Boys of Summer

Based on previous behaviour and current roster holes, we will see some activity by the Edmonton Oilers over the summer. The signing of Ryan Strome yesterday gets us closer, with Darnell Nurse the last man standing from last season’s roster. There is plenty of room on the 50-man and many rumours that the game is afoot. Is it? And if so, how many balls are in the air? Is Milan Lucic still in play? Let’s have a look at the possibilities.

THE ATHLETIC!

Give The Athletic as a gift or get it yourself and join the fun! Offer is here, less than $5 a month! I find myself reading both the hockey (Willis, Dellow, Pronman, et cetera) and the baseball coverage a lot, it’s a pure pleasure to visit. We’ll sell you the whole seat, but you’ll only need the edge.

Colborne is big (6.05, 221) and 28, with a lot of NHL experience (295 games). Per 82gp, he has scored 12-20-32 in the world’s best league. He can play a rugged, skill game but he is inconsistent and has had injury issues. I think there’s an excellent chance he will be in Oilers camp this fall. Folks, this is a Chiarelli type player.

PROJECTED ROSTER 2018-19

Oilers don’t have room to go long on Nurse and can’t sign Patrick Maroon, so we may be getting a close facsimile of the opening night roster. The Milan Lucic chatter won’t go away, more in a minute. Folks, this is us. I wrote about value contracts in the latest item for The Athletic, searching for value contracts. Edmonton needs six, please and thanks.

Strome’s scoring numbers were down slightly year over year, the 5-on-5 shooting percentage contributed. Better in possession (he was 50.28 Corsi for 5-on-5 without Connor McDavid) and there were some interesting little bits about his numbers (scored over 2.00/60 5-on-5 with McDavid, Leon, Puljujarvi and 1.82 with Jujhar Khaira). I thought he was gone for sure mid-season, but he came on and found a role later in the year.

AVAILABLE FREE AGENTS

It’s amazing how many players are still available. We’ve gone from black and white to shades of gray, meaning all available players range in price from $1 million to a training camp invite, but there are several players worth mentioning at this point.

  • L Lance Bouma. He’s a fringe NHL player but intangibles get mentioned when his name comes up. There isn’t an obvious statistical reason to retain his services.
  • L Mike Cammalleri. The Oilers know him, he did contribute after coming over via trade. A long shot for sure.
  • R Alex Chiasson. I thought he had signed but CapFriendly has him as still available. Gigantic winger with some skill and a lot of grit.
  • L Jason Chimera. Probably an invite, he is 40 and played seven minutes a night last season but guys who can skate (and he still can) often have long coda’s. He might have another season in him.
  • LC Joe Colborne. Mentioned above, he might be an invite, Peter Chiarelli was general manager in Boston when the Bruins drafted him. If he’s healthy, Colborne could see NHL time.
  • LD Brandon Davidson. He struggled last season and the injuries are piling up but as a low risk investment or invite Davidson has all kinds of appeal.
  • RD Dylan DeMolo. Remains unsigned, he is absolutely worth an NHL contract.
  • RD Cody Franson. Math still loves him and that’s good enough for me. His speed means he may not get a contract but might land an invite.
  • LC Derek Grant. Effective in a very specific role, the Oilers could use more depth at center but Grant should easily get an attractive offer that sees him in the everyday lineup.
  • R Jannik Hansen. I’d completely forgotten about him but he’s a fine player who should have some track left. Bob mentioned him yesterday.
  • L Scott Hartnell. He’s aging now and hasn’t been mentioned a lot but did score 13 goals while playing 12 minutes a night last season.
  • R Ales Hemsky. He’s 34, has been hurt a lot, but I would be happy to see him get an invite to training camp. One of my all-time favourite Oilers.
  • L Tomas Jurco. Good size, good speed, he’s been pushing for regular duty in the NHL for several seasons.
  • L Patrick Maroon. He still hasn’t signed, if I were a suspicious sort this blog would drill down on the idea Mr. Maroon is waiting for the Oilers to find cap room. He has chem with 97, it would allow Nuge to move down to the Draisaitl line. It probably won’t happen.
  • L Benoit Pouliot. A team looking for a player who can help them will check out Pouliot. He forechecking alone is worth having him on the roster.
  • RC Nick Shore. An attractive set of skills, righty center who can help out in multiple areas. Why is he still available?
  • L Scottie Upshall. His name has been mentioned a few times in connection with the Oilers over the last several days. Maybe he’s the one.
  • R Tommy Wingels. He’s a utility forward but can help in multiple areas. How much is left?

