The Advanced Stats Thread Episode IX

[silverfish]

I wasn't saying they don't watch--to the contrary, they do watch. I have no doubt. I just get the sense that there are quite a few people who watch the games, come out of it without a strong opinion on the performance of a given player, and then develop that opinion after the fact only once they consult the numbers. It's like, "I don't really know what I saw regarding Player X, seemed pretty uneventful. Numbers, tell me what I saw!"

If people are livestreaming their disdain for Neal Pionk or Jimmy Vesey or whoever throughout the game and then support that with some statistics afterwards, great. No complaints for me. But I know quite a few people who actively participate in the GDT, say nothing about a particular guy, then start trashing him with everything they can find from NaturalStatTrick as soon as the game ends.

That's all. I'm not a "watch the games, nerd" guy, you know?

Ah, gotcha.

I know everyone is watching.

Not everyone

 

[nyr__1994]

I wasn't saying they don't watch--to the contrary, they do watch. I have no doubt. I just get the sense that there are quite a few people who watch the games, come out of it without a strong opinion on the performance of a given player, and then develop that opinion after the fact only once they consult the numbers. It's like, "I don't really know what I saw regarding Player X, seemed pretty uneventful. Numbers, tell me what I saw!"

If people are livestreaming their disdain for Neal Pionk or Jimmy Vesey or whoever throughout the game and then support that with some statistics afterwards, great. No complaints for me. But I know quite a few people who actively participate in the GDT, say nothing about a particular guy, then start trashing him with everything they can find from NaturalStatTrick as soon as the game ends.

That's all. I'm not a "watch the games, nerd" guy, you know? I know everyone is watching.

This - I bet if you dug hard enough you could disect a GDT where someone was speaking highly of how player A played because they had a goal or a nice play. Then goes and looks at the numbers and realizes that they got their teeth kicked in numbers wise and goes on to bash them.

Like everything the truth is in the middle somewhere....

 

[silverfish]

 

[silverfish]

New aesthetic on the TOI vis requested by @SA16 , minute labels.

 

[Machinehead]

Lettieri is truly awful. Spooner is worse. McLeod is McLeod. I guess it doesn't matter right now but this team desperately needs to address their 4th line going forward.

@Mac n Gs wasn't it you who said Namestnikov was the only thing making our 4th line not an AHL line? Well Namestnikov moved up and here it is!

Also, can we talk about Shattenkirk's +8.04 corsi rel and +10.38 xGF rel? I've officially retired from defending him in any thread but this one. It's just not worth it.

 

[Oscar Lindberg]

Agreed on Lettieri. I think if he was left handed, he would not be on the team right now.

Also Shattenkirk is bad, haven't you heard? @Machinehead

 

[Machinehead]

I mean, Shattenkirk should have more points. That's absolutely on him.

Granted, we are playing Neal Stionk on the 1PP over him for some reason, where Stionk has been an absolute train wreck. Unfortunately, that Bobby Orr goal he scored (WHICH WAS AT EVEN f***ING STRENGTH) is going to buy him at least three months of being better offensively than Shattenkirk even if he doesn't score another point. This is why I hate the eye test.

Speaking of, no, to be completely honest, Shattenkirk doesn't really pass the eye test for me. But when you drive play like that, I reach a point where I literally don't care. Sorry.

 

[Mac n Gs]

Lettieri is truly awful. Spooner is worse. McLeod is McLeod. I guess it doesn't matter right now but this team desperately needs to address their 4th line going forward.

@Mac n Gs wasn't it you who said Namestnikov was the only thing making our 4th line not an AHL line? Well Namestnikov moved up and here it is!

Also, can we talk about Shattenkirk's +8.04 corsi rel and +10.38 xGF rel? I've officially retired from defending him in any thread but this one. It's just not worth it.

I didn’t use that phrasing exactly, but it was clear he was the one making plays all over the ice.

As to Shatty, it’s a moot point because people will only look at his point totals and plus minus

 

[Machinehead]

*Buch takes a penalty late in the game*

Everyone: "Buchnevich, ya stupid!!"

 

[Blue Blooded]

Lettieri is truly awful. Spooner is worse. McLeod is McLeod. I guess it doesn't matter right now but this team desperately needs to address their 4th line going forward.

@Mac n Gs wasn't it you who said Namestnikov was the only thing making our 4th line not an AHL line? Well Namestnikov moved up and here it is!

Also, can we talk about Shattenkirk's +8.04 corsi rel and +10.38 xGF rel? I've officially retired from defending him in any thread but this one. It's just not worth it.

I said that! (@Mac n Gs may also have said it at some point)

Shatty is playing great, though Smith is somehow putting up a godawful RelxGF% despite being his most common partner. Claesson-Shattenkirk have been sublime as a pairing.

 

[Machinehead]

I said that! (@Mac n Gs may also have said it at some point)

Shatty is playing great, though Smith is somehow putting up a godawful RelxGF% despite being his most common partner. Claesson-Shattenkirk has been sublime.

I remember now!

 

[Machinehead]

One of the biggest hurdles facing the acceptance of analytics: armchair statisticians who think saying the word "causation" makes everything they disagree with just go away.

Folks need to pick up a textbook (I have one they can borrow) and read about correlation vs causation. Learn how it actually works. It's valuable material. What it definitely isn't, is a magic bean that makes the bad man telling you you're wrong go away.

