Jump to content
The Official Site of the Vancouver Canucks
Canucks Community

Interesting Canucks Advanced Stats


JamesB

Recommended Posts

Now that the NHL is posting advanced stats it is possible to get a different look at the performance of Canuck players. (Of course advanced stats have been available from a variety of sources for a while, although different sources did different things with the data and reliability was always an issue.) In any case, here are some facts that I find interesting.

The points per 20 minutes played number shows that, not surprisingly, among Canuck regulars, the top three are Henrik at 0.96. Daniel and 0.95 and Vrbata at 0.87. But what is surprising is that next are Horvat at 0.68 and Kassian at 0.64.

Then we have Burrows, Richardson, and Hansen. Then, interestingly, Higgins, Bonino, Matthias, and Vey are all tied at 0.57. McMillan, with 3 pts in 52 games this year (mostly with Arizona) has an absurdly low number.

Of course the main innovation of advanced stats is to look at shot attempts. There are disadvantages as it does not take shot quality into account (and it does not account for strength of competition). But there are some interesting numbers.

The the team relative shot performance while a given player is on the ice for 5 on 5 play is a key stat (sometimes called relative CORSI), Once again McMillan's numbers are extremely low. Horvat is also fairly low, reflecting the fact for the first part of the year his line generated relatively few shots -- focusing largely on avoiding defensive mistakes and just dumping the puck into the offensive zone instead of generating shot attempts. That has changed a lot in the past 10 or 15 games.

But Kassian's numbers are pretty good -- certainly above the team average. (The top 3 are Henrik, Kennins (but be wary of small sample size), and Daniel.)

One overall conclusion from the advanced stats is that Kassian is pretty good.

A skeptic might argue that Kassian's stats benefit from weak quality of competition so I checked out the quality of competition numbers at Behind the Net.

Not surprisingly, the Canucks who have faced the toughest quality of competition are the Sedins, Vrbata, Tanev and Edler. (They are the top 5.) Surprisingly Corrado is 6th and despite that his advanced stats are good (once again with a small sample size).

Back to Kassian, his quality of competition is only slightly below average and above that of both Vey and Richardson. and not far below Horvat and Matthias.

My takeaways include:

1. Corrado and Kennins look very good but that is based on small sample sizes. (And Clendenning looks like a good pickup).

2. Kassian is surprisingly good. Playing McMillan ahead of him can only be described as absurd.

3. The top line and the top D pairing are perhaps even better than we give them credit for.

4. CDC favorites Horvat, Dorsett, Richardson, and Matthias are not bad but are perhaps not quite as good as average CDC opinion would suggest. (But don't get me wrong. The fact that Bo is doing as well as he is as 19 year old is great. But I am a bit less sold on Dorsett, Richardson, and Matthias than I was, and they are all UFAs.).

Link to comment
Share on other sites

Corsi-wise, there's a pretty big split in our roster. Matthias, Dorsett, Richardson, Hansen and the rest of our bottom-6 are constantly getting shot at. It's always been noticeable on the ice, watching them get hemmed into their own zone at the beginning of the season, but Matthias and Hansen have turned their game around since a few roster changes have been made (mainly Richardson coming out and Kenins in). They play with a much more offensive-mentality now so I'm sure their Corsi ratings will improve.

Horvat has been impressive as a rookie but I was surprised to see Kassian rated so highly.

Defensively, there's no doubt that Edler and Tanev have been our best defencemen and that quite frankly everyone else has been a solid negative Corsi player all year long. I haven't been pleased with any of our bottom-4 defencemen and hopefully Benning addresses this, because their shocking Corsi ratings confirm how badly they've been getting shot at when on the ice. Sbisa, Stanton, Hamhuis and Bieksa have been brutal in their own zone. They can't seem to move the puck up the ice without turning it over or getting caught which is why their Corsi ratings are so bad. Now Clendening walks into the lineup and has a pretty decent Corsi rating because he has a superb first pass out of his own zone, and can actually let a shot go in the offensive zone.

