HFBoards

Go Back   HFBoards > NHL Eastern Conference > Atlantic Division > Montreal Canadiens
Mobile Hockey's Future Become a Sponsor Site Rules Support Forum vBookie Page 2
Notices

Sabremetrics/Microstats in Hockey

Reply
 
Thread Tools
Old
04-04-2012, 10:49 PM
  #176
Talks to Goalposts
Registered User
 
Join Date: Apr 2011
Location: Edmonton
Country: Canada
Posts: 3,610
vCash: 500
Quote:
Originally Posted by Lafleurs Guy View Post
Can you do us a favour? Can you post some of the analytic stats with definitions here? And dumb it down for us. I've seen some out there and found it interesting but I'd like to see simple definitions and your opinion on how effective each one is.
Getting started on this.


Shot attempts: Shots + Shots attempted blocked + Shot attempted missed.

Corsi: Shot attempts for versus shot attempts against for a particular game situation (i.e. 5 on 5, 5 on 4, 4 on 4 etc). Usually expressed on a per 60 minutes of playing time basis or as a percentage with 50% being even. Originally devised as a measure of the amount of work a goaltender would have to per game. It was discovered that it correlated extremely strongly with amount of time spent in each zone when the NHL tracked that information and that on a team basis it was the strongest predictor of future wins for a team (better than goal differential which was better than current record).

Because the time of possesion and zone time stats aren't tracked its seen as a proxy for puck possession and territorial control. One of its chief advantages is that it tracks so many events so its easy to get a large sample size to work with. The other is that it is one of the most stable aspects of the game (which is probably related to the large sample size). Corsi for a team/player is very good at predicting future Corsi while things like shooting/save% don't predict well. As such 5 on 5 corsi is a very good predictor of future 5 on 5 goals +/-.

Fenwick: Corsi's brother with blocked shots ignored. This gives somewhat better information at the cost of fewer events. Removing blocked shots probably allows it to better ***** defensive/offensive zone play because its has best correlation to "scoring chance" rates.

Scoring Chances: There is some quibiling over what the exact definition of a scoring chance should be but the basic definition is an unblocked shot attempt from within the "homeplate area" consisting of the defensive zone contained by a line from the goalpost to the faceoff dot with another to the ringette line using the ringette line as the top of the area. Some allow for well screened point shots as chances and some do not. Non-official stat tracked by people other than the NHL. Scoring chance data are usually seen as the best for micro-stat analysis (and most often what NHL coaches themselves seem track from information publicly availible) but it isn't publicly available for all teams currently. Oliver of http://enattendantlesnordiques.blogspot.ca/ tracks chances for Montreal.

Scoring chance differentials typically track very well with Fenwick/Corsi and track well with future situation goal differentials.

Score tied Corsi/Fenwick: One of the discoveries on the behavior of NHL teams is the effect of the score on how shot differentials. Teams with the league tend to be outshot, teams behind tend to outshoot. This is the answer the the paradox that teams that outshoot in general tend to win the most games but who had more or less shots in a particular game doesn't line up well with who won. Score tied Corsi/Fenwick is generally the metric most used to describe a team's 5 on 5 ability. More recently people have started to instead use a score situation adjusted shot rate which is probably a slight improvement.

Team shooting percentage: Basically team goals divided by team shots. A team being higher or lower than league average in this is an excellent sign that their are going to be better/worse in the future because there is very little evidence that a team can control their 5 on 5 shooting percentage in the long run.

On a team level its very rare to beat out your score tied 5 on 5 Corsi or do worse than it. Those that have mostly accomplish this by superior goaltending (save percentage) although Vancouver has done better offensively than their shot rates by virtue of their powerplay conversion rate. Similarly Columbus has done worse than average on the powerplay resulting in worse goals for their shots. 5 on 5 though, it seems that in the modern NHL every team comes back to the mean.

The caveat modern NHL is crucial because teams like the 80's Oilers produced their offense on percentages rather than shot rates. But there doesn't seem to be any team that can manage it these days.

Personal shooting percentage: Percentage of shots taken by a player that become goals. Extremely volatile but converges on a player's career ability. This is extremely volatile but on the long term will converge with a player's career talent. From how it progresses over a career in the NHL and the degree it translates from the AHL to the NHL (almost entirely), in my opinion its less of a talent as it is a tendency which reflects a style of play.

Team on ice shooting percentage: The percentage of shots by his team while a player is on that become goals. Much like personal shooting percentage is heavily volatile but converges on a mean, in this case usually league average. Almost all players tend to be in a very narrow band of 7-8.5% team on-ice shooting with changes year to year being fairly random. Their seems to be a very small cadre of players that do better to average year to year. Usually these are player that play consistently on very good lines such as Sedin-Sedin and Stamkos-St. Louis. There are so few of them its debatable if the ability to be significantly better than average in this regard is real or just the mark of their being hundreds of players and a few are going to be better than average over multiple years by shear population dynamics.

Talks to Goalposts is online now   Reply With Quote
Old
04-04-2012, 10:51 PM
  #177
MathMan
Registered User
 
Join Date: Jan 2006
Posts: 17,025
vCash: 500
Quote:
Originally Posted by Lafleurs Guy View Post
The reason why Price isn't blamed is because the club leaves him to twist in the wind once we get a lead. It's that simple. We get the lead, the other team adjusts and our team can't handle it.
Except that's not happening. Montreal does a good job of limiting chances against. There's more shots against but this happens with every team in the league. They allow more shots but proportionately fewer goals.

But Montreal has been allowing the goals. Price isn't getting the flack because he's gone from Scapegoat to "the only thing keeping the Habs from sucking utterly (tm)", even though neither characterization is close to true, so the defense gets it, even if the data tells us they don't particularly deserve it.

Quote:
Originally Posted by Lafleurs Guy View Post
Well, if Phil Kessel doesn't go to the net against the Bruins, I'd say that's proof enough. It certainly does have an effect.
Are you sure Kessel goes to the net less against Boston than other teams? Did you count and compare, or is this just a general impression? Do you have data, or just anecdote? Is this a real thing or just something you think?

Quote:
Originally Posted by Lafleurs Guy View Post
For the purposes of this conversation... it doesn't matter. The Habs had Savard, Dryden, Lafleur, Robinson... They still count as a club packed with stars. Also notice that teams packed with superstars won multiple cups.
Well duh. Superstars are good players and teams packed with multiple such players are going to be good. You can't build teams like that anymore.

The problem isn't the assumption that good players help win cups, the problem is with the assumption that they're necessary, and especially, that they have a special quality that makes them so -- like how Pronger was mostly responsible for going to the finals three times. I think he just happened to be a very good player on three teams that happened to make the finals.

My question about Gainey is meaningful because it falls into the whole "definition of a superstar" thing. I'm trying to find out if it's all about offense for you.

MathMan is offline   Reply With Quote
Old
04-05-2012, 02:38 AM
  #178
Miller Time
Registered User
 
Miller Time's Avatar
 
Join Date: Sep 2004
Posts: 8,184
vCash: 500
Quote:
Originally Posted by Mathletic View Post

Funny you mention the classical music and wine tasting examples. In the book Super Crunchers there's the story of an analyst who built models to predict the price of wine simply based on data. Like quantity of rain and whatnot. Been a while since I read the book but the guy basically beat every wine taster and elitist of the milieu at predicting how much a wine would sell for.
sorry, but you lost me here...

how is that related to the example I gave?

predicting the price that a wine will sell for has nothing to do with quality or taste experience of the wine.

not to get into wine details, but as with any luxury type item, price has very little to do with actual quality. Both are subjective, but price is driven far more by branding/reputation & the sad state of "ranking/scoring" which both the average & the pretentious wine drinker relies on. Further, the reality of China's growing "appetite" for first growth bordeaux has completely thrown the pricing of wine into a ludicrous, albeit highlight predicable(for now), state. Sort of like predicting wether or not the latest Rhianna single is going to hit top-5 on the charts... the herd is predictable... but the herd is about as far removed from the true expert as one can get.


Quote:
Originally Posted by Mathletic View Post
Also, although you may not be aware of it but musical analysis actually started with Pythagoras. A greek mathematician, known for the Pythagorean Theorem. Musical analysis at its root is a study of fractions and proportions. Obviously, music is taught by words and not maths with words like triads, augmented triad, harmony and what have you but deep down it's still maths.
is music math or is math music... that's for another debate...

regardless, i don't see how your comment address what I'm talking about?

math is very much behind what makes Player 'A' get to the puck faster than Player 'B' (or shoot harder, or pass more accurately, and so on and so on).

sports, at the physical level, is about human movement, bodies in motion operate on a plane/field of play, angles/velocity/acceleration et. are all measurable components that will explain "WHAT" is happening.

I've never stated otherwise.

but find me the mathematical formula or statistical breakdown that explains why a player chokes under pressure, why a player struggles for one coach but shines for another, why one player shines in a high-stress envrionment like montreal whereas another wilts...

find me the stat package that will predict which junior star will get intimidated going into the corners vs chara and the mathematical formula that will predict which average junior player will break out and become the next Martin St.Louis...

call it intangibles if you want, but from my own experience with sport, I've directly seen how easy it is for an expert observer to recognize successful behaviors/intentions/actions in a junior athlete despite otherwise lackluster "measurable" indicators.

great scouts can watch a player play and see past "what" he does & understand "why" & "how" he's doing it (or not doing it), and that's the kind of insight that mathematical formulas don't cover, at least not yet.

