bit, but Jagr didn't hit at all. Is there maybe some value to hitting over a 7 game series where you're facing the same team over and over, and players might be worn down and force themselves into bad plays? Absolutely. I'm just not sure this effect is real.

 

[TBF1972]

In the Southern Hemisphere they spin the other way

Ushuaia Devils

 

[Its Always Sundstrom]

And now another episode of a day in the life of Patrik Elias…

Patty is sitting in the TV room sulking. He’s trying to figure how to tell his wife he knows she watched the finale of Only Murders in the Building without him but without sounding upset.

 

[devilsblood]

Thing of beauty.

Ya, that is a very nice touch.

 

[devilsblood]

I'm not as up on hockey stats as I am on baseball ones (very different sport to try and create metrics for, of course), but I think the conversation around them hits a similar hiccup that sometimes impacts baseball talks.

Namely, I think that advanced stats in baseball absolutely have the right idea in terms of emphasizing what's the most important aspects of success in the game, and the initial statistical revolution in the game was huge for very good reasons. For example, the initial things that really jumped out was how sabermetrics emphasized the importance of walks and on base percentage, downplayed the importance of raw RBI numbers (they're heavily reliant on the team surrounding you, so not a good metric of an individual player's ability), and made people reconsider the risks of using bunts and stolen base attempts due to the 27 outs available to you in a game being very precious and not usually worth giving up outside of some obvious situations.

That all said, some people internalized these so much that I think they lost sight of the fact that, yeah, stolen bases can be very useful, as long as you can maintain a steal success rate of about 80%; walks are great, but what if you're being "overly patient" and passing up pitches to hit too often?; and while RBIs are a weak stat for measuring individual player performance, is there not something to be said for a player having a good "RBI approach" when runners are on base, e.g. being focused on driving the ball, getting a sac fly, focusing on contact over selling out fully for power, etc.?

Basically, the advanced stats get at the heart of what you need to build a successful roster, but that doesn't mean the other stuff has no utility; it's just that old school thinkers would take the other stuff and play it up at the expense of the core essentials you need for success, which is not the way to go.

But wasn't OBP always a thing in baseball? Slugging as well. Combining the two is a neat little invention though.

And while chicks may dig the long ball, I have a place in my heart for guys who can go the other way and don't strike out much.

 

[MachoDiablo]

But wasn't OBP always a thing in baseball? Slugging as well. Combining the two is a neat little invention though.

And while chicks may dig the long ball, I have a place in my heart for guys who can go the other way and don't strike out much.

It's always been a thing, but for ages the emphasis was on the Triple Crown stats of average, homers, and RBIs; the saber revolution is what led to a lot of teams realizing that OBP could be a bigger deal than average, among other shifts in perspective. The thing that brought about the Moneyball movement was that teams with less financial advantages, in this case the Oakland A's, realized that the way to compete with the big boys was to take advantage of stats and aspects of the game that were being undervalued by other clubs; instead of pursuing a guy with a .300 average (who you likely won't be able to afford anyway) who gets on base at a .330 clip, get the guy who hits .260 but gets on base at a .360 rate. You can't beat, say, the Yankees at their own game, so take advantage of the things teams like the Yankees aren't prioritizing.

What that eventually led to was every team beginning to understand how important OBP is, but one can argue that it also led to teams going too far in deemphasizing how helpful a higher average can be. One reason the Mets have spent so long in first place this season is they've built a lineup that's not afraid to go the other way, and has also cut down on their strikeouts overall, and it might be working because a lot of other clubs have gone too far in the "three true outcomes" direction of homers, strikeouts, and walks.

Regardless, I think that's just something to keep in mind as hockey's advanced stats grow and become more and more well-developed; use them, learn to prioritize what they point toward, but try not to swing too far away from other things that work unless you really have no choice.

 

[Eggtimer]

It's always been a thing, but for ages the emphasis was on the Triple Crown stats of average, homers, and RBIs; the saber revolution is what led to a lot of teams realizing that OBP could be a bigger deal than average, among other shifts in perspective. The thing that brought about the Moneyball movement was that teams with less financial advantages, in this case the Oakland A's, realized that the way to compete with the big boys was to take advantage of stats and aspects of the game that were being undervalued by other clubs; instead of pursuing a guy with a .300 average (who you likely won't be able to afford anyway) who gets on base at a .330 clip, get the guy who hits .260 but gets on base at a .360 rate. You can't beat, say, the Yankees at their own game, so take advantage of the things teams like the Yankees aren't prioritizing.

