Spin-Off Semin Discussion (from draft thread)

[StormCast]

Defensive responsibilities of a C and a W are different. It's harder to play defense as a C than as a W. I never denied that. I've maintained throughout this entire thread that Bergeron is considerably better than Semin without the puck and that I consider Bergeron the best two way forward in the world.

I am 19 and am already getting paid to write about hockey. So if I "don't know what I'm watching" whatever, joke's on the people that give me money I guess.

Yes it is "misguided" to say that good positioning alleviates having to use a lot of effort. Yes you are so right. That totally isn't common sense at all.

Here's the thing about advanced stats, believe in 'em or don't, I'm gonna be right more often than you are on a regular basis

Wow, hold the phone. Your original contention about Semin compared to Bergeron was "he's not far off." Now your say "Bergeron is considerably better than Semin without the puck."

Also, you do realize that you have to use a lot of effort to consistently get into position? I mean, it is a game played across 3 zones last I checked. So in other words, if my D pinches and I'm the high forward having to go back the other way, playing a tight gap while skating backgrounds requires little effort? The things you learn here. Much like your use of stats, you can't use make universal conclusions without full context.

And a 19 year old who knows more than others as proven by the fact that he gets paid. Wow, color me shocked. Welcome to the working world. But tell us how interning for Eklund really is.

 

[StormCast]

I mentioned this before. "They" don't seem interested in these kinds of things.

Correct and when you combine observation bias while using such a group metric to try to discern individual performance, the results are often muddled.

Maybe we need a selfie competition to determine the winner.

Well, I have a Semin selfie stick but I have trouble keeping it on the ice.

 

[Ole Gil]

No one has answered me so I'm assuming my definition is right. Even strength GF vs GA represented in a %. That's +/- (slightly improved to remove ENG) in a form that looks more advanced. If they retro'd Ron Francis' GF% he'd be under 50%. Hell the stat says in the time period referenced that the Sedins are 1 and 2. Does that mean they're in the same league as Bergeron away from the puck too, or better?

The applications of Corsi and fenwick I get. Makes perfect sense to me why those numbers are being used and what merit they have. Trotting out GF% as if that's some kind of major reference is asinine.

This is the result of getting our statistical analysis filtered through journalism majors and ex hockey players. For people who aren't particularly interested in it, or those who are, but don't have the background for it, the results are quite the mixed bag.

 

[Boom Boom Apathy]

This is the result of getting our statistical analysis filtered through journalism majors and ex hockey players. For people who aren't particularly interested in it, or those who are, but don't have the background for it, the results are quite the mixed bag.

Wally with shots fired.

 

[StormCast]

No one has answered me so I'm assuming my definition is right. Even strength GF vs GA represented in a %. That's +/- (slightly improved to remove ENG) in a form that looks more advanced. If they retro'd Ron Francis' GF% he'd be under 50%. Hell the stat says in the time period referenced that the Sedins are 1 and 2. Does that mean they're in the same league as Bergeron away from the puck too, or better?

The applications of Corsi and fenwick I get. Makes perfect sense to me why those numbers are being used and what merit they have. Trotting out GF% as if that's some kind of major reference is asinine.

Yes, but what you're missing, Joe, is that RF can't hold a candle to Semin. Interestingly, he's tried holding a candle to his backside but that hasn't worked either.

Now that...is a hot take.

 

[Blueline Bomber]

I'm also getting paid to write hockey and clearly SC and I are on different sides of this Semin debate. So I'm starting to believe that's got nothing to do with it.

 

[Joe McGrath]

I'm also getting paid to write hockey and clearly SC and I are on different sides of this Semin debate. So I'm starting to believe that's got nothing to do with it.

People who get paid (presumably the most) to write about hockey voted for Ovechkin to be on to year end all-star teams two years ago because they didn't know which wing he played on. I don't think I get paid for it wins anyone any arguments.

 

[Blueline Bomber]

People who get paid (presumably the most) to write about hockey voted for Ovechkin to be on to year end all-star teams two years ago because they didn't know which wing he played on. I don't think I get paid for it wins anyone any arguments.

It's actually quite easy to become a part of the PHWA, so I'm not sure how much one can take from those votes. Obviously, you've got some really respected names in the group, but at the same time, there are reporters who can't even (apparently) cover their own team properly, so expecting them to pay attention to the rest of the league is a tall order.

 

[Joe McGrath]

It's actually quite easy to become a part of the PHWA, so I'm not sure how much one can take from those votes. Obviously, you've got some really respected names in the group, but at the same time, there are reporters who can't even (apparently) cover their own team properly, so expecting them to pay attention to the rest of the league is a tall order.

I agree, but they do get paid to write about hockey. Which is my point.