The 50-man is at 44, I think Evan Bouchard signs and makes the big club. That’s 45, and the Oilers will want to start the season shy of 50. Let’s say there are three more spots. From this list, I choose Dylan DeMolo, Nick Shore and Benoit Pouliot. My guess is the Oilers choose Joe Colborne.

GRAVEL ROAD

I haven’t spent much time on Kevin Gravel so far, but promise to drill down on the new Oilers defenseman in the coming days. An interesting article worth your time is this item from Paul Gazzola. Interesting player, the numbers skewed from a year ago and we’ll chat in the next week or so.

MILAN LUCIC

Frank Seravalli was my guest on the Lowdown Thursday, we chatted about Milan Lucic and his future. Based on current events, my read is that things have settled down, the market wasn’t there, Edmonton and Lucic move on together.

Frank had different ideas.

I just can’t imagine going into next season with all the issues they had to close out last season more or less and Peter Chiarelli basically saying, ‘Well, it’s up to you guys to figure it out.’ I mean, that’s not really  how it works. It’s up to the manager to figure out a solution and find a way to make this team better. Because this team, on paper at least, certainly isn’t better.”

The other curio is Patrick Maroon, still out there flapping in the wind. Teams have apparently pursued him but he remains free as a bird. Could we see a Lucic (plus a pick) offload, a Maroon signing and a Nurse extension in rapid order? I’m inclined to say no, but Frank doesn’t pull this stuff from nowhere.

One thing I did want to point out here: When Frank says the team on paper isn’t better, that is true. However, making a bet on improved health and a recovery in net is a reasonable attack when you’ve no cap room and cannot afford to bleed even more talent.

Peter Chiarelli improved the penalty kill, grabbed some secondary scoring and may well be forcing his coach to bet on Jesse Puljujarvi. I’m onboard with all of it, no matter the optics. The window of opportunity to win the Stanley Cup should be open now, but the Edmonton Oilers badly need Jesse Puljujarvi, Kailer Yamamoto, Evan Bouchard and others to step up this year. Betting on the in-house option is somewhat novel for the Oilers, who have been sending good young players to other NHL teams for all of this century. Draft and develop, get good players, keep good players, same as it ever was. This is the way and the light.

David Staples wrote about the Seravalli interview I did on the Lowdown Thursday. His article is here.

written by

The author didn‘t add any Information to his profile yet.
Related Posts

233 Responses to "Boys of Summer"

« Older Comments
  1. Woodguy v2.0 says:

    Georges: I think Russell shouldn’t have sniped Brossoit in that Leafs game. That was just nasty.

    That’s the list of comps you come up with for Russell, eh? I’m curious. What have you learned from your years of study of defensemen? Is it trust your eyes? Is it Klefbom is a 1D?

    The HC shouldn’t still be the HC after a season like that. There’s not much point in drawing inferences on players when you have such wonky decision-making behind the bench. 2 out of 3 seasons with CMD we finish under 80 points. Eyebrows raised. Apparently we have an HC in waiting now. I’ll take 3 as the over/under for the longest losing streak for the HC to be fired. I’d prefer 1.

    Ahh, I missed Georges’ snark.

    That’s the list of FA’s Dmen who signed in the past 2 months and where they were slotted on their teams/where they’ve had success.

    I looked at their TOI/gm vs their team mates, and success with the various partners and eyeballed on which paid they played with each partner.

    Some guys, like Schenn I have as 5/6 played a bit in the top 4, but only when pressed up there due to injuries.

  2. Bank Shot says:

    OriginalPouzar: I think that burning that first year of the ELC can, in many cases, make the 2nd contract cheaper as the player won’t be as established and productive with the ELC is finished.

    The year towards free agency (40 games on the roster threshold) is the more material threshold, in my opinion.