What that eventually led to was every team beginning to understand how important OBP is, but one can argue that it also led to teams going too far in deemphasizing how helpful a higher average can be. One reason the Mets have spent so long in first place this season is they've built a lineup that's not afraid to go the other way, and has also cut down on their strikeouts overall, and it might be working because a lot of other clubs have gone too far in the "three true outcomes" direction of homers, strikeouts, and walks.

Regardless, I think that's just something to keep in mind as hockey's advanced stats grow and become more and more well-developed; use them, learn to prioritize what they point toward, but try not to swing too far away from other things that work unless you really have no choice.

I wish the Jays could learn something from the Mets . It is so frustrating watching average - below average pitching dominate Bluejays hitters . All the have to do is pitch everyone down and away . If they have a RHP that can throw a slider . All Jays hitters worth a shit are righties and they cannot and will not adjust to how teams pitch them . It’s insane .

 

[Eggtimer]

I'm ok with analytics, I love advanced stats, I just don't trust them overall, without eye test. A lot of players makin impact without good possession metrics, someone have them but it doesn't work. Some could be meh in 60% of time on the ice, but their decision making in offensive zone help them to create really dangerous chances.
There are too many factors in hockey, you can't count or/and balanced them. And different players are playing different roles. Someone could be dominant driving force but can't find or create open space in cycling, other one could move his body well and create space for partners, but useless in transition. What's about defensive forwards, who is very good against the cycling, close gaps and corridors, but isn't good with the puck or can play on the slot only. Or could play rebounder role and score 10+ goals on rebounds but isn't fast enough to make the rushes and bee your transition guy in your bottom. Metrics of that kind of guy could be bad, but "this" guy could make very important job.

Advanced metrics shows only what counted person wants to show you. May be he doesn't understand different things, but it is what it is. Ofcourse it helps, and we should use it. But without eye test and understanding of different aspects of the game, all this stories about how good Tatar and look at his metrics, cost nothing. Because when you watch him, you can clearly understand why stat people love him, but coaches sit him on the bench, and gm's traded him.

I’d love to see what NHL teams actually use for analytics. For sure they are different than what we have access to . It would be very interesting to see what teams value what stats over others. And how much emphasis is put into basing decisions off of those stats .
Example - egghead stats person pushes fir player X in the lineup and put on line 3 with player Y and Z as stats show it “ should” be the most effective . How much weight dues that hold ? Or are stats used more do for just drafting and developing ? Or drafting and scouting when potentially trading for a player ?
What I don’t get , is when a player clearly is getting demolished by every stat metric , yet the team continues to play that player . Prime example - Ty Smith last season . From some stats the average joe had access to , he was one of the worst D men in the league ( with any sort of playing minutes ) . Why would he be trotted out there as much as he was then ? Stubbornness? Trying to see if he could fight his way through things and figure it out ? No other better options ?

 

[Guadana]

I’d love to see what NHL teams actually use for analytics. For sure they are different than what we have access to . It would be very interesting to see what teams value what stats over others. And how much emphasis is put into basing decisions off of those stats .
Example - egghead stats person pushes fir player X in the lineup and put on line 3 with player Y and Z as stats show it “ should” be the most effective . How much weight dues that hold ? Or are stats used more do for just drafting and developing ? Or drafting and scouting when potentially trading for a player ?
What I don’t get , is when a player clearly is getting demolished by every stat metric , yet the team continues to play that player . Prime example - Ty Smith last season . From some stats the average joe had access to , he was one of the worst D men in the league ( with any dirt of playing minutes ) . Why would he be trotted out there as much as he was then ? Stubbornness? Trying to see if he could fight his way through things and figure it out ? No other better options ?

