GDT: wardo vs new york rangers

[Navin R Slavin]

In fact, let's see if there's a Team Hex metric. Oh, look, there is!

http://war-on-ice.com/hexteams.html

Now select Carolina.

Of course, we already know that Carolina should be well above average in Shots Against, since the team leads the NHL in that category. And the graph reflects that; it's completely blue. But where are the shots coming from?

As it turns out, the Canes are quite a bit above average in preventing shots from the slot! Not just light blue, but moderate blue!

So in looking at these two graphs, we see the following data -- and this is data, not interpretation:

* The Canes are better than league average at preventing shots.
* The Canes are way better than league average at preventing shots from high-risk areas.
* Cam, as a goaltender, is NHL average in stopping shots from high-risk areas.
* Cam, as a goaltender, is lower than NHL average in stopping shots from anywhere else.

Now, how does one interpret that data? I don't know how else to interpret it, except that Cam is mostly to blame for his own poor save percentage. He sees fewer high quality shots than most goalies in the league, and he surrenders more goals on poor quality shots than most goalies in the league. That combination is pretty damning.

 

[Navin R Slavin]

i agree. However, right now, i think one of the conclusions i saw earlier was that the hurricanes are a great hockey team that's just unlucky.

Assuming you disagree with that assessment, we'd agree that something is afoot statistically in carolina.

---
hank,
"this year, cam ward is objectively a below-average nhl goaltender."

it isn't objective, because what one views as the criteria for a below-average goaltender is subjective. Objectively, cam has a below average sv%. Subjectively, that makes him a below average goaltender.

Just like i don't think you'd say the canes are objectively a very good team because of their possession stats. I wouldn't anyways.

¯\_(ツ)_/¯

 

[Navin R Slavin]

And high risk/low risk is simply decided by the shot location?

Can you come up with a better metric that's quantifiable in any way with the data we have?

 

[Anton Dubinchuk]

devils advocate (alternative, even if unlikely to be accurate way to interpret data)

canes defense is bad in such a way that turns "less high risk" areas into higher risk ones. example could be failure to clear the front of the net, making point shots more dangerous, or the like.

note: i dont actually think this is the case (maybe to a degree but not enough to say ward is an average goaltender), but just offering some fuel for the fire.

 

[Bub]

Not to change the subject, but to change the subject...

... Even other coaches/announcers have stated that (I usually watch the other teams feeds) ...

Something I wanna start doing too. Other than Jack Edwards (which doesn't matter since I can't get Boston games due to local blackouts), who are the arch-homers to avoid?

 

[Anton Dubinchuk]

Now, how does one interpret that data? I don't know how else to interpret it, except that Cam is mostly to blame for his own poor save percentage. He sees fewer high quality shots than most goalies in the league, and he surrenders more goals on poor quality shots than most goalies in the league. That combination is pretty damning.

link wont open on my computer, does the link suggest that the shots that the canes DO give up are more from lower risk areas? or is it the same ratio of high risk to low risk shots, and the canes just give up less all around?

because if the latter, the canes ability to prevent shots shouldn't affect the sv%, just the gaa

 

[Stickpucker]

I would rather Start Greiss or Neuverth than Ward.

 

[Navin R Slavin]

devils advocate (alternative, even if unlikely accurate way to interpret data)

OK.

canes defense is bad in such a way that turns "less high risk" areas into higher risk ones. example could be failure to clear the front of the net, making point shots more dangerous, or the like.

In this particular case, there could be two reasons the shot is "more dangerous": it's tipped, or it's screened. Tips end up being shots from the slot anyway. So the only way this case holds up is if the Canes see a crazy number of screened shots by comparison to other teams. I think the eye test is pretty reliable here, and I haven't seen that an abnormal number of goals caused by screens. But: ¯\_(ツ)_/¯

 

[Navin R Slavin]

link wont open on my computer, does the link suggest that the shots that the canes DO give up are more from lower risk areas? or is it the same ratio of high risk to low risk shots, and the canes just give up less all around?

because if the latter, the canes ability to prevent shots shouldn't affect the sv%, just the gaa

The Canes give up fewer shots overall, and markedly fewer shots in high risk areas, than other teams do.

 

[Blueline Bomber]

Can you come up with a better metric that's quantifiable in any way with the data we have?

If that is the metric they're using to determine risk, then I'd say it's a flawed one. Shot location is only a small part of what determines if a shot is "dangerous".

A Gleason shot from the point doesn't have the same risk as a Faulk shot from the same location, for example. Or two shots come from the same location in the low slot (a high risk area), but due to the way the respective shooter is positioned, one shot is able to be lifted, while the other hits pad. Same location, but one is decidedly more dangerous.

