Wanna blog? Start your own hockey blog with My HockeyBuzz. Register for free today!
 
Forums :: Blog World :: Sean Maloughney: G68: Leafs @ Oilers - Revenge
Author Message
Oilers4Life14
Detroit Red Wings
Location: ON
Joined: 03.06.2013

Mar 9 @ 10:26 PM ET
A - Hall
B - Stone
C - Gaudreau
D - McDavid
E - Kucherov
F - Johansen
G - Tarasenko
H - Wheeler
I - Ovi

I see your arguement, however, you have one fatal flaw from my perspective. Why relGF% and not rel. GF/60? RelGF% can be exaggerated by sample size and anomolies. Rel GF/60 on the other hand amounts for this to an extent (from my understanding). Overall, I just like how GF/60 rel. standardizes the data set.

In fact, if you look at the rel. GF% data that represents Kucherov and compare it to his forward teammates, Danick Martel (only 78 minutes of hockey played w/TBL) leads this category with a whooping 36.59 GF% rel... which indirectly may represent how an outlier highly influences this stat. Meanwhile, Nikita Kucherov is in the middle of the pack (compared to his teammates) with respect to that stat.

Furthermore, if you switch to rel. GF/60, Ovi (leads), Kucherov (leads), McDavid (leads), Gaudreau (2nd), Johansen (4th), Tarasenko (4th), Hall (1st), Stone (Leads with Ottawa, but more on this later) and Wheeler (2nd) all lead the team or are top 4 in that stat department, not to mention everyone else near the top can be considered the team's best players (subjectively), thus proving its viability and accuracy with respect to predicting or correlating who is the best in terms of offensive contribution.

Meanwhile, relGF% has all kinds of anomalies causing nuisances with visualizing the team in that specific department or adding onto the bigger picture. TBL has Danick Martel leading the way, yet he is last in rel GF/60. Nashville's relGF% is a joke of a stat with the range being from 56.92 to -100, with Johansen at -9.49. All of the team's best players (subjectively), including Forsberg, Johansen, Turris, Arvidsson, and others are ranked near the bottom and have negative relGF% values. Winnipeg's players relGF% is also quite conflicting with respect to where your top players are. Wheeler sits in the middle of the pack (while is 2nd in rel GF/60), while Copp and Appleton lead the way. Issues are also evident in the other teams of the players you mentioned when it comes to relGF%. By utilizing rel GF/60 instead, there is far less variation in terms of data sets along with a far more accurate portrayal of the elite as actually elite with respect to who you would expect.

Lastly, none of the players you mentioned have a negative rel GF/60 value. But do you know who does with his new team after 5 games? Mark Stone. -0.23. Ranked 12/19 in that department. Lowest value and rank out of all of them when it comes to their current rel GF/60 (rel GF/60 value with their current team), despite his 28.57 relGF% which leads the Knights. Just something interesting to consider.

*Parameters are even-stength, forwards only, and the source is natural stat trick.

- EdmHockeyMan


Both of you are (frank)ing nerrrrrrrds holy poop
EdmHockeyMan
Referee
Location: Lumbridge, AB
Joined: 06.24.2013

Mar 9 @ 10:29 PM ET
Both of you are (frank)ing nerrrrrrrds holy poop
- Oilers4Life14




RatedR80
Edmonton Oilers
Joined: 10.17.2013

Mar 9 @ 10:37 PM ET
Nucks are getting slapped right now....Stone with his 1st as a Knight
MaximumBone
Edmonton Oilers
Joined: 06.15.2012

Mar 9 @ 11:12 PM ET
A - Hall
B - Stone
C - Gaudreau
D - McDavid
E - Kucherov
F - Johansen
G - Tarasenko
H - Wheeler
I - Ovi

I see your arguement, however, you have one fatal flaw from my perspective. Why relGF% and not rel. GF/60? RelGF% can be exaggerated by sample size and anomolies. Rel GF/60 on the other hand amounts for this to an extent (from my understanding). Overall, I just like how GF/60 rel. standardizes the data set.

- EdmHockeyMan

I mainly chose to use relGF% as it captures a snapshot idea as to the overall impact the player has in his time on the ice. Meanwhile, relGF/60 is a more limited expression of the player's offensive impact. In evaluating the value of Stone as an elite player Vs Kucherov, I think taking his defensive impacts into account as well given that they're a big part of his appeal as a player.

