Parise for Pouliot, Jacques

[paxtang]

Think about that for a couple of minutes. If you could lineup Parise against Forsberg and have Parise win the out-scoring battle, wouldn't it make sense to do that? The only guaranteed way to win a hockey game is have your first line outscore their firstline, your 2nd line outscore their 2nd line, your 3rd line outscore their 3rd line and your 4th line outscore their 4th line. Once you start matching your 3rd line agaisnt their 1st line, you have to hope your 1st line can outscore their 3rd line by at least as much as their 1st line outscores your 3rd line.

We in Edmonton saw how this worked with Doug Weight. Moreau-Marchant-Grier always matched up against the opponents best line, and we hid the Doug Weight line from everyone. What happened? The Oilers were a medicore team the saw Doug Weight inflate his point totals to get a fat contract then leave. Our checking line always matched up against the oppositions best and we were a 90 point team. Now look at our team. We've lost guys like Doug Weight and Mike Comrie, and are still a 90 point team. The reason being is the Oilers now match up their best outscorers (Smyth, Horcoff, Dvorak, York) against the other teams best players. We have no where near the name recognition that we used to, but we are a better team (our schedule is harder now than it was when Weight was an Oiler).

I mean you could try and use Parise's offense against weaker players to try and use it to your advantage... but to be honest, that only really works if you have a John Madden or a Jere Lehtinen on the team (they are about the only 2 players in the league who you can match a checking line vs top line and come out ahead on most nights). Other than that, you aren't giving yourself the best chance to win.



You have this one-dimensional view of hockey. It's all about offensive and defensive players to you. In an ideal situation, a coach matches top line vs top line. Forsberg never plays against an oppositions 4th line, he's always out there against their very best. He'd get a tonne more points playing agaisnt a 4th line, but it wouldn't necessarily help his team as much. The best recipe for success is to outscore every line, and most teams don't have 3rd lines that can outscore first lines (in fact, unless you consider Madden and Lehtinen 3rd line players, there aren't any lines, at least when playing good teams). Ryan Smyth is not a great defensive player (he's probably slightly above average, but in terms of reading plays in his end, or knowing when to double team the puck, he does not compare to a guy like Lehtinen). He's a great out-scorer, but he isn't a great defensive player. I mean he is effective because most of the time when he is on the ice, the puck is in the other teams zone. They can't score when the puck is in their end of the ice.

Guys like Lehtinen and Madden are unique because they can outscore the top lines... not just hold them close. If you can get that out of players, then running good offensive players out against the crappier lines is a solid move. But that's why teams like the Devils or Stars didn't fret when guys like Mogilny or Hull left... you can get similar outscoring against weaker opposition for a fraction of the cost.

I don't know, it seems to me, at least in the Atlantic division, that a lot of teams used a shut down line on the top line. I know Primeau was nearly exclusivly matched up against the 1st line of the opposing teams. I'm not saying everyone does it, but there's more than just Madden and Lehtien who do it, although I'm sure Hitch's experience in Dallas influenced his use of Primeau.

I think if you get down to it, most NJ Dev fans wouldn't really disagree a whole lot with the evaluations of the Oiler fans on Parise, save for the inevitable homer tint.

And New Jersey sucks.

 

[oil slick]

Statistical significance isn't the same as causality
An experiment is ideally designed so that (hypothetically) the independent variable(s) represent factors that cause change in the dependent variable(s). However, statistical inference makes no claims of causality. None at all. All that is being done is a computation of the probability that your null hypothesis is true. In scientific research, causality is established by

Having an explanatory theory of causality. An experimental result consistent with such a theory is a good thing, it's just not proof all by itself
Careful controls for biases. The more carefully you control for biases, the higher your confidence is that your experiment is really focused on the relationship you want to explare. Replication.

If there really is a causal relationship there, the experiment should be repeatable. This is a lot easier to do properly in chemistry or physics than it is in psychological experimentation, but it needs to be done in any case.

