Larry Brooks: Marc Staal wants to stay in New York

[Richter Scale]

John Moore is a better possession player than Klein. And I'm no fan of Moore, but the results are clear.

Moore without Klein: 52.2%
Moore with Klein: 49.7
Klein without Moore: 45.9%

Klein has a unique talent for getting his team pinned in its own zone.

This is what I'm talking about. You're just saying "look stats, so ___ is true" without digging any deeper and examining why those stats may have turned out the way they did. In my opinion, that's lazy.

Was Moore playing sheltered minutes when he wasn't on the ice with Klein? Was Klein playing against a higher quality of competition when he wasn't on the ice with Moore? Was Moore getting a higher percentage of offensive zone starts compared with Klein? When Moore was separate from Klein, was he generally paired with better defensemen than Klein was when he was separate from Moore? When they were separate, was one of them playing with poorer quality forwards more often than the other? Was John Moore kept away more from important and difficult defensive situations and zone starts than Klein was? Were the times when they weren't on the ice together moments of the game when one of them either desperately needed a change, or they were the only d-man able to make a change so they were stuck out there with 4 other tired guys? And did that happen more often for one of them based on their ability to make a change or the d-men they were replacing's ability to make a change (these last two, I don't know of any metric for measuring)?

There are so many external factors that play into the reasons behind why the advanced stats end up the way they are that using them to prove a point by themselves is flawed. Ok, I'll stop asking questions and answer a few of them.

Here are some potential explanations for the disparity you posted above:

- Moore received sheltered minutes. Over the course of last season, Moore faced a far lower quality of competition than Klein did.

- Klein also was playing with players of a far lower quality than Moore did throughout the course of the year.

- Moore also had hugely sheltered zone starts. He had the highest percentage of offensive zone starts of any Rangers defenseman last year at over 63 % offensive zone starts. Klein's was right about at 50 %. So is it reasonable to conclude that the times when Moore was away from Klein, he was more often being started in the offensive zone than Klein was when he was away from Moore? How else do we get that disparity?

- Klein saw far fewer minutes paired with high quality players as his defensive partners when he was away from Moore than Moore did when he was away from Klein. Moore played nearly 6 hours of minutes with MDZ, 2 hours of minutes with Stralman, and nearly 2 hours of minutes with Girardi. Klein's most common defensive partner after John Moore - while he was on the Rangers - was Raphael Diaz.

- And that doesn't even consider the guys he was paired with during his time in Nashville. Could it be that because Klein played for a team that was struggling last year for over half his regular season playing time, that this impacted his corsi? Did it help John Moore's corsi that the times he was "away" from Kevin Klein while they were playing on separate teams, that he was on the arguably better team?


(You can find my source material for these items here: http://stats.hockeyanalysis.com/ratings.php?db=201314&sit=5v5&type=corsi&teamid=20&pos=defense&minutes=100&disp=1&sort=HARTp&sortdir=DESC & http://www.behindthenet.ca/nhl_statistics.php?ds=63&s=13&f1=2013_s&f2=5v5&f4=D&f5=NYR&c=0+1+3+5+4+6+7+8+32+33+34+45+46+11+12+13+14+15+16+29+30+31+35+36+37+38+39+40+47+48+49+50+51+52+53+54+55+56+63+67+57+58+59+60+61+62+64+65+66# & http://www.progressivehockey.com/p/blog-page.html)

 

[Richter Scale]

Very coincidental unless you can show us otherwise. Winning as a team is not proof Klein is the cause of winning just because it started happening when he showed up.

Just like pointing to corsi, without any context or explanation of why it is significant, is not proof of Klein (or any other player) being "bad."

 

[Zil]

They also believed that prospects like Pavel Brendl, Jamie Lundmark and Hugh Jessiman were worthy of first round picks.

If you're going to bring up Lundmark, Jessiman, and Pavel ****ing Brendl there's no point in discussing this with you. If you want to be taken seriously then don't confuse picks made under Smith and Maloney with those made under Clark and Gorton.

That's complete crap.

Klein can control the quality of shots that Lundqvist faces. That's the whole point. A good defenseman prevents high quality scoring chances.

No, what more importantly he can control is the quality of shots Lundqvist faces. Everyone noticed the D settle down when he was brought it. And that's because DZ sucked bad and he was replaced by a steady d-man that plays a decent game.

