Corsi and Advanced Stats

[BroadwayJay*]

Haha no problem at all. The more the merrier.

Regarding the fourth line, yeah I think that might have been due to Martin's absence/injury. It's always tricky interpreting single-game stats though.

I'd say it is almost worthless to try and interpret a single-game's stats as anything other than a single game.

They had a bad night, they're a fourth line. It happens. They might have a great night against Boston and even it out.

The bigger question, one which I think remains an open one, is whether this fourth line brings more value than our third line. I am getting the distinct impression that some folks are leaning yes with that question. I'm still leaning no, but every game provides additional data which eventually may lead to an answer to that question (or at least a consensus recommendation).

 

[charlie1]

I'd say it is almost worthless to try and interpret a single-game's stats as anything other than a single game.

They had a bad night, they're a fourth line. It happens. They might have a great night against Boston and even it out.

The bigger question, one which I think remains an open one, is whether this fourth line brings more value than our third line. I am getting the distinct impression that some folks are leaning yes with that question. I'm still leaning no, but every game provides additional data which eventually may lead to an answer to that question (or at least a consensus recommendation).

Really? I think that's the wrong question. The main question for me is whether we can quantify their contribution beyond standard possession metrics.

 

[BroadwayJay*]

Really? I think that's the wrong question. The main question for me is whether we can quantify their contribution beyond standard possession metrics.

I suppose I'm looking at it from the perspective of decision-making, which I think covers your inquiry as well.

If the question is whether they're outperforming the third line, and the answer is yes, then the decision is to give that line more ice time to bring it in line with third-line ice time.

You don't think your question is covered by that same inquiry? As in, your question is something that is a question we ask before reaching the ultimate question for a decisionmaker, which is my question.

 

[charlie1]

I suppose I'm looking at it from the perspective of decision-making, which I think covers your inquiry as well.

If the question is whether they're outperforming the third line, and the answer is yes, then the decision is to give that line more ice time to bring it in line with third-line ice time.

You don't think your question is covered by that same inquiry? As in, your question is something that is a question we ask before reaching the ultimate question for a decisionmaker, which is my question.

I just want to see if we quantify their contribution in a novel way. What people do with that information is up to them. And I don't know why we would take ice time from the 3rd line, and not the 2nd or 1st. Heck, I don't even know which line IS the third line. That's just a very specific question you're asking.

 

[BroadwayJay*]

I just want to see if we quantify their contribution in a novel way. What people do with that information is up to them. And I don't know why we would take ice time from the 3rd line, and not the 2nd or 1st. Heck, I don't even know which line IS the third line. That's just a very specific question you're asking.

I don't know which line is the third line, I just know that it is the line that gets the third-most ice time, whereas the fourth line gets the fourth-most icetime. If they deserve more ice time, and therefore should be flipflopped with the third-line; that would be the event I'm describing.

Would it have made more sense to say the inquiry is whether the fourth line should be used more?

 

[BroadwayJay*]

Or rather, used more at the expense of other lines? And then a second inquiry of how much more?

I tried to encapsulate both ideas into one and that may have been a mistake. I'm not always as good at conveying complex ideas in bite-size statements as I'd like to be. I'm trying to get better though.

 

[IceAce]

I'd be more interested in his relative CF and relative FF. They're both just slightly above average for the team. That's not a red flag so long as his team is doing well. If the team is doing awful and the relative corsi is not good, that's a concern.

If we were to acquire him, I'd do a full analysis. Until that happens, I don't think it is worth the time.

any clue which one here is the relative number you're talking about?

http://stats.hockeyanalysis.com/rat...ters&minutes=300&disp=1&sort=PCT&sortdir=DESC

 

[BroadwayJay*]

CF%RelTM

Corsi For Percentage Relative to Team

 

[charlie1]

Advanced stat of the day

A quick comparison of Cam Talbot vs. Henrik Lundqvist:

Talbot and Lundqvist have almost identical even-strength Sv%: 0.933 (Talbot) vs. 0.931 (Lundqvist).

BUT the locations from which they excel at stopping shots are very different. Lundqvist is excellent at stopping shots from the high-percentage "home plate" area. While Talbot is much weaker there, but is stronger at stopping perimeter shots.

So teams that can penetrate into the high-percentage area will have more success against the Rangers while Lundqvist is out, while weaker teams that are constrained to outside shots will be less successful.