    Yeah that’s true. Although sometimes you are playing a kid before he is truly ready, and you can miss out on a guy outperforming his entry level contract to a large degree.

    Draisaitl for instance could theoretically be on the last year of his entry level contract this season, and the Oilers would have an extra $5 million to play with.

    The other aspect about trying to delay starting an ELC is that you will have a better idea of what you have if the player is a little older. The Hall contract for example turned out great, but the RNH contract is just starting to look like fair market value 5 years into the deal.

  3. pts2pndr says:

    Richard S.S.:
    Contrary to popular belief, Milan Lucic will be on the Top Six.Peter Chiarelli trusts Lucic, trusts Lucic’s Agent as the straight shooter he is.IMO Milan Lucic will be in the Top Six until he plays his way out of it.I understand some posters have an anti-Lucic bias and can’t wait to tell everyone about it, again and again and again. So unless there is a new opinion, please be sparse with the tedious rhetoric.

    Chiarelli doesn’t make out the line up! He also doesn’t make the sun come up and is not responsible for all that ails the Oilers! McLellan was not his hire and marches to his own drummers! That is why Kassian gets paid third line wages and plays on the fourth line.

  4. Woodguy v2.0 says:

    Wilde: Hard disagree. Top-three moment of the season.

    This is dumb. The point of the post is that every player that’s in Puljujarvi’s age group that has outscored him has done so with much greater opportunity. That is indeed what the numbers say and is undeniable.

    The only question is whether or not he will have similar results with similar opportunity.

    Not actually sure where you’re disagreeing with me, or if you are at all.

    I do think that it looks like you’re characterising it as a player breaking out resulting in greater opportunity, whereas I’d say it’s greater opportunity resulting in the player breaking out.

    Edit – had the wrong filter on.

    Will re-post with correct numbers

  5. Woodguy v2.0 says:

    The worst thing about JP’s year last year was the 5v4 production.

    Only 2 goals and no assists in 48 minutes.

    I hope that mostly had to do with him not knowing what to do due to a language barrier.

  6. Southern Oil says:

    Gerta Rauss: Bouchard’s cap hit will only be $925k, so Gravel or the 14F (if they run 8D) will go to Bakersfield. (23 man max roster incl min 2 goalies)

    We can (and will) sign Bouchard without having to make a roster move, his contract will be pencilled against to the 50 man limit, and if he plays less than 9 games this season, his contract will slide and not count against the 50 man

    *edit-lol, I was typing as fast as I could and I still couldn’t post before OP

    Thanks to both you and Original Pouzar

  7. JimmyV1965 says:

    Wilde: The point isn’t really to investigate if other players had years of muted production similar to 98 just prior to their breakout year.

    The point is more to illustrate what /should/ happen in 2018-19.

    I’m ragging on how 2017-18 was handled, sure, but I meant for the message to be ‘if you want x, do y’ and not ‘you did z instead of y!!”.

    By comparing their years to Puljujarvi’s, what we’re doing is showing that it’s not that Puljujarvi has failed to have his breakout year, it’s that he hasn’t had a fair shot at it.

    Basically, looking forward to next year, if he’s on his way to 800+ minutes and 500+ with 29 or 97 and powerplay time, expect 50 points or more.

    If he’s with Strome and Lucic, expect bust-label, trade rumours, and ~30 points.

    But maybe JP wasn’t ready to play with the top line players. Did any of the guys you listed have low production to start with, and then start producing, or did they all click right away, like Drai did when he was called up a couple years ago? John Shannon said that JP was actually worried about playing with McDavid and Drai. He didn’t have the confidence to play with those two.

  8. Wilde says:

    Woodguy v2.0: Edit – had the wrong filter on.

    Will re-post with correct numbers

    I’m frightened already.

    Woodguy v2.0:
    The worst thing about JP’s year last year was the 5v4 production.

    Only 2 goals and no assists in 48 minutes.

    I hope that mostly had to do with him not knowing what to do due to a language barrier.

    Check out Brandon Saad 17-18.

    One for one.

