The Advanced Stats Thread Episode IX

[PlamsUnlimited]

I went to Corsica

What browser do you use

 

[Machinehead]

What browser do you use

Chrome

 

[Irishguy42]

I've been able to load Corsica on Firefox easily just now.

 

[PlamsUnlimited]

Wtf, it does not load for me on Chrome for a lot of live games.

 

[Machinehead]

Corsica has improved. It's still not great.

It's Ristolainen now instead of Pionk.

 

[DanielBrassard]

Is it any surprise the team's xGF% has started to tank? When you get never have the puck, you aren't going to be able to generate offense, and opposing teams will break you down eventually

 

[Ola]

Is it any surprise the team's xGF% has started to tank? When you get never have the puck, you aren't going to be able to generate offense, and opposing teams will break you down eventually

Yeah, the problem for us lately is that our leadership haven’t picked up when it has been needed. I have nothing against Zibanejad, but the difference between us and some teams not trending the wrong way is those teams core players putting the team on their backs when it’s needed and getting pts. Kreider is still streaky and Ziba isn’t winning games for us. Buch is hurt and Zucc is hurt. This should be Zuccs final season in NY. Hayes is our most consistent forward, by my eye test at least.

Unless someone takes charge there could be a wide spread energy leakage that will hit across the board on performance.

 

[Ola]

Hayes is an interesting case study. Very consistent shut down role for two years, especially that year with Grabner. In which they got so many D zone starts. Played well and produced well, but very bad metrics. This year? He gets away from the shut down role and his CF goes up a lot.

Did he improve a ton?

Is it the coach?

Didn’t adjustment methods in place catch the negatives of his role the two previous roles?

He is another example of a player where I think the metrics failed as a stand alone tool to evaluate a player. And as to the topic if Pionk can improve? Hayes CF rel is up 14% from two years ago.

 

[Machinehead]

Hayes is an interesting case study. Very consistent shut down role for two years, especially that year with Grabner. In which they got so many D zone starts. Played well and produced well, but very bad metrics. This year? He gets away from the shut down role and his CF goes up a lot.

Did he improve a ton?

Is it the coach?

Didn’t adjustment methods in place catch the negatives of his role the two previous roles?

He is another example of a player where I think the metrics failed as a stand alone tool to evaluate a player. And as to the topic if Pionk can improve? Hayes CF rel is up 14% from two years ago.

His zone starts are more offensive this year, but still not favorable. Always felt zone starts were a boogeyman anyway.

Quality of Competition is the same.

He's still our most defensively used center, just not to the same extreme.

I think he's improved a lot.

 

[Mac n Gs]

Hayes is an interesting case study. Very consistent shut down role for two years, especially that year with Grabner. In which they got so many D zone starts. Played well and produced well, but very bad metrics. This year? He gets away from the shut down role and his CF goes up a lot.

Did he improve a ton?

Is it the coach?

Didn’t adjustment methods in place catch the negatives of his role the two previous roles?

He is another example of a player where I think the metrics failed as a stand alone tool to evaluate a player. And as to the topic if Pionk can improve? Hayes CF rel is up 14% from two years ago.

There are more advanced player evaluation metrics and models that showed Hayes to be one of the most valuable players to this team last year and this year. Even by the eye test it was apparent that he was excelling in the shut down role and still driving play with poor linemates- it's why it was so laughable when some people on these boards (not you) were saying that Hayes was stat padding at the end of last year. No, ya dope, he got better minutes, actual linemates, and powerplay time. What a shocker that a skilled offensive player started scoring more when used that way.

The other thing I've been meaning to ask you is your thoughts on my posts about Quinn collapsing these guys down low too far in the d-zone. I saw you liked a post from a GDT last week or so, but never got to ask you further about the topic.

 

[aufheben]

Are there any threads here where guys can take personal grievances and just duke it out? Because I'd read that ****.

It’s seriously pissing in the wind trying to get this board to stop insulting each other.

 

[aufheben]

Corsica doesn't work for me very much.

Where do you go? I feel like we are deeper than the N train tunnels on the xGF% list now.

The Broadway line has the deepest tunnels in the MTA. #MTAfunfacts

I've been able to load Corsica on Firefox easily just now.

Corsica has improved. It's still not great.

It's Ristolainen now instead of Pionk.

Corsica still runs like crap on Firefox for me, although I haven’t tried it today.

 

[PlamsUnlimited]

It’s seriously pissing in the wind trying to get this board to stop insulting each other.

