Jump to content
The Official Site of the Montréal Canadiens
Canadiens de Montreal

#6 Shea Weber 2016-17


habs_93
 Share

Recommended Posts

2 hours ago, ramcharger440 said:

but he is way better than many felt at the time of the trade and he is what this team needs now and for the next few years.

I'll speak for myself, based on my pre-season assessment. No, he isn't, he's exactly as expected: one of our worst possession players and awash in power play points. Shooting and power play percentages are heavily influenced by luck, which is subject to change at any time. The underlying numbers of this team are still extremely bad, and Weber is contributing to them being bad. We are yet again sailing aloft on the Icarusian wings of PDO. Given the fresh memories of last season, slight caution about projecting October over 5 months seems reasonable.

What the team needs now is what every NHL team needs if it wants to win: productive time in the offensive zone and possession of the puck. We're getting that—from Jeff Petry.

Link to comment
Share on other sites

On 10/25/2016 at 4:26 AM, ramcharger440 said:

Pretty happy so far and being as we are pretty much in our supposed window now is when we need to be happy.

I agree.  

Look, you can say +/- is meaningless but usually the extremes are a pretty good indicator of something.  Weber is leading the entire NHL in +/- which is frankly unheard of for a big minute defensman not named Orr or Robinson.  


That said we're talking about a VERY small sample size but rather than argue 'well he might come crumbling down" or "this isnt sustainable" lets just sit back and enjoy.  Right now he's playing some fantastic hockey and, more importantly, our whole roster seems to be gaining confidence. I am still not positive what to make of Emelin-Weber (although they didnt look too bad last night) but if they can work,  Im not sure we've had this good a second and third pairing in the past 30 years. 

Link to comment
Share on other sites

The fact he is playing in the east is a plus too, less travel smaller teams all helping to contribute to an all world start to the season. I for one am not worried in the slightest about any collapse of Webber or the team. last year was the anomaly in my opinion, a bunch of injuries and a lack of something perhaps it was leadership or grit or some issues in the room who knows none of us were in it so we will never know for sure. this is a team with quite a few new parts playing themselves into a cohesive unit that should have another great year after a snake bitten one last year.

Link to comment
Share on other sites

I love what I've seen so far. He brings a strong defensive game along with the feeling of confidence. Some what like Carey's calming influence. Like some who don't care about plus minus . I think the advanced possession stats are way overrated. He does bring the intangibles. Like leadership and strength. I like what I've seen so far. I may be alone but I liked the trade to start with myself.

Link to comment
Share on other sites

3 hours ago, maas_art said:

Look, you can say +/- is meaningless but usually the extremes are a pretty good indicator of something.

Extreme quantities of meaningless figures are just as meaningless as ones near the mean, though. If you're in the mood for a bit of a scholarly read, here's some material by actual professional statisticians on the topic of plus/minus: Gramacy, Robert B., Matt Taddy, and Sen Tian. "Hockey Player Performance via Regularized Logistic Regression."

27 pages is a lot, so I'll excerpt a bit (emphasis mine):

Quote

The [+/-] score represents what statisticians call a marginal effect: the average change in some response (goals for-vs-against) with change in some covariate (a player being on the ice) without accounting for whatever else changes at the same time. It is an aggregate measure that averages over the contributions of other factors, such as teammates and opponents...

...since it measures a marginal effect, the plus-minus is impacted by many factors beyond player ability, which is the actual quantity of interest. The ability of a player’s teammates, or the quality of opponents, are not taken into account. The amount of playing time is also not factored in, meaning plus-minus stats are much noisier for some players than others. Finally, goalies, teams, coaches, salaries, situations, special teams, etc. – which all clearly contribute to the nature of play, and thus to goals – are neither accounted for when determining player ability by plus-minus, and nor are they used to explain variation in the goals scored against a particular player or team.

+/- does not merely poorly describe a player's total skill level, it is entirely illusory. Not mentioned in that excerpt is that goals in the NHL are relatively rare, as far as events go. Which makes it all the more useless, because it's influenced by random noise. It's one of those "not even wrong" situations. Here's another article, for good measure: http://www.arcticicehockey.com/2014/6/5/5602668/why-plus-minus-is-the-worst-statistic-in-hockey

Link to comment
Share on other sites

7 hours ago, habs_93 said:

I'll speak for myself, based on my pre-season assessment. No, he isn't, he's exactly as expected: one of our worst possession players and awash in power play points. Shooting and power play percentages are heavily influenced by luck, which is subject to change at any time. The underlying numbers of this team are still extremely bad, and Weber is contributing to them being bad. We are yet again sailing aloft on the Icarusian wings of PDO. Given the fresh memories of last season, slight caution about projecting October over 5 months seems reasonable.

