Dave Nonis: [mod: Commentary on advanced stats] (Video)

[tarheelhockey]

If this is the point he's trying to make, it's not a legitimate point. If your goalie sucks, you can only LESSEN the impact of his lack of skill by improving your possession game.

I don't think he would disagree with that, based on what he said. He emphasized that common sense would support the idea that possession drives success.

The problem is without analyzing the outputs, you have no way of evaluating the inputs. Determining that Coach Bob using this sort of X and O's structure is completely useless if you don't bother to determine whether that X and O's structure results in having the puck more often than the other team.

Again, common sense tells us that possession drives success. That much is obvious.

What a GM wants to know is: to what extent will Coach Bob's system enhance the ability of the players on my roster? Hockey's advanced stats community has yet to discover a method for answering that kind of question. Nonis even left the door open for future discovery -- he explicitly stated that he believes there could be a statistical model for management-level decision making, it just hasn't crossed his desk yet.

Instead of taking that as an insult to the community, take it as a challenge.

It seems to me like you're missing a key link in the equation. In order to get information, you have to be willing to actually collect the information. Your argument, in a financial context, would boil down to something like this: "I want to know how much money I have, but I don't really think it's important that anyone bothers to count my money." Or, alternatively "I want to have more money, but I don't think it's important to know how much money I have right now or what impact my decisions have on whether I make more money or not."

I want to improve puck possession, but it's not useful to know how often I have the puck, or how well potential new players drive puck possession.

The bolded is a completely inaccurate analogy. Someone in Nonis' position is not interested in being told how much money he already has. He is already tracking that penny-by-penny on a daily basis.

In regard to the second analogy, I'm not sure why you think Nonis doesn't care how much money he has. His job is to make more money, by acquiring the correct combination of employees who will combine to maximize his company's income. Simply counting the amount of money that his employees are currently making, or comparing it to the figures of employees at other companies, won't get that job done. He needs to know EXACTLY what those people are doing to create income, because otherwise he will just have a collection of guys who were formerly successful in other places. That might work in a highly quantifiable field like finance, but it doesn't translate to something like building a hockey team where the quantifiables are all propped up by unquantifiable factors.

 

[tarheelhockey]

So tell me again why it's unimportant to understand how many shots you're giving up?

That's a strawman that nobody including Nonis has proposed.

 

[PorkChopSandwiches]

Corsi has been proven to have predictive value regarding a team's chances of winning. I'm not so sure you can say it has been proven to increase the chances of winning through some sort of application.

most correct statement ive seen re: Corsi / Fenwick

 

[TKB]

1) Hitting more leads to winning more.
2) Blocking more shots leads to winning more.
3) Guys who turn the puck over more are worse players.
4) Dumping and chasing is better than carrying the puck in.

and the list goes on.

Your statements are far to vague and/or imo not accurate in terms of being considered "common hockey knowledge beliefs" related to winning.

1) and 2) - Hitting more than what? Blocking more shots than what? Are you suggesting that a player in position to deliver a check or block a shot should not do so? If you are talking about team philosophies such as a "shot-blocking" team or a "hard hitting" team, the fact that different teams take different approaches to these aspects of the game would suggest that there is not a universal hockey belief that a team has to be a hard hitting and/or shot blocking team to win.

3) Again vague. What player and what kind of turnover. Turns it over more than who and is worse than who?. If an 8 minute night guy makes 4 turnovers at the blue line (offensive or defensive) he is probably packing his bags the next day. On the other hand everyone in hockey understands that guys such as Patrick Kane are going to make turnovers, he still (appropriately) gets 20 minutes a night and awarded the Conn Smythe.

4) I don't agree at all that it is common hockey knowledge that dump and chase is universally considered better than carrying the puck in. Puck possession (as s style -not just a statistic) has been in the North American Game a lot longer than the wide-spread use of "hockey analytics".