Quote:
Originally Posted by Mathletic View Post
None of Pythagoras nor the guy from Super Crunchers, nor was Bill James a great analyst in the discipline each studied.

Today mathematicians build models to help golfers make better putts and read greens. None of them are any good at golf yet they can teach a young golfer how to putt in 15 minutes more than that golfer would learn by himself doing all sorts of mystical abracadabras like that guy for I don't know how many years.
do you not see how far removed a golf swing is from the constant action/adjustments that a fluid sport like hockey represents?

granted, during the actual swing there is tons of micro-adjustments going on, but all of them occur without any outside human influence.

Golf, even more so than baseball, is OBVIOUSLY a sport where mathematical formulas work very very well at both improving & predicting performance... the variables are contained and minimal.

Hockey is dynamic, always changing. A player will never in his career shoot exactly the same way as he just did, never. The mechanics will never be as close to a previous shot as a golfer's is to his previous swing.

and it's this huge variability, this immense amount of unpredictability (and the mental/emotional ability to adjust to it) that an expert observer provides the best "filter" for...

in a way, it's the ability of the human observer to comprehend and "compute" unpredictability (and thus seek answers for) that give it the advantage over the mathematical model which implicitly "expects" predictability.

not sure if i explained that last bit well, but basically it's the awarness of the "unkown unknown", and being sensitive to it, that makes the difference.

Miller Time is offline   Reply With Quote
Old
04-05-2012, 03:14 AM
  #179
Mathletic
Registered User
 
Mathletic's Avatar
 
Join Date: Feb 2002
Location: St-Augustin, Québec
Country: Canada
Posts: 11,872
vCash: 500
Quote:
Originally Posted by Miller Time View Post
sorry, but you lost me here...

how is that related to the example I gave?

predicting the price that a wine will sell for has nothing to do with quality or taste experience of the wine.

not to get into wine details, but as with any luxury type item, price has very little to do with actual quality. Both are subjective, but price is driven far more by branding/reputation & the sad state of "ranking/scoring" which both the average & the pretentious wine drinker relies on. Further, the reality of China's growing "appetite" for first growth bordeaux has completely thrown the pricing of wine into a ludicrous, albeit highlight predicable(for now), state. Sort of like predicting wether or not the latest Rhianna single is going to hit top-5 on the charts... the herd is predictable... but the herd is about as far removed from the true expert as one can get.
you're right, 2 different things. I couldn't remember what the chapter was about in the book, so I got a little refresh.

http://www.oppapers.com/essays/Super...ummarys/956346

* Orley Ashenfelter- Uses statistics to extract hidden info from large datasets

* Orley’s Wine Predictions

* What really got him in trouble is crunching numbers to assess the quality of Bordeaux wines. He found that low levels of harvest rain/high avg. summer temps produce the greatest wines

Reduced his wine theory to a formula; he can predict the general quality of any vintage by plugging weather statistics for a given year into formula. NOT well received by traditional wine critics

BUT, unlike the traditional critics who have to wait months to taste the wines, Orley could compute his predictions right away (he published them in a newsletter called Liquid Assets)

1990 New York Times Article on his prediction machine greatly intensified traditional critic’s animosity towards Orley-not only did his ’86 prediction contradict the critic’s (Parker), but he predicted ‘89’s wine to be “wine of the century”.

1990’s wine was predicted to be even better than ’89-both predictions proved




Quote:
Originally Posted by Miller Time View Post
is music math or is math music... that's for another debate...

regardless, i don't see how your comment address what I'm talking about?

math is very much behind what makes Player 'A' get to the puck faster than Player 'B' (or shoot harder, or pass more accurately, and so on and so on).

sports, at the physical level, is about human movement, bodies in motion operate on a plane/field of play, angles/velocity/acceleration et. are all measurable components that will explain "WHAT" is happening.

I've never stated otherwise.
The tools used to analyze music are mathematical in the first place. Not only that but "musical temperament", that is why the C is tuned at a certain frequence and so on, is based on maths. Musical analysis began with a mathematician. What makes an harmony or melody work is the arrangement or proportions between notes. You may use fancy words to describe the harmony but at its basics it's explained by mathematics.

Also, you can even assess the quality of a performance by a musician through maths. Such a model was developped at McGill's university by I don't remember whom.

There are even models that can predict whether or not a song will be a hit. There was an interview going on youtube with Malcolm Gladwell and some other guy from a company called "blue intelligence" or something like it. It showed he could analyze performance, predict whether the song would be a hit and so on in various styles. The guy was also featured in a documentary running on PBS with Professor Daniel Levitin.

found a link

http://www.mikemccready.com/2007/06/...er-conference/

apparently the company is called Music Xray ... I had in mind something about blue intelligence but anyways


Quote:
Originally Posted by Miller Time View Post
but find me the mathematical formula or statistical breakdown that explains why a player chokes under pressure, why a player struggles for one coach but shines for another, why one player shines in a high-stress envrionment like montreal whereas another wilts...

find me the stat package that will predict which junior star will get intimidated going into the corners vs chara and the mathematical formula that will predict which average junior player will break out and become the next Martin St.Louis...

call it intangibles if you want, but from my own experience with sport, I've directly seen how easy it is for an expert observer to recognize successful behaviors/intentions/actions in a junior athlete despite otherwise lackluster "measurable" indicators.

great scouts can watch a player play and see past "what" he does & understand "why" & "how" he's doing it (or not doing it), and that's the kind of insight that mathematical formulas don't cover, at least not yet.



do you not see how far removed a golf swing is from the constant action/adjustments that a fluid sport like hockey represents?

granted, during the actual swing there is tons of micro-adjustments going on, but all of them occur without any outside human influence.

Golf, even more so than baseball, is OBVIOUSLY a sport where mathematical formulas work very very well at both improving & predicting performance... the variables are contained and minimal.

Hockey is dynamic, always changing. A player will never in his career shoot exactly the same way as he just did, never. The mechanics will never be as close to a previous shot as a golfer's is to his previous swing.

and it's this huge variability, this immense amount of unpredictability (and the mental/emotional ability to adjust to it) that an expert observer provides the best "filter" for...

in a way, it's the ability of the human observer to comprehend and "compute" unpredictability (and thus seek answers for) that give it the advantage over the mathematical model which implicitly "expects" predictability.

not sure if i explained that last bit well, but basically it's the awarness of the "unkown unknown", and being sensitive to it, that makes the difference.
Like I said earlier, there are studies showing the effect of "clutchness". It probably doesn't even work you think it does. You want mathematical models to express reality as you see it or scouts see it. That's not the point of science. Stats analysis are there to present reality as precisely as possible. I'll make the analogy to fall in crime rate and abortion as described in Freakonomics. Police departments will always pat themselves in the back whenever there's a crime reduction in their county. Most often cases there are plenty of other explanations. In the book freakonomic they present such an alternative.

Book authors came to the conclusion that abortion plays a bigger role in the reduction of crime than policemen do. They did not prove that X guy was afraid of the police by mathematically proving that the guy had increased blood pressure when thinking of the police or what have you. Sometimes there are just different explanations to what you think is happening.

http://en.wikipedia.org/wiki/Legaliz...d_crime_effect

In hockey and sports there are plenty of different explanations, you just have to approach analytics with an open-mind. Like the birthday effect in hockey.

There are models that predict how well a player from an inferior league will do in the upper league. They are not deterministic obviously. But they will beat the NHL index or NBA index if we may call them that way.

Also, I'm not sure how Martin St. Louis proves your point. All he had going for himself was production. So-called experts never gave him a shot to start with.

Not to quote the "Moneyball" movie but I'll basically say the same as Billy Beane talking to his head scout. "You have no clue how a guy will do in the future". Stats analysis shows that the n'1 overall pick of any draft is only 55% of the time better than the 2nd overall pick. There's tons of luck that goes into drafting.

The point of the maths in golf is not to help on the swing. They exist to help you read greens. In case you never played golf. Greens change a lot. There's an infinity of situations you can find yourself in. Golfers spend their lifetime learning how to read greens. It has always been taught as an art that only innate skills and practice can help you with. The maths have helped golfers generalizing them in order to find the better trajectories as they play on new greens each and everytime.

The same goes with hockey, sure the game changes but as in anything you can break it into building blocks.

Not sure I understand the last part. If it's so unpredictable that we can't analyze it and break it down into parts, then there's surely no way anyone could analyze it even an expert.

If hockey is so chaotic, there's no way we could predict anything. It would be the same as predicting a hurricane based off the butterfly effect.

Also, I hope you realize GM's give contract almost entirely based on goals, assists, time-on-ice and +/-. You can run a regression and you'll predict salaries based on those variables. Stats analysis will tell you that these stats are not even the best predictors of future performances. You can do the same for basketball. GM's give contracts based on your average boxscore stat. We're far from the guru analytics that only an über-expert could make.