 

[sbjnyc]

One of the biggest hurdles facing the acceptance of analytics: armchair statisticians who think saying the word "causation" makes everything they disagree with just go away.

Folks need to pick up a textbook (I have one they can borrow) and read about correlation vs causation. Learn how it actually works. It's valuable material. What it definitely isn't, is a magic bean that makes the bad man telling you you're wrong go away.

If correlation and causation were the same, weather forecasts would be causing the weather. I'm not sure what you're suggesting here. xG (or whatever) models clearly have no causation in and of themselves, but neither do actual goals (or wins or whatever) so the idea (I assume) is to determine, where possible, what factors have a causal relationship with variables of interest that can also explain current goals (or wins or whatever) and capture them in your model. So if a model has a higher correlation to future goals, wins or whatever than actual goals, then the question that remains is whether the factors in the xG model have a causal relationship. This isn't a given. If a causal relationship can't be shown then what does the model actually tell you?

 

[Machinehead]

If correlation and causation were the same, weather forecasts would be causing the weather. I'm not sure what you're suggesting here. xG (or whatever) models clearly have no causation in and of themselves, but neither do actual goals (or wins or whatever) so the idea (I assume) is to determine, where possible, what factors have a causal relationship with variables of interest that can also explain current goals (or wins or whatever) and capture them in your model. So if a model has a higher correlation to future goals, wins or whatever than actual goals, then the question that remains is whether the factors in the xG model have a causal relationship. This isn't a given. If a causal relationship can't be shown then what does the model actually tell you?

Who said correlation and causation are the same?

The point I'm making is, when one hockey event is correlating with another hockey event, it's reasonable (sure, not a given) that we have causation.

Let's say a player has played 10 years in the NHL. For 10 years, his team consistently performs poorly in shot metrics when he's on the ice. You can't just dismiss that as correlation. It's very reasonable that he's causing that.

 

[Filthy Dangles]

slick breakout by Freddie.

 

[sbjnyc]

Who said correlation and causation are the same?

The point I'm making is, when one hockey event is correlating with another hockey event, it's reasonable (sure, not a given) that we have causation.

Let's say a player has played 10 years in the NHL. For 10 years, his team consistently performs poorly in shot metrics when he's on the ice. You can't just dismiss that as correlation. It's very reasonable that he's causing that.

You are saying that correlation is not the same as causation and then in the next sentence say that if 2 events are correlated then there is causation. So no, what you are saying is not reasonable at all, IMO. I mean, it might be true but you can't just assume it.

 

[Machinehead]

You are saying that correlation is not the same as causation and then in the next sentence say that if 2 events are correlated then there is causation. So no, what you are saying is not reasonable at all, IMO. I mean, it might be true but you can't just assume it.

I'm not saying there is causation. I'm saying there could be.

People too readily assume that there isn't because "correlation does not equal causation" has become a meme.

There's causation probably more often than there's not.

 

[sbjnyc]

I'm not saying there is causation. I'm saying there could be.

People too readily assume that there isn't because "correlation does not equal causation" has become a meme.

There's causation probably more often than there's not.

Maybe I didn't get your original comment. Of course there can be but it's not something you can assume either.

 

[Machinehead]

Maybe I didn't get your original comment. Of course there can be but it's not something you can assume either.

My original comment was that people who know nothing about statistics say "you're wrong because causation" because causation has become a meme.

There's an assumption among advanced stats opponents that we can take any two things that correlate and just say there's no causation even when causation is very reasonable.

 

[DanielBrassard]

slick breakout by Freddie.

That was a great play. Makes all the difference in the world, most defenseman would go right off the glass in the situation.

 

[sbjnyc]

My original comment was that people who know nothing about statistics say "you're wrong because causation" because causation has become a meme.

There's an assumption among advanced stats opponents that we can take any two things that correlate and just say there's no causation even when causation is very reasonable.

I mean, it is a logical fallacy. I understand what you're saying (I think), but there's really no way to show it. What you seem to be saying is something akin to "I can't define it but I know it when I see it".

 

[Machinehead]

I mean, it is a logical fallacy. I understand what you're saying (I think), but there's really no way to show it. What you seem to be saying is something akin to "I can't define it but I know it when I see it".

What I'm saying is everyone knows the rule that we can't assume causation, and now people are doing the opposite and just assuming there isn't.

 

[silverfish]

 

[Machinehead]

Kreider was drunk tonight.

Also, Pionk...man.

Absolutely horrendous CF% player, hits the ice at the first sign of trouble, getting the hardest minutes despite the fact that he's getting face-washed, doing little things to make the fans, coaches, and FO forget about it.

Who the f*** does that sound like?

I'm trying real hard to like him because he can do highlight reel shit, but I do not need another Dan Girardi in my life. I don't need it.

I'm getting it, aren't I?

 

[gorangers0525]

I didn’t use that phrasing exactly, but it was clear he was the one making plays all over the ice.

As to Shatty, it’s a moot point because people will only look at his point totals and plus minus

Don’t even bring up +-. If you went by it, Shatty would be the 2nd best D on the team, and Stionk would be the worst player on the team by far.

Even meaningless stats aren’t in Pionk’s favor. Literally all Pionk has going for him is a few flashy offensive plays, he is egregiously bad defensively. I haven’t even followed metrics closely this year, he just looks absolutely putrid out there when not showing some offensive flash.