ANOTHER GRETZKY?

by Lowetide

”Everyone has a five-year plan, and so did we. It’s not like it fell exactly into place, but we were adding certain elements to get to this point.” -Peter Chiarelli, Behind the Moves (June 2011)

The Flaming rumor is interesting (Guy doesn’t break a lot of NHL stuff but his sources in hockey are outstanding). Gretzky is director of amateur scouting for the Bruins and of course has strong connections to Peter Chiarelli. The people I know in the industry tell me Gretzky is a bright guy who did good work with his staff in Boston (the Coyotes years were difficult, not sure about the dynamic there).

How many people does a general manager need to help in evaluation of players? He would need a capable amateur scouting director and staff; a pro scouting staff, that would include NHL observers and also men who could scout Sweden, Finland, Russia et cetera. Plus you would need an assistant GM and then one or two men who would form the internal brain trust. Right? I believe there will be another shoe to drop if this story has legs.

I wrote on this subject last week:

  • One of the things this blog remains concerned about is Peter Chiarelli’s staff. If you drop by the Oilers management page, the names and faces staring back at you look similar to the last regime and the regime previous to that and the one before that one, too. If we are to believe the Edmonton Oilers, the forensic files found that the true culprits for the last decade were Dallas Eakins, Stu MacGregor, Morey Gare, Brad Davis, Kent Hawley, Billy Moores—and Tyler Dellow.
  • I am not buying it. Peter Chiarelli is the general manager of the Edmonton Oilers, an extremely difficult job with enormous pressure. I hope this blog has established over these years that we find no joy in watching people lose their jobs, but at the same time that linked management page is atrocious based on what has happened in the last decade. Mr. Chiarelli, consider tearing down that management page—for your own sake.
  • Source

If the Gretzky hiring occurs, we can assume someone is moving out from the inner circle, the brain trust. I am not going to guess—if you have been following the Oilers for any length of time you know where the smart money is being laid.

CHIARELLI’S BIG EDMONTON MOVES

  • Trading for Griffin Reinhart
  • Trading for Cam Talbot
  • Signing Andrej Sekera
  • Trading Taylor Hall for Adam Larsson
  • Signing Milan Lucic

It would be ridiculous to imply these moves are anyone else’s responsibility—this is Peter Chiarelli’s Edmonton track record. That said, replacing some of the inner circle management in Edmonton is an idea that is long overdue. We hope they choose well, and assume the new hire will know why we have a Gretzky drive in the city.

85 comments
0

You may also like

0 0 votes
Article Rating
85 Comments
Newest
Oldest Most Voted
Inline Feedbacks
View all comments
Kosmic Burrito

Lowetide,

What you said is definitely true. I just think that I’m still stunned with some of their surprise moves during the 2015 draft. They definitely reached with a few picks. But in all fairness, you’re right. You can’t judge a draft only a year removed from it. At least a portion of my dreams came true. No more Howson at least. And Gretzky definitely had a hand in some good picks in Arizona as well. Only time will tell…I just hope that it’s a damn good time brother

New Improved Darkness

Woodguy:
Mangers in all industries will hire people they know.

From Give and Take by Adam Grant, chapter 3, concerning 38,577 procedures performed by 203 cardiac surgeons at forty-three different hospitals, over a two-year period.

When Huckman and Pisano examined the data, they discovered a remarkable pattern. Overall, surgeons didn’t get better with practice. They only got better at the specific hospital where they practiced. For every procedure they handled at a given hospital, the risk of patient mortality dropped by 1 percent. But the risk of mortality stayed the same at other hospitals. The surgeons couldn’t take their performance with them. They weren’t getting better at performing coronary artery bypass grafts. They were becoming more familiar with particular nurses and anesthesiologists, learning about their strengths and weaknesses, habits, and styles. This familiarity helped them avoid patient deaths, but it didn’t carry over to other hospitals. To reduce the risk of patient mortality, the surgeons needed relationships with specific surgical team members.

But wait, it gets worse.