Link to comment
Share on other sites

Defensively, there's no doubt that Edler and Tanev have been our best defencemen and that quite frankly everyone else has been a solid negative Corsi player all year long. I haven't been pleased with any of our bottom-4 defencemen and hopefully Benning addresses this, because their shocking Corsi ratings confirm how badly they've been getting shot at when on the ice. Sbisa, Stanton, Hamhuis and Bieksa have been brutal in their own zone. They can't seem to move the puck up the ice without turning it over or getting caught which is why their Corsi ratings are so bad. Now Clendening walks into the lineup and has a pretty decent Corsi rating because he has a superb first pass out of his own zone, and can actually let a shot go in the offensive zone.

If SAT is the same thing as Corsi, then Hamhuis is positive. Probably more impressive since other than Corrado the partner's he's played 5 on 5 the most with have a negative SAT, and a fairly large one at that.

Hamhuis: +6

Bieksa: -70

Sbisa: -92

Link to comment
Share on other sites

Corsi-wise, there's a pretty big split in our roster. Matthias, Dorsett, Richardson, Hansen and the rest of our bottom-6 are constantly getting shot at. It's always been noticeable on the ice, watching them get hemmed into their own zone at the beginning of the season, but Matthias and Hansen have turned their game around since a few roster changes have been made (mainly Richardson coming out and Kenins in). They play with a much more offensive-mentality now so I'm sure their Corsi ratings will improve.

Horvat has been impressive as a rookie but I was surprised to see Kassian rated so highly.

Defensively, there's no doubt that Edler and Tanev have been our best defencemen and that quite frankly everyone else has been a solid negative Corsi player all year long. I haven't been pleased with any of our bottom-4 defencemen and hopefully Benning addresses this, because their shocking Corsi ratings confirm how badly they've been getting shot at when on the ice. Sbisa, Stanton, Hamhuis and Bieksa have been brutal in their own zone. They can't seem to move the puck up the ice without turning it over or getting caught which is why their Corsi ratings are so bad. Now Clendening walks into the lineup and has a pretty decent Corsi rating because he has a superb first pass out of his own zone, and can actually let a shot go in the offensive zone.

This is a better summary of the situation with our D than I provided. However, it is a bit tough on Hamhuis. He has not been as good as in the past two years, but his percentage CORSI (SAT% on the NHL site) is not bad (above 50%) and his strength of opposition is quite high. And he has usually been paired with someone who is not a legitimate top 4 D.

The Canucks actually have 5 defencemen with decent numbers. Edler and Tanev are of course very good. Corrado (based on a small sample) looks good and Hamhuis is also pretty good -- certainly legitimate top 4. And Glendenning looks good, for the reasons you indicate.

That means we really only need one more solid D, preferably a left handed shot. Stanton and Sbisa have both struggled, but I think both have improved. And on the right side Bieksa was having a bad year and Weber is marginal.

Right now, assuming everyone was healthy Edler and Tanev would be the top pairing, Hammer and Corrado would be the second pairing and I would go with Stanton and Glendenning as the 3rd pairing with Bieksa as 7th man and Sbisa as 8th man. I know this won't happen however,

And while so many Ds are injured I would like to see Sanguinetti get a shot.

Link to comment
Share on other sites

Right now, assuming everyone was healthy Edler and Tanev would be the top pairing, Hammer and Corrado would be the second pairing and I would go with Stanton and Glendenning as the 3rd pairing with Bieksa as 7th man and Sbisa as 8th man. I know this won't happen however,

Bieksa deserves more credit than that. He played more minutes and against tougher competition than all of them except the top 3.

Not to mention the other things he brings, which can't really be measured via stats.

Link to comment
Share on other sites

One overall conclusion from the advanced stats is that Kassian is pretty good.

Hence advanced stats are merely one of many tools in the tool box to judge a hockey player

It tells you something and depending on the information given, you may want to keep in mind, but you can just as easily shrug it off when the information given by it contradicts all other evidence - for example, when it tells you that Chris Tanev is better than Shea Weber

Link to comment
Share on other sites

Nice analysis of the stats.

I found Kenins to have many very good numbers, including his points-per-20 mins of ice time. As you said, however, it's a small sample size. Still, it's exciting to see young players on the team performing so well.

Link to comment
Share on other sites

Advanced stats say that players like Stanton and Sbisa aren't good, but they play defensive minutes so of course they will be outshot. Besides, goals against is more important than shots against. Look at Stanton's plus/minus, despite playing many games in the top 4 (or getting top 4 minutes) as a 3rd pair D-man in his sophomore season.