I had to do that yesterday. I was scared it was going to slant it back on me

 

[silverfish]

 

[Ola]

There are more advanced player evaluation metrics and models that showed Hayes to be one of the most valuable players to this team last year and this year. Even by the eye test it was apparent that he was excelling in the shut down role and still driving play with poor linemates- it's why it was so laughable when some people on these boards (not you) were saying that Hayes was stat padding at the end of last year. No, ya dope, he got better minutes, actual linemates, and powerplay time. What a shocker that a skilled offensive player started scoring more when used that way.

The other thing I've been meaning to ask you is your thoughts on my posts about Quinn collapsing these guys down low too far in the d-zone. I saw you liked a post from a GDT last week or so, but never got to ask you further about the topic.

I don't really have any insight in this area other than the very basic stuff. In the modern game you have overloaded more and more and moved away from the man-man approach we used to see. These pictures kind of illustrates, to the left, how everyone played 20 years ago, and to the right, how most play today (Y=winger, B=D):

Like the idea behind this development is not hard to get. The defending Y's job is to remove the passing lane from yellow C to B. By moving down this deep you do that and you also really take away the ice and passing lanes in more dangerous positions. Often the phrase "we left a player wide open in front of our net" is used, but its not true that often. Stuff happens and players reacts and have to react to it. If you play like the picture to the left and yellow C dekes the pants off his guy the other D have to leave his guy and try to handle both. Any of the defending Ys will not have time to drop down and take away the Y that now is wide open.

This is what you mean when you talk about layers in hockey. In the picture to the right the blue Ys can easily drop down and take out the yellow Ys if required. But nothing comes without a price. Attacking forwards are really good at getting the puck to unmarked Ds. You often see forwards send the puck around the entire attacking zone along the boards to the D at the far end, that can be done in the picture to the right but not left. Zucc uses this often, and its interesting how he doesn't look if we have a D open there or not, its hard to get time and see that, but he just sees how far down the defending winger is, if he is deep Zucc throws a no look pass around the boards to the D at the far point. And of course, every once in a while that D is changing and the puck goes out of the zone...

Under AV we played much more like the picture to the left and didn't quite get those layers. We didn't just opt to leave guys open, but when situations occurred we got exposed. You undoubtedly must play like more to the right in todays game, forwards are so skilled that the Ds need back-up and you must take away the ice. OTOH you of course both want to eat the cookie and keep it, and in all these instances in hockey you get into the magical word used -- "decision making". Making the small decisions. Like there is no reason for the defending wingers to drop down when the Ds have the guy with the puck under control? Right? So you leave it up to the winger to decide, when its needed you collapse deep, when its not you stay up. Then you just try to educate your personell, and that certainly is a work in progress. The term decision making is used by coaches in relation to a situation where its best from one perspective to play one way but from another perspective its better to play another way, instead of picking your poison you try to play both ways and have the players making individual decisions on the ice as to when to play each respective way.

Just like you I think we over committed on the back check when you made that post. Like we really had the forward under control but the wingers went down real deep and that forward could more or less fire the puck hard along the boards -- of course unstoppable for our defenders, and the puck was picked up by un-marked Ds on the points.

 

[Ola]

In addition to the above, the ability of teams to collapse deeper with the forwards and leave the points open to a larger degree has also improved with the players ability to block shots in the game today.

And what kind of is unclear to me is what direction the above is heading in the last years.

 

[Ola]

You have a video example so I can see exactly the kind of the play you're referring to? Also, this seems like something that would be handled in the prescout/film room, not sure how it relates to analytics or statistics

I can show it with these two drawings.

This is basically a very standardized transition play, i.e. the left wing leaves a little early and if you can't make a stretch pass immidiately the left wing turns into center ice and heads back towards his D. Its hard to take away the passing lane to the LW and the LW can direct the puck out towards the C on the left side or the RW on the right side.

This is what Hackstol had his team doing instead, simply, instead of going up along the boards and then cutting to the middle, the LW cuts into central ice and then turns out towards the boards.

Like said I haven't seen a ton of Philly, but when I have seen them teams have had a really hard time containing them, it feels like the pucks are always bouncing their way, but that winger coming in from like the blind side of the D -- more or less -- engaging in that puck battle with speed is of course hard to handle.

The reason I was posting it here was because I was wondering if it was showing in the data that it is effective in any way? Naturally Philly is a struggling team that hardly have impressive fundamentals. But I recon that there isn't much breakout data available, I've seen references to breakout stats but it doesn't seem like they are public.

 

[Filthy Dangles]

I can show it with these two drawings.

