What Corsi really translates to (in numbers that are easy to grasp)

[RageQuit77]

Maybe it's only me, but I get absolutely too often in these discussions about stats an impression that how people generally handle with their statistical knowledge suggest that there would be some kind mathematical conclusiveness in all of these numbers directly translatable to on-ice performances implying that there would be some kind "optimal strategy" for every circumstances and situations in a rink.

Even then all relevant information would be available (that certainly isn't the case now, and I seriously doubt never will) for statistical analysis, individual games differ from each others that many ways that what can be reasonably concluded from one, cannot be easily applied to other. The process of Statistical analysis itself is the worst reason of lost information (that could be useful for the statistics if counted/measured) as the statistical system (scientific paradigm) itself direct people's perception on the game, limiting what we consider worth for stats, but also what we can even think on about this worth. Stats, if they in best scenarios reveal a lot of more or less hidden aspects of the game's dynamics, they can also act like a filter for what we look at. In theory relatively simple (in sense what can be numerated/formulated) things, patterns and phenoms could be left unnoticed due 'wrong' scope of statistical analysis, possibly even misinterpreted in wrong way in their meaning. If nobody ever ask a good questions, neither statistics can develop further. Bad application doesn't help spotting good questions.

It's not a set of static and rigid dogmas - it's very complex living process. For both hockey, and - ideally - for hockey statistics.

 

[Elvs]

Never said any of this. These threads always devolve into madness because they're one assumption after another.

Maybe I'm mistaken (sorry if that's the case), but I thought your view on Fowler is that he hurts his team more than he helps?

Now, if that's what you are saying, is it your belief that Fowler having a +20 goal differential during the last four seasons (including playoffs) is an anamoly, and that the luck will change any day now?

 

[Machinehead]

Maybe I'm mistaken (sorry if that's the case), but I thought your view on Fowler is that he hurts his team more than he helps?

Now, if that's what you are saying, is it your belief that Fowler having a +20 goal differential during the last four seasons (including playoffs) is an anamoly, and that the luck will change any day now?

Dan Girardi is great at +/- also and I think he's one of the worst players I've ever seen play.

I hate +/- for defensemen because it's primarily forwards who score goals and goalies who prevent goals so it's really a stat that leaves defenseman behind.

Also, the Ducks have been good and +20 isn't really good on that team. 17 Ducks have a better rating over that period. You know, just FWIW. I think goal differential for defensemen is useless anyway for the reason I gave.

 

[struckbyaparkedcar]

Maybe I'm mistaken (sorry if that's the case), but I thought your view on Fowler is that he hurts his team more than he helps?

Now, if that's what you are saying, is it your belief that Fowler having a +20 goal differential during the last four seasons (including playoffs) is an anamoly, and that the luck will change any day now?

Cam Fowler is a plus whatever because he plays on the Ducks. He's actually a negative player relative to his team in terms of goal differential.

 

[Creativero]

Fantastic post.
A lot of the stats blogosphere has great application of math, poor application of the scientific method and critical thinking. Many of the "conclusively proven" x has no impacts are based on a couple of horribly flawed circular conclusions.

I'm not going to rehash the same argument. I'm just going to say I agree with what you say about variables and application, but the evidence against Fowler is conclusive.

This is exactly the sort of statement that makes people question if you're just blindly accepting whatever the "advanced stats community" tells you. He's making $4 a year on an RFA contract so I guess the Ducks didn't think it was conclusive.

 

[Machinehead]

This is exactly the sort of statement that makes people question if you're just blindly accepting whatever the "advanced stats community" tells you.

Another assumption.

You think I'm just blindly accepting a notion that I haven't studied? I was the first one on this site to point Fowler out, so there wasn't much for me to "accept." I started this particular trend.

I don't mean to be like Walter White "I am the one who knocks" but in this case I was.

 

[RageQuit77]

Funny thing is that seldom people realize that statistics can "make" a player worse then he is. It's in-built inevitability of stats. One has to be worse then other. One player is better that other in some other statistical category. No way can be situation everybody is levelled to the same...

Poor Fowler. No matter how loved he is within his team, no matter how he tries best he can, no matter what is his impact in the locker room to his fellow teammates... etc. all this is rendered meaningless because some stats says so... "conclusively". Lol.

Roughly half of the league perform worse then average. Kick them away!

 

[Sojourn]

Cam Fowler is a plus whatever because he plays on the Ducks. He's actually a negative player relative to his team in terms of goal differential.

That seems to suggest he benefits from playing on a good team, while ignoring that he contributes to them being a good team. Fowler isn't just some passenger. He's one of Anaheim's most important and heavily used defenseman.

 

[Creativero]

Another assumption.

You think I'm just blindly accepting a notion that I haven't studied? I was the first one on this site to point Fowler out, so there wasn't much for me to "accept." I started this particular trend.

I don't mean to be like Walter White "I am the one who knocks" but in this case I was.