IMO, advanced stats are only sometimes relevant.

Link to comment
Share on other sites

Now that the NHL is posting advanced stats it is possible to get a different look at the performance of Canuck players. (Of course advanced stats have been available from a variety of sources for a while, although different sources did different things with the data and reliability was always an issue.) In any case, here are some facts that I find interesting.

The points per 20 minutes played number shows that, not surprisingly, among Canuck regulars, the top three are Henrik at 0.96. Daniel and 0.95 and Vrbata at 0.87. But what is surprising is that next are Horvat at 0.68 and Kassian at 0.64.

Then we have Burrows, Richardson, and Hansen. Then, interestingly, Higgins, Bonino, Matthias, and Vey are all tied at 0.57. McMillan, with 3 pts in 52 games this year (mostly with Arizona) has an absurdly low number.

Of course the main innovation of advanced stats is to look at shot attempts. There are disadvantages as it does not take shot quality into account (and it does not account for strength of competition). But there are some interesting numbers.

The the team relative shot performance while a given player is on the ice for 5 on 5 play is a key stat (sometimes called relative CORSI), Once again McMillan's numbers are extremely low. Horvat is also fairly low, reflecting the fact for the first part of the year his line generated relatively few shots -- focusing largely on avoiding defensive mistakes and just dumping the puck into the offensive zone instead of generating shot attempts. That has changed a lot in the past 10 or 15 games.

But Kassian's numbers are pretty good -- certainly above the team average. (The top 3 are Henrik, Kennins (but be wary of small sample size), and Daniel.)

One overall conclusion from the advanced stats is that Kassian is pretty good.

A skeptic might argue that Kassian's stats benefit from weak quality of competition so I checked out the quality of competition numbers at Behind the Net.

Not surprisingly, the Canucks who have faced the toughest quality of competition are the Sedins, Vrbata, Tanev and Edler. (They are the top 5.) Surprisingly Corrado is 6th and despite that his advanced stats are good (once again with a small sample size).

Back to Kassian, his quality of competition is only slightly below average and above that of both Vey and Richardson. and not far below Horvat and Matthias.

My takeaways include:

1. Corrado and Kennins look very good but that is based on small sample sizes. (And Clendenning looks like a good pickup).

2. Kassian is surprisingly good. Playing McMillan ahead of him can only be described as absurd.

3. The top line and the top D pairing are perhaps even better than we give them credit for.

4. CDC favorites Horvat, Dorsett, Richardson, and Matthias are not bad but are perhaps not quite as good as average CDC opinion would suggest. (But don't get me wrong. The fact that Bo is doing as well as he is as 19 year old is great. But I am a bit less sold on Dorsett, Richardson, and Matthias than I was, and they are all UFAs.).

Your analysis lines up with basically everything I've been saying about our team since the beginning of the season. The reason people think Kassian is invisible is because he gets very little ice time, but in the ice time he gets, even if he's not checking the boards and slamming his stick around like Dorsett, he is still pushing possession positively. Kassian gives the puck away a lot in the neutral and defensive zone, but he also helps get it in to the offensive zone, and contributes in keeping it there.

Link to comment
Share on other sites

I'm not totally sold on this advanced stats stuff yet. It seems interesting but I don't think you can use them alone to get concrete evidence about a player or a team. It seems to be better suited for a sport like baseball where everything is more "measurable".

Having more information available about players and a team is definitely a good thing and it's cool to see how players compare using advanced stats. I still think watching the games is the best way to judge who's playing well.

Link to comment
Share on other sites

So how does advanced stats measure huge hit by Sbisa to get the team going? How about Dorset going a million miles a minute to show our rookies and vets what effort looks like? How about Juice dropping the gloves to protect his teammates? How about being on the ice when miller is having one of his bad games ( he always does when we loose) can't control a rebound or stop a beach ball leading to more goals and shots against.

Oh wait, how about our PK going into a tailspin since Richardson has been gone?

You let me when Coris and your "advanced whatever" can measure all that then I will pay some attention. Hockey is a team sport. In tennis you can pay attention to stats to tell the whole story as its one on one. In baseball the play initiates with only two people playing (batter and pitcher). It doesn't work that way in football, hockey or basketball.