What the team needs now is what every NHL team needs if it wants to win: productive time in the offensive zone and possession of the puck. We're getting that—from Jeff Petry.

Petry is playing pretty damn good but not as good as Weber. I don't know why your so down on Weber but Subban is gone forget about him. I don't think Weber even wanted to come to Montreal but he's certainly making the most of it. and in my opinion possession is just as much a useless stat as +/-. His job as a d-man is to limit good scoring opportunities. and that's exactly what he is doing. plus pitching in a little on offence.  give him credit while he is playing good if he starts playing bad then crap on him.

Link to comment
Share on other sites

2 hours ago, CaptWelly said:

I love what I've seen so far. He brings a strong defensive game along with the feeling of confidence. Some what like Carey's calming influence. Like some who don't care about plus minus . I think the advanced possession stats are way overrated. He does bring the intangibles. Like leadership and strength. I like what I've seen so far. I may be alone but I liked the trade to start with myself.

You are far from alone. I called this from the moment the trade was made. They will all come around when the obvious comparison in the win loss column starts taking over. We are doing great with Carey and with Al so very little to argue over. Weber is a beast in his own end and awesone as far as offense goes. All i have to say is P.K. wish you luck but later dude

Link to comment
Share on other sites

 I suppose that leading the nhl in points by a defenseman, or leading our team in scoring or being a top 10 ice time player are all moot or irrelevant too. 

I disliked the trade too. I still think its going to be very painful in a few years but the negativity towards Weber at this point is frankly looking pretty silly.  He's been a beast.  Will it change? Will he plummet to earth? Maybe.  But if the Norris trophy was given out right now he'd be the unanimous winner.   Every player out there has their weaknesses and Weber is no exception but his strengths are clearly outweighing his weaknesses right now. 

Further that, since all the animosity towards him seems to be a direct result of the trade: so far this year, he's played significantly better than Subban who has been disastrous. Its not just that he's not scoring he has looked horrible at times in Nashville.  I have caught about 4 of their 6 games & he looks nothing like the PK we know.  Im absolutely sure he'll level out & get back to normal but right now he's be caught out of position on nearly every other shift it seems. 

 

Link to comment
Share on other sites

7 hours ago, habs_93 said:

I'll speak for myself, based on my pre-season assessment. No, he isn't, he's exactly as expected: one of our worst possession players and awash in power play points. Shooting and power play percentages are heavily influenced by luck, which is subject to change at any time. The underlying numbers of this team are still extremely bad, and Weber is contributing to them being bad. We are yet again sailing aloft on the Icarusian wings of PDO. Given the fresh memories of last season, slight caution about projecting October over 5 months seems reasonable.

What the team needs now is what every NHL team needs if it wants to win: productive time in the offensive zone and possession of the puck. We're getting that—from Jeff Petry.

pre-season assessment come on. how many did he actually play and here's a little article from 2015 on possession. http://www.thehockeynews.com/news/article/how-did-the-best-possession-team-in-the-nhl-miss-the-playoffs-and-how-did-one-of-the-worst-get-in

Link to comment
Share on other sites

2 hours ago, maas_art said:

He's been a beast.  Will it change? Will he plummet to earth? Maybe.  But if the Norris trophy was given out right now he'd be the unanimous winner.   Every player out there has their weaknesses and Weber is no exception but his strengths are clearly outweighing his weaknesses right now. 

I just can't be optimistic about it at all when his strength appears to be playing in front of Carey Price. #12 in the league in PDO (111.87) among all players with 60 or more minutes, #1 defenceman in PDO, all while sitting on 50.48% adjusted 5-on-5 Corsi For. (data from Corsica.hockey). Corsica's Expected PDO model for Weber is 98.66. That's not a small difference. There's regression, and then there's what is all but likely to happen next. The team gets cratered and generates almost nothing 5-on-5 when he's on the ice. It's not a good situation, because everyone has to know the Michel Therrien playbook by now. When adversity hits—and with a team PDO of 106.41, it most certainly will—the response will be: more Weber!