  9. sumaclab says:

    I agree on Maroon. He is waiting. He knows something that we do not know. I would bet that Lucic is not an Oiler in 2 weeks. I could see the Rangers looking to get Lucic to reach the Cap floor. Perhaps even Vegas would take a looksee after being physically pushed around in the final by the Caps. Losing Perron. Neal and not yet being able to land Karlsson may push them to make a move to add Lucic for cap reasons and just for his veteran presence. Id bet on Vegas if he were to go anywhere.

  10. Woodguy v2.0 says:

    Wilde: Hard disagree. Top-three moment of the season.

    This is dumb. The point of the post is that every player that’s in Puljujarvi’s age group that has outscored him has done so with much greater opportunity. That is indeed what the numbers say and is undeniable.

    The only question is whether or not he will have similar results with similar opportunity.

    Not actually sure where you’re disagreeing with me, or if you are at all.

    I do think that it looks like you’re characterising it as a player breaking out resulting in greater opportunity, whereas I’d say it’s greater opportunity resulting in the player breaking out.

    To that end:

    JP’s 5v5 Pts/60 with his 6 most common forward line mates:

    Lucic – 466 min – 1.16/60
    McDavid – 257 min – 1.87/60
    Strome – 237 min – 1.26/60
    Khaira – 136 min – 0.88/60
    Nuge – 117 min – 0.51/60
    Draisaitl – 101 min – 2.37/60

    JP only played 13 minutes with 97 &29
    JP only played 6 minutes with 97 & 93
    JP only played 1 minute with 29 & 93

    So we can be sure that the TOI with the 97/29/93 were when they were all at C and not on a combo line with two of them.

  11. flyfish1168 says:

    Woodguy v2.0:
    The worst thing about JP’s year last year was the 5v4 production.

    Only 2 goals and no assists in 48 minutes.

    I hope that mostly had to do with him not knowing what to do due to a language barrier.

    Who were JP linemates mostly when he got 5×4 playing time?

  12. Bank Shot says:

    sumaclab:
    I agree on Maroon. He is waiting. He knows something that we do notknow. I would bet that Lucic is not an Oiler in 2 weeks. I could see the Rangers looking toget Lucic to reach the Cap floor. Perhaps evenVegas would take a looksee after being physically pushed around in the final by the Caps. Losing Perron. Neal and not yet being able to land Karlsson may push them to make a move to add Lucic for cap reasons and just for his veteran presence. Idbet on Vegas if he were to go anywhere.

    It’s either that or he wants to get paid like a 27 goal scorer and all the teams are saying “Piss Off…. McDavid.”

  13. Wilde says:

    JimmyV1965: But maybe JP wasn’t ready to play with the top line players. Did any of the guys you listed have low production to start with, and then start producing, or did they all click right away, like Drai did when he was called up a couple years ago? John Shannon said that JP was actually worried about playing with McDavid and Drai. He didn’t have the confidence to play with those two.

    Some of those players not only didn’t click right away, but were straight up bad outside of being paired up with skilled vets. Panarin carried PLD possession-wise, with the latter still only scoring 1.52/60.

    Hakstol played Konecny with Filpulla and they were dogshit(TK at 0.88/60), until he gave him the 1RW job with Couts and Giroux for free. Worked out. Pretty well the opposite of earning it or proving he was ready.

    I think the confidence will come. Just run it.

  14. Wilde says:

    sumaclab:
    I agree on Maroon. He is waiting. He knows something that we do notknow. I would bet that Lucic is not an Oiler in 2 weeks. I could see the Rangers looking toget Lucic to reach the Cap floor. Perhaps evenVegas would take a looksee after being physically pushed around in the final by the Caps. Losing Perron. Neal and not yet being able to land Karlsson may push them to make a move to add Lucic for cap reasons and just for his veteran presence. Idbet on Vegas if he were to go anywhere.

    All of these teams are trying to be the 3rd on the Karlsson deal first. Both NJD and NYR are looking to take salary to facilitate the deal as per Pagnotta this morning.

    Bet on whoever gets left out of that party starting the Lucic conversation back up afterwards. The price paid to offload Callahan/Girardi/etc will probably set the rate for dumping Lucic.

    Also apparently the Islanders are in on Karlsson. Wonder where the Sens had Brady Tkachuk’s old UDNTDP teammate Bode Wilde on their draft board?

