PERSONAL OPINION

Every now and then, someone will ask me to post my personal opinion about the Oilers and what they should do moving forward. The blog, as you know, is kind of driven by a combination of prevailing wisdom and trying to figure out what the hell this team is trying to do at any given time.

I don’t often prattle on about what I think for several reasons. First, I don’t have all of the information. Second, when I do make a suggestion (find balance, sign Demers, keep your powder dry) it never works that way so at some point it becomes a colossal waste of time. Too many options, plus I always remember the time Earl Weaver and Whitey Herzog won World Series in completely different ways.

I write mostly about possible options, and after that wait to see what the people in charge do to fix the problems. I hope one day to see the team achieve balance, that is one thing that has eluded the town team for a decade and a little bit. That said, here are 10 things I think Peter Chiarelli should do between now and the trade deadline.

  1. Improve the backup goalie situation. It is very dangerous to be exposed at such an important position. Edmonton has a chance at the playoffs, goalies are close to free, and Jonas Gustavsson does not have the confidence of the coach.
  2. Trade Kris Russell for a second-round pick. The coach is developing an attachment to him, and Chiarelli is actually doing a reasonable job of collecting defensemen (Larsson, Benning added to Sekera, Klefbom, Davidson, Nurse). In today’s NHL you need a pick every round just to keep pace, and PC has the dual job of improving the NHL team and building a prospect cupboard that is badly lacking. I believe he has the scouting staff, wise for him to invest in that group with a timely trade of Russell.
  3. Run the Nuge inside the top 6F for the rest of the season. Pretty clear, that for all of his offensive acumen, 93 is no longer focusing enough on that side of the game. Since checkers are easier to find, Chiarelli would be wise to find one—Hanzal if you can—and see what a top 6F of Maroon—McDavid—Leon, Lucic—Nuge—Eberle can do over the last 46 games.
  4. Recognize that the high risk trade is not working for him. He can recover from the Reinhart and Hall trades, but is burning daylight on this rebuild. If Chiarelli gets less than 100 cents on the dollar for Nuge or Klefbom—and those are his next available assets—he may not be around for the good times. No matter what the reputation (Stanleys have high value, and Chiarelli has one), you can’t keep bleeding out on the big trades.
  5. Push the coach to play Jesse Puljujarvi. I expect the decision has already been made, but honestly this kid can play right now. I have a growing suspicion we are dealing with a 25-25-50 winger instead of a 35-35-70, but his two-way acumen will be valuable all the same. I look at JP the same way Sather looked at the 1994 draft: Edmonton didn’t need to hit a home run there, just needed to make damn good and sure that the picks can play. Puljujarvi can play, let him do it.
  6. Send down Drake Caggiula. He is a little lost. No harm, no foul, I really like this young player. He is not helping, is all.
  7. Get creative on the RH shot for the power play. I would rather see Mr. Chiarelli acquire Brandon Montour than Dennis Wideman, and would further consider it reasonable for him to invest some of the future to get said player. Looking for a long-term solution is still the right thing to do for this building team.
  8. Keep the powder dry on rentals at the deadline. Dealing for a Montour makes sense, that player will hang around for a long time (potentially). I think it is unwise (goalie slot aside) to spend draft picks inside the top 100 for short-term improvements. Are we sure Jarome Iginla (who I love as a player, btw) will score more than Anton Slepyshev? Why risk making a move with low payoff now and later? One exception: Hanzal.
  9. Sell off all available UFA veterans as soon as possible. If the Oilers can send away Matt Hendricks now, they should do so, allowing Anton Lander and or Jujhar Khaira to stay here the rest of the year.
  10. Find the analytics guy who tells you what you don’t want to hear, not the one who agrees with you. I have no idea if PC listened to Tyler Dellow, but am sure as hell certain the analytics guy was giving strong readings on the tea leaves when in Edmonton’s employ. This isn’t some simple freaking task, some pretty smart people are wrong every damned day on this stuff and get exposed in the years that follow. This is damned serious, way more serious than some verbal spat over Corsi online. If you are wrong on a player, you are wrong forever. The most important thing about the analytics guy is that he is reading the data correctly. Seriously. That’s it. Peter Chiarelli’s recent quotes on Kris Russell should be making the sweat run down the crack of your ass 24-7 if you are an Oilers fan.

These are my opinions.

written by

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

111 Responses to "PERSONAL OPINION"

  1. commonfan14 says:

    Can’t really double down on both Reinhart and Russell, so something will have to give there.

  2. Centre of attention says:

    You nailed it on all counts, LT. Zero argument from me on any of these opinions.

  3. stush18 says:

    Hmmm.

    Quick disagreement LT. Calling JPs offensive potential at this point seems a little early. He’s doing better than drai had as a first year.

    I really think they’re giving him the same Leon treatment. He’s too strong to be in junior. I think it’s obvious they want him to stay on North American ice.

    I expect him to be moved to the AHL sometime near his 40 game mark.

  4. Centre of attention says:

    Hall just did that inside-out move, but kept the puck this time. He also sauced the puck threw a defender to his teammate for a tap in right afterwards.

    Le sigh.

  5. Justthestatsman says:

    Wow, from the Fred Astaire comparison this morning to “making the sweat run down the crack of your ass” this afternoon, and not a word out of place in either. You sir, are a 5-tool blogger. (I’m not literate enough to know what those 5 tools are, but I’m sure you have them!)

  6. Lowetide says:

    stush18:
    Hmmm.

    Quick disagreement LT. Calling JPs offensive potential at this point seems a little early. He’s doing better than drai had as a first year.

    I really think they’re giving him the same Leon treatment. He’s too strong to be in junior. I think it’s obvious they want him to stay on North American ice.

