Player Underlying Stats Summary

[Rick M.]

Originally Posted by CaptainCrunch

I love Dube and what he does, he seems a bit snake bitten. Would it be smart to send him to the AHL for a bit to get his scoring confidence up and bring up someone like a Mangiapane for a look see?

I think Dube will be a good NHLer but his lack of physical strength is a problem for him currently. It is too easy for opposing players to staple him to the boards and keep him there until the play moves on. One more off-season of strength building should do wonders.

[DeluxeMoustache]

Originally Posted by TheScorpion

One thing I get out of this -- Ryan is the Flames' most exposed forward, yet he's second only to Czarnik in terms of allowing the least amount of chances.

That's good stuff.

That is good. The coach trusts him, he is dependable.

Imagine how much more energy he would have to apply to perfecting his checking game if he wasn’t deployed on the PP!

[edslunch]

Originally Posted by Bingo

Team Chart:
Flames 5on5 / 60 minutes VS the 16th best team (bubble) in each category.

Really points to where the team needs to improve, and the fact that they've taken steps in the right direction.

I don’t know how to interpret this chart

[Kovaz]

Originally Posted by Bingo

Team Chart:
Flames 5on5 / 60 minutes VS the 16th best team (bubble) in each category.

Really points to where the team needs to improve, and the fact that they've taken steps in the right direction.

[chart]

I'm totally on your side as far as the importance of data and analytics in hockey, but I've gotta be honest: that's a bad chart. If you're trying to communicate something through data visualization it's your job to make it easy and intuitive to interpret.

Specifically:

1. Lines shouldn't be used to connect discrete, independent data points; they should be used to connect samples of a continuous value. The use of a line implies that you can interpolate between values on the line. So if a line chart measuring temperature has a data point at [2PM, 12 degrees] and at [6PM, 6 degrees], you can interpolate and say at 4PM it was probably ~9 degrees. In this case there's no intermediate value between [Game 2, 8 Scoring Chances/60 above median] and [Game 3, -8 Scoring Changes / 60 above median]
2. Different measurements sharing the same axis. The bars being right next to one another implies they should be compared, but they're measuring different things. Visually, seeing the orange bar above the silver bar feels like it should mean something, but it's actually meaningless because Corsi events and scoring chances are different things. If you want to compare them on the same chart, some sort of normalization should be applied first (Maybe plotting the z-score of that measurement).

In general, I've found it much more effective to use Small Multiples when you've got a bunch of related, but different data to display.

[bubbsy]

Originally Posted by Erick Estrada

Can someone remind me why Hathaway has been in the lineup in the first place? He was a nice story a few years ago when inserted in the lineup a bit of a good luck charm but over the course of last season and into this one he's the very definition of replacement player. He should really only get in the lineup in the case of injury replacement or when one of the other forwards is downright terrible and deserves a night in the pressbox. He should not be a regular player for any NHL team.

For me, this is where the stats and eye test just don't match. I've been pleasantly happy with Hathaway this year. Finishes checks, gets into scrums. Does he generate offense, no, and the line he's playing gets stuck in their own zone a bit, but id wager the same would be true of his replacement. He's the 12th forward and not really a detriment.

It's nice to think about having 4 skilled scoring lines, heck i dreamed about that all summer but the truth of the matter is outside of our top line and tkatchuk, nobody can score on this team. Given that, having some grit is fine by me.

Unrelated question : what does "differential" mean in the stats shared?

[GioforPM]

Originally Posted by CaptainCrunch

I love Dube and what he does, he seems a bit snake bitten. Would it be smart to send him to the AHL for a bit to get his scoring confidence up and bring up someone like a Mangiapane for a look see?

Listening between the lines to an interview with the Stockton coach it doesn't sound like Mangiapane has earned a trip north. He damned him and his linemates with faint praise like "starting to figure it out/settle in" or something like that. He seemed to like Lazar a whole lot more.

If there were going to send Dube down they'd have done it before his contract kicked. And I don't think he has a confidence issue, he has a bad luck/close call issue. He's been right there for goals. If he does have a confidence issue I don't think a demotion will help.

