Kukla's Korner

The Puck Stops Here

Point Shares: A Sabermetric Goal

One of the big achievements in baseball sabermetrics is Bill James win shares system.  This is a system that assigns wins to the individual players on baseball teams. It works very well to rate individual players.  One goal in hockey sabermetrics is to come up with a similar system.  Of course there are differences between hockey and baseball which make building a hockey system more difficult, but that hasn`t stopped Justin Kubatko at hockey-reference.com from trying.  I am going to look at his system and its results.  I think it is an important goal, but we are not able to do it properly yet.  As a result, he has built an approximate system which can give unreasonable results in some cases.

Kubatko makes three conceptual changes from the Bill James system.  First in the baseball system, a win is worth 3 win shares.  In the hockey system one point in the standings is worth one point share.  The hockey version of this is problematic because of point scheme that the NHL uses.

In the NHL, some games are worth two total points and others are worth three.  The extra point occurs only if the game is tied at the end of regulation time.  This is not something that is caused by any individual player, yet there is an extra point available that will be credited to individual players becomes available.  This point is often given out for pretty arbitrary reasons - winning a shootout.  I think that it is a failure to try to link up a point share system with the NHL`s point system because it is a poorly defined system.  It is probably most accurate to link it to a more logical point system for teams.  Perhaps two points for a win, one point for a game that goes to a shootout (win or loss) and no points for loss (including an overtime loss) is more meaningful.

Kubatko also breaks the perfect link between wins and win shares.  In baseball, a win is worth exactly three win shares.  A team that wins 100 games has exactly 300 win shares among their players.  In the hockey system this is approximate.  A team with 100 points will have about 100 point shares, but it may be a slightly higher or lower total.  This is a necessity in part because of the NHL point system.  It isn`t possible to meaningfully have some games worth two point shares and others worth three for a reason so arbitrary as if it is tied at the end of regulation.

The third difference is that Kubatko allows for players to have negative point shares.  If a player plays poorly enough to cost his team wins (for example Brian Elliott), he should be given a negative point share total because he cost his team points.  In the baseball system, the worst total a player can have is zero win shares regardless of how poorly he plays.  This is something problematic with the baseball system that is potentially improved here.

The calculation of point shares is approximate and it becomes more and more approximate as we go back into history where sufficient stats (such as ice time) were not properly recorded.  In the future I will discuss the details of these calculations and look at some of its results and the systematic biases that were involved.

Filed in: | The Puck Stops Here | Permalink



A total aside, I would like to know how sabremetrics would have worked for teams like the 80s Oilers who had multiple hall of famers at the same position. Would it have theoretically favored one of Gretzky or Messier due to differences in quality of opposition?

Clearly you wouldn’t have this data, but would it be fair to assume this would relate similarly to Malkin and Crosby, Kesler and Sedin?

Posted by Brad from Calgary on 08/25/11 at 03:25 PM ET


I don’t know, but the game was different back then. Those Oilers only gathered 51% of the shots.

Posted by Ralph on 08/25/11 at 07:53 PM ET

PuckStopsHere's avatar

Sufficient data does not exist to do some of the things with statistics that we do today with the 1980s.  Figuring out Corsi ratings etc from the 80s would require a lot of work going through game tapes and would likely still be incomplete. 

Some preliminary studies have been done (I am not aware of any that incorporate the majority of their games), but I think the moral of the story is that Edmonton dominated by any possession metric when the game was close or tied, but they played a lot of time with huge leads and like any team, the Oilers sat on those leads and took very few shots (only the high quality ones) while letting their opponents take a lot generally low percentages shots.

That said, I don’t think it answers Brad’s question.  I think he is asking what happens if a team has a lot of all star calibre players at the same position.  No one player would get as much ice time because of the presence of the others and thus the totals of any one player would be less than it would be without the others.  Depending upon the way they are used, we might see that one player gets to play against weak opposition or against strong opposition or we might not.  I think a lot of what we would see is how well these players work together.  Presumably these players would know that they would have significantly more playing time in essentially any other situation in the league and their reaction to that situation would influence their numbers.  If the player was unhappy in that situation, their numbers would decline and if the player remained happy they wouldn’t.

Posted by PuckStopsHere on 08/25/11 at 09:19 PM ET

Lindas1st's avatar

I would like to know how sabremetrics would have worked for teams like the 80s Oilers who had multiple hall of famers at the same position. 
Posted by Brad

Here’s the 83-84 Oiler’s (highest scoring team of all-time and SC Champ) player stats with each player’s Points Share listed last.

Player gp, g-a-pts, +/-, ps
Gretzky 74, 87-118-205, +76, 19.6
Coffey 80, 40-86-126, +52, 14.3
Kurri 64, 52-61-113, +38, 10.1
Messier 73, 37-64-101, +40, 8.0
Anderson 80, 54-45-99, +41, 8.8
Linsman 72, 18-49-67, +30, 4.1
Hughs 77, 27-28-55, +18, 3.4
Hunter 80, 22-26-48, +25, 2.6
Lowe 80, 4-42-46, +37, 5.5
Huddy 75, 8-34-42, +50, 6.2
Gregg 80, 13-27-40, +40, 5.9
Lindstrom 73, 22-16-38, +18, 2.0
Pouzar 67, 13-19-32, +17, 1.2
McClellend 52, 8-20-28, +9, 0.9
Lumley 56, 6-15-21, +14, 0.2
Fogolin 80, 5-16-21, +33, 3.6
Jackson 60, 8-12-20, +28, 3.5
Semenko 52, 6-11-17, +9, 0.0
Fuhr 45, 0-14-14,  E, 8.8*
Roulston 24, 5-7-12, E, 0.3
Conacher 45, 2-8-10, -2, -0.8
Chartraw 24, 2-6-8, +5, 1.0
Berry 13, 2-3-5, +6, 0.2
Summanen 2, 1-4-5, +4, 0.4
Gorence 12, 1-1-2, E, -0.2
Playfair 2, 1-1-2, +4, 0.4
Habscheid 9, 1-0-1, -3, -0.3
Sherven 2, 1-0-1, +1, 0.1
Blum 4, 0-1-1, E, 0.1
Moog 38, 0-1-1, E, 7.0*
Strueby 1, 0-1-1, +2, 0.1
Clark 1, 0-0-0, E, 0.0
Cote 13, 0-0-0, -5, -0.6
Graves 2, 0-0-, E, -0.1
Jalonen 3, 0-0-0, -2, -0.2
Kerr 3, 0-0-0, -3, 0.2

Talk about an offensive juggernaut!
A 200 point first line center with a 100 point second line center. 2 50-goal scoring right wingers. And a d-man with 40 goals and 120+ points. No wonder they loved 4 on 4.

* goalie points share

Posted by Lindas1st from New England on 08/25/11 at 09:34 PM ET

Add a Comment

Please limit embedded image or media size to 575 pixels wide.

Add your own avatar by joining Kukla's Korner, or logging in and uploading one in your member control panel.

Captchas bug you? Join KK or log in and you won't have to bother.


Notify me of follow-up comments?


Most Recent Blog Posts

About The Puck Stops Here

imageThe Puck Stops Here was founded during the 2004/05 lockout as a place to rant about hockey. The original site contains over 1000 posts, some of which were also published on FoxSports.com.

Who am I? A diehard hockey fan.

Why am I blogging? I want to.

Why are you reading it? ???

Email: y2kfhl@hotmail.com