    I expect him to be moved to the AHL sometime near his 40 game mark.

    Yeah, growing suspicion but hope to see the offense blossom. Certainly still in an adjustment period.

  7. kinger_OIL says:

    Centre of attention:
    You nailed it on all counts, LT. Zero argument from me on any of these opinions.

    – Well if Russel, signed on a 1 year contract, could garner a 2nd, then I agree.

    – Not going to happen though.

    – Between getting Sekera, Larsson, Benning, properly slotting Nurse not being stubborn on Reinhart, flushing Hunt, signing Russel for a year, Gryba for a year, I think he has done a good to great job on D management. More than anything else (save David), this team is much better as a result of having a very good D (RHD Cannon notwithstanding)

  8. stush18 says:

    What is everyone okay with in regards to a Russell extension?

    I’d be okay with 2 yrs ~ 3 mill.

    Then you can protect 7 forwards, 3 dmen, 1 goalie.

    Nuge, lucic, ebs, drai, maroon, pitlick, letestu

    Klef, sekera, Larsson

    Talbot

    Now you have pouliot, kassian, fayne, lander, Russell, davidson exposed to be picked. You are either going to lose davidson or Russell, imo, which allows the other to be kept and play the following year. If Russell (hopefully) is grabbed, then next year you can run

    Maroon-mcdavid-drai
    Lucic-nuge-ebs
    Pou-cags-JP
    JJ-lander-pitlick
    Slepy

    Klef-Larsson
    Sekera-nurse
    Davidson/Russell-benning
    Talbot
    Brossoit

    More or less this years team, but more experienced.

  9. Centre of attention says:

    I don’t think the Bruins have gained possession in the offensive zone yet.

    Just desperately dog piling their own crease, helplessly watching Blue Jacket players stroll in seemingly out of nowhere and score into open nets.

  10. Centre of attention says:

    Ok, Bobrovski is not invincible. That much is clear now.

  11. Lowetide says:

    Stush: If they sign Russell to a two-year, $6 million deal, I don’t think anyone would have much to quarrel over. PC would probably trade one of Sekera, Nurse, Klefbom or Davidson to get a RHD to go along with Larsson and Benning, but as a third pair D at that price you can live with it for sure.

    I would give that job to Brandon Davidson, though.

  12. Woodguy says:

    Get creative on the RH shot for the power play. I would rather see Mr. Chiarelli acquire Brandon Montour than Dennis Wideman, and would further consider it reasonable for him to invest some of the future to get said player. Looking for a long-term solution is still the right thing to do for this building team

    The coach kept running 14 out there with 15, 42, and 98 on the bench.

    I agree he needs to get creative with another RHS but the solution might be on the roster.

  13. Woodguy says:

    stush18,

    I don’t see any reason to extrend the 5th best LHD on the roster.

    Sekera
    Klefbom
    Nurse
    Davidson
    Russell

    is how I would rank them this year.

  14. Glass says:

    What would it take to acquire Montour, or better yet, what would you be willing to give up? Reinhart 1 for 1 for Montour? Lastly, if you had to choose one guy aside from Montour & Wideman, who would it be?

  15. Lowetide says:

    Woodguy:
    stush18,

    I don’t see any reason to extrend the 5th best LHD on the roster.

    Sekera
    Klefbom
    Nurse
    Davidson
    Russell

    is how I would rank them this year.

    PC would have to trade Klefbom, I would think. Get a RHD, lose Davidson in the expansion draft.

  16. Lowetide says:

    Glass:
    What would it take to acquire Montour, or better yet, what would you be willing to give up? Reinhart 1 for 1 for Montour? Lastly, if you had to choose one guy aside from Montour & Wideman, who would it be?

    I would do a Glen Sather, get someone the other team maybe soured on a little. Ryan Pulock, that kind of guy. Get someone who could turn out, there is time to experiment.

  17. LadiesloveSmid says:

    Klefbom, Larsson, Sekera, Benning, Davidson, and Nurse should all be starting 6 D next season. Russell makes no damn sense, I will be mad even if they sign him to a 1-yr 2M deal. I’d prefer Gryba or Fayne on the last year of his deal.

    Benning and Nurse being exempt from expansion is big, really hope Davidson brings it like last year so he forces them to protect him (not Kassian, Maroon, Letestu like Stauff says)

  18. Centre of attention says:

    Lowetide: PC would have to trade Klefbom, I would think. Get a RHD, lose Davidson in the expansion draft.

    This chain of moves does not please me at all.

    Not one bit.

    How can you look at that depth chart and tell me Klefbom and Davidson are the ones leaving?

    If that is the way Chiarelli is leaning I give up. This team is beyond hopeless if that is their train of thought. “Gee, our defense isn’t quite perfect. I know, lets trade our second best defenseman!”

  19. slopitch says:

    LT, like it when you do this. You should do it more (2-3x a year).

    While your at it 1) can this cluster win a Stanley without a number 1 d (assuming Nurse, Klefbom, Larsson top out in the 2-3 range) 2) seems like you want to build for next year (year 3 McDavid ELC). I tend to agree. If that’s the case, how important is making the playoffs this year?

    I’m with ya on Hanzal.

  20. Lowetide says:

    Centre of attention: This chain of moves does not please me at all.

    Not one bit.

    How can you look at that depth chart and tell me Klefbom and Davidson are the ones leaving?

    If that is the way Chiarelli is leaning I give up. This team is beyond hopeless if that is their train of thought. “Gee, our defense isn’t quite perfect. I know, lets trade our second best defenseman!”

    That’s why the Russell quote from Chiarelli is a bit of a mystery. If you look at this roster rationally, Russell works as 3LHD, and Edmonton has better and cheaper. That said, the only way signing him makes sense imo is