Shot quality, expected goals and the Royal Road

[Roof-Daddy]

Originally Posted by Realtor 1

IMO there is no such thing as a high / medium / low scoring zone on the ice.

There absolutely is high, medium and low danger zones on the ice an it's undeniable.

Now, not every shot from a high danger zone has the same chance of going in as other shots from a high danger zone, but on average your chances of scoring from high danger zone are much better than scoring on a shot from a medium or low danger zone.

[Bingo]

Originally Posted by Itse

Have to say, the names on that list make me seriously doubt the methodology. Sure, Murray didn't have the best regular season, but he's no Scott Darling.

The Penguins are off the charts on almost all metrics this year but didn't have a dominant season, and their goaltending was certainly one of the things holding them back.

[gargamel]

Originally Posted by Bingo

All high danger metrics are based on an addition or subtraction equation based on ...

1) the shot has to be in home plate
2) and get a situational add to it to move it from a scoring chance to a high danger event

I have yet to find what the formula is, and what the adds and subtracts are, but it isn't as simple as every shot in a zone gets to be a high danger, and I suspect the royal road does come into the equation making more valid than many of you are suggesting.

From the Corsica site:

Here’s what my model does account for:

Shot type (Wrist shot, slap shot, deflection, etc.)
Shot distance (Adjusted4 distance from net)
Shot angle (Angle in absolute degrees from the central line normal to the goal line)
Rebounds (Whether or not the shot was a rebound)
Rush shots (Whether or not the shot was a rush shot)
Strength state (Whether or not the shot was taken on the powerplay)

[DeluxeMoustache]

There are a few random thoughts that have come to mind

- For shots where a goalie is set, they may miss or allow a perfect shot. Statistically this is expected to happen about once every twenty shots. So when you see a goalie beat cleanly, not every time is it ‘oh, he’s got to have that’

- when we were watching this year and people were saying saying too many shots were coming from the outside, people brought up the heat maps of shot location. Heat maps of shot location certainly reflect the location of the shot but not the context.

- Hartley’s emphasis on quick transition and activating the D tries to create an outnumbered situation, 4 on 3 if the opposing wingers are caught deep, which tries to create time in the O zone where there is space available for east-west movement

- moving as a 5 man unit, you have to try to create outnumbered situations in a more contained area

- Valiquette doesn’t account for 100 percent of goals in these 7 categories, but does put the 76 percent of goals scored on what he calls green shots into those 7 buckets. It is great to observe and to incorporate the knowledge in to a strategy, but it isn’t, as far as I can see, ready to be used predictively (also of course due to what data is actually tracked and how it can be translated in to pre shot situations)

[Bingo]

^ well I think we can all abandon the notion that his stats are too simplified then!

[GranteedEV]

Originally Posted by Cali Panthers Fan

It tells me that when Gaudreau and Monahan are sometimes guilty of overpassing the puck, it's because they realize this is how to best score goals. When I see Bennett fly down the wing and shoot it into the crest, I now know that it's just not an effective strategy to score goals, and he needs to work on finding his teammates more..

With a player as coachable as Bennett, I worry that there is a message to not try to make a play if there is a risk of a giveaway. Bennett had the least giveaways on the team last year among the regulars and I do not think that is a good thing. Bennett has shown high end vision and passing touch - but his royal road crossing play numbers are awful.

I fully agree that passes accross the royal road need to be utilized and I strongly worry that the messages of using the boards and managing the puck are having a detrimental effect on the younger players' offensive development. Not just Bennett but Jankowski, Mangiapane... basically anyone who didn't already establish themselves under Bob.

[The Boy Wonder]

Still would like to see them take into account things like number of defenders back, whether it’s a 2on1 or 3on2, shot speed, time since last shot, length of shift

I think getting into these types of details would truly give more accurate stats but this royal road is about the best I’ve seen yet

[DeluxeMoustache]

Originally Posted by Bingo

All high danger metrics are based on an addition or subtraction equation based on ...

1) the shot has to be in home plate
2) and get a situational add to it to move it from a scoring chance to a high danger event

I have yet to find what the formula is, and what the adds and subtracts are, but it isn't as simple as every shot in a zone gets to be a high danger, and I suspect the royal road does come into the equation making more valid than many of you are suggesting.