If you're going to try and call me out at least do some basic research. There have been tons of studies on this. Klein or any skater for that matter has very, very little if any control over the quality of shots his opponents generate. Goalie save percentage does not correlate with skaters from year to year. It doesn't. Whatever positive effects he does have are so tiny as to be nearly negligible. Here's a little reading:

Here’s an example: Does a good defenseman help his goaltender record a better save percentage? Of course! A great defensive defenseman pounces on rebounds and clears the front of the net and doesn’t let his opponents get great scoring chances. He makes his goaltender better. Anyone who watches hockey knows this.

Except … well, this. The numbers don’t play nice with common sense.

Once you go down the advanced stats rabbit hole, this stuff happens all the time. Sometimes it’s big things and sometimes it’s little things, but it’s not rare and it takes some getting used to.

http://grantland.com/the-triangle/the-fakers-guide-to-advanced-stats-in-the-nhl/

http://www.broadstreethockey.com/2013/7/4/4487304/save-percentage-variability-regression-defense

http://hockey-graphs.com/2014/07/07...-no-sustainable-control-over-save-percentage/

 

[Zil]

It doesn't matter which team Klein was on. His teammates' corsi numbers almost always go up when not paired with him: http://stats.hockeyanalysis.com/showplayer.php?pid=698&withagainst=true&season=2013-14&sit=5v5

 

[Thirty One]

Just like pointing to corsi, without any context or explanation of why it is significant, is not proof of Klein (or any other player) being "bad."

Of course it isn't.

But given that:
1. When Klein is on the ice more shots are directed at his goalie
2. Defenseman show no sustainable ability to affect shooting percentage

What makes him good?

 

[Zil]

Of course it isn't.

But given that:
1. When Klein is on the ice more shots are directed at his goalie
2. Defenseman show no sustainable ability to affect shooting percentage

What makes him good?

Shhh. When I tell them that they curse at me.

 

[Thirty One]

Here's Klein's 5-on-5 On-Ice SV%s by season:

2013-14 0.9249
2012-13 0.9174
2011-12 0.9232
2010-11 0.933
2009-10 0.906
2008-09 0.9377
2007-08 0.8906

Notice how it's (a) wildly random and (b) not very good?

You would think a shot-quality suppressing defenseman would post consistently good numbers, no?

 

[Richter Scale]

Of course it isn't.

But given that:
1. When Klein is on the ice more shots are directed at his goalie
2. Defenseman show no sustainable ability to affect shooting percentage

What makes him good?

First off - I never argued he was "good." Just that 1. He isn't bad, and 2. that just pointing to his corsi does not prove that he is bad.

Not seeing how what you said is any different than pointing to corsi in the way Zil did. What I am saying is, fine, 1 is true, and perhaps 2 is true as well (I would debate this, but this wasn't a major point I was making). But what is the reason they are true? Is it because he is a bad player? Or are there other factors at play that produce those results, and would do so whether he was bad, good, or maybe just "not bad"?

Putting it another way (and taking it to the extreme to demonstrate the point I'm making):

Take any given defenseman and put them out there in two different scenarios. 1. In which they are facing toughest possible opponents, are playing with the worst players on their team, and are getting only defensive zone starts. 2. In which they are facing the easiest possible opponents, are playing with the best players on their team, and are getting only offensive zone starts.

Which one of those scenarios will likely produce the worse corsi? And does that mean that the very same defenseman was somehow a worse player in that first scenario than in the second scenario?


You can have a player whose team sees more shots against while they are on the ice who is no worse than another player whose team sees less shots against while they are playing - and the sole basis for that could simply be the situations they are put out on the ice for, and not the quality of the player. It can't be looked at in a vacuum.

I would argue that defensive defensemen are relied on more in tougher defensive situations which would inherently impact their corsi. Doesn't mean they're bad - it just means the coach has more confidence in them to handle those situations over other players.

--

Now, I'm not trying to prove Klein is god's gift to defense (quite the contrary, but I would argue he is a very serviceable 3rd pairing dman). But when you only look at these advanced stats to compare him to other defenseman, you have to make sure the comparison is a comparable one. The Moore-Klein comparison is not comparable, for the reasons I laid out above.

Also not saying there aren't defensemen out there who are better than Klein, just that a sole and blinding reliance on corsi to make that argument (without looking at all those other factors), is not making a convincing argument. It isn't looking at apples to apples. Its looking at apples to oranges.