[Bingo]

Originally Posted by Textcritic

I agree. I have never been impressed with Hathaway, but also thought he looked good in his first game or two this season. But he really should not be in any more than a handful of games unless filling in for an injured player. My hope is that Czarnik gets back into the lineup, and Hathaway takes a seat for a while.

The one thing he does very well, and more consistently this year is be an ass.

There is value in that.

But he's pulling away as the worst player on the ice every night when it comes to what happens when he's out there.

Only three players are sub 50% CF, and he's at 44%

[Bingo]

Originally Posted by edslunch

I don’t know how to interpret this chart

You want everything above zero ... the thick grey line.

Every column above zero means the team is doing better offensively than the 16th place team in that category.

Every line point above zero means the team is doing better defensively than the 16th place team in that category.

[Bingo]

Originally Posted by Kovaz

I'm totally on your side as far as the importance of data and analytics in hockey, but I've gotta be honest: that's a bad chart. If you're trying to communicate something through data visualization it's your job to make it easy and intuitive to interpret.

Specifically:

1. Lines shouldn't be used to connect discrete, independent data points; they should be used to connect samples of a continuous value. The use of a line implies that you can interpolate between values on the line. So if a line chart measuring temperature has a data point at [2PM, 12 degrees] and at [6PM, 6 degrees], you can interpolate and say at 4PM it was probably ~9 degrees. In this case there's no intermediate value between [Game 2, 8 Scoring Chances/60 above median] and [Game 3, -8 Scoring Changes / 60 above median]
2. Different measurements sharing the same axis. The bars being right next to one another implies they should be compared, but they're measuring different things. Visually, seeing the orange bar above the silver bar feels like it should mean something, but it's actually meaningless because Corsi events and scoring chances are different things. If you want to compare them on the same chart, some sort of normalization should be applied first (Maybe plotting the z-score of that measurement).

In general, I've found it much more effective to use Small Multiples when you've got a bunch of related, but different data to display.

Some good feedback, thanks

[Bingo]

Originally Posted by GioforPM

Listening between the lines to an interview with the Stockton coach it doesn't sound like Mangiapane has earned a trip north. He damned him and his linemates with faint praise like "starting to figure it out/settle in" or something like that. He seemed to like Lazar a whole lot more.

If there were going to send Dube down they'd have done it before his contract kicked. And I don't think he has a confidence issue, he has a bad luck/close call issue. He's been right there for goals. If he does have a confidence issue I don't think a demotion will help.

Dube's contract wasn't slide possible ... he never had a ticking time date as I understand.

[GioforPM]

Originally Posted by Bingo

The one thing he does very well, and more consistently this year is be an ass.

There is value in that.

But he's pulling away as the worst player on the ice every night when it comes to what happens when he's out there.

Only three players are sub 50% CF, and he's at 44%

Hathaway is great in short doses. Peters needs to keep an eye on whether his energy starts to drop, his physical play lessens, etc.

[squiggs96]

Originally Posted by Bingo

I wish the words "advanced" and "fancy" were never added to something so bloody simple.

I don't think advanced stats are so bloody simple. I say this as a guy who has watched hockey for many years, has been a STH for over a decade, and has excelled in many mathematics courses. The reason they aren't straight forward is they get presented like this:

Originally Posted by Bingo

Team Chart:
Flames 5on5 / 60 minutes VS the 16th best team (bubble) in each category.

Really points to where the team needs to improve, and the fact that they've taken steps in the right direction.

When I see that chart I picture this:


Even if I understood what each item stood for, which I don't, it's really hard to read what's going on. From looking at that I have zero idea what the team needs to improve on. I don't know how anyone would interpret that chart without an in depth knowledge of advanced analytics. I don't even think it's comparing apples to apples. It might not even be apples to oranges. It could be apples to vehicles.



This, to me, is the major flaw of advanced stats. They are presented in a way that's no immediately grasped, as opposed to points, goals, save percentage, and the advanced stats people get irritated when they have to explain things over and over. If it was that simple, it would be simple.

[Makarov]

Based on these statistics, I would really like to see them consistently try Bennett-Backlund-Tkachuk as the second line.

