• Hey, guest user. Hope you're enjoying NeoGAF! Have you considered registering for an account? Come join us and add your take to the daily discourse.

NHL Off-Season 2014 |OT2| - Dan Cleary

S

Samyy

Member
calder said:
Important note, because my first post was short and vague: I don't think anyone argues that shot quality doesn't exist - obviously a tap-in to an empty net is different from a dump in from centre ice. But attempts to *measure* shot quality end up being less useful than you'd think, because of all the reasons mentioned already (how to define all the elements, how to gauge one shot from a nearly identical one etc) and a growing suspicion that shot quality for and against is pretty random. A few years ago the advanced stats crowd was going nuts for shot quality but after a lot of attempts I don't think anyone was able to come up with a system that showed a meaningful difference from team to team never mind player to player. And if there's no real difference (ie a difference beyond noise/luck) in a stat between good teams and bad then it's not worth tracking.

The other aspect of shot quality that has gone out of favour is the notion that good defensive teams force the other team to take 'bad' shots from the outside. While you can see a connection between good defensive teams and puck possession there isn't one with shot quality, which is to say that good defensive teams give up high quality shots in a seemingly random fashion, or at least the numbers between good and bad quality shots against are so close it may as well be random That's another knock against shot quality as a useful metric.
Click to expand...

The only measureable way I can think of for approximating Shot Quality is by using position on the ice from where the shots are taken from and the % of the time a shot from that position goes in.
ie. % of shots from the slot that result in a goal vs. from the half-boards

But yea I don't know if that would really tell you anything
 
Q

Quick

Banned
The real stat I care about is the number of days left before the season starts in relation to the number of Kings fans here dropping off.

Some other QuickStats:

Brown knees per 5 games played (BK5)

Smelly racism per black player (SR/BP)

Water bottles per Clarkson (WBC)

Awkward Kessel Percentage (UmYouKnow%)

Communism Percentage (HammerSickle%)
 
D

DopeyFish

Not bitter, just unsweetened
Samyy said:
The only measureable way I can think of for approximating Shot Quality is by using position on the ice from where the shots are taken from and the % of the time a shot from that position goes in.
ie. % of shots from the slot that result in a goal vs. from the half-boards

But yea I don't know if that would really tell you anything
Click to expand...

It was because of that stat and looking at how NHL was rating shot quality as the primary reason for me forming this stat.

Shot distance is a poor metric, because of how important angles are. Close up shots (less than 5 feet) when goalie in position will have likely be a very poor quality. (as shot angle potential make it easier for goalie to save)

Slot and out of slot is poor because of the amount of goals scored at the hash marks and point corners.

Point shots have the benefit of obstructed shots. Closer shots have the benefit of less reaction time.
 
S

Samyy

Member
DopeyFish said:
It was because of that stat and looking at how NHL was rating shot quality as the primary reason for me forming this stat.

Shot distance is a poor metric, because of how important angles are. Close up shots (less than 5 feet) when goalie in position will have likely be a very poor quality. (as shot angle potential make it easier for goalie to save)

Slot and out of slot is poor because of the amount of goals scored at the hash marks and point corners.

Point shots have the benefit of obstructed shots. Closer shots have the benefit of less reaction time.
Click to expand...

Dopey you can't account for every single possible variable that could affect the chances of a shot going in or not.

Why not just split the O-Zone into a grid, and use that to look at % of total shots resulting in goals. Deflections and reaction time are variables you can't really account for practically in any way.

I mean I guess what you could do is determine the total # of deflected SOGs in a game - and then the # of total goals from deflect shots, then look at the proportion of deflected shots resulting from different positions on the ice and perhaps if you collect enough data you can adjust shot position [point vs. slot] that way, but again is it going to give you something more meaningful then the basic distance/position examination that would make it worth all that extra work? I doubt it.
 
C

calder

Member
Are the Jets tanking for Connor McDavid?

