08-24-2013, 04:25 PM
|
#41
|
Franchise Player
|
Quote:
Originally Posted by Justin Azevedo
same principle. thornton dominates possession, phaneuf is a 27 minute a night defenseman.
oh, and since you're going to be a condescending dick about it - their.
man it would be great if someone of your intelligence level could find some way to differentiate guys who have skill taking away the puck and pluggers taking it away more due to their possession levels. also toews has traditionally played easier competition.
the point here is that this stat basically takes the least useful players on the team in terms of possession (and winning games) and makes them "good". the players aren't as good and they are more of a liability the higher they score on this arbitrary scale
fundamentally untrue. the rangers have a much higher shot count because they play at msg 41 games per year. just because they are official nhl stats doesn't mean that they're right.
aww man i don't care at all.
|
dude What the heck are you even looking at this thread???
Quote:
this is one of the least-needed threads in the history of the internet.
|
and yet here you are not only lurking but feeling a compulsion to post.
You should have a lot to talk about at your next session
Quote:
oh, and since you're going to be a condescending dick about it - their.
|
Their what?
Quote:
aww man i don't care at all
|
and you are just waaaaaay too cool or your Caps is broken. Must of taken a force of will to work the " for whatever emphasis on whatever point you were trying to make.
|
|
|
08-24-2013, 04:49 PM
|
#42
|
First Line Centre
|
Quote:
Originally Posted by ricardodw
My hypothesis is that teams can only have so many soft players before they begin to become ineffective.
Chicago was 30th in road hits and total hits..... however they only had 2 players that would fit in the soft category : Kane and Keith.
Sort of a gang mentality .. everyone plays hard and they don't get pushed around. The same works in reverse with so many small soft guys it becomes too hard and courageous for the remaining gang members to stand up on a continuous basis when you can see 5-6 guys never stepping up.....
The soft guys HAVE to justify their role in the gang by being exceptional at what they do good. Cammalleri and Hudler (and Tanguay and Cervenka) have to be 30-40 goal scorers to justify their role in the gang.
Stajan used to be a soft skilled guy but had a much better season 2012-13 when he competed harder and grittier. If he isn't going to kick in 60 pts then at least he is making other contributions.
|
I would argue that hits will generally inversely related to possession. Since Chicago generally has the puck a lot, it is to be expected that they hit less. I think you'll get a better indicator of grit if you normalize by a possession metric. For example, to see how much a team hits, find out hits given/possession. Higher is better. To see how well they can take hits, find out possession/hits received. Again, higher is better as it indicates they can take a hit without losing the puck.
|
|
|
The Following 2 Users Say Thank You to sworkhard For This Useful Post:
|
|
08-24-2013, 05:37 PM
|
#43
|
First Line Centre
|
Justin, I must ask, do you still go around bars trying to convince people you're this Justin Azevedo, telling them you played for Kitchener in the OHL and represented Team Canada, despite being a different "shape" than one would expect from a pro athlete? I remember when you did that to me a couple years ago, in trying to promote how Flames Nation had tactical analysis above other sites because of your playing experience...weak.
Well that's my nothing contribution, other than agreeing with those - Justin included - who believe this to be a great measure of picking out players who correlate to losing, and that the stats that make up this measure are wildly inconsistent and misleading. I hope going forward the Oilers, Leafs, and Canucks aim to make their entire rosters consist of high "RGI" players, ignoring other more meaningful and successful measures.
__________________
"I think the eye test is still good, but analytics can sure give you confirmation: what you see...is that what you really believe?"
Scotty Bowman, 0 NHL games played
|
|
|
The Following User Says Thank You to Lanny_McDonald For This Useful Post:
|
|
08-24-2013, 06:16 PM
|
#45
|
Draft Pick
|
Quote:
Originally Posted by united
Justin, I must ask, do you still go around bars trying to convince people you're this Justin Azevedo, telling them you played for Kitchener in the OHL and represented Team Canada, despite being a different "shape" than one would expect from a pro athlete? I remember when you did that to me a couple years ago, in trying to promote how Flames Nation had tactical analysis above other sites because of your playing experience...weak.
|
unfortunate for you that you messed up this sad attempt at libel by not realizing that i've only been affiliated with flamesnation for the past year and a bit.
but please, feel free to prove that i've ever done this.
otherwise, lol. thanks for coming out.