 

[iCanada]

1) Hitting more leads to winning more.
2) Blocking more shots leads to winning more.
3) Guys who turn the puck over more are worse players.
4) Dumping and chasing is better than carrying the puck in.

and the list goes on.

You are making yourself look pretty ignorant here...

You come in here all off the rails because Nonis says Corsi doesn't help him do his job. Then you throw out the bolded nugget; If you dump the puck in you get better Corsi/Fenwick. If you carry it in there is no change.

Your own metric says dumping and chasing is better.

 

[hatterson]

You are making yourself look pretty ignorant here...

You come in here all off the rails because Nonis says Corsi doesn't help him do his job. Then you throw out the bolded nugget; If you dump the puck in you get better Corsi/Fenwick. If you carry it in there is no change.

Your own metric says dumping and chasing is better.

Really?

I've seen data that says the exact opposite.

 

[iCanada]

Really?

I've seen data that says the exact opposite.

I could be wrong on that.

I havn't looked at a whole lot of the data to be honest, was a logical assumption on my part. Does a dump in not count as a shot attempt? Even if fired directly wide? If a Shea Weber shoots from the point wide on purpose that counts.

I'm totally willing to eat crow if I am wrong. I suppose it depends on the frequency of you winning that race to the puck and the impending puck battle, if you just give up possession thats probably bad for your corsi.

 

[YoSoyLalo]

"It doesn't dictate who's gonna win"

Uh, yes it does. Sure, there are exceptions, but what teams that have won the cup the past few years have been poor puck possession teams? The answer is none.

What an idiot. There are posters on HFBoards who could probably run a team better than this moron.

 

[hatterson]

I could be wrong on that.

I havn't looked at a whole lot of the data to be honest, was a logical assumption on my part. Does a dump in not count as a shot attempt? Even if fired directly wide?

I'm totally willing to eat crow if I am wrong.

A dump in would not be counted as a shot attempt unless fired at the net. A large majority of dumpins are not directed towards the net, especially as part of a dump and chase strategy, to prevent the goalie from being able to play the puck.

Granted zone entries isn't an area I've studied much, I've generally just skimmed the articles, but from what I've seen it's anywhere from 1.5 to 2 times as effective to carry the puck in as opposed to dump and chase.

 

[fedfed]

1) Hitting more leads to winning more.
2) Blocking more shots leads to winning more.
3) Guys who turn the puck over more are worse players.
4) Dumping and chasing is better than carrying the puck in.

and the list goes on.

I'd add ones about fighting impacting momentum and +/- defining defensive abilities.

 

[iCanada]

A dump in would not be counted as a shot attempt unless fired at the net. A large majority of dumpins are not directed towards the net, especially as part of a dump and chase strategy, to prevent the goalie from being able to play the puck.

Granted zone entries isn't an area I've studied much, I've generally just skimmed the articles, but from what I've seen it's anywhere from 1.5 to 2 times as effective to carry the puck in as opposed to dump and chase.

Okay, but what is the difference between a Dman purposely shooting wide, and a dump in? How do you know the intention of the player?

Thats one thing that I dont really like about Corsi as a stat, it is very finnicky...

One guy might say the D just missed the net, another might say it isn't counted as a chance at all. Same with a puck battle between two players in the slot; Player A tries to shoot, Player B pokechecks and strips him as the shot is being released. Is that a blocked shot, or is that no shot at all?

Very subjective in my opinion.

 

[tarheelhockey]

Granted zone entries isn't an area I've studied much, I've generally just skimmed the articles, but from what I've seen it's anywhere from 1.5 to 2 times as effective to carry the puck in as opposed to dump and chase.

I'm curious how contextual variables are controlled in those studies. I mean, if you have 5 defenders standing you up at the blue line it's probably not going to be very effective to try and carry the puck through.

 

[hatterson]

I'm curious how contextual variables are controlled in those studies. I mean, if you have 5 defenders standing you up at the blue line it's probably not going to be very effective to try and carry the puck through.

