HFBoards

Go Back   HFBoards > NHL Western Conference > Central Division > Winnipeg Jets
Mobile Hockey's Future Become a Sponsor Site Rules Support Forum vBookie Page 2
Notices

Optimizing the Jets Roster Series

Reply
 
Thread Tools
Old
09-29-2014, 11:44 AM
  #1
garret9
AKA#VitoCorrelationi
 
garret9's Avatar
 
Join Date: Mar 2012
Location: Vancouver
Posts: 17,587
vCash: 500
Optimizing the Jets Roster Series

Hey guys,

I have a 5 part series that runs starting today. The series uses multiple statistics such as turnovers per puck touch, successful zone exits, zone entries, zone entry break-ups, face off conditional Corsi%, and other numbers to find how the Jets could improve their results without having to add more talent.

The first article in the series is Left-Handed Defensemen.
Quote:
Over last week we've shown the effects of zone deployment with comparing Byfuglien the forward vs the defender. We continue this in depth study by turning our focus to the left side defensemen.
The second article in the series is Right-Handed Defensemen.
Quote:
The Jets have three right-handed defensemen with NHL experience under contract. In descending 5v5 TOI per a game last season, they are: Zach Bogosian, Jacob Trouba, and Paul Postma. The first two defensemen have sufficient sampling for all discussions to follow, while the later one will be a lot more limited. There's also Dustin Byfuglien... but we'll ignore him for now.
The third article in the series is Centres.
Quote:
The Winnipeg Jets have four centres with NHL experience under contract who look to have earned full time spots. In descending 5v5 TOI per a game last season, they are: Bryan Little, Mark Scheifele, Mathieu Perreault, and Jim Slater.

Perreault comes as a new addition for the Jets, so his numbers are skewed from playing under a different system, with different teammates, and also being compared against other players.
The fourth article in the series is Left Wing.

The fifth article in the series is Right Wing.

As always: thoughts, comments, critiques, questions, etc. are encouraged.


Last edited by garret9: 10-07-2014 at 04:14 PM.
garret9 is offline   Reply With Quote
Old
09-29-2014, 12:46 PM
  #2
Unholy goalie
Registered User
 
Unholy goalie's Avatar
 
Join Date: Jul 2011
Posts: 499
vCash: 862
Quote:
Originally Posted by garret9 View Post
Hey guys,

I have a 5 part series that runs starting today. The series uses multiple statistics such as turnovers per puck touch, successful zone exits, zone entries, zone entry break-ups, face off conditional Corsi%, and other numbers to find how the Jets could improve their results without having to add more talent.

The first article in the series is Left-Handed Defensemen.


The second article in the series is Right-Handed Defensemen.

The third article in the series is Centres.

The fourth article in the series is Left Wing.

The fifth article in the series is Right Wing.

As always: thoughts, comments, critiques, questions, etc. are encouraged.
Great stuff! Can't wait for the rest!

I'm liking this focus on zone deployment more and more as it allows for much more nuanced analyses that connect nicely with coaching options.

Now that you have been doing this for a while, could you comment on what might constitute a significant difference between individual/team averages when it comes to the post-faceoff data? Not sure if this has already been addressed elsewhere.

I'm also wondering a bit more about the zone entry analysis. You ask why Pardy has had a higher percentage of controlled entries than Stuart, but it seems obvious when you consider how much Pardy carries the puck whereas Stuart dumps. It seems that the percentages for controlled entry are calculated by dividing the carries by total entries. This biases carries (rightfully). In my mind maybe there should be some allowance for dump-ins that have been retrieved to be considered "controlled entry", or else the answer to your question is simply that Stuart should stop dumping and start carrying the puck over. If we also accept that Pardy and Stuart are near equal in skating, maybe skating isn't as much of a factor in carrying in?

Unholy goalie is offline   Reply With Quote
Old
09-29-2014, 12:54 PM
  #3
garret9
AKA#VitoCorrelationi
 
garret9's Avatar
 
Join Date: Mar 2012
Location: Vancouver
Posts: 17,587
vCash: 500
Significance is calculated using standard tests to figure out what's the probability of that difference just by chance.

