Where Players Spend their Time in the Offensive Zone

by Hawerchuk on Jul 30, 2010 9:00 AM EDT Comment 11 comments

I've been looking at the origins of high (and low) shooting percentage, and one of the interesting outcomes was that shooting percentage is deterministically driven by where a player positions himself on the ice. So if we look at even-strength shots taken by Andrew Brunette over the last five seasons, we see a player who spends his time very close to the goal (0,0) - it's no surprise that he led the league in shooting percentage last season.

Ilya Kovalchuk, on the other hand, is all over the map. He's usually between the dots and below the top of the faceoff circles, and he doesn't really get in close to the goal.

Chris Pronger has the typical pattern of a left defenseman, with a small amount of time on the right side.

It's key to keep in mind that defensemen really don't pinch in very far to shoot. The defenseman who gets closest to the goal on average - Dan Boyle - shoots from further out than the least offensive forward.

0 recs | Comment 11 comments |

Story-email Email Printer Print

Comments

This is really interesting (particularly Kovalchuk’s), although I don’t think “where players spend their time in the offensive zone” and “where players shoot from” is exactly synonymous.

A Capital Wasteland - art & hockey from Washington, D.C.

by Jake Shapiro on Jul 30, 2010 9:05 AM EDT reply actions

Gabe

what’s really interesting to me is where you find the data to compile this graph, and how easy it would be to compile for other players.

I’m sure doing a whole team’s worth is a tedious exercise, but I’d love to do it, on my own, for the Avs,

Intelligent opinion does not follow the transitive property
Jibblescribbits: C'mon over and waste some time

by Jibblescribbits on Jul 30, 2010 10:57 AM EDT reply actions

Give this man the tools!

by Scott Reynolds on Jul 30, 2010 2:21 PM EDT up reply actions

Agreed

I’d love to see Vancouver’s.

'Nucks Misconduct - Housing Swedish Millionaires Since 2000.

by Yankee Canuck on Jul 31, 2010 7:48 AM EDT up reply actions

Would it be possible to use this data for goalies? I’m curious if this could help illustrate SV% on chance shots and non-chance shots. But it would probably be a ton of work…

Hockey Blogger at Pensburgh.com

by GoPens! on Jul 30, 2010 12:38 PM EDT reply actions

Any way to see what it looks like with different linemates?

There are a handful of centers who always seem to increase the goals scored by their linemates (Joe Thornton comes to mind). Is there a way to do a comparison of what a player’s shots look like when he’s on the ice with such a player versus other times?

by Bourque77 on Jul 30, 2010 1:34 PM EDT reply actions

Great job Gabe! This is really interesting stuff and it seems like we’re scratching the surface in terms of quantifying player “styles” from some of this information.

by Scott Reynolds on Jul 30, 2010 2:22 PM EDT reply actions

This is the type of data that it would seem that teams would have

I know in baseball, there are advanced scouting reports that reveal a hitters “hot” and “cold” zones over the course of his career, and then teams can attack a player based on that data. The type of pitches they struggle with and in what counts would also be useful info. And, in basketball, I know that teams who utilize stats often look at what areas of the court an opponent is most effective at and on what types of plays they are most successful at. Likewise, they take note of where there “cold” zones are and where they are the most inefficient and they try to push them into that area of the court.

It would be interesting to see if a similar pattern could be found for hockey players. What area of the ice do they tend to shoot the best? What area of the offensive zone do they struggle with and/or where are they most likely to give the puck up? Of course, there could be small sample size issues present with things like that, and teams would need to be cognizant of that.

I’m not sure how in depth teams get with things like that, but it would appear that data like this could be a form of quantitative scouting.

by Scottwood on Jul 30, 2010 7:18 PM EDT reply actions

Similar to the goaltender's idea

how difficult would it be to run a defensive equivalent looking at shot locations AGAINST a given defender.

Could you feasibly show a consistent defensive ability to force shooters to the outside and thus reduce expected shooting percentages? I’d assume it’s possible if you had data for the entire NHL in terms of which players shared the ice, and all the shooting info for the NHL in a given season.

I’m wondering how much distinction there is between two given defenders in this regard… if there’s little in the way of difference between an average vs. an elite defender in this regard, then I’m curious what may distinguish an elite from an average defender.

"Success is the ability to go from one failure to another with no loss of enthusiasm."

- Sir Winston Churchill

I'm pretty sure he's talking about the Leafs.

by Steve Burtch on Jul 31, 2010 11:17 AM EDT reply actions