Even if you have all the above, your model may not explain everything and need to be refined. For example, the model of an indivisable atom was exceedingly useful for 19th century research into chemical reactions,a nd the development of the periodic table of elements. However, it wasn't sufficient to explain why if you had a lump of uranium it would emit radioactive particles and eventually transmute into lead. This lead to a new theory of the nucleus to explain these phenonema. This doesn't mean the old theory is useless -- the old theory probably explains everything you ever did in high school chemistry and most of what you did in college chemistry, (assuming you continued to take any chemistry, that is). However, scientific models attempt to explain all measured phenomena as accurately as possible

Thanks for the update. Although I work with stats all the time, and even gave a lecture to a graduate course on pattern recognition, it is fantastic to be tutored on the fine points of first year undergraduate studies.

Having worked in brain imaging research for three years, I can tell you that although you strive to remove biasis, you cannot do this. It is impossible. Your stats professors should be telling you that the real world sometimes cannot supply the unbiased, independent data that would be ideal. In these cases, you produce correlations that may have dependencies within them, but you try to analyse how strong a likelihood that a result you found is accurate.

In brain imaging, when doing classifications using Bayes Law, we should have roughly 100*#features*#classes in order to produce ideal results. This is infeasible, seeing as you have to actually take MRI scans, which take many hours per set of scans. So, we produced results using fewer scans. Were these results absolutely correct? No. Were they still worthwhile looking at? Absolutely. We did statistical analysis of the data, and showed the limitations, and we made no claims that the results were difinitive... but it was of interest to many.

You are applying the rigour of Physics to the real world, but what you don't understand is that noone here is saying that this little stat is as valid as some physics experiment that has been done a million times, just that there is probably some correlation.

 

[JimEIV]

Thanks for the update. Although I work with stats all the time, and even gave a lecture to a graduate course on pattern recognition, it is fantastic to be tutored on the fine points of first year undergraduate studies.

Having worked in brain imaging research for three years, I can tell you that although you strive to remove biasis, you cannot do this. It is impossible. Your stats professors should be telling you that the real world sometimes cannot supply the unbiased, independent data that would be ideal. In these cases, you produce correlations that may have dependencies within them, but you try to analyse how strong a likelihood that a result you found is accurate.

In brain imaging, when doing classifications using Bayes Law, we should have roughly 100*#features*#classes in order to produce ideal results. This is infeasible, seeing as you have to actually take MRI scans, which take many hours per set of scans. So, we produced results using fewer scans. Were these results absolutely correct? No. Were they still worthwhile looking at? Absolutely. We did statistical analysis of the data, and showed the limitations, and we made no claims that the results were difinitive... but it was of interest to many.

You are applying the rigour of Physics to the real world, but what you don't understand is that noone here is saying that this little stat is as valid as some physics experiment that has been done a million times, just that there is probably some correlation.

"Today, the statistical community generally recognizes that these approaches are inappropriate in an era when anyone with a computer and a statistical software package can attempt to be his/her own statistician. "Cause & effect" must be among the first things that are addressed because this is what most people will use statistics for! Newspapers, radio, television, and the Internet are filled with claims based on some form of statistical analysis. Calcium is good for strong bones. Watching TV is a major cause of childhood and adolescent obesity. Food stamps and WIC improve nutritional status. Coffee consumption is responsible for heavens knows what! All because someone got hold of a dataset from somewhere and looked for associations. Which claims should be believed? Only by understanding what it takes to establish causality do we have any chance of being intelligent consumers of the "truths" the world throws at us. "

Freedman points out that statistical demonstrations of causality are based on assumptions that often are not checked adequately. "If maintained hypotheses A,B,C,... hold, then H can be tested against the data. However, if A,B,C,... remain in doubt, so must inferences about H. Careful scrutiny of maintained hypotheses should therefore be a critical part of empirical work--a principle honored more often in the breach than in the observance." That is, an analysis could be exquisite and the logic could be flawless provided A,B,C hold but the same attention is rarely paid to checking A,B,C as goes into the analysis that assumes A,B,C hold.


By the way I am a developer for a pharmaceutical company not a student.