Last edited by Mathletic: 04-05-2012 at 04:26 AM.
Mathletic is online now   Reply With Quote
Old
04-05-2012, 12:24 PM
  #180
Lafleurs Guy
Registered User
 
Lafleurs Guy's Avatar
 
Join Date: Jul 2007
Posts: 20,998
vCash: 500
Quote:
Originally Posted by Talks to Goalposts View Post
Getting started on this.
Cool, I'll have a look later. Thanks.

Quote:
Originally Posted by MathMan View Post
Except that's not happening. Montreal does a good job of limiting chances against. There's more shots against but this happens with every team in the league. They allow more shots but proportionately fewer goals.

But Montreal has been allowing the goals. Price isn't getting the flack because he's gone from Scapegoat to "the only thing keeping the Habs from sucking utterly (tm)", even though neither characterization is close to true, so the defense gets it, even if the data tells us they don't particularly deserve it.
We have the gametime threads here man... happened all the time. We get the lead, the other team wakes up and is all over us. We collapse and they win. You think this is Price's fault, I've got to disagree with you. Do you have the numbers broken down by period? It matters when those shots come too because if you face 40 shots in a game but all 40 come in the last ten minutes, you're probably going to let in more goals than you otherwise might right? Doesn't that make sense?


Quote:
Originally Posted by MathMan View Post
Are you sure Kessel goes to the net less against Boston than other teams? Did you count and compare, or is this just a general impression? Do you have data, or just anecdote? Is this a real thing or just something you think?
I've seen it myself and everytime the Leafs play the Bruins it's mentioned by the announcers. The guy doesn't go near the net when he plays them. Feel free to check your stats cause I sure don't have them.


Quote:
Originally Posted by MathMan View Post
Well duh. Superstars are good players and teams packed with multiple such players are going to be good.
So then stop arguing with me dude!
Quote:
Originally Posted by MathMan View Post
You can't build teams like that anymore.
Funny thing, back when they were putting Pittsburgh together everyone said the cap would keep them from keeping that core together... didn't happen. They're still there and there's no sign of them breaking up. So what you're saying here is BS. You absolutely can build teams and keep that core together. You may not win the cup as much as teams back in the 70s did because there are more teams out there and the path is harder but you can still keep those teams together.

Quote:
Originally Posted by MathMan View Post
The problem isn't the assumption that good players help win cups, the problem is with the assumption that they're necessary, and especially, that they have a special quality that makes them so -- like how Pronger was mostly responsible for going to the finals three times. I think he just happened to be a very good player on three teams that happened to make the finals.
So the Oilers make the finals without Pronger? Are you for real? Gretzky just happened to be with the Kings when they suddenly had a 180 degree turnaround (that just so happens to coincide with his arrival)?

Wow. I'm actually at a loss for words here. I have no idea what to say to you. If you actually believe that those guys just happened to be there... those teams might not have even made the playoffs without those guys.

And you keep referring to 'magic' or 'special quality'... dude, it's not magic. Better players give you an advantage over other teams. Not sure how this simple concept is eluding you. The BEST players will help you more than the good ones will. You can try to make it up in the aggregate (ie. depth) but without superstars there, clubs haven't really been able to win. We could say that maybe Carolina did it and that's about it.

Nobody is claiming that a superstar can win it by themselves. But it's hard to win when you don't have them.
Quote:
Originally Posted by MathMan View Post
My question about Gainey is meaningful because it falls into the whole "definition of a superstar" thing. I'm trying to find out if it's all about offense for you.
Personally, I think our HOF is totally screwed up. The reason I don't mention Gainey here is because some may debate whether or not he was a superstar. You'll notice above that I draw red circles around guys like Newendyke, Hossa, Staal and St. Louis because I'm trying to hold a high standard to make my point to you. Those clubs didn't just have stars on their teams, they had SUPERSTARS. Yzerman is a superstar no matter how you slice it. Ditto with Gretzky, Lafleur, Bossy, Lemieux etc... that's why I don't want to get into a Bob Gainey type debate on this.

As for my opinion on Gainey, he was a legit HOF player. I actually think Carbonneau should be considered for the HOF too as he was far and away the best defensive forward of his generation. Do I think of them as 'superstars' though? No. I think you need these kinds of players on your team to go along with superstars if you want to win though. You can't just have a bunch of Guy Lafleurs and expect to win. Tampa is a good example of this now... a select group of great forwards but no goaltending and weak D. Until they get some depth to surround St. Louis and Stamkos with, they aren't going to win. Ditto with Edmonton.

I think Edmonton has three potential superstars on that club. But they have no defense and little depth. Their management are a bunch of idiots so I'm not sure where they're going. BUT (and this is a big BUT) they have done the hard part. They have the superstars. Getting multiple superstars isn't easy but once you have them its easier to work backwards. It's a lot easier to get a Rick Green, Brian Skrudland or Craig Ludwig than it is to find a Patrick Roy. Nobody is going to trade you a Patrick Roy unless something weird happens (and it did). But you can get a Rick Green at a fair price from a club going nowhere. For Edmonton, having those three (potential anyway) superstars in your lineup is huge. If they could get a decent GM in place over there they could set themselves up well for a long time.


Last edited by Lafleurs Guy: 04-05-2012 at 12:35 PM.
Lafleurs Guy is offline   Reply With Quote
Old
04-05-2012, 01:01 PM
  #181
MathMan
Registered User
 
Join Date: Jan 2006
Posts: 17,025
vCash: 500
Quote:
Originally Posted by Lafleurs Guy View Post
We have the gametime threads here man... happened all the time. We get the lead, the other team wakes up and is all over us. We collapse and they win. You think this is Price's fault, I've got to disagree with you.
No no no no no. I don't think it's Price's fault, myself. I'm just surprised he hasn't been blamed.

Quote:
Originally Posted by Lafleurs Guy View Post
Do you have the numbers broken down by period? It matters when those shots come too because if you face 40 shots in a game but all 40 come in the last ten minutes, you're probably going to let in more goals than you otherwise might right? Doesn't that make sense?
Actually score effects work like this: if you're ahead, you tend to allow more shots, but each shot is less likely to go in. And if you're behind, you tend to allow fewer shots, but each is more likely to go in. This is a trend observed league-wide, it's not just the Habs.

Intutively it makes a certain amount of sense: teams that are behind press more, so they get more shots, against a team that collapses against their goalie. And if they are behind, when they give up a shot it more often is an odd-man rush due to over-aggressive play.

Quote:
Originally Posted by Lafleurs Guy View Post
I've seen it myself and everytime the Leafs play the Bruins it's mentioned by the announcers. The guy doesn't go near the net when he plays them. Feel free to check your stats cause I sure don't have them.
I don't either, but announcers saying something is hardly proof that it actually happens. Those guys create engaging narratives rather than give dispassionate analysis, and really, in a way that's their job. Dispassionate analysis makes for crummy TV and it's really hard to do on the fly.

Quote:
Originally Posted by Lafleurs Guy View Post
Funny thing, back when they were putting Pittsburgh together everyone said the cap would keep them from keeping that core together... didn't happen. They're still there and there's no sign of them breaking up.
They only have a couple of superstars though, namely Crosby and Malkin. And surrounded them with good depth and good coaching. They aren't the 70s Canadiens.

Quote:
Originally Posted by Lafleurs Guy View Post
So the Oilers make the finals without Pronger? Are you for real? Gretzky just happened to be with the Kings when they suddenly had a 180 degree turnaround (that just so happens to coincide with his arrival)?
Do I think they helped get there? Of course. Do I think that they had some special quality that made their team better in a way a player of equal ability would not have? Not for a moment. Do I think Gretzky's presence singlehandedly turned a 68-point team to a 92-point one? No, I don't think it was the only thing. Do I think he improved that club? Of course he did.

Do I think Pronger's presence was the main cause of three Cup Finals? No, because I don't think any of Pronger's teams were among the two best clubs in the NHL those years. They were pretty good clubs (well, not the Oilers) that did well in the playoffs those years. Swap Pronger with Lidstrom on all these clubs and they all get better; their odds of making the Cup Finals wouldn't have been worse. Beyond his superior hockey ability, Pronger doesn't have a special quality that makes his teams more likely to reach the Cup finals any more than Ty Conklin does.

It's not the notion that superstars are good players that I have trouble with. Obviously if you remove the best player from a team that team is much worse. It's the notion that they have some sort of "special quality" beyond being better at hockey.

It's like you think that superstars have some sort of mystique. When they're really just especially good hockey players.

Quote:
Originally Posted by Lafleurs Guy View Post
As for my opinion on Gainey, he was a legit HOF player. I actually think Carbonneau should be considered for the HOF too as he was far and away the best defensive forward of his generation. Do I think of them as 'superstars' though? No.
Why wouldn't they be superstars, if they were good enough to be Hall of Famers?

Is being a superstar all about offense for a forward? What about defensemen? Are you defining superstar forwards solely in terms of point production? Does defense matter at all?

Is Rick Nash a superstar? Ryan Getzlaf? Corey Perry? Ryan Kesler?