I can give you an example that Corsica would not capture as far as I know. Puck carrier one on one with the goalie. Scenario a, he comes straight in and shoots. Scenario b, he goes wide, cuts across the middle, and gets the goalie moving laterally. Corsica wouldn’t distinguish. But if I read correctly, cutting across the middle raises likelihood of scoring from ~5 percent to 33 percent

Definitely it is an improvement, what they are trying to do on Corsica, to try to improve categorization, but it doesn’t seem to be where it needs to be

[Bingo]

Originally Posted by DeluxeMoustache

I can give you an example that Corsica would not capture as far as I know. Puck carrier one on one with the goalie. Scenario a, he comes straight in and shoots. Scenario b, he goes wide, cuts across the middle, and gets the goalie moving laterally. Corsica wouldn’t distinguish. But if I read correctly, cutting across the middle raises likelihood of scoring from ~5 percent to 33 percent

Definitely it is an improvement, what they are trying to do on Corsica, to try to improve categorization, but it doesn’t seem to be where it needs to be

Totally agree, it's like your golf game it can always get better.

But given

a) the fact that it's not just every shot from within say home plate and
b) it's applied objectively across every game and every team

I wouldn't say it can be discounted as a pretty damn good summary either.

[DeluxeMoustache]

Originally Posted by Bingo

^ well I think we can all abandon the notion that his stats are too simplified then!

I don’t understand this remark

[edslunch]

Originally Posted by Bingo

^ well I think we can all abandon the notion that his stats are too simplified then!

But interestingly doesn't include the passing/goalie movement factor we're discussing here.

[Bingo]

Originally Posted by DeluxeMoustache

I don’t understand this remark

There were a lot of replies to the corsica data suggesting it was too simple or flawed.

[ricardodw]

Originally Posted by Roof-Daddy

There absolutely is high, medium and low danger zones on the ice an it's undeniable.

Now, not every shot from a high danger zone has the same chance of going in as other shots from a high danger zone, but on average your chances of scoring from high danger zone are much better than scoring on a shot from a medium or low danger zone.

Have to disagree. Having someone either skilled at dealing off the puck to high scoring areas AND/OR having team defense not controlling the passing lanes on a regular basis greatly changes the averages.

That is why coaches go to great lengths to match up defensemen who control the passing lanes against the other teams skilled puck movers/ shooters lines.

Gaudreau/ Monahan would dominate even more than they do if they get out against the other teams non-defense players.

I do not have any fancy stats to back it up (except for +/-) but I would venture to guess that there a lot more quality scoring chances generated against the Flames top-defense pair and best defense line than average. This puts extra pressure on the goalie to be outstanding.

Inference from playing time suggests that Hamilton is really bad at controlling passing lanes as he does not get put out on the PK.

[DeluxeMoustache]

Originally Posted by Bingo

Totally agree, it's like your golf game it can always get better.

But given

a) the fact that it's not just every shot from within say home plate and
b) it's applied objectively across every game and every team

I wouldn't say it can be discounted as a pretty damn good summary either.

See this is what I wonder. Say an average goalie throws up a .915. People generally say .905 is bad and .925 is great. You are looking at what causes that 1 percent deviation from average.

In the example I gave you, the Corsica model can’t, as far as I know, predict whether 8 percent of goals are 5 percent likely to happen or 33 percent.

Many people have thrown up their hands and said there is no such thing as shot quality, but we know that’s not true. Still their best models had enough noise to drive the R2 into meaningless territory.

I acknowledge that they are trying to put some context so it is not simply shot location, note that I didn’t dispute that.

I just don’t know, from that list, that what they are capturing is adequate to capture context, and generate the most meaningful results.

[Roof-Daddy]

Originally Posted by ricardodw

Have to disagree. Having someone either skilled at dealing off the puck to high scoring areas AND/OR having team defense not controlling the passing lanes on a regular basis greatly changes the averages.

That is why coaches go to great lengths to match up defensemen who control the passing lanes against the other teams skilled puck movers/ shooters lines.

Gaudreau/ Monahan would dominate even more than they do if they get out against the other teams non-defense players.

I do not have any fancy stats to back it up (except for +/-) but I would venture to guess that there a lot more quality scoring chances generated against the Flames top-defense pair and best defense line than average. This puts extra pressure on the goalie to be outstanding.

Inference from playing time suggests that Hamilton is really bad at controlling passing lanes as he does not get put out on the PK.

It doesn't matter if you disagree or not.

It isn't an opinion, it's a fact.

