Blockbuster TradeBreaker VAN/OTT

[MSP4LYFE]

I also believe you need to look at 50 point players before calling Bozak miscast in a top 6. But once again, stats and rational analysis have never been your thing. FWIW 38th ranked Centre for points. Right there with Grabo, Stepan, Kesler, Hudler, Lecavalier, Gagner, Little, Richards, Roy, *Hodgson*, Laich & Ott - and only Vinny played less games. So no he's not a legit #1 C, however lots of legit #1 C only managed 60pts in 80 games last season; so yes, i'd appreciate some reasoning showing his numbers representing a miscast top 6 player.

Nick, any idiot could cite TOI and points, but the reality is that neither are very good measures of a players skill, ability or overall production in isolation.

For instance, looking at TOI in isolation one could rationally conclude that Bozak was the second most trusted player on the Maple Leafs roster, given that he clearly wasn't given as much ice as he did due to his offensive prowess. However, a closer look at QUALCOMP shows that Tyler Bozak ranked seventh among all Maple Leafs forwards with at least 60GP in 2011-12. So although Bozak played the second most minutes on the Maple Leafs (by a matter of seconds), it is evident that his minutes were sheltered and not against the top competition. Those minutes went to the Grabovski line, and Kessel and Lupul.

Furthermore, Tyler Bozak averaged 18:50 of TOI/G last year, I listed every center listed by nhl.com with at least 18:50 of TOI/G and 70 GP. Of the 29 centers that met this criteria, only 5 produced less points than Tyler Bozak (47): Mike Richards (44), Kyle Brodziak (44), Derek Roy (44), Matt Cullen (35), and Shawn Horcoff. Of the aforementioned players, only two can be classified as top 6 players, Mike Richards and Derek Roy. It is interesting to note that Richards production last year was the worst since his sophomore year, and Roy's production last year was the worst since his rookie year. Further consider that neither Roy nor Mike Richards played with linemates as talented or productive as Phil Kessel and Joffrey Lupul last season.

[Robin Hood]

I also believe you need to look at 50 point players before calling Bozak miscast in a top 6. But once again, stats and rational analysis have never been your thing. FWIW 38th ranked Centre for points. Right there with Grabo, Stepan, Kesler, Hudler, Lecavalier, Gagner, Little, Richards, Roy, *Hodgson*, Laich & Ott - and only Vinny played less games. So no he's not a legit #1 C, however lots of legit #1 C only managed 60pts in 80 games last season; so yes, i'd appreciate some reasoning showing his numbers representing a miscast top 6 player.

Nick, any idiot could cite TOI and points, but the reality is that neither are very good measures of a players skill, ability or overall production in isolation.

For instance, looking at TOI in isolation one could rationally conclude that Bozak was the second most trusted player on the Maple Leafs roster, given that he clearly wasn't given as much ice as he did due to his offensive prowess. However, a closer look at QUALCOMP shows that Tyler Bozak ranked seventh among all Maple Leafs forwards with at least 60GP in 2011-12. So although Bozak played the second most minutes on the Maple Leafs (by a matter of seconds), it is evident that his minutes were sheltered and not against the top competition. Those minutes went to the Grabovski line, and Kessel and Lupul.

Furthermore, Tyler Bozak averaged 18:50 of TOI/G last year, I listed every center listed by nhl.com with at least 18:50 of TOI/G and 70 GP. Of the 29 centers that met this criteria, only 5 produced less points than Tyler Bozak (47): Mike Richards (44), Kyle Brodziak (44), Derek Roy (44), Matt Cullen (35), and Shawn Horcoff. Of the aforementioned players, only two can be classified as top 6 players, Mike Richards and Derek Roy. It is interesting to note that Richards production last year was the worst since his sophomore year, and Roy's production last year was the worst since his rookie year. Further consider that neither Roy nor Mike Richards played with linemates as talented or productive as Phil Kessel and Joffrey Lupul last season.

Checkmate. As Steve said:

i'd trade Bozak for Hodgson 7 days a week, and if we came up with an 8th day, i'd do it on that day too.