This has been veering off-topic for a while, but I've always been genuinely curious, because I've never quite grokked how you define a superstar.

MathMan is offline   Reply With Quote
Old
04-05-2012, 02:53 PM
  #182
Lafleurs Guy
Registered User
 
Lafleurs Guy's Avatar
 
Join Date: Jul 2007
Posts: 20,998
vCash: 500
Quote:
Originally Posted by MathMan View Post
No no no no no. I don't think it's Price's fault, myself. I'm just surprised he hasn't been blamed.

Actually score effects work like this: if you're ahead, you tend to allow more shots, but each shot is less likely to go in. And if you're behind, you tend to allow fewer shots, but each is more likely to go in. This is a trend observed league-wide, it's not just the Habs.

Intutively it makes a certain amount of sense: teams that are behind press more, so they get more shots, against a team that collapses against their goalie. And if they are behind, when they give up a shot it more often is an odd-man rush due to over-aggressive play.
Okay.
Quote:
Originally Posted by MathMan View Post
I don't either, but announcers saying something is hardly proof that it actually happens. Those guys create engaging narratives rather than give dispassionate analysis, and really, in a way that's their job. Dispassionate analysis makes for crummy TV and it's really hard to do on the fly.
From the games I've seen that's how it's looked. And his lack of success against them sure seems to correspond with it. I don't know what the analytical stats would say... but I suspect it would reinforce what I said.
Quote:
Originally Posted by MathMan View Post
They only have a couple of superstars though, namely Crosby and Malkin. And surrounded them with good depth and good coaching. They aren't the 70s Canadiens.
First, the Montreal Canadiens are the best team of all-time. Will we see that club again? Man, I don't know how you could. Ken Dryden just might be the best goalie of all-time and Guy Lafleur was the best offensive player in the game. Throw in a supporting cast of stars and it would be awfully tough to replecate in any era. Detroit came pretty close about ten years ago though with Yzerman, Fedorov, Lidstrom, Shanahan, Robitaille, Hasek, Hull... I think it would be tough to do again but certainly not impossible. That's not the standard I'm saying that cup teams have to have though. The Montreal Canadiens of the 70s are the best team of all time.

I try to limit examining cup winning teams to examples that are pre-2000. Some of these players are really young. Kris Letang looks like a stellar defenseman right now but he's been hurt a lot. I don't know if he's going to have a superstar career because it's just beginning. He could very well turn out to be another one though as he looks great. Chicago had Keith, Toews and Kane... very young in their careers but all are building great resumes with Toews being arguably the best. Vancouver's got at least three superstars now but those are vets. Edmonton has three potential superstars and my have a fourth on the way... who knows what they can put together with what they have over there? The cap is definitely a factor but some guys might be willing to play for less to win cups. We've seen it happen before.

Point is that most clubs have at least two superstars on them if not more. If you don't have at least one... good luck. Because your team without superstars on it is going to have to face at least three rounds with teams that probably do have at least one on them. If you get past OV, you face Crosby, get past him you face Pronger, get past him you face Lidstrom... not easy to do without a superstar of your own to help you win.
Quote:
Originally Posted by MathMan View Post
Do I think they helped get there? Of course. Do I think that they had some special quality that made their team better in a way a player of equal ability would not have? Not for a moment. Do I think Gretzky's presence singlehandedly turned a 68-point team to a 92-point one? No, I don't think it was the only thing. Do I think he improved that club? Of course he did.
Right, I get it you acknowledge that they were big parts of the team... let's not bother with what we both agree with because it's a waste of time.

Straight up... do you think the Kings make it to the finals without Gretzky? If Jimmy Carson remains with that club and continues putting up 100 points a year does that club make the finals?

Do you think the Oilers even make the playoffs (let alone the finals) without Pronger?
Quote:
Originally Posted by MathMan View Post
Do I think Pronger's presence was the main cause of three Cup Finals?
No, because I don't think any of Pronger's teams were among the two best clubs in the NHL those years. They were pretty good clubs (well, not the Oilers) that did well in the playoffs those years. Swap Pronger with Lidstrom on all these clubs and they all get better; their odds of making the Cup Finals wouldn't have been worse. Beyond his superior hockey ability, Pronger doesn't have a special quality that makes his teams more likely to reach the Cup finals any more than Ty Conklin does.
So then you'd trade Pronger for Conklin? Do you see how silly you're being here?

Of course Pronger has a special quality... it's called being one of the best players in the game.

Quote:
Originally Posted by MathMan View Post
It's not the notion that superstars are good players that I have trouble with. Obviously if you remove the best player from a team that team is much worse. It's the notion that they have some sort of "special quality" beyond being better at hockey.

It's like you think that superstars have some sort of mystique. When they're really just especially good hockey players.
Dude, they are superstars because they ARE better at hockey. There's no mystique and there's no magic. Better players just give you a better chance to win. What's the mystery here? We've seen it done with Patrick Roy.
Quote:
Originally Posted by MathMan View Post
Why wouldn't they be superstars, if they were good enough to be Hall of Famers?
Because the HOF allows players who aren't superstars. We've seen some pretty stark examples of them letting in lesser players due to personal relationships too. Cam Neely gets in on his 2nd try but Tim Kerr doesn't sniff the HOF? Adam Oates is still waiting? Doesnt' make any sense. How does Bernie Federko make it?

If you want to use the HOF as a standard you certainly can if you wish. I just choose not to because I think the Hall's standards for who they let in are too low and they are inconsistent as well.
Quote:
Originally Posted by MathMan View Post
Is being a superstar all about offense for a forward? What about defensemen? Are you defining superstar forwards solely in terms of point production? Does defense matter at all?

Is Rick Nash a superstar? Ryan Getzlaf? Corey Perry? Ryan Kesler?

This has been veering off-topic for a while, but I've always been genuinely curious, because I've never quite grokked how you define a superstar.
I think a superstar is a guy who's at the top echelon of the game. I don't try to supply a hard definition because hockey's a fluid game and there will always be debate. I think each player can be debated on via an individual basis and that's why I excluded the 'maybe a superstar' types from our conversation. I wanted to show you that even if I exclude it to the absolute elite players, cup winning teams STILL have them on their roster. Forget about Marian Hossa, forget about Joe Newendyke or Theo Fleury. I've also ruled out guys like Robinson who were past their prime when winning ('86.) Look at the players who've won:

Yzerman, Sakic, Messier, Lafleur, Bossy, MacInnis, Coffey, Belfour, Trottier, Gretzky, Lemieux, Jagr, Crosby, Hull, Malkin... Goes on and on man.

As for superstars today? Sedins, Crosby, OV, Stamkos, Lidstrom, Pronger, Toews, Chara, Iggy, Malkin... All those guys are modern superstars because year they consistently dominate the game. Some (Igninla, Pronger, Lidstrom) are fading but they have had superstar careers. Rising superstars who may get there or who might already be there? Corey Perry, Karlsson, Keith, Pieterangelo, Tavares... Great potential to get there and some may have already realized it. We have to wait and see if they can repeat their success or acheive their potential though. Karlsson is having a superstar type year but it's been one season.

The one thing that I think is interesting is the goalie position. There hasn't been the dominant goalies that we've seen in the past for long stretches. Lundqvist and Luongo are probably the only two superstars (along with the fading Brodeur) right now. Rinne is debatable, Quick and Price may get there but I think there's just the two guys right now.

Lafleurs Guy is offline   Reply With Quote
Old
04-05-2012, 03:31 PM
  #183
MathMan
Registered User
 
Join Date: Jan 2006
Posts: 17,025
vCash: 500
Quote:
Originally Posted by Lafleurs Guy View Post
I try to limit examining cup winning teams to examples that are pre-2000.
You know, that might be the bulk of the problem right there. That was pre-lockout, pre-salary cap, pre-new NHL, pre-dead puck era, pre-parity, pre-almost everything that matters to modern team-building. There's only so much you can apply from team-building in previous eras to the modern day.

Quote:
Originally Posted by Lafleurs Guy View Post
Do you think the Oilers even make the playoffs (let alone the finals) without Pronger? So then you'd trade Pronger for Conklin? Do you see how silly you're being here?
You're missing my point. You can't remove a team's best player, replace it with nothing and expect it to play at the same level, that is silly; but that doesn't mean that player has a special quality. Edmonton probably doesn't make the finals without Pronger, but they also don't make them without Roloson, without Pisani, without Horcoff, without Hemsky, or without Smyth. Heck you remove Spacek or Samsonov and they probably don't either.

On the other hand, you can replace Pronger and MAB with two first-pairing D-men that don't have the regard Pronger does and they're probably just as good if not better.

Besides, Edmonton was a Cinderella club. They really weren't all that good even with Pronger. And the Flyers making it to the finals arguably had more to do with Halak than with Pronger (as he spared them the need to face the East's two best clubs).

Quote:
Originally Posted by Lafleurs Guy View Post
Dude, they are superstars because they ARE better at hockey. There's no mystique and there's no magic.
Well, then maybe you need to reframe your argument, because you certainly come across as thinking that they have some quality that's virtually necessary. The more talent you have the better your odds are, obviously, but that doesn't mean that "having a superstar" is a prerequisite -- "having enough talent" is (on a practical level), and the latter does not necessarily require the former. Though it helps.

