Did a quick look at the NHL remaining schedule, in my opinion the Flyers better have control over their own destiny before the last handful of games. Quick look but seems they have a tougher go of it at the end.

All teams struggle, not 9-0 struggle but struggle. Boston was going to run away with it, now they have had a bad stretch. Their own bias blog said the other day “maybe the Bruins just are not that good after the first line”. Isles started slow, then fast and now they will have to deal with some injuries, like losing a Coots. Pens are MEH and now Malkin is week to week.

Why they play all the games. Whoever is playing their best hockey down the stretch and the healthiest, with a little bit of luck will go the furthest.

Get it all out
- wcorvette

After their big 3 and krejci, there next guys are debrusk and Ritchie, you’re not going too far with that

After their big 3 and krejci, there next guys are debrusk and Ritchie, you’re not going too far with that
- ClaudeFather

Ritichie is actually having a pretty solid year. Krejci needs to score some goals. He has 1 and Coyle and Debusk are who have to pick it up.

That's incorrect. Again the flaw in the stat is that it cannot differentiate between more routine shots that are considered a high danger chance and far more dangerous high danger chances. For example, if a wrist shot is taken from a high danger chance area where the goalie is stationary and out on his angle. That is a far less dangerous shot than a shot in a high danger area with an open net for the shooter to look at.
If a team defensively gives up the same rate of high danger chances as a previous season but the chances are far more dangerous, than that has a big effect on the goaltending HDSV%. How can we know if that is the case? We can watch the game! It's lunacy to blindly look at a stat, come to a flawed conclusion and ignore watching the game!

- MJL

Lets say we divide high danger chances into 3 categories:

Really high danger chances (RHDC)
Average high danger chances (AHDC)
Garden variety high danger chances (GHDC).

A team that commits more big defensive errors to give up more RHDC is also likely to commit more small defensive errors to give up more GHDC. The two will cancel out.

The problem with the eye test is this: we would notice the greater # of RHDC because they are big screw ups. We may not notice the greater # of GHDC because they are more subtle and more quick.

Also in the RHDC, it is far easier to notice that say Myers blew coverage, then say that Hart's glove was two inches below where it should be. So again, the eye test may fail us.

Ritichie is actually having a pretty solid year. Krejci needs to score some goals. He has 1 and Coyle and Debusk are who have to pick it up.
- MJL

He is, yea it’s the same thing every year with them, will they get the secondary scoring they need. With their losses on D and those guys struggling I really don’t see them getting far

Lets say we divide high danger chances into 3 categories:

Really high danger chances (RHDC)
Average high danger chances (AHDC)
Garden variety high danger chances (GHDC).

A team that commits more big defensive errors to give up more RHDC is also likely to commit more small defensive errors to give up more GHDC. The two will cancel out.

The problem with the eye test is this: we would notice the greater # of RHDC because they are big screw ups. We may not notice the greater # of GHDC because they are more subtle and more quick.

Also in the RHDC, it is far easier to notice that say Myers blew coverage, then say that Hart's glove was two inches below where it should be. So again, the eye test may fail us.
- PT21

They don't automatically cancel out. You're just making it up that they do to fit a premise. The eye test can be flawed but so can the analytics or I should say the analysis of the analytics. The data is the key. It's not about the science of it or the math. It's about the data.

They don't automatically cancel out. You're just making it up that they do to fit a premise. The eye test can be flawed but so can the analytics or I should say the analysis of the analytics. The data is the key. It's not about the science of it or the math. It's about the data.
- MJL

No no. Not making it up. Not at all.

Think of what it would mean if they did not cancel out, over a large enough sample. That would mean, by definition, that the typical high danger chance would be more dangerous than the middle layer I talked about (what I called the Average HDC or AHDC). Then the typical HDC would lie in the RHDC.

That would mean that when people talk about HDC, they are really usually talking about RHDC. That would be a systematic structural problem with the definition, which would mean there should be a revision of the concept.

I would doubt there is such a structural problem in such a commonly used concept.It would be like saying a fair coin, when tossed enough # of times, would tend to skew heads.

Now, of course, there might be variations from sample to sample, but I think you would need more than an "eye test" to prove it because of the biases of the eye.

No no. Not making it up. Not at all.

Think of what it would mean if they did not cancel out, over a large enough sample. That would mean, by definition, that the typical high danger chance would be more dangerous than the middle layer I talked about (what I called the Average HDC or AHDC). Then the typical HDC would lie in the RHDC.

That would mean that when people talk about HDC, they are really usually talking about RHDC. That would be a systematic structural problem with the definition, which would mean there should be a revision of the concept.

I would doubt there is such a structural problem in such a commonly used concept.It would be like saying a fair coin, when tossed enough # of times, would tend to skew heads.