[shooker]

maybe i am permanently high but nick,which side u on? r u a fan of bozak or not? what u mean by "(fucking terrible)"?

hahaha liked this one

[bills09]

lol @ McClement = Terrible Grinder....

[kimmer]

maybe i am permanently high but nick,which side u on? r u a fan of bozak or not? what u mean by "(fucking terrible)"?

hahaha liked this one

pretty much a case of tl;dr btwn the convo of karim/nick

Just to get it straight, nick thinks bozak is trash, and not a valid top 6, kareem thinks vice versa?

[MSP4LYFE]

maybe i am permanently high but nick,which side u on? r u a fan of bozak or not? what u mean by "(fucking terrible)"?

hahaha liked this one

pretty much a case of tl;dr btwn the convo of karim/nick

Just to get it straight, nick thinks bozak is trash, and not a valid top 6, kareem thinks vice versa?

Nick thinks Bozak is a legitimate top 6 forward, presumably on the second line, I on the other hand think Bozak is miscast as a top 6 forward. and is best suited for a third line role. We both agree that Bozak is a better player than Jay McClement.

[The BBKL Insider]

I agree with kareems statement above.

Bozak is a nice player to have; he has solid vision, solid speed, decent on the dot, he's fairly talented. However I think he's a 3rd line C on a good team.

[KapG]

bozak is definitely a tad underrated I think.

He may not be a legit 1st line centre, but he can fill in that spot fine for the time being. Ideally I see him as a 2nd/3rd line C like you guys do, I still love his skill set and his faceoff ability though. I'd still trade him for Tarasenko 100 times out of 100 '_'

[Nick]

QUALCOMP is a meaningless stat. You'll find me saying that as far back as 2007/08 on HF Boards. But thankfully a statsitically analysis also shows this: http://hockeyanalysis.com/2012/01/25/im ... teammates/

There is very little variation in quality of opposition, but significant variation in quality of teammate.

If you have a great offensive player, the theory is your opponents will want to match up their great defensive players against him. But, at the same time you are trying to match up your great offensive player against their weakest defensive players. When at home, you get the line matching advantage, while on the road your opponent does. When all is said and done everything more or less evens out.

Bozak benefits from strong quality of teammates, but even that you'll see that effects the margins, and not the grand scheme of things. Unless as your previous statement indicates, you don't think he played with Kessel and Lupul, in which case his quality of teammate would adjust downwards if anything.

So although Bozak played the second most minutes on the Maple Leafs (by a matter of seconds), it is evident that his minutes were sheltered and not against the top competition. Those minutes went to the Grabovski line, and Kessel and Lupul.

It wasn't a matter of seconds... it was over 70 minutes more, and in less games. in terms of toi/GP, yes him and Lupul were very close 18:50 vs 18:36 - Lupul played no PK, Bozak played a little. Lupul played major PP, Bozak played reasonable ( 40sec/gp more then Grabo) - nearly identical ES TOI/GP with Lupul which suggests they were line-mates, most of the time (so how is he not facing the same opponents?), and this was supported by what I saw when watching the games.

Saying Kessel, Grabo, & Lupul played against the other teams top offensive players doesn't make sense - Grabo didn't play with Phil and Joff - Bozak did.

and not that it matters, but 18:50 is an odd time to chose, as it's not #1 C toi, more in line with #2 toi (17-19min), also worth nothing in Bozak's 73 GP he produced 47, yes guys that played 82 and 18:50 produced more - comfortably within the margins of a conservative scale for Bozak. Not to mention Little whom you omitted.

To be clear, I'm not calling him a #1 centre, I'm saying he did well and fits (numbers supported) in a top 6 TOI role, at the centre position. That was my opinion when watching him, and is supported by the hard-unadjusted #, as well as a more indepth comparison across other top 6 C. No he doesn't content with the top 20 C in the NHL, but neither does his TOI.

anyways, clearly a agree to disagree stage (something we did before it came here), and time will tell. Shiv's attempted Troll's have been hilarious, weak and petty as of late, I hope you have more in the ammo box for when our team's face Shiv.