As for the definition of superstar, I think we have different criteria of player evaluation (personally I value two-way play/outscoring ability over raw offensive total, especially in the power-on-power NHL, and put no value on draft pedigree or reputation) so I doubt we'll ever agree on that. In general, I think raw offensive total is grossly overrated, while you seem to value it very highly (it seems to me, above all else). I do think it's significant that the last two Cup winners had two 30-goal men and one 70-point guy between them, however.

MathMan is offline   Reply With Quote
Old
04-05-2012, 03:37 PM
  #184
Goldthorpe
Meditating Guru
 
Join Date: Jan 2003
Location: Montreal
Posts: 4,305
vCash: 500
Quote:
Originally Posted by Lafleurs Guy View Post
Do you think the Oilers even make the playoffs (let alone the finals) without Pronger?
Maybe not. But say instead of 1 playoff of 4 series, the Oilers played 10. Each time, we set the clock back and all the 16 teams who made the playoff this year try again to win the cup, with games obviously having different outcomes. Out of ten, how many times do you think the Oilers makes the final even with Pronger playing (and hopefully not getting injured)?

Goldthorpe is offline   Reply With Quote
Old
04-05-2012, 04:13 PM
  #185
Lafleurs Guy
Registered User
 
Lafleurs Guy's Avatar
 
Join Date: Jul 2007
Posts: 20,998
vCash: 500
Quote:
Originally Posted by Talks to Goalposts View Post
Getting started on this.


Shot attempts: Shots + Shots attempted blocked + Shot attempted missed.

Corsi: Shot attempts for versus shot attempts against for a particular game situation (i.e. 5 on 5, 5 on 4, 4 on 4 etc). Usually expressed on a per 60 minutes of playing time basis or as a percentage with 50% being even. Originally devised as a measure of the amount of work a goaltender would have to per game. It was discovered that it correlated extremely strongly with amount of time spent in each zone when the NHL tracked that information and that on a team basis it was the strongest predictor of future wins for a team (better than goal differential which was better than current record).

Because the time of possesion and zone time stats aren't tracked its seen as a proxy for puck possession and territorial control. One of its chief advantages is that it tracks so many events so its easy to get a large sample size to work with. The other is that it is one of the most stable aspects of the game (which is probably related to the large sample size). Corsi for a team/player is very good at predicting future Corsi while things like shooting/save% don't predict well. As such 5 on 5 corsi is a very good predictor of future 5 on 5 goals +/-.

Fenwick: Corsi's brother with blocked shots ignored. This gives somewhat better information at the cost of fewer events. Removing blocked shots probably allows it to better ***** defensive/offensive zone play because its has best correlation to "scoring chance" rates.

Scoring Chances: There is some quibiling over what the exact definition of a scoring chance should be but the basic definition is an unblocked shot attempt from within the "homeplate area" consisting of the defensive zone contained by a line from the goalpost to the faceoff dot with another to the ringette line using the ringette line as the top of the area. Some allow for well screened point shots as chances and some do not. Non-official stat tracked by people other than the NHL. Scoring chance data are usually seen as the best for micro-stat analysis (and most often what NHL coaches themselves seem track from information publicly availible) but it isn't publicly available for all teams currently. Oliver of http://enattendantlesnordiques.blogspot.ca/ tracks chances for Montreal.

Scoring chance differentials typically track very well with Fenwick/Corsi and track well with future situation goal differentials.

Score tied Corsi/Fenwick: One of the discoveries on the behavior of NHL teams is the effect of the score on how shot differentials. Teams with the league tend to be outshot, teams behind tend to outshoot. This is the answer the the paradox that teams that outshoot in general tend to win the most games but who had more or less shots in a particular game doesn't line up well with who won. Score tied Corsi/Fenwick is generally the metric most used to describe a team's 5 on 5 ability. More recently people have started to instead use a score situation adjusted shot rate which is probably a slight improvement.

Team shooting percentage: Basically team goals divided by team shots. A team being higher or lower than league average in this is an excellent sign that their are going to be better/worse in the future because there is very little evidence that a team can control their 5 on 5 shooting percentage in the long run.

On a team level its very rare to beat out your score tied 5 on 5 Corsi or do worse than it. Those that have mostly accomplish this by superior goaltending (save percentage) although Vancouver has done better offensively than their shot rates by virtue of their powerplay conversion rate. Similarly Columbus has done worse than average on the powerplay resulting in worse goals for their shots. 5 on 5 though, it seems that in the modern NHL every team comes back to the mean.

The caveat modern NHL is crucial because teams like the 80's Oilers produced their offense on percentages rather than shot rates. But there doesn't seem to be any team that can manage it these days.

Personal shooting percentage: Percentage of shots taken by a player that become goals. Extremely volatile but converges on a player's career ability. This is extremely volatile but on the long term will converge with a player's career talent. From how it progresses over a career in the NHL and the degree it translates from the AHL to the NHL (almost entirely), in my opinion its less of a talent as it is a tendency which reflects a style of play.

Team on ice shooting percentage: The percentage of shots by his team while a player is on that become goals. Much like personal shooting percentage is heavily volatile but converges on a mean, in this case usually league average. Almost all players tend to be in a very narrow band of 7-8.5% team on-ice shooting with changes year to year being fairly random. Their seems to be a very small cadre of players that do better to average year to year. Usually these are player that play consistently on very good lines such as Sedin-Sedin and Stamkos-St. Louis. There are so few of them its debatable if the ability to be significantly better than average in this regard is real or just the mark of their being hundreds of players and a few are going to be better than average over multiple years by shear population dynamics.
I've read through this. Nice concise explanation. Thanks. I'm going to digest this over the next little bit and will probalby ask more questions of you later.

I'm also going to copy and paste your post into the first post of this thread so folks can refer to it. Thanks.

Lafleurs Guy is offline   Reply With Quote
Old
04-05-2012, 04:16 PM
  #186
Lafleurs Guy
Registered User
 
Lafleurs Guy's Avatar
 
Join Date: Jul 2007
Posts: 20,998
vCash: 500
Quote:
Originally Posted by Goldthorpe View Post
Maybe not. But say instead of 1 playoff of 4 series, the Oilers played 10. Each time, we set the clock back and all the 16 teams who made the playoff this year try again to win the cup, with games obviously having different outcomes. Out of ten, how many times do you think the Oilers makes the final even with Pronger playing (and hopefully not getting injured)?
Maybe none. But without Pronger it's definitely none.

Lafleurs Guy is offline   Reply With Quote
Old
04-05-2012, 04:28 PM
  #187
Talks to Goalposts
Registered User
 
Join Date: Apr 2011
Location: Edmonton
Country: Canada
Posts: 3,610
vCash: 500
Difficulty of minutes measurements:

In my opinion the more important way of analyzing the modern NHL game isn't in whether you accept shot based metrics or not (goal based ones generally get you close enough on that front anyway, at least if you accept goaltending as a factor which is uncontroversial) its acknowledging the major role that what type of minutes a player plays is a major factor in both how he looks to the eye and what kind of results they put up. NHL most NHL coaches manage their bench for particular territorial or line matchup roles and if there are a few that don't they'll get matched by the opposition resulting in largely the same effect.

Doing so may not have the direct intuitive effect you'd expect, i.e. starting in the offensive zone more might not directly raise a player's offensive stats or playing tougher minutes might not result in more goals against since such things will also depend on how a player responds to different situations but the effect will be strong on a player's differentials.

Two flavours of quality of minute situations are in current use. Quality of Compition (strength of opponent) and zone starts (location of starting shifts). An obvious third catagory would be strength of teammates but no one has really found a reliable measurement for that, its extremely hard to disentangle the strength of a player's linemates versus his own contribution. For that kind of thing I'd suggest looking at most common linemates by ice time and forming your own conclusion although you can do a more labourious "With You Without You" analysis (WOWY).

Quality of competition is measured two ways currently in the mainstream. Opposition +/- (QUALCOMP) and opposition relative Corsi (relCorsi QoC). The methodology of both is a bit weak so the straight numbers aren't always reliable comparing between teams, largely because each team has a different schedule but they work well within a team. Generally relCorsi QoC is the more reliable one when the two disagree for basically the same reasons Corsi is more reliable than +/-. For following a particular team though I recomend also looking for yourself what matchup are used game per game. For Montreal, Oliver's scoring chance reports for each game will also have matchup information while Habs Eyes on the Prize game threads will have a like to timeoneice.com script for a breakdown on minutes played per opponent each game. From tracking matchups the past couple years I can tell you that relCorsi QoC corresponds very strongly with who gets what kind of opponent but your welcome to test it out on your own. I didn't believe they worked until I did that myself.

Measuring where a player starts his shifts is generally done by way of Ozone % which is simply the porportion of non-neutral zone faceoffs (neutral zone faceoffs are considered neutral events for this) taken in the offensive zone. Unlike quality of competition which basically every team will have a spread on, whether to control your player's starting position is a decision made by a particular coach so not all teams will have a meaningful difference in zone starts. Vancouver in particular embraces this method, with Malhotra basically taking only defensive draws while Sedin takes tons of offensive zone ones. Chicago is a slightly less extreme adherent with the much under-rated David Bolland taking the heavy defensive duties while guys like Kane and Toews generally get a favourable starting point.