 

[Thirty One]

First off - I never argued he was "good." Just that 1. He isn't bad, and 2. that just pointing to his corsi does not prove that he is bad.

Not seeing how what you said is any different than pointing to corsi in the way Zil did. What I am saying is, fine, 1 is true, and perhaps 2 is true as well (I would debate this, but this wasn't a major point I was making). But what is the reason they are true? Is it because he is a bad player? Or are there other factors at play that produce those results, and would do so whether he was bad, good, or maybe just "not bad"?

Putting it another way (and taking it to the extreme to demonstrate the point I'm making):

Take any given defenseman and put them out there in two different scenarios. 1. In which they are facing toughest possible opponents, are playing with the worst players on their team, and are getting only defensive zone starts. 2. In which they are facing the easiest possible opponents, are playing with the best players on their team, and are getting only offensive zone starts.

Which one of those scenarios will likely produce the worse corsi? And does that mean that the very same defenseman was somehow a worse player in that first scenario than in the second scenario?


You can have a player whose team sees more shots against while they are on the ice who is no worse than another player whose team sees less shots against while they are playing - and the sole basis for that could simply be the situations they are put out on the ice for, and not the quality of the player. It can't be looked at in a vacuum.

I would argue that defensive defensemen are relied on more in tougher defensive situations which would inherently impact their corsi. Doesn't mean they're bad - it just means the coach has more confidence in them to handle those situations over other players.

--

Now, I'm not trying to prove Klein is god's gift to defense (quite the contrary, but I would argue he is a very serviceable 3rd pairing dman). But when you only look at these advanced stats to compare him to other defenseman, you have to make sure the comparison is a comparable one. The Moore-Klein comparison is not comparable, for the reasons I laid out above.

Also not saying there aren't defensemen out there who are better than Klein, just that a sole and blinding reliance on corsi to make that argument (without looking at all those other factors), is not making a convincing argument. It isn't looking at apples to apples. Its looking at apples to oranges.

Obviously player usage will have an impact.

Would you believe that Moore-Klein started less shifts in the defensive zone compared to Moore-Klein?

 

[Mikos87]

You can point to the nerd stats with Klein, but the eye test says that he gets pinned because he isn't the most mobile skater out there.

Unless teams play that LA/STL/2012NYR style of hockey, the more immobile defenseman on possession based teams have bad corsi numbers because they can't get the puck out as quickly as some of their counterparts.

Klein is a mistake free defenseman that stabilized JMoore, and the team defense GAA after that trade, alongside the "Winning" Percentage show that.

He doesn't "suck", he's just a little slow eh.

 

[TheRedressor]

I thought this tread was about Staal re-signing or not.......

 

[Richter Scale]

Here's Klein's 5-on-5 On-Ice SV%s by season:

2013-14 0.9249
2012-13 0.9174
2011-12 0.9232
2010-11 0.933
2009-10 0.906
2008-09 0.9377
2007-08 0.8906

Notice how it's (a) wildly random and (b) not very good?

You would think a shot-quality suppressing defenseman would post consistently good numbers, no?

I see 2 seasons where it was "not very good," one where it was fine but not stellar, and 4 where it was pretty damn good. I also notice that the two worst seasons came at the beginning of his career. Could it be that he wasn't as good of a defenseman when he was first starting in the league as he was after a few years of development?

Another thing I notice is that the save % while he was on the ice is - aside from those 2 bad early seasons - generally higher than the save % for the season of the goalie(s) in front of whom he played (admittedly by small margins). That might seem to indicate the opposite of what you're trying to convey... that perhaps there was something about Klein's presence on the ice that did impact the goalies' save percentages. Would that be the shot quality they faced, or would you argue some other reason for that?

Again though - this is not what I was arguing earlier. But you brought it up, and its another thing with which I think I would disagree, so I'm addressing it.

 

[Richter Scale]

Would you believe that Moore-Klein started less shifts in the defensive zone compared to Moore-Klein?

Is this a trap?

 

[Thirty One]

I see 2 seasons where it was "not very good," one where it was fine but not stellar, and 4 where it was pretty damn good. I also notice that the two worst seasons came at the beginning of his career. Could it be that he wasn't as good of a defenseman when he was first starting in the league as he was after a few years of development?