Lundqvist's Hextally plot (blue is better than league-average Sv%, green is league-average):



Talbot's Hextally plot:

 

[camper*]

Have to give you a bit of credit todate...your understanding of what is on paper has been rather good todate(not thats its overwelming)..unlike some who's hair on fire try babbling graph facts.....theres always more to these things for sure...yet todate...stress todate you weigh/balance statements.....our p.p. may have diff. under one of the graphs shown.(being perimeter possesion guys)
Nothing wrong with throwing all these graphs out there some stuff is interesting and beneficial....its keeping the look what I read guys from catching fire

 

[Bones45]

Camper -- just a word of advice here. Just trying to help.

I don't know what you are saying for two reasons. One -- and most importantly, I know nothing about advanced stats or Corsi/Fenwick.

But secondarily -- the way you write --- is exhausting to read. You could be talking about any topic, and I would skip over it.

I'm not saying to look to me for how to express yourself -- hell no. But it looks like you have some hockey knowledge and much to say -- it would be shame to miss it.

Just my .02.

 

[Islanders1932]

A quick comparison of Cam Talbot vs. Henrik Lundqvist:

Talbot and Lundqvist have almost identical even-strength Sv%: 0.933 (Talbot) vs. 0.931 (Lundqvist).

BUT the locations from which they excel at stopping shots are very different. Lundqvist is excellent at stopping shots from the high-percentage "home plate" area. While Talbot is much weaker there, but is stronger at stopping perimeter shots.

So teams that can penetrate into the high-percentage area will have more success against the Rangers while Lundqvist is out, while weaker teams that are constrained to outside shots will be less successful.

Lundqvist's Hextally plot (blue is better than league-average Sv%, green is league-average):



Talbot's Hextally plot:

Nevermind, I figured it out.

 

[charlie1]

Nevermind, I figured it out.

I missed your Q. Glad you figured it out. Let me know if you have any others...

 

[BroadwayJay*]

Is it possible that the "elite" nature of Henrik Lundqvist's goaltending is his ability to make saves in the "home plate" and that is what makes his save percentage so high?

Is that the conclusion we're drawing from these charts? Seems like it is to me.

 

[PK Cronin]

Is it possible that the "elite" nature of Henrik Lundqvist's goaltending is his ability to make saves in the "home plate" and that is what makes his save percentage so high?

Is that the conclusion we're drawing from these charts? Seems like it is to me.

For whatever reason, if he's good at making saves in those areas teams are likely playing into an advantage for him. Everyone wants to get to that real estate for great opportunities, but perhaps against Lundqvist it might be better to just fire away from the points.

 

[IceAce]

maybe people get mesmerized by his dreamy good looks in close and can't shoot straight?

 

[PK Cronin]

maybe people get mesmerized by his dreamy good looks in close and can't shoot straight?

I didn't even think of that...wonder what would happen if we put a 2 way mirror on his mask so players can't see him. I bet he'd be like Poulin in there.

 

[BroadwayJay*]

I can think of a few guys who aren't going to like this:

http://hockey-graphs.com/2015/02/09/the-usefulness-or-lack-thereof-of-hit-totals/

The hit statistic rivals the faceoff in praise by some more traditional hockey analysts.

Interestingly, hit differentials almost have an equal and opposite relationship with goal differentials as shot differentials have with goals.

It’s not that hits or hitting is bad; it’s quite the opposite. There are reasons why teams hit after all. It is just that the hit statistic is bad. It doesn’t really tell you anything except the team that is hit more. A team can out hit because they never had the puck, or they can out hit because they are an effective team in the forecheck. More often than not, a team’s hit differential is telling you more about the former than the later.

I've never thought "hits" was a very useful statistic so this comes as no surprise to me. I also recall Dellow's work on this from some time ago that was taken down when he joined the Oilers.

Does this mean that hitting is useless? No. Only that the hit statistic is not correlative with success. There is always a place for physicality, I just don't think it is pronounced as people think and I don't think the collection of mindless hits has much value either.

An interesting little piece here.

 

[IceAce]

 

[BroadwayJay*]

I think that's going to give me a seizure.

 

[The Underboss]

http://hfboards.mandatory.com/showthread.php?p=96962243#post96962243


Someone already started Part II. Link above. Take the party there.