OccamsRazor Posted December 9, 2014 Share Posted December 9, 2014 I get where you are coming from. Don't drag me back into this i'm done my head is going to explode!!!!! Waves the white flag!!! The Alamo is yours!!! Link to comment Share on other sites More sharing options...
phlfly Posted December 9, 2014 Share Posted December 9, 2014 most stat is suck anyway, like Ov is great player but his +_ is really bad, but put him hawks between Kane and Toews and he will score may be 60 or 70. And one more example Nash 's stat were terrible in columbus since the team is really bad, now he is 18 goal +9. So as I said before I would rather to have Ov over Giroux since at least he is exited to watch, Ov and his one timers are at least hitting the net when G's are missing most time. Link to comment Share on other sites More sharing options...
idahophilly Posted December 9, 2014 Share Posted December 9, 2014 most stat is suck anyway, like Ov is great player but his +_ is really bad, but put him hawks between Kane and Toews and he will score may be 60 or 70. And one more example Nash 's stat were terrible in columbus since the team is really bad, now he is 18 goal +9. So as I said before I would rather to have Ov over Giroux since at least he is exited to watch, Ov and his one timers are at least hitting the net when G's are missing most time. Phlfly, you never disappoint. Thanks. Link to comment Share on other sites More sharing options...
J0e Th0rnton Posted December 9, 2014 Share Posted December 9, 2014 Don't drag me back into this i'm done my head is going to explode!!!!! Waves the white flag!!! The Alamo is yours!!! Hey. I am not trying to browbeat like some others lol. I enjoy advanced stats, and think they have some value, but I do not think they are the end all to discussions like some do. I just know they correctly portray Brad Stuart was a train wreck for the Sharks unless Braun was covering his arse. Link to comment Share on other sites More sharing options...
Podein25 Posted December 9, 2014 Share Posted December 9, 2014 Here's what stats can't measure, except in the aggregate: the player who makes the players around him better. Not every player, even pros, can say they do that.Of course, there are those players whose main quality is in making the players around them *worse*. We like to call them "Umbergers"in Flyerland Link to comment Share on other sites More sharing options...
OccamsRazor Posted December 9, 2014 Share Posted December 9, 2014 I enjoy advanced stats, and think they have some value, but I do not think they are the end all to discussions like some do. I agree completely and waaaaaay back that is all i was trying to say....but this has got out of hand and by all means i don't want to ruin everyone's else's fun i'm just saying i've said my part just get exhausted going over it over and over with folks not you in particular i'm just saying. I've just read about as much as i can on this stuff and feel it has it's place but its not the be all end all tool. Link to comment Share on other sites More sharing options...
idahophilly Posted December 9, 2014 Share Posted December 9, 2014 I agree completely and waaaaaay back that is all i was trying to say.... That's what was/is so frustrating about it. I have been saying it has a place but I'm still wrong. You either agree 100% or your wrong. Then when it's denied that people make the decisions WITH reasoning involved, well then you have no where to go. Link to comment Share on other sites More sharing options...
aziz Posted December 9, 2014 Share Posted December 9, 2014 Here's what stats can't measure, except in the aggregate: the player who makes the players around him better. and that's what these relative applications of corsi and fenwick try to do. they just start with such garbage data. Link to comment Share on other sites More sharing options...
brelic Posted December 9, 2014 Share Posted December 9, 2014 There is no reasonable discussion to be had in this situation when adv stats folks are firmly entrenched. I'm not an advanced stats guy, and I'm not entrenched. I don't care much for advanced stats in hockey - I just watch the games and use traditional stats to talk shop. Nothing special. I know about corsi. I've read about corsi. But I never really looked or followed that measurement at all. I do use stats on a regular basis at work, though, and they're important to me in that world. Anyway, I apologize if you got the impression that I wasn't being reasonable - it was not my intention. Link to comment Share on other sites More sharing options...
idahophilly Posted December 9, 2014 Share Posted December 9, 2014 and that's what these relative applications of corsi and fenwick try to do. they just start with such garbage data. garbage in - garbage out. Yet if it could be used correctly it might be more useful. Oh well... Link to comment Share on other sites More sharing options...
aziz Posted December 9, 2014 Share Posted December 9, 2014 That's what was/is so frustrating about it. I have been saying it has a place but I'm still wrong. You either agree 100% or your wrong. Then when it's denied that people make the decisions WITH reasoning involved, well then you have no where to go. you're right, and it shouldn't be that way. these advanced stats have their place and their purpose, but some people need to view them as small parts of the puzzle while others need to view them as parts of the puzzle in general. they are there, and they contain information. definitionally vague information, but still information. they should inform opinion, but not form it. as you say, people seem to want to go full tilt in one direction or another. Link to comment Share on other sites More sharing options...