 

[oil slick]

"Today, the statistical community generally recognizes that these approaches are inappropriate in an era when anyone with a computer and a statistical software package can attempt to be his/her own statistician. "Cause & effect" must be among the first things that are addressed because this is what most people will use statistics for! Newspapers, radio, television, and the Internet are filled with claims based on some form of statistical analysis. Calcium is good for strong bones. Watching TV is a major cause of childhood and adolescent obesity. Food stamps and WIC improve nutritional status. Coffee consumption is responsible for heavens knows what! All because someone got hold of a dataset from somewhere and looked for associations. Which claims should be believed? Only by understanding what it takes to establish causality do we have any chance of being intelligent consumers of the "truths" the world throws at us. "

Freedman points out that statistical demonstrations of causality are based on assumptions that often are not checked adequately. "If maintained hypotheses A,B,C,... hold, then H can be tested against the data. However, if A,B,C,... remain in doubt, so must inferences about H. Careful scrutiny of maintained hypotheses should therefore be a critical part of empirical work--a principle honored more often in the breach than in the observance." That is, an analysis could be exquisite and the logic could be flawless provided A,B,C hold but the same attention is rarely paid to checking A,B,C as goes into the analysis that assumes A,B,C hold.

OK - so analyse where A,B,C fail with respect to Igor's little stats. I'm sure you'll find a few, and I'm sure I'll agree with them. If you could not show some potential flaws in A,B,C, then I would say that the stat was difinitive... which has not been said.

To dismiss the statistic by saying the it sounds like astrology is not what Freedman and your stats Prof are talking about.

a)Igor has proposed that outscoring oponents is a desirable thing. Argue that he's a pp specialist or something.

b)Igor has proposed that a good indicator of this is how well you do against opponents with given +/-'s.

c)Igor has proposed that the fact that Parises linemates do better when he's not around is a knock on Parise.

There's plenty of stuff to dispute, but saying it's Astrology when it's obviouse there is some potential for truth to it, is silly.

 

[JimEIV]

c)Igor has proposed that the fact that Parises linemates do better when he's not around is a knock on Parise.

There's plenty of stuff to dispute, but saying it's Astrology when it's obviouse there is some potential for truth to it, is silly.

Fair enough,

But C is what I have a problem with. Mainly because there are so many permutations of what his teammates do without Parise that I do not believe cause and effect can be shown.

Kind of like a study that looks at 100 coal-mine workers; all coal-mine workers smoke therfore cigarettes causes lung problems......The study never looks to see if it is the mine that is causing the sickness!

I believe that is what we are seeing here. When Parise plays with certain players is he asked to play more of a defensive role??? When he plays with others is he being asked to provide more offense? The stats do not show Cause and Effect, pure and simple.

 

[Asiaoil]

Fair enough,

But C is what I have a problem with. Mainly because there are so many permutations of what his teammates do without Parise that I do not believe cause and effect can be shown.

Kind of like a study that looks at 100 coal-mine workers; all coal-mine workers smoke therfore cigarettes causes lung problems......The study never looks to see if it is the mine that is causing the sickness!

I believe that is what we are seeing here. When Parise plays with certain players is he asked to play more of a defensive role??? When he plays with others is he being asked to provide more offense? The stats do not show Cause and Effect, pure and simple.

The minute you subject your own line of reasoning (roughly summarized as I believe what I see) to the same level of scrutiny as igor's stats approach is when I will step back into this debate.

Other than that - just want to tell the oiler guys that you have shown why that board is one of the most interesting around - very impressive guys. A thread like this is typical most days.

 

[JimEIV]

The minute you subject your own line of reasoning (roughly summarized as I believe what I see) to the same level of scrutiny as igor's stats approach is when I will step back into this debate.

Other than that - just want to tell the oiler guys that you have shown why that board is one of the most interesting around - very impressive guys. A thread like this is typical most days.

I think you just said that until I disprove the world is flat, the world is still flat.

Ok sounds good to me.

 

[Asiaoil]

I think you just said that until I disprove the world is flat, the world is still flat.

Ok sounds good to me.

No I said that I am very willing to discuss the pros and cons of Roger Nielsen's approach and subject it to evaluation. I also think that some folks in this debate should consider the possibilty that what they want to believe about Parise is coloring what they are actually seeing on the ice as they watch the game. Humans are subjective creatures.

An old very smart prof of mine used to mutter this at me in the field when I came up some some smart-ass concept: "never would have seen it if you didn't believe it first". I think it may apply here.

 

[dawgbone]

Fair enough,

But C is what I have a problem with. Mainly because there are so many permutations of what his teammates do without Parise that I do not believe cause and effect can be shown.

