HFBoards

HFBoards (http://hfboards.hockeysfuture.com/index.php)
-   Detroit Red Wings (http://hfboards.hockeysfuture.com/forumdisplay.php?f=34)
-   -   Corsi, +/- and other Jamesian Ideology, wrt to the Wings (http://hfboards.hockeysfuture.com/showthread.php?t=753857)

Winger98 03-22-2010 08:12 AM

Corsi, +/- and other Jamesian Ideology, wrt to the Wings
 
Quote:

Originally Posted by Fugu (Post 24646624)
With Stuart, I see a guy who is doing the right thing the majority of the time. I like his mobility. I like his physical play. He fits in nicely with the offense and on attacks. He's good at keeping the pucks in on the point, or in getting back quickly if needed. That doesn't mean he doesn't make mistakes, but they aren't a matter of course every time he's on the ice. I do think a lot of situations would be worse if he hadn't been in the mix at all.

In baseball stats, there are separate stats to effectively measure a pitcher's luck, good or bad, in an attempt to show how accurate other stats such as ERA and WHIP are at reflecting their performance. A few years ago, Armando Galarraga had a great season for the Tigers with a good WHIP and ERA, but some of his peripherals showed that he was also unnaturally lucky that season with balls in play resulting in outs instead of hits, etc. His counting stats (wins, ERA) have since moved into more accordance with his peripherals.

I wonder if we wouldn't see something similar with Stuart. Not sure how to gauge how "lucky" a hockey player is (but maybe something involving quality/shots allowed, quality of comp/teammates, and goals allowed) but I wonder what such a thing would say about Stuart. Like you, I like him on the ice. He rarely appears to make a clear mistake -at least in the Lilja My-God-What-Are-You-Doing? vein and he actually hits people. Maybe he's just absurdly unlucky. Though it should also be pointed out that he hasn't always had the best +/- in the past, either. So maybe we're both just missing something.

Roy S 03-22-2010 10:44 AM

Quote:

Originally Posted by Winger98 (Post 24646963)
In baseball stats, there are separate stats to effectively measure a pitcher's luck, good or bad, in an attempt to show how accurate other stats such as ERA and WHIP are at reflecting their performance. A few years ago, Armando Galarraga had a great season for the Tigers with a good WHIP and ERA, but some of his peripherals showed that he was also unnaturally lucky that season with balls in play resulting in outs instead of hits, etc. His counting stats (wins, ERA) have since moved into more accordance with his peripherals.

I wonder if we wouldn't see something similar with Stuart. Not sure how to gauge how "lucky" a hockey player is (but maybe something involving quality/shots allowed, quality of comp/teammates, and goals allowed) but I wonder what such a thing would say about Stuart. Like you, I like him on the ice. He rarely appears to make a clear mistake -at least in the Lilja My-God-What-Are-You-Doing? vein and he actually hits people. Maybe he's just absurdly unlucky. Though it should also be pointed out that he hasn't always had the best +/- in the past, either. So maybe we're both just missing something.

As for turning the Wings off, I sort of agree with you. Not sure I'd do it for the same reason but, since moving to Cleveland, I catch FAR fewer games than before and it hasn't killed me.

I like where your head is at and this will be a very long post...

CORSI is a stat that looks at the number of shots generated for and against when a player is on the ice at even strength (shots on net, missed shots, blocked shots, etc.) Its been shown that shooting percentage and save percentage have a large degree of variation every year amongst teams and players and total shot differential is the best predictor of future production and/or tries to take some of the luck out of the equation. Scoring chances also can have a large degree of variance.

As a team, we have the 3rd best shot differential in the entire NHL and have a team wide shooting percentage that is unsustainable and will not stay that low for long. Zetterberg, Lidstrom, Williams, Bertuzzi and Cleary are some of the guys who have been a bit unlucky in that department. In terms of team Corsi %, we have been in the top 5 every year (including this year) post lockout and had the best team post lockout score in 07-08. This is a stat that doesn't look at goalie play and mostly tells you how good a team's position players are at ES, PP and PK. So, that does need to be considered, but a goalie's save percentage has a high degree of variability each season and can be pretty random. Howard's save percentage, for example, is due for regression as his current save percentage is unsustainable over the long run unless he is Dominik Hasek in his prime.