While Huckman and Pisano were collecting their hospital data, down the hall at Harvard, a similar study was underway in the financial sector. … Star analysts carry superior knowledge and expertise that they should be able to use regardless of who their colleagues are. … “Analysts are one of the most mobile Wall Street professions because they expertise is portable. I mean, you’ve got it when you’re here and you’ve got it when you’re there. The client base doesn’t change. You need your Rolodex and your files, and you’re in business.” … Over the nine-year period, 366 analysts—9 percent—moved, so it was possible to see whether the stars maintained their success in new firms. … When star analysts moved to a different firm, their performance dropped, and it stayed lower for at least five years. … On average, firms lost about $24 million by hiring star analysts. Contrary to the beliefs of Fraenkel and other industry insiders, Groysberg and his colleagues conclude that “hiring stars is advantageous neither to stars themselves, in terms of their performance, nor to hiring companies in terms of their market value.”

The result from this second study, in particular, would need to control for regression to the mean, and that’s not discussed in this account (I have mixed feelings about this book, which I’ve elaborated previously).

I think this is just another example of how armchair analytics falls into the hockey card fallacy: that a team is the linear sum of separable parts.

I think the linear model is a fabulous thing to pursue, to inform the beginning of the conversation. But then we work so hard to get any kind of serviceable model, it becomes easy to mistake the linear model for the destination itself.

In particular, you see coaching and management people take heaps of abuse they don’t deserve. What’s not to like about thoughtless rage enablement? Seven billion brainstems can’t be wrong.

Story #1: the linear model dominates ultimate performance, and if the coaches could just figure this out and assign minutes accordingly, all good things would flow.

Story #2: the only purpose of non-linearity is to dupe the stupid people in coaching and management on nearly every other team (legendary figures and recent champions excepted so as to maintain the flogging standard). Hence horse-trader wiles: the showcase, the pump and dump, and keeping your cards close to the vest on organizational needs, even those more obvious to every observer than an Rwandan drummer missing both hands.

The truth is that management is about 80% coping with non-linear challenges, and the non-linear considerations are anywhere from 3 to 10 times intrinsically harder than managing linear terms.

No one really knows just how hard management is until they’ve had to manage thirty people who all adhere to the linear fallacy (multiply by another factor of two if you took the job because you yourself endorsed the linear fallacy—my how the clueless sleep less during that first difficult year).

One of the most significant forms of non-linearity concerns promotion: having someone in the organization who has built up large amounts of relationship equity performing in that environment and deciding how high and how soon to promote this person. Just a little bit too far, and soon you find your (formerly) esteemed teammate presiding over a raging tire fire. Drucker blames the person who made the promotion, and says that the fix is not to fire Tire Fire McBob, but to put TFM back into a position commensurate with the person’s true ability, to see whether this person can return to previous form (presumably this person got the promotion in the first place for demonstrated achievement).

Now of course, not all tire fires are created equal, and if you truly believe that TFM started the tire fire by carelessly playing with matches (that the worst of the tire fire was self-inflicted) then one must take that into account. Perhaps demonstrated achievement was a Potemkin village. There are many scenarios to contemplate.

But the first assumption after a tire fire—according to Drucker—should be “one promotion too many” and to put the person back into a position where they can regain success and become a valued team member once again.

The person who rages most foul against Drucker’s simple wisdom is the Queen of Taint. Management by bad smell. Purification rituals, just because. You owe me a rolling head, etc. etc. The narcissism is strong in this one.

Because, you see, the simple solution to purging the organization of its organization continuity and expertise is to procure more hockey cards. Just poach some star coach or GM from some other sad-sack team, mission accomplished. (Data not included, or even considered.)

Where non-linearity ascends all the way to the second-order batty rafters, is when you get into perception management (yet another term in a laundry list of “reasons other than winning” as defined competently and up close). “Without rolling some heads, we can’t be respected.” “By whom, precisely?” I’m inclined to ask back. “The face in the mirror?”