Montreal under Martin seemed to favour this strategy to start the season (a departure from is normal modus opperendi of pursuing straight power versus power matchups). With Desharnais getting in the range of a 60% Ozone while Plekanec (and usually Gionta) was around 42-45%. This was pretty wise considering that early season Desharnais produced lots of offense when it the offensive zone but had a weak possession/territorial game and the bad tendency to get pinned in his own zone when starting their facing a good line. One time when this broke down due to injury in a game against Pittsburgh, Desharnais basically got pinned in his own zone the entire game and ended up with something like a -20 even strength shot differential for the game. This kind of set up became less pronounced as the season continued but fortunately Desharnais developed the ability to face better competition and was paired with a very good two-way winger in Cole to compensate.

Subban also tends to be used a lot in the defensive zone, pretty much since he took over the first pairing last season when he moved up to play with Gill. That he generates team leading results in 5 on 5 shots and goal differential with unfavourable zone starts and very high quality of competition measurements is another reason for considering him to be an exceptional player.




For example of how this kind of analysis can be helpful, look at how a few of us were arguing at the beginning of the season that Cole was an awesome sigining while Leino was terrible. It was based largely on the principle that while they had similar surface stats, Cole was getting it done in a difficult situation while Leino had probably the easiest job in the league. Buffalo couldn't put Leino in a similar situation as he had in Philadelphia and he sunk. Cole ended up playing in an easier situation in Montreal and flourished.

Meanwhile the major drop in both Cammalleri and Plekanec's offensive totals since 2009-10 co-incides with when they got moved to playing a heavy shutdown role to start last season. Previously both tended to play about 2nd line competition, albeit with a major defensive zone faceoff responsibility, and both were on pace for a very good ~45 even strength points the previous season in that role. For reference, 40+ ES points is pretty much only achieved these days by top scoring line players. 45 points was good for 33rd in the league that season, 40 points 55th.

Talks to Goalposts is online now   Reply With Quote
Old
04-05-2012, 04:33 PM
  #188
MathMan
Registered User
 
Join Date: Jan 2006
Posts: 17,025
vCash: 500
TTG, when your primer is done, you need to grab it and post it someplace we can link to. Great work.

What's next on the menu -- per-icetime rate stats?

MathMan is offline   Reply With Quote
Old
04-05-2012, 04:41 PM
  #189
Lafleurs Guy
Registered User
 
Lafleurs Guy's Avatar
 
Join Date: Jul 2007
Posts: 20,998
vCash: 500
I'm going to answer this one more time and then I'm going to bail on this conversation. Not because I don't think it's worthy but because we're so far off topic and I really want to learn more about the SABR stuff. Besides everyone else has given up.
Quote:
Originally Posted by MathMan View Post
You know, that might be the bulk of the problem right there. That was pre-lockout, pre-salary cap, pre-new NHL, pre-dead puck era, pre-parity, pre-almost everything that matters to modern team-building. There's only so much you can apply from team-building in previous eras to the modern day.
Lockout or not, it still applies.

Post lockout there have been Hurricanes, Ducks, Wings, Pens, Bruins, Hawks.

Let's assume, Carolina has zero (despite Staal's 100 points and 45 goals.)

Ducks, Wings, Pens all have multiple superstars who are vets. The Bruins have one superstar on it with another having a superstar season... Thomas is strange. I don't know whether or not to call him a superstar because his career started so late but I don't see how anyone could say that he hasn't at least played like a superstar over the past few years. So out of the 6 four have multiple superstars.


That leaves the Hawks who are a young team and have Keith (who's already considered one of the best blueliners in the game) and Toews who was 2nd in goals before going down to injury. Superstars now? They're young players who are emerging as such but they aren't the vets that we see on other teams. Superstars? I think so. Maybe too soon to say. But Toews and Keith are certainly right up there and the Hawks also had 'star' players like Hossa, Kane and Campbell. They're similar to Calgary. Lots of star players to go along with two top guys.

It still applies man.



Quote:
Originally Posted by MathMan View Post
You're missing my point. You can't remove a team's best player, replace it with nothing and expect it to play at the same level, that is silly; but that doesn't mean that player has a special quality. Edmonton probably doesn't make the finals without Pronger, but they also don't make them without Roloson, without Pisani, without Horcoff, without Hemsky, or without Smyth. Heck you remove Spacek or Samsonov and they probably don't either.

On the other hand, you can replace Pronger and MAB with two first-pairing D-men that don't have the regard Pronger does and they're probably just as good if not better.

Besides, Edmonton was a Cinderella club. They really weren't all that good even with Pronger. And the Flyers making it to the finals arguably had more to do with Halak than with Pronger (as he spared them the need to face the East's two best clubs).
Like I said... no player can make it on his own. Yzerman used to miss the playoffs regularly as did Sakic and Lemieux. That's not the point.

The point is that if you add a superstar to an average team, you just might be able to at least make it to the finals with them. Pronger, Gretzky, Roy... we've seen it happen in the past.

We haven't seen a team WITHOUT one of those players win. Teams without those players don't get as far as teams with these players do.
Quote:
Originally Posted by MathMan View Post
Well, then maybe you need to reframe your argument, because you certainly come across as thinking that they have some quality that's virtually necessary. The more talent you have the better your odds are, obviously, but that doesn't mean that "having a superstar" is a prerequisite -- "having enough talent" is (on a practical level), and the latter does not necessarily require the former. Though it helps.
I've said the same thing about superstars that I have about top picks. They are a KEY ingredient to winning. Top picks become superstars with greater frequency and often lead or help lead teams to cups. It only stands to reason that we should try to get these kinds of players. And they're found mostly in the top part of the draft.
Quote:
Originally Posted by MathMan View Post
As for the definition of superstar, I think we have different criteria of player evaluation (personally I value two-way play/outscoring ability over raw offensive total, especially in the power-on-power NHL, and put no value on draft pedigree or reputation) so I doubt we'll ever agree on that. In general, I think raw offensive total is grossly overrated, while you seem to value it very highly (it seems to me, above all else). I do think it's significant that the last two Cup winners had two 30-goal men and one 70-point guy between them, however.
Bob Gainey was a great player. So was Guy Carbonneau. I used to love the way Carbo played and would dive in front of pucks. That being said... he's not a superstar.

What makes up a superstar? Tough to define. I'd say it's a player who can dominate consistently and that kind of talent is rare. If you've got a guy who can finish top five in goals or points every year it's hard to find that kind of talent. Why? Because there are only maybe four other players who can do this in the entire league. As much as I really liked Carbo, I can replace him with Skrudland or Kasper or some other checker who can do the job well enough.

Seriously would you take Gainey or Carbo over:
Gretz
Bossy
Trottier
Yzerman
Lemiex
Sakic?

No? Neither would I.

Thing about superstars too... they make the players around them better. I've wondered in the past what Kurri's career would've been like without Gretzky. Would Messier have become the player he did? Messier got to play a 2nd line role without having to face hard checking his whole career. He also had the benefit of playing with Gretz on occassion and padding his stats. Playing with superstars helps your development in all kinds of ways. Look at John Leclair... the very day he gets traded from us and plays with Lindros he scores something like 9 goals in his first 6 games.

Lafleurs Guy is offline   Reply With Quote
Old
04-05-2012, 04:42 PM
  #190
Lafleurs Guy
Registered User
 
Lafleurs Guy's Avatar
 
Join Date: Jul 2007
Posts: 20,998
vCash: 500
Quote:
Originally Posted by Talks to Goalposts View Post
Difficulty of minutes measurements:

In my opinion the more important way of analyzing the modern NHL game isn't in whether you accept shot based metrics or not (goal based ones generally get you close enough on that front anyway, at least if you accept goaltending as a factor which is uncontroversial) its acknowledging the major role that what type of minutes a player plays is a major factor in both how he looks to the eye and what kind of results they put up. NHL most NHL coaches manage their bench for particular territorial or line matchup roles and if there are a few that don't they'll get matched by the opposition resulting in largely the same effect.

Doing so may not have the direct intuitive effect you'd expect, i.e. starting in the offensive zone more might not directly raise a player's offensive stats or playing tougher minutes might not result in more goals against since such things will also depend on how a player responds to different situations but the effect will be strong on a player's differentials.

Two flavours of quality of minute situations are in current use. Quality of Compition (strength of opponent) and zone starts (location of starting shifts). An obvious third catagory would be strength of teammates but no one has really found a reliable measurement for that, its extremely hard to disentangle the strength of a player's linemates versus his own contribution. For that kind of thing I'd suggest looking at most common linemates by ice time and forming your own conclusion although you can do a more labourious "With You Without You" analysis (WOWY).