Corsi On tells you what happens when a player is on the ice, but that is largely dependent on who he is playing with. In general, the better measure is a Corsi +- that also looks at what happens with him off the ice. Overall, Stuart has a positive Corsi number both this season and last season. On the other hand, our entire team (except Draper) had a positive Corsi number last year. This year, only Maltby and Brad May had a negative Corsi number.

If we look at Corsi the last 2 years, then Stuart still does not look good. He had the lowest Corsi ON score among our d men last year and the lowest Corsi +- score. This year, he again has the lowest Corsi On score and has the second worst Corsi +- score among our D men (Big E is worse). Essentially, Stuart has a much lower shot differential than is normal for our team.

Last year, his GA looked worse than it should have b/c of poor goalie play and/or bad luck. His save percentage against last year of .897 is unsustainable over a large sample size and made him look worse than he was last season. This year, though, his save percentage against is .918. So, he has received good goalie support this season.

Another area to look at is defensive zone starts. Players who start more in the defensive zone will likely have lower Corsi scores and be underrated (defensive forwards come to mind). However, Stuart sees higher offensive zone starts than defensive zone starts and starts in the offensive zone more so than Lidstrom, Rafalski, and Ericcsson do and basically the same as Kronwall.

One underrated aspect of his game is penalties. He has drawn 13 penalties this season and only taken 6. As a result, he has given us 7 more PP opportunities. Given an average success rate on those PP, he likely gave us an additional goal or two that won't show up in the box score. Pavel has drawn 26 and only taken 3. With those 23 extra PP opportunities, he has probably given us 5 extra goals. On the other end of the spectrum, Bertuzzi has probably costs us about 3 goals b/c of his disparity between penalties drawn and taken.

All in all, there is probably some element of bad luck involved with his numbers. But, for the most part, he allows too many shots against and hasn't shown the offensive game to make up for that at ES. With LA, in 07-08, his relative Corsi +- compared to the rest of the team was right around 0. So, I'm not sure if his play just dropped when he came here, doesn't fit the style of play, or if he is made to look worse statistically for reasons one could only see through observation. I'm sure some of the latter is at play as I really think he was hurt by playing much more this season and should do better in the coming games as Babcock can pair him with Kronwall more often and Stuart won't have to cover as much for him as he would with an inferior partner. In general, I think he is better as a complimentary player over the long run and gets exposed when he has to play more of a lead role.

As for guys who might appear to be getting unlucky or underrated by the surface stats... you need to look at more than 1 year of data, but Franzen had the best Corsi +- on our team last year (by a slim margin) and does again this year in a limited sample of games. He was a major, major loss and is every time he misses a game. I guess a good measure is to see who we struggle the most with when they are on off the ice and/or who we get by the best with them off the ice. These need to be placed in proper context as Abdelkader, Maltby and Meech are both overrated in this stat b/c of how weak there competition was. If you exclude them then it gives you an interesting list, imo, and shows the worth of Drew Miller defensively.

http://www.behindthenet.ca/2009/new_...&team=DET&pos=

In general, the best measure will be Corsi +- as long as it is placed in the proper context. Its not saying Holmstrom is a better player than Z or Pavel. He goes against easier competition and is more of a complimentary player. It just says that he is great in his role and there is a noticeable impact when he, or that line, is not on the ice. This will tend to underrate defensive forwards (like Miller) and overrate guys who go against very easy competition (like Jason Williams). One needs to look at all 3 Corsi numbers (Rel, On, Off), the quality of competition and teammates, offensive or defensive zone percentage and penalties drawn and taken to get a more complete view of ES statistics.

http://www.behindthenet.ca/2009/new_...&team=DET&pos=

Jussha 03-22-2010 10:51 AM

I know a lot of people like to use CORSI stats to get an overall measure of a player's worth and contributions to the team. I don't know too much about CORSI, but I know a boatload of oiler fans used to post CORSI stats non stop to show that Horcoff was a first line centre and worth the huge contract he got. Having lived in Edmonton my whole life and not being able to stand the type of player horcoff is (I truly think the guy sucks and when I see him on the ice I see nothing in his skill set that shows me he is a top 6 forward in the NHL), I don't think CORSI should be a be all and end all for confirming someone's overall contribution to a team, and whether he is good or bad. But like I said, I'm not a CORSI guru and don't understand it completely, maybe I should look more into it when I get a bit of free time.

