Player Discussion Caps D over 6 last seasons

[francaisvolantsparis]

I continued my research up to the 12 last seasons. Updated my formula. Now it is called UARPDPEDIBOCPP* 'Awesomeness 2.0'.

Here are the results**:

Player	UARPDPEDIBOCPP* 'Awesomeness 2.0'	Grade
Dmitry Orlov	61	Elite
John Carlson	60	Elite
Mike Green	59	Elite
Matt Niskanen	49	Good
Brian Pothier	47	Good
Nate Schmidt	45	Good
Roman Hamrlik	44	Good
Dennis Wideman	42	Good
Tom Poti	42	Good
Michal Kempny	39	OK
Karl Alzner	38	OK
Tyler Sloan	35	OK
Brooks Orpik	34	OK
Madison Bowey	31	OK
Milan Jurcina	28	AHL
Taylor Chorney	28	AHL
John Erskine	27	AHL
Jeff Schultz	26	AHL
Scott Hannan	26	AHL
Shaone Morrisonn	25	AHL

[TBODY] [/TBODY]

(*) UARPDPEDIBOCPP - Ultimate Awesomeness Rating Points Developed for Passionate Eyewatchers with Durable Irrelevant Biased Opinions on Caps Player Performances
(**) A player need to play >30 games per season to score the UARPDPEDIBOCPP points. Only Caps performances are counted. If a player quits the team 'Awesomeness 2.0' is no longuer updated.

 

[txpd]

Matt Niskanen is only 2 points better than Brian Pothier???? Do you have my idea how stupid that is?

 

[francaisvolantsparis]

Matt Niskanen is only 2 points better than Brian Pothier???? Do you have my idea how stupid that is?

It is not stupid, it is Science

 

[Hivemind]

It is not stupid, it is Science

Junk science.

Show a measured correlation between your Awesomeness stat and winning hockey games, and then I'll pay attention to it.

 

[PB12]

This thread was used to pimp Orlov. We all know JC is the teams most valuable defensemen. God help Samsonov as his GAA goes up because Orlov is too much of a chicken **** to hit someone or block a shot himself. If he uses that wimpy *** pokecheck again, i'm going to break all his sticks in rage. Maybe we can tie him to net like Goldberg in the Mighty Ducks and have Ovechkin and Bondra take slapshots at him.

 

[twabby]

This thread was used to pimp Orlov. We all know JC is the teams most valuable defensemen. God help Samsonov as his GAA goes up because Orlov is too much of a chicken **** to hit someone or block a shot himself. If he uses that wimpy *** pokecheck again, i'm going to break all his sticks in rage. Maybe we can tie him to net like Goldberg in the Mighty Ducks and have Ovechkin and Bondra take slapshots at him.

Did Orlov do something to you personally?

 

[PB12]

Did Orlov do something to you personally?

Yeah. Its called getting better. With Orpik and Nisky gone, he's going to be one of the leaders. With a new goaltender coming in, its time to man up.

 

[francaisvolantsparis]

Junk science.

Show a measured correlation between your Awesomeness stat and winning hockey games, and then I'll pay attention to it.

You know that D is only a part of a team and the correlation can not be like 0.99999

But let's do it. Here you have the 11 points for last 11 seasons without the 2013 lockout season:

(190,94) (189,108) (199,121) (191,107) (192,92) (216,90) (226,101) (234,120) (252,118) (229,105) (242,104)

Pairs are ('Awesomeness 2.0' sum for first four best D, Caps season total points).

Pearson's Correlation is 0.34333

For example the same correlation for (Caps Total Points, Team Goals For) is 0.64700

 

[Hivemind]

You know that D is only a part of a team and the correlation can not be like 0.99999

But let's do it. Here you have the 11 points for last 11 seasons without the 2013 lockout season:

(190,94) (189,108) (199,121) (191,107) (192,92) (216,90) (226,101) (234,120) (252,118) (229,105) (242,104)

Pairs are ('Awesomeness 2.0' sum for first four best D, Caps season total points).

Pearson's Correlation is 0.34333

For example the same correlation for (Caps Total Points, Team Goals For) is 0.64700

That Pearson Correlation is roughly half of Corsi%'s relationship to winning (which ranges between 0.6 - 0.8 depending on season). Your metric is very far from the best way to evaluate players, so I don't see why anyone should put stock in it.