The Stuart bit is me trying to guide people to figure out things on his own while I guide them in stead of telling them what is.

garret9 is offline   Reply With Quote
Old
09-29-2014, 01:24 PM
  #4
Unholy goalie
Registered User
 
Unholy goalie's Avatar
 
Join Date: Jul 2011
Posts: 499
vCash: 862
Quote:
Originally Posted by garret9 View Post
Significance is calculated using standard tests to figure out what's the probability of that difference just by chance.
Sure.. so maybe I didn't articulate myself properly. Have you looked into establishing a more concrete threshold at which point we could say that CF/60 and WPG CF/60 are close enough to be "pretty much the same" or far enough to be "significantly less/more"? Not a criticism, but just a question since you're pretty thorough. Significance tests are at the limit of my understanding so maybe I'm shooting in the dark here.. feel free to call a spade a spade.

Quote:
Originally Posted by garret9 View Post
The Stuart bit is me trying to guide people to figure out things on his own while I guide them in stead of telling them what is.
Oops.. Probably more effective than my blundering idiocy..

It would be interesting nonetheless to see if there's some value in reframing retreived dump-ins. Maybe a good defender carries in more often than not, but a better defender can also read when there's a high probability of retrieval, thus effectively starting the zone time closer to the net. Just a thought..

Anyway, overall this is a must-read for the board imo. Interested to see the RHD article re: Bogo.

Unholy goalie is offline   Reply With Quote
Old
09-29-2014, 02:02 PM
  #5
pucka lucka
Registered User
 
pucka lucka's Avatar
 
Join Date: Apr 2010
Location: Ottawa
Posts: 4,219
vCash: 50
So glad we signed Stuart for 3 more years. We need intangibles that don't provide any tangible benefit.

pucka lucka is offline   Reply With Quote
Old
09-29-2014, 04:30 PM
  #6
garret9
AKA#VitoCorrelationi
 
garret9's Avatar
 
Join Date: Mar 2012
Location: Vancouver
Posts: 17,587
vCash: 500
I've shown Stuart has some uses

garret9 is offline   Reply With Quote
Old
09-29-2014, 05:19 PM
  #7
CaptainChef
Registered User
 
CaptainChef's Avatar
 
Join Date: Jan 2014
Location: Bedroom Jetsville
Country: Canada
Posts: 5,864
vCash: 500
Quote:
Originally Posted by supahdupah View Post
So glad we signed Stuart for 3 more years. We need intangibles that don't provide any tangible benefit.
Hey I'm a heck of a lot happier that they signed Stu for three than Thorbs. At least Stu provides them some positive results (if he is used in the right situations) and we do need some depth at that position. Will be interesting to see where Thorbs excels -- I suggest nowhere but maybe that's my bias

CaptainChef is online now   Reply With Quote
Old
09-29-2014, 05:32 PM
  #8
Hunter368
Canada
 
Hunter368's Avatar
 
Join Date: Nov 2011
Posts: 8,288
vCash: 500
Quote:
Originally Posted by garret9 View Post
I've shown Stuart has some uses
Paul talked about Stu being much more productive/successful in certain situations.

Hunter368 is offline   Reply With Quote
Old
09-29-2014, 06:14 PM
  #9
garret9
AKA#VitoCorrelationi
 
garret9's Avatar
 
Join Date: Mar 2012
Location: Vancouver
Posts: 17,587
vCash: 500
Quote:
Originally Posted by troubabooster View Post
Hey I'm a heck of a lot happier that they signed Stu for three than Thorbs. At least Stu provides them some positive results (if he is used in the right situations) and we do need some depth at that position. Will be interesting to see where Thorbs excels -- I suggest nowhere but maybe that's my bias
I don't like giving the punchlines for free but:

Statistically
Stuart overall is a middling 3rd pair d-man
Thorburn isn't really a 4th line player

garret9 is offline   Reply With Quote
Old
09-30-2014, 11:21 AM
  #10
garret9
AKA#VitoCorrelationi
 
garret9's Avatar
 
Join Date: Mar 2012
Location: Vancouver
Posts: 17,587
vCash: 500
Right handed defensemen:
http://www.arcticicehockey.com/2014/...ing-defensemen

garret9 is offline   Reply With Quote
Old
09-30-2014, 12:31 PM
  #11
pucka lucka
Registered User
 
pucka lucka's Avatar
 
Join Date: Apr 2010
Location: Ottawa
Posts: 4,219
vCash: 50
You really want an NHL job. I love the data appealing to coaches. I think if you became more edgy you'd have great success. By edgy I mean rude

pucka lucka is offline   Reply With Quote
Old
09-30-2014, 12:36 PM
  #12
castle
Registered User
 
castle's Avatar
 
Join Date: Dec 2011
Country: Canada
Posts: 1,509
vCash: 1000
About Enstrom....