In fact, if you look at the rel. GF% data that represents Kucherov and compare it to his forward teammates, Danick Martel (only 78 minutes of hockey played w/TBL) leads this category with a whooping 36.59 GF% rel... which indirectly may represent how an outlier highly influences this stat. Meanwhile, Nikita Kucherov is in the middle of the pack (compared to his teammates) with respect to that stat.
- EdmHockeyMan

If you look at relGF% strictly without context then yeah, you'll fall victim to these such "anomalies". As I note in the original post, I'm aware that relGF% has a tendency to penalize teams of extreme depth (Kucherov and Johansen) and paint players on teams with a dearth of talent (McDavid, Hall and Stone) but that's precisely why I used it.

I'm not intending to disparage Kuch or Johansen in posting those numbers; it's just detailing that their team as a whole is posting comparable or better numbers without them so the position that they're driving the results has to be questioned through another lens or two.

Furthermore, if you switch to rel. GF/60, Ovi (leads), Kucherov (leads), McDavid (leads), Gaudreau (2nd), Johansen (4th), Tarasenko (4th), Hall (1st), Stone (Leads with Ottawa, but more on this later) and Wheeler (2nd) all lead the team or are top 4 in that stat department, not to mention everyone else near the top can be considered the team's best players (subjectively), thus proving its viability and accuracy with respect to predicting or correlating who is the best in terms of offensive contribution.
- EdmHockeyMan

But we're not trying to prove that each player leads their team; we're both already aware of that reality. I chose each of them because they were objectively one of the main drivers of their teams (except Johansen, but more on that later). Precisely my point is that, because Stone not only statistically fits in with this group across a wide array of metrics but exceeds many of his cohorts, he is completely worth the contract he was just given. If you'll recall, it was your original post that called that into question.

Meanwhile, relGF% has all kinds of anomalies causing nuisances with visualizing the team in that specific department. TBL has Danick Martel leading the way, yet he is last in rel GF/60. Nashville's relGF% is a joke a stat with the range being from 56.92 to -100, with Johansen at -9.49. All of the team's best players (subjectively), including Forsberg, Johansen, Turris, Arvidsson, and others are ranked near the bottom and have negative relGF% values. Winnipeg's players relGF% is also quite conflicting with respect to where your top players are. Wheeler sits in the middle of the pack (while is 2nd in rel GF/60), while Copp and Appleton lead the way. By utilizing rel GF/60 instead, there is far less variation in terms of data sets along with a far more accurate portrayal of the elite as actually elite with respect to who you would expect
- EdmHockeyMan

And this precisely the point I mentioned earlier regarding remembering context in the evaluation. No stat is perfect or ideal in isolation. You've gotta ask yourself why does player A exceed player B in this stat? Situational context? Quality of linemates? Deployment? This is precisely what I was starting to hint at with respect to linemates. Stone's linemates were the two rookies while Kuch's main linemates were a proven ~50 pt player and a fellow ascending elite talent in Point.

How about deployment? Stone's line was tasked with a significantly greater proportion of defensive starts as one of his many talents involves driving the play in a positive direction while Kuch's line had some favourable deployment, as it should (can quote numbers if we care to go further). This isn't to take away from Kuch so much as to further underscore the impact Stone has on his team. To go back to the stat I changed, his actual stat line had him at an absurd +18.1% relGF (4.23 GF/60 and 3.31 GA/60) while playing primarily with two rookies (Tkachuk and White). Skilled as those rookies may be, they're not even in the same stratosphere as Johnson and Point.

Lastly, none of the players you mentioned have a negative rel GF/60 value. But do you know who does with his new team after 5 games? Mark Stone. -0.23. Ranked 12/19 in that department. Lowest value and rank out of all of them when it comes to their current relGF%, despite his 28.57 relGF% which leads the Knights. Just something interesting to consider.

*Parameters are even-stength and the source is natural stat trick.

- EdmHockeyMan

Way to cite a tiny 5 game sample size in an effort to support your point. Here's mine:

He's been on-ice for another goal tonight (two if he hadn't changed just before the Stastny goal) making his GF/60 3.48 while his team's will be seeing another great increase thanks to those 4 other 1st period 5-on-5 goals. Meanwhile, he has STILL only been on the ice for 1 goal against his 86mins of ice time which means his GA/60 is a whopping... 0.7 and his relGA/60 is -1.27 meaning their team GA/60 with him off the ice has been 1.97. Elite. #worth9point5
EdmHockeyMan
Referee
Location: Lumbridge, AB
Joined: 06.24.2013

Mar 9 @ 11:12 PM ET
Nucks are getting slapped right now....Stone with his 1st as a Knight
- RatedR80


It's looking ugly for them.
oil90
Edmonton Oilers
Location: ON
Joined: 12.05.2010