The league average for save percentage drops as shots move from low danger, to medium danger, to high danger.

Like I said, not all low danger shots are equal. Not all medium danger shots are equal. Not all high danger shots are equal.

But on average, the closer to the net the shot happens the better chance it's results in a goal.

2016/17

League average High danger sv% - 81.17%
League average Med danger sv% - 92.50%
League average Low danger sv% - 97.91%

...and it's like that every year.

[Bingo]

Originally Posted by ricardodw

I do not have any fancy stats to back it up (except for +/-) but I would venture to guess that there a lot more quality scoring chances generated against the Flames top-defense pair and best defense line than average. This puts extra pressure on the goalie to be outstanding.

Inference from playing time suggests that Hamilton is really bad at controlling passing lanes as he does not get put out on the PK.

Pretty sure that would be a bad guess.

Underlying stats that do exist, including from that same Corsica site have Hamilton 4th in the league in expected % of goals when he's on the ice. Giordano is 2nd.

The 3M line are all ranked between 45th and 75th in the league for forwards, so certainly where you'd expect with 31 teams.

[Bingo]

Originally Posted by DeluxeMoustache

See this is what I wonder. Say an average goalie throws up a .915. People generally say .905 is bad and .925 is great. You are looking at what causes that 1 percent deviation from average.

In the example I gave you, the Corsica model can’t, as far as I know, predict whether 8 percent of goals are 5 percent likely to happen or 33 percent.

Many people have thrown up their hands and said there is no such thing as shot quality, but we know that’s not true. Still their best models had enough noise to drive the R2 into meaningless territory.

I acknowledge that they are trying to put some context so it is not simply shot location, note that I didn’t dispute that.

I just don’t know, from that list, that what they are capturing is adequate to capture context, and generate the most meaningful results.

Absolutely, but its' deeper than just counting stats with no differentiation. With every step forward in differentiation it has to take on more meaning doesn't it?

Fans can be myoptic as hell, we all see things based on biases and team preferences.

I like having something to pull that doesn't have a rooting interest.

[ricardodw]

With the quality analysis and hockey theory so readily available, analyzed, studied and implemented I do not see the edge in having a coach who has as his qualification expertise in strategies and systems. ie GG and Peters.

Anyone making his living as a hockey coach should have a far greater understanding of this than anyone posting on this board.

The qualities of successful coaches on winning teams is the ability to adapt strategies often on the fly and motivate players.

The Flames have a veteran team (above the league average age) of highly paid very successful individual players that all have been exposed to various strategies and systems. I can't believe that they are so for the lack of a better word stupid that they wouldn't understand any system and the need to play as a team.

This team needs a leader and psychologist rather than a tactician.

[ricardodw]

Originally Posted by Bingo

Pretty sure that would be a bad guess.

Underlying stats that do exist, including from that same Corsica site have Hamilton 4th in the league in expected % of goals when he's on the ice. Giordano is 2nd.

The 3M line are all ranked between 45th and 75th in the league for forwards, so certainly where you'd expect with 31 teams.

You would have to agree that controlling the passing lanes is the mark of any superior defensive player. There is a reason that the prime goal of a defenseman on a 2 on 1 is to take away the pass.

What corisa stat touches on how well a player controls passing lanes?

It would seem that you have basically pointed out the flaw with this analysis.

If Hamilton and Gio were in fact league leading defenseman the Flames as a team would not be a the very best with all things going well and no injuries a bubble team.

[Bingo]

Originally Posted by ricardodw

You would have to agree that controlling the passing lanes is the mark of any superior defensive player. There is a reason that the prime goal of a defenseman on a 2 on 1 is to take away the pass.

What corisa stat touches on how well a player controls passing lanes?

It would seem that you have basically pointed out the flaw with this analysis.

If Hamilton and Gio were in fact league leading defenseman the Flames as a team would not be a the very best with all things going well and no injuries a bubble team.

When almost every counting and underlying stat goes a players way, he's probably a pretty good player.

You've disliked Hamilton for three years, so I get that you're coming at with a view and you are hoping to back that view up with some sort of data to make it relevant.

It doesn't really exist.

However I don't think the player is perfect, he lacks in the physical side, and he may have a blind spot to passing lanes as you suggest. I do think you're probably right about his abilities and how they translate to penalty killing, but five on five the guy gets it done to an elite level and its pretty hard to refute.