Roy S 03-22-2010 11:30 AM

Quote:

Originally Posted by Jussha (Post 24648614)
I know a lot of people like to use CORSI stats to get an overall measure of a player's worth and contributions to the team. I don't know too much about CORSI, but I know a boatload of oiler fans used to post CORSI stats non stop to show that Horcoff was a first line centre and worth the huge contract he got. Having lived in Edmonton my whole life and not being able to stand the type of player horcoff is (I truly think the guy sucks and when I see him on the ice I see nothing in his skill set that shows me he is a top 6 forward in the NHL), I don't think CORSI should be a be all and end all for confirming someone's overall contribution to a team, and whether he is good or bad. But like I said, I'm not a CORSI guru and don't understand it completely, maybe I should look more into it when I get a bit of free time.

It isn't the end all. But, its one of the best stats available. You need to look at multiple stats and multiple years and place them in the proper context. Here is a link to a primer on it.

http://www.behindthenethockey.com/20...estions-3-what

Hockey analysis is probably 15 or so years behind baseball and doesn't lend itself as well to statistical analysis. At the winter meetings, there was an article somewhere that said that all but 2 MLB teams had a stats department. Most teams have consultants in the NHL, but I doubt many (or any) have a full team of stat guys. B/c of the free flowing nature of the game, it is hard to measure everything and its hard to separate a player's contributions from his teammates. Shot differential is the best predictor of future success and the best way we currently have to evaluate a player is through that and looking at what happens with him on the ice and without him on the ice. You can make some inferences based on that and it is better than just looking at PTS and +-. Players also tend to be fairly consistent across teams with their ratings and that gives some more credibility to them. Bertuzzi and Eaves both had negative Corsi +- ratings last year and do again this season. Williams was slightly below average last year and slightly above average this season in Corsi +-, but most of that difference can be attributed to the much easier competition he's been asked to play against this season compared to last and the difference in teammate quality.

Tom Awad's GVT is an attempt to put a player's total contributions into one neat statistic. Its not a bad stat to look at, but you need to look at much more.

At this stage of hockey analysis, observation/scouting is of more importance in projecting future performance, but I think the stats do a good job of detailing what happened in the past.

RedWingsNow* 03-22-2010 08:59 PM

Quote:

Originally Posted by Scottwood (Post 24649198)

At this stage of hockey analysis, observation/scouting is of more importance in projecting future performance, but I think the stats do a good job of detailing what happened in the past.

In other words, believe your eyes as it happens.
Believe the stats from the past.

Seems like an illogical point made by someone trying to find a Bill James for hockey.

Roy S 03-22-2010 09:31 PM

Quote:

Originally Posted by Captain Bob (Post 24661589)
In other words, believe your eyes as it happens.

No. Our eyes deceive us and play tricks on us and we sometimes only see what we want to see. See the illogical Nick Lidstrom discussions early in the year about him getting owned defensively. I would never say that you should completely trust what your eyes tell you. Eye witness accounts, even of crime scenes, are incredibly sketchy.

Quote:

Believe the stats from the past.
I wish you would have when we had the discussions about Lidstrom's defense.

Quote:

Seems like an illogical point made by someone trying to find a Bill James for hockey.
I fail to see what I said was illogical. Stats are a tool and so to is human observation. Both should be used in looking at past performance and in identifying future performance. Depending on the circumstances, one can be more useful than the other but they both need to be considered. Research can also be very important in identifying trends.

Fugu 03-23-2010 11:57 AM

I wanted to save this portion of the discussion from the Homer Thread Gone Awry.... Scott wrote far too much to leave it to die on that vine.

We also need to introduce people to Poisson, and reconcile how much you can use statistical measures to a game many claim is mainly random, while difficult to tweeze apart the team:individual aspects.

Harnessed in Slums 03-23-2010 10:00 PM

Quote:

Originally Posted by Scottwood (Post 24649198)
B/c of the free flowing nature of the game, it is hard to measure everything and its hard to separate a player's contributions from his teammates. Shot differential is the best predictor of future success and the best way we currently have to evaluate a player is through that and looking at what happens with him on the ice and without him on the ice.