There is a lot to be said for bringing in new blood, especially if you’ve got an established positive working relationship, you know the context you’re bringing the person into—because you patiently assessed the dynamics for a year—and you’re filling the right hole for the right reason.

Or you can simply pronounce “finally the penny dropped for Chia, who now recognizes the incompetence of those he inherited”. Right. Because no-one has ever seen management by flamethrower go horribly wrong when a person gifted in the art of mulching red tunics rises to the top of a giant enterprise.

———

I listened to the following EconTalk yesterday:

Robert Solow on Growth and the State of Economics

What’s great about this episode is that Solow is 90 years old and seems not to have forgotten a thing.

For instance, just to give one example, a good friend of mine, Padma Desai, tried to use the empirical apparatus. She was a student of the Soviet Union, what was then the Soviet Union. And what she found was that, unlike the United States–and by the way, you might remember that in the Post-War years, the fact that the Soviet economy was eventually going to collapse was not so clear, and Khrushchev’s remark about how we’ll bury you, and he meant economically—but Padma Desai found that the Soviet Union was just the reverse of the United States, and that what increases in income per person had been achieved there, had been achieved by grinding down consumption and investing in heavy industry, especially an enormous amount of the national income. Which suggested right away that over the long haul, Khrushchev was not going to bury us, but the other way around. Which turned out to be the case. So, this got to be a popular thing.

Solow won a Nobel prize in part, I suspect, for his work overturning a traditional economic fable about the capital investment cycle.

And it’s amazing to think about now, but what I was able to find, for the numbers that I needed to put flesh on that theory, covered the period in the United States from 1909 to 1949. … But when I analyzed those time series, those data, it turned out that what the history of that period seemed to be saying was that in fact, almost all—more than 80%—of the long-term average growth of output per person, or income per person, could be traced to this broad sense of technology progress. And that even between 1909 and 1949, very little of the fairly enormous improvement in income per head and the standard of living in the United States, even over that period, only a relatively small amount came from the nitty gritty of saving and investment.

It’s generally easy for bullies to appear successful over shorter time frames, which provides one of the major foundation stones for scapegoat theory.

Without scapegoat theory, maybe only 10% of the population can claim any natural talent in the management profession. With scapegoat theory, 50% of the population can claim natural talent in the management sphere (many of these self-inclusionistas don’t even get so far as to view management as a intrinsically challenging profession). That’s a potent vanity multiplier of five fold. You do the math.

———

I read book after book after book, and they all seem to corroborate my dismal view of subject. I even go to the trouble of not believing half of what I read, no matter how eminent the source, and still it comes out the same.

I get weary at times. It’s almost enough to drive a person into the poisonous arms of post-modern critical analysis: that the actual product of professional sports is to enable those with more bravado than clue to feather themselves with a headdress of vanity bangles.

———

Human folly is surely the masterclass of the human condition. Not with another thousand productive years could I complete this study (my first step in the Methuselah Matriculation: invite Robert Solow to enroll as a freshman at the Institute of Finally Getting Down to It—IFI for short).

———

First semicentennial semester: The Confounds of Vanity 1001.

Solow: Don’t you think it’s kind of embarrassing at my age to be taking a course at the thousand level? Couldn’t we do something about the course numbering?

Me: I think I saw in the marketing faculty an introductory course offered as The Psychology of Influence 999.95. Maybe you should switch programs?

Solow: Have you checked your progressives lately? Was that the price or the course number?

Me: Pretty sure that was the course number.

Solow: Damn! Guess I’m going to have to suck it up and count my blessings, small though they appear to be. You think—are they going to offer up every course under a course number consisting solely of ones and nines and fives?

Me: Personally, I wouldn’t bet against it. They might even go so far as so construct a millennial grading system with no C or B students at all [*]. Anyway, welcome to your next 900. Glad you could join us.

Solow: My lord, I don’t have even the first idea how to become 900-years more gruff and cranky!

Me: I’m frightened, too.

