NHL now officially tracking advanced stats

[MathMan]

We are 10th for SAT for - SAT against when the game is close (1 goal difference).

I hate to call into question stats from nhl.com, but multiple sites have the Habs around 22nd in the league in this exact metric and they have the same definition of "close". I would very much like to know how they calculated that, what is and is not included, and got to that number because the difference from basically everyone else is quite significant.

 

[LaP]

I hate to call into question stats from nhl.com, but multiple sites have the Habs around 22nd in the league in this exact metric and they have the same definition of "close". I would very much like to know how they calculated that, what is and is not included, and got to that number because the difference from basically everyone else is quite significant.

I believe more NHL.com official stats than some 3rd party web sites. I think NHL.com stats are actually in line with what i see when i watch the game.

We start slowly. We **** when the score is 0-0 and that what NHL.com stats are saying. Then in the 2nd we start playing better (score is close). Then in the 3rd we are one of the best team (we are behind).

 

[MathMan]

I believe more NHL.com official stats than some 3rd party web sites.

I wouldn't be so quick. The 3rd party sites have contacts with the people who originated the numbers, have been doing this for far longer, have validated their numbers/method against each other and and are generally far more robust as a result.

I know from experience that it's very easy to get this wrong, which is why I would like to understand how they arrived at their numbers. I can't really trust this right now, that SAT close thing is way out of line with everything that we already know.

We start slowly. We **** when the score is 0-0 and that what NHL.com stats are saying. Then in the 2nd we start playing better (score is close). Then in the 3rd we are one of the best team (we are behind).

The Habs' tied numbers are bad, their numbers when trailing by 1 are great, their numbers when ahead by 1 are awful, but somehow the sum of all that ends up different on nhl.com than everywhere else and it's not a small discrepancy.

Certainly to suggest the Habs are the 10th-best possession club in the league fails the eye test.

 

[Yep]

I would spend my time debating with other HF posters about a myriad of stats just so we concluded that Carey is the team, DD sucks, and MT should be fired?

CAREYSTEAM
DDSUCKS
FIREMT​

T-shirt potential right there.

Well done.

 

[ThaDevilGirl]

CAREYSTEAM
DDSUCKS
FIREMT​

T-shirt potential right there.

Well done.

Lol, can I get a percentage of the sales?

 

[LyricalLyricist]

Depends on the team. If I play one shift for Buffalo and get 10 shots against (because I'm bad at hockey), my SAT Rel would be around 1300, first in the league.

SAT Rel can lead to strange results. Not sure it's that useful to look at it league wide. Depends on a lot on the team.

But yes, you are right, on a good team, it will help if you play more.

Is it pro-rated by ice time though? That's what I'm confused about.

No -- this is basically the team's SAT while the guy is on the ice, minus the team's SAT while the guy is on the bench.

In a very, VERY rough way it shows the impact of a player on his team's SAT, but keep in mind that this is heavily impacted by usage, teammates, matchups, and etc. Use with a lot of caution. But it's a pretty good clue that Subban is really good.

EDIT: Or are they not prorating by icetime? If not then yeah, that's pretty useless.

I'm not sure if they are doing SAT Rel as a function of ice time or just counting the games you play only.

Pateryn is 19 SAT for but -4 SAT REL.

They don't go into specifics about how they reach that -4. Only that it's when he's not on the ice. Is it adjusted to reflect his ice time? Doubtful IMO.

Basically, I figure he played 2 GP and the team had a 23 SAT for when he's not on the ice during the game, resulting in 19-23.

I find it kind of pointless though. If not adjusted per ice time the numbers get inflated dramatically for good players who play more.

 

[MathMan]

Is it pro-rated by ice time though? That's what I'm confused about.

They have a separate page for per-icetime stats and relative stats are not on it (weird as this is where they would actually be useful).

Right now the REL thing is completely useless. Most of the league leaders are Sabres because the team is so baaaaaaaaaaaaaad in the 40-50 minutes the player isn't on that it doesn't matter that it is baaaaaaaaaaaaaaad while they are on.