Now, of course, there might be variations from sample to sample, but I think you would need more than an "eye test" to prove it because of the biases of the eye.
- PT21

Yes, you are making it up. You're also not understanding the data. Falsely comparing a coin flip where there are but two possibilities. Tossing a coin is random. High danger chances and the quality of high danger chances is not a random flip of the coin. All of the commonly used analytics in hockey are flawed. Again, not the science in a vacuum away from the variables of the game. The issue is the data. All analytics and metrics need analysis. That same bias is present in that as we've seen today. If it wasn't, then everyone would have the same conclusion when looking at any metric.

Yes, you are making it up. You're also not understanding the data. Falsely comparing a coin flip where there are but two possibilities. Tossing a coin is random. High danger chances and the quality of high danger chances is not a random flip of the coin. All of the commonly used analytics in hockey are flawed. Again, not the science in a vacuum away from the variables of the game. The issue is the data.
- MJL

Ok let me try again. Its not hard.

1. When one says 'HDC', what does the phrase typically conjure up for fans, across all teams, several seasons? Whatever that image is, that is the 'average' or median HDC. Lets fix that concept.

2. This means by construction, about an equal % of chances are better quality than the image, and an equal % are worse.

3. Note: This is not a premise! This is by definition, since the first line indicates our image is the median. If 2 failed, our image/idea of a HDC in 1 is not the median image. The of course we need to correct that, have in mind the right image, and then proceed from step 1 again.

4. Now, you claim that for the Flyers in 20-21, the fraction of the HDC that are higher than average quality is abnormally high. In other words, the Flyers median HDC against is higher quality than what is commonly understood in 1.

5. That could well be true. But in general, since there are a lot of HDC in each game, and we have had 25+ games in a season, the sample size is now quite large. When sample sizes are large like that, the sample average tends to get closer and closer to the population average.

6. Does that mean you cannot have a large sample size that has anomalous values? Sure you can. But you would have to prove that with a little more than saying, I know it because I see it. Because of the biases I described.

Ok let me try again. Its not hard.

1. When one says 'HDC', what does the phrase typically conjure up for fans, across all teams, several seasons? Whatever that image is, that is the 'average' or median HDC. Lets fix that concept.

2. This means by construction, about an equal % of chances are better quality than the image, and an equal % are worse.

3. Note: This is not a premise! This is by definition, since the first line indicates our image is the median. If 2 failed, our image/idea of a HDC in 1 is not the median image. The of course we need to correct that, have in mind the right image, and then proceed from step 1 again.

4. Now, you claim that for the Flyers in 20-21, the fraction of the HDC that are higher than average quality is abnormally high. In other words, the Flyers median HDC against is higher quality than what is commonly understood in 1.

5. That could well be true. But in general, since there are a lot of HDC in each game, and we have had 25+ games in a season, the sample size is now quite large. When sample sizes are large like that, the sample average tends to get closer and closer to the population average.

6. Does that mean you cannot have a large sample size that has anomalous values? Sure you can. But you would have to prove that with a little more than saying, I know it because I see it. Because of the biases I described.
- PT21

I stopped reading after the first few comments because I couldn't stop laughing. Again, you're inventing things again and not actually looking at what is actually going on, on the ice. Apparently, it is too hard.

I stopped reading after the first few comments because I couldn't stop laughing. Again, you're inventing things again and not actually looking at what is actually going on, on the ice. Apparently, it is too hard.
- MJL

Very well then. 😀

Very well then. 😀
- PT21

I spent some time reading your posts about these stats (back and forth) and I'm hoping you are both just doing this for giggles, cause if not you guys need some serious hobbies outside of watching hockey and posting on hockey buzz. Very funny tho'. As Maximum might say" ARE YOU NOT ENTERTAINED"?

I spent some time reading your posts about these stats (back and forth) and I'm hoping you are both just doing this for giggles, cause if not you guys need some serious hobbies outside of watching hockey and posting on hockey buzz. Very funny tho'. As Maximum might say" ARE YOU NOT ENTERTAINED"?
- DANCER

Both MJL and I are actually HockeyBuzz bots, Dancer. Bots don't have bodies, bots don't have hobbies.

Hope you got the Covid jab btw. At your age, you should be eligible wherever you are. If you haven't, stop being a dang fool and get it.

Nighty night.

The Rangers are (frank)in useless

The Rangers are (frank)in useless
- SuperSchennBros

Caps are going to make it. It would have been nice to see the caps lose but there is value in the Rangers losing also

Both MJL and I are actually HockeyBuzz bots, Dancer. Bots don't have bodies, bots don't have hobbies.

Hope you got the Covid jab btw. At your age, you should be eligible wherever you are. If you haven't, stop being a dang fool and get it.

Nighty night.
- PT21

There you go again Sheldon. Making crap up for attention. You must be lonely.