  15. Woodguy v2.0 says:

    flyfish1168: Who were JPlinemates mostly when he got 5×4 playing time?

    With TOI With
    Cam Talbot 36.8
    Milan Lucic 20.9
    Patrick Maroon 19.8
    Ryan Strome 19.1
    Mike Cammalleri 18.5
    Ryan Nugent-Hopkins 17.3
    Darnell Nurse 14.7
    Oscar Klefbom 13.0
    Matthew Benning 10.8
    Jujhar Khaira 8.2
    Leon Draisaitl 7.3
    Al Montoya 7.1
    Pontus Aberg 6.9
    Andrej Sekera 6.7
    Drake Caggiula 5.4
    Connor McDavid 5.2
    Mark Letestu 5.0

    You can find it here: http://naturalstattrick.com/playerreport.php?season=20172018&stype=2&sit=pp&stdoi=oi&rate=n&v=t&playerid=8479344

  16. Drew says:

    Woodguy v2.0: Here’s the post from Russell signing his $4×4 13 months ago.

    The majority of the posters disliked the signing:https://lowetide.ca/2017/06/23/oilers-sign-russell/

    respectfully, i said a group. the signal to noise ratio was large around that time.
    I remember getting into it with a couple of posters when i suggested that i believed that a group (dellow, lowetide, gmoney, vic ferr. and you) could do a better job than some of the old timey gm’s in the league. Was called all sorts of names etc.

  17. Georges says:

    Woodguy v2.0: Ahh, I missed Georges’ snark.

    That’s the list of FA’s Dmen who signed in the past 2 months and where they were slotted on their teams/where they’ve had success.

    I looked at their TOI/gm vs their team mates, and success with the various partners and eyeballed on which paid they played with each partner.

    Some guys, like Schenn I have as 5/6 played a bit in the top 4, but only when pressed up there due to injuries.

    Awww, and I missed the WG take on numbers.

    I think you did a bit about not liking Faulk and his Rel GF%. That’s a step. If you do a little bit more work, you’ll be able to see if there’s enough evidence to determine if any skater (defenseman or forward) has a significant effect on his team’s GF% when he’s on the ice. You need more years of data but if you do the calculations, you’ll find defensemen usually don’t make a difference. They provide a context or background for forwards. Forwards are more likely to make a difference.

    What I notice about defensemen is that they seem to be defined partly by GF%, more by TOI. HC’s place defensemen in tiers. If they get the tier wrong and move a defenseman higher in the order than he can handle, GF% suffers. So coaches try to find the right level for each D. There are very few defensemen who can capably handle 1D minutes. Our HC, and apparently you, believed Klefbom was a 1D. Anyone can see from the public data that 1D’s don’t perform like Klefbom did to that point in his career. Klefbom’s results didn’t predict future 1D. It was a bad bet. The coach made a lot of bad bets. He failed miserably in a season with high expectations. And he’s still here.

    Those guys are bad comps for Russell, either because of TOI or Rel GF%. Russell has averaged over 22 minutes the past 5 seasons. His Rel GF% is a modest +0.6. That’s not a 4/5. He had a bad season. A lot of our players did. Still don’t know what that tells us about our players.

  18. Wilde says:

    Georges:

    What I notice about defensemen is that they seem to be defined partly by GF%, more by TOI.

    I knew you were a burner account Georges!!

  19. OriginalPouzar says:

    Bank Shot: Yeah that’s true. Although sometimes you are playing a kid before he is truly ready, and you can miss out on a guy outperforming his entry level contract to a large degree.

    Draisaitl for instance could theoretically be on the last year of his entry level contract this season, and the Oilers would have an extra $5 million toplay with.

    The other aspect about trying to delay starting an ELC is that you will have a better idea of what you have if the player is a little older. The Hall contract for example turned out great, but the RNH contract is just starting to look like fair market value 5 years into the deal.

    Absolutely – you lose out on the potential value contract aspect and, frankly, with our current cap structure, adding value contracts is imperative to future success that I think patience with the prospects is essential.

    Yes, good point on Drai but it cuts both ways – assuming another solid to very good season (lets say he gets 80 points with little time with McDavid which is no unreasonable given he’ll have better PP numbers) – he’d probably come in higher than $8.5M the way contracts are inflating currently.

