Don Granato Interim Coach for Sabres/Ellis and Girardi to the Bench. Steve Smith Fired

[LTS]

6 minutes ago, Curt said:

How?

EDIT:  not snarky, a honest and serious question.

"Luck is what happens when preparation meets opportunity." - Roman Philosopher, Seneca.

In short, the Sabres are now preparing, through practice, to have the opportunity, through games, to improve their luck

[LGR4GM]

19 minutes ago, Curt said:

Don’t you think that they have looked better though?

Even if the xGF% doesn’t reflect it, they seem to be playing looser and with more confidence.

Maybe it’s just some combination of facing bad teams/goaltending, Ullmark being back, extra pep from the new coach, and the results of puck luck coloring my perception a bit.  I don’t know.

Yes, they do look better and I think they are playing better. I also think that Ullmark being back is a major factor in winning. On top of that the bolded is a very important factor in all this. Philly, Devils, and the Rangers aren't great teams. The Rangers are probably the best of the group. 

[darksabre]

3 minutes ago, LTS said:

"Luck is what happens when preparation meets opportunity." - Roman Philosopher, Seneca.

In short, the Sabres are now preparing, through practice, to have the opportunity, through games, to improve their luck

I like that.

I never for a second bought into the argument that the Sabres were good but just unlucky. The stat models don't tell the whole story. You watch that team on the ice and you see that they aren't doing the things they need to do to win and there's no way you can just call it bad luck. 

[Curt]

7 minutes ago, LTS said:

"Luck is what happens when preparation meets opportunity." - Roman Philosopher, Seneca.

In short, the Sabres are now preparing, through practice, to have the opportunity, through games, to improve their luck

What you describe is not luck.  You are describing playing better hockey.

[Doohickie]

20 minutes ago, SwampD said:

You make your own luck. 

This.  Look at the Asplund goal.  High skill play for sure, but he had room because he and Mitts used speed to create it.  That's an example of making your own luck. They just didn't move their feet like that under Krueger.

[Curt]

1 minute ago, Doohickie said:

This.  Look at the Asplund goal.  High skill play for sure, but he had room because he and Mitts used speed to create it.  That's an example of making your own luck. They just didn't move their feet like that under Krueger.

I don’t get it.  Who would call that lucky?

That’s not luck!  That’s playing good hockey!  They weren’t making luck.  They were making a good hockey play.

[Doohickie]

How is PDO calculated?  Does it take into account the speed of the player taking the shot?  They increase their chances (and their luck) by moving.

When they're static set pieces it's easier to defend so their shots, even from the same areas, are not as likely to go in.

[LTS]

25 minutes ago, Curt said:

What you describe is not luck.  You are describing playing better hockey.

Exactly.  That's the point of the quote.

There is no such thing as luck.  If you get analytical about anything you can find something that would explain how the outcome of a situation was impacted by a certain factor.

If people want to call it "luck" that's fine but there is no such real thing.  If I enter the lottery and win, it's not luck.  It took a conscious action of entering a contest into which there was a sleight probability of me winning. Yet, there was a probability of that outcome and as such, am I "lucky" it happened?  Sure, but it was a potential outcome, however improbable, that was only brought about by my preparation "the choosing to buy a ticket".

Now, winning the lottery without ever entering, that would be something.

 

[DarthEbriate]

In my experience there's no such thing as luck. Good against the Devils is one thing. Good against the Islanders? That's something else.

[Curt]

11 minutes ago, LTS said:

Exactly.  That's the point of the quote.

There is no such thing as luck.  If you get analytical about anything you can find something that would explain how the outcome of a situation was impacted by a certain factor.

If people want to call it "luck" that's fine but there is no such real thing.  If I enter the lottery and win, it's not luck.  It took a conscious action of entering a contest into which there was a sleight probability of me winning. Yet, there was a probability of that outcome and as such, am I "lucky" it happened?  Sure, but it was a potential outcome, however improbable, that was only brought about by my preparation "the choosing to buy a ticket".

Now, winning the lottery without ever entering, that would be something.

 

You take certain actions.  Those actions have a chance of a positive or negative result, depending on things that are out of your control.  If the things out of your control result in a positive outcome, that is lucky.  Jeebus.