He needs to be deployed differently. We're winning games right now despite the amount of minutes he's playing because Al Montoya (.963) and Carey Price (.954) are #4 and #5 in 5-on-5 adjusted save percentage and the shooting percentages are in low earth orbit. We would be a significantly different team with Pekka Rinne (.911 adj 5-on-5 SV%) and Marek Mazanec (.870 adj 5-on-5 SV%). League mean is closer to .911. This is a house of cards.

 

2 hours ago, xxdocxx said:

and in my opinion possession is just as much a useless stat as +/-.

Why is that?

 

2 hours ago, xxdocxx said:

but Subban is gone forget about him.

Thanks for the suggestion on what I should do.

Link to comment
Share on other sites

34 minutes ago, habs_93 said:

I just can't be optimistic about it at all when his strength appears to be playing in front of Carey Price. #12 in the league in PDO (111.87) among all players with 60 or more minutes, #1 defenceman in PDO, all while sitting on 50.48% adjusted 5-on-5 Corsi For. (data from Corsica.hockey). Corsica's Expected PDO model for Weber is 98.66. That's not a small difference. There's regression, and then there's what is all but likely to happen next. The team gets cratered and generates almost nothing 5-on-5 when he's on the ice. It's not a good situation, because everyone has to know the Michel Therrien playbook by now. When adversity hits—and with a team PDO of 106.41, it most certainly will—the response will be: more Weber!

He needs to be deployed differently. We're winning games right now despite the amount of minutes he's playing because Al Montoya (.963) and Carey Price (.954) are #4 and #5 in 5-on-5 adjusted save percentage and the shooting percentages are in low earth orbit. We would be a significantly different team with Pekka Rinne (.911 adj 5-on-5 SV%) and Marek Mazanec (.870 adj 5-on-5 SV%). League mean is closer to .911. This is a house of cards.

But its chicken and Egg. 

You could just as easily argue that our goalies have such high percentages because our defense is keeping the shooting to a minimum.  

One of the reason advanced stats fail is that they treat all shots the same way.  If you could go back in time & look at advanced stats in our late 80s early 90s teams you'd be horrified because our whole system was based around "let Patrick see the shot - as many shots as the other team wants to shoot - and clear the rebound"

 

Im not sure what to think at this point, and if it is or isnt sustainable, but what i do know is that Im going to happily enjoy our wins, not look for reasons why "we shouldnt have won." 

Link to comment
Share on other sites

Just now, maas_art said:

You could just as easily argue that our goalies have such high percentages because our defense is keeping the shooting to a minimum. 

That's the thing, though: we're not. We're 23rd in the league in 5-on-5 shots against per 60 (according to Corsica.hockey).

 

9 minutes ago, maas_art said:

One of the reason advanced stats fail is that they treat all shots the same way.

That's done because shots (and shot events, in general) are so common. Putting all shots in the same bin is one of those things that doesn't want to go down easily. It viscerally feels wrong because we've all loved and watched the game for years and years, and it feeds a confirmation bias. There are people (like me) who try to take timing (inferring rebounds, etc) and the NHL's primitive location data into account, which when tested seems to provide more useful information. But when you deal with the volume of events that the modern NHL does, the basic shot information has empirical meaning even if the outliers are memorable.

Link to comment
Share on other sites

So this is my view on Corsi and PDO and advanced stats in generalized statements:

- Corsi is a more valuable tool than +/-, as 93 alluded to, because you're measuring many more events than just goals. When you measure fewer events, there is more chance that random luck and noise dirty your assessment. When you measure shots taken while a player is on, that gives you more events than goals and gives you a higher chance of getting a true evaluation of the player's impact on the game. And when you measure shot attempts, it gives you even more data points, which gives you even more chance at getting to the truth.

- Now the question is whether shot attempts really matter. Why should a blocked shot matter as much as a goal, which is captured by +/- alone? Well maybe it does and maybe it doesn't. A goal definitely matter more to the outcome of the game, but we capture a lot of a player's ability to contribute to scoring by measuring traditional stats like  goals and assists. Does +/- really add that much more? The thing about goals and +/- is that there's more chance those numbers won't be reproducible unless the things driving scoring are being reproduced too.