Yeah this right here does a good job encapsulating my general view on stats. First off Baseball is so much easier to chart because it is so discrete (meaning easy to break down into specific plays). The pitcher pitches the ball. The batter either hits the ball or he doesn't. The fielder either makes the play or he doesn't. Hockey is rarely that straightforward. I mean Tomas Holmstrom deserves a crapload of assists for goals where he's never factored onto the scoresheet. Is that worthless? If not, is there really a way to track how much his rear end contributed to a goal being scored by blocking the goalie's view? Or the defensive player guarding him being in the goalies way or inadvertently deflecting a puck into his own net? It's just so much more messier than baseball given how many players can be in on any one given data point.

With that being said the biggest stat in my opinion is how much of a shot differential your team has as far as shots you put on goal and the number you allow. But it's a team-wide stat that involves 18 players - three forwards at a time and two defenseman at a time as a five man unit going up against another five man unit.

So in my opinion that's the one stat I think is really worth paying attention to and the CORSI of course touches on that. My only qualification would be that the team-wide CORSI is much more important than the individual CORSI. I mean Peter Regin and Tomas Holmstrom are two of the top 3 CORSI guys in the league.

RedWingsNow* 03-23-2010 10:21 PM

He plays defense like a man.

This nerdification of hockey is useless.

What was his CORSI when Babcock called him the best defenseman on the team earlier this year?

Fugu 03-23-2010 10:53 PM

fishy things
 
http://canuckscorner.com/tombenjamin/?p=378


http://canuckscorner.com/tombenjamin/?p=379


http://canuckscorner.com/tombenjamin/?p=382

Roy S 03-23-2010 10:53 PM

Quote:

Originally Posted by Captain Bob (Post 24687745)
What was his CORSI when Babcock called him the best defenseman on the team earlier this year?

Of course, what coaches say to the media and what they actually believe can sometimes be 2 completely different things. And, if we are using that argument, then why hasn't he said it lately (when his play, statistically speaking, completely dropped off)?

Winger98 03-23-2010 11:07 PM

A lot to read/digest, frankly not the energy for it tonight. Great posts, though, Scott. I've seen the Corsi #s before but didn't know what it related to, after reading your explanation, I like it even more.

Without going into more numbers, Stuart still bugs me because my eyes say he's good back there. But I wonder how much can be read into fewer shots generated while he's on the ice. Without looking anything up, I'm assuming Stuart sees a lot of ice time with our "second" line, but I wonder if his pairing also soaks up a lot of ice time with our "fourth" line (to shield our third pairing a bit while keeping Lids/Raffy fresh to head out with our top line). Without stats, such wandering might be out of place in a stat thread, but Stuart's numbers bug me.

nik jr 03-23-2010 11:42 PM

Quote:

Originally Posted by Captain Bob (Post 24687745)
He plays defense like a man.

This nerdification of hockey is useless.

What was his CORSI when Babcock called him the best defenseman on the team earlier this year?

but not necessarily a smart man


i haven't based my opinion of stuart on exotic stats. i have just noticed how often he gets scored on, and how much he is out of position. i noticed those things when he played for other teams.

same for kronwall

TheMoreYouKnow 03-24-2010 06:04 AM

I'm skeptical of the stathead philosophy being applied to hockey for all the obvious reasons, the fluidity of the game, the difficulty of quantifying player action and the dynamic between them etc. so I can't say I'm that impressed by these statistics.

Sports like hockey and soccer will always have to rely more on scouting and more or less subjective opinions in determining player quality than other sports like baseball or football.

I personally think Stuart is a 2nd pairing guy and in a solid role with the Wings, he is nowhere near as unreliable as your typical 3rd pairing guys in my view. I think the problem with him is that he was once projected as a star and then he merely turned into an okay NHLer.

norrisnick 03-24-2010 06:31 AM

Quote:

Originally Posted by Captain Bob (Post 24687745)
He plays defense like a man.

This nerdification of hockey is useless.

What was his CORSI when Babcock called him the best defenseman on the team earlier this year?