Quote:
Another area where some video work could be focused is on the first 10 seconds of an offensive zone face off win.
It seems to me that the reason Enstrom shows his sustained differential in CF is because he doesn't always shoot the puck right away after a OZ faceoff. If he started doindg that more, the Jets lose possession more and his later CF differential would start to disappear. So the real question is... do teams score more often per shot in the first 10 seconds, or more often after passing the puck around a bit? It would be interesting to know the actual numbers. We're all familiar with the quick shot goals, but what's the ratio of those to the 'puck goes off a leg/stick/skate/arm and the other team is breaking out'?

castle is offline   Reply With Quote
Old
09-30-2014, 12:40 PM
  #13
pucka lucka
Registered User
 
pucka lucka's Avatar
 
Join Date: Apr 2010
Location: Ottawa
Posts: 4,219
vCash: 50
Quote:
Originally Posted by castle View Post
About Enstrom....



It seems to me that the reason Enstrom shows his sustained differential in CF is because he doesn't always shoot the puck right away after a OZ faceoff. If he started doindg that more, the Jets lose possession more and his later CF differential would start to disappear. So the real question is... do teams score more often per shot in the first 10 seconds, or more often after passing the puck around a bit? It would be interesting to know the actual numbers. We're all familiar with the quick shot goals, but what's the ratio of those to the 'puck goes off a leg/stick/skate/arm and the other team is breaking out'?
That's not really what happens. His non-shooting should improve possession (shot totals) in aggregate for those playing with him. Controlled possession leads to more shots. Shooting in general doesn't improve possession numbers. Better puck possession improves shooting numbers. You have it backwards.

pucka lucka is offline   Reply With Quote
Old
09-30-2014, 01:03 PM
  #14
CaptainChef
Registered User
 
CaptainChef's Avatar
 
Join Date: Jan 2014
Location: Bedroom Jetsville
Country: Canada
Posts: 5,864
vCash: 500
Quote:
Originally Posted by supahdupah View Post
You really want an NHL job. I love the data appealing to coaches. I think if you became more edgy you'd have great success. By edgy I mean rude
I agree. Lets hear what you really think on some of these political/odd decisions rather than fence-sitting.

CaptainChef is online now   Reply With Quote
Old
09-30-2014, 01:07 PM
  #15
CaptainChef
Registered User
 
CaptainChef's Avatar
 
Join Date: Jan 2014
Location: Bedroom Jetsville
Country: Canada
Posts: 5,864
vCash: 500
and I have to say that I would feel so much better if you were on the Jets staff because I think it is becoming more and more obvious that they have nobody on their staff that gives them these sort of analyses

CaptainChef is online now   Reply With Quote
Old
09-30-2014, 01:14 PM
  #16
Hank Chinaski
Global Moderator
No mistakes...
 
Hank Chinaski's Avatar
 
Join Date: May 2007
Location: YTH
Country: Canada
Posts: 16,924
vCash: 50
Quote:
Originally Posted by supahdupah View Post
You really want an NHL job. I love the data appealing to coaches. I think if you became more edgy you'd have great success. By edgy I mean rude
C'mon now, TNSE loves their folksy, understated people. We want him working for them, not in Toronto or Vancouver.

Great stuff as always, garret.

Hank Chinaski is offline   Reply With Quote
Old
09-30-2014, 01:22 PM
  #17
CaptainChef
Registered User
 
CaptainChef's Avatar
 
Join Date: Jan 2014
Location: Bedroom Jetsville
Country: Canada
Posts: 5,864
vCash: 500
Here's a thought based on the article. is there any way we could make a trade and get big Buff the defenceman?

Pretty evident he is dominant, in fact head and shoulders better than either Bogo or Trouba in certain situations.

CaptainChef is online now   Reply With Quote
Old
09-30-2014, 01:26 PM
  #18
garret9
AKA#VitoCorrelationi
 
garret9's Avatar
 
Join Date: Mar 2012
Location: Vancouver
Posts: 17,587
vCash: 500
Quote:
Originally Posted by castle View Post
About Enstrom....



It seems to me that the reason Enstrom shows his sustained differential in CF is because he doesn't always shoot the puck right away after a OZ faceoff. If he started doindg that more, the Jets lose possession more and his later CF differential would start to disappear. So the real question is... do teams score more often per shot in the first 10 seconds, or more often after passing the puck around a bit? It would be interesting to know the actual numbers. We're all familiar with the quick shot goals, but what's the ratio of those to the 'puck goes off a leg/stick/skate/arm and the other team is breaking out'?
There are times I know the answer but I'm trying to shape the information and questions so people can figure it out on their own without me having to bat them on the head with it.

