General Leafs Talk

[Corn Flake]

A little off-topic so my apologies. At first I thought Lupul's most recent tweet was a dig toward Bozak but realize now it's just him criticizing advanced stats for whatever reason. Ha.

Joffrey Lupul ‏@JLupul 2h
If you bring certain attributes and you play to win. I'll take you on my team 7 nights a week. Lets not look at this like Moneyball.

Joffrey Lupul ‏@JLupul 2h
contracts aren't awarded by this CORSI i am hearing all about. They are awarded for an equal value of skill and depth (at a certain position

This gave me a good chuckle and set a lot of adv. stats purists spinning.

Someone I follow tweeted this, which I agree with wholeheartedly...

@MapleLeafsHH I don't question the validity of adv. stats, what I have an issue with is the complete dismissal of everything else that can't be measured.

 

[Corn Flake]

He's just saying that there is more to the game than just numbers.

Which to a certain faction of adv. stats loyalists is not true.  ie: Lupul sucks because his CORSI blah blah numbers are bad.

You know what team had really high Corsi and Fenwick numbers? Some of the best adv stats in the league? The 2009 Toronto Maple Leafs.. the team that finished 29th.

 

[Heroic Shrimp]

A little off-topic so my apologies. At first I thought Lupul's most recent tweet was a dig toward Bozak but realize now it's just him criticizing advanced stats for whatever reason. Ha.

Joffrey Lupul ‏@JLupul 2h
If you bring certain attributes and you play to win. I'll take you on my team 7 nights a week. Lets not look at this like Moneyball.

Joffrey Lupul ‏@JLupul 2h
contracts aren't awarded by this CORSI i am hearing all about. They are awarded for an equal value of skill and depth (at a certain position

Offering his support to Grabovski?

 

[GhostofPotvin29]

He's just saying that there is more to the game than just numbers.

Which to a certain faction of adv. stats loyalists is not true.  ie: Lupul sucks because his CORSI blah blah numbers are bad.

You know what team had really high Corsi and Fenwick numbers? Some of the best adv stats in the league? The 2009 Toronto Maple Leafs.. the team that finished 29th.

And just like anyone who advocates "advanced stats" they would tell you that that was a positive aspect of the 2009-10 Leafs but that it is stupid to look at it out of context.  Their goaltending was terrible.  Why must people always talk about these stats like they're being advocated as being the be-all end-all.  People who are passionate about them are passionate about using them as one of the tools in player evaluation, but it often becomes having to defend them because a lot of people want to inherently criticize their existence.

There's nothing advanced about Corsi.  It's named after a person and it tracks shots for and against.  Nothing advanced about it.  It's no more advanced than tracking SV%.  It's helpful, it's a good thing to be a positive in. 

We're going to have a full season of derisive remarks about "advanced stats" coming up on this board.

 

[GhostofPotvin29]

A little off-topic so my apologies. At first I thought Lupul's most recent tweet was a dig toward Bozak but realize now it's just him criticizing advanced stats for whatever reason. Ha.

Joffrey Lupul ‏@JLupul 2h
If you bring certain attributes and you play to win. I'll take you on my team 7 nights a week. Lets not look at this like Moneyball.

Joffrey Lupul ‏@JLupul 2h
contracts aren't awarded by this CORSI i am hearing all about. They are awarded for an equal value of skill and depth (at a certain position

This gave me a good chuckle and set a lot of adv. stats purists spinning.

Someone I follow tweeted this, which I agree with wholeheartedly...

@MapleLeafsHH I don't question the validity of adv. stats, what I have an issue with is the complete dismissal of everything else that can't be measured.

Who completely dismisses everything that can't be measured?

Most advocates of advanced stats don't talk about things that can't be measured because THEY CAN'T MEASURE THEM - so how could you talk about things about a player that you can't quantify?  How can you talk with any authority about a person's character when you don't know enough about them personally?

It's not discounting that element to not discuss it, it is acknowledging that it is impossible to do so there's no point.  If you're a GM/scout/etc where you can make judgments on a more personal level, then you can.

 

[Beowulf]

A little off-topic so my apologies. At first I thought Lupul's most recent tweet was a dig toward Bozak but realize now it's just him criticizing advanced stats for whatever reason. Ha.

Joffrey Lupul ‏@JLupul 2h
If you bring certain attributes and you play to win. I'll take you on my team 7 nights a week. Lets not look at this like Moneyball.

Joffrey Lupul ‏@JLupul 2h
contracts aren't awarded by this CORSI i am hearing all about. They are awarded for an equal value of skill and depth (at a certain position

This gave me a good chuckle and set a lot of adv. stats purists spinning.

