Explain/discuss this. Habs have 6 D in the top 20 for Giveaways/60

[Milhouse40]

But what are the stats for these other young defensemen? It's fine saying that they are young.

The problem is not that they are young technically, it's young player playing with other inexperience players, something most teams don't do on defense. And if you look at other young D in the NHL, they are for the most part top pick, none of our players are either

 

[LaP]

The guy who count the giveaways at the bell center just doesn't know what a giveaway is. Weber giveaways doubled after being traded to montreal. It's been this way for a long time. Why do i know this? Because people were using giveaways to bring down Subban and to me it just did not make any sense. While Subban was not the best one defensively his giveaways stat was not matching the eye test. It was clearly inflated. So i gave a look at the stats and yep Markov and Petry giveaways were very high too. Petry, markov and Subban were all in the top 10 for giveaways from 2013 to 2016.

The guy at the bell center will count like a dman dumping the puck in the opponent zone for a line change as a giveaway and while technically speaking you'ee giving the puck away it's done on purpose. He will also count one of the lob Subban was doing to put pressure on a dman forcing him to control a bouncing puck landing in front of him before Patch could pressure him as giving the puck away even if the dman would control the puck after battling it with a fore checker. Etc ...

Giveaways is a useless stat honestly like many other subjective stats.

 

[Jigger77]

This is just my take on this, but it's not just a D thing imo. The D ends up looking bad but often the forwards are the initial cause. Plays like this were rampant this year. If Caufield who is under no pressure at all just chips this out of the zone off the wall this whole sequence and goal doesn't happen. I shot this video because I was watching them do this over and over again and it was killing me. I'm a strong proponent of spending as little time in your D zone as humanly possible. That decision to pass it back to Barron there was either a bad read or bad coaching. That puck needs to be moved up ice there. Yes that likely means giving up possession but at least you're in a position where they can regroup. Passing it back to your D in this case it just asking for trouble. This happened way too much this year. I get it. They want to play a possession game. If they insist on doing this then they need the entire team to make better reads. There are situations where I could be convinced passing it back deep in your own zone is a good play, but 90% of the time I think that the puck needs to be out of the zone when you have the chance. They spend way too much time in their own zone, and in cases like this for self-imposed reasons. I've watched playoff teams closely to see if any of them do this. I saw Vancouver do it once or twice, no other teams do this regularly like Montreal did this year. Just my 2 cents.

 

[nhlfan9191]

They don’t play structured hockey. A lot of the time the forwards are all over the place and not open for an easy outlet or short pass outside of when Suzuki/Caufield/Slafkovsky are on the ice.

 

[Habs Halifax]

This is just my take on this, but it's not just a D thing imo. The D ends up looking bad but often the forwards are the initial cause. Plays like this were rampant this year. If Caufield who is under no pressure at all just chips this out of the zone off the wall this whole sequence and goal doesn't happen. I shot this video because I was watching them do this over and over again and it was killing me. I'm a strong proponent of spending as little time in your D zone as humanly possible. That decision to pass it back to Barron there was either a bad read or bad coaching. That puck needs to be moved up ice there. Yes that likely means giving up possession but at least you're in a position where they can regroup. Passing it back to your D in this case it just asking for trouble. This happened way too much this year. I get it. They want to play a possession game. If they insist on doing this then they need the entire team to make better reads. There are situations where I could be convinced passing it back deep in your own zone is a good play, but 90% of the time I think that the puck needs to be out of the zone when you have the chance. They spend way too much time in their own zone, and in cases like this for self-imposed reasons. I've watched playoff teams closely to see if any of them do this. I saw Vancouver do it once or twice, no other teams do this regularly like Montreal did this year. Just my 2 cents.

It's likely a combo of many factors.

* Too much youth learning on the Job
* Aside from Suzuki, no real stability at center. Evans don't count lol. Hopefully Dach changes this
* Aggressive counter for the Habs home games.
* Wingers not helping as much as they should.
* Injuries and not much consistent pairings within our D.

 

[The Gr8 Dane]

Maybe the 8 shitters out of the 12 forwards on the team can try to get open for them on the breakout , I'm looking at you Anderson and Gallagher , blowing the zone like you're gonna do anything with the puck if you do get it.

 

[Habs13]

Lack of center depth - Dach played two games, Dvorak was out a long time, and Newhook is a winger. Evans is a good bottom six option but the fact that the only top six center on the team is Suzuki means a lack of transition game.

Playing pucks off the boards to our useless wingers doesn't help... wonkey donkey and the ghost of Gallagher, Pearson, White, eeesssh!

Then, consider how young the D core is - and how bad Matheson is in the defensive side of the game. Solid offensively, but a nightmare keeping pucks in at the blueline. I'm surprised his isn't higher up the list, to be honest.

 

[tooji]

Our D is a bunch of rookies and mid-tier vets

 

[Beer and Chips]

Our D is a bunch of rookies and mid-tier vets

Ghule, Harris, Kovacevic, and Xhekaj will be starting their third year.

 

[OnTheRun]

Deep neural networks for Human Pose Estimation could measure this accurately from video, in real time. I could put a PhD student on it if the Habs wanted to fund it.

This is a bit off topic but, pretty sure SportLogiq have been doing this for age.

 

[Habitant#1]

I would take any stat that isn't centrally counted with massive, massive piles of salt.

Actually, no. I would just throw these numbers in the trash.

 

[Sasha Orlov]

Our team is garbage

 

[morhilane]

This is a bit off topic but, pretty sure SportLogiq have been doing this for age.

I'm not sure why the NHL isn't doing it. They already set up cameras all over for NHL Edge metrics (since 3-4 years ago), but a lots of stats are still man counted. Edge only checks shots and movement/time spent on the ice.

 

[Habs Halifax]

I would take any stat that isn't centrally counted with massive, massive piles of salt.

Actually, no. I would just throw these numbers in the trash.

Nah, you just need to be smart enough on how to use the stat. There are general trends you can take from it.... even when the stats are not 100% accurate.

I don't believe it's one extreme to the other. Be obsessive with it or completely trash it.

 

[Habs Halifax]

I'm not sure why the NHL isn't doing it. They already set up cameras all over for NHL Edge metrics (since 3-4 years ago), but a lots of stats are still man counted. Edge only checks shots and movement/time spent on the ice.

From what I have comprehended, the NHL wants each team to do their own programing/software on how to create reports from the available Advance stats they provide with puck tracking and player tracking.

They are still doing the old school stats but also digging deeper into the advance stats. Remember, Bettman is a business man... he rather the teams spend the management recourses on it.

 

[VirginiaMtlExpat]

This is a bit off topic but, pretty sure SportLogiq have been doing this for age.

I'm familiar with them. They segment and track players in real time, but they do not perform human pose estimation, i.e. stick figures, based on video. That said, if they track the puck carrier as well (i.e. segment the puck in real time), they can probably do a basic version of that. But without the stick figures based on HPE, they may run into ambiguity on who the puck carrier is and thus how many giveaways take place.

Edit: the reason that it's ambiguous is that with just the segmentation based on a box, without the pose estimation, a player A deking another player B, with the puck going between B's legs, would actually look like a change of possession based on proximity to their algorithm, then B losing it to A again according to proximity, though a hockey fan can clearly see that A is retaining possession throughout. A stick figure approach can deduce that; what SportLogiq has cannot, and would tend to inflate those numbers.

I remember their CEO Mitchell Wasserman from Marianopolis. Incredibly bright guy. I'm linked in with Mehrsan Javan. He and I chatted about this stuff.