 

[Anton Dubinchuk]

the journalism majors part is pretty right

stats in the real world are NEVER as definitive as most of the people would like to think

theres a reason 538 exists, and its not because their staff has the particular and rare skill of being able to click the "sort by corsi" tabs at the top of stats columns. to make definitive statements regarding hockey players output, you have to do a good bit (as in, 20x) more statistical work than i have ever seen done on here.

most of the time when people say things like "these two stats prove (X extrapolation)," all im reading is "my B in high school statistics is the extent of my legitimate knowledge of the subject." advanced stats "have their place," as is the popular debate nowadays, but its not like it was billy bean himself doing the math. if you look throughout the history of sports analytics, particularly starting in the 60s with the cowboys, it has ALWAYS been the mathmaticians/stats guys that have done the work. this isnt some bizarre movement coming out of nowhere that you either hop on or you dont. this is a new line of mathematical and statistical application that really does require mathematical or statistical knowledge. thats why whenever i see corsi numbers or zone starts or this or that or whatever, i always look and say "interesting," without ever trying to make definitive statements. there are definitive statements to be made, but it requires work. and sorting by corsi or making those little bubble graphs might feel legit, but unless youve got the understanding of statistics to know what it means and what it doesnt, its little more than non definitive (but interesting) speculation

 

[Anton Dubinchuk]

http://www.ibtimes.com/dallas-cowbo...r-pradesh-helped-create-americas-team-1441358

this is interesting and i referred to it in my last post

im taking a sports history class at tech right now, its fascinating

 

[Finlandia WOAT]

The trend I find humorous is that almost every thread that involves "Stats vs. Eye Test" devolves into the following argument:

"Well [X] and [Y] say this!"

"Yes, but do you watch the games? You can't say anything because you don't watch the games!!!!!"

rinse and repeat for 20 pages.

Especially with defense from forwards....most people don't actually watch games of other teams outside of when they are playing their favorite team. Of those that do, defense is hard to detect because, as Stephen said, often the most noticable plays are not the best indicators of defensive play.

And most casual fans aren't going to pay attention to the play away from the puck...now, StormCast may do that, I don't know, but he hasn't provided any concrete examples of why Semin's play away from the puck is worse than Bergeron's. Not that it isn't, that's just the problem with "Eye Test" arguments... they become little more than appeals to authority.

 

[Anton Babchuk]

if semin isn't great away from the puck, then why was he named captain?

 

[Anton Dubinchuk]

if semin isn't great away from the puck, then why was he named captain?

francis doesnt watch the games

 

[nobuddy]

i never claimed that semin is close to bergeron because i get paid to write about hockey

seriously people good lord reading comprehension is not that hard, but i guess all you statistic major experts don't really know how to read and comprehend english properly (am i doing the condescending thing about area of expertise right)

when i said that i was responding to him saying that i don't know what i'm watching when i watch hockey. he was critiquing me directly there, not my opinion of semin being not way worse than bergeron away from the puck.

i'm not a ****ing idiot who thinks all of my opinions are right because i get paid.

 

[Anton Dubinchuk]

i am not a statistics major

 

[What the Faulk]

the journalism majors part is pretty right

stats in the real world are NEVER as definitive as most of the people would like to think

theres a reason 538 exists, and its not because their staff has the particular and rare skill of being able to click the "sort by corsi" tabs at the top of stats columns. to make definitive statements regarding hockey players output, you have to do a good bit (as in, 20x) more statistical work than i have ever seen done on here.

most of the time when people say things like "these two stats prove (X extrapolation)," all im reading is "my B in high school statistics is the extent of my legitimate knowledge of the subject." advanced stats "have their place," as is the popular debate nowadays, but its not like it was billy bean himself doing the math. if you look throughout the history of sports analytics, particularly starting in the 60s with the cowboys, it has ALWAYS been the mathmaticians/stats guys that have done the work. this isnt some bizarre movement coming out of nowhere that you either hop on or you dont. this is a new line of mathematical and statistical application that really does require mathematical or statistical knowledge. thats why whenever i see corsi numbers or zone starts or this or that or whatever, i always look and say "interesting," without ever trying to make definitive statements. there are definitive statements to be made, but it requires work. and sorting by corsi or making those little bubble graphs might feel legit, but unless youve got the understanding of statistics to know what it means and what it doesnt, its little more than non definitive (but interesting) speculation

This isn't ****ing rocket science. Understanding and applying these statistics is really not difficult.

"Sorting by corsi" yes that's all anyone does around here.

 

[the halleJOKEL]

alexander semin

 

[Anton Dubinchuk]

This isn't ****ing rocket science. Understanding and applying these statistics is really not difficult.