This isn't figure skating. There are no style points. Effectiveness is what matters, not how pretty they are doing whatever it is they're doing.

Fugu 03-24-2010 06:53 AM

Quote:

Originally Posted by norrisnick (Post 24691364)
This isn't figure skating. There are no style points. Effectiveness is what matters, not how pretty they are doing whatever it is they're doing.


Well, isn't that what el Capitan is saying-- real men aren't pretty skaters? :sarcasm:


Remember that signature Heaton put up about someone's note to self about crossing the ice when Kronwall was out, and keeping your head up (Scott Nichols??)?

That doesn't show up on any stat-- getting into a guy's head.

Still, what no one has addressed yet is the team nature of hockey with highly dependent variables. Randomness seems to dictate more the results than what anyone does on the ice (as painful as that seems to consider).

Personally, I'd like to hear Scott's reconciliation of the Poisson model cited above and how CORSI or any other stat fits in, from his perspective.

Roy S 03-24-2010 09:42 AM

Quote:

Originally Posted by Fugu (Post 24691431)

Still, what no one has addressed yet is the team nature of hockey with highly dependent variables. Randomness seems to dictate more the results than what anyone does on the ice (as painful as that seems to consider).

Personally, I'd like to hear Scott's reconciliation of the Poisson model cited above and how CORSI or any other stat fits in, from his perspective.

Goal scoring is incredibly random and there is a ton of luck involved. Team shooting percentage and save percentage has a lot of variation in it each season and so do individuals shooting percentage and save percentage. That is what Corsi strips out. It looks at the total shot differential with a player on the ice and is a far better predictor of future success than goals for and goals allowed.

Its why we never spent big money on a goalie until Hasek in 2002 and Joseph for a few seasons. Save percentage is pretty random and unless you have an elite goalie, like Hasek was, then you are going to be paying a premium for something that you really shouldn't. The difference among most goalies in the NHL is not that large. The difference from a 1st line forward compared to a 4th line grinder is extremely different. Thus, we spend all of our money on forwards and defenseman.

We also knew ahead of time the importance of shot differential and that if the other team doesn't have the puck, then they won't score that much. Hence, our love for puck moving d-men before it was the trendy thing to do. We also knew ahead of time how important faceoff % can be and how hurtful extra penalties can be.

Its also why Holland never panicked when we had an early round exit in the playoffs and dissembled the whole roster like some were calling for. The playoffs are basically a crapshoot given the small sample size that a 7 game series is, and you need a lot to break your way to win it all. The talent disparity in the league is much closer together now than what it was in the 80's with the Oilers or any other dynasty. So, when we had great regular seasons and didn't win it all in the postseason, it didn't necessarily mean that we were not a good playoff team. Certainly, some personnel is better suited for postseason play than others and you need to identify those players. Overall, though, if you are a great regular season team year after year, like we have been, then you are going to be the recipients of some early round upsets and if you have the talent and team good enough to win it all every year, then you are going to cash in at some point. We've cashed in 4 times.

RedWingsNow* 03-24-2010 10:16 AM

Let me clean this up for you ...


Quote:

Originally Posted by Scottwood (Post 24692928)
Goal scoring is incredibly random and there is a ton of luck involved. Team shooting percentage and save percentage has a lot of variation in it each season and so do individuals shooting percentage and save percentage. That is what Corsi strips out. It looks at the total shot differential with a player on the ice and is a far better predictor of future success than goals for and goals allowed.

This stat measures carrying the play more than it does scoring goals.


Quote:

Its why we never spent big money on a goalie until Hasek in 2002 and Joseph for a few seasons. Save percentage is pretty random and unless you have an elite goalie, like Hasek was, then you are going to be paying a premium for something that you really shouldn't. The difference among most goalies in the NHL is not that large. The difference from a 1st line forward compared to a 4th line grinder is extremely different. Thus, we spend all of our money on forwards and defenseman.
Even though Ozzie had hall of fame stats, he wasn't a real Hall of Fame goalies. But real Hall of Fame goalies are few and far between. The difference between mediocre goalie and good goalie isn't that much.
And that, (with the exception of the Belfour contract offer) is why we don't break the bank on goalies in the salary cap era.

