Oh I know, definitely not disagreeing with that hahah one thing I will say, his blogs/comments are always entertaining, given how outspoken and over the top he is. Might not always agree with him, but it is entertaining.
- j.boyd919

Very true

I trust that the numbers are accurste. I also believe that Corsi is in fact important. But how one can sit here and say over a cDiourse of a season Reilly and Gardiner even out, when Reilly has played an entire games worth of ice time more against Sidney freaking Crosby than Gardiner is mind bottling. Maybe the difference is over stated, but it's a difference, and that cannot be denied
- sbroads24

100% there's a difference...a big difference in some cases.

You lost your entire argument when you said this.

Delete your account.
- Aetherial

Nah. Not really, there's been tons of studies, research, and articles (that I referenced) that back that up.

I trust that the numbers are accurste. I also believe that Corsi is in fact important. But how one can sit here and say over a course of a season Reilly and Gardiner even out, when Reilly has played an entire games worth of ice time more against Sidney freaking Crosby than Gardiner is mind bottling. Maybe the difference is over stated, but it's a difference, and that cannot be denied
- sbroads24

I think those stats geeks just underrate the difference between the top echelon of players and the second line guys. Yeah they're close and there's the odd team with two elite players on different lines, but the typical gap between them is the difference between 80-90 pt seasons and 60 pt seasons. Fairly large difference there even if their skills are virtually the same.

100% there's a difference...a big difference in some cases.
- Garnie

Lots of factors, including home/road games, last change that over a large sample size, can even out. What people who rely on stats can't seem to realize is that in hockey, coaches can't rely on statistical means or averages. It takes one shift to change the outcome of a game. If it didn't matter, coaches wouldn't work so hard to find combinations and match ups that make a difference. In the real world, on the ice, quality of competition matters a lot.

I think people don't realize that those of us who use stats also love and watch a ton of hockey, read every blog, article and magazine, etc. We're nerds.

But to answer your question I'd turn it around: give me an example in the world, it can be anything, of a thing you would evaluate, analyze or compare where there is an obviously better way to do so than measuring and recording the data and then anaylyzing it.

So, while ideas, tracking, methods, theory, analysis etc can and will improve, there will never be a better way to evaluate anything than by measuring, recording and analyzing the data. It is an impossibility.

I also don't think anything is "inarguable" but if you don't agree 5v5 offense and improving the play of your linemates are two of the most valuable criteria to judge a hockey player on, then what is?

It is inarguable that Hall is among the best of the best at those two things. Add n the fact that he is a top 5 fastest skater and decent at defense, and what would your argument that he is bad, or that I'm overrating him even be?
- James_Tanner

I mean you completely sidestepped what I thought was a pretty reasonable question and pretty much just got up on a soapbox. I'm not, and didn't, argue against anything in any way, shape, or form.

My question was simply: "Could you concede the possibility that maybe advanced statistics play a slightly less useful role than you and other staunch proponents think?" I'm not even saying that I believe that, but I'm curious to hear if you think that there is even the possibility of wiggle room for how useful these things are.

Oh I know, definitely not disagreeing with that hahah one thing I will say, his blogs/comments are always entertaining, given how outspoken and over the top he is. Might not always agree with him, but it is entertaining.
- j.boyd919

This.

Tanner is simply an entertainer. Tanner is to Bob McKenzie (real hockey minds) what Wrestling is to Football, Hockey, Baseball (real sports).

No, sir. You have the predetermined opinion that you're unwilling to change.

In the face of so many people determined to bring common sense to your writing with facts, data and heated debate: you don't even have the humility to admit you were way off from time to time.

You never own up to the mistakes you make and you don't let common sense get in the way once many people prove you wrong in the comments section. Arrogant, Cloutier-like writing.
- fry

Well that just isn't true, I learn in public, which can be embarrassing sometimes, but whatever, I am happy to change my mind whenever I am proven wrong.