Quality of competition is measured two ways currently in the mainstream. Opposition +/- (QUALCOMP) and opposition relative Corsi (relCorsi QoC). The methodology of both is a bit weak so the straight numbers aren't always reliable comparing between teams, largely because each team has a different schedule but they work well within a team. Generally relCorsi QoC is the more reliable one when the two disagree for basically the same reasons Corsi is more reliable than +/-. For following a particular team though I recomend also looking for yourself what matchup are used game per game. For Montreal, Oliver's scoring chance reports for each game will also have matchup information while Habs Eyes on the Prize game threads will have a like to timeoneice.com script for a breakdown on minutes played per opponent each game. From tracking matchups the past couple years I can tell you that relCorsi QoC corresponds very strongly with who gets what kind of opponent but your welcome to test it out on your own. I didn't believe they worked until I did that myself.

Measuring where a player starts his shifts is generally done by way of Ozone % which is simply the porportion of non-neutral zone faceoffs (neutral zone faceoffs are considered neutral events for this) taken in the offensive zone. Unlike quality of competition which basically every team will have a spread on, whether to control your player's starting position is a decision made by a particular coach so not all teams will have a meaningful difference in zone starts. Vancouver in particular embraces this method, with Malhotra basically taking only defensive draws while Sedin takes tons of offensive zone ones. Chicago is a slightly less extreme adherent with the much under-rated David Bolland taking the heavy defensive duties while guys like Kane and Toews generally get a favourable starting point.

Montreal under Martin seemed to favour this strategy to start the season (a departure from is normal modus opperendi of pursuing straight power versus power matchups). With Desharnais getting in the range of a 60% Ozone while Plekanec (and usually Gionta) was around 42-45%. This was pretty wise considering that early season Desharnais produced lots of offense when it the offensive zone but had a weak possession/territorial game and the bad tendency to get pinned in his own zone when starting their facing a good line. One time when this broke down due to injury in a game against Pittsburgh, Desharnais basically got pinned in his own zone the entire game and ended up with something like a -20 even strength shot differential for the game. This kind of set up became less pronounced as the season continued but fortunately Desharnais developed the ability to face better competition and was paired with a very good two-way winger in Cole to compensate.

Subban also tends to be used a lot in the defensive zone, pretty much since he took over the first pairing last season when he moved up to play with Gill. That he generates team leading results in 5 on 5 shots and goal differential with unfavourable zone starts and very high quality of competition measurements is another reason for considering him to be an exceptional player.




For example of how this kind of analysis can be helpful, look at how a few of us were arguing at the beginning of the season that Cole was an awesome sigining while Leino was terrible. It was based largely on the principle that while they had similar surface stats, Cole was getting it done in a difficult situation while Leino had probably the easiest job in the league. Buffalo couldn't put Leino in a similar situation as he had in Philadelphia and he sunk. Cole ended up playing in an easier situation in Montreal and flourished.

Meanwhile the major drop in both Cammalleri and Plekanec's offensive totals since 2009-10 co-incides with when they got moved to playing a heavy shutdown role to start last season. Previously both tended to play about 2nd line competition, albeit with a major defensive zone faceoff responsibility, and both were on pace for a very good ~45 even strength points the previous season in that role. For reference, 40+ ES points is pretty much only achieved these days by top scoring line players. 45 points was good for 33rd in the league that season, 40 points 55th.
Thanks. I'll post this in the first post as well.

Lafleurs Guy is offline   Reply With Quote
Old
04-05-2012, 05:40 PM
  #191
Ubercron9000
Registered User
 
Ubercron9000's Avatar
 
Join Date: Aug 2009
Posts: 782
vCash: 500
Does anybody know if trevor Timmins uses sabremetrics or microstats as a guide to draft players? I mean how did duncan Siemens get drafted 11th overall but darren dietz gets drafted in the 5th round?

Im not good at math but if you draft a 3rd pairing guy in the fith round assuming he is on a stacked team putting up good numbers with his toi. Example if his toi is 5 minutes and he puts up 15 points. So you draft him in the 5th round and next year you speculate he might get 15 minutes toi would it be approximate that he gets 45 points? I have no idea but I have a feeling TT uses something of that variety when scouting players especially for later rounds.

sorry if I sound stupid or make no sense lol

Ubercron9000 is offline   Reply With Quote
Old
04-05-2012, 05:50 PM
  #192
Roke
Registered User
 
Roke's Avatar
 
Join Date: Jul 2003
Location: Winnipeg
Country: Canada
Posts: 1,889
vCash: 500
Quote:
Originally Posted by Ubercron9000 View Post
Does anybody know if trevor Timmins uses sabremetrics or microstats as a guide to draft players? I mean how did duncan Siemens get drafted 11th overall but darren dietz gets drafted in the 5th round?

Im not good at math but if you draft a 3rd pairing guy in the fith round assuming he is on a stacked team putting up good numbers with his toi. Example if his toi is 5 minutes and he puts up 15 points. So you draft him in the 5th round and next year you speculate he might get 15 minutes toi would it be approximate that he gets 45 points? I have no idea but I have a feeling TT uses something of that variety when scouting players especially for later rounds.

sorry if I sound stupid or make no sense lol
I'd say it's very unlikely, at least the type of analysis Talks to Goalposts is talking about. The statistical data in junior hockey just isn't there - you can't even get individual shot totals. Heck, the AHL just started tracking individual shot totals around the time Pacioretty's first stint in the AHL. I don't even think there's time on ice data for individual players.

One area where it could help when physically watching a guy for scouting is keeping track of where he starts his shifts on a faceoff (like the Zonestarts stuff in Talks to Goalposts last mini-primer). A guy putting up points while starting often in the defensive zone is probably better than the guy starting often in the offensive zone.


Last edited by Roke: 04-05-2012 at 05:56 PM.
Roke is offline   Reply With Quote
Old
04-05-2012, 06:03 PM
  #193
Ubercron9000
Registered User
 
Ubercron9000's Avatar
 
Join Date: Aug 2009
Posts: 782
vCash: 500
Quote:
Originally Posted by Roke View Post
I'd say it's very unlikely, at least the type of analysis Talks to Goalposts is talking about. The statistical data in junior hockey just isn't there - you can't even get individual shot totals. Heck, the AHL just started tracking individual shot totals around the time Pacioretty's first stint in the AHL. I don't even think there's time on ice data for individual players.

One area where it could help when physically watching a guy for scouting is keeping track of where he starts his shifts on a faceoff (like the Zonestarts stuff in Talks to Goalposts last mini-primer). A guy putting up points while starting often in the defensive zone is probably better than the guy starting often in the offensive zone.
thanks for clarifying that for me

Ubercron9000 is offline   Reply With Quote
Old
04-05-2012, 06:18 PM
  #194
Talks to Goalposts
Registered User
 
Join Date: Apr 2011
Location: Edmonton
Country: Canada
Posts: 3,610
vCash: 500
Quote:
Originally Posted by Ubercron9000 View Post
Does anybody know if trevor Timmins uses sabremetrics or microstats as a guide to draft players? I mean how did duncan Siemens get drafted 11th overall but darren dietz gets drafted in the 5th round?

Im not good at math but if you draft a 3rd pairing guy in the fith round assuming he is on a stacked team putting up good numbers with his toi. Example if his toi is 5 minutes and he puts up 15 points. So you draft him in the 5th round and next year you speculate he might get 15 minutes toi would it be approximate that he gets 45 points? I have no idea but I have a feeling TT uses something of that variety when scouting players especially for later rounds.

sorry if I sound stupid or make no sense lol
There is really very little information to go on in most minor leagues to do this kind of analysis. The best league out there for this is actually the QMJHL who have a fairly sophisticated statistical suite for a junior league, most impressively the track how many scoring chances a player produces ("dangerous shots" is the term they use). Interestingly, Louis Leblanc, despite less impressive point scoring was an absolute beast by the Q's micros. Best personal shot rate in the league and one of the best personal scoring chance rates. Bournival is similarly good this season but not quite to the same extent.

The best underlying stat you can look at for the AHL is just shot rate, something that Leblanc has also excelled at this season (almost 4 per game, 2nd best of his age cohort, Pacioretty was really strong by this measure too). I've personally done some preliminary work at seeing how shot rates and shooting percentage translate from the AHL to the NHL. The preliminary answer is shooting rate does seem like a translatable skill with rates AHL rates being about half the corresponding NHL rate on average. Shooting percentage wildly fluctuates like it does in the NHL from year to year but on the whole they look pretty similar in each league (best guess currently its retained by ~90% with low year to year individual correlation)

The most advanced general use and peer reviewed (i.e. open source methodology, not some guys home-brew stats work they're using for their own purposes or trying to sell) work I'm aware of for basic NHL scoring equivalencies, i.e. the average rate that point and goal scoring from one league translates from a feeder league to the NHL. The best established relation is from the leagues that regularly send a lot of players to the NHL at similar ages, namely the CHL and AHL. CHL on average you can roughly predict NHL scoring by multiplying by .29 (some split between OHL+WHL and the Q with .30 for the former and .27 for the later but a significant difference between the two populations hasn't been established) and AHL rates by .44.

This relationship is pretty rough, the best corrections are made by adjusting for age of player (younger being generally better for the player). I know that for those that have looked at junior scoring, age down to the month is seen as significant.

The AHL projections are probably the most accurate you can find anywhere, for example you could have pretty closely predicted NHL scoring rates for Pacioretty in each of his 3 AHL-NHL split years by converting his AHL scoring rates.