 

[Canadienna]

One thing I notice that I think is very important is our SAT ahead being SO BAD. We have nobody, and I mean NOBODY who is in the positive when were ahead. I think that shows what we know, that this team likes to sit on a lead and cut down on the forechecking, but it's nice to see some proof of it.

 

[optimus2861]

I know from experience that it's very easy to get this wrong, which is why I would like to understand how they arrived at their numbers. I can't really trust this right now, that SAT close thing is way out of line with everything that we already know.

The raw SAT counts amongst the three sites I'm looking at are fairly close, so it doesn't seem to be a raw data problem (all numbers are 5-on-5):

stats.hockeyanalysis.com: 1835-1856 Fenwick , 2569-2621 Corsi
naturalstattrick.com: 1876-1902 Fenwick, 2632-2694 Corsi
NHL: 1842-1893 Fenwick, 2580-2684 Corsi

I'd guess that NHL isn't computing "Close" correctly and that it's counting "within 1, third period" as "Close". That would likely vault Montreal up several spots on the chart.

 

[DAChampion]

When I was skimming this thread earlier I thought that the SAT scores of American born players might have been released somehow. Could look something like this:

Ryan Malone
Verbal: 400/800
Math: 340/800

 

[Hoople]

I don't analyze the games, I watch them. Much more enjoyable.

Best post in the thread.

 

[M.C.G. 31]

I don't follow advanced stats *too* much, but the new stats layout on NHL.com (the points, goals, assists, etc.) looks really nice and sleek.

 

[Brainiac]

I don't follow advanced stats *too* much, but the new stats layout on NHL.com (the points, goals, assists, etc.) looks really nice and sleek.

Yeah, it's pretty sleek.

Also, just the name change will help a lot, IMO. You can't argue against SAT. It's shot attempts, that's what was measured and that's what we analyze and talk about. Simple, clear, straight to the point.

 

[DAChampion]

Yeah, it's pretty sleek.

Also, just the name change will help a lot, IMO. You can't argue against SAT. It's shot attempts, that's what was measured and that's what we analyze and talk about. Simple, clear, straight to the point.

I agree with that actually.

The prior name for the stats, "Corsi", "Fenwick", et cetera was deliberately obscurantist, it increased confusion rather than clarity, as part of an agenda to feign sophistication.

By giving the stats names that make sense, the NHL is undermining the bloggers.

 

[MathMan]

I'd guess that NHL isn't computing "Close" correctly and that it's counting "within 1, third period" as "Close". That would likely vault Montreal up several spots on the chart.

I'll run that query when I get home. My raw numbers don't line up with the NHL's (they line up with war-on-ice) but if that's really the factor I should be able to see it.

Note that their tooltip indicates that they only use tied third period though.

 

[Brainiac]

By giving the stats names that make sense, the NHL is undermining the bloggers.

Touché!

The NHL's guys are professionals, after all.

 

[LaP]

Yeah, it's pretty sleek.

Also, just the name change will help a lot, IMO. You can't argue against SAT. It's shot attempts, that's what was measured and that's what we analyze and talk about. Simple, clear, straight to the point.

Totally agree. I have no problem with advanced stats. My problem is when people call them puck possession stats. This is not what they are. Saying those are puck possession stats you assume the team attempting the most shots have the puck most often. It might be true 90-95% of the time but there will always be time where it's not true. This is not how you are doing stats. There's only one puck possession stat. It's a stat where you time how long a player/team have the puck in a game.

Some players prefer to keep the puck, cycle and attempt only high % shots/plays. Couple of games ago Sekac and Eller spent like 1 minutes in the offensive zone without attempting a single shot. Good possession number, no so good advanced stats number.

 

[MathMan]

The prior name for the stats, "Corsi", "Fenwick", et cetera was deliberately obscurantist, it increased confusion rather than clarity, as part of an agenda to feign sophistication.