This is basically a very standardized transition play, i.e. the left wing leaves a little early and if you can't make a stretch pass immidiately the left wing turns into center ice and heads back towards his D. Its hard to take away the passing lane to the LW and the LW can direct the puck out towards the C on the left side or the RW on the right side.

This is what Hackstol had his team doing instead, simply, instead of going up along the boards and then cutting to the middle, the LW cuts into central ice and then turns out towards the boards.

Like said I haven't seen a ton of Philly, but when I have seen them teams have had a really hard time containing them, it feels like the pucks are always bouncing their way, but that winger coming in from like the blind side of the D -- more or less -- engaging in that puck battle with speed is of course hard to handle.

The reason I was posting it here was because I was wondering if it was showing in the data that it is effective in any way? Naturally Philly is a struggling team that hardly have impressive fundamentals. But I recon that there isn't much breakout data available, I've seen references to breakout stats but it doesn't seem like they are public.

Hockey analytics are far, far away from saying how successful that particular set play that might be. There are those blueline stats that attempt to say how good or bad a team is at giving and gaining the lines with the puck but I highly doubt the people who record it account for particular set plays on breakouts or what routes or configurations of the chess pieces might have ran in the neutral zone. You need live puck tracking to try and solve that.

Anyway, it's a smart play you'll see teams use in control breakouts or regroups, maybe PHI uses it more than others. It forces a tough decision on the opposing RD. If he should cross the red line and try to gap up on their LW or back off so as to not get caught. And the Philadelphia LD will have to fill that lane if LW gets caught on the boards.

 

[Machinehead]

I don't get Washington. Their underlying numbers suck again. They're in first place again.

 

[GeorgeKaplan]

I don't get Washington. Their underlying numbers suck again. They're in first place again.

Are they just kind of like how AVs Rangers were in that they have a really good goalie and they’re efficient at scoring but bad at everything else?

 

[SA16]

You don't need great underlying numbers when you have possibly the best goal scorer of all time.

It's not their goalie though Holtby has been bad this year (until this past week) and was bad last year.

 

[sbjnyc]

I can show it with these two drawings.

This is basically a very standardized transition play, i.e. the left wing leaves a little early and if you can't make a stretch pass immidiately the left wing turns into center ice and heads back towards his D. Its hard to take away the passing lane to the LW and the LW can direct the puck out towards the C on the left side or the RW on the right side.
View attachment 158797

This is what Hackstol had his team doing instead, simply, instead of going up along the boards and then cutting to the middle, the LW cuts into central ice and then turns out towards the boards.
View attachment 158799

Like said I haven't seen a ton of Philly, but when I have seen them teams have had a really hard time containing them, it feels like the pucks are always bouncing their way, but that winger coming in from like the blind side of the D -- more or less -- engaging in that puck battle with speed is of course hard to handle.

The reason I was posting it here was because I was wondering if it was showing in the data that it is effective in any way? Naturally Philly is a struggling team that hardly have impressive fundamentals. But I recon that there isn't much breakout data available, I've seen references to breakout stats but it doesn't seem like they are public.

I swear I saw similar diagrams in a physics class.

 

[Ola]

I don't get Washington. Their underlying numbers suck again. They're in first place again.

Sometimes I feel like the entire league suck...

But seriously, it’s not like it takes a ton to win in this league right now. Look at Montreal...

 

[Ola]

Hockey analytics are far, far away from saying how successful that particular set play that might be. There are those blueline stats that attempt to say how good or bad a team is at giving and gaining the lines with the puck but I highly doubt the people who record it account for particular set plays on breakouts or what routes or configurations of the chess pieces might have ran in the neutral zone. You need live puck tracking to try and solve that.

Anyway, it's a smart play you'll see teams use in control breakouts or regroups, maybe PHI uses it more than others. It forces a tough decision on the opposing RD. If he should cross the red line and try to gap up on their LW or back off so as to not get caught. And the Philadelphia LD will have to fill that lane if LW gets caught on the boards.

Another thing I was wondering about was what numbers people have for the PPs with the sticks out towards the boards? Elias Pettersson and co loves those PPs. But they just seem so backwards. High left shooting LW and high right shooting RW and a point.

But sure, it’s hard to take away those wristers...

 

[ManUtdTobbe]

Another thing I was wondering about was what numbers people have for the PPs with the sticks out towards the boards? Elias Pettersson and co loves those PPs. But they just seem so backwards. High left shooting LW and high right shooting RW and a point.

But sure, it’s hard to take away those wristers...

The thought behind it is better puck management because it's easier to keep the passes away from the PKers, the clear negative is that there's no one timers.