I disagree wholeheartedly. All they do is quantify what you see during the game in different ways. Can they be helpful if you only have but so much time to dissect video? Yes. But a (good) coach or whoever needs to see the video finds the time to properly dissect it.

They have limited uses and there popularity has become almost annoying. I've found most people don't even understand them all that well, as they try to use them to make certain arguments hit home that make absolutely no sense.
- mochoson

Question: Why can't a coach/GM/fan watch the tape AND have extensive numbers on hand?

I disagree wholeheartedly. All they do is quantify what you see during the game in different ways. Can they be helpful if you only have but so much time to dissect video? Yes. But a (good) coach or whoever needs to see the video finds the time to properly dissect it.

They have limited uses and there popularity has become almost annoying. I've found most people don't even understand them all that well, as they try to use them to make certain arguments hit home that make absolutely no sense.
- mochoson

Dead on! And that has been illustrated perfectly here in the last few days.

Question: Why can't a coach/GM/fan watch the tape AND have extensive numbers on hand?
- Tomahawk

He can, that's not the argument. The argument is looking at the numbers in the proper context and interpreting them properly. And taking into consideration the human element that comes into play, such as confirmation bias and correlation versus causation. They can certainly tell a Coach what happened, but they can't tell the Coach why they happened. For that, more indepth analysis is required that advanced stats simply can't equal. As much as 40% of the data in Corsi can be attributed to false positives, and false negatives. It does a better job then traditional plus/minus because it increases the sample size. But it has the same pratfall that plus/minus has.

It's ridiculous.

With a few cities ready to support a team, no matter how good or bad they are in waiting this is a joke.

We missed an entire season & another 1/2 a year of the same rhetoric just to see things never really changed regarding the health of all the franchises.
But then again, it never really was about all that, it was about guys like Snider & Jacobs ensuring their pile of money continues to grow while other teams exchange tickets for dumpster material.
- puckhead17

There are only a handfull of teams that will continue to generate money and draw fans even if they are losing. Even some original six teams like chicago were finding it hard during tough times...

There will never be a time when all teams will be making money. Maybe teams like the Flyers, Ranger, Habs, and Toronto can continue to make money despite losing seasons, but most other teams cant. Any small market team needs to have a winning team, plus a good marketing strategy... Like Nashville. Of every other team, its ebbs and flow, and the only way to get attandance is to win.

Appandax... there is only 1 team that dispite having such a tradition of winning still does not have a fanbase, the Devils. They are the only CITY that DONT deserve a franchise. While there are OWNERS that dont deserve teams.. Likethe former panthers and former Phoenix owners.

Question: Why can't a coach/GM/fan watch the tape AND have extensive numbers on hand?
- Tomahawk

No ones saying they can't.

But a smart coach/GM/fan will watch the tape, so that when their handed any data they know precisely how to analyze it. The data is a cliff note of what happened in the tape.

It's like me saying I'm going to read only 1 in every 3 chapters of a 30 chapter novel. I may get the basic story by the books end, but there's so much substance in the middle that I've missed that there's no way I didn't miss something critical.

Basically, the only people that I'd ever except advanced stats from is from someone who has seen actual video of whatever they are applying the stats too. If not, all your doing is regurgitating a fraction of the whole story to me, something I could have done myself.

Part of the problem is, unless you have:

- birds-eye view of the full ice surface (TV crops a lot of the action out),
- the ability to follow all 12 players on the ice simultaneously,
- watched every game, for every team, across all seasons,
- have a encyclopedic, photographic memory of every little play (not just memorable goals, hits, gaffes)
... you're going to need stats. Advanced stats are no different than regular stats... they just provide more specific answers to more specific questions.

They do not replace watching the games. But they do tell you a whole heck of a lot more about the games/plays/seasons you missed/forgot than the traditional stat line.
- Tomahawk

That is why teams have scouts. They go to the games, and GMs talk to these guys who give an evalutaion. I actually highly doubt a GM gives more then a peak at any advanced stats when determining trade value. And that alone should tell you a lot.

Why can't hockey players just be robots? That way, statistical analysis would be 100% valid and the be-all, end--all?