The guy makes 4 million a year on an RFA contract and played for the national team in the Olympics. I just don't see any independent confirming evidence that doesn't require me to circle back to just accepting Corsi when it doesn't seem to mesh with the eye test or scouting report. I mean maybe the guy just doesn't fit well with the Ducks?

 

[Givememoneyback]

That seems to suggest he benefits from playing on a good team, while ignoring that he contributes to them being a good team. Fowler isn't just some passenger. He's one of Anaheim's most important and heavily used defenseman.

Apparently, the Ducks might be better offer relying on Clayton Stoner.

 

[Machinehead]

The guy makes 4 million a year on an RFA contract and played for the national team in the Olympics. I just don't see any independent confirming evidence that doesn't require me to circle back to just accepting Corsi when it doesn't seem to mesh with the eye test or scouting report. I mean maybe the guy just doesn't fit well with the Ducks?

I don't put much stock in coach and GM sections. Jack Johnson was on that team over Keith Yandle. Yandle is better than three Johnsons.

There's plenty of examples of executives in hockey making catastrophic mistakes. Not saying Fowler was one, but it happens. These guys are human. At best their opinions aren't worth any more than numbers.

 

[Sojourn]

Apparently, the Ducks might be better offer relying on Clayton Stoner.

I'm sure other teams would love that.

 

[BurgoShark]

Not contributing to the argument, but thanks to the OP for that explanation. I have only a vague understanding of fancy stats so it was educational for myself to have those numbers explained (put in to perspective).

 

[struckbyaparkedcar]

******** on Cam Fowler aside, there is a legitimate issue with hockey stats vs other sports where we don't have the same ability to meaningfully quantify the strategic application of a player vs their skillset.

 

[SladeWilson23]

Funny thing is that seldom people realize that statistics can "make" a player worse then he is. It's in-built inevitability of stats. One has to be worse then other. One player is better that other in some other statistical category. No way can be situation everybody is levelled to the same...

Poor Fowler. No matter how loved he is within his team, no matter how he tries best he can, no matter what is his impact in the locker room to his fellow teammates... etc. all this is rendered meaningless because some stats says so... "conclusively". Lol.

Roughly half of the league perform worse then average. Kick them away!

Agreed. Adam Larsson has a -2 CFRel. He's a beast of a defenseman.

 

[Elvs]

Girardi makes blantant mistakes that leads directly to goals against his team. One mistake like that can elimate you from the playoffs, when it's just best out of 7.

Also, the Ducks have been good and +20 isn't really good on that team. 17 Ducks have a better rating over that period. You know, just FWIW. I think goal differential for defensemen is useless anyway for the reason I gave.

So, Fowler is hurting Anaheim, because a better defenseman could be more than +20?

If I hide $10 in my left pocket, and I hide $15 in my right pocket... Then I let you pick, and you point to the $10 pocket... Do you now consider yourself poorer?

 

[RageQuit77]

So, Fowler is hurting Anaheim, because a better defenseman could be more than +20?

Hard to know because it is not that easy to gather adequately comparable sample sets with other players.

That is case with Gretzky too. We will never know how his dynasty era would've gone in other team, with other buddies. He certainly hurt the league and most players in it.

Same can be said on many other guys, also for those who were hurt by other teams regardless of their core importance and impact to their own team's play.

Corsi can bluff a lot if one let it mesmeraze too deeply. In a league where defenders won't get really hurted by Corsi, would be very dull to watch. Score board would be as useless as it would be blank white.

 

[Ol' Jase]

If we're talking about zone starts, and you argue with me using something that isn't a zone start, I'm going to dismiss it. Sorry.

What he said might be reasonable and we can debate his theory, but his theory has absolutely nothing to do with what we were talking about.

I'm talking about zone starts. He's talking about something he literally made up.

I didn't make up anything. YOU literally spelled out what a zone start is and what it isn't.

You then claimed that it was a "case closed" that zone starts have no affect on Corsi

Then, in an incrediblely unbelievable mathematic folly, stated 60% of recorded Corsi situations don't count towards Zone Starts.

Most mind boggling is that you seem to want to represent the advanced stat community and can't see how statistically ridiculous this is.

60%. 60 ****ing percent. Your math, not mine.

 

[Creativero]

I don't put much stock in coach and GM sections. Jack Johnson was on that team over Keith Yandle. Yandle is better than three Johnsons.

There's plenty of examples of executives in hockey making catastrophic mistakes. Not saying Fowler was one, but it happens. These guys are human. At best their opinions aren't worth any more than numbers.

The bar for saying something is not conclusive isn't the same as saying something is conclusive. If professional hockey GMs who, I would think spend a lot of time thinking about these things don't seem to be agreeing with a theory, that's enough for me to say the evidence is at minimum non-conclusive.

I do have to give you some credit for consistency.