Someone I follow tweeted this, which I agree with wholeheartedly...

@MapleLeafsHH I don't question the validity of adv. stats, what I have an issue with is the complete dismissal of everything else that can't be measured.

Who completely dismisses everything that can't be measured?

Most advocates of advanced stats don't talk about things that can't be measured because THEY CAN'T MEASURE THEM - so how could you talk about things about a player that you can't quantify?  How can you talk with any authority about a person's character when you don't know enough about them personally?

It's not discounting that element to not discuss it, it is acknowledging that it is impossible to do so there's no point.  If you're a GM/scout/etc where you can make judgments on a more personal level, then you can.

Very well said on both posts.

 

[bustaheims]

My issue with some of the advanced stats is that I don't think they really measure what they're trying to measure - at least, not on an individual basis. Some of them, like zone starts and QoC are useful in terms of putting a player's performance in context of how they were used by their coach, but others, like Corsi and Fenwick, suffer from a similar flaw as +/-, in that they're trying to relate a collective performance to the individual. That's somewhat mitigated by some of the deeper analysis people do with them, but, those same things can be done with +/-, though no one seems to care to do so. Things like Corsi and Fenwick seem like they'd be useful in measuring the performances of line combinations or defensive pairings or both together, but, on a player to player level, I just can't give them much weight.

 

[GhostofPotvin29]

My issue with some of the advanced stats is that I don't think they really measure what they're trying to measure - at least, not on an individual basis. Some of them, like zone starts and QoC are useful in terms of putting a player's performance in context of how they were used by their coach, but others, like Corsi and Fenwick, suffer from a similar flaw as +/-, in that they're trying to relate a collective performance to the individual. That's somewhat mitigated by some of the deeper analysis people do with them, but, those same things can be done with +/-, though no one seems to care to do so. Things like Corsi and Fenwick seem like they'd be useful in measuring the performances of line combinations or defensive pairings or both together, but, on a player to player level, I just can't give them much weight.

That's why you need to look at them and then look at them again over longer sample sizes.  Corsi is basically just time on attack/possession of the puck, which if it is a solid number for a long enough period of time, should be able to tell you something positive about the player/the line/the team.  That's why people compare their numbers with and without certain linemates, too.

Again, I think it is helpful and useful.

 

[bustaheims]

Again, I think it is helpful and useful.

I'm not disagreeing that they can be helpful and/or useful, I just question how helpful/useful they really are. I don't give them the same weight that you seem to. I also don't think NHL teams necessarily use the same advanced stats or all of the advanced that are available to us, though I know they do use them - reportedly, they all tend to have their own advanced measures, etc.

 

[GhostofPotvin29]

Again, I think it is helpful and useful.

I'm not disagreeing that they can be helpful and/or useful, I just question how helpful/useful they really are. I don't give them the same weight that you seem to. I also don't think NHL teams necessarily use the same advanced stats or all of the advanced that are available to us, though I know they do use them - reportedly, they all tend to have their own advanced measures, etc.

I think if the stat were actually something like the EA NHL videogame series has, the time on attack stat, that it would be thought of higher in the mainstream of stats.  But I think it's basically saying the same thing, and that Corsi numbers generally reflect scoring chances (if one is high the other is high), and that I think it is next to impossible to have sustained success without driving possession.

I don't know why it seems it has to be either/or for some.  It should be adding to the discussion of a player or a team.  There seems to be a real anger that this stuff is thought highly of.

 

[bustaheims]

I don't know why it seems it has to be either/or for some.  It should be adding to the discussion of a player or a team.  There seems to be a real anger that this stuff is thought highly of.

Well, the same thing happened in baseball with the whole Moneyball phenomenon. Traditionalists railed against until teams openly using those measures heavily showed some signs of success. That hasn't happened in hockey yet, and, the impression I get is that hockey people are even more set on traditional measures. That being said, part the anger also comes from the way some people in the advanced stats community carry themselves. They do sometimes give them impression that they feel they're on a higher level than others, etc, and that rubs some people the wrong way. And, honestly, when you still have mainstream hockey media people going on and on about things like heart and "compete level" and other such ethereal measurements, are you really surprised that the average fan would look at the advanced stats community with derision? It's basically the jocks versus the nerds.

 

[maplesyrup]

... As the stoners point and laugh.

 

[Mike1]

They also never really bothered to address their weaknesses. Lack of toughness was not the reason we lost to the Bruins. Our defense hasn't been upgraded, at center we got worse or have committed to the status quo, which was a problem. David Clarkson isn't going to stop this team from getting constantly outshot or prevent them from spending way too much time in their own end.