Quote:

We also knew ahead of time the importance of shot differential and that if the other team doesn't have the puck, then they won't score that much. Hence, our love for puck moving d-men before it was the trendy thing to do. We also knew ahead of time how important faceoff % can be and how hurtful extra penalties can be.
The best defense is a good offense (red wings fans have known this 4ever).
Faceoffs are important.

Quote:

Its also why Holland never panicked when we had an early round exit in the playoffs and dissembled the whole roster like some were calling for. The playoffs are basically a crapshoot given the small sample size that a 7 game series is, and you need a lot to break your way to win it all. The talent disparity in the league is much closer together now than what it was in the 80's with the Oilers or any other dynasty. So, when we had great regular seasons and didn't win it all in the postseason, it didn't necessarily mean that we were not a good playoff team. Certainly, some personnel is better suited for postseason play than others and you need to identify those players. Overall, though, if you are a great regular season team year after year, like we have been, then you are going to be the recipients of some early round upsets and if you have the talent and team good enough to win it all every year, then you are going to cash in at some point. We've cashed in 4 times.
Anything can happen in a seven game series.

I think you're letting the Wings off the hook a bit to easy here.
Quote:

2006 -- lost to Edmonton, 4-2, Western Conf. quarterfinals

2004 -- defeated Nashville, 4-2, Western Conf. quarterfinals
-- lost to Calgary, 4-2, Western Conf. semifinals

2003 -- lost to Anaheim, 4-0, Western Conf. quarterfinals
I think those three seasons, taken together, showed that there was something fundamentally wrong with the roster.

Between 05-06 and 06-07, this team dropped Williams, Wooley SHanahan and Yzerman and replaced Legace with Hasek.
The next year, we lost Lang and Schneider

There was a big transformation in team speed.

Roy S 03-24-2010 10:37 AM

[QUOTE=Captain Bob;24693442]

Quote:

This stat measures carrying the play more than it does scoring goals.
And carrying the play translates to scoring goals.


Quote:

Even though Ozzie had hall of fame stats
No, he didn't.


Quote:

The best defense is a good offense (red wings fans have known this 4ever).
Faceoffs are important.
So too is not taking penalties.

Quote:

Anything can happen in a seven game series.

I think you're letting the Wings off the hook a bit to easy here.


I think those three seasons, taken together, showed that there was something fundamentally wrong with the roster.

Between 05-06 and 06-07, this team dropped Williams, Wooley SHanahan and Yzerman and replaced Legace with Hasek.
The next year, we lost Lang and Schneider

There was a big transformation in team speed.
There probably was something wrong with the roster. But, they were better than those teams they lost to and that is why we continued to have success beyond those postseason failures and why those teams didn't (Anaheim retooled quickly when they got Pronger but there was a reason they missed the playoffs the following season after their near cup run in 03).

Fugu 03-24-2010 12:41 PM

Quote:

Originally Posted by Scottwood (Post 24692928)
Goal scoring is incredibly random and there is a ton of luck involved. Team shooting percentage and save percentage has a lot of variation in it each season and so do individuals shooting percentage and save percentage. That is what Corsi strips out. It looks at the total shot differential with a player on the ice and is a far better predictor of future success than goals for and goals allowed.


See, I still think this ignores the fundamentals that you seem to accept earlier--- huge variability simply due to random events. We're taking a group measure (statistical measures generally are), and then trying to attribute them to an individual.

The ultimate measure of success is goals scored. Goal scoring seems to be a random event. All the rest of it is an attempt to minimize goals scored, but that's fighting..... entropy? (Not sure what to call it.)

Mind you, I think some stats are necessary, but taken alone or even as a primary tool seems to accept that hockey is static when in fact it's one of the most dynamic sports around.

How about this... are there good players buried on very bad teams? How do their stats compare to the players that are on the best teams?

norrisnick 03-24-2010 05:53 PM

Quote:

Originally Posted by Fugu (Post 24691431)
Well, isn't that what el Capitan is saying-- real men aren't pretty skaters? :sarcasm:


Remember that signature Heaton put up about someone's note to self about crossing the ice when Kronwall was out, and keeping your head up (Scott Nichols??)?

That doesn't show up on any stat-- getting into a guy's head.