Your hypothesis on Enstrom holding being reluctant to shoot does seem reasonable. Teams usually try yo get the puck to then defenders right away to shoot, with the forwards moving towards the net to screen, deflect, or pick up the rebound. From my video analysis it does seem very plausible. Enstrom prefers to move the goalie laterally one way then send a cross ice pass to Buff or the RWer for a one timer.

We do know that sh% is slightly higher post faceoff and gradually moves back to norm. However, this is due to the way most teams play. Since Enstrom plays differently it is possible the curve is different for him.

I don't currently agree with your hypothesis on if Enstrom shot initially his sustain effect would disappear (unless he started shooting recklessly). From my observations one of the big reasons is he holds the blue line well and he can join the cycle. Very few Jets do that as often and only Buff did it more effectively.

garret9 is offline   Reply With Quote
Old
09-30-2014, 04:33 PM
  #19
truck
HFB Partner
 
truck's Avatar
 
Join Date: Jun 2012
Posts: 9,464
vCash: 500
Every Jets fan should be reading every article to get smarter.

truck is offline   Reply With Quote
Old
09-30-2014, 10:44 PM
  #20
garret9
AKA#VitoCorrelationi
 
garret9's Avatar
 
Join Date: Mar 2012
Location: Vancouver
Posts: 17,587
vCash: 500
Thanks for the kind words guys. It is a lot of work.
Centres tomorrow.

garret9 is offline   Reply With Quote
Old
10-02-2014, 12:25 AM
  #21
StefanW
HFB Partner
 
StefanW's Avatar
 
Join Date: Mar 2013
Location: Ottawa
Country: Canada
Posts: 6,277
vCash: 50
This was an excellent read. You did a good job integrating the new microstats that have recently become available with other measures. Post faceoff rates are always intriguing, although I still haven't decided whether D Zone faceoff rates are the best measure for D who do not take draws or go out to cover the point after a faceoff loss. It does kind of have to be there for purposes of what you are doing though.

Solid, and lots of useful information.

StefanW is offline   Reply With Quote
Old
10-02-2014, 12:36 AM
  #22
garret9
AKA#VitoCorrelationi
 
garret9's Avatar
 
Join Date: Mar 2012
Location: Vancouver
Posts: 17,587
vCash: 500
Quote:
Originally Posted by garret9 View Post
Thanks for the kind words guys. It is a lot of work.
Centres tomorrow.
Delay due to life going upside down.

garret9 is offline   Reply With Quote
Old
10-07-2014, 04:15 PM
  #23
garret9
AKA#VitoCorrelationi
 
garret9's Avatar
 
Join Date: Mar 2012
Location: Vancouver
Posts: 17,587
vCash: 500
Sorry about the delay... that's life...


Centres:
http://www.arcticicehockey.com/2014/...mization-right

garret9 is offline   Reply With Quote
Old
10-07-2014, 04:27 PM
  #24
YWGinYYZ
Mod Supervisor
 
YWGinYYZ's Avatar
 
Join Date: Jul 2011
Location: Toronto
Country: Canada
Posts: 23,259
vCash: 50
Optimizing our centres is easy - clone Little 3 times, and roll him on all 4 lines.

Nice article - it'll be interesting to see how Scheifele does this year, especially with 'new' linemates as you suggest. It'll also be interesting to see if Perreault can keep up his very good stats. As you mention, his numbers were also good in WSH, so it might not just be from playing in Anaheim.

YWGinYYZ is online now   Reply With Quote
Old
10-07-2014, 04:33 PM
  #25
Master Scheif
he's cookin'
 
Master Scheif's Avatar
 
Join Date: Jun 2014
Location: Winnipeg
Country: Canada
Posts: 1,577
vCash: 700
Great article! I'm not sure if you can edit them after posting, but I couldn't help but notice both the "We continue this in depth study, now turning our focus to the right side defensemen." at the beginning, and the part about Denial of Opposition Zone Entry where that block just trails off with the last word being "they".

I think this will be interesting to revisit next year when we can observe Scheif's development. I'm also very happy with the Perreault acquisition.

Master Scheif is online now   Reply With Quote
Reply

Forum Jump


Bookmarks

Thread Tools

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off



All times are GMT -5. The time now is 11:14 AM.

monitoring_string = "e4251c93e2ba248d29da988d93bf5144"

vBulletin Copyright ©2000 - 2016, Jelsoft Enterprises Ltd.
HFBoards.com is a property of CraveOnline Media, LLC, an Evolve Media, LLC company. 2016 All Rights Reserved.