Mar 9 @ 11:12 PM ET
Why do they always play 20mins or less of every game. If only this team could play full 60.
MaximumBone
Edmonton Oilers
Joined: 06.15.2012

Mar 9 @ 11:13 PM ET
Both of you are (frank)ing nerrrrrrrds holy poop
- Oilers4Life14

Not wrong. Also not bothered, so deal with it, cuckmuffin
MaximumBone
Edmonton Oilers
Joined: 06.15.2012

Mar 9 @ 11:14 PM ET
Why do they always play 20mins or less of every game. If only this team could play full 60.
- oil90

I'd say they were playing plenty fine. They were just clearly outskilled by a dominant offensive team and were let down by an ill-timed penalty (that wouldn't get called in a West V West game) and the subsequent goal against.
RatedR80
Edmonton Oilers
Joined: 10.17.2013

Mar 9 @ 11:23 PM ET
It's looking ugly for them.
- EdmHockeyMan


The Flames better finish 1st or they are (frank)ed if they play Vegas
RatedR80
Edmonton Oilers
Joined: 10.17.2013

Mar 9 @ 11:25 PM ET
I'd say they were playing plenty fine. They were just clearly outskilled by a dominant offensive team and were let down by an ill-timed penalty (that wouldn't get called in a West V West game) and the subsequent goal against.
- MaximumBone


I thought 11-3-1 was being optimistic but from some of the reactions I have been seeing I was too generous....stupid Oilers should have been 15-0 to finish the season
RatedR80
Edmonton Oilers
Joined: 10.17.2013

Mar 9 @ 11:26 PM ET
The Condors destroyed San Jose 7–1.....wish I didn’t miss that one
MaximumBone
Edmonton Oilers
Joined: 06.15.2012

Mar 9 @ 11:31 PM ET
The Flames better finish 1st or they are (frank)ed if they play Vegas
- RatedR80

That's one impressive set of forwards that Vegas managed to piece together in less than two seasons of work. A bunch of good, "under-the-radar" additions in JAM, Smith, Tuch and even Stastny. People were weirdly soft on Stastny last offseason but he's still one of the most effective possession Cs in the league. Despite his play with the Jets, people seemed to sleep on him. It's not often a high-end top-6 C like him reaches UFA.

That Pacioretty-Stastny-Stone line has all the makings of an elite postseason line that I'll enjoy following closely to see if they can dominate like they seem built to.
MaximumBone
Edmonton Oilers
Joined: 06.15.2012

Mar 9 @ 11:34 PM ET
The Condors destroyed San Jose 7–1.....wish I didn’t miss that one
- RatedR80

Looks like Yamo got injured last game or something. Didn't see that until just now. Wonder what happened.
RatedR80
Edmonton Oilers
Joined: 10.17.2013

Mar 9 @ 11:38 PM ET
That's one impressive set of forwards that Vegas managed to piece together in less than two seasons of work. A bunch of good, "under-the-radar" additions in JAM, Smith, Tuch and even Stastny. People were weirdly soft on Stastny last offseason but he's still one of the most effective possession Cs in the league. Despite his play with the Jets, people seemed to sleep on him. It's not often a high-end top-6 C like him reaches UFA.

That Pacioretty-Stastny-Stone line has all the makings of an elite postseason line that I'll enjoy following closely to see if they can dominate like they seem built to.

- MaximumBone


They hit a bunch of homeruns since day one and they will be set for years because of it....this team is going nowhere and I wouldn’t be surprised if they keep finding ways to get even better....you can bet Seattle is taking lots of notes on this team
EdmHockeyMan
Referee
Location: Lumbridge, AB
Joined: 06.24.2013

Mar 9 @ 11:39 PM ET


He's been on-ice for another goal tonight (two if he hadn't changed just before the Stastny goal) making his GF/60 3.48 while his team's will be seeing another great increase thanks to those 4 other 1st period 5-on-5 goals. Meanwhile, he has STILL only been on the ice for 1 goal against his 86mins of ice time which means his GA/60 is a whopping... 0.7 and his relGA/60 is -1.27 meaning their team GA/60 with him off the ice has been 1.97. Elite. #worth9point5

- MaximumBone


I'll yield with some consideration and remarks.

Firstly, I did some editing on the initial post in order to prove more clarity in what I was trying to say (my ideas were not changed, just how they were presented) and to fix up some mistakes.