Last edited by Justin Azevedo; 08-24-2013 at 06:18 PM.
|
|
|
08-24-2013, 08:24 PM
|
#46
|
Franchise Player
|
Quote:
Originally Posted by thymebalm
Yea, but the thing is, every team plays in every building
|
Every team plays in ONE building for 41 games, and no two teams play 41 games in the same building. When the statistics you are using are counted differently in each building, that alone is enough to make a valid comparison impossible.
I like the idea of compiling the stat using road games only, and also the idea of correcting the stat with a puck-possession factor. However, those two changes will only fix two of the most obvious problems with a ginned-up number that, I'm afraid, doesn't really mean anything.
Last edited by Jay Random; 08-24-2013 at 08:28 PM.
|
|
|
08-25-2013, 12:48 AM
|
#47
|
tromboner
Join Date: Mar 2006
Location: where the lattes are
|
Quote:
Originally Posted by New Era
Since this stat is based off of subjective measures, such as what is a take away/give away or what is a hit, why not just say who is soft and who has grit?
|
Because then you could make whoever you want gritty or not gritty? Those second assists can be kind of iffy, so screw points, why not just say who's a good scorer and who isn't?
E.g. Alex Tanguay is a gritty guy who but he's no good offensively. Tim Jackman is a soft guy who scores a lot. Because I said so. Q.E.D.
====================
Quote:
Originally Posted by Jay Random
I like the idea of compiling the stat using road games only, and also the idea of correcting the stat with a puck-possession factor. However, those two changes will only fix two of the most obvious problems with a ginned-up number that, I'm afraid, doesn't really mean anything.
|
It might mean something. It hasn't really been proven meaningless or meaningful.
Last edited by SebC; 08-25-2013 at 01:03 AM.
|
|
|
08-25-2013, 01:21 AM
|
#48
|
Franchise Player
|
Quote:
Originally Posted by SebC
It might mean something. It hasn't really been proven meaningless or meaningful.
|
It's an omnium gatherum of questionable numbers that are only questionably related to each other, all added up without any scaling or weighting factor. (Are 1 hit, 1 blocked shot, and 1 takeaway all equally valuable in winning hockey games? I have reason to doubt it.)
If ricardodw (or anyone else) can demonstrate a strong correlation between his composite number and goal-scoring or goal prevention, then we can start to talk about how meaningful it is. As it stands, it's just his personal taste in players (basically, 'I like tough guys') dressed up in the form of an arbitrary number.
|
|
|
The Following 3 Users Say Thank You to Jay Random For This Useful Post:
|
|
08-25-2013, 02:26 AM
|
#49
|
tromboner
Join Date: Mar 2006
Location: where the lattes are
|
Quote:
Originally Posted by Jay Random
It's an omnium gatherum of questionable numbers that are only questionably related to each other, all added up without any scaling or weighting factor. (Are 1 hit, 1 blocked shot, and 1 takeaway all equally valuable in winning hockey games? I have reason to doubt it.)
|
Agreed on all counts. However, if a "weighted toughness winning factor" were to exist (say, a goals equivalency for hits, blocks, takeways, and giveaways), then the unweighted equivalent would reflect the weighted version to some extent and thus have value.
|
|
|
08-25-2013, 03:17 AM
|
#50
|
Franchise Player
Join Date: May 2004
Location: Helsinki, Finland
|
Quote:
Originally Posted by Jay Random
As it stands, it's just his personal taste in players (basically, 'I like tough guys') dressed up in the form of an arbitrary number.
|
I don't think that's fair at all. Imperfections can be argued, but it's clearly an attempt at finding a somewhat objective combined number.
|
|
|
08-25-2013, 03:43 AM
|
#51
|
First Line Centre
|
Quote:
Originally Posted by ricardodw
Iginla does not block shots.... therefore he doesn't kill penalties..... feeds to a lower ARGI score. Point scoring is very very important.......
|
Huh? Not following your logic there.