Not very impressed with Dave Nonis at all. He completely missed the boat on what this team needed. He wasted assets/depth & cap space upgrading what really didn't need upgrading. Why add a goalie when the one you have got you the playoffs? Why give a winger a big 7-year contact when there are a plethora of wingers out there you could signed for less dollars/term & similar production? Why buy-out a center & then keep another one who has a worse offensive ceiling?

This has not been a good off-season at all...it's not like my expectations were sky high. Lord only knows, for the Leafs, they never are.

I disagree with pretty much all of this.

While its true that they haven't upgraded the defensive personnel, aside from a legitimate top pairing player, they are pretty much on par with the rest of the league in that regard. Outside of Letang, who the Leafs obviously couldn't afford, nobody of that calibre was readily available. The off season isn't over but the likelihood of the Leafs acquiring such a player this season is remote.

At centre, I believe the Leafs significantly upgraded. While Grabovski is quite a talented player, where does he fit? You Can't just look at raw numbers when building a team. It takes more than just talent. You have to find the right mix of role players and skill to balance your team. Grabovski was an outcast. He lost his job to Kadri and couldn't fill the role that management believes Bolland can. While Bozak may not have the offensive potential that Grabovski does, he FITS better in his role on that line and with the team in general. It's better for overall team chemistry.

The reason the Leafs got outshot so much last season is because they played so much without the puck. Turnover machines like Grabovski and MacArthur didn't help that cause, and they didn't produce enough offence to make those turnovers something easy to ignore. The idea of acquiring players like Clarkson and Bolland is to possess the puck more, generate turnovers and just generally spend more time in the offensive zone. After all, the easiest way to play defence is to spend the majority of the game with the puck in the other team's end.

Nonis definitely addressed the team's needs. Not all of them, mind you, but that's not really a simple task to complete. I think Nonis was bang on in identifying the Leafs' weaknesses and he did a decent job addressing them. Only time will tell whether or not he was successful.

As for Bernier, he's insurance. Reimer has to fight for his job. Scrivens never really provided a challenge. This can only serve to prove whether or not Reimer is actually a number 1 goalie.

I don't know how you can say Grabovski is a better fit than Bozak when the Leafs never really gave the guy any serious opportunities with someone like Kessel or Lupul. Even in spite of that, he still was a more productive player overall than Bozak, playing with lesser players.

Saying that Grabovski & MacArthur were hemmed in their own end all the time is just bogus. If anything Bozak was, in spite of playing with the best wingers on the team, constantly stuck in his own zone trying to get the puck out. Why do you think Carlyle put Grabovski in the defensive zone all the time last year in the first place? He knew he couldn't count on Kadri because he was young & starting Bozak there would have been a huge disaster because he was already spending way too much time in the defensive zone, even with all the offensive zone face-offs.

Nonis didn't really address any of the needs that this team had. He lowered the talent level at center which wasn't good enough. He added a goaltender when goaltending was a strength of the team already. The defense which struggled is the same. The team's depth upfront is thinner than last year with underrated guys like Frattin & Komarov not around. Not a good off-season at all.

 

[Corn Flake]

I don't know how you can say Grabovski is a better fit than Bozak when the Leafs never really gave the guy any serious opportunities with someone like Kessel or Lupul. Even in spite of that, he still was a more productive player overall than Bozak, playing with lesser players.

Grabbo played mostly with linemates he was very familiar with and had produced his best numbers with in years prior.

Not a good off-season at all.

It's July 7.

 

[TML fan]

That depends on your definition of productive. If you're looking at it from a purely statistical standpoint, sure, but as I've said before, there is more to the game than numbers and more involved in building a team than just raw talent.

Grabovski got more defensive assignmenys because that was the role he was put into, based on the fact that Kadri essentially stole his job. Cutting Grabovski loose is just the natural progression of a growing hockey team.

I never singled out MacArthur and Grabovski for getting hemmed in the defensive zone. The entire team was guilty of that. It had less to do with personnel than it did with the overall defensive philosophy of the team which I criticized pretty much the entire season.

Adding a goaltender like Bernier just strengthened their depth at that position. Adding to a strength is never a bad thing. He may have lowered the talent level in terms of raw skill but ideally, and in Nonis' mind, he actually strengthened it by acquiring a player who better fits the role that Grabovski was put into. The players currently on the roster, in my mind, have more clearly defined roles and are ideally suited to play those roles.

 

[Corn Flake]

I'm going off memory which is always dangerous but I'm fairly certain when the Leafs were bulking up in the early 00's with Corson, Roberts, Tucker, etc, they became statistically worse as well. I think the difference between the '00 team that went to the semis vs. Buffalo and the team that featured those 3 additions had a big drop in simple stats.