I'll clarify. Let's look at three players from the Habs last year... Pacioretty, Weise, and Galchenyuk. Now to start the year, we could look at a guy like Pacioretty and say, there's a player who's scoring goals, but he's doing so because his line maintains good possession and he gets lots of shots and scores on some of them. Weise also scored a lot of goals, but he often had bad possession numbers, and he had a really high shooting percentage and PDO. Galchenyuk, at the start of the year, had really great possession numbers but wasn't scoring as much as we would have liked. Pacioretty is the type of player whose worth is measurable in terms of basic statistics but it's also reflected in pretty good possession numbers and advanced stats too. Supporters of both groups would be right to say their stats show he's a good player. In Weise's case, though, advanced stats supporters would say they don't like the look of Weise's stats whereas basic stats supporters would say 10 goals in 30 games looks great any way you cut it. The reverse would be true of Galchenyuk, where basic stats would be asking why he wasn't scoring as much as we would have liked but advanced stats would show he had great possession numbers and a low PDO, meaning he was suffering from his share of bad luck. So what happened? Weise's luck bottomed out, while Galchenyuk soared as the season progressed. Now these are maybe somewhat extreme examples, but the premise is that the basic stats did a good job at telling us what a player has done but the advanced stats did a better job at telling us what they were likely to do. Mitchell has 4 goals in 8 games this year but awful possession numbers. Do we expect him to score 40 goals, as his pace would suggest, or do we expect him to level off? If your expectation is the latter, why is that?

- So let's come back to Corsi or Fenwick... measuring all types of shot attempts, no matter whether they hit the net or not, no matter whether they go in or not. Now maybe there should be a measure that gives more weight to a goal or a shot on target than one that's blocked or misses, but in terms of what we have, Corsi does a fair job at measuring possession. The fact is that players, for the most part, tend to try to shoot when they're in a position to hit the net or at least have a chance at scoring. So shot attempts, to a large degree, reflects how often a player's line is in possession of the puck in the offensive zone, which is a pretty good surrogate for how well they're playing, or at least as good a surrogate as we're probably going to get without getting into very complicated calculations that the average Joe won't be able to follow.

- Is Corsi perfect? Absolutely not. But possession data seems to be a better predictive tool of how players/teams will do in the future than just how many goals a player has or what +/- they have. I've asked this before, but how does Pacioretty go from being +38 to -10 one year later? Did he go from being a star to being an awful player? Did he forget how to play defence? Or is it that that stat has a lot of variability and random chance affecting it? In Pacman's case, he went from playing on a team where the goalie stopped everything to a team where the goalie stopped a lot less, and that probably accounts for 75% of the change in his +/-. But the fact Price stops lots of pucks and Condon/Scrivens stopped less doesn't make Pacman a better or worse defensive player. The fact Pacman hopped onto the ice half a second before the opposing team roofs a breakaway into the back of our net has pretty little to do with him as well. Now obviously shot attempts can happen in situations that have very little to do with the actual player too. But because there are so many more shot attempts than goal events, the white noise of events that don't have anything to do with the player are drowned out more. Including all types of shot attempts probably still gives us a useful reflection of possession and value while also giving us more data points to get closer to the player's real worth.

- Last thing I want to address is the story about how Corsi can't be useful because a good Corsi team missed the playoffs. Again, Corsi is not perfect. No stat is perfect. But a good Corsi team probably has a better shot at longer-term success than a poor Corsi team, like the Leafs or Avs of a couple of seasons ago or our team last year. Let's use a baseball analogy... imagine you have a player who hits for a .350 average and another guy who hits for .200. Now in any one at bat, either hitter could come up with a hit. But just because in one inning the .200 hitter gets a hit and the .350 hitter strikes out, it doesn't mean that batting average is a useless stat or isn't predictive. There are going to be outliers, there are going to be instances where despite the odds, sometimes things don't go as planned. But that's sports. Sometimes the team with a 30% chance of winning a game beats the favorite. But just because the underdog upsets the favorite in one game, doesn't mean they would do it again if the game were replayed. Corsi tells us the odds a player or team might do well in the future. PDO tells us how likely a team is to regress or improve towards the mean. In Weber's case, he's put up great numbers thus far, but his possession numbers at ES are suboptimal. Now Carey playing phenomenal hockey will mask a low Corsi's effects to some degree, but is Weber likely to end up with a 100-point season? No. Is he likely to continue to benefit from a PDO of 112? No. The odds are high that his play and his success will drop off, and that's fine. He can still be a good player even if that happens, but advanced stats are just telling us he's not going to be as good as some might think he's going to be based on the small sample size of games we've had thus far.