Blocking shots isn't necessarily a requirement for forwards killing penalties. A lot of it depends on the system. Take Michael Grabner. He's second among Isles forwards in PK time and he's 10th among Isles forwards in blocked shots.
Quote:
Originally Posted by ricardodw
So is physically controlling the game.....
Pittsburgh had far better scoring stats and Boston was far grittier. Boston 4 Pittsburgh 0. The Leafs are not nearly as talented as Pittsburgh but forced Boston to resort to a miracle on ice finish. Why? The ARGI might provide some insight as to why Toronto was better than Pittsburgh in the playoffs.
|
I'll let you do the math on this one but Boston vs Pittsburgh. Hits/Blocked shots/takeaways/giveways.
Game 1: Bruins win 3-0
Pens: 34/10/9/8
Bruins: 19/11/5/1
Game 2: Bruins win 6-1
Pens: 37/9/5/12
Bruins: 19/16/5/2
Game 3: Bruins win 2-1
Pens: 46/24/10/8
Bruins: 34/25/13/10
Game 4: Bruins win 1-0
Pens: 33/12/1/8
Bruins: 29/19/5/11
The stats suggest that the Bruins were hardly grittier. As you can see, the Pens actually out hit the Bruins every game and by a significant margin. Blocked shots were generally pretty close except for the 6-1 game. Similar figures can be found in the LA vs Chicago series where LA outhit Chicago every game and blocked slightly less shots. Looking at the stats for both series, it seems the winner was generally the team that had the better takeaway/giveaway ratio.
In the regular season, I think Colorado had one the best "RGI" index score yet they were one of the worst teams in the league.
|
|
|
The Following 2 Users Say Thank You to FAN For This Useful Post:
|
|
08-25-2013, 05:02 AM
|
#52
|
First Line Centre
Join Date: Sep 2004
Location: Haifa, Israel
|
Quote:
Originally Posted by theinfinitejar
You can do all the work you want, but this metric is useless from the word go.
To win hockey games you need score more goals than the other guys, and the best way to do that is to have the puck more than they do. So if you have a metric where "positive" events can only happen when the other team has the puck, how can it possibly be useful at predicting success?
|
This metric is measuring "grit" and grit always happens when the other team has the puck (except the rare case of hitting a guy while skating with the puck, like young Ovechkin used to do). So, the issue you have is not with this metric, but rather with "grit" itself. And that's silly issue, because what you are saying is essentially that it's useless to have ANY metric that is designed to measure what players are doing when the other team has the puck.
|
|
|
08-25-2013, 09:51 AM
|
#53
|
Appealing my suspension
Join Date: Sep 2002
Location: Just outside Enemy Lines
|
would
Quote:
Originally Posted by Jay Random
It's an omnium gatherum of questionable numbers that are only questionably related to each other, all added up without any scaling or weighting factor. (Are 1 hit, 1 blocked shot, and 1 takeaway all equally valuable in winning hockey games? I have reason to doubt it.)
If ricardodw (or anyone else) can demonstrate a strong correlation between his composite number and goal-scoring or goal prevention, then we can start to talk about how meaningful it is. As it stands, it's just his personal taste in players (basically, 'I like tough guys') dressed up in the form of an arbitrary number.
|
Yeah that would be my issue here. Sarich destroying Marleau counts the same as Butler bumping into Havlat. Much like Jackman winning the puck on the boards so he can fire it down the ice counts the as Datsyuk jumping a defenceman at his own blue line for a goal. Big impact plays are the ones that make a difference in games. If those could accurately be recorded and factored...than there might be something here. All those Penguins bumping Chara likely had as much impact as one Chara or Lucic hit.