[shooker]

If you have a great offensive player, the theory is your opponents will want to match up their great defensive players against him. But, at the same time you are trying to match up your great offensive player against their weakest defensive players. When at home, you get the line matching advantage, while on the road your opponent does. When all is said and done everything more or less evens out.

I don't get this exactly... how does it not matter who you play against? They both seem very relevant to me. In fact it seems to me that an award such as a selke usually goes to a player while taking into consideration who he plays against, how well he shuts them down and how many points he puts up against harder opposition. I am no statistician but to me this seems like something rather important when evaluating a player. You talk about earned ice time, players earn that not only for scoring but for not allowing goals to be scored with the opposition on the ice. I understand that over a year it should average out that you face the same difficultly of player throught the year, as in 1st lines play checking players yada yada yada, but that doesnt change the fact that better players face tougher opposition night in and night out than lesser players. I just dont see how you can totally dismiss Qualcomp, seems like an opinion rather than fact to me.

not trying to be controversial I just think dismissing it is rather bias.

[Nick]

So you're the coach for the leafs, you want to get Kessel out away from the matchup as much as possible, but you're also trying to get a certain line out against their top line.

so your trying for:
Your top line versus their 2nd (neither offensive or defensively dominant)
Your 2nd line versus their 3rd line (their shutdown line)
and your third line versus their 1st (offensive)

their Coach is trying for the opposite.

On a statistical breakdown over the entire season, on a player to player or line to line basis, there is virtually no effect from quality of competition,

[shooker]

So you're the coach for the leafs, you want to get Kessel out away from the matchup as much as possible, but you're also trying to get a certain line out against their top line.

so your trying for:
Your top line versus their 2nd (neither offensive or defensively dominant)
Your 2nd line versus their 3rd line (their shutdown line)
and your third line versus their 1st (offensive)

their Coach is trying for the opposite.

On a statistical breakdown over the entire season, on a player to player or line to line basis, there is virtually no effect from quality of competition,

but there are teams like Van and LAK who want their 2nd line out vs the opposition 1st lines or teams like Det who want their 1st line vs the other teams. So many teams differ in tactics because of the hand they were dealt team wise that I don't think you can say this all averages out. Let's say the Canucks play the Ducks in ANA. Sure Kelser may not get out to face Getz, Perry + 100% of the shifts, but I am betting he gets out there 70+ % to me that is still a large portion of the time against a tougher opposition. The number might even be higher on how many shifts they get against them, I constantly see stats like 18 out of 22 shifts vs such and such line.

I dont know. I would need to see stats like how much time did Kelser play vs opposition + where he started his shift + quality of linemates etc before I can dismiss any of the Stat based theories.

[MSP4LYFE]

QUALCOMP is a meaningless stat. You'll find me saying that as far back as 2007/08 on HF Boards. But thankfully a statsitically analysis also shows this: http://hockeyanalysis.com/2012/01/25/im ... teammates/

Disagreeing with a statistic, and concluding that it is meaningless are two very different things that should not be confused with one another. QUALCOMP does not vary a great deal from player to player, but it still has it's usefulness, and you would be hard pressed to argue that Bozak's low QUALCOMP is a result of Carlyle/Wilson trying to get him out against weaker lines. Particularly when Kessel and Lupul have considerably higher QUALCOMP ratings.

It wasn't a matter of seconds... it was over 70 minutes more, and in less games. in terms of toi/GP, yes him and Lupul were very close 18:50 vs 18:36 - Lupul played no PK, Bozak played a little. Lupul played major PP, Bozak played reasonable ( 40sec/gp more then Grabo) - nearly identical ES TOI/GP with Lupul which suggests they were line-mates, most of the time (so how is he not facing the same opponents?), and this was supported by what I saw when watching the games.

It is a matter of seconds in TOI/G, I'm not sure why you would reference Total TOI, and no Bozak did not play less games than Joffrey Lupul, he played seven more. Eight if you consider that Lupul went down early against the Bruins.

I should also add that Bozak's ice time rose considerably in the dying weeks of the season, either due to small sample size variance, or as a result of the injuries to Grabovski and Lupul.