Again, I'm not saying this is the case for Ward's play this year, but you can't simply say "Well, all shots that come from this location are low risk, so if a goaltender isn't stopping them, he's below average."

 

[Anton Dubinchuk]

The Canes give up fewer shots overall, and markedly fewer shots in high risk areas, than other teams do.

the wording of that answer still leaves me unsure...

edit: never mind, i finally got it to load (that is a data-heavy website my goodness). looks like the canes do indeed give up a lower ratio of high-risk shots to regular shots than the rest of the league does, as well

 

[Navin R Slavin]

If that is the metric they're using to determine risk, then I'd say it's a flawed one. Shot location is only a small part of what determines if a shot is "dangerous".

A Gleason shot from the point doesn't have the same risk as a Faulk shot from the same location, for example. Or two shots come from the same location in the low slot (a high risk area), but due to the way the respective shooter is positioned, one shot is able to be lifted, while the other hits pad. Same location, but one is decidedly more dangerous.

OK, that's possible. But we're definitely getting into Occam's Razor territory here.

One of these hypotheses -- "Cam isn't a very good goalie" -- has a lot of evidence backing it up, and relies on a few simple and testable assumptions.

Whereas the competing hypothesis -- "Cam is a decent goalie whose numbers are suffering because of the team in front of him" -- doesn't have a lot of evidence backing it up, and relies on some complex and untestable assumptions.

Again, I'm not saying this is the case for Ward's play this year, but you can't simply say "Well, all shots that come from this location are low risk, so if a goaltender isn't stopping them, he's below average."

I think you can say exactly that, until someone produces a more compelling hypothesis.

 

[Bub]

Hmm... (scroll down to #5)

http://sports.yahoo.com/blogs/nhl-puck-daddy/understanding-goalies--cam-ward-and-protecting-players--wednesday-countdown-205054270.html

 

[Boom Boom Apathy]

Something I wanna start doing too. Other than Jack Edwards (which doesn't matter since I can't get Boston games due to local blackouts), who are the arch-homers to avoid?

I don't mind the homers as I just kind of ignore the over the top stuff.

I don't know if they are always this way, but the Flyers announcers have been the most homer of the ones I've watched recently. Not terrible, but enough so I noticed it. In the past, I've noticed the Pens announcers as well.

None of the other ones I've seen this year have stood out to me as being overly homer, at least for the games I've watched.

 

[Boom Boom Apathy]

Hmm... (scroll down to #5)

http://sports.yahoo.com/blogs/nhl-puck-daddy/understanding-goalies--cam-ward-and-protecting-players--wednesday-countdown-205054270.html

Ouch.

Ward is clearly what he is: A goaltender with no ability to stop the puck at an NHL level any more.

Carolina's coaches and front office have to understand fundamentally that Ward has long since been passed by as a viable option. Are we really saying a Conn Smythe a decade ago is the reason he still gets trotted out.

 

[VAcaniac]

I don't mind the homers as I just kind of ignore the over the top stuff.

I don't know if they are always this way, but the Flyers announcers have been the most homer of the ones I've watched recently. Not terrible, but enough so I noticed it. In the past, I've noticed the Pens announcers as well.

None of the other ones I've seen this year have stood out to me as being overly homer, at least for the games I've watched.

They definitely are always that way.

Also Florida. I don't know if I have ever watched a Panthers game in the last 12 years where the announcers didn't complain about officiating (not just when the Canes dive either lol). I guess when you consistently suck as much as they have for decades you have to find a scapegoat though...

 

[Boom Boom Apathy]

They definitely are always that way.

Also Florida. I don't know if I have ever watched a Panthers game in the last 12 years where the announcers didn't complain about officiating (not just when the Canes dive either lol). I guess when you consistently suck as much as they have for decades you have to find a scapegoat though...

I think this might qualify as throwing stones while living in glass houses.

Yeah, I haven't seen the Florida crew in a long time so forgot about them.

 

[Blueline Bomber]

OK, that's possible. But we're definitely getting into Occam's Razor territory here.

One of these hypotheses -- "Cam isn't a very good goalie" -- has a lot of evidence backing it up, and relies on a few simple and testable assumptions.

Whereas the competing hypothesis -- "Cam is a decent goalie whose numbers are suffering because of the team in front of him" -- doesn't have a lot of evidence backing it up, and relies on some complex and untestable assumptions.

Well, seeing as how very few people are saying he's a decent goaltender whose numbers are suffering because of the team in front of him, there doesn't seem to be a reason to worry about that aspect.

My issue was with your (or rather, the data your using's) assumption that shot location is the determining factor in risk. Every shot is unique, because every shot presents itself in a unique situation. You could take video of 1000 in-game shots from the same location on the ice, and each one of them will have something different from every other one that will determine whether the shot should be more difficult or easier to save.