I still think you should have included rel GF/60 into the mix just to provide some form of standardization when it comes to relative metrics and rel. GF/60 does just that. Not to mention it provides another lens to see whether or not a specific player is driving results. Sure it may represent a limited offensive view, but it does provide a good perspective to consider and analyze along with the bigger picture.

As such, I'm not trying to isolate specific data sets, but rather to critique your choice of stats. I understand that you cannot abide by a single stat without looking at the context, but I just thought you could have drawn up a better picture by including a more standardized statistic.

As for that 5 game sample , I had doubts on putting that in there. For one, it did show how relGF% can be bloated while rel GF/60 has some parameters of standardization, but then I thought, it's from a 5 game sample, which kind of goes against what I'm trying to say.

Overall, you bring up good points and as such, I'll be reconsidering my opinion and analysis of the deal.
EdmHockeyMan
Referee
Location: Lumbridge, AB
Joined: 06.24.2013

Mar 9 @ 11:41 PM ET
The Flames better finish 1st or they are (frank)ed if they play Vegas
- RatedR80


Pretty much.
MaximumBone
Edmonton Oilers
Joined: 06.15.2012

Mar 9 @ 11:41 PM ET
Lagesson with 13 points (3G and 10A), 19 shots and +14 in his last 16 games to go along with his usual excellent defensive play. I'm really liking what I'm seeing from him; might have a perfect fit for our bottom pairing in the NEAR future.
EdmHockeyMan
Referee
Location: Lumbridge, AB
Joined: 06.24.2013

Mar 9 @ 11:46 PM ET
Lagesson with 13 points (3G and 10A), 19 shots and +14 in his last 16 games to go along with his usual excellent defensive play. I'm really liking what I'm seeing from him; might have a perfect fit for our bottom pairing in the NEAR future.
- MaximumBone


Not to mention Benson has been doing really well for his first full season in the AHL. So far, lots of good news when it comes to this team's future.
RatedR80
Edmonton Oilers
Joined: 10.17.2013

Mar 9 @ 11:55 PM ET
Lagesson with 13 points (3G and 10A), 19 shots and +14 in his last 16 games to go along with his usual excellent defensive play. I'm really liking what I'm seeing from him; might have a perfect fit for our bottom pairing in the NEAR future.
- MaximumBone


I would prefer year after next but I guess that depends on what happens in the off season
MaximumBone
Edmonton Oilers
Joined: 06.15.2012

Mar 9 @ 11:55 PM ET
I'll yield with some consideration and remarks.

Firstly, I did some editing on the initial post in order to prove more clarity in what I was trying to say (my ideas were not changed, just how they were presented) and to fix up some mistakes.

I still think you should have included rel GF/60 into the mix just to provide some form of standardization when it comes to relative metrics and rel. GF/60 does just that. Not to mention it provides another lens to see whether or not a specific player is driving results. Sure it may represent a limited offensive view, but it does provide a good perspective to consider and analyze along with the bigger picture.

As such, I'm not trying to isolate specific data sets, but rather to critique your choice of stats. I understand that you cannot abide by a single stat without looking at the context, but I just thought you could have drawn up a better picture by including a more standardized statistic.

As for that 5 game sample , I was doubts on putting that in there. For one, it did show how relGF% can be bloated, while rel GF/60 has some parameters of standardization, but then I thought, it's from a 5 game sample, which kind of goes against what I trying to say.

Overall, you bring up good points and I'll reconsider my opinion and analysis of the deal.

- EdmHockeyMan

I don't think I fully understand what you mean by standardizing.

The reason I chose to use a mix of relGF% and raw GF/60 was to paint a picture of both the players' straight up scoring prowess as well as framing it within the context of the team they played on.

Had I just done GF/60 and relGF/60, I would've missed out on defensive impact (a seminal point in my argument for his value) and, in other analyses, that combination tends to skew the data for players on super lines where team scoring depth is limited (Colorado, Edmonton, New Jersey and kinda Boston).

Had I just had relGF/60 and relGF%, I wouldn't have an objective metric by which to compare the players' actual on-ice production which was another point to my argument of Stone's value. Despite all the adversity he has faced in his linemates, deployment, team quality and the likes, he still manages to produce 4.23 GF/60.

