A system of grading ice quality would have to be created to begin with. Each NHL rink has unique weather, refrigeration systems, maintenance teams and many other things that would affect the starting product.

The quality deteriorates as the game goes on. At it's most basic, this could roughly accounted for by breaking down every period into smaller segments, allowing for more snow and ruts. For more accurate results, analysts would need to find a better way to measure the degradation.

The ice continues to degrade (slightly) every period of the game. The short time that the ice gets to "set up" between periods is not as effective as the long time that the ice gets between events. Ever hear announcers mention bad ice caused by last night's concert or circus at the arena?

Is the game fast paced with a lot of stops and starts? That's going to degrade the ice faster than a slow, methodical game will.

Get the picture? Even the small things need to be measured and included in analysis. At least they do if you want an accurate analysis. Right now nobody is measuring anything except the very largest pieces of data. They are ignoring all of the underlying qualifiers.

I want to get those answers that analytics experts are promising. The problem is, nobody has enough data to give reliable and accurate analysis.
- sportsjunkie007

All i can see is that you are complaining about unquantifiable variables.

As for the bad ice, yes, i understand what you are getting at, but bad ice effects all the players.

Actually, its not a coin toss, if you actually bothered to read his blog.
- JShack

Actually it is when you are talking about guys that are only going to play 16-18minutes.

I don't know about that, with the spotlight not so much on CrabMan, he may look like one of the best D you have after the nine games.

I think he stays, Drysattle too.
- Byfuglien Ate Me

- saskoil21

I was going to laugh, but didn't want to pump his tires too much.

I don't know that Nurse even gets 9, but what the hell happens if he does end up staying? Really throws a wrench in to the "plan" for the defense.

I don't think he should stay pass those 9, but if he brings a physical brand of defense while not hurting the team it's going to be hard to send him back.
- Lahey

Please explain what a "physical" brand of defence does for a team? Do the San Jose Sharks play a physical brand of defense? is that why they are just an amazing team. Is that why every single time the oilers play the Sharks we get absolutely dominated and NEVER leave our own end. Ya... thats probably it.

I was going to laugh, but didn't want to pump his tires too much.
- Lahey

It's not funny when that's the name people call your wife.

Actually it is when you are talking about guys that are only going to play 16-18minutes.
- Lahey

You think that Marincin was only going to play 16 minutes this year? Well, only a terrible coach would play, arguably, their best puck moving defenceman for 16 minutes. But i guess for those 16-18 minutes we should employ reallllllllly good looking dmen that will bring some cool physicality to the position. Because physicality wins games! ask the leafs

Please explain what a "physical" brand of defence does for a team? Do the San Jose Sharks play a physical brand of defense? is that why they are just an amazing team. Is that why every single time the oilers play the Sharks we get absolutely dominated and NEVER leave our own end. Ya... thats probably it.
- JShack

Having a d-man that punishes the opposition goes a long way. Players love playing against us as we are too easy to play with. There is no concern of going in the corners or making a high risk play with your head down or back turned.

We just don't have enough guys that wear the other team down.

You think that Marincin was only going to play 16 minutes this year? Well, only a terrible coach would play, arguably, their best puck moving defenceman for 16 minutes. But i guess for those 16-18 minutes we should employ reallllllllly good looking dmen that will bring some cool physicality to the position. Because physicality wins games! ask the leafs
- JShack

Ah so you thought he was going to play more minutes with the addition of Fayne and Nikitin?

I also love how the guy with 6 assists last year is our best puck mover. Especially with that Schultz kid on this team.

MacT is a lucky bumb,Pitlick and Pinnizorro clear waivers....

You think that Marincin was only going to play 16 minutes this year? Well, only a terrible coach would play, arguably, their best puck moving defenceman for 16 minutes. But i guess for those 16-18 minutes we should employ reallllllllly good looking dmen that will bring some cool physicality to the position. Because physicality wins games! ask the leafs
- JShack

Okay I think marincin is better than klefbom right now, but you make it seem like we are going from karlsson to Fraser

MacT is a lucky bumb,Pitlick and Pinnizorro clear waivers....
- AlEx_OiL

Have you started day drinking? HA isn't even that bad.

Have you started day drinking? HA isn't even that bad.
- ChetManly

I don't know about that.

Ah so you thought he was going to play more minutes with the addition of Fayne and Nikitin?
- Lahey

If hes better, i ABSOLUTELY think that he will play more... you clearly know notihng if you think big hard hitting Dmen are still important in todays game. Please ask yourself this. Who is harder to play against. Erik Karllson or kevin Bieksa. Now if you dont like that one. How about this. Who would you rather match up against as a forward. Lindstrom or darrien hatcher.

If you like playing in your own d zone, then im guessing youll choose karllson, lindstrom and marincin.

You are brilliant

So everyone here stops arguing Yes the MM to OKC was a DUMB mistake by MacT And Eakins end of discusion.

If hes better, i ABSOLUTELY think that he will play more... you clearly know notihng if you think big hard hitting Dmen are still important in todays game. Please ask yourself this. Who is harder to play against. Erik Karllson or kevin Bieksa. Now if you dont like that one. How about this. Who would you rather match up against as a forward. Lindstrom or darrien hatcher.

If you like playing in your own d zone, then im guessing youll choose karllson, lindstrom and marincin.

You are brilliant
- JShack

I was talking about a #6 d-man, but you obviously can't comprehend. Why not compare Karlsson to Weber if you are going to make a comparison. Not a #4 d-man on his way down.