This is the 3rd post in a row that you made that completely missed the boat. When comparing the rate, now understand what rate is, from last year to this year and seeing that it is similar. Then drawing from that a conclusion that goaltending is the issue and not the defense in front of it, well that's not understanding the stat. You're arguing from the wrong point of view and are confusing rate with the difficulty of a quality chance and the variables of the difficulty. The stat cannot measure that. The weakness has always been there.
- MJL

But NO ONE is making this conclusion ... EXCEPT YOU!

I'll put it simply like this and if you disagree, then that just proves what I've already said. The stat you posted does not just point to goaltending. It points to overall team defense and goaltending.
- MJL

Yes I agree with this statement

It goes back to your trend of wanting to single out a player for a team wide issue. When it comes to analysis of the stat, the difference in quality and difficulty between even a subset of data points such as high danger chances is extremely relevant. Again, if you disagree then that again proves what I've already said.
- MJL

When you are talking about a subset of data for any metric, the variability in that data is much higher, and therefore less conclusive. However if you look at the data as a whole, i.e. your sample size gets large enough, the variability gets smoothed out and the metric has more meaning. So I am not sure what you are getting at here, are you saying that 28 games this year is not enough sample size versus 69 games last year and therefore you really can't compare this year to last year yet? If so, you have a point and I understand your argument. If however you mean something different, I am not sure what it is and thus cannot agree with you yet.

I stopped reading after the first few comments because I couldn't stop laughing. Again, you're inventing things again and not actually looking at what is actually going on, on the ice. Apparently, it is too hard.
- MJL

Its a shame you stopped reading the post, because he actually knows what he's talking about in this regard.

That's incorrect. Again the flaw in the stat is that it cannot differentiate between more routine shots that are considered a high danger chance and far more dangerous high danger chances. For example, if a wrist shot is taken from a high danger chance area where the goalie is stationary and out on his angle. That is a far less dangerous shot than a shot in a high danger area with an open net for the shooter to look at.
If a team defensively gives up the same rate of high danger chances as a previous season but the chances are far more dangerous, than that has a big effect on the goaltending HDSV%. How can we know if that is the case? We can watch the game! It's lunacy to blindly look at a stat, come to a flawed conclusion and ignore watching the game!
Did you feel the same way about Sanheim and Myers last season as you do currently? How about with Provorv? Should I refresh your memory of your recent comments about the Flyers defense? I'm not singling them out here. I'm using it as an example of how I think your conclusion is flawed. You can't eliminate the team's play in front of the goaltending from the equation when looking at goaltending numbers. To do so is lunacy!
- MJL

I don't think you understand how metrics and data collection work. Maybe this will help you:

Scoring Chances and High-Danger Scoring Chances
As originally defined by War On Ice, 'Scoring Chances' indicate shots attempts that are taken from areas of the ice where goals are more likely to be scored. Attempts made from the attacking team's neutral or defensive zones are excluded.

Inside the zone, a shot is assigned a value of 1,2 or 3, depending on where it was from. A rebound shot (defined as any attempt made within 3 seconds of another blocked, missed or saved attempt without a stoppage in play in between) adds a point to this value. A blocked shot decreases the value by 1.

'Scoring Chances' are any shot attempts with a final value of 2 or higher. 'High-Danger Scoring Chances' are any shot attempt with a final value of 3 or higher.

So imagine shots from the point are 1 point shots, shot from the center circles are 2 point shots and shots from the slot area are 3 point shots. I couldn't find an actual map anywhere to picture this, if anyone else has it please post it. Rebounds add another point value.

The point is that what you are talking about in terms of variance goes away as your sample size increases, in other words sometimes the goalie will be set for the shot and other times they will be diving to bloc an open net. But the number of each of these specific examples averages out over time and comes closer to the median. I believe 28 games (1/3 of a typical season) is enough of a sample size to learn something from the data, which again is all the data is for.

How do I feel about Sanheim and Myers? I feel they have regressed significantly from last year. Their confidence is shot, and as a result they are trying to do too much and its killing them. But I can say in general that this also applies to Provorov, Hart, Elliott, probably Ghost, and so on. The overall team has regressed in the way they are playing this year. It seems to me that the forwards want to play pond hockey instead of playing the system that AV wants them to, which is maybe a boring system for them, you know, clogging up the neutral zone and playing a patient game. To me the players are not bought in and that is why everything is unraveling.

Last game against the Islanders showed two things: 1. The Flyers can play a sound defensive game when they put their mind to it and 2. they are an extremely fragile team that can quickly unravel when something does not go their way, and in this case it was a bad goal given up by Hart to make the score 3-1.

Did a quick look at the NHL remaining schedule, in my opinion the Flyers better have control over their own destiny before the last handful of games. Quick look but seems they have a tougher go of it at the end.