__________________
"Some guys like old balls"
Patriots QB Tom Brady
Last edited by Sylvanfan; 08-25-2013 at 09:56 AM.
|
|
|
The Following User Says Thank You to Sylvanfan For This Useful Post:
|
|
08-25-2013, 12:26 PM
|
#54
|
Acerbic Cyberbully
Join Date: Aug 2003
Location: back in Chilliwack
|
Quote:
Originally Posted by Itse
I don't think that's fair at all. Imperfections can be argued, but it's clearly an attempt at finding a somewhat objective combined number.
|
But it is "objective" only insofar as ricardodw has selected those events that he has interpreted as indications of grit. I believe that his selection is arbitrary and impressionistic, and quite simply boils dow to this: When he watches hockey games "gritty players" are those players who hit, block shots, and are the beneficiaries of more take-aways than they are victimised for giveaways. It is a metric built on a handful of one observer's favourite qualities in a player to the exclusion of virtually everything else.
|
|
|
The Following 3 Users Say Thank You to Textcritic For This Useful Post:
|
|
08-25-2013, 02:53 PM
|
#55
|
Franchise Player
|
Quote:
Originally Posted by Textcritic
But it is "objective" only insofar as ricardodw has selected those events that he has interpreted as indications of grit. I believe that his selection is arbitrary and impressionistic, and quite simply boils dow to this: When he watches hockey games "gritty players" are those players who hit, block shots, and are the beneficiaries of more take-aways than they are victimised for giveaways. It is a metric built on a handful of one observer's favourite qualities in a player to the exclusion of virtually everything else.
|
I would like to be able to use dump ins from centre and going for a line change with 30 seconds into a 50 second shift .... but I just don't see that stat.
I would like to use puck battles lost Or non-contested when they could be.
I would like to see a stat on players circling away after their goalie has been run.
If my selection of these statistics are "arbitrary and impressionistic" then you must have some in mind to statistically evaluate the toughness or more specifically the softness of individual players.
I don't mind soft players and every team has a few and usually they are extremely skilled. The trouble with these players is that if they are not controlling the play AND scoring the goals they are not contributing to a team's success.
I think that I have stated my theory that there are no successful teams that have more than 3 soft players getting significant ice-time.
No matter how many Jackman's and Begin and even McGrattons you have playing gritty in their 8-10 minutes the overall lack of physicality of 3+ soft players hand-cuffs the teams ability to compete and win games.
The problem is that when I label Hudler as a soft player there are poster's that admire his ability to go into the tough areas. This is mostly anecdotal evidence... He went in the slot a few times and people were amazed that such a small player would venture in there.
When I said that 2011-12 Stempniak was soft people posted the greatest hits from his past. He did have the size and ability to move out of the soft category when he played like he did in 2012-13. Similarly with Stajan and Backlund.
I have never stated that Chicago wins because they are more physical than other teams.... contrary they as a team are not..... However they just have 2 soft players on their roster.... Kane and Keith who are off the charts skillful.
A great part of my concern was that when Gaudreau declares he wants to play in the NHL he would become the 4th or 5th soft player on the Flames roster.
My grit analaysis has changed my attitude.... The Flames have jettisoned 3 of their softest 5 players from last season.... Bouwmeester, Tanguay and Cervenka.
All that is left is Cammalleri and Hudler.
The statistics say that Brodie, Butler, Baertschi and Backlund are not as soft as they appeared from watching them playing on a too-soft team.
There will be room for Gaudreau and even Granlund if they show that they have the extreme skills that soft player should be expected to have. Byron never had a chance with the Flames NHL roster loaded with soft skilled guys...... Maybe this year.
Vancouver fans will say that the Sedins are not soft....... they are .... the Canucks understood this and have been successful but can't have Mason Raymond, Roy in the lineup along with the Sedins. ... It led to playoff disaster this year.
It seems that teams are using some sort of grit test when going after these small players..... It is not a question if Mason Raymond or Kyle Wellwood has top-6 skills..... it becomes do they have more skills than our softest 3 players.
Hodgson has way more skills than Kassian...... Buffalo was an unsuccessful team as they already had Vanek, Gerbe and Ennis... To get back down to 3 softies Buffalo bought out Gerbe.