You said that there is no such thing as luck, then outlined an exact scenario that demonstrated the existence of luck.  Luck is not magic, it’s just probabilities working out favorably.

Winning the lottery without entering would not be luck.  That would be magic.  Lol

Edited April 7, 2021 by Curt

[JoeSchmoe]

I'm an engineer that works in a world of numbers all day long.

GF is a product of an infinite number of variables. XGF is a product of a finite number of variables.

The eye test has the uncanny ability to process a nearly infinite number of variables that XGF cannot. It's kind of like how humans can beat a supercomputer at chess.

Eye test wise, they're playing much better than before, and deserving of the 8pts in 5 games. (Though NJ could have maybe had a few more goals last night.)

[darksabre]

15 minutes ago, LTS said:

Exactly.  That's the point of the quote.

There is no such thing as luck.  If you get analytical about anything you can find something that would explain how the outcome of a situation was impacted by a certain factor.

If people want to call it "luck" that's fine but there is no such real thing.  If I enter the lottery and win, it's not luck.  It took a conscious action of entering a contest into which there was a sleight probability of me winning. Yet, there was a probability of that outcome and as such, am I "lucky" it happened?  Sure, but it was a potential outcome, however improbable, that was only brought about by my preparation "the choosing to buy a ticket".

Now, winning the lottery without ever entering, that would be something.

 

This is why I think the fancy stats stuff is still lagging behind in providing context. They might be able to look at things like shot quality and say "this player is just unlucky" but they really still aren't getting into the meat of what makes hockey so interesting from an analytics standpoint: the flow of the play.

I still believe that until we are modeling player movement on the ice and analyzing how the way one player moving around the ice and taking certain actions influences the play of the other players on the ice, we can't really look at stuff like Expected Goals and come to strong conclusions.

So when the stats crowd was hemming and hawing about players just being unlucky it made me go "hmmm" because all that said to me is that the stats they were using to make that argument were inadequate.

Edited April 7, 2021 by darksabre

[Torpedo Forecheck]

Gave up bothering with arguing with the analytics crowd a long time ago, and in general I'm a numbers guy. If you've got something worthwhile let me know and I'll use it because I want to win. The luck argument (and other concepts) get taken way to far IMO. What do you do with it?.....Oh we've been lucky now we're gonna lose some. Or we've been unlucky and we're better than our record and we will win when its our time....which of course runs the risk of becoming excuse making. Please don't act like biases people have with their eyes are so limited as to mean nothing. And, please don't act like the numbers capture everything we need to know, and give the missing not measured variables some credit too.

[LGR4GM]

19 minutes ago, JoeSchmoe said:

I'm an engineer that works in a world of numbers all day long.

GF is a product of an infinite number of variables. XGF is a product of a finite number of variables.

The eye test has the uncanny ability to process a nearly infinite number of variables that XGF cannot. It's kind of like how humans can beat a supercomputer at chess.

Eye test wise, they're playing much better than before, and deserving of the 8pts in 5 games. (Though NJ could have maybe had a few more goals last night.)

I don't agree with how you are wording this or what you are saying. xGF is expected goals for based on multiple variables that do in fact matter. It is predictive in nature. Teams with higher xGF% are better and teams with worse xGF% are worse. Your eyes only tell you what you have watched in that game, xGF models take into account all games. Yes they are playing better but your baseline is the Sabres and xGF% baseline is the league. 

https://www.naturalstattrick.com/teamtable.php

[LGR4GM]

7 minutes ago, Torpedo Forecheck said:

Gave up bothering with arguing with the analytics crowd a long time ago, and in general I'm a numbers guy. If you've got something worthwhile let me know and I'll use it because I want to win. The luck argument (and other concepts) get taken way to far IMO. What do you do with it?.....Oh we've been lucky now we're gonna lose some. Or we've been unlucky and we're better than our record and we will win when its our time....which of course runs the risk of becoming excuse making. Please don't act like biases people have with their eyes are so limited as to mean nothing. And, please don't act like the numbers capture everything we need to know, and give the missing not measured variables some credit too.