I believe good teams have better stats programs, pay people to count more specific things, and use it in different context. For example having good transition metrics against better competition is much harder. But I think it will ends to watch specific player and understand what he did right or wrong. Its not that hard for specialists.
In Ty case, you want to have him on the roster to have his price on the right height. May be he can bounce back, gain form, caught some hot strick or whatever. On my taste he wasn't good in his first year at all. But you just don't trade the guy for the wrong player if the guy is cost something. Marino is the right guy on paper. I wouldn't be surprised if Fitz would not trade him, if the right deal wouldn't on the table. Young guys are becoming good players with expirience, this is what they gave to him. You can sit your two way deal player or cheap vet, if he svck, but if you pay actual money or want your player to develope, give him some time.
They gave him time and gain more context to make decision about him. I like this decision. Now pens will take more their own info about him. May be their coaches will help him with his weaknesses, and they understand that their coaches could do it. Their prospect pool is thin, Smith is a good addition for them. May be they have their own specific stats where Smith was good at, and they understand hiw they can use him.

 

[devilsblood]

I’d love to see what NHL teams actually use for analytics. For sure they are different than what we have access to . It would be very interesting to see what teams value what stats over others. And how much emphasis is put into basing decisions off of those stats .
Example - egghead stats person pushes fir player X in the lineup and put on line 3 with player Y and Z as stats show it “ should” be the most effective . How much weight dues that hold ? Or are stats used more do for just drafting and developing ? Or drafting and scouting when potentially trading for a player ?
What I don’t get , is when a player clearly is getting demolished by every stat metric , yet the team continues to play that player . Prime example - Ty Smith last season . From some stats the average joe had access to , he was one of the worst D men in the league ( with any sort of playing minutes ) . Why would he be trotted out there as much as he was then ? Stubbornness? Trying to see if he could fight his way through things and figure it out ? No other better options ?

Hoping Smith could figure it out was definitely a big part of the decision making. I doubt there was some stat suggesting he was actually playing well. The fact that he is now gone strongly suggests they knew he wasn’t playing well.

But Sal will give us a glimpse at some interesting stats which i imagine teams use. Like shots in the slot or cross ice pass completion percentage.

 

[njdevils1982]

Hoping Smith could figure it out was definitely a big part of the decision making. I doubt there was some stat suggesting he was actually playing well. The fact that he is now gone strongly suggests they knew he wasn’t playing well.

But Sal will give us a glimpse at some interesting stats which i imagine teams use. Like shots in the slot or cross ice pass completion percentage.

pretty sure thats why @Better Call Dano calls it "salanlysis" (i think thats how its spelled...but most likely not)

my version 'saltistics' was inspired by the above mentioned

 

[Xirik]

Fellow Devils fans! in two years we can finally fix one of the teams biggest plunders of the 90s.


Not Drafting Miroslav Satan!

Two years from now his son will be draft eligible where we can then draft him. Satan will finally come home.

 

[Guttersniped]

I’d love to see what NHL teams actually use for analytics. For sure they are different than what we have access to . It would be very interesting to see what teams value what stats over others. And how much emphasis is put into basing decisions off of those stats .
Example - egghead stats person pushes fir player X in the lineup and put on line 3 with player Y and Z as stats show it “ should” be the most effective . How much weight dues that hold ? Or are stats used more do for just drafting and developing ? Or drafting and scouting when potentially trading for a player ?
What I don’t get , is when a player clearly is getting demolished by every stat metric , yet the team continues to play that player . Prime example - Ty Smith last season . From some stats the average joe had access to , he was one of the worst D men in the league ( with any sort of playing minutes ) . Why would he be trotted out there as much as he was then ? Stubbornness? Trying to see if he could fight his way through things and figure it out ? No other better options ?

They have access to a lot of data we don’t.

There are private companies like Sport Logiq, which is where Sal gets his fun stats for the TV broadcast. Teams also produce their own.

And there’s the new data from the NHL’s puck tracking technology, not sure what that is or if/when the public will get to see some of it.

Some public stat folks track stats themselves for macro stats, often for players and prospects in other leagues too. People look at a lot aspects of the game. I do dislike when people think analytics ignore the complexities of the game when it’s quite the opposite.

Analytics are going to play some role in every team, how and to what degree is the only question.

With Smith the team didn’t really have much choice, there wasn’t any veteran depth options. I don’t think playing prospects would have worked out better. Smith’s role got more sheltered. Once the playoffs were out of hand you might as well play him, the other choice would have been to send Ty down to the AHL.