———

[*] Available grades being 99.95, 99, 95, 91, 59, 55, 51, 19, 15, 11, 9, 5, and 1.

McSorley33

G Money,

My memory is bad was it his time on the farm or his partner?

I thought he played with JO after the farm. And had a slow footed partner prior to the farm.

Either way, GR needs a puck moving partner…..

Don’t get me wrong the time on the farm did not hurt GR either. …

G Money

semi legendary rot lobster,

Ah, sorry my good man, er lobster. I do not know what Reinhart eats, but I agree this factor needs to be at the top of the evaluation list.

G Money

VOR,

Smart. If I ever find my interest in hockey stats compromises my enjoyment of the game, I too will stop. As it is, I’m lucky in that I remain very distinctly a ‘fan’ and a ‘fanalyst’ with the first driving the second.

Honestly, the Oilers themselves remain a vastly bigger risk to my fandom than do the stats.

Re: WoodMoney, I should give a shoutout to you, because at least from my perspective, I’m certain it was a discussion with you (here) that helped solidify my areas of concern regarding weaknesses in modern possession/shot metrics.

You in particular were very vocal about how ineffective they could be with respect to defensemen. (Turns out that every stat, and the eyeball, all tend to have a tough time with defensemen …)

That extended (multi thread) conversation wasn’t the only thing, but it certainly played a part in convincing me I should set off on the path of understanding what worked and didn’t work w.r.t. to hockey metrics, and try to do something about it.

Both DangerFen and WoodMoney sit on that path (of which we’re probably at the middle at best).

I look forward to seeing your use of boolean algebra in scouting!

VOR

GMONEY, thanks for the explanation and the great link. That is one of the best statistical explanations I have ever read. Simple and clear without being simplified. I haven’t been following Manny’s work at all. I have to admit that I haven’t even been paying that much attention to yours and Woodguy’s. Yes, I know, a mortal sin.

Maybe more than most people here I get what a shit load of work you’ve put in. I’m deeply impressed by the commitment and the curiosity it had to take to do something on that scale. And it sounds like you are pressing on, pushing the envelope still further which I think is great, and kudos to all the people who have offered to help.

However, I came to realize that my obsession with possession metrics were becoming the lens through which I viewed hockey. It was ruining my joy as a fan. Worse it was obstructing my work on scouting and drafting, I effectively had blinders on. So for about 9 months now I haven’t read anything to do with possession metrics. I haven’t thought about or commented on other people’s work. I have stopped trying to analyze and gone back to trying to synthesize.i am back to actually enjoying watching hockey.

bobinyvr

Lowetide: Senior VP of European Hallway Celebration would be a dandy title.

How about Senior VP of Bold Moves and Hyperbole

Little Poteet

https://scoutingpost.com/2016/08/02/report-keith-gretzky-to-oilers-as-assistant-gm/
This article by Kirk Luedeke has good things to say about Keith Gretzkys time in Boston while questioning a couple choices and calls the coyotes years a mixed bag. A more optimistic take than many here have had

Kosmic Burrito

As long as MacT, or any of the old regime is shown the door…I’ll be ok with this. But if Gretzky had something to do with the Bruins last two drafts, well then I’ll be worried lol…

kinger_OIL

– PC says: ”Everyone has a five-year plan, and so did we. It’s not like it fell exactly into place, but we were adding certain elements to get to this point.”

– Mike Tyson says: “Everybody has a plan until they get punched in the mouth.””

semi legendary rot lobster

G Money,

u r forgeting most important question

does griffindor eat seafood

G Money

AsiaOil,

Not sure how easy this is to read, but this is a scuzzy little chart I threw together of Reinhart’s DangerFen over the course of the season.

http://i.imgur.com/8m2zozQ.png

The raw game by game DFF% is in the background in grey. The blue trace is his DFF smoothed with a Gaussian(2) smoother, which is the most awesome smoothing algorithm I’ve ever come across (as effective as a 25 game rolling average, yet somehow *way* quicker – and no startup delay).

