The Arbitrary/Subjective Nature of Assists

KingArthursCourt

pronouns: he/him/his
Nov 11, 2019
252
798
British Columbia
The goal is, obviously, the most clear cut and objective counting statistic in hockey (setting aside fringe cases like figuring out if the puck grazed a player or not and so whose goal it should be). The assist, on the other hand, while tracked as a counting statistic in the same way and even given equal value in the context of points (up to the tiebreaker) is an artificial construct that features at least a few points of subjectivity:

1. The awarding of assists at all is at the discretion of scorekeepers who must make qualitative determinations on whether 0, 1, or 2 assists should be awarded for each goal.

2. The secondary assist is a totally subjective cut-off point; any argument for why there should be more than one assist awarded on a goal ("sometimes the secondary assist is the most important part of the goal!") can easily be extended to any number of assists - if 5 players team up on a goal, who's to say the 4th assist isn't the "most important" part of the scoring play?

So my question is this: has anyone attempted to analytically examine what would happen under different possible formulations of the assist, such as not awarding secondary assists (the data for this obviously exists - primary assists - but more what it would do to scoring tables and career stats), or awarding up to 5 assists on a goal, if the scorekeeper determined it justified (this would of course require going through many games and making judgment calls on each goal)? I think it would be interesting to see the effect of 1st assists, 2nd assists, 3rd assists, etc.
 

Michael Farkas

Grace Personified
Jun 28, 2006
13,350
7,832
NYC
www.HockeyProspect.com
Interesting thought project. Because of the ease of access to goal video and how quickly (relative to entire game tape) one could pour through them fairly regularly. Of course, you're adding a layer of subjectivity to it (which I find to be a positive), but it might prove quite interesting.

There's a lot of criteria that would need to be drawn out. From the "look back", to the degree in which the puck strikes/is controlled by the other team for an amount of time, are these "adjusted assists" still going to be counted in the order in which the puck was touched or in the order of importance, would there be any valuable to more obvious non-touching plays that helped facilitate the goal (i.e. screens, picks, etc.)...there's a lot to discuss...
 

Hockey Outsider

Registered User
Jan 16, 2005
9,056
13,986
So my question is this: has anyone attempted to analytically examine what would happen under different possible formulations of the assist, such as not awarding secondary assists (the data for this obviously exists - primary assists - but more what it would do to scoring tables and career stats), or awarding up to 5 assists on a goal, if the scorekeeper determined it justified (this would of course require going through many games and making judgment calls on each goal)? I think it would be interesting to see the effect of 1st assists, 2nd assists, 3rd assists, etc.

Has anyone counted up 3rd (or 4th and 5th) assists to see how that would impact the scoring race? Not to my knowledge (this would be a massive project). That being said, way back in the early days of the NHL - I don't recall the exact years - there was an unlimited number of assists that could be awarded, but in practice this was very rarely done. (Even secondary assists were much less common than they are today).

I've looked at what the scoring race would look like if secondary assists were removed - sort of. See Examining the value of the secondary assist. I showed what ES scoring among forawrds would look like weighing secondary assists at about two-thirds of a goal (I use two statistical methods to prove why that's a reasonable number). There are some changes, of course, but all things considered, they're quite marginal. The top 25 players are the same; 23 of them change no more than two spots in the rankings.
 

Chili

En boca cerrada no entran moscas
Jun 10, 2004
8,440
4,269
Watching games how many times do we see an effective screen as the key part of a goal? (i.e. taking away the goaltender's view).

It's usually unrewarded statistically but it's an important part of the game.
 
  • Like
Reactions: Hockey Outsider

Choralone

Registered User
Oct 16, 2010
4,998
3,932
Burbank, CA
I wonder instead if it would be possible to characterize the type of assist, perhaps the zone where the pass was made and the zone where it was received, or if the shot was off the pass (one-timer) or not.
 

SotasicA

Registered User
Aug 25, 2014
8,489
6,402
The +/- is the most clear cut statistic in hockey. Don't even have to watch replays to see who scored the goal. I wouldn't be against it replacing points. Though there used to be an award for best plus/minus anyway, so it should probably just stay separate.
 

Uncle Rotter

Registered User
May 11, 2010
5,974
1,037
Kelowna, B.C.
The +/- is the most clear cut statistic in hockey. Don't even have to watch replays to see who scored the goal. I wouldn't be against it replacing points. Though there used to be an award for best plus/minus anyway, so it should probably just stay separate.
Ahem.
edmonton-oilers-charlie-huddy-19881.jpg
 

Ad

Upcoming events

Ad

Ad

-->