 

[francaisvolantsparis]

That Pearson Correlation is roughly half of Corsi%'s relationship to winning (which ranges between 0.6 - 0.8 depending on season). Your metric is very far from the best way to evaluate players, so I don't see why anyone should put stock in it.

I see, you do not understand what correlation is.

Let's do an exercise.

-Exercise-

Find three vectors D, F and TP such that:

- Correlation between (D+F) and TP is 0.7
- Correlation between D and TP is 0.3
- Correlation between F and TP is 0.4

 

[twabby]

Yeah. Its called getting better. With Orpik and Nisky gone, he's going to be one of the leaders. With a new goaltender coming in, its time to man up.

There are plenty of players who need to play better in order to win a championship again, but you always seem to single out Orlov like he personally murdered your family. Your criticisms of his game are petty, and if you're saying he needs to "man up" I'm sure you also agree that everyone else on the D-corps needs to "man up" as well because as a unit they have a lot of room for improvement, including Carlson who I do agree is their best defenseman.

Your criticism is also especially odd because by most accounts Orlov was their best defenseman against Carolina so why would he be the target of your ire? I'd get the criticism if Orlov's series against Carolina was a flash-in-the-pan similar to DSP in 2018, but it wasn't. Orlov has been very good for years now and it's likely that Niskanen's sharp decline was a major factor in Orlov's 2018-19 season being subpar.

Just say you don't like the guy and be honest with yourself and everyone else.

 

[francaisvolantsparis]

My forecast for the 2019/20 season:

Player	Awesomeness 2.0
Dmitry Orlov	65
John Carlson	64
Radko Gudas	57
Michal Kempny	48
Nick Jensen	47
Christian Djoos	38
Jonas Siegenthaler	25

[TBODY] [/TBODY]

Best lines:

Dmitry Orlov - Radko Gudas
Christian Djoos - John Carlson
Michal Kempny - Nick Jensen

 

[Calicaps]

This thread:

 

[Hivemind]

I see, you do not understand what correlation is.

Let's do an exercise.

-Exercise-

Find three vectors D, F and TP such that:

- Correlation between (D+F) and TP is 0.7
- Correlation between D and TP is 0.3
- Correlation between F and TP is 0.4

It's cute that you've just taken your first statistics class and want to try and come up with ways to apply it. I'm not going to solve your homework problem for you to prove that I understand correlation (nor does your particular problem even employ the same understanding of correlation that we're discussing).

To put it bluntly for you, Team Corsi% has a much stronger linear relationship with team points% than your Awesomeness Metric does. Your awesomeness metric is not a particularly strong tool at defining the relationship between a hockey teams' skills and winning hockey games, and given that we have superior tools (which have also demonstrated predictive value, which you have yet to demonstrate), I don't see why we would bother to take your stat seriously.

 

[francaisvolantsparis]

It's cute that you've just taken your first statistics class and want to try and come up with ways to apply it. I'm not going to solve your homework problem for you to prove that I understand correlation (nor does your particular problem even employ the same understanding of correlation that we're discussing).

To put it bluntly for you, Team Corsi% has a much stronger linear relationship with team points% than your Awesomeness Metric does. Your awesomeness metric is not a particularly strong tool at defining the relationship between a hockey teams' skills and winning hockey games, and given that we have superior tools (which have also demonstrated predictive value, which you have yet to demonstrate), I don't see why we would bother to take your stat seriously.

Do that exercise. Trust me, it will make you smarter. First you will show everyone, that you really understand what you are saying and second you will understand why you are wrong.

 

[Holtbyisms]

This thread:

Agreed, this is a giant steaming pile of statistics trolling poop.

 

[Hivemind]

Do that exercise. Trust me, it will make you smarter. First you will show everyone, that you really understand what you are saying and second you will understand why you are wrong.

You think anyone in this forum is going to be persuaded by your homework?
Look, you've only shown something with a relatively weak (compared to other metrics) correlation to winning hockey games. Heck, you even used the same evaluation of comparing pearson coefficients to demonstrate your statistic vs goals for. You have done nothing to indicate any causal relationship between your made up statistic, or its component statistics, and winning hockey games. You have done nothing to demonstrate future predictive value. You haven't even made any basic attempts to correct for scorekeeper biasing (such as the oft-used road statistics, to eliminate any bias that comes from playing in front of the same scorekeeper for 41/82 games out of the season). You have yet to demonstrate how you've determined or assigned weighting.