Who the shooter is, who's defending him, the position of...just about every player on the ice, whether the shot comes off cleanly, etc, etc. Dozens of factors come into play with every shot.

I think you can say exactly that, until someone produces a more compelling hypothesis.

Well, you CAN say it, but you'd be wrong. Because again, making blanket statements like "All shots from X location are low risk" is just asking for millions of examples on why that's false.

 

[Finlandia WOAT]

Do you have any proof that the hurricanes are giving up shots to better quality players moreso than other teams? That seems unlikely given they give up less shots than most of the league.

 

[Boom Boom Apathy]

My issue was with your (or rather, the data your using's) assumption that shot location is the determining factor in risk. Every shot is unique, because every shot presents itself in a unique situation. You could take video of 1000 in-game shots from the same location on the ice, and each one of them will have something different from every other one that will determine whether the shot should be more difficult or easier to save.

I think you are now making it overly complicated and suffering from analysis paralysis. It's no different than pitchers and hitters in baseball, QB's in football, kickers in football, basketball players shooting, etc... no two events are ever EXACTLY the same, but that doesn't mean the data is invalid or you can't draw conclusions from it and that's what using statistics are all about. Over a large enough samples size, it's reasonable (but not always 100%) to assume that those things normalize / even out. Ward has seen almost 170 Low Danger ES shots this year so far. At some point, you can't say he's the only one seeing a higher number of rockets from Chara.

 

[VAcaniac]

I think this might qualify as throwing stones while living in glass houses.

Yeah, I haven't seen the Florida crew in a long time so forgot about them.

We get lucky and make deep runs every now and then.

 

[Blueline Bomber]

Do you have any proof that the hurricanes are giving up shots to better quality players than other teams? That seems unlikely given they give up less shots than most of the league.

Never said they were.

I think you are now making it overly complicated and suffering from analysis paralysis. It's no different than pitchers and hitters in baseball, QB's in football, kickers in football, basketball players shooting, etc... no two events are ever EXACTLY the same, but that doesn't mean the data is invalid or you can't draw conclusions from it and that's what using statistics are all about. Over a large enough samples size, it's reasonable (but not always 100%) to assume that those things normalize / even out. Ward has seen almost 170 Low Danger ES shots this year so far. At some point, you can't say he's the only one seeing a higher number of rockets from Chara.

Never said he was. In fact, I stated the exact opposite. Ward's numbers are terrible this year because he's giving up goals on shots that any NHL goaltender should stop. The team in front of him has nothing to do with it.

What I am saying is that shot location alone is a poor way to determine how dangerous a shot is, which that data seems to be saying.

I'm also attempting to kill the last hour and a half of work. If this conversation goes on for 10 more minutes, I'll have succeeded.

 

[Finlandia WOAT]

Never said they were.

Then what relevance does questioning the application of hex shots have to do with cam ward?

 

[Blueline Bomber]

Then what relevance does questioning the application if hex shots have to do with cam ward?

Where did I say it had anything to do with Ward?

 

[RodTheBawd]

Is there a single person saying that statistically, Ward isn't a horrible goalie? Those of us that are saying "they aren't telling the whole story" or "something doesn't add up" are met with the same exact stats that we already know are terrible. My question, which no one on the "[he is] a goaltender with no ability to stop the puck at an NHL level any more" (which is ****ing absurd) side will answer, is of the 46 goals he's allowed this season, how many of those goals are actually bad/soft? He's not above reproach, I don't see anyone arguing that. He's playing small, is continuing to struggle with fighting through traffic, and has regressed a bit from last year in terms of overcomitting. His 13-14 season, he made those highlight reel saves because he pulled himself out of position constantly. Last season, he was a lot more controlled. This season, he isn't as bad as 13-14, but is worse than 14-15. He has plenty of warts. But every game I watch I look for a reason to pin everything I can on him (or Lack) to counter my goalie apologist bias, and I can't say that those numbers are reflective of his (or Lack's more recent) play. Last game, which is fresh in everyone's mind, he gave up 2 laughably bad goals. Which has turned into "HE'S LET UP SOFTIES ALL SEASON." Where? Which games? Show me the clips of him giving up tons of soft goals this year. Sure, he's had a few, as does every goalie. Why the **** do coaches and players watch hours of gameplay if all they need to do, according to our armchair experts, is look at the numbers. Statistics are an invaluable tool in sports, and I love to see them grow and evolve in hockey. Wallym has done a much better job than myself explaining a similar view to mine. I don't have a number to give you as to why I don't think Ward has been abysmal this year. I don't think that invalidates my position, it just makes it much more difficult to argue it