How the heck do you measure how many cross checks Alex Burrows is willing to take in front of the net in comparison to anyone else on the team. The guy takes Thomas Holmstrom esque punishment.

Why do you think the team drafts character guys? Because if you have a team of hardworking players from top to bottom then that effort leads players to push beyond their limits. Linden Stanley Cup Finals 1994. Character guy playing when he should've been in a hospital bed leading an above average team to a Nathan Lafayette goal away from glory. You bet that his grit, character, hard work inspired his teammates to play better even when he wasn't on their line.

ALSO

You get sheltered minutes not just against opposition which is measured BUT you get sheltered minutes when you are not on the ice in a challenging game where the team is loosing or holding onto a lead which is not measured. You put out your Richardson's, Burrows, Hansen and Higgins when the other team has momentum. That leads to negative advanced stats. I don't know if you have watched a hockey game but momentum ebbs and flows. Very rarely does a team dominate from start to finish. If most of your icetime comes when we have positive momentum then obviously your numbers are going to be better. Coaches favorites would have an inverse relationship here. The more you are relied upon in a tough situation the lower your corsi or whatever would be. Who cares what your quality of opposition is if you are on the ice when the team is up by 3? In those chaotic last couple of minutes before a game is over and we are clinging to the lead do you think the favorites are just racking up the positive shot totals for and against?

The advanced stats stuff will be a lot more interesting when you can measure a persons whole career together. That should even out a lot. In the meanwhile it is just like plus/minus. I honestly think that it has no greater relevancy than that.

Link to comment
Share on other sites

I'm not totally sold on this advanced stats stuff yet. It seems interesting but I don't think you can use them alone to get concrete evidence about a player or a team. It seems to be better suited for a sport like baseball where everything is more "measurable".

Having more information available about players and a team is definitely a good thing and it's cool to see how players compare using advanced stats. I still think watching the games is the best way to judge who's playing well.

Quite right, people tend to get immersed in the stats and forget about the context around them.

Link to comment
Share on other sites

This is a better summary of the situation with our D than I provided. However, it is a bit tough on Hamhuis. He has not been as good as in the past two years, but his percentage CORSI (SAT% on the NHL site) is not bad (above 50%) and his strength of opposition is quite high. And he has usually been paired with someone who is not a legitimate top 4 D.

The Canucks actually have 5 defencemen with decent numbers. Edler and Tanev are of course very good. Corrado (based on a small sample) looks good and Hamhuis is also pretty good -- certainly legitimate top 4. And Glendenning looks good, for the reasons you indicate.

That means we really only need one more solid D, preferably a left handed shot. Stanton and Sbisa have both struggled, but I think both have improved. And on the right side Bieksa was having a bad year and Weber is marginal.

Right now, assuming everyone was healthy Edler and Tanev would be the top pairing, Hammer and Corrado would be the second pairing and I would go with Stanton and Glendenning as the 3rd pairing with Bieksa as 7th man and Sbisa as 8th man. I know this won't happen however,

And while so many Ds are injured I would like to see Sanguinetti get a shot.

Corsi sucks and is for nerds to split the atom over.

Link to comment
Share on other sites

These stats aren't totally useless, but without context you can't make an argument for or against any player.

For example, Higgins leads the Canucks in 'USAT close'. Does that make him a great scoring threat in close games? Or is he just taking weak shots on goal instead of looking for better plays like Vrbata, who is dead-last in that category for the Canucks.

But there are some things about these stats that make you go 'hmmmmmm'...

Like USAT ahead vs USAT behind. It looks like whenever we're ahead in the score, opposing teams are able to take control of the game, because virtually ALL the Canucks have poor USAT ahead numbers. The Canucks as a team are 20th in USAT ahead. Also revealed: The Canucks are ultra-reliant on the Sedins whenever they are behind in the score. (No, really.)

Teams accross the league with good USAT ahead: Islanders, TB, St. Louis, Pittsburgh, LA, Minnesota and Chicago. LA led the NHL last season. Chicago led it in 2013. Not saying the Islander will win the cup, but it does make you go 'hmmmm'.

As per the norm, none of these stats are all too revealing. The Sedins are still our best offensive players. The Canucks have trouble holding a lead. And no specific knowledge on any player can be gained using these stats, but they can be applied to what we already know through just watching how the player performs.

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

  • Recently Browsing   0 members

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