The key thing to note is that, although the data is thin (only 29 games), there is a very distinct difference in character between his first 15 games (12-gap-3) and his last 14 games. The first 15, note the highs are lower, and the lows are lower.

The last 14 games, the highs are higher, and the lows are also (significantly) higher.

The smoother reflects that in being below 50% almost the entire pre-AHL split, and above 50% almost the entire post-AHL split.

His time on the farm made a *big* difference, folks.

There’s a player there. I don’t think he’ll be top pairing material, but Top 4 is more than possible. He may be stuck in the AHL next season because of leftorium and ELC bonus issues, but don’t write him off.

G Money

VOR,

Manny’s work is all multiple regression based. So I’m 100% certain that when he describes it as Boolean, it means he is treating special teams state not just as a categorical variable, but more specifically a dichotomous predictor variable (http://www.psychstat.missouristate.edu/multibook/mlt08m.html#BM02).

He’ll code it as 1 or 0 and throw that into the regression. I’m certain special teams is not the only such categorical variable that he uses.

If I recall correctly, since he’s predicting goals, I think he sets it up so his dependent variable is *also* categorical (goal or not goal), so he’s actually doing: a logistic multiple regression with multiple dichotomous predictor variables.

SwedishPoster

Isn’t Keith Gretzky the exec pretty much at the same level as Keith Gretzky the player, the difference being that as an exec you can get to the NHL on name alone.

AsiaOil

I’m not – he’s going to be a guy you can play with an offensive RHD on the 2nd pair and still come out ahead in your own end.

Woodguy: Oilers Dmen Expected Goals Against/60 4v5
(xGA/60 is a better than shots against per 60 as it weight the shots on type (slap, wrist etc) and location.)

PlayerxGA60
GRIFFIN.REINHART4.09
ANDREJ.SEKERA5.24
MARK.FAYNE6.58
BRANDON.DAVIDSON6.68
DARNELL.NURSE6.78
ERIC.GRYBA6.97
OSCAR.KLEFBOM7.15
JUSTIN.SCHULTZ7.83

Based on this metric, I ‘d argue that Gryba wasn’t one of the better PK defenders.

Really surprised where GR ranked.

VOR

Woodguy,

A Boolean variable is either true or false. How that relates to the data set in question I can’t imagine. But binary logic systems allow only two states at any moment. On-Off, Yes-No, etc. I have loved Boolean algebra since I first learned it at university eons ago. It is at the heart of the long term project I am doing on scouting.

Woogie63

The two managers PC has taken from Boston;

Wayne’s brother and his brother.

JD_Wry

I think that’s the thread here, folks. Chiarelli is unable to talk to Keith Gretzky or Scott Bradley at this time, so there’s a period of time where status quo makes sense. Fair?

From: https://lowetide.ca/2015/05/01/make-it-so/

May 1, 2015

spoiler

I’ve always thought this particular hire has been baked in the cake since Chia took over the steam engine.

russ99

Any move by Chiarelli to get his own guys in and get MacT and Howson’s voices out of the room is fine by me.

Even if it’s Wayne’s brother, which may be a selling point for Katz.

Lenny

Nostradumbass,

Yes he is, though Gretzky is actually the most positive about Seguin among him, Neely and Bradley (who I just saw has since been promoted to AGM). Bradley is the one saying Seguin doesn’t fit the culture or play the way the Brunis play, lacks toughness etc. Also says “I don’t like the way his game is going” after Seguin had a 67 point season in Year 2 of his NHL career, and 32 points in 48 games in year 3. That video is painful to watch.

Someone should have shown them this list of 20 year old seasons since 2000 where a player scored over 65 points:
Alex Ovechkin
Steven Stamkos
Ilya Kovalchuk
Evgeni Malkin
Anze Kopitar
Patrice Bergeron
Sidney Crosby
Patrick Kane
Nicklas Backstrom
Jonathan Toews
Matt Duchene
Tyler Seguin
John Tavares
Marian Gaborik

Mr DeBakey

Lowetide: Ha. No I have been doing all kinds of things around the house, but had no Saturday show and did not work today so extra time. ?