He can, that's not the argument. The argument is looking at the numbers in the proper context and interpreting them properly. And taking into consideration the human element that comes into play, such as confirmation bias and correlation versus causation. They can certainly tell a Coach what happened, but they can't tell the Coach why they happened. For that, more indepth analysis is required that advanced stats simply can't equal. As much as 40% of the data in Corsi can be attributed to false positives, and false negatives. It does a better job then traditional plus/minus because it increases the sample size. But it has the same pratfall that plus/minus has.
- MJL

Why do you act like Corsi is the only stat? Forget about shots. You still never provided any explanation as to why Grossmann is on the ice for so many more even strength goals than his regular partner Streit, and why he is the worst defensemen on the team at making successful passes out of the defensive zone.

Why can't hockey players just be robots? That way, statistical analysis would be 100% valid and the be-all, end--all?
- jmatchett383

Well, what we need to do is if team A has 6 20 goal scorers, and team B only has 3... We need to give team B a 20 goal scorer from team A to make everything fair.

Why do you act like Corsi is the only stat? Forget about shots. You still never provided any explanation as to why Grossmann is on the ice for so many more even strength goals than his regular partner Streit, and why he is the worst defensemen on the team at making successful passes out of the defensive zone.
- Feanor

I think a lot of people have given the answer... its because Grossmann is struggling in those areas of his game this season. I don't need advanced stats to tell me that. But by reading those numbers alone, you throw out the blocked shots, and other good stuff he does.

Why do you act like Corsi is the only stat? Forget about shots. You still never provided any explanation as to why Grossmann is on the ice for so many more even strength goals than his regular partner Streit, and why he is the worst defensemen on the team at making successful passes out of the defensive zone.
- Feanor

I'd love to hear this explanation, because this already makes absolutely no sense.

He can, that's not the argument. The argument is looking at the numbers in the proper context and interpreting them properly. And taking into consideration the human element that comes into play, such as confirmation bias and correlation versus causation. They can certainly tell a Coach what happened, but they can't tell the Coach why they happened. For that, more indepth analysis is required that advanced stats simply can't equal. As much as 40% of the data in Corsi can be attributed to false positives, and false negatives. It does a better job then traditional plus/minus because it increases the sample size. But it has the same pratfall that plus/minus has.
- MJL

Because everything you see with your eyes are exempt from confirmation biases?



Case in point: is that a water stain or a miracle image of the virgin mary? It would depend on who you ask, how they were raised, the ideas in their heads.

The difference between stats and eyeball test can be boiled down to one thing: numbers provide a universal starting point for debate and interpretation. Whereas, even if you're watching the same play, from the same viewpoint, two people can feel entirely different about the same play.

I'm not a fan of advanced stats. I do think they can be effective and helpful, but I will take the eyeball test any day, in a game like hockey. For instance, advanced stats may show grossman as the worst defensemen, but it doesn't take account how many times Grossman is on tehe ice with players. A, b, c and d. Or how those players were playing defensively.

I'm not discounting them because I don't know many details of them, I just think I can watch te game and know Grossman isn't as bad as been said. Just my opinion I'm not taking sides lol

I'd love to hear this explanation, because this already makes absolutely no sense.
- mochoson

Just off the top of my head, maybe because the defensive pairs have been broken up a lot over the season... so Grossmann could have had those goals go in while with a different defensive partner..

Or Grossmann is a PKer and a lot of goals may be scored on a shift right after a PK ends, before the normal partner is back on the pairing.

Or Streit plays on the PP and might take a shift off while Grossmann plays with someone else, and sometime a team responds after a good kill.

That is why teams have scouts. They go to the games, and GMs talk to these guys who give an evalutaion. I actually highly doubt a GM gives more then a peak at any advanced stats when determining trade value. And that alone should tell you a lot.
- youarewrong

Teams only employ so many scouts... that's a limited number of eyeballs to see everything with. That's especially evident with prospects, who only get a dozen or so viewings from a team's scouts if they're lucky. Later-round guys only get probably 1-2 viewings.

I don't know about you, but I'd like as detailed numbers as possible for the large swaths of the season where my scouts weren't able to be there.

Because everything you see with your eyes are exempt from confirmation biases?

Case in point: is that a water stain or a miracle image of the virgin mary? It would depend on who you ask, how they were raised, the ideas in their heads.

The difference between stats and eyeball test can be boiled down to one thing: numbers provide a universal starting point for debate and interpretation. Whereas, even if you're watching the same play, from the same viewpoint, two people can feel entirely different about the same play.
- Tomahawk

That is why they hire neutral 3rd parties to research the players live... And these scouts give the review to the GMs. It takes out a lot of the bias.

Teams only employ so many scouts... that's a limited number of eyeballs to see everything with. That's especially evident with prospects, who only get a dozen or so viewings from a team' scouts if they're lucky. Later-round guys probably 1-2 viewings.