Could I have included all three? Sure and I had planned to do just that in the next post detailing deployment, usage and linemates in greater detail but I preferred to keep the stats sections smaller and more concise (also not taking up more than one line).
RatedR80
Edmonton Oilers
Joined: 10.17.2013

Mar 9 @ 11:57 PM ET
Not to mention Benson has been doing really well for his first full season in the AHL. So far, lots of good news when it comes to this team's future.
- EdmHockeyMan


I don’t want to be a downer because I love the Condors but I see more of a group of guys that might excel at bottom 6 or bottom pairing D (minus Jones I think he is top 4 material).....Benson could be a good top 6 player along with Yamamoto but who knows at this point
MaximumBone
Edmonton Oilers
Joined: 06.15.2012

Mar 10 @ 12:04 AM ET
I don’t want to be a downer because I love the Condors but I see more of a group of guys that might excel at bottom 6 or bottom pairing D (minus Jones I think he is top 4 material).....Benson could be a good top 6 player along with Yamamoto but who knows at this point
- RatedR80

Which is precisely one of the things we desperately need. A productive bottom-6 alleviates some of the burden of scoring from the top-6 and helps drive teams to success.

If we're talking long-term impact players, I see Jones, Bouchard and one of Yamamoto or Benson achieving such out of the players currently in the system with an outside shot to Maksimov and Samorukov. The rest seem well-positioned to fill out our depth admirably and hopefully save us from the pit of mediocrity that is overpaying depth UFAs.

Russell and Gambardella are safe bets to compete for a bottom-6 spot next year (could probably do it now) while Benson and Marody should probably see another start in Bako unless they take their skating to a new level over the summer. Jones and Lagesson could likely do well in a bottom pairing role next year and, if it weren't for our LHD depth, they might already.

I'll just say, don't be surprised if this Bakersfield team graduates 3 or 4 players to start next year and another 1 or 2 by season's end.
MaximumBone
Edmonton Oilers
Joined: 06.15.2012

Mar 10 @ 12:10 AM ET
I would prefer year after next but I guess that depends on what happens in the off season
- RatedR80

Russell trade/buyout, Sekera can stay as he's proven to be his usual effective self, bring in a RD (Gudas) and then bring up one of the two. Draft Byram.

Klefbom- Larsson
Nurse- Gudas
Sekera- Benning
Gravel, Lagesson

Klefbom- Larsson
Sekera- Barrie (Nurse for Barrie?)
Lagesson- Benning
Gravel

Jones*, Bouchard, Byram, Bear, Samorukov, Berglund and Kemp in the system.

* only because I prefer him dominating offensively w/ Bouchard to floating between 6 & 7 D
Oilers4Life14
Detroit Red Wings
Location: ON
Joined: 03.06.2013

Mar 10 @ 12:11 AM ET
Which is precisely one of the things we desperately need. A productive bottom-6 alleviates some of the burden of scoring from the top-6 and helps drive teams to success.

If we're talking long-term impact players, I see Jones, Bouchard and one of Yamamoto or Benson achieving such out of the players currently in the system with an outside shot to Maksimov and Samorukov. The rest seem well-positioned to fill out our depth admirably and hopefully save us from the pit of mediocrity that is overpaying depth UFAs.

Russell and Gambardella are safe bets to compete for a bottom-6 spot next year (could probably do it now) while Benson and Marody should probably see another start in Bako unless they take their skating to a new level over the summer. Jones and Lagesson could likely do well in a bottom pairing role next year and, if it weren't for our LHD depth, they might already.

I'll just say, don't be surprised if this Bakersfield team graduates 3 or 4 players to start next year and another 1 or 2 by season's end.

- MaximumBone


No Bear?
EdmHockeyMan
Referee
Location: Lumbridge, AB
Joined: 06.24.2013

Mar 10 @ 12:17 AM ET
I don't think I fully understand what you mean by standardizing.

The reason I chose to use a mix of relGF% and raw GF/60 was to paint a picture of both the players' straight up scoring prowess as well as framing it within the context of the team they played on.

Had I just done GF/60 and relGF/60, I would've missed out on defensive impact (a seminal point in my argument for his value) and, in other analyses, that combination tends to skew the data for players on super lines where team scoring depth is limited (Colorado, Edmonton, New Jersey and kinda Boston).

Had I just had relGF/60 and relGF%, I wouldn't have an objective metric by which to compare the players' actual on-ice production which was another point to my argument of Stone's value. Despite all the adversity he has faced in his linemates, deployment, team quality and the likes, he still manages to produce 4.23 GF/60.

Could I have included all three? Sure and I had planned to do just that in the next post detailing deployment, usage and linemates in greater detail but I preferred to keep the stats sections smaller and more concise (also not taking up more than one line).

- MaximumBone


Standardization as in account for anomalies and outliers, and minimize their impact in the grand scheme of things.

If you had used all three, that would have been far better in terms of analysis, but I understand your constraints and perspective.
Page: Previous  1, 2, 3, 4, 5  Next