At my house, all sorts of yard work opportunities present themselves.
I look forward to living in a retirement community.
Soon.
Damn soon.

[…] Oilers blogger Lowetide speculates that a Keith Gretzky hiring could being a domino effect as Chiarelli swaps out pieces of the old guard for “his guys.” […]

Diesel

Nostradumbass:
Isn’t Keith Gretzky the “he won’t pay the price” guy in the Seguin trade video?

ughhhhhhhhh(h x 100000)

stevezie

Woodguy: I’m not saying its right, just that its not exclusive to Oiler management.

i completely agree.

Gayfish

Metrics are bs. There is a reason football coaches don’t extensively code every game and look for patterns. The can just have a feeling. (Or film hand signals)

Nostradumbass

Isn’t Keith Gretzky the “he won’t pay the price” guy in the Seguin trade video?

JD_Wry

flyfish1168:
Since Gretz is currently employed by the Bruins, does this mean we owe the Bruins another draft pick? If he is hired

The rule has been 86’d, but teams owing picks are still being held to that.

G Money

Woodguy: G,

If we run 5v4 numbers, I wonder if we need to change the “elite” dynamic? (i.e, does 1 elite and 2 middle still equal elite?)

I wonder if we need to use purely 5v4 metrics to create an “elite 5v4” list?

As we know many “elite” players aren’t good on the PP. Hall being a prime example.

Iz goot question!

I think for sure special teams need their own special criteria and processing because, you know, they are *special*.

Ryan

Woodguy:
Ryan,


How is it that you make me feel like you’re an old friend and we’ve never met?

Because I’m sweet and unassuming.

Also, Reinhart’s shots against /60 the previous season was 83.4 or also worst on the Islanders.

You know he played under 5 minutes on the PK that year right?

I know I like small samples too much, but that’s infinitesimal.

This is a funny place.

I know your thought tendancies better than people whom I actually know.

Since you’re so busy writing checks, I’ll buy the first few rounds when we actually do go for a beer sometime.

Ryan

Woodguy: Small sample (50 minutes) and played 2PK and having a good run probably make up most of it.

I wouldn’t bash McLellan for not playing him more.

Also,

Since you are quoting TOI/60, I’m going to assume you are looking at behindthenet.ca

Adam Pardy’s TOI/60 includes his WPG games.

Klefbom played while Reinhart was on the farm.

So Reinhart was #4PK (on PK2)while he was on the roster.

I wasn’t bashing Mclellan.

Agreed small sample and 2 Pk and having a good run makes the most of it.

Funny to notice the venerable shot suppression beast, Taylor Chorney had 34.1 shots against / 60 while playing 54 games but somewhat less toi/60 than GR.

Woodguy

Ryan:
Anyway, either Reinhart is just an absolute beast of an NHL penalty killer or something else is driving his results.Given that he was sixth in Pk toi/60 on the Oilers, maybe Mclellan is too stupid to recognize his gord like Pk abilities or some other factors are in play.

Small sample (50 minutes) and played 2PK and having a good run probably make up most of it.

I wouldn’t bash McLellan for not playing him more.

Also,

Since you are quoting TOI/60, I’m going to assume you are looking at behindthenet.ca

Adam Pardy’s TOI/60 includes his WPG games.

Klefbom played while Reinhart was on the farm.

So Reinhart was #4PK (on PK2) while he was on the roster.

Victoria Oil

LT – you’ve been uber productive lately with all your regular blog postings plus a bunch of ON posts. We greatly appreciate your work – but have you been bored this weekend in E-town?

Woodguy

Ryan,


How is it that you make me feel like you’re an old friend and we’ve never met?

Because I’m sweet and unassuming.

Also, Reinhart’s shots against /60 the previous season was 83.4 or also worst on the Islanders.

You know he played under 5 minutes on the PK that year right?