But these "common sense" arguments were disputed in the post I linked to written by the guy with the PHD from Harvard and the job in the NHL. So, my apologies about not having the humility to assume every argument is a popularity contest.

This.

Tanner is simply an entertainer. Tanner is to Bob McKenzie (real hockey minds) what Wrestling is to Football, Hockey, Baseball (real sports).
- NugentHallberle

Is Bob McKenzie a "great hockey mind" or a reporter with awesome connections? I actually think McKenzie probably has some weird outdated opinions about stuff, but who knows, one reason everyone loves him so much is he doesn't really get too opinionated. He shoots right down the middle when he does.

Lots of factors, including home/road games, last change that over a large sample size, can even out. What people who rely on stats can't seem to realize is that in hockey, coaches can't rely on statistical means or averages. It takes one shift to change the outcome of a game. If it didn't matter, coaches wouldn't work so hard to find combinations and match ups that make a difference. In the real world, on the ice, quality of competition matters a lot.
- MJL

I wish I could play poker with you.

This.

Tanner is simply an entertainer. Tanner is to Bob McKenzie (real hockey minds) what Wrestling is to Football, Hockey, Baseball (real sports).
- NugentHallberle

Bob McKenzie is a great hockey mind? All he does is report from sources lol he never gives his opinion on anything hahahaha

My question was simply: "Could you concede the possibility that maybe advanced statistics play a slightly less useful role than you and other staunch proponents think?" I'm not even saying that I believe that, but I'm curious to hear if you think that there is even the possibility of wiggle room for how useful these things are.
- eichiefs9

No. People like Tanner just abuse them. Lets put a very fast, very random game played by humans into numbers, and assume the numbers perfectly, objectively paint a picture of what's happening on the ice. There are so many factors going on unaccounted for that there will always be exceptions to the norm and numbers that move away from the mean. There's so much you have to look at besides the hard and soft numbers to get any sort of accurate picture, which is why large sample sizes help mitigate some of these issues.

For what it's worth, the hyper-statistical people who seem to think they are objectively quantified perfectly to the real world...have never played hockey from my experience. never have, never will. and that's okay. But, experience counts for something and playing the game does make you view certain nuances in different lights. You now know why that guy seemed lazy as (frank) on the back check...he was at the end of his shift. Not something a lot of normal hockey fans would ever think about and would just be roasting the lazy backchecker.

Well that just isn't true, I learn in public, which can be embarrassing sometimes, but whatever, I am happy to change my mind whenever I am proven wrong.

But these "common sense" arguments were disputed in the post I linked to written by the guy with the PHD from Harvard and the job in the NHL. So, my apologies about not having the humility to assume every argument is a popularity contest.
- James_Tanner

The guy has as PHd in Chemistry and a BA in Physics and just recently got that job in the NHL. More credible than you or I, but far from an expert. Just cause I have an engineering degree and knew a guy with mental illness doesn't mean I can diagnose your bipolar disorder.

I can also almost guarantee he hasn't played one shift of hockey in his life (similar to Tanner).

100% there's a difference...a big difference in some cases.
- Garnie

This.

I’m guessing that QoC probably evens out for 60%-80% of the league. But there are many exceptions; you can’t just use it as a blanket statement to cover every player in the league.

I also like hero charts for a quick over view of a player. Gardiner’s hero chart appears out of whack with what you see on the ice. When something like this happens, further research can be helpful.

Gardiner’s QoC is clearly below Rielly’s and probably a big reason why his Hero Chart looks out of whack.

In any of these stats does it take into consideration the goalie they were playing against?

I only ask this because I remember when the Hawks sucked they always seemed to play against the backup goalie. So say Hall faced a backup 40% of the games will his stats suffer when he is on a better team and faces the starting goalie?

Does anyone get a weird psycho-sexual vibe from all these bloggers talking about men countless times a week/year?

I studied Psychology for a year, this sounds a lot like Freud.