We aren't talking about luck. We are talking about expected goals for percentage. The Sabres were able to score more goals than expected while simultaneously allowing fewer goals than expected. Since xGF involves average save percentage in the same situation, we are seeing Ullmark being a better goalie than Blackwood. Hence why we could run a negative xGF% and still win last night. In the end though to be a long term successful team you want to get your xGF% positive because you end up relying on your goalie saving you less. 

xGF is more of a benchmark to consider and it isn't about luck. For example Chicago has the worse xGF% in the league. Either they are getting great goaltending, outperforming their xGF or some combination but just like the Sabres getting shutout constantly and have decent xGF but getting almost nothing, eventually regression happens. I think we are seeing regression from Buffalo right now and yes part of that is Granato opening it up (I think when plays have more puck movement or happen on rushes your xGF goes up and the stat only kinda of incorporates pre shot movement from my understanding) and Ullmark being a good goalie. 

xGF% certainly does not capture everything we need to know. 100% agree

[JoeSchmoe]

2 minutes ago, LGR4GM said:

I don't agree with how you are wording this or what you are saying. xGF is expected goals for based on multiple variables that do in fact matter. It is predictive in nature. Teams with higher xGF% are better and teams with worse xGF% are worse. Your eyes only tell you what you have watched in that game, xGF models take into account all games. Yes they are playing better but your baseline is the Sabres and xGF% baseline is the league. 

https://www.naturalstattrick.com/teamtable.php

I'm saying as much as people want to live in a world where they can use finite stats to exactly predict outcomes, the game of hockey is too "analog" for this to more accurate than the human eye, which can account for far more variables with far more dynamic weighting.

[LGR4GM]

1 minute ago, JoeSchmoe said:

I'm saying as much as people want to live in a world where they can use finite stats to exactly predict outcomes, the game of hockey is too "analog" for this to more accurate than the human eye, which can account for far more variables with far more dynamic weighting.

Stats don't exactly predict outcomes. 

[bunomatic]

They’re simply working harder at their craft and believe in themselves again. Confidence. Ralph wanted them to buy into something that they had a difficult time believing would work. Granato offered them a simpler version. They bought in.

Edited April 7, 2021 by bunomatic

[pi2000]

It's easier to bury your chances when you're in better physical shape.   I think part of the poor puck luck was guys just being tired and missing their opportunities.   

They look lighter on their feet now, don't slow down as much during a shift, which tells me whatever Granato is doing in practice to condition these guys is working.   Maybe taking shorter shifts too.

[WildCard]

1 minute ago, pi2000 said:

It's easier to bury your chances when you're in better physical shape.   I think part of the poor puck luck was guys just being tired and missing their opportunities.   

They look lighter on their feet now, don't slow down as much during a shift, which tells me whatever Granato is doing in practice to condition these guys is working.   Maybe taking shorter shifts too.

Just putting in more effort for a better coach IMO. I don't think a month of conditioning really makes them that much more physical/faster, I think they're just trying harder.

[LGR4GM]

Fun fact, Buffalo is 12th in high danger corsi for chances. 

[WildCard]

2 minutes ago, LGR4GM said:

Fun fact, Buffalo is 12th in high danger corsi for chances. 

What were we under Ralph vs Granato? Any chance that's available? 

[LGR4GM]

5 minutes ago, WildCard said:

What were we under Ralph vs Granato? Any chance that's available? 

It probably is somewhere but I don't have that info on hand. 

Wait... when was Ralph fired?

[LGR4GM]

Krueger, 5v5 numbers, 28 games

235 hdcf = 48.55 HDCF%

48.05 xGF 53.25 xGA = 47.43 xGF%

90.23 sv%

 

Granato, 5v5 numbers, 10 games

83 hdcf = 51.23 HDCF%

15.82 xGF 18.03 xGA = 46.27 xGF%

93.62 sv%

Are there other numbers ppl want?

[Rasmus_]

8 minutes ago, LGR4GM said:

Krueger, 5v5 numbers, 28 games

235 hdcf = 48.55 HDCF%

48.05 xGF 53.25 xGA = 47.43 xGF%

90.23 sv%

 

Granato, 5v5 numbers, 10 games

83 hdcf = 51.23 HDCF%

15.82 xGF 18.03 xGA = 46.27 xGF%

93.62 sv%

Are there other numbers ppl want?

 

Definitely looks a lot more impressive.  I think he's let his players play as well.  Looking at Mittelstadt, Skinner, Montour and Dahlin in particular, and you see them having less stress on their play style.  If you could explain HDCF for me, I don't know that stat type.