Last edited by ricardodw; 08-25-2013 at 02:57 PM.
|
|
|
08-25-2013, 06:00 PM
|
#56
|
Franchise Player
|
Quote:
Originally Posted by Pointman
This metric is measuring "grit" and grit always happens when the other team has the puck (except the rare case of hitting a guy while skating with the puck, like young Ovechkin used to do). So, the issue you have is not with this metric, but rather with "grit" itself. And that's silly issue, because what you are saying is essentially that it's useless to have ANY metric that is designed to measure what players are doing when the other team has the puck.
|
The issue is not that it happens when you haven't got the puck. The issue is that it happens most often when your team is so bad that you NEVER have the puck. If you have a game between two teams with equally 'gritty' players, and one of them controls the puck for 36 minutes and the other for 24 minutes, then the team that only has the puck for 24 minutes will seem 1.5 times as gritty as the other. To that extent, a high 'grit' score is a consequence of failure. And if it's measuring failure, it isn't measuring grit.
A metric that measures one thing is good. A metric that measures two things, and gives no way of distinguishing between them, is worse than useless.
Last edited by Jay Random; 08-25-2013 at 06:05 PM.
|
|
|
The Following User Says Thank You to Jay Random For This Useful Post:
|
|
08-25-2013, 06:04 PM
|
#57
|
Franchise Player
|
Quote:
Originally Posted by SebC
Agreed on all counts. However, if a "weighted toughness winning factor" were to exist (say, a goals equivalency for hits, blocks, takeways, and giveaways), then the unweighted equivalent would reflect the weighted version to some extent and thus have value.
|
I can agree with this. However, ricardodw hasn't even begun to do the statistical legwork required. He hasn't established that there is any equivalency between hits, blocks, takeaways, or giveaways and goals for or against. I can't even say he's confusing correlation with causation — because he hasn't even shown the correlation. The scaling factors would arise naturally, if he first proved that there is something to scale.
|
|
|
08-25-2013, 10:17 PM
|
#58
|
Franchise Player
|
Quote:
Originally Posted by Enoch Root
But if more giveaways are awarded in Edmonton, shouldn't Oiler players also benefit from getting more takeaway stats, offsetting the difference?
Or are you suggesting that the statistician is biased against the local boys?
|
I'm not sure if the same guy awards giveaways takeaways, but one doesn't always mean there is another. Takeaways seem to be awarded "normally".
No bias. Just that the guy hands out buttloads of giveaways for both teams. If you look through the stats you will see similar stuff around the league. One team's guy might give out a hit every time guys brush shoulders together in the corner. These stats are so subjective and poorly kept, they aren't all that useful IMO.
|
|
|
The Following 3 Users Say Thank You to Oil Stain For This Useful Post:
|
|
08-26-2013, 09:16 AM
|
#59
|
Franchise Player
|
Quote:
Originally Posted by Oil Stain
I'm not sure if the same guy awards giveaways takeaways, but one doesn't always mean there is another. Takeaways seem to be awarded "normally".
No bias. Just that the guy hands out buttloads of giveaways for both teams. If you look through the stats you will see similar stuff around the league. One team's guy might give out a hit every time guys brush shoulders together in the corner. These stats are so subjective and poorly kept, they aren't all that useful IMO.
|
There are definitely differences from rink to rink.
I like the suggestion of only keeping the stats for road teams - that would normalize things quite a bit.
But at the end of the day, they are secondary stats and fans should be aware that there are discrepancies in the administration of them.
|
|
|
08-26-2013, 09:26 AM
|
#60
|
Franchise Player
Join Date: May 2004
Location: Helsinki, Finland
|
If you used ratios like hits for / against like you do with giveaways / takeaways, instead of just the positive numbers, the problem of differences between rinks should mostly go away.
Generally speaking, stats like these are never going to be perfect, but as long as you know the weaknesses, they can still provide some information value.
You're never going to come up with one stat to tell the definitive story of how good a team is (outside of points).
|
|
|
Posting Rules
|
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts
HTML code is Off
|
|
|
All times are GMT -6. The time now is 08:12 PM.
|
|