Confirmed with Link: [NYR/NSH] Michael Del Zotto for Kevin Klein

[McRanger]

I think you may find this interesting.

Klein's played over 100 5-on-5 minutes with 12 defensemen from 2009-2013. Two of them had a better CF% with him than without him. (I took out ice time shared with Weber/Suter to make sure it wasn't because of that.)

In total, they had a weighted average change of -2.6%. Del Zotto's is -0.6%.

How do you factor in quality of competition and offensive zone starts? How does the quality of the forwards factor in?

Seems difficult to compare two players from two different teams playing two different roles.

 

[Cassano]

I think you may find this interesting.

Klein's played over 100 5-on-5 minutes with 12 defensemen from 2009-2013. Two of them had a better CF% with him than without him. (I took out ice time shared with Weber/Suter to make sure it wasn't because of that.)

In total, they had a weighted average change of -2.6%. Del Zotto's is -0.6%.

Yeah thanks for bringing this up.

I was very well aware that Klein's presence on the ice stagnated the play of his defense partners. Which is why I'm utterly confused how anyone can remotely suggest he's a great defensive player.

While MDZ did drag the play of his defense partners, at least he had offensive upside, was a proven point producer and was 6 years younger than Klein. As per your chart, he also didn't drag the play of his linemates to the extent Klein has.

This is pretty much the anti-moneyball move. Oh well. While Klein may be the inferior player, he at least he'll be

• ©Grittyâ„¢
• ©Character Guyâ„¢
• ©Tough As Nailsâ„¢

 

[Thirty One]

Could that be because of the forwards they played with?

Still trying to acclimate to exactly how to view advanced stats (WAR equivalent, plz. Get it done, Slats) and isn't Corsi dependent on anybody on the team stopping shots and taking shots?

As an MDZ advocate who thinks his defensive woes were overblown, this is interesting.

Yeah, but I think that over 100+ minutes Bouillon-Klein would play with a similar group of forwards as Bouillon-Josi or Bouillon-Gill, no?

 

[Vickers8]

He was pretty good I thought. Had a nervous first shift, where he almost coughed up the puck, then he settled down to play his game, a couple of blocks, a couple of hits, a shot on goal. Nothing flashy.

Steady and reliable

 

[SnowblindNYR]

No, because of course, watching him play three games is always better than three years of data.

Um, I never said that. Predators fans seem to think he's a good defensive player and are upset to see him go, I'm sure they watch him play more than that.

 

[Khelvan]

As I've never taken a look at these advanced stats before, what do CF, CA, and CF% mean?

 

[Thirty One]

As I've never taken a look at these advanced stats before, what do CF, CA, and CF% mean?

CF = Corsi For = All shot attempts (goals, saves, blocks, misses) directed towards the opponent's net while player X is on the ice.
CA = Corsi Against = All shot attempts (goals, saves, blocks, misses) directed by the opponent towards player X' net while player X is on the ice.
CF% = Corsi-For% = CF / (CF + CA)

And here's why Corsi is better than +/- AINEC: http://rangersunlimited.com/2014/01/03/girardi-lundqvist-chickens-and-eggs/
#shamlessselfpromotion

 

[SnowblindNYR]

Hey you guys ever read the book "When Genius Failed"? If not it's about people smarter I'm sure than most here that were mathematicians and geniuses, basically they revolutionized the field of computational finance. Anyway, Hedge Fund Long Term Capital Management hired these guys and bet a lot and leveraged even more on the ruble. Of course, unforeseen circumstances happened and the fund lost like 98% of its value (maybe a little less) and had to be bailed out by the government and huge investment banks.

These are far less complicated, but the point remains, you can't be married to your statistics.

 

[aufheben]

Advanced stats are useful within contexts, which -31- and Blue Blooded usually provide.

 

[Thirty One]

Hey you guys ever read the book "When Genius Failed"? If not it's about people smarter I'm sure than most here that were mathematicians and geniuses, basically they revolutionized the field of computational finance. Anyway, Hedge Fund Long Term Capital Management hired these guys and bet a lot and leveraged even more on the ruble. Of course, unforeseen circumstances happened and the fund lost like 98% of its value (maybe a little less) and had to be bailed out by the government and huge investment banks.

These are far less complicated, but the point remains, you can't be married to your statistics.

Have you every read Where Eyes Failed?

Again, no one is suggesting that the conversation ends with metrics. But I haven't seen a single attempt to explain why the stats presented aren't a useful tool.

 

[Tawnos]

Have you every read Where Eyes Failed?

Again, no one is suggesting that the conversation ends with metrics. But I haven't seen a single attempt to explain why the stats presented aren't a useful tool.

They are a useful tool, but only to the point where they disagree with the unquantifiable aspects of defensive play, of which there are plenty. In other words, their main use is to inform what we see, rather than be a whole package to evaluate a player.

