What advanced stats would you track?

Randomtask68

Registered User
Jul 19, 2010
612
1
Burlington, MA
Other than the "traditional" advanced stats like Corsi, Fenwick, PDO, Zone starts etc, what are some others events in games that can be quantified in such a manner? If you were in an NHL front office, what would you want to track?

Examples off the top of my head:

- Dumping the puck in the zone versus skating it into the zone. There has been some work done on this suggesting skating the puck into the zone is something like twice as effective as dumping it in. But I think this can be expanded like what extraskater does in breaking down situation like 5 on 5, 5 on 5 close, and so on. I would be intrigued as to how teams fair on the powerplay when they are forced to dump it in versus gaining entry through skating.

- In regards to defensemen, if there was someway to judge how good/bad they are in getting the puck out of the zone when they have a legitimate chance of clearing the zone. No idea how you would measure this, but maybe just measure how much time a defenseman spends time in his own end, which would be influenced by zone starts of course.

- How effective teams are at executing or defending each type of odd man break. Breakaway, 2 on 1, 3 on 2. You could see how goalies do on each type of shot, how defenseman do in defending each situation, and how the offense does in either getting shots or goals from these chances. Not sure if these are frequent enough to make anything out of them, but it could show how opportunistic teams are and how many of their even strength goals they get from odd man rushes.

Just some thoughts, curious to think what other posters can think of to be quantified in an advanced stats kind of manner.
 

PepsiCenterMagic

Food is Great
Jul 17, 2013
651
44
Gvt.

[Mod edit] : in other words , the Goals Versus Threshold stats
 
Last edited by a moderator:

nobuddy

Registered User
Oct 13, 2010
17,994
97
Nowhere
- Dumping the puck in the zone versus skating it into the zone. There has been some work done on this suggesting skating the puck into the zone is something like twice as effective as dumping it in. But I think this can be expanded like what extraskater does in breaking down situation like 5 on 5, 5 on 5 close, and so on. I would be intrigued as to how teams fair on the powerplay when they are forced to dump it in versus gaining entry through skating. .

@ShutdownLine on twitter is working on this for this past season and will make the data available when completed for a price.



As for your question, distance skated, average speed skated, average speed of shot/pass, etc.
 

member 105785

Guest
None to add, but people should stop tracking takeaways, they're completely meaningless
 

Filatov2Kovalev2Bonk

Effortless sexy.
Jul 13, 2006
12,733
1,061
Cumberland
Total Possession Time (TPT):

The amount of time the puck spends on a player's stick. Any puck that: bounces away, is shot, is taken away or is not directly in contact with a player's stick and moving forward ends total possession time.

This would show actual puck possession, which is likely not more than a few seconds per player shift. :nod:
 

BrokenFace

Registered User
Aug 15, 2010
1,571
1,735
STL
Heat charts to show where players spend most of their time, like they have in soccer. I think this is on the way.

More than anything though, I would love it if someone in the advanced stats community really figured out a way to quantify shot quality. I know people have tried before and over the course of a season the "quality" of shots taken by a player/team evens out and it's the quantity that can be used to make predictions, but I'm still a bit skeptical of the idea the some players don't consistently generate better shots.
 

PuqTalk

I love Cogliano
Jun 24, 2012
1,866
0
Texas
Heat charts to show where players spend most of their time, like they have in soccer. I think this is on the way.

That's an excellent idea. Also a kind of deviation... heat charts on which holes players score most of their goals on. And which amount are by slap/snap/wrist/backhand shots. Not sure if that complex of an analysis would matter significantly, but it'd be cool to see.
 

Randomtask68

Registered User
Jul 19, 2010
612
1
Burlington, MA
amount of time spent in each zone.

Now would you differentiate between time spent in the offensive zone while Team A is possessing the puck on the attack and when the Team B is possessing the puck in their own end? Would they both count as time spent in Team B's zone or is there a difference depending on who has the puck, or is that too hard to time accurately?
 

DaveG

Noted Jerk
Apr 7, 2003
51,266
48,837
Winston-Salem NC
zone start/finish percentage, and could even expand that to zone start/finish percentage of the team with the player both on and off the ice
 

Thrice

To Be Everywhere...
Sep 27, 2007
2,129
2
Some sort of defensive zone breakout rating for defensemen. This would include a failed breakout rating/percentage.
 

PSGJ

Registered User
May 19, 2012
833
0
Sweden
I'd like to see someone cobble together a stat for power play creation. You'd need penalties taken (minus situations where players from both teams get penalties) and penalties drawn. The stat would show a plus if you create more power plays and a minus if you create more box plays.
 

SimplySensational

Heard of Hough
Mar 27, 2011
18,839
6
VA
I'd like to see someone cobble together a stat for power play creation. You'd need penalties taken (minus situations where players from both teams get penalties) and penalties drawn. The stat would show a plus if you create more power plays and a minus if you create more box plays.

Already exists...

Matt Duchene lead the league with 24 more penalties drawn than taken... Phaneuf took 29 more than he drew...
 

tigervixxxen

Optimism=Delusional
Jul 7, 2013
53,061
6,158
Denver
burgundy-review.com
Scoring chances, zone time and zone entires/exits. I know teams and individuals already track these things, it should be universal. Scoring chances vs shots and shot attempts would be interesting. Plus why use shot attempts to indicate possession when there could be an actual measure of time of possession.
 

Sidney the Kidney

One last time
Jun 29, 2009
55,782
46,875
Not necessarily an advanced stat, but I'd like to see it officially and readily available the number of scoring chances each team gives up and creates per game.

Scoring chances is just a bit more in depth for analyzing shots on goal, and can give a better idea of how quality those shots a team gives up/a goalie faces really are.

As it is, shot totals can be misleading since currently a soft wrister from the point with no screen counts just as much as a point blank shot from the slot area.

Note: I understand this stat is already kept track of at some level. I'm advocating it being more readily available for the casual fan to view on any site that lists the various other stats.
 

sharks9

Registered User
Jan 16, 2012
16,444
2,604
Canada
Expanding more on which teams give up more quality chances would help us evaluate goalies better.
 

Wats

Error 520
Mar 8, 2006
42,017
6,689
I'm not sure why simple stats like icetime spent in offensive zone/neutral zone/defensive zone aren't already readily available.

Shots should have a distance subsection where you can see who leads in shots 1-20 feet from net vs further away. This is also already tracked by NHL so shouldn't be too difficult to make into a database as opposed to finding people to track it.

Zone clears might be interesting - the number of times a player gets puck out of defensive zone.

Last player to have puck before offsides/icing is worth knowing as well since I find the same people tend to do both on some teams.
 
Last edited:

eklunds source

Registered User
Jul 23, 2008
8,323
0
Ed Snider's basement
amount of time spent in each zone.

That has about a 90-95% correlation to shot attempts. People don't bother tracking it because it doesn't provide more useful information than what we already have.

Leafs time on attack

Most post hits in a season would be interesting.
The amount of posts hit is similar to missed shots, in that it's very random year-to-year. Some games a player might hit the post 4 times, then not again for a month. Some years a player will hit the post a ridiculous amount of times, and some years he'll shoot post-and-in a ridiculous amount of times.



I'd like to see shooting percentage expressed as a percentage of shots taken that we're not blocked - it seems silly to "lower" a players shooting percentage if the goalie makes a save, but to not affect it at all if he misses the net.

Also, excluding shots that were blocked because players have historically shown a talent for getting shots through or not - ie the number of shots they take that get blocked are predictive of how often their shots will be blocked in the future.
 

Ad

Upcoming events

Ad

Ad