 

[Machinehead]

So, Fowler is hurting Anaheim, because a better defenseman could be more than +20?

If I hide $10 in my left pocket, and I hide $15 in my right pocket... Then I let you pick, and you point to the $10 pocket... Do you now consider yourself poorer?

Poorer than I would have been if I picked $15. But we're rooting for the Ducks in this scenario which means there's money to spare anyway.

Then there's being a Rangers fan which is like living in a dumpster until the dumpster gets repossessed. That makes the selection a little more important. That $5 is another meal now.

It all depends on team.

 

[Machinehead]

I didn't make up anything. YOU literally spelled out what a zone start is and what it isn't.

You then claimed that it was a "case closed" that zone starts have no affect on Corsi

Then, in an incrediblely unbelievable mathematic folly, stated 60% of recorded Corsi situations don't count towards Zone Starts.

Most mind boggling is that you seem to want to represent the advanced stat community and can't see how statistically ridiculous this is.

60%. 60 ****ing percent. Your math, not mine.

Yes. Zone starts don't affect corsi precisely BECAUSE 60% of the readings have nothing to do with zone starts.

If I wanted to find out how many goals Chris Krieder has scored against the Atlantic Division, I wouldn't count games against the Islanders.

 

[Ol' Jase]

Yes. Zone starts don't affect corsi precisely BECAUSE 60% of the readings have nothing to do with zone starts.

If I wanted to find out how many goals Chris Krieder has scored against the Atlantic Division, I wouldn't count games against the Islanders.

Because you are segregating the goals Kreider scores. You are NOT segregating overall Corsi compared to Zone Starts.

Holy ****, I can't believe this conversation. You are applying 60% of one variable when another variable isn't applied and calling it proof.

Just amazing.

 

[Machinehead]

Because you are segregating the goals Kreider scores. You are NOT segregating overall Corsi compared to Zone Starts.

Holy ****, I can't believe this conversation. You are applying 60% of one variable when another variable isn't applied and calling it proof.

Just amazing.

You just don't get it.

60% of the time there is no zone start. NO ZONE START. Again, the average player only has a zone start in 40% of his shifts, i.e. after a faceoff. That's what a zone start is. That's what a zone start is according to the National Hockey League.

Whatever kinda **** you wanna come up with about where the puck is when you jump over the boards, is your project. There's no data on it and nobody talked about it until you did. You invented it. It's your idea. Congrats.

I'm fine with that. Track it and show us the data, I'm very interested. But it's not zone starts.

I'm talking about zone starts, which don't apply to 60% of a player's statistical results. That's a fact. I'm not ignoring anything. I'm not applying anything. 60% of shifts do not begin with a zone start. If you think they do, then you don't understand what a zone start is.

Zone start. Zone. Start. It's two words. It's a literal translation: STARTING a shift in a ZONE. When you jump over the boards, you're joining a shift that has already started. It is not your zone start.

One more time, there is is no zone start for 60% of a player's shifts. Zone starts are not a thing for 60% of shifts. Not every shift has a zone start, but every shift counts towards corsi.

If you disagree with that, then take it up with the NHL because that's what a zone start is.

 

[Ol' Jase]

You just don't get it.

60% of the time there is no zone start. NO ZONE START. Again, the average player only has a zone start in 40% of his shifts, i.e. after a faceoff. That's what a zone start is. That's what a zone start is according to the National Hockey League.

Whatever kinda **** you wanna come up with about where the puck is when you jump over the boards, is your project. There's no data on it and nobody talked about it until you did. You invented it. It's your idea. Congrats.

I'm fine with that. Track it and show us the data, I'm very interested. But it's not zone starts.

I'm talking about zone starts, which don't apply to 60% of a player's statistical results. That's a fact. I'm not ignoring anything. I'm not applying anything. 60% of shifts do not begin with a zone start. If you think they do, then you don't understand what a zone start is.

Zone start. Zone. Start. It's two words. It's a literal translation: STARTING a shift in a ZONE. When you jump over the boards, you're joining a shift that has already started. It is not your zone start.

One more time, there is is no zone start for 60% of a player's shifts. Zone starts are not a thing for 60% of shifts. Not every shift has a zone start, but every shift counts towards corsi.

If you disagree with that, then take it up with the NHL because that's what a zone start is.

You are applying a situation where one variable is only applied 40 percent of the time to another variable 100 percent of the time and citing that the 40 percent variable has no affect on the 100 percent variable and calling it proof. More ridiculously, you're calling it accurate.

It's absolute basic mathematics. It doesn't work, no matter how you want to present it, which is beyond ludicrous.

On top of that, I specifically asked you if you had segregated the scenarios and you said you did. So now I guess you made that up?

 

[Machinehead]

You are applying a situation where one variable is only applied 40 percent of the time to another variable 100 percent of the time

Because that's how hockey works.

40% of all shifts have a zone start.

100% of shifts have corsi.