I don't know about you, but I'd like as detailed numbers as possible for the large swaths of the season where my scouts weren't able to be there.
- Tomahawk

Scouts also get information from other scouts. They cant be at every game, but its normal that one guy is. If homer is interested in a player the first thing he is going to do is call his scout. The first thing the scout is going to do, is call another scout that sees that guy a lot, and get a baseline on the guy. The second thing he is going to do is go see the guy live and back up what the first scout said.

Look, I'm not saying advanced stats dont do anything. I think for the most part the re-inforce what people are seeing. Such as a coach sees that Grossmann is struggling in certain areas... He has some nerd run the advanced stats and it confirms his suspicions. They are a useful tool to re-inforce IMO. When it becomes an issue to me, is when someone just takes an advanced stat (Reporter/Fan) and draws conclusions from that.

No ones saying they can't.

But a smart coach/GM/fan will watch the tape, so that when their handed any data they know precisely how to analyze it. The data is a cliff note of what happened in the tape.
- mochoson

You kind of implied they couldn't... like a good coach would watch ALL the tape and eschew any cliff notes.

In practice, it's probably a little more complicated than that. Tape gets processed, flagged and indexed in real-time by the video staff... coach can then watch the entire thing, or just the angles/clips that he wants. If a team is keeping track of metrics, somebody's entering on-ice events into a spreadsheet as well, from which reports are generated for the GM/Coach to consume later.

With so little time between games/periods, the teams with the best cliff notes have a big advantage.

Why do you act like Corsi is the only stat? Forget about shots. You still never provided any explanation as to why Grossmann is on the ice for so many more even strength goals than his regular partner Streit, and why he is the worst defensemen on the team at making successful passes out of the defensive zone.
- Feanor

Sure I have, I replied to those posts on the other thread when you asked them then. I guess you missed them. And it's for the same basic reason. Hockey is a team game, and none of the stats can isolate one player. How many of those ES goals scored when Streit wasn't his partner, were Grossmann's fault? A defenseman is not the only player involved in puck clears. It's the same flaw with Corsi and any of the derivatives. Only one way to truly tell. Watch the game.

Bill is Fired

Sure I have, I replied to those posts on the other thread when you asked them then. I guess you missed them. And it's for the same basic reason. Hockey is a team game, and none of the stats can isolate one player. How many of those ES goals scored when Streit wasn't his partner, were Grossmann's fault? A defenseman is not the only player involved in puck clears. It's the same flaw with Corsi and any of the derivatives. Only one way to truly tell. Watch the game.
- MJL

Thats another good one... A player would allow more goals because he isnt comfortable with the other partner... they may have more miscues....

Because everything you see with your eyes are exempt from confirmation biases?



Case in point: is that a water stain or a miracle image of the virgin mary? It would depend on who you ask, how they were raised, the ideas in their heads.

The difference between stats and eyeball test can be boiled down to one thing: numbers provide a universal starting point for debate and interpretation. Whereas, even if you're watching the same play, from the same viewpoint, two people can feel entirely different about the same play.
- Tomahawk

I agree totally. But with a sound knowledge of hockey principals and defensive zone coverage, the percentage of determining the cause of a play gone wrong are infinitely higher then simply using Corsi or any other analytic.
And at least we're making progress here and it's been admitted that Corsi based stats require interpretation and context, rather then simply being black and white as one poster states they are.

I agree totally. But with a sound knowledge of hockey principals and defensive zone coverage, the percentage of getting the cause of a play gone wrong are infinitely higher then simply using Corsi or any other analytic.
And at least we're making progress here and it's been admitted that Corsi based stats require interpretation, rather then simply being black and white as one poster states they are.
- MJL

I could be wrong, but from following the debate, I don't think he meant the conclusions/interpretations were black & white, only that the numbers themselves were self-evident.

I don't believe in the 100% infallibility of these new metrics -- there's ample room for improvement -- however people who refuse to allow for the benefit of analytics in hockey are just as blind as these so-called people who 'don't watch the games'.

Was going to sell my tickets for Thursday's game, but I think I might want to go if they do something to commemorate Keith Allen's passing.

I think a lot of people have given the answer... its because Grossmann is struggling in those areas of his game this season. I don't need advanced stats to tell me that. But by reading those numbers alone, you throw out the blocked shots, and other good stuff he does.
- youarewrong

But he only blocks shots because he gave the puck away. That is FACT!