I wish I could believe the Jets had an actual plan like that.
jquIRYFZjWKXJ.png

*dare to dream*
 
D

DopeyFish

Not bitter, just unsweetened
Samyy said:
Dopey you can't account for every single possible variable that could affect the chances of a shot going in or not.

Why not just split the O-Zone into a grid, and use that to look at % of total shots resulting in goals. Deflections and reaction time are variables you can't really account for practically in any way.
Click to expand...

What do you think i'm doing? Lol im separating the ozone into zones and then using variables for the purposes of weighting.

I can look at everything from the highest level (shots coming from certain places on ice) or look with more details to see variances with those variables applied.

Im not tracking super detailed info simply because the data required would be too much... to the point that i'd be sourcing data from different eras and that data isn't usable at all.

The website i'm designing for this project will allow you to choose the variables you want to use. I said before... it's going to be a candy store. The amount of stuff i intend to show and the flexibility of that data will be without equal... on any sports website.
 
C

calder

Member
DopeyFish said:
Lol im separating the ozone into zones and then using variables for the purposes of weighting.
Click to expand...

That's an insane amount of work, but IF you could do it I bet it would make shot quality potentially useful. I just don't think it's possible for even a dedicated group of volunteers to track it, this would be much more difficult to do than Bill James' Project Statsheet was back in the 80's because tracking baseball plays is inherently easier.

Really, what you should focus on is trying to figure out a way to use technology to do that shit for you. If you could so something similar to what's done in the NBA, using video to track players, then you wouldn't need to do a thing with the data itself. Simply having a better, more consistent way to track ice time and shot attempt locations would be a massive evolution to hockey analytics, because even today most hockey stats are tracked by hand. Are you a video data expert by any chance? ;)
 
S

Socreges

Banned
calder said:
That's an insane amount of work, but IF you could do it I bet it would make shot quality potentially useful. I just don't think it's possible for even a dedicated group of volunteers to track it, this would be much more difficult to do than Bill James' Project Statsheet was back in the 80's because tracking baseball plays is inherently easier.

Really, what you should focus on is trying to figure out a way to use technology to do that shit for you. If you could so something similar to what's done in the NBA, using video to track players, then you wouldn't need to do a thing with the data itself. Simply having a better, more consistent way to track ice time and shot attempt locations would be a massive evolution to hockey analytics, because even today most hockey stats are tracked by hand. Are you a video data expert by any chance? ;)
Click to expand...
But do you think subjectivity could be eliminated? Ignoring the astronomical amount of work required, that's the biggest problem for me.
 
S

Samyy

Member
Socreges said:
But do you think subjectivity could be eliminated? Ignoring the astronomical amount of work required, that's the biggest problem for me.
Click to expand...

No I think dividing the offensive area into equal sized "zones", collecting shot data from each of those zones and then % of goals from those shots from each zone is pretty objective, its more about whether it tells you anything useful.

The other things (like amount of people in front of the net) and whatever else don't think thats possible.
 
S

Socreges

Banned
Samyy said:
No I think dividing the offensive area into equal sized "zones", collecting shot data from each of those zones and then % of goals from those shots from each zone is pretty objective, its more about whether it tells you anything useful.

The other things (like amount of people in front of the net) and whatever else don't think thats possible.
Click to expand...
Right. Those are objective fields. If that's what Dopey has resigned to, then I'll leave the rest to you guys.
 
C

calder

Member
Socreges said:
But do you think subjectivity could be eliminated? Ignoring the astronomical amount of work required, that's the biggest problem for me.
Click to expand...

I was being a bit polite, it's a huge - probably impossible - "if". I think the work involved is almost unfathomable for even a dedicated group of volunteers to do for anything more than a single team, *and* I think the subjective nature of determining how to measure "quality" means that any numbers you did theoretically get from that group of volunteers would have a lot of noise. Which is what stats guys have found the last few years, leading back to my comments about how shot quality numbers look as close to random as to be pointless.