Honestly I'm not entirely sure. When I get home I'll dig back through the articles I saw on it and find out, although I know zone entries is still being actively researched so it's possible the answers could change in the coming months/years.

Another question I have regarding it is how causative it is. Better possession teams carry the puck in more often than the dump it in. Well is carrying vs. dumping actually helping them be better possession teams, or are they just straight up better at hockey and thus have more ability to actually carry the puck in as opposed to getting shut out at the line. If a dman has me rushing at him, he has zero fear of me blowing past him so can easily step up and either force a turnover or force me to dump it in. However if it's Crosby coming at him that's an entirely different scenario.

 

[KaseMeOutside]

If I was a team with a "smart" front office, i'd be happy that someone like nonis is a GM. It's someone who can be taken advantage of

 

[Yossarian54]

Honestly I'm not entirely sure. When I get home I'll dig back through the articles I saw on it and find out, although I know zone entries is still being actively researched so it's possible the answers could change in the coming months/years.

Another question I have regarding it is how causative it is. Better possession teams carry the puck in more often than the dump it in. Well is carrying vs. dumping actually helping them be better possession teams, or are they just straight up better at hockey and thus have more ability to actually carry the puck in as opposed to getting shut out at the line. If a dman has me rushing at him, he has zero fear of me blowing past him so can easily step up and either force a turnover or force me to dump it in. However if it's Crosby coming at him that's an entirely different scenario.

Here you go:

http://nhlnumbers.com/2012/7/9/how-important-is-neutral-zone-play

http://nhlnumbers.com/2012/7/11/more-on-the-advantages-of-puck-possession-over-dump-and-chase

It seems the amount of shots and goals generated by a carry-in or a pass-in are literally more than double those generated by a dump- or tip- in.

Tyler Dellow's also done a ton of stuff focused specifically on Taylor Hall - how an increased dump-in percentage by the Oiler's when he's on the ice has lead to a precipitous fall in his Corsi%.

 

[hullsy47]

1) Hitting more leads to winning more.
2) Blocking more shots leads to winning more.
3) Guys who turn the puck over more are worse players.
4) Dumping and chasing is better than carrying the puck in.

and the list goes on.

why people are trying to defend nonis /burke is beyond me ......hed have been fired next year anyways when this team didnt play for the next coach u bring in
problem with nonis is he hasnt got burke shooting off his yap for him ......until burke can weasel him back to calgary
even calgary board wont endorse nonis

 

[tarheelhockey]

Here you go:

http://nhlnumbers.com/2012/7/9/how-important-is-neutral-zone-play

http://nhlnumbers.com/2012/7/11/more-on-the-advantages-of-puck-possession-over-dump-and-chase

It seems the amount of shots and goals generated by a carry-in or a pass-in are literally more than double those generated by a dump- or tip- in.

Tyler Dellow's also done a ton of stuff focused specifically on Taylor Hall - how an increased dump-in percentage by the Oiler's when he's on the ice has lead to a precipitous fall in his Corsi%.

This is exactly the kind of thing Nonis is talking about.

This article, which is linked about 1/3rd of the way into the first one you linked, actually lays out the specific conclusions of zone-entry analysis:

This leaves us with the following picture:

• Carrying the puck in definitely generates more offense, and should be attempted at every opportunity when trailing.
• The top 9 should usually try to carry the puck in whenever they can, and top-9 players who often carry the puck in are helping the team with their aggressive play. However, with a lead towards the end of the game there may be cause to run designed dump-in plays that give up fewer counterattacks.
• The fourth line should carry the puck in when they have a clear opportunity, but should be a bit more cautious with borderline plays.

Now, I fully agree that from our perspective as fans this is all interesting stuff. Quantifying the impact of carrying the puck vs. dumping it is useful when we comb over the details of a particular game or series. It allows us to describe the teams' performance with much greater accuracy than before.