Saying Kessel, Grabo, & Lupul played against the other teams top offensive players doesn't make sense - Grabo didn't play with Phil and Joff - Bozak did.

I clearly stated "the Grabovski line" emphasis on line.

and not that it matters, but 18:50 is an odd time to chose, as it's not #1 C toi, more in line with #2 toi (17-19min), also worth nothing in Bozak's 73 GP he produced 47, yes guys that played 82 and 18:50 produced more - comfortably within the margins of a conservative scale for Bozak. Not to mention Little whom you omitted.

Then you missed the point of the exercise, my goal was to prove that Bozak's production does not warrant his ice time, or anything close to it. A comparison to most second line centers would have been pointless given that Bozak plays considerably more than most second line centers, and with better quality of competition.

To be clear, I'm not calling him a #1 centre, I'm saying he did well and fits (numbers supported) in a top 6 TOI role, at the centre position. That was my opinion when watching him, and is supported by the hard-unadjusted #, as well as a more indepth comparison across other top 6 C. No he doesn't content with the top 20 C in the NHL, but neither does his TOI.

If Bozak is a #2 center, he is a mediocre one, his production does not compare with the good to great #2's in the game. Including Grabovski, which is why he profiles as a #3 on a good team.

As for the bolded portion of your post, that is far from fact. The only thing you have supported with your numbers is that Bozak played alot of minutes, and produced mediocre second line numbers alongside two top 10 scorers in ppg.

anyways, clearly a agree to disagree stage (something we did before it came here), and time will tell. Shiv's attempted Troll's have been hilarious, weak and petty as of late, I hope you have more in the ammo box for when our team's face Shiv.

Fair enough.

[Nick]

Disagreeing with a statistic, and concluding that it is meaningless are two very different things that should not be confused with one another.

Finding the construct validity of a measure too low, aka useless, does in fact lead to the non-conclusion in regards to that statistic. I'm fairly well grounded in this statement. Qualcomp is basically even across the entire NHL - especially over a 30+ game sample.

[MSP4LYFE]

That blog that you posted does not conclude that QUALCOMP is "basically even across the NHL." That is a gross exaggeration. Anyways, keep clinging onto that argument, it's the only one that holds any amount of weight.

[Nick]

That blog that you posted does not conclude that QUALCOMP is "basically even across the NHL." That is a gross exaggeration.

With these advanced stats we're changing from a straight numbers comparison and instead we are grading on a curve – judging players not by their absolute performance but by their performance relative to what we expect from the average player put in those situations. The most common things to look at when evaluating a player’s usage are the quality of his teammates, the quality of his competition, and how often he started in the offensive or defensive zone. There is a logical foundation which leads us to believe factors have an impact on the player’s outcomes, and as statistical analysts we seek to quantify that impact so we can correct for it when evaluating performance.

[Nick]

I don't mean to say that quality of opponents isn't a very important measure - instead that the statistic QUALCOMP, is useless. It has no variance and as such can predict nothing outside of it's margin for error.

The difference between being 95th percentile in QUALCOMP and 13th percentile is a scarcely noticeable shift in their ice time - it measures nothing.

[kimmer]

Man I take a nice saturday quality nap from like 1pm til now - and u niggas still debatin this shit

[MSP4LYFE]

That is great, but your comment above is still a gross exaggeration, and when citing QUALCOMP it is useless to compare players on different teams (I.E. the average player put in those situations), most advanced stats still aren't refined enough to compare across different teams, or even leagues. Which is why I used those numbers for the Leafs and the Leafs only.

Anyways, I'm done with this argument. If you want to cling to QUALCOMP, that is your prerogative, there are far more justifications listed above.

[MSP4LYFE]

I don't mean to say that quality of opponents isn't a very important measure - instead that the statistic QUALCOMP, is useless. It has no variance and as such can predict nothing outside of it's margin for error.

The difference between being 95th percentile in QUALCOMP and 13th percentile is a scarcely noticeable shift in their ice time - it measures nothing.

That is only meaningful if we were comparing two different players on two different teams. This is the real last post on the subject!