I have watched a LOT of Kevin Klein, particularly in person. Admittedly less over the last 2 years than the two years prior to that, when I lived in Nashville. From my perspective, what the advanced stats are saying about the player is very misleading.

 

[Thirty One]

They are a useful tool, but only to the point where they disagree with the unquantifiable aspects of defensive play, of which there are plenty. In other words, their main use is to inform what we see, rather than be a whole package to evaluate a player.

I have watched a LOT of Kevin Klein, particularly in person. Admittedly less over the last 2 years than the two years prior to that, when I lived in Nashville. From my perspective, what the advanced stats are saying about the player is very misleading.

I'm not sure why defensive play is unquantifiable. I think defensive play is an area where the eye test fails miserably.

I think a good defensive player is one who allows less shots on goal under the same conditions. Tell me why I'm wrong.

 

[SnowblindNYR]

Have you every read Where Eyes Failed?

Again, no one is suggesting that the conversation ends with metrics. But I haven't seen a single attempt to explain why the stats presented aren't a useful tool.

Trick question, Eyes never failed!

 

[SnowblindNYR]

At the end of the day can we let the guy play more than a game before we criticize him?

 

[Thirty One]

Trick question, Eyes never failed!

Yes, that explains why the best mutual funds are run by managers who don't understand math and simply have "good feelings about this one".

 

[Ail]

CF = Corsi For = All shot attempts (goals, saves, blocks, misses) directed towards the opponent's net while player X is on the ice.
CA = Corsi Against = All shot attempts (goals, saves, blocks, misses) directed by the opponent towards player X' net while player X is on the ice.
CF% = Corsi-For% = CF / (CF + CA)

And here's why Corsi is better than +/- AINEC: http://rangersunlimited.com/2014/01/03/girardi-lundqvist-chickens-and-eggs/
#shamlessselfpromotion

Great entry. Makes a ton of sense.

 

[Thirty One]

At the end of the day can we let the guy play more than a game before we criticize him?

He's played many games. Life doesn't begin when traded to the Rangers.

 

[Cassano]

Yes, that explains why the best mutual funds are run by managers who don't understand math and simply have "good feelings about this one".

Man, I was gonna post the exact same thing.

 

[Thirty One]

Great entry. Makes a ton of sense.

 

[NYR Viper]

I'm not sure why defensive play is unquantifiable. I think defensive play is an area where the eye test fails miserably.

I think a good defensive player is one who allows less shots on goal under the same conditions. Tell me why I'm wrong.

Because there are a lot of outside factors you don't take into account.

Malik played a TON with Jagr when he was here. Therefore I am sure his rating was very high. Malik was a decent 2nd pairing d-man who probably was rated as one of the better d-men in the league because of his supporting cast. What stat shows that?

And I should note, I am not against utilizing stats at all, I just don't like to have them thrown around and weighted more than watching a player play.

 

[McRanger]

I'm not sure why defensive play is unquantifiable. I think defensive play is an area where the eye test fails miserably.

I think a good defensive player is one who allows less shots on goal under the same conditions. Tell me why I'm wrong.

How does quality of competition weigh in? What about role? Are defenders that play a more defensive role with bottom six forwards (and vice versa) judged the same way as defensemen that play with top six forwards? Does it factor quality of shots? Are all attempted shots equal or are they weighted?

 

[Cassano]

The biggest flaw with this 'I'd rather be watching a player play' is that we are all biased.

I can say Stu Bickel is a great defenseman because my eyes tell me so. Don't you dare back anything with facts to disapprove that!

 

[NYR Viper]

The biggest flaw with this 'I'd rather be watching a player play' is that we are all biased.

I can say Stu Bickel is a great defenseman because my eyes tell me so. Don't you dare back anything with facts to disapprove that!

And that's why you should use both.

 

[Thirty One]

How does quality of competition weigh in? What about role? Are defenders that play a more defensive role with bottom six forwards (and vice versa) judged the same way as defensemen that play with top six forwards? Does it factor quality of shots? Are all attempted shots equal or are they weighted?

Because there are a lot of outside factors you don't take into account.

Malik played a TON with Jagr when he was here. Therefore I am sure his rating was very high. Malik was a decent 2nd pairing d-man who probably was rated as one of the better d-men in the league because of his supporting cast. What stat shows that?

And I should note, I am not against utilizing stats at all, I just don't like to have them thrown around and weighted more than watching a player play.

By same conditions, I mean same Quality of Teammates/Competition and Zone Starts those are all factors that can be measured and accounted for.

Defenders have been shown to have little impact on shot quality.

 

[NYR Viper]

By same conditions, I mean same Quality of Teammates/Competition and Zone Starts those are all factors that can be measured and accounted for.

Defenders have been shown to have little impact on shot quality.

Hey, maybe I'm wrong, I have been before. I just feel like in a game that is as fluid and has so many outside factors, it's hard to rely totally on statistics in a major team game. Baseball is another story.