    RFA vs. UFA, I know, but Logan Coture just got $8M X 8.

  20. flyfish1168 says:

    Woodguy v2.0: WithTOI With
    Cam Talbot36.8
    Milan Lucic20.9
    Patrick Maroon19.8
    Ryan Strome19.1
    Mike Cammalleri18.5
    Ryan Nugent-Hopkins17.3
    Darnell Nurse14.7
    Oscar Klefbom13.0
    Matthew Benning10.8
    Jujhar Khaira8.2
    Leon Draisaitl7.3
    Al Montoya7.1
    Pontus Aberg6.9
    Andrej Sekera6.7
    Drake Caggiula5.4
    Connor McDavid5.2
    Mark Letestu5.0

    You can find it here: http://naturalstattrick.com/playerreport.php?season=20172018&stype=2&sit=pp&stdoi=oi&rate=n&v=t&playerid=8479344

    Thanks for the site. Data Galore

  21. OriginalPouzar says:

    pts2pndr: Chiarelli doesn’t make out the line up!He also doesn’t make the sun come up and is not responsible for all that ails the Oilers!McLellan was not his hire and marches to his own drummers!That is why Kassian gets paid third line wages and plays on the fourth line.

    I don’t buy the “McLellan was not Chiarelli’s hire”.

    McLellan was hired after Chiarelli and I don’t believe the Peter agreed to come to the Oilers without the authority to hire his own coach in to a vacant spot. He was an established and respected manager and not desperate that he would accept a job without such authority.

    Also, Kassian signed his contract after playing the year on the 4th line and the playoffs on the 4th line. While he got a very appreciative contract, I don’t believe the manager signed him to be a third line player when he had just come off an entire season and playoffs on the 4th line.

  22. OriginalPouzar says:

    sumaclab:
    I agree on Maroon. He is waiting. He knows something that we do notknow. I would bet that Lucic is not an Oiler in 2 weeks. I could see the Rangers looking toget Lucic to reach the Cap floor. Perhaps evenVegas would take a looksee after being physically pushed around in the final by the Caps. Losing Perron. Neal and not yet being able to land Karlsson may push them to make a move to add Lucic for cap reasons and just for his veteran presence. Idbet on Vegas if he were to go anywhere.

    The Rangers are $3M below the cap floor but with a roster of 16 players – they have 7 players to fill out the roster including signing Spooner, Haynes and Vessey.

    They do not need Lucic to reach the cap floor.

  23. Woodguy v2.0 says:

    Georges,

    Lots to unpack here, so I’ll do it point by point:

    Awww, and I missed the WG take on numbers.

    I think you did a bit about not liking Faulk and his Rel GF%. That’s a step. If you do a little bit more work, you’ll be able to see if there’s enough evidence to determine if any skater (defenseman or forward) has a significant effect on his team’s GF% when he’s on the ice. You need more years of data but if you do the calculations, you’ll find defensemen usually don’t make a difference. They provide a context or background for forwards. Forwards are more likely to make a difference.

    That depends on what you’re looking at.

    If you are grouping the forwards with the dmen into one one pile, then yes more forwards have higher impact. That’s because you are measuring a forward’s impact on Dmen who play with all the lines while the forward play with their own line only and all the Dmen.

    Forwards and Dmen need to be examined separately because of how different their usage is. Forwards play with all the Dmen, Dmen play with all the forwards. Don’t group those results in one pile.


    What I notice about defensemen is that they seem to be defined partly by GF%, more by TOI. HC’s place defensemen in tiers. If they get the tier wrong and move a defenseman higher in the order than he can handle, GF% suffers. So coaches try to find the right level for each D. There are very few defensemen who can capably handle 1D minutes. Our HC, and apparently you, believed Klefbom was a 1D. Anyone can see from the public data that 1D’s don’t perform like Klefbom did to that point in his career. Klefbom’s results didn’t predict future 1D. It was a bad bet. The coach made a lot of bad bets. He failed miserably in a season with high expectations. And he’s still here.

    I don’t think Klef is a 1/2, but a 2/3 as he should be the 2nd best partner on pairing.