"Sorting by corsi" yes that's all anyone does around here.

appropriate statistical analysis takes a lot more work than many people here think.

all anyone here is qualified to do is make general observations of the numbers, as we would treat goals or assists.

it is clear from what many people post here that they think they are more qualified than they are.

the conclusions people make in a definitive or absolute manner often cannot be appropriately made from the data they provide.

 

[StormCast]

And most casual fans aren't going to pay attention to the play away from the puck...now, StormCast may do that, I don't know, but he hasn't provided any concrete examples of why Semin's play away from the puck is worse than Bergeron's. Not that it isn't, that's just the problem with "Eye Test" arguments... they become little more than appeals to authority.

Everyone, including even the OP now, has claimed Bergeron is "considerably better" than Semin without the puck. And flawed, misapplied stat analyses hardly constitute concrete examples.

when i said that i was responding to him saying that i don't know what i'm watching when i watch hockey. he was critiquing me directly there, not my opinion of semin being not way worse than bergeron away from the puck.

As it relates to Semin's play without the puck vs. Bergeron, anyone who tries to make the case they aren't far apart is indeed missing something. As noted though, I sense there is confirmation basis at work with you and Semin. It appears you've now even realized the comparison was overstated.

all anyone here is qualified to do is make general observations of the numbers, as we would treat goals or assists.

it is clear from what many people post here that they think they are more qualified than they are.

the conclusions people make in a definitive or absolute manner often cannot be appropriately made from the data they provide.

Agreed and too often in the race to make the stats "prove" something, the missing link is context and the questions of credibility are then rightly asked.

 

[What the Faulk]

appropriate statistical analysis takes a lot more work than many people here think.

all anyone here is qualified to do is make general observations of the numbers, as we would treat goals or assists.

it is clear from what many people post here that they think they are more qualified than they are.

the conclusions people make in a definitive or absolute manner often cannot be appropriately made from the data they provide.

I don't think anyone is making definitive or absolute conclusions any more so than the people who go "I watched the Hurricanes play 85% of the season, I'm right no matter what anyone says." We're all just idiots talking on a message board. Some of us get paid to do it for other outlets, but we're all on a pretty level playing field I think.

 

[Anton Dubinchuk]

I don't think anyone is making definitive or absolute conclusions any more so than the people who go "I watched the Hurricanes play 85% of the season, I'm right no matter what anyone says." We're all just idiots talking on a message board. Some of us get paid to do it for other outlets, but we're all on a pretty level playing field I think.

i agree completely. didnt even address the "eye test" in my post because the differences in the conclusions that can be made from that are obvious.

the appeal of the advanced stats "movement" (in quotes because its pretty much been a steady process in sports since the 60s) is that it can objectively quantify things that were previously left to subjective opinion. this is why moneyball was so great and made a great movie, this is why bill James is famous. i think that most people, especially those particularly on the "fancy stats side," get this. what i think a lot of people dont get is that these are still just stats. without the objective knowledge necessary to make these appropriate objective conclusions, "fancy stats" are becoming just another de facto source of subjective opinion.

if anything, im incredibly on the "left" as far as fancy stats go, probably more than most of you, in a way. i think that developed properly they can begin to start make legitimate and game changing objective conclusions about players, similar to WAR or another advanced stats in baseball (although not to the same extent, just the same objectivity). but it requires actual statistical analysis, not something an amateur can throw together in a message board post.

i suppose there is an analogy. i would say that in most cases, someone here claiming to be making a statistical claim (when compared to the work of a legitimate statistician) is similar to the people who watch a few checkers games and say theyre excited about the potential of an up and coming dman (compared to a veteran nhl scout). its uninformed, mostly based on stuff theyve heard from other sources, and may or may not be right. id argue that the statistical stuff is worse though, because with that there is actually a legitimately objective and correct way to do it, whereas "eye test" scouting is still inexact at its highest levels. i can luck my way to a positive player evaluation, i thought john tavares would be a great nhl player and i was right. thats why we are all here, to share our relatively uninformed subjective opinions with one another, and cheer for the same team. its tough to luck your way to getting a math problem right though. but even if you do, if you havent objectively used the stats right on the way there, its not really a valuable opinion. its why math teachers make you show your work. you could guess that the answer is 3 and be right, but if your work leading there was two equations that didnt have anything to do with the problem and then a few question marks, youre not getting credit.

 

[Anton Babchuk]

alex semin is better with and without the puck than patrice bergeron. i have watched both of them play at least three times each.

 

[Boom Boom Apathy]

alex semin is better with and without the puck than patrice bergeron. i have watched both of them play at least three times each.

I think that was Marc-Andre Bergeron you watched.

 

[Anton Babchuk]

I think that was Marc-Andre Bergeron you watched.

this just shows how flawed the eye test is. i couldn't even remember who i watched. how am i supposed to remember details?