Another thing I notice is that the save % while he was on the ice is - aside from those 2 bad early seasons - generally higher than the save % for the season of the goalie(s) in front of whom he played (admittedly by small margins). That might seem to indicate the opposite of what you're trying to convey... that perhaps there was something about Klein's presence on the ice that did impact the goalies' save percentages. Would that be the shot quality they faced, or would you argue some other reason for that?

Again though - this is not what I was arguing earlier. But you brought it up, and its another thing with which I think I would disagree, so I'm addressing it.

No, I posted 5-on-5 SV%. Comparing it to his starting goalie's total SV% that year doesn't really work.

If you want to see Klein vs. his team on the whole:

Again, I don't think this is meaningful. It's his shot differentials that are troubling.

Is this a trap?

Opps.

 

[Richter Scale]

No, I posted 5-on-5 SV%. Comparing it to his starting goalie's total SV% that year doesn't really work.

If you want to see Klein vs. his team on the whole:

You're right. I noticed that when I was looking at your trap post, and was going to edit the post to indicate that - but you beat me to it. My mistake; missed that it was 5-on-5 only.

 

[aemoreira1981]

No he's goin to Carolina

After Jim Rutherford was fired from his job in Carolina and ended up in Pittsburgh, I doubt that that happens now.

 

[Ail]

Just like pointing to corsi, without any context or explanation of why it is significant, is not proof of Klein (or any other player) being "bad."

However doing so in context, is. Which has been done here many times.

 

[Doctyl]

Yea I see what all these numbers say, but does anyone actually believe we lost the Klein trade?

 

[Thirty One]

Yea I see what all these numbers say, but does anyone actually believe we lost the Klein trade?

Yes.

 

[Richter Scale]

However doing so in context, is. Which has been done here many times.

It has?

Where has someone pointed to his corsi, accounted for possible external factors which impact that corsi, and compared him to other players who: 1. play on 3rd pairings, 2. play against comparable competition, 3. play with comparable quality of teammates, and 4. receive comparable zone starts?

I think you don't know what context means.

The closest anyone has come to that is Zil's last post on the topic which read:

It doesn't matter which team Klein was on. His teammates' corsi numbers almost always go up when not paired with him

But even that has problems. What if Nashville used him similarly to the way NYR has, and in a way similarly different from his teammates there? (I haven't checked this - so I don't know if that's the case) But if it is, then you're still not comparing apples to apples.

EDIT: I would also point out that, at least for Rangers players in Zil's list above, 12 of them had higher corsis away from Klein, and 11 of them had lower corsis away from Klein. The 12 with higher corsis away from Klein played about 1100 minutes with him; the 11 with lower corsis away from Klein played about 550 minutes with him. But still - not exactly as clear of a picture as is being painted.

 

[Cake or Death]

I come in to take a quick peak if there were any Staal updates. I saw like nothing about Staal and almost all Klein

 

[Thirty One]

I came here expecting to read about how Klein solidified the third pair, and everyone's talking about how we're not talking about Staal smh.

 

[lucky13]

I came here expecting to read about how Klein solidified the third pair, and everyone's talking about how we're not talking about Staal smh.

I came here to find out if klein stabilized staal ...instead i found out that Finckel was Einhorn...

 

[Ail]

It has?

Where has someone pointed to his corsi, accounted for possible external factors which impact that corsi, and compared him to other players who: 1. play on 3rd pairings, 2. play against comparable competition, 3. play with comparable quality of teammates, and 4. receive comparable zone starts?

I think you don't know what context means.

The closest anyone has come to that is Zil's last post on the topic which read:



But even that has problems. What if Nashville used him similarly to the way NYR has, and in a way similarly different from his teammates there? (I haven't checked this - so I don't know if that's the case) But if it is, then you're still not comparing apples to apples.

EDIT: I would also point out that, at least for Rangers players in Zil's list above, 12 of them had higher corsis away from Klein, and 11 of them had lower corsis away from Klein. The 12 with higher corsis away from Klein played about 1100 minutes with him; the 11 with lower corsis away from Klein played about 550 minutes with him. But still - not exactly as clear of a picture as is being painted.

The four comparisons you listed are your own arbitrary conditions for context and if you require a comparison of Klein with every 6 or 7th defensemen in the league, then maybe you should do it and let us know what you find, not that it matters, but good luck anyway.

 

[Ail]

I came here to find out if klein stabilized staal ...instead i found out that Finckel was Einhorn...

Dude... spoilers.