Kind of like a study that looks at 100 coal-mine workers; all coal-mine workers smoke therfore cigarettes causes lung problems......The study never looks to see if it is the mine that is causing the sickness!

Except this isn't what is happening. It looks at Parise with and without various linemates, and it looks at various linemates with and without Parise. So it's getting both the coal miners who smoke, the ones who don't, and the people who aren't coalminers and smoke, and the ones who aren't coalminers and don't smoke. Smoking or coal mining do not cause lung cancer... they just increase your risk of getting it.

A good outscorer will always have one trait... bringing up the outscoring numbers of the players he plays with. It doesn't really matter what else his teammates do. You don't really need cause and effect because it's not an exact science (in other words it's very hard to predict how far a good outscorers numbers will drop when paired with a poor one, or vice versa). I mean I'm sure if you analyze enough data, you could probably make a decent enough chart with approx. values, but who has the time for that?

What we do know, is that as your level of competition (or "tough minutes") decreases, your out-scoring numbers decrease, and your point production increases. And as your level of competition increases, your out-scoring numbers decrease and your point production decreases (edited).

Take a look at scoring around the NHL. Goal differential hasn't changed much over the past 5 years, but no one gets 50+ goals or 100+points. Ask yourself why that is? How can scoring remain almost the same over the past 5 years, yet the point totals for leading scorers gets lower and lower? Take a look at team lists, and I'm sure you'll find fewer and fewer 80-100 point players and more and more 20-30 point players (that is unresearched and just an educated guess). That's more high-end guys playing against tougher competition, and more lower-end players playing against weaker competetion as opposed to facing off against the oppositions top lines.

I believe that is what we are seeing here. When Parise plays with certain players is he asked to play more of a defensive role??? When he plays with others is he being asked to provide more offense? The stats do not show Cause and Effect, pure and simple.

So what you are hypothesizing, is that when Parise is asked to play a more defensive role, he is a better outscorer than when he is given free-reign to do whatever he likes? Or is it the other way around?

I mean is he asked to play a more offensive role with McAmmond and Rheaume, or is that when he's asked to play more defense.

Either way, it doesn't matter as he isn't really effective in either. If he's asked to play in a defensive role with McAmmond and Rheaume, he drags their outscoring numbers down. If he's asked to play in an offensive role, he's still dragging their numbers down.

Regardless of whether he's asked to be more offensively involved, or more defensively reliable, he's not bringing up anyone else's outscoring numbers, so either way he's not getting the job done.

 

[Crosbyfan]

Smoking or coal mining do not cause lung cancer... they just increase your risk of getting it.

.

Smoking or coal mining without proper protection (inhaling coal dust etc.) very likely causes cancer in some individuals in some circumstances. Thus the increased risk.

A week ago I had a few vodka and orange juice and woke up with a headache the next day.
6 days ago I had a few rye and orange juice and woke up with a headache the next day.
5 days ago I had a few gin and orange juice and woke up with a headache the next day.
4 days ago I had a few rum and orange juice and woke up with a headache the next day.
3 days ago I had a few scotch and orange juice and woke up with a headache the next day.
I've only drunk water in the last couple of days and feel much better.

That orange juice was nasty stuff!

 

[JimEIV]

Smoking or coal mining without proper protection (inhaling coal dust etc.) very likely causes cancer in some individuals in some circumstances. Thus the increased risk.

A week ago I had a few vodka and orange juice and woke up with a headache the next day.
6 days ago I had a few rye and orange juice and woke up with a headache the next day.
5 days ago I had a few gin and orange juice and woke up with a headache the next day.
4 days ago I had a few rum and orange juice and woke up with a headache the next day.
3 days ago I had a few scotch and orange juice and woke up with a headache the next day.
I've only drunk water in the last couple of days and feel much better.

That orange juice was nasty stuff!

Exactly!!

 

[Asiaoil]

Smoking or coal mining without proper protection (inhaling coal dust etc.) very likely causes cancer in some individuals in some circumstances. Thus the increased risk.

A week ago I had a few vodka and orange juice and woke up with a headache the next day.
6 days ago I had a few rye and orange juice and woke up with a headache the next day.
5 days ago I had a few gin and orange juice and woke up with a headache the next day.
4 days ago I had a few rum and orange juice and woke up with a headache the next day.
3 days ago I had a few scotch and orange juice and woke up with a headache the next day.
I've only drunk water in the last couple of days and feel much better.