Unless they find a way to automate it, either with cameras or sensors in the equipment and pucks to constantly get telemetry from everything moving, I don't think it works well enough to get usable data for something like shot quality. And I'm sure the NHL and most of the teams are working on ways to track stuff like that without guys in the video room making notes by hand for hours during and after every game, which is basically what they do now.
 
S

Samyy

Member
calder said:
I was being a bit polite, it's a huge - probably impossible - if.

Unless they find a way to automate it, either with cameras or sensors in the equipment and pucks to constantly get telemetry from everything moving, I don't think it works well enough to get usable data for something like shot quality. And I'm sure the NHL and most of the teams are working on ways to track stuff like that without guys in the video room making notes by hand for hours during and after every game, which is basically what they do now.
Click to expand...

Those camera's already exist, and are used in the NBA, track everything every player does. Apparently also used in soccer.
 
S

Sanjuro

Member
Quick said:
Jets are gonna relocate to Seattle.

Tabris guarantee.
Click to expand...

Seattle probably deserves a team the most in a striking while the iron is hot manner. Should be able to latch onto the great atmosphere with MLS/NFL and create a possibly rivalry with Vancouver.
 
Q

Quick

Banned
SmackCrackleNPop said:
NHL 2014-2015 |OT1| Dan Clearsi. Now and Forever
Click to expand...

zroid said:
NHL 2014-2015 |OT2| Dan Clearsi. Now and Forever 2
Click to expand...

Are we doing a general OT vs. monthly OT now?

Gonna need a catch-up on how to read and evaluate Clearsi ratings. Too much intangibles.

Sanjuro said:
Seattle probably deserves a team the most in a striking while the iron is hot manner. Should be able to latch onto the great atmosphere with MLS/NFL and create a possibly rivalry with Vancouver.
Click to expand...

I posted that partly as a joke, but I do want Seattle to get a team. And I want the Sonics back, too.
 
M

MacAttack

Member
CrazedArabMan said:
So, new CoD MP looks actually good, may pick it up down the line on sale as my MP for this year will be Halo for competitive and Destiny for co-op.
Click to expand...

I think this map pack may be the best one of the set of four released for Ghosts. The map, Subzero, is set in Canada complete with hockey locker rooms, Yeti's and freezers stocked with beer.

Gold Rush is the only "big" map in this pack and it reminds me of Big Thunder Mountain at WDW with rideable mine carts that complete a circuit around the map. You can tell they spent a lot of time on this map, its gorgeous.

Arena is a typical symmetrical death match arena. A level specific care package will unleash all manner of traps. Fun map but very chaotic and hectic.

There is also a smallish Chinese village map that looks pretty and is pretty solid but is nothing special in terms of layout or surprises.

All the maps except Gold Rush are small to not quite medium size that keep the action moving along and make it difficult to camp any particular area. With Gold Rush, the multiple levels, caves and mountainous terrain keeps people moving around. For better or worse, run and gun maps are what works in this iteration of CoD and these maps play to that strength.
 
S

Socreges

Banned
I always loved him as an actor and comedian. Didn't care that his comedy was ADHD chaos - that's what made him brilliant. He seemed like an amazing person too.

MacAttack said:
Bad timing for my Call of Duty map pack review, which seems trivial in comparison to this news.

RIP Robin Williams.
Click to expand...
I thought the same, so this made me laugh
 
You must log in or register to reply here.

Similar threads

NHL 2023-2024 Season |OT| My Other Ride’s a Zamboni
2 3 4
184
6K
Trogdor1123 replied
Suicide Squad: Kill the Justice League Season One Kicks Off on March 28
News
45
2K
Kagoshima_Luke replied
Slayer Revealed in Trailer for Guilty Gear Strive; May Release Date, Season Pass 4 Confirmed
News Trailer
5
732
Barakov replied
Jujutsu Kaisen confirmed for season 3
News
13
4K
NightmareFarm replied
RTTP: Twin Peaks season 3
36
2K
Krathoon replied
Top Bottom