But... from the perspective of a GM, in what way are these analyses supposed to be useful? Carrying the puck leads to more offense than dumping it in -- yeah, no ****. Anyone watching a game for the first time has an intuitive negative reaction to dump-ins, because they're visibly less effective than rushes with possession. That isn't a helpful observation for a pro. "Hey Gretz, you score a lot more when you carry the puck across the blue line! You should carry it more!"... "Thanks kid."

The basic problem is that those analyses lack a control for the circumstances under which a player decides to dump the puck. The event doesn't happen in a vacuum. Until someone comes up with a model that can be translated into actual game strategy and personnel decisions, a GM is just going to respond to these kinds of numbers with a polite "oh that's interesting" and keep moving on.

 

[TKB]

This is exactly the kind of thing Nonis is talking about.

This article, which is linked about 1/3rd of the way into the first one you linked, actually lays out the specific conclusions of zone-entry analysis:



Now, I fully agree that from our perspective as fans this is all interesting stuff. Quantifying the impact of carrying the puck vs. dumping it is useful when we comb over the details of a particular game or series. It allows us to describe the teams' performance with much greater accuracy than before.

But... from the perspective of a GM, in what way are these analyses supposed to be useful? Carrying the puck leads to more offense than dumping it in -- yeah, no ****. Anyone watching a game for the first time has an intuitive negative reaction to dump-ins, because they're visibly less effective than rushes with possession. That isn't a helpful observation for a pro. "Hey Gretz, you score a lot more when you carry the puck across the blue line! You should carry it more!"... "Thanks kid."

The basic problem is that those analyses lack a control for the circumstances under which a player decides to dump the puck. The event doesn't happen in a vacuum. Until someone comes up with a model that can be translated into actual game strategy and personnel decisions, a GM is just going to respond to these kinds of numbers with a polite "oh that's interesting" and keep moving on.

It astounds me that this does not get addressed (in a meaningful way) in all these zone entry "analysis" discussions.

I bet Boone Jenner woke up this morning wishing he hadn't opted for the controlled entry last night.

 

[Cap'n Flavour]

That's a strawman that nobody including Nonis has proposed.

He said Corsi is meaningless. Did anyone actually listen to the clip?

 

[tarheelhockey]

He said Corsi is meaningless. Did anyone actually listen to the clip?

Yes, I listened to the clip. That's how I know that he also said that it's important to have better possession and more shots than the other team.

 

[Cap'n Flavour]

Yes, I listened to the clip. That's how I know that he also said that it's important to have better possession and more shots than the other team.

So tell me again why it's unimportant to understand how many shots you're giving up?

That's a strawman that nobody including Nonis has proposed.

Nonis quite literally said that a measure of shots (given up and taken) is meaningless. I don't think it's a strawman to infer that he doesn't think it's important to understand Corsi/Fenwick, considering that he said that it's meaningless.

He also said they were 'interesting', in the sense of providing pretty pictures, which may as well be saying that they're not interesting at all.

 

[SaskRinkRat]

This is exactly the kind of thing Nonis is talking about.

This article, which is linked about 1/3rd of the way into the first one you linked, actually lays out the specific conclusions of zone-entry analysis:



Now, I fully agree that from our perspective as fans this is all interesting stuff. Quantifying the impact of carrying the puck vs. dumping it is useful when we comb over the details of a particular game or series. It allows us to describe the teams' performance with much greater accuracy than before.

But... from the perspective of a GM, in what way are these analyses supposed to be useful? Carrying the puck leads to more offense than dumping it in -- yeah, no ****. Anyone watching a game for the first time has an intuitive negative reaction to dump-ins, because they're visibly less effective than rushes with possession. That isn't a helpful observation for a pro. "Hey Gretz, you score a lot more when you carry the puck across the blue line! You should carry it more!"... "Thanks kid."

The basic problem is that those analyses lack a control for the circumstances under which a player decides to dump the puck. The event doesn't happen in a vacuum. Until someone comes up with a model that can be translated into actual game strategy and personnel decisions, a GM is just going to respond to these kinds of numbers with a polite "oh that's interesting" and keep moving on.