[Bingo]

Originally Posted by squiggs96

I don't think advanced stats are so bloody simple. I say this as a guy who has watched hockey for many years, has been a STH for over a decade, and has excelled in many mathematics courses. The reason they aren't straight forward is they get presented like this:

Even if I understood what each item stood for, which I don't, it's really hard to read what's going on. From looking at that I have zero idea what the team needs to improve on. I don't know how anyone would interpret that chart without an in depth knowledge of advanced analytics. I don't even think it's comparing apples to apples. It might not even be apples to oranges. It could be apples to vehicles.



This, to me, is the major flaw of advanced stats. They are presented in a way that's no immediately grasped, as opposed to points, goals, save percentage, and the advanced stats people get irritated when they have to explain things over and over. If it was that simple, it would be simple.

Well you can certainly shoot the graph as a culprit for sure, to each their own.

But the stats themselves aren't that difficult.

Someone counts how many times the Flames attempt a shot for in a game when five on five, and how many times their opposition does.

Inside that are shades ... a scoring chance within a set section of the ice, and a high danger chance that occurs in that same area, but is generated by a list of factors like a pass or rebound or deflection that make it more dangerous.

There's no formula or hypothesis at work, just a count. Not that difficult.

If the graph made it confusing that's on me not the stat.

But as I said ... everything above zero is good. Below zero is bad.

The columns are mostly above = Flames good at offence.
The lines are mostly below = Flames bad at defense.

[GioforPM]

Originally Posted by Bingo

Dube's contract wasn't slide possible ... he never had a ticking time date as I understand.

Thanks, I heard the ten days didn't apply for being sent back (because he's done junior) but I thought I heard it still applied to the slide.

Either way, I'd keep him up. If a guy deserves to be on the team I think it does more harm than its worth to artificially extend his contract one more year anyway.

[DeluxeMoustache]

If I am reading it correctly, I think they need to work on games 3, 8, 9 and 11

[Bingo]

Originally Posted by DeluxeMoustache

If I am reading it correctly, I think they need to work on games 3, 8, 9 and 11

They were awful in what they gave up in games 3 (Preds W), 6 (Bruins W), and 8 (Rangers W)

Had leads in all of them and let their goaltender do too much work.

[DeluxeMoustache]

Makes sense.

In terms of these metrics, game 10 (9-1 Pitt) wasn’t so bad. Just another day at the rink. Quality wins out over quantity in that one. Calgary didn’t spend much time in the d zone because it felt like every time Pitt came in, they just scored and it went back out to center. Didn’t spend time racking up shots, just scored.

[edslunch]

Originally Posted by Bingo

Well you can certainly shoot the graph as a culprit for sure, to each their own.

But the stats themselves aren't that difficult.

Someone counts how many times the Flames attempt a shot for in a game when five on five, and how many times their opposition does.

Inside that are shades ... a scoring chance within a set section of the ice, and a high danger chance that occurs in that same area, but is generated by a list of factors like a pass or rebound or deflection that make it more dangerous.

There's no formula or hypothesis at work, just a count. Not that difficult.

If the graph made it confusing that's on me not the stat.

But as I said ... everything above zero is good. Below zero is bad.

The columns are mostly above = Flames good at offence.
The lines are mostly below = Flames bad at defense.

Honestly a good way to start demystifying these stats is call them what they are like ‘shot attempt %’ instead of ‘Corsi’ which is the name of a guy. Same with PDO:

“PDO is not actually an acronym for anything. It comes from the online handle of Brian King, the first to propose it, for forums and Counter-Strike.[8]”

[SeanCharles]

Originally Posted by GioforPM

Peters put Neal on the Backlund line halfway through the game and to my eye it looked pretty good. I'd put him there for a while to see what happens.

I wouldn't mind seeing the following deployed for awhile:

Gaudreau-Monahan-Lindholm
Tkachuk-Backlund-Neal
Bennett-Jankowski-Dube
Frolik-Ryan-Hathaway

I've been disappointed in Ryan and Jankowski as I don't think either are playing at a 3rd line center level.

I wonder how Czarnik would look up the middle?