    His career RelTGF% are a concern for sure.

    Which of Klef’s results are you referring to? The ones this past year where he had 33% GF in the first two months and then 50% GF after that, or the year before before he was 53% or more?


    Those guys are bad comps for Russell, either because of TOI or Rel GF%. Russell has averaged over 22 minutes the past 5 seasons. His Rel GF% is a modest +0.6. That’s not a 4/5. He had a bad season. A lot of our players did. Still don’t know what that tells us about our players.

    That is the market for UFA Dmen. The NHL pays based on points and usage. They don’t all fit into the same pile, but I used the latest market to show his value.

    Also, I don’t use straight Rel GF%’s, I only use RelT GF%. Straight Rel’s are fraught with usage influence and should be discarded completely. Rel T GF% is the WOWYs, showing how the player affects each team mate individually, which removes a lot of the usage influence.

    Don’t use straight Rels, they lead to poor conclusions based on results having not much to do with the player.

    Its also funny that you look at “5 years of Russell” and decided he’s ok because it comes out close in the end.

    Here’s Russell’s RelGF% (for you) over the last 5 years and his RelTGF% over the last 5 years (what you should use)

    RelGF%
    13/14 -0.31
    14/15 +11.5
    15/16 +1.27
    16/17 -1.66
    17/18 -6.33

    You seem to want to throw out last year as an “off year”, but don’t mention his serious outlier year 4 years ago which impacts his overall number much more and also fail to mention he’s been negative his both years as an Oiler.

    He’s a 4/5 because he’s a drag when in the top 4.

    A player needs to be a positive contributor in the top 4 to not be a 4/5 imo.

    I don’t take his straight TOI as the gospel of what a player is or can play, its just where the coach played him.

    Whether or not he succeeded relative to his peers in that spot is what leads me to rate the player.

    Scandella was a 2nd pair Dman forever on MIN. Because he got moved to 1st pair doesn’t make him a 1st pairing Dman. It means that where he played.

    YMMV

    Lastly,

    Most people doing player ratings only use the past 3 years of results as those further back do not predict the future as much.

    The weighting is usually:

    3years – 20%
    2 years – 30%
    1 year – 50%

    A non-weighted Russell Rel GF% for 3 years is -1.6% GF

    A weighted Russell Rel GF% for 3 years is -3.46% GF

    I agree that last year was particularly bad and don’t see him as a -6.33 Rel GF% player

  24. Woodguy v2.0 says:

    flyfish1168: Thanks for the site. Data Galore

    Take some time to explore everything and pay close attention to the options of which kind/type of results to show.

    Data galore indeed.

  25. Bulging Twine says:

    Munny:
    Watching the PVR of the Croatia-Russia match and have to say the crowd in Sochi is simply redeckulous. Might be the loudest crowd I’ve ever heard.

    That’s what I was thinking too, especially with one particular clap/chant then they showed the crowd and barely anyone was clapping/chanting during that particular moment. Pretty sure they are pumping in sound which isn’t uncommon, many NFL teams do that. Maybe the Oilers should try that.

  26. Melvis says:

    “Darcy, what’s all that riffraff you dragged in doing passed out on the lawn? What will the neighbours think? I won’t have it…I just won’t have it. Never again!”

    #mrswoodguy

  27. Woodguy v2.0 says:

    Melvis:
    “Darcy, what’s all that riffraff you dragged in doing passed out on the lawn? What will the neighbours think? I won’t have it…I just won’t have it. Never again!”

    #mrswoodguy

    There is a 0% chance I’m having this at my house for that exact reason

  28. Melvis says:

    “Darcy, that rusty old dodge van out front. it’s been there for four days. There’s a greasy pizza box taped to the window. It says Do Not Disturb. Melvis. Can you arrange to have it towed?

    ” Is it giving off a weird smell?”

    “I don’t know and I don’t care. Just get it out!”

    “Jeeez”

  29. Georges says:

    Woodguy v2.0,

    The ole “lots to unpack” WG wall of text. Really missed that.

    Here’s my favorite:

    “Also, I don’t use straight Rel GF%’s, I only use RelT GF%. Straight Rel’s are fraught with usage influence and should be discarded completely. Rel T GF% is the WOWYs, showing how the player affects each team mate individually, which removes a lot of the usage influence.