idahophilly Posted December 9, 2014 Share Posted December 9, 2014 I'm not an advanced stats guy, and I'm not entrenched. I don't care much for advanced stats in hockey - I just watch the games and use traditional stats to talk shop. Nothing special. I know about corsi. I've read about corsi. But I never really looked or followed that measurement at all. I do use stats on a regular basis at work, though, and they're important to me in that world. Anyway, I apologize if you got the impression that I wasn't being reasonable - it was not my intention. No harm no foul. I swim in stats. I can't get away from them. But I learned to be VERY careful with them. If I get it wrong it most likely costs 6 to 7 figures and months of development time. Hence the careful dance with them. And they can lie. My bosses expect me to point out where the stats are not telling the whole story. Anyway, no issue here Brelic... Link to comment Share on other sites More sharing options...
Podein25 Posted December 9, 2014 Share Posted December 9, 2014 and that's what these relative applications of corsi and fenwick try to do. they just start with such garbage data.What would good data look like? I'm not convinced it's measurable. And yet observable... Link to comment Share on other sites More sharing options...
brelic Posted December 9, 2014 Share Posted December 9, 2014 as you say, people seem to want to go full tilt in one direction or another. Are you talking about the push in the league for teams to hire stats nerds on staff and churn out this stuff? Where did this push even come from? Link to comment Share on other sites More sharing options...
idahophilly Posted December 9, 2014 Share Posted December 9, 2014 What would good data look like? I'm not convinced it's measurable. And yet observable...It's kinda like a black hole then. It can't be directly observed (in this case measured) yet you know its there by watching what happens around it..... Link to comment Share on other sites More sharing options...
idahophilly Posted December 9, 2014 Share Posted December 9, 2014 Are you talking about the push in the league for teams to hire stats nerds on staff and churn out this stuff? Where did this push even come from? Both I guess. Teams wouldn't be doing due diligence if they didn't hire some experts to break it down. I think it is the fans that are getting carried away with it. But, that is what fans do. We are allowed to get carried away so long as anyone is prepared to have someone disagree with them. However, I don't think the teams are using it the way the fans are. No proof of that I realize but i suspect they wouldn't give away their methodology publicly anyway. Link to comment Share on other sites More sharing options...
aziz Posted December 9, 2014 Share Posted December 9, 2014 What would good data look like? I'm not convinced it's measurable. And yet observable... good data for that? i agree, i don't know that it exists. i mean, other than just looking at linemate production before/after the addition of the guy being looked at. then you have to factor in all the other things that might have an influence...... really, eyeballs are your best friend there. and point totals. i don't now what gives better insight into how a line combination works than watching them play and seeing how they produce. people will try with a bunch of squint-inducing statistics, but there isn't a real one-for-one proxy. Link to comment Share on other sites More sharing options...
idahophilly Posted December 9, 2014 Share Posted December 9, 2014 Another unknown (at least to me) is this: Are the corsi stat crunchers the teams are hiring hockey people or are they just "anything" stats guys? I could see that making a difference, good and bad.. Link to comment Share on other sites More sharing options...
Podein25 Posted December 9, 2014 Share Posted December 9, 2014 It's kinda like a black hole then. It can't be directly observed (in this case measured) yet you know its there by watching what happens around it.....Yeah, you watch the tape with the coaches and realize you were out of position, or should have been more decisive, or had a brain cramp, or fanned on it, or the ref got in your way, or you were just plain unlucky.I'm not convinced any of that can be "measured". But it can be observed on video and adjustments can be made etcLike they have since Beta came on the market. Link to comment Share on other sites More sharing options...
idahophilly Posted December 9, 2014 Share Posted December 9, 2014 good data for that? i agree, i don't know that it exists. i mean, other than just looking at linemate production before/after the addition of the guy being looked at. then you have to factor in all the other things that might have an influence...... really, eyeballs are your best friend there. and point totals. i don't now what gives better insight into how a line combination works than watching them play and seeing how they produce. people will try with a bunch of squint-inducing statistics, but there isn't a real one-for-one proxy. That's what i was saying about the human influence in the decision process. When a coach makes line adjustments he is going mostly off what he sees, not what corsi says the past three months were and what the player might do 3 weeks from now. Link to comment Share on other sites More sharing options...