I love hockey stats, but this is just masturbatory math with no basis or predictive quality. This is the type of crap that gives statistical analysis a bad name. This is the type of crap that people who really want a "WAR" for hockey keep shitting out, that ends up indicating that Tyler Sloan is a better defenseman than Brooks Orpik.

 

[PB12]

There are plenty of players who need to play better in order to win a championship again, but you always seem to single out Orlov like he personally murdered your family. Your criticisms of his game are petty, and if you're saying he needs to "man up" I'm sure you also agree that everyone else on the D-corps needs to "man up" as well because as a unit they have a lot of room for improvement, including Carlson who I do agree is their best defenseman.

Your criticism is also especially odd because by most accounts Orlov was their best defenseman against Carolina so why would he be the target of your ire? I'd get the criticism if Orlov's series against Carolina was a flash-in-the-pan similar to DSP in 2018, but it wasn't. Orlov has been very good for years now and it's likely that Niskanen's sharp decline was a major factor in Orlov's 2018-19 season being subpar.

Just say you don't like the guy and be honest with yourself and everyone else.

Twabby guess what. He played well in SEVEN games. I want to see him play well in more than 7 games. You and the Orlov pep squad can hype of Orlov with your "stats" as much as you want. If you ask most Caps fans to take an eye test on Orlov, they will tell you he needs to work on his defensive game.

 

[PB12]

Can we just lock this thread. Its a waste of time.

 

[Silky mitts]

Hockey and football are more team sports than baseball and basketball where if the team is able to use the player’s strengths he’s more valuable, like the player that produces the most might not always be the player that is the most valuable whereas in baseball and basketball you’d always want the guy that produces the most. So a unified statistic is less realistic. But if the points, 5 on 5 goal differential, and Corsi overwhelmingly favor the same guy and a statistic says he’s not the best defenseman I dunno.

 

[Midnight Judges]

I think you guys are being a bit crotchety.

I agree the stats don't accurately reflect the awesomeness, but maybe it's not the end of the world to begin the conversation with stats, and then take a look at where the stats are steering the wrong way. Might be interesting.

Stats for defensemen in the NHL have always been problematic in that offensive stats don't necessarily reflect the best players, and defensive stats can also be equally if not even more misleading.

 

[twabby]

Twabby guess what. He played well in SEVEN games. I want to see him play well in more than 7 games. You and the Orlov pep squad can hype of Orlov with your "stats" as much as you want. If you ask most Caps fans to take an eye test on Orlov, they will tell you he needs to work on his defensive game.

He’s been good for the vast majority of his ~500 game NHL career. That was my exact point in saying “his postseason wasn’t a flash in the pan like DSP’s last year.”

 

[francaisvolantsparis]

You think anyone in this forum is going to be persuaded by your homework?

Why do you think others cannot judge by themselves? Face it you are wrong. Someone who like math has already an opinion. Someone who simply read your posts has a good guess as well.

You should not aggress people but reason logically to persuade someone.

 

[RandyHolt]

Open ice hits boost awesomeness. I wish they were tracked over the decades. I guess there is no point to start now with one of the last throwbacks retiring. One hit early in a game 1 can make that guy and his flaky teammates decide to dump it in the rest of the series.

#ClearingTheLDPipe

 

[CapitalsCupReality]

There are plenty of players who need to play better in order to win a championship again, but you always seem to single out Orlov like he personally murdered your family. Your criticisms of his game are petty, and if you're saying he needs to "man up" I'm sure you also agree that everyone else on the D-corps needs to "man up" as well because as a unit they have a lot of room for improvement, including Carlson who I do agree is their best defenseman.

Your criticism is also especially odd because by most accounts Orlov was their best defenseman against Carolina so why would he be the target of your ire? I'd get the criticism if Orlov's series against Carolina was a flash-in-the-pan similar to DSP in 2018, but it wasn't. Orlov has been very good for years now and it's likely that Niskanen's sharp decline was a major factor in Orlov's 2018-19 season being subpar.

Just say you don't like the guy and be honest with yourself and everyone else.

He’s a vocal critic, but it’s never personal in his comments. Honestly, I feel like Orlov should be more offensively than he is. His D had come around to pretty solid, but never stand out to you in a game good (the much too rare explosive hit aside). I will be very interested to see how his pair does this year with him as the lead dog. I’m expecting him to not be able to carry the pair, which will probably be needed somewhat IMO. Hope he proves me wrong or that his partner can rise to the challenge.