I know I like small samples too much, but that’s infinitesimal.

flyfish1168

Since Gretz is currently employed by the Bruins, does this mean we owe the Bruins another draft pick? If he is hired

Ryan

Anyway, either Reinhart is just an absolute beast of an NHL penalty killer or something else is driving his results. Given that he was sixth in Pk toi/60 on the Oilers, maybe Mclellan is too stupid to recognize his gord like Pk abilities or some other factors are in play.

Water Fire

Ryan: How is it that you make me feel like you’re an old friend and we’ve never met?

I saw you headed for the shots against / 60 4v5 before I finished my post.

While I was already aware of Reinhart’s results for the Oilers, I maintain that his success is driven by other factors… Small sample, second pp units, luck etc.

Also, Reinhart’s shots against /60 the previous season was 83.4 or also worst on the Islanders.

The call is coming from in the house!

Ryan

Woodguy:
Same list of Dmen 4v5.

TOI 4v5

PlayerTOI
DARNELL.NURSE150.11
ANDREJ.SEKERA139.44
BRANDON.DAVIDSON135.12
ERIC.GRYBA126.21
MARK.FAYNE122.27
OSCAR.KLEFBOM81.81
JUSTIN.SCHULTZ56.45
GRIFFIN.REINHART50.86

Shots Against/60 4v5
PlayerSA60
GRIFFIN.REINHART35.39
ANDREJ.SEKERA41.31
MARK.FAYNE47.11
DARNELL.NURSE49.96
BRANDON.DAVIDSON50.18
OSCAR.KLEFBOM53.54
ERIC.GRYBA56.57
JUSTIN.SCHULTZ57.4

4v5 Fenwick Against/60 (I like Fenwick as a PK metric as it gives credit for blocked shots)
PlayerFA60
GRIFFIN.REINHART51.91
ANDREJ.SEKERA53.36
MARK.FAYNE62.81
DARNELL.NURSE71.15
ERIC.GRYBA74.64
BRANDON.DAVIDSON75.49
OSCAR.KLEFBOM76.27
JUSTIN.SCHULTZ79.72

Note to Ryan: Both shots and Fenwick are SV% independent too

How is it that you make me feel like you’re an old friend and we’ve never met?

I saw you headed for the shots against / 60 4v5 before I finished my post.

While I was already aware of Reinhart’s results for the Oilers, I maintain that his success is driven by other factors… Small sample, second pp units, luck etc.

Also, Reinhart’s shots against /60 the previous season was 83.4 or also worst on the Islanders.

Water Fire

frjohnk: What’s league average and what did Larsson post?
I’m on my EBERLE phone.

EBERLE hates poutine.

Woodguy

frjohnk: What’s league average and what did Larsson post?
I’m on my EBERLE phone.

League median (100+ min on PK for Dmen) for xGA/60 is 6.25, Larsson is 6.74

Also,

When I sorted the list by xGA/60 I noticed a lot of Dmen who *I know* play 2PP populated the top of the list (least xGA/60)

That makes intuitive sense as the 1PP is usually more loaded with talent than 2PP, so 2PK will probably put up better numbers.

Probably plays a role in Reinhart’s numbers as well (as well as his small 50 min sample)

G,

If we run 5v4 numbers, I wonder if we need to change the “elite” dynamic? (i.e, does 1 elite and 2 middle still equal elite?)

I wonder if we need to use purely 5v4 metrics to create an “elite 5v4” list?

As we know many “elite” players aren’t good on the PP. Hall being a prime example.

Woodguy

Lowetide:
Note: Woodguy posted the McKenzie item before me, but I changed his time so mine shows first. I am very immature, but promise I have not done this before.

My post is SV% independent too.

Woodguy

@TSNBobMcKenzie
Just saw some Keith Gretzky to EDM as AGM rumors. It’s my understanding he will, in fact, be joining EDM in that capacity in near future.

Woodguy

G Money:
Woodguy,

It means special teams were a categorical variable in the regression.So in theory he should have differentiated results for PP/PK than for EV, which is certainly much better than not including it at all.