Link to comment
Share on other sites

Having a great year so far with whoever he gets paired with, heck he is even making Emelin look good! that in and of itself is some cause for celebration! it is true PK is sucking right now but I am sure that will turn around at some point and Webber is rocking for us so no matter what side of the coin you find yourself on lets just be happy that we are winning and see where it gets us.

Link to comment
Share on other sites

2 hours ago, ramcharger440 said:

Goal scoring machine! what a player even with a tough week behind him he still had plenty of jump for this game. his partner was weak though.

Interestingly, there seems to be a growing number of Nashville fans who, while happy with PK, think that losing Weber has been a very big blow to the team.  

One of the Nashville papers called it the "worst trade in Nashville Predators history" over the weekend.  A little premature one might think - and I am sure the ship will right itself - but its certainly interesting to see the value they placed on Weber over there. 

Link to comment
Share on other sites

19 minutes ago, maas_art said:

Interestingly, there seems to be a growing number of Nashville fans who, while happy with PK, think that losing Weber has been a very big blow to the team.  

One of the Nashville papers called it the "worst trade in Nashville Predators history" over the weekend.  A little premature one might think,and I am sure the ship will right itself but its certainly interesting to see the value they placed on Weber over there. 

It's hard not to like Weber, just as it's hard not to like Subban IMO. They're both great players. For the Nashville media, they must look at the success Weber and Montreal are having and then look at their own team's start this year and wonder what's going wrong. Nashville has a strong team though, and that ship will be righted. They're just in a really tough division, and their goaltending hasn't been up to snuff, especially compared to Carey. I wonder, if the Leafs had scored 5 goals last night and won the game (not unreasonable with how they played if Carey hadn't been in goal) and Weber had ended up -3 whether there would be as much negativity in Nashville.

To the eye test, you remember the big plays in games... Weber makes big hits, he takes booming shots and scores goals, he's on the ice a lot, so you notice him doing a lot of good things. He also doesn't control the puck a lot (per ice time, he touches the puck less than most of the other D on the team) and he doesn't try as many tough or creative passes as someone like Subban, Markov, Petry, or Beaulieu so you don't see him making big turnovers. So to the eye test, you notice a lot of good things and not a lot of bad ones. But if you look a little closer, you also note to yourself that he doesn't control the puck a lot, that he doesn't get onto loose pucks very often, that he often makes the hit but lets the puck stay with the opposition, and that he allows a lot of shots and shot attempts when on the ice.

That being said, I do like a lot of what Weber has done here. It's hard not to love what a threat he is on the PP, the way Sheldon Souray was. I still absolutely believe Subban is the better on-ice player and much better value when you compare the remaining lengths of contract. But if Carey can stay healthy, I'm coming around to the fact that it's okay for a player like Weber to allow shot attempts so long as he's limiting second chances and rebounds against, because Carey is good enough to stop most of those first-shot attempts. And what Weber does for the PP means we can often get the 1-2 goals we need to win a game with Carey in net. But subtract Carey and a Weber-Condon duo or even a Weber-Montoya duo over a full season could be a disaster.

 

Link to comment
Share on other sites

35 minutes ago, maas_art said:

Interestingly, there seems to be a growing number of Nashville fans who, while happy with PK, think that losing Weber has been a very big blow to the team.

I think forcing their skaters to play in front of this:

...is a significantly bigger factor to their early season struggles than anything else. Nashville's goaltending is a garbage fire.

Link to comment
Share on other sites

Yeah i agree that its silly at this point to even try to pin the blame (or success) on either player. Its just interesting to see how differently its being viewed over there. Thankfully they still seem to think that PK is a great player (which he is) and havent turned on him yet but ive read a quite a few articles/comments with people questioning whether Weber was actually making Josi look better than he is, rather than what the stats made many believe (that Josi was propping up Weber).  Maybe Weber is so old-school that he just defies advanced stats :P   We'll have to make up a series of Weber jokes along the lines of chuck norris.   "Advanced stats are so scared of Weber they Fenwick themselves." 