The guy has as PHd in Chemistry and a BA in Physics and just recently got that job in the NHL. More credible than you or I, but far from an expert. Just cause I have an engineering degree and knew a guy with mental illness doesn't mean I can diagnose your bipolar disorder.

I can also almost guarantee he hasn't played one shift of hockey in his life (similar to Tanner).
- NugentHallberle

One would say mentioning someone's degrees to help prop up their argument is a pretty common logical fallacy. The funny thing being that it wasn't even necessary. Let the argument speak for itself.

From Eric Tulsky, ...

The analogous situation in zone starts would be if everyone’s offensive zone start percentage were between 48% and 52%; such small corrections are scarcely worth the effort, and a person who ignored competition when evaluating players would not be wrong by much....
- James_Tanner

Rielly falls outside of this range for zone starts.

But you THINK its BS. You have a predetermined opinion you are not willing to change.

It's proven. The study is from 2012 and many, many people have replicated it. Almost no players face extreme competition to the point where you have to adjust for it in the analysis of one player vs another.

Despite your examples (and I swear, the next time someone accuses me of cherry picking....) over the course of the season, they faced a minute tenth of a percentage or so difference in the CF% of their competition.
- James_Tanner

clearly rielly faces 1st line competion and Gardiner faces 2nd line competion.

To me, this is pretty significant difference, despite it not showing up in the Corsi averages.

You ever think that looking at averages could have some flaws?

Lots of factors, including home/road games, last change that over a large sample size, can even out. What people who rely on stats can't seem to realize is that in hockey, coaches can't rely on statistical means or averages. It takes one shift to change the outcome of a game. If it didn't matter, coaches wouldn't work so hard to find combinations and match ups that make a difference. In the real world, on the ice, quality of competition matters a lot.
- MJL

My issue is when people insinuate that advanced stats are the end all-be all. They aren't, and although I do enjoy learning about them and believe they are a very good predicator of success, I do realize that winning is the end all-be all, but advanced statistics are just another tool to evaluating players, teams, league trends, etc. The more tools you have, the better chance you have at creating a superior team IMO, so why throw away more tools?

This.

I’m guessing that QoC probably evens out for 60%-80% of the league. But there are many exceptions; you can’t just use it as a blanket statement to cover every player in the league.

I also like hero charts for a quick over view of a player. Gardiner’s hero chart appears out of whack with what you see on the ice. When something like this happens, further research can be helpful.

Gardiner’s QoC is clearly below Rielly’s and probably a big reason why his Hero Chart looking out of whack.
- Steven_Stamkos

Same as players on most crappy teams....good Corsi = easier comp and bad corsi = harder comp...but lets take a couple guys with PHd's who think they can dream up a formula to make them smarter than coaches/GM's. It's entertaining what kind people it's brought out of the woodwork.

One would say mentioning someone's degrees to help prop up their argument is a pretty common logical fallacy. The funny thing being that it wasn't even necessary. Let the argument speak for itself.
- Blackstrom2

Chiarelli has a job in the NHL, so theres that. And for some reason Ken Holland and Jim Benning still have jobs...

Tanner, Hall is known around the NHL as a 'partier'....if you catch my drift. This is a substantiated fact, and the Oilers wanted to get him as far from McDavid, Draisaitl et al as fast as possible.

Eberle is also one of the gang....but the Oil are hoping that by removing Hall things will fix themselves.

This shouldn't come as a surprise to anyone. He has been labeled. It happens in all sports.

Same as players on most crappy teams....good Corsi = easier comp and bad corsi = harder comp...but lets take a couple guys with PHd's who think they can dream up a formula to make them smarter than coaches/GM's. It's entertaining what kind people it's brought out of the woodwork.
- Garnie

rel corsi has it flaws. good for team analysis and comparing players on the same team - to a certain extent. even the guy who runs ownthepukck states it's use:


"it is more appropriate to think of RelTM metrics as a measure of a skater’s ability relative to teammates playing the same position."