That orange juice was nasty stuff!

Hey I admit it is possible that Parise's outscoring numbers may be due to bad luck, poor coaching, crappy linemates, Jupiter aligning with Mars, or an underworld conspiracy. This is possible but unlikely - and perhaps a simpler explanation is that he just needs to work on his defensive game. I think we have defended our position quite adequately - and I have seen nothing even remotely similar from the other side.

Time to step up to the plate or walk out of the arena - and I'll be happy to hear you guys offer some real evidence for what you think you see.

 

[Crosbyfan]

Hey I admit it is possible that Parise's outscoring numbers may be due to bad luck, poor coaching, crappy linemates, Jupiter aligning with Mars, or an underworld conspiracy. This is possible but unlikely - and perhaps a simpler explanation is that he just needs to work on his defensive game. I think we have defended our position quite adequately - and I have seen nothing even remotely similar from the other side.

Time to step up to the plate or walk out of the arena - and I'll be happy to hear you guys offer some evidence for what you think you see.

7 goals.

How many times did his team pull their goaltender?

How many times was Parise chosen to be on the ice when they pulled their goaltender?

Who is more likely to score? The team with the goalie? The team with the extra forward?

7 goals... on a slightly minus team.

How many times was his team down a goal or two late in the game?

How many times was Parise chosen to be on the ice if somewhat rested when they were down a goal or two late in the game?

Who is more likely to score? The team desperate for goals? The team that can trap and counterattack?

7 goals... on a slightly minus team.


If you are trying to say that Parise is not a "shutdown center" you have "evidence" that he was not "used" (or did not perform) that way on a regular basis. (the -7 combined with the good offence stats)

I agree that the +/- is one of the better stats available but if you are trying to say he is a defensive liability I don't think 7 goals is enough.

I don't think 7 goals is enough to say Parise is a defensive liability.

 

[DARKSIDE]

Oiler fans try calming down. I mean just because Lou picked Lowe's pocket is no reason to to get your panties all bunched up. As for the most over hyped prospect, that award goes to Jeff Carter!

 

[Asiaoil]

Oiler fans try calming down. I mean just because Lou picked Lowe's pocket is no reason to to get your panties all bunched up. As for the most over hyped prospect, that award goes to Jeff Carter!

Hey if we could get any 5 on 5 numbers out of the OHL I would love to see what guys like Carter and our own lovable little PP whore Shremp are doing with respect to out-scoring. The results may make some people unhappy - or raise the quality of the debate above "he-said" she-said"

You make it sound like Oilers fans on a mission to justify trades - not so - we are quite happy to tear our own prospects to shreds if the evidence points that way. Drop by our board for a visit and see for yourself. This thread is about prospect evaluation techniques and we could use any player - just happened to be Parise this time.

 

[igor*]

Majoring in Comp Sci with a minor in Statistics.........This sounds like astrology to me.

If you ignore my own view on the game for a minute ... do you have anything to disagree with in what Dawgbone has said in this thread? I mean he's describing the tactics that NHL coaches employ in this era, and why they do it. And accurately IMO. I hope I didn't get your back up, and perhaps I'm wrong, but you seem to be kicking at this thread on sheer principle now.

When the NHL starts again, humour me ... watch a game that you are dispassionate about, where you don't hate or love either team ... BOS vs PHI would be good for HF (if you don't hate Philly). And count the scoring chances in the game, just by your own judgment, vs who was on the ice for either team. Tally 'em up ... pretend you're a coach for a day . If you have PVR it's very easy, and in any case it's not difficult ... relative to the time many of us spend thinking and talking about hockey it's a drop in the bucket.

That should be an eye opener. But what will blow you away is the next step; when you go to the PHI and BOS boards on here and ask them what happened in the game (be sure to kiss arse first, that's how Lowetide does it ). Priceless stuff, everyone should do it at least once.

 

[igor*]

7 goals.

How many times did his team pull their goaltender?

How many times was Parise chosen to be on the ice when they pulled their goaltender?

Who is more likely to score? The team with the goalie? The team with the extra forward?

7 goals... on a slightly minus team.