I still contend this trade may end up being a win for us - but only if we win the cup in the next couple of years.  If that doesnt happen, its most likely a disaster.  Whether we win because of Weber or not is moot - the trade will still be heralded a win, just like Neiwendyk for Iginla, even though Jerome went to to have so many great years in Calgary, Dallas won that cup at least in part because of Joe.  


its a very risky move though, because if we dont win - and the odds arent with us - then its likely Bergevin's job (which it should be of course). 

Link to comment
Share on other sites

2 minutes ago, maas_art said:

Yeah i agree that its silly at this point to even try to pin the blame (or success) on either player. Its just interesting to see how differently its being viewed over there. Thankfully they still seem to think that PK is a great player (which he is) and havent turned on him yet but ive read a quite a few articles/comments with people questioning whether Weber was actually making Josi look better than he is. 

I still contend this trade may end up being a win for us - but only if we win the cup in the next couple of years.  If that doesnt happen, its most likely a disaster.  Whether we win because of Weber or not is moot - the trade will still be heralded a win, just like Neiwendyk for Iginla, even though Jerome went to to have so many great years in Calgary, Dallas won that cup at least in part because of Joe.  


its a very risky move though, because if we dont win - and the odds arent with us - then its likely Bergevin's job (which it should be of course). 

Well just look at the Chelios-for-Denis-Savard trade. We lost that trade by a wide margin, yet the Habs won the Cup that very season, so the talk about how bad the trade was never materialized in the short term and is only really something we talk about in retrospect. But it didn't take much longer for Serge Savard and Jacques Demers to pay the price for the consequences of not having Chelios in the line-up and relying on goaltending and a lot of luck to win them a Cup. That's pretty much the formula Therrien and Bergevin are using now. The question today is the same as it was looking at the Chelios trade: yes, we won a Cup with the guys we had, but could we have been even better for longer and won multiple Cups if we hadn't made that deal? We may well win a Cup in the next year or two on Carey's back if we're able to get lucky enough, but that doesn't mean the trade was a good one nor does it mean we can't improve upon our head coach or other areas of deficiency.

Link to comment
Share on other sites

2 minutes ago, BigTed3 said:

We may well win a Cup in the next year or two on Carey's back if we're able to get lucky enough, but that doesn't mean the trade was a good one nor does it mean we can't improve upon our head coach or other areas of deficiency.

That's my problem with the trade right now. It happened, whatever, we're here. The issue is that the organization flatly refuses to recognize the reality it has for itself now.

A properly deployed Shea Weber with the right kind of system to play to his strengths and limit the impact of his weaknesses is one thing. We're not getting that. Instead, we're getting something very predictable: the last three years' Canadiens with P.K. Subban removed and Shea Weber inserted with almost no tangible change in how the team plays hockey. And it's just going to get worse as the minutes pile up and the season wears on. Even if you think Shea Weber is the best defenceman in the league, you can't be happy with a system that leads to opponents having 64% of the puck possession when he's on the ice 5-on-5, even if you (quite reasonably) think Carey Price can probably handle it. So let's fix it.

"Strong defence" is dead. We hit people when hitting people generates positive puck possession, otherwise we do the next best thing which results in the Habs skating up ice. Standing around near the crease is a useless endeavour; pressure the puck carrier, force a turnover, and move. Weber is a large individual, and he can make passing lanes less inviting while doing other things, too. He's a smart guy, so it'll be like having 6 skaters on the ice against lesser skilled teams.

We've got Carey Price, so let's burn up the scorekeepers' desk. High event hockey every night. Foot speed is not Weber's thing. Luckily, high event hockey doesn't have to be "fast" hockey, at all, so there's no reason it should be a problem. In fact, quite the opposite. The tendency to automatically lose possession when entering the offensive zone means the opponent's stretch passes will connect more often, which makes a slower player look way worse than they actually are. Carry the puck in and establish a battery. Don't pass three times when you can pass once and then shoot. Shoot wide to play for rebounds if there's too much traffic. And so on.

Link to comment
Share on other sites

Guest
This topic is now closed to further replies.
 Share

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...