I don't think there was an agenda. I think that there just wasn't a whole lot of thought put into naming the metrics -- Vic Ferrari got things started with Corsi and it became traditional to name them after their inventors.

By giving the stats names that make sense, the NHL is undermining the bloggers.

I doubt it; the blogs were already debating whether it was better to rename the metrics before the NHL got in it. I suspect we're going to see "SAT/Corsi" and "USAT/Fenwick" in articles for a while.

 

[Brainiac]

I doubt it; the blogs were already debating whether it was better to rename the metrics before the NHL got in it. I suspect we're going to see "SAT/Corsi" and "USAT/Fenwick" in articles for a while.

Forget it. All the mainstream journalists will pick up on SAT right away. Clarity. It's important when you're out there on TV talking about something. You don't want to say 'Corsi' and then have to spend 10 minutes to explain what it is.

Everybody knows what a goal is and what an assist is. Even people who don't follow hockey at all can grasp the idea from a generic knowledge of sports. Everybody knows what a shot is. So when you say "shot attempts", it's clear as day what you're talking about.

It might look like a detail to you, because you're inside that community. But the way you name things is extremely important when your job is to communicate information.

 

[Brainiac]

The Habs' tied numbers are bad, their numbers when trailing by 1 are great, their numbers when ahead by 1 are awful, but somehow the sum of all that ends up different on nhl.com than everywhere else and it's not a small discrepancy.

Certainly to suggest the Habs are the 10th-best possession club in the league fails the eye test.

Here is the definition, from NHL's website (http://www.nhl.com/stats/enhanced)

SAT% close: shot attempts % (SAT%) when team is within one goal of their opponent in periods 1 and 2, or tied in period 3.

Is it different with the way some of you guys track it? Might explain some of the discrepency.

 

[Captain Mountain]

Forget it. All the mainstream journalists will pick up on SAT right away. Clarity. It's important when you're out there on TV talking about something. You don't want to say 'Corsi' and then have to spend 10 minutes to explain what it is.

Everybody knows what a goal is and what an assist is. Even people who don't follow hockey at all can grasp the idea from a generic knowledge of sports. Everybody knows what a shot is. So when you say "shot attempts", it's clear as day what you're talking about.

It might look like a detail to you, because you're inside that community. But the way you name things is extremely important when your job is to communicate information.

Most mainstream journalists don't pick up anything right away.

 

[Andy]

Here is the definition, from NHL's website (http://www.nhl.com/stats/enhanced)

SAT% close: shot attempts % (SAT%) when team is within one goal of their opponent in periods 1 and 2, or tied in period 3.

Is it different with the way some of you guys track it? Might explain some of the discrepency.

Makes sense, the Habs play well when behind, we already knew that.

I also wonder how many people who scoffed at "corsi cup" are slowly going to start accepting these stats now that they have become an official part of the nhl website.

 

[Ohashi_Jouzu*]

I also wonder how many people who scoffed at "corsi cup" are slowly going to start accepting these stats now that they have become an official part of the nhl website.

They mean nothing more today than they did yesterday. They're simply more accessible and sortable, and still can't tell us which team actually had the puck for most of the 60 minute game.

 

[Brainiac]

Most mainstream journalists don't pick up anything right away.

François Gagnon already did. As a recognized member of the media, he probably had access to a beta version of this new part of NHL.com.

If François Gagnon understands something, I'm pretty sure a bunch of people do.

 

[Andy]

All the site does is confirm what I see with my eyes. The Habs have terrible starts, pick up the pace when their backs are against the wall. In other words, they can play a possession game when they have to, but seem content in getting outplayed for most of the game first. This only reinforces my belief that this is a coaching issue.

Their SAT while ahead is brutal also. They are 25th in league, suggesting the have a hard time at preventing their opponents from picking up the pace when they have the lead. Only Columbus, Buffalo, the Leafs and Calgary are worse. yeesh.