Have you started day drinking? HA isn't even that bad.
- ChetManly

Nope just coffee here at my office,I just didn't like the Pitlick call,you got a pb with that?

Great blog.

I was, like most people, a big fan of Klefboms from stuff I heard and seen from the WJC and his stint in the SEL. However, upon actually watching him play and following him in the AHL he has been disappointing. Marincin has been exactly the opposite, while not perfect by any means he has impressed at every level.

Even in preseason it seemed fairly obvious to me Marincin is head and shoulders above Nurse, Hunt, Aulie, and Klefbom.

I agree it is a foolish move for a coach/GM to get rid of one of their best defenseman to start the season.

However. I can see why they have done it.

I sincerely doubt Klefbom will see much, if any playing time. Even with the injury to Nikita the amount of actual playing time any of our depth defenseman will see will be minimal. Nikita could actually start the season from comments made by Eakins.

Nurse will get a legit show casing this year and IMO deserves it. Besides Marincin Nurse was clearly the next best and getting better every game. But barring the guy scoring a point per game and being +20 he will be sent down for development after his 9 game stint. If Marcincin was on the opening roster it's very tough to rationalize playing Nurse over him.

Klefbom and Hunt get their cup of coffee during the easiest part of a difficult NHL schedule. Then Klefbom heads back to the AHL to work on his game playing big minutes and Hunt rarely sees the outside of the pressbox.

Could be wrong but won't the Oilers probably run 14 forwards 7 defense and 2 goalies once Nikita and Gazdic are free from injuries?

Ah so you thought he was going to play more minutes with the addition of Fayne and Nikitin?

I also love how the guy with 6 assists last year is our best puck mover. Especially with that Schultz kid on this team.
- Lahey

Schultz played 25 minutes per night, is 3 years older and got ALL the PP time. how are their offensive numbers even comparable? Im not here to argue if marincin is more offensively gifted than Schultz. But hes a bigger asset to our d core than hot body Klefbom

Okay I think marincin is better than klefbom right now, but you make it seem like we are going from karlsson to Fraser
- CrustyRooster

More like Chara to Karlsson

If hes better, i ABSOLUTELY think that he will play more... you clearly know notihng if you think big hard hitting Dmen are still important in todays game. Please ask yourself this. Who is harder to play against. Erik Karllson or kevin Bieksa. Now if you dont like that one. How about this. Who would you rather match up against as a forward. Lindstrom or darrien hatcher.

If you like playing in your own d zone, then im guessing youll choose karllson, lindstrom and marincin.

You are brilliant
- JShack

And Karlsson isn't hard to play against. The guy gets easy zone starts and bleeds goals.

I was talking about a #6 d-man, but you obviously can't comprehend. Why not compare Karlsson to Weber if you are going to make a comparison. Not a #4 d-man on his way down.
- Lahey

Ahhhh so klefbom and his ZERO points and 5 shots in preseason make him a weber comparable... Great

how can Klefbom and Marincin currently be on the same level in some minds? Marincin was playing top lines (again, most common opponents Getzlaf, Perry, D Sedin) and performing better than the rest of the club for over half a season. Klefbom played 17 games against softer competition and was at about the average team level.

If you're basing anything off of 6 pre season games then don't even bother responding

Ahhhh so klefbom and his ZERO points and 5 shots in preseason make him a weber comparable... Great
- JShack

You need to go back and start re-reading. You are going to steinley levels.

Somehow Marincin and Klefbom = Weber and Karlsson. You really are a new level on here

A GOOD statistician will acknowledge the fact that their analysis is subject to misinterpretation. They will also admit that their analysis is only as good as the data that they have to work with.

Far too many analytics supporters want to use only the information readily available at their fingertips, ignoring an extremely large number of unanswered questions and then come to absolute conclusions.

So many things that affect each play are never accounted for. THAT is the shortcoming of analytics. Ice degradation, player injuries during game, muscle fatigue, mental issues and hundreds of other things will make a difference on every single play. None of these things are currently accounted for by analytics.

Analytics experts need FAR more quantifiable data if they want to claim accurate results. I'm not talking about basic mathematic facts, I'm talking about true analysis. Until analysts can actually gather more information to build an accurate picture, perhaps they should stop trying to claim that they can tell the rest of us the future.
- sportsjunkie007

you cant measure everything. But you can measure indicators....indicators of positive outcomes. Nothing that's measured anywhere in the world is done by measuring all influential components of the event process. You use statistics to find indicators that seem to be related to a positive outcome, even though we may not know why it happens like that or are able to explain any part of it. Science works that way, medicine works that way. We don't always know why, or the exact mechanics of why something happens, we just know it IS, and it DOES, and that we can use it for reliable prediction of the future. Waiting for every part of that relationship to be completely explored and explained before acting on a positive predictive trend will hinder progress. This is what is happening with corsi and a lot of other advanced stats. Its an indicator of positive outcome, saddled with a large black box. We don't know what's going on in there, but we know its generally accurate, so why do we care about whats in that black box? Sure work on figuring out what in there, but don't wait for those things to be sorted out before you use it. If it proved to be significantly predictive, use it! Should we spend 20 years collecting better data to explain all the reasons why it works first? Or should we just spend those 20 years using it to pick better players.


In a randomized controlled trial of a drug, if unexpected positive or negative outcomes occur that are proven to be real, the trial is immediately stopped or changed even if determinants of these outcomes remain unknown. If science and medicine can do this, why cant hockey. I literally cannot understand this.