 

[mr grieves]

Adding a goaltender like Bernier just strengthened their depth at that position. Adding to a strength is never a bad thing. He may have lowered the talent level in terms of raw skill but ideally, and in Nonis' mind, he actually strengthened it by acquiring a player who better fits the role that Grabovski was put into. The players currently on the roster, in my mind, have more clearly defined roles and are ideally suited to play those roles.

I agree that Bolland is better suited to the role of shutdown C... But as a judgment about the roster generally? This will, for the foreseeable future, have to be followed by an asterisk denoting "except, of course, for perpetual temporary 1C Tyler Bozak, unless 'keep Kessel comfortable' is a 'clearly defined role' in a hockey team's line-up."

 

[Kin]

My issue with some of the advanced stats is that I don't think they really measure what they're trying to measure - at least, not on an individual basis. Some of them, like zone starts and QoC are useful in terms of putting a player's performance in context of how they were used by their coach, but others, like Corsi and Fenwick, suffer from a similar flaw as +/-, in that they're trying to relate a collective performance to the individual. That's somewhat mitigated by some of the deeper analysis people do with them, but, those same things can be done with +/-, though no one seems to care to do so. Things like Corsi and Fenwick seem like they'd be useful in measuring the performances of line combinations or defensive pairings or both together, but, on a player to player level, I just can't give them much weight.

This is true. One of the things that I think is the fundamental flaws with the advanced stats we're getting out of hockey right now is that they ignore a lot of the lessons we've learned from analytics in other sports.

For years, for instance, Bill James thought that raw defensive statistics in baseball were useful. Things like Range Factor which was just a simple number of plays made per game were used to make definitive statements about a player's defensive range. But eventually James noticed something pretty significant, that some first basemen who were quick and smart had lower assist totals than ones who were slow and lousy fielders. That's because it ignored a simple truth about the game, namely that on plays to first, some first basemen would be adamant that a pitcher cover the bag and some wouldn't. The raw statistics really didn't reflect anything other than a choice or a preference. There was no commonality in the aggregate, some guys did things one way, some another. Using raw statistics didn't account for the influence of other players on the diamond and was not a measure of quality.

Nowadays, nobody in the baseball community thinks much of raw defensive statistics. Things like bWAR and fWAR are all based on things like DRS or UZR which are qualitative judgments made by examining the actual plays a player makes and grading them.

The advanced numbers movements in Basketball and Football learned from that debate. They realized that you can't really just assume a similarity in the aggregate. Not everyone takes the same percentage of open jumpers, not every running back gets the same quality of blocking on a sweep and so on.

In Hockey, though, the people trying to push advanced stats have ignored that lesson entirely. Just about everything they push is "Well, yeah, clearly the team and preference factors into it but there's an assumed truth in the aggregate".

Trying to divide people who like the advanced stats in hockey and those don't as those who are willing and receptive to discussing new ideas just isn't doesn't hold water. I don't dislike the advanced statistics because I don't like fancy new numbers and love compete level. Compete level makes me roll my eyes like anyone else.  I dislike the advanced statistics being pushed in hockey because I've followed the rise of Sabrmetrics since I was a little kid and I've learned to spot good stats from bogus ones.

 

[Kin]

I don't know why it seems it has to be either/or for some.  It should be adding to the discussion of a player or a team.

You're right, it should. But the reality is that it doesn't and that's mainly on the people who like them. Because they're the ones who, typically, are taking these flawed, imperfect numbers and using them to make really definitive, discussion killing statements like "Dion Phaneuf plays against the other teams top lines because QOC" and "Mikhail Grabovski wasn't given a chance offensively because he only started 36.7% of his face-offs...."

Now, does either number have some value in our understanding of how a player plays? Maybe. Does either number have enough behind it to make either of those statements? No. I'm fine with either one adding to the discussion, I'm not fine with either one settling the discussion.

 

[TML fan]

Adding a goaltender like Bernier just strengthened their depth at that position. Adding to a strength is never a bad thing. He may have lowered the talent level in terms of raw skill but ideally, and in Nonis' mind, he actually strengthened it by acquiring a player who better fits the role that Grabovski was put into. The players currently on the roster, in my mind, have more clearly defined roles and are ideally suited to play those roles.

I agree that Bolland is better suited to the role of shutdown C... But as a judgment about the roster generally? This will, for the foreseeable future, have to be followed by an asterisk denoting "except, of course, for perpetual temporary 1C Tyler Bozak, unless 'keep Kessel comfortable' is a 'clearly defined role' in a hockey team's line-up."

If you have a 40 point centre and two 70 point wingers, or a 70 point centre and two 40 point wingers, which is better?