idahophilly Posted December 9, 2014 Share Posted December 9, 2014 If I see Berube calling his time out to look up the next best corsi numbers for new line combo's I'll freak out! Link to comment Share on other sites More sharing options...
brelic Posted December 9, 2014 Share Posted December 9, 2014 If I see Berube calling his time out to look up the next best corsi numbers for new line combo's I'll freak out! Hextall publicly stated that Raffl had the best WOWY numbers with Giroux and Voracek last year so they were slotting him in as top line winger. Link to comment Share on other sites More sharing options...
brelic Posted December 9, 2014 Share Posted December 9, 2014 good data for that? i agree, i don't know that it exists. i mean, other than just looking at linemate production before/after the addition of the guy being looked at. then you have to factor in all the other things that might have an influence...... really, eyeballs are your best friend there. and point totals. i don't now what gives better insight into how a line combination works than watching them play and seeing how they produce. people will try with a bunch of squint-inducing statistics, but there isn't a real one-for-one proxy. I think it can be done; it's just that it hasn't yet. I can imagine some sort of player meta-index that is built precisely with the 'line unit' as a whole. So, for example, when 10 is on the ice, and also 28 and 93, here is how he performs (could be everything from G, A, PTS per 60, or PIMs/60, or SOG/60, blocked shots/60, whatever) combined with a volatility index (i.e. the degree to which a player's output varies with different linemates). That would be a clue as to how much a player is dependent on particular linemates versus someone who slots in well almost anywhere. I imagine a guy like G would have a low volatility index, and a guy like Simmonds would have a very high volatility index. Use the volatility index as a multiplier for the meta-index (which I think would be similar to the WOWY idea but without needing a matrix table to look things up... I want ONE helpful value). Link to comment Share on other sites More sharing options...
pilldoc Posted December 9, 2014 Share Posted December 9, 2014 Ok, I'm gonna through my hat into this ring for just this one time. I will admit I know very little of all the complexities of the Fenwick / Corsi numbers. I read throught he majority of this thread and did some research and reading on my own in order to try to understand this whole advance statistics idea. Personally, I think there is a place for advanced statistics, however, it should be used as tool. It is not to replace sound hockey judgement. Just like at my job, we have tools to aid us in the proper dosing of certain antibiotics (ie. Vancomycin), and we use a formula to help predict the drug level. Just because the drug level may tell us one thing, a provider may choose to ignore that level, and dose the drug based upon the patient's current condition. Same with advance hockey stats. I think they do have a place in the hockey world, BUT, they have to be understood and used by the right people in the right circumstance. Personally, I get what the Corsi / Fenwick scores are trying to accomplish, but to be perfectly honest, I think there are way too many variables. I get the basis of Corsi is 5 on 5 hockey but does Corsi take into account these situations: (i'm asking becaue I really do not know)5 on 44 on 45 on 36 on 5 (extra skater end of game)6 on 4 (extra skater at end of game when the other team is a man down)I will admit there is alot about Corsi I do not understand, though I am trying. I agree with Idaho though, you can't jsut put 100% faith in what the Corsi number tells you. It is a guide, but it should not be the sole determinant of a players worth. That comes from scouting a player correctly, speaking with the teammates / coach (especially if the player is a junior player or not), and speaking with other scouts. I do believe advance stats do have a place in the NHL in spotting trends, but they cannot be the sole basis of a players worth. Simply they cannot replace sound hockey judgement. ok..that is my 2 cents worth...I'm not even sure I made any sense....but at least I tried. Link to comment Share on other sites More sharing options...
brelic Posted December 9, 2014 Share Posted December 9, 2014 Personally, I think there is a place for advanced statistics, however, it should be used as tool. you can't jsut put 100% faith in what the Corsi number tells you. I don't think anyone said otherwise. It's not that numbers lie, like the cliche goes - it's that numbers are just numbers. They're decontextualized. So in your case, if a certain medication says that 10% of patients with diabetes suffer from side effect A, and you know that the patient has already exhibited side effects with a similar medication in the past due to his diabetes, you're probably going to recommend against taking that medication. The numbers don't lie - 10% suffer side effects. But you have additional information that contextualizes it. I thought this thread was discussing whether or not advanced statistics were valid models, not whether or not they should form 100% of your decision-making. Link to comment Share on other sites More sharing options...
Recommended Posts
Archived
This topic is now archived and is closed to further replies.