All teams struggle, not 9-0 struggle but struggle. Boston was going to run away with it, now they have had a bad stretch. Their own bias blog said the other day “maybe the Bruins just are not that good after the first line”. Isles started slow, then fast and now they will have to deal with some injuries, like losing a Coots. Pens are MEH and now Malkin is week to week.

Why they play all the games. Whoever is playing their best hockey down the stretch and the healthiest, with a little bit of luck will go the furthest.

Get it all out
- wcorvette

Now its the Bruins' turn to go through a Covid out break, I just hope the league makes them play with their top 6 forwards out the way they forced the Flyers to!

GEEZ


The beginning of the season the Team wasn’t playing well but goaltending was, the Flyers were winning. Then the team play got better and goaltending sagged and it was a mixed bag. Then both goaltending and team play was bad and it was all losing.

One affects the other, at times one has carried the other. It would be nice to see both consistent.

I found it!



In this example from WAR on Ice, shots from the yellow zone indicate low danger chances, shots from the pink zone indicate medium danger chances, and shots from the blue zone indicate high danger chances. In calculating SCF, shots from the yellow zone are assigned a score of one, pink assigned a score of two, and blue assigned a score of three. Rush chances and rebounds are assigned an additional point. A blocked shot lose one point. Scoring chances are counted as any event that is above a score of two.

For example, only rebounds or rush chances from the yellow zone that hit the net would be counted as scoring chances. Any unblocked shots from the pink zone, a blocked rebound or rush chance from the pink zone, or any shot from the blue zone would also count as scoring chances.

But NO ONE is making this conclusion ... EXCEPT YOU!
- jd250

You made the statement that the stat points to the goaltending not playing as well as it did last year. My point is that the stat can't show that. it can only tell you what has happened on the ice. The stat just shows a lower HDCA SV%. Why that is takes analysis. My analysis is that the result is as much to do with the team defense in front of the goaltending as it is with the goaltending play. If not more.

Now its the Bruins' turn to go through a Covid out break, I just hope the league makes them play with their top 6 forwards out the way they forced the Flyers to!
- jd250

If it’s just those guys and they are indeed positive....yes the Bruins will have to return and get 5-6 different guys like the Flyers did. Don’t forget the Flyers had a couple of games cancelled before they were forced to play. This is a different year. You just ever know what’s around the corner.

I don't think you understand how metrics and data collection work. Maybe this will help you:

Scoring Chances and High-Danger Scoring Chances
As originally defined by War On Ice, 'Scoring Chances' indicate shots attempts that are taken from areas of the ice where goals are more likely to be scored. Attempts made from the attacking team's neutral or defensive zones are excluded.

Inside the zone, a shot is assigned a value of 1,2 or 3, depending on where it was from. A rebound shot (defined as any attempt made within 3 seconds of another blocked, missed or saved attempt without a stoppage in play in between) adds a point to this value. A blocked shot decreases the value by 1.

'Scoring Chances' are any shot attempts with a final value of 2 or higher. 'High-Danger Scoring Chances' are any shot attempt with a final value of 3 or higher.

So imagine shots from the point are 1 point shots, shot from the center circles are 2 point shots and shots from the slot area are 3 point shots. I couldn't find an actual map anywhere to picture this, if anyone else has it please post it. Rebounds add another point value.

The point is that what you are talking about in terms of variance goes away as your sample size increases, in other words sometimes the goalie will be set for the shot and other times they will be diving to bloc an open net. But the number of each of these specific examples averages out over time and comes closer to the median. I believe 28 games (1/3 of a typical season) is enough of a sample size to learn something from the data, which again is all the data is for.

How do I feel about Sanheim and Myers? I feel they have regressed significantly from last year. Their confidence is shot, and as a result they are trying to do too much and its killing them. But I can say in general that this also applies to Provorov, Hart, Elliott, probably Ghost, and so on. The overall team has regressed in the way they are playing this year. It seems to me that the forwards want to play pond hockey instead of playing the system that AV wants them to, which is maybe a boring system for them, you know, clogging up the neutral zone and playing a patient game. To me the players are not bought in and that is why everything is unraveling.

Last game against the Islanders showed two things: 1. The Flyers can play a sound defensive game when they put their mind to it and 2. they are an extremely fragile team that can quickly unravel when something does not go their way, and in this case it was a bad goal given up by Hart to make the score 3-1.
- jd250

Actually it is you who doesn't understand which is illustrated by your comments. The variance does not go away as the sample size increases. That is a fallacy. In fact with a team like the Flyers and how they have been as a team defensively, the higher quality danger chances keep piling up. So while the rate of high danger chances may be similar to last year, the quality is not. You proved my point with your comments on Sanhiem and Myers. It also shows that you don't understand what AV's system is. It is not a boring clog up the neutral zone system.