How many times was his team down a goal or two late in the game?

How many times was Parise chosen to be on the ice if somewhat rested when they were down a goal or two late in the game?

Who is more likely to score? The team desperate for goals? The team that can trap and counterattack?

7 goals... on a slightly minus team.


If you are trying to say that Parise is not a "shutdown center" you have "evidence" that he was not "used" (or did not perform) that way on a regular basis. (the -7 combined with the good offence stats)

I agree that the +/- is one of the better stats available but if you are trying to say he is a defensive liability I don't think 7 goals is enough.

I don't think 7 goals is enough to say Parise is a defensive liability.

Well, although I don't have "evidence" that you are a "genuine dumbass" ... I'm pretty sure that you are.

Peace out.

 

[igor*]

Doesn't work that way dude, you guys are the one that introduce it, so you need to show first that it is accurate, not the other way around. BTW, I like igor's stats and numbers (still trying to figure where finds the time to do this ), I just don't think it makes a case of Parise being a defensive liabilty as alot of folks are suggesting.

And I for one have never claimed him to be a shut down center (which could be the case down the road, still early), really don't think anybody has from our side, but did it ever occur that if he used in soft situations it's probably an attempt of the coach to take advantage of his offensive skills and not so much an attempt to hide any flaws in Parise defensive game, if there is any? I mean does any coach look to put his best offensive players head to head to the opponents best defensive players, hell no. They are always going to try to put them against a line that have the best chance of producing.

As for doing this stuff ... very simple, and really a community effort in a lot of ways. A guy on OilFans wrote a script to dump absolutely everything off of the NHL.com database. Another to convert it to CSV. And while I'm NOT a programmer ... even I could manage to write a few simple scripts for it. With those written ... some other clever bugger was able to dump the AHL stuff into virtually the same format so I could use the same scripts (and he did it in minutes ). Very basic, heavy handed stuff on my part.

I was looking at the Oilers AHL team for the F of it, and because it took about 10 seconds to do (and because the Parise/Pouliot stuf keeps coming over there far too much too) I thought I'd run Albany as well just cuz.

Then Mr Bugg told me about this thread and sort of goaded me onto it ... much to my regret. Thankfully the hardarse posters from the Oilers board haven't noticed this or descended, because we're talking about hockey basics now, and there are some pretty soft targets here.

On the Parise thing ... I'm not pissing on the kid. Wish him the best, seriously. But sensibly he's not looking to be a world beater, not if history is any gauge. Should be expected to be a decent NHLer though, that seems pretty reasonable. And for that matter neither is Pouliot. If EITHER team gets a guy who averages 40 points a year and can at least resemble an outscorer in their prime ... then collectively N.J and EDM have beaten the curve I think.

 

[Lou is God]

As for doing this stuff ... very simple, and really a community effort in a lot of ways. A guy on OilFans wrote a script to dump absolutely everything off of the NHL.com database. Another to convert it to CSV. And while I'm NOT a programmer ... even I could manage to write a few simple scripts for it. With those written ... some other clever bugger was able to dump the AHL stuff into virtually the same format so I could use the same scripts (and he did it in minutes ). Very basic, heavy handed stuff on my part.

That's what I figured it had to be something like that, I mean to do it manually would be incredibly time consuming, you would have time for nothing else but that.

BTW, I'm done here, I think were all beating a dead horse now.

 

[igor*]

...
BTW, I'm done here, I think were all beating a dead horse now.

Yup. Me too. Good talking though.

 

[Crosbyfan]

Well, although I don't have "evidence" that you are a "genuine dumbass" ... I'm pretty sure that you are.

Peace out.

Igor, keep your brain together Igor. I know you can do it.

 

[Jason MacIsaac]

Hey I admit it is possible that Parise's outscoring numbers may be due to bad luck, poor coaching, crappy linemates, Jupiter aligning with Mars, or an underworld conspiracy. This is possible but unlikely - and perhaps a simpler explanation is that he just needs to work on his defensive game. I think we have defended our position quite adequately - and I have seen nothing even remotely similar from the other side.

Time to step up to the plate or walk out of the arena - and I'll be happy to hear you guys offer some real evidence for what you think you see.

You have defended nothing, you used stats to prove a point where everyone else who seen him used personal opinions. You could allways just as a Bingo fan or Rochester what his game is like.