It's hard for me to believe that you are being authentic in your criticism. Your argument suggests that "knowing what works" is not a pre-requisite for "doing what works".

Eventually, when the world eradicates cancer, the cure is going to be heavily based in a phenomenally thorough understanding of what causes cancer.

The analytics community is completely open about the fact that more work needs to be done to understand what specific skills / behaviours promote the puck possession style (that results in more goals for and less goals against, other things equal). But if you're a GM in the league, and you believe the mounting evidence that shot attempts are directly correlated with puck possession and that puck possession drives wins, you are a complete moron if you don't hop on the bandwagon and start trying to understand what behaviours/systems/skills drive puck possession. It would be absolutely inexcusable in a private business setting, and you would be instantly fired.

"Anyone watching a game for the first time has an intuitive negative reaction to dump-ins, because they're visibly less effective than rushes with possession."

Then why do teams still dump the puck in at a horrendously high rate?

 

[hatterson]

Nonis quite literally said that a measure of shots (given up and taken) is meaningless. I don't think it's a strawman to infer that he doesn't think it's important to understand Corsi/Fenwick, considering that he said that it's meaningless.

He also said they were 'interesting', in the sense of providing pretty pictures, which may as well be saying that they're not interesting at all.

He also said:

"Any person with a brain would say you want to have the puck more than you don't."

"We'd like to have more shots on net, we'd like to possess the puck more than we do right now."

"Puck possession alone doesn't mean anything. Do you want to have it more than the other team? Of course you do." (Emphasis added by me)

"The simple thing about possession is you want to have the puck more than the other team [but] it doesn't dictate who's gonna win the game"

It's clear he's not just sitting there saying that it's ideal the Leafs only had a corsi for percentage in the 41% range. However he's saying that the goal shouldn't be to improve your corsi stat, it should be to improve your hockey team.

 

[hatterson]

But if you're a GM in the league, and you believe the mounting evidence that shot attempts are directly correlated with puck possession and that puck possession drives wins, you are a complete moron if you don't hop on the bandwagon and start trying to understand what behaviours/systems/skills drive puck possession.

Why do you believe Nonis doesn't understand this? The video has multiple mentions of his saying that possession is important and you want to have the puck more.

"Anyone watching a game for the first time has an intuitive negative reaction to dump-ins, because they're visibly less effective than rushes with possession."

Then why do teams still dump the puck in at a horrendously high rate?

Perhaps because carrying the puck in is harder? It means you're potentially taking a hit instead of giving on and it gives the other team a chance to poke it off your stick and counter attack.

 

[SaskRinkRat]

He also said:

"Any person with a brain would say you want to have the puck more than you don't."

"We'd like to have more shots on net, we'd like to possess the puck more than we do right now."

"Puck possession alone doesn't mean anything. Do you want to have it more than the other team? Of course you do."

"The simple thing about possession is you want to have the puck more than the other team [but] it doesn't dictate who's gonna win the game"

It's clear he's not just sitting there saying that it's ideal the Leafs only had a corsi for percentage in the 41% range. However he's saying that the goal shouldn't be to improve your corsi stat, it should be to improve your hockey team.

Two things:

1) Yes, it does dictate who is going to win the game over time. Think about it like a poker hand. If you're playing Hold'em and you have AA, you're not going to beat 72 off-suit every hand, but over time (over a series of hands) having AA is going to demolish 72.

2) Improving your hockey team will, in all likelihood, improve your Corsi %. The best known way to improve your hockey team is to increase how often you have the puck. The more you have the puck, the better your Corsi% will be.

Hence my conclusion that Nonis fundamentally misunderstands relationships between all of these concepts. It's the same sort of logic that gets certain players (guys like Subban, Campbell, etc.) as less desirable because they are "defensively irresponsible". If you understand that playing in the other teams end all of the time reduces the need to be defensively responsible (because you have to play less defence), then you understand the value in a player like Subban or Campbell.