Of course there will always be guys like Maxwell that can't successfully make the jump from being good in the A to good in the NHL. False negatives seem to be pretty rare though, you generally have to be pretty good in lower leagues to score right away in the best league.

European league stuff tends to be a lot more complicated although the general principle that there is a correlation between scoring there and scoring in the NHL seems to hold.

Talks to Goalposts is online now   Reply With Quote
Old
04-05-2012, 06:23 PM
  #195
MathMan
Registered User
 
Join Date: Jan 2006
Posts: 17,025
vCash: 500
Quote:
Originally Posted by Roke View Post
I'd say it's very unlikely, at least the type of analysis Talks to Goalposts is talking about.
Although a wealthy pro team could afford to hire someone to look at the games on tape and collect the data -- and it doesn't have to be shots, it could be anything Timmins considers important.

MathMan is offline   Reply With Quote
Old
04-05-2012, 06:32 PM
  #196
Mathletic
Registered User
 
Mathletic's Avatar
 
Join Date: Feb 2002
Location: St-Augustin, Québec
Country: Canada
Posts: 11,872
vCash: 500
Quote:
Originally Posted by Ubercron9000 View Post
Does anybody know if trevor Timmins uses sabremetrics or microstats as a guide to draft players? I mean how did duncan Siemens get drafted 11th overall but darren dietz gets drafted in the 5th round?

Im not good at math but if you draft a 3rd pairing guy in the fith round assuming he is on a stacked team putting up good numbers with his toi. Example if his toi is 5 minutes and he puts up 15 points. So you draft him in the 5th round and next year you speculate he might get 15 minutes toi would it be approximate that he gets 45 points? I have no idea but I have a feeling TT uses something of that variety when scouting players especially for later rounds.

sorry if I sound stupid or make no sense lol
From what I heard Timmins is more into fitness tests and how they correlate to future performance. Not so much data analysis based on production and whatnot. I bought a few of those articles a while back. Apparently Timmins is onto those things.

One such article is

RELATIONSHIP OF PHYSICAL FITNESS TEST RESULTS AND HOCKEY PLAYING POTENTIAL IN ELITE-LEVEL ICE HOCKEY PLAYERS

JAIME F. BURR, RONI K. JAMNIK, JOSEPH BAKER, ALISON MACPHERSON, NORMAN GLEDHILL, AND E. J. MCGUIRE

Human Performance Laboratory, York University, Toronto, Canada

...

as for your hypothesis, no it's not stupid no hypothesis is stupid. Best researchers look for all possible avenues. The more you do, the more you might find what you're looking for. That said, no, minutes extrapolation at such small samples doesn't explain a whole lot. That said, Dietz was a great pick as he had a great defensive performance last year.

Mathletic is online now   Reply With Quote
Old
04-05-2012, 06:35 PM
  #197
SouthernHab
Registered User
 
SouthernHab's Avatar
 
Join Date: Mar 2011
Location: USA
Country: United States
Posts: 10,160
vCash: 500
Quote:
Originally Posted by Miller Time View Post





math is very much behind what makes Player 'A' get to the puck faster than Player 'B' (or shoot harder, or pass more accurately, and so on and so on).

sports, at the physical level, is about human movement, bodies in motion operate on a plane/field of play, angles/velocity/acceleration et. are all measurable components that will explain "WHAT" is happening.

I've never stated otherwise.

but find me the mathematical formula or statistical breakdown that explains why a player chokes under pressure, why a player struggles for one coach but shines for another, why one player shines in a high-stress envrionment like montreal whereas another wilts...

find me the stat package that will predict which junior star will get intimidated going into the corners vs chara and the mathematical formula that will predict which average junior player will break out and become the next Martin St.Louis...

call it intangibles if you want, but from my own experience with sport, I've directly seen how easy it is for an expert observer to recognize successful behaviors/intentions/actions in a junior athlete despite otherwise lackluster "measurable" indicators.

great scouts can watch a player play and see past "what" he does & understand "why" & "how" he's doing it (or not doing it), and that's the kind of insight that mathematical formulas don't cover, at least not yet.



do you not see how far removed a golf swing is from the constant action/adjustments that a fluid sport like hockey represents?

granted, during the actual swing there is tons of micro-adjustments going on, but all of them occur without any outside human influence.

Golf, even more so than baseball, is OBVIOUSLY a sport where mathematical formulas work very very well at both improving & predicting performance... the variables are contained and minimal.

Hockey is dynamic, always changing. A player will never in his career shoot exactly the same way as he just did, never. The mechanics will never be as close to a previous shot as a golfer's is to his previous swing.

and it's this huge variability, this immense amount of unpredictability (and the mental/emotional ability to adjust to it) that an expert observer provides the best "filter" for...

in a way, it's the ability of the human observer to comprehend and "compute" unpredictability (and thus seek answers for) that give it the advantage over the mathematical model which implicitly "expects" predictability.

not sure if i explained that last bit well, but basically it's the awarness of the "unkown unknown", and being sensitive to it, that makes the difference.
Best post in the entire thread. Great job!!!!!

SouthernHab is offline   Reply With Quote
Old
04-05-2012, 06:51 PM
  #198
Talks to Goalposts
Registered User
 
Join Date: Apr 2011
Location: Edmonton
Country: Canada
Posts: 3,610
vCash: 500
Quote:
Originally Posted by Miller Time View Post
I've never stated otherwise.

but find me the mathematical formula or statistical breakdown that explains why a player chokes under pressure, why a player struggles for one coach but shines for another, why one player shines in a high-stress envrionment like montreal whereas another wilts...

find me the stat package that will predict which junior star will get intimidated going into the corners vs chara and the mathematical formula that will predict which average junior player will break out and become the next Martin St.Louis...

call it intangibles if you want, but from my own experience with sport, I've directly seen how easy it is for an expert observer to recognize successful behaviors/intentions/actions in a junior athlete despite otherwise lackluster "measurable" indicators.

great scouts can watch a player play and see past "what" he does & understand "why" & "how" he's doing it (or not doing it), and that's the kind of insight that mathematical formulas don't cover, at least not yet.
I think your letting the perfect be the enemy of the good here.

You don't have to have the Grand Unified Theory of hockey that explains everything for analytic method to be useful. This the basic way practical science works, progressively looking for a better explanation than the previous one.

The mark of a good theory or system isn't that it will explain anything to be useful. It just needs to be something that you can use to make testable predictions which are as or more accurate as ones you can make by other means.

Its not like NHL scouts are consistently good at finding St. Louis types. And the track recorded at drafting junior players isn't much better than just taking the guy with the best junior scoring rates adjusted for age (might be worse, I know that's something that's been looked at).

What you're asking for is awfully close to asking for Laplace's demon which is a practical impossibility. What you should be asking for is something better than the collective wisdom of experts like NHL GMs.

As for finding a St. Louis by numbers, I suggest not ignoring guys that can manage 2+ points per game in NCAA hockey at 19 regardless of size to be a pretty good start.

As for finding clutchness, its a pretty open question if that's a skill an individual possesses (i.e. the continual ability to do better in important situations). Its pretty easy to find clutch events, harder to be sure about clutch individuals. After all, something like doing well or poorly in a couple of high pressure situations is going to be practically indistinguishable from random chance in the sample sizes most people talk about.

Roy is talked about as the pinnacle of a big game goalie but he's had his fair share of important situation were he failed to step up. I'd attribute his success more towards being good enough to have had lots of opportunities to succeed that the times he did he was remembered for them.


Last edited by Talks to Goalposts: 04-05-2012 at 06:58 PM.
Talks to Goalposts is online now   Reply With Quote
Old
04-05-2012, 07:06 PM
  #199
Mathletic
Registered User
 
Mathletic's Avatar
 
Join Date: Feb 2002
Location: St-Augustin, Québec
Country: Canada
Posts: 11,872
vCash: 500
on a side note, nothing like being a moneyball fan now and watching the Jays vs Cleveland lol! Seriously, watching the team trying to screw their chances by bunting and trying to steal bases. Hearing a clueless commentator on how they have to manufacture run. Trying to screw things up again by having Arencibian bunting ... and then, JP hits a home run ... priceless lol!

Mathletic is online now   Reply With Quote
Old
04-05-2012, 10:24 PM
  #200
Ubercron9000
Registered User
 
Ubercron9000's Avatar
 
Join Date: Aug 2009
Posts: 782
vCash: 500
thanks roke, talks to goal posts, mathletic and mathman for the responses and helping me understand better.

Ubercron9000 is offline   Reply With Quote
Reply

Forum Jump


Bookmarks

Thread Tools

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off



All times are GMT -5. The time now is 06:37 PM.

monitoring_string = "e4251c93e2ba248d29da988d93bf5144"
Contact Us - HFBoards - Archive - Privacy Statement - Terms of Use - Advertise - Top - AdChoices

vBulletin Copyright ©2000 - 2014, Jelsoft Enterprises Ltd.
HFBoards.com is a property of CraveOnline Media, LLC, an Evolve Media, LLC company. ©2014 All Rights Reserved.