    Don’t use straight Rels, they lead to poor conclusions based on results having not much to do with the player.”

    Delicious. Very WG.

    I also liked this:

    “Most people doing player ratings only use the past 3 years of results as those further back do not predict the future as much.

    The weighting is usually:

    3years – 20%
    2 years – 30%
    1 year – 50%”

    Have you used that recipe to “predict the future”?

  30. rickithebear says:

    Georges:
    The game is GF and GA. You win with superior goal differential.
    You want the best goal diff players you can get.
    So the line and Ppair accumulation gives you the best win potential.

    Do you think the goal diff is the same for the same GF%
    Cause that is how you treat them.
    It is a basic Scientific Methad question.
    It would be embarrassing to present varied data as the same.

    I will look at real data from 17-18 forwards.
    These 6 forwards have 40.0 GF%
    Values in per 60 minute rates.
    Player; EVGF; EVGA; Goal differential; (12 min x 80gm unit diff)
    Dickson; 1.05; 1.58; -.53; (-8.5)
    G. bourque; 1.38; 2.07; -.64; (-10.3)
    Baptiste; 1.62; 2.44; -.82; (-13.1)
    Kassian; 1.77; 2.65; -.88; (-14.1)
    Sorensen; 2.01; 3.02; -1.01; (-16.2)
    Caggulia; 2.12; 3.18; -1.06; (-17.0)

    Does GF% properly differentiate for
    HD ratio.
    Open shot exclusion
    NZ trap failure
    Offensive skater misses, turnover ratios

    Did you check for standard HD corsi/shot density and open corsi/shot density for every position in the 512 variant situations. Cause those would be the baselines to measure against.

    A seasons Accumulation of data.
    To which real differential numbers can be measured against.

    Which is why I state performance in relative comp ranges.

    Real GF%
    scientific method is laughing at you!

    Please give Accurate data for scoring mechanism!

  31. Georges says:

    rickithebear,

    Hi Ricki. Best wishes for your health.

    You picked one season’s worth of data to make the G+/- vs. GF% argument. There’s not enough there to achieve significance on whether the players you picked are negative players or not. They could be positive players who had bad luck.

    I think what you’re trying to say is if two players have the same below 50 GF%, the higher event player will cost his team more in the long run. So you need to look at rate differential, not just proportion. Sure. That arithmetic is clear.

    A few observations:

    – you need more data to do what you tried to do; I think you might just be seeing small sample effects in the list you provided

    – you need to build confidence intervals for your rates; because n * lambda is low, you probably shouldn’t use the normal approximation to the poisson

    – you should use empirical Bayes estimation to adjust proportions for players with different event totals; again, more data provides more evidence

    – a quick look at the data over the past 3 years suggests players in the 40 GF% range may differ in +/- rates by about 0.6 at the 2-sigma level; this works out to six goals over the course of a season at the extremes (these types of players are unlikely to play the full 12 min x 80)

    – 6 goals may not exceed the confidence intervals; I doubt it would.

    But it’s a valid point and one to check whenever you make player comparisons. Using proportions allows for simpler statistical tests. I’m not sure you lose enough in the translation to warrant the uptick in complexity. I may have a look.

    Cheers.

  32. Andy Dufresne says:

    edmoil3:
    So two years ago Frank picked oilers to miss the playoffs, they won a round, almost two. Last year they were Franks cup winners, they missed the playoffs. Frank has no ability to read the situation in Edmonton. Based on his track record, I’m actually excited that frank is crapping on them this summer.

    Oilers holes at the end of the season were coaching, P.K/Pp, goaltending, RW, puck moving D men.

    Peter drafted two RW wingers two years in a row. Then Bouchard. Signed Koskinen. Put together a nice balanced coaching staff to help with tactics, special teams and develop of youth especially the D. He signed Brodziak and Rieder for depth and PK tools. Our Gm is well aware of the holes and he addressed them without giving up assets.

    The voice of reason.
    Post more often.
    Please.

« Older Comments

Leave a Reply

Want to join the discussion?
Feel free to contribute!
© Copyright - Lowetide.ca