I stoped in this thread long ago because debating someone who can only use stats will go nowhere. I will leave it at that.

 

[dawgbone]

Smoking or coal mining without proper protection (inhaling coal dust etc.) very likely causes cancer in some individuals in some circumstances. Thus the increased risk.

That's not cause and effect. Cause and effect would be everyone who smoked or mined coal getting lung cancer. There are very few things that cause lung cancer (in other words if you do it, you will get lung cancer), but there's a list of stuff long enough to reach the moon that can increase your risk of acquiring lung cancer.

So let's make up our mind here kids... are we talking about cause and effect, or are we talking about likely outcomes? Nothing like changing the song when you have no idea what the words are.

A week ago I had a few vodka and orange juice and woke up with a headache the next day.
6 days ago I had a few rye and orange juice and woke up with a headache the next day.
5 days ago I had a few gin and orange juice and woke up with a headache the next day.
4 days ago I had a few rum and orange juice and woke up with a headache the next day.
3 days ago I had a few scotch and orange juice and woke up with a headache the next day.
I've only drunk water in the last couple of days and feel much better.

That orange juice was nasty stuff!

That's cute... idiotic... but cute. Fitting.

 

[dawgbone]

7 goals.

How many times did his team pull their goaltender?

How many times was Parise chosen to be on the ice when they pulled their goaltender?

Who is more likely to score? The team with the goalie? The team with the extra forward?

7 goals... on a slightly minus team.

How many times was his team down a goal or two late in the game?

How many times was Parise chosen to be on the ice if somewhat rested when they were down a goal or two late in the game?

Who is more likely to score? The team desperate for goals? The team that can trap and counterattack?

7 goals... on a slightly minus team.


If you are trying to say that Parise is not a "shutdown center" you have "evidence" that he was not "used" (or did not perform) that way on a regular basis. (the -7 combined with the good offence stats)

I agree that the +/- is one of the better stats available but if you are trying to say he is a defensive liability I don't think 7 goals is enough.

I don't think 7 goals is enough to say Parise is a defensive liability.

Sweet Jesus man... I'll respond here because Igor's said his piece.

Ignore the +/- stat you see on the AHL website. That's garbage and totally on the wrong path. That factors in SH gf/a, empty netters f/a, etc.

This isn't what his stat in the +/- column says. It's what the ES +/- says (5 on 5 goals for and goals against). That is what the majority of the game is played at. So Parise could be on the ice for 40 empty net goals against, and it wouldn't make a lick of difference, as it wouldn't count against his ES +/-.

It's not 7 goals we are talking about... it's every single even strength goal for and against that he is on for, which is a much bigger sample size. It's how he plays during the most common part of the game, which is 5 on 5.

The evidence that he is not a shutdown centre is in the events... in other words goals. Unless he played a tonne of time against top opposition where neither team scored (very unlikely), calling him a shutdown centre just isn't true.

It's not rocket science... it's just looking beyond the most readily available stats. I don't do nearly the work igor does in terms of pulling numbers from the AHL/NHL, but I do a lot of stats work in the coaching I do, and it's the same principle, and it's proven to be greatly effective. It sucks having to change everything you know about hockey, but it's incredibly effective, and highly addictive.

I don't expect you to beleive me... you won't until you spend some time and try it out. It's difficult to grasp, and your natural reaction is to beleive everything you see in game or read in the paper or on the internet that pertains to your belief. Spend time watching what coaches do... especially at home (where they get the matchups they want). Even better is a home and home series, where you can count matchups. Watch a Colorado-Dallas game and watch how many times Modanno lines up against Forsberg (both home and away). Look for the good outscorers in the league and when two teams face off, how often the best on each team lines up against each other (Dallas-Colorado is a good one).

 

[Bryanbryoil]

DAMN is this a long thread!!! IMO the best thing to do is see what kind of AHL rookie seasons Pouliot and Jacques have next year. The Runners won't be the greatest team and Pouliot maybe playing behind Brodziak or another vet "I hope not!!!". So we shall see, until then IMO this discussion is just repetitious and flat out annoying!!! Parise is a good prospect as is Pouliot and Jacques. Until there's a comperable basis "Next Season" this is pointless.