You don't think Nick gets in forwards' heads? See arice's sig and Brown's comments on the matter. Nash and Crosby as well had statements recently about how ridiculous it is to play against a D like Nick where it's a lot simpler and more straight forward to play against a physical guy. Yet Nick doesn't "play defense like a man"...

Roy S 03-24-2010 06:56 PM

Quote:

Originally Posted by Fugu (Post 24696060)
See, I still think this ignores the fundamentals that you seem to accept earlier--- huge variability simply due to random events. We're taking a group measure (statistical measures generally are), and then trying to attribute them to an individual.

The ultimate measure of success is goals scored. Goal scoring seems to be a random event. All the rest of it is an attempt to minimize goals scored, but that's fighting..... entropy? (Not sure what to call it.)


I'm not sure I understand what you're arguing here. Shot differential is not random. That is all Corsi is measuring... shot differential with a player on the ice. You then need to use other stats (like competition level and offensive zone %) to give it meaning and put it in proper context. Maybe Corsi works bests at the team level and not as well at the individual level (not that I necessarily agree with that)? Although, I don't see what would be wrong with that, either. That is all we really care about... how our team is doing and how they are likely to do in the future.

Quote:

How about this... are there good players buried on very bad teams? How do their stats compare to the players that are on the best teams?
Of course there are good players on bad teams. Relative Corsi +- is a stat designed to see how important they are to their specific team.

http://www.behindthenet.ca/2009/new_...=12&team=&pos=

This is not a ranking of the best players in the NHL this season. Its not saying Matt Niskanen is a better player than Nick Lidstrom. You just have to know how to read the stat and put it in a proper context, and what its pros and cons are. Some don't like doing that and prefer a "correct answer" when there really isn't one. I don't mind thinking critically and realizing that there is often more than meets the eye and questioning conventional wisdom. Its not for everyone. But, I enjoy it.

Roy S 03-27-2010 01:03 PM

Quote:

Originally Posted by Captain Bob (Post 24693442)

This stat measures carrying the play more than it does scoring goals.

I found what I was looking for.

http://objectivenhl.blogspot.com/201...oting-and.html

Quote:

Specifically, he looked at aggregated EV goal and shot ratios for each team over the last three seasons as a whole (2007-08, 2008-09 and 2009-10, I presume). He found that the teams with the best EV goal ratios during this period were overwhelmingly teams that outshot the opposition at EV.

I think that his point is an important one. While the relationship between outshooting and outscoring may not be apparent over brief periods, the teams that succeed at even strength over the long run are those that spend more time in the opposition's end than their own.
Carrying the play/ out shooting the opposition= scoring more goals.

ArGarBarGar 03-27-2010 01:14 PM

Quote:

Originally Posted by nik jr (Post 24689456)
but not necessarily a smart man


i haven't based my opinion of stuart on exotic stats. i have just noticed how often he gets scored on, and how much he is out of position. i noticed those things when he played for other teams.

same for kronwall

What I came in to post. Last night showed a perfect example of that on the Minnesota breakaway attempt. Stuart was the last defenseman back and instead of hanging back for a 1 on 1 (possible 2 on 2) he pinched and tried to keep the puck in and was easily passed up.

He seems to be very inconsistent at times.

Roy S 03-27-2010 01:18 PM

And a statistical look at the most valuable players to their respective teams (not neccessarily a ranking of the best players, but just a look at those who had the most positive impact on their teams performance).

That Lidstrom guy sucks. He comes in at #2 in the entire NHL among both forwards and defenseman.

http://www.ontheforecheck.com/2010/3...say-nhl-mvp-is
Quote:

Given all the issues Detroit has had to struggle through this season, the Red Wings have had to rely on Lidstrom perhaps more than they ever have. His penalty killing work is particularly telling. When he's on the ice, the Red Wings give up 3.45 goals per 60 minutes, a rate that is better than the top overall PK teams in the league. When he's on the bench, the Wings give up 8.36 goals per 60 minutes, which would rank among the worst.


All times are GMT -5. The time now is 09:09 PM.

vBulletin Copyright ©2000 - 2014, Jelsoft Enterprises Ltd.
HFBoards.com, A property of CraveOnline, a division of AtomicOnline LLC ©2009 CraveOnline Media, LLC. All Rights Reserved.