I’m of the opinion based on my review of shot locations and probabilities that special teams are an entirely different animal, so I will continue to be skeptical that that alone will give you good results.And as always, is entirely reliant on actually working the right process for multiple regression, which … well, you know.

Yeah, probably best.

Should crack the whip on our regression specialist to do that for PP/PK DFF.

Woodguy

Same list of Dmen 4v5.

TOI 4v5

Player TOI
DARNELL.NURSE 150.11
ANDREJ.SEKERA 139.44
BRANDON.DAVIDSON 135.12
ERIC.GRYBA 126.21
MARK.FAYNE 122.27
OSCAR.KLEFBOM 81.81
JUSTIN.SCHULTZ 56.45
GRIFFIN.REINHART 50.86

Shots Against/60 4v5
Player SA60
GRIFFIN.REINHART 35.39
ANDREJ.SEKERA 41.31
MARK.FAYNE 47.11
DARNELL.NURSE 49.96
BRANDON.DAVIDSON 50.18
OSCAR.KLEFBOM 53.54
ERIC.GRYBA 56.57
JUSTIN.SCHULTZ 57.4

4v5 Fenwick Against/60 (I like Fenwick as a PK metric as it gives credit for blocked shots)
Player FA60
GRIFFIN.REINHART 51.91
ANDREJ.SEKERA 53.36
MARK.FAYNE 62.81
DARNELL.NURSE 71.15
ERIC.GRYBA 74.64
BRANDON.DAVIDSON 75.49
OSCAR.KLEFBOM 76.27
JUSTIN.SCHULTZ 79.72

Note to Ryan: Both shots and Fenwick are SV% independent too

G Money

Woodguy,

It means special teams were a categorical variable in the regression. So in theory he should have differentiated results for PP/PK than for EV, which is certainly much better than not including it at all.

I’m of the opinion based on my review of shot locations and probabilities that special teams are an entirely different animal, so I will continue to be skeptical that that alone will give you good results. And as always, is entirely reliant on actually working the right process for multiple regression, which … well, you know.

G Money

Mr DeBakey,

I think I’m beginning to understand his pen name. It’s not a reference to the Oilers, but a reference to an actual event*.

“Hey, the SCADA system was showing massive pressure fluctuations on the pipeline segment about 90km from Peace River all night! Why didn’t you hit the alarm!?!”

“Skay da? Hoozat?”

“Huh? What the hell’s the matter with you? That’s the system that records and displays all the flow and pressure sensors on our pipelines. You’re supposed to be monitoring it. It’s the backbone of our entire optimization and HSE program.”

“Monitoring? You means it like measures things and I have to watch those? Like metrics?”

“Yeah. How can you possibly not know this by now? The metrics are showing huge pressure problems. We almost for sure have a spill on our hands.”

“Bah! Fool! Who use metrics! I seen that pipeline good just last week! Spent almost ten minutes!”

“WTF? You can’t see a pipeline good. That fucker is hundreds of kilometres long, and carries almost 200,000 barrels a day!”

“Metrics er useless, dummy. I know bettern you. It’s obvierse to anyones. Smrt guys like me who know everythin’ about pipelines know only fools use metrics.”

“Jesus F Christ. We’re in deep shit.”

*One massive oil spill later*

“So… Mr. ‘metrics are useless’ … got any thoughts on the massive spill we just had that we could have prevented IF YOU’D PAID ATTENTION TO THE FUCKING METRICS?”

“What? I didn’t see any oil, so therefore there wunt no spill. It’s eyes that matter, not your stupid numbers! How many times I gotta tell you!”

“You know what? You’re union so I can’t fire you. But from now on, I want you to name yourself after this incident, just to remind everyone not to listen to a word you say or put you in charge of anything important. Now get the fuck out of here.”

“Hmmf. All u stupids. I gonna go on Lowetide and explain those dummies why they doan need metrics either.”

*Only the spill is true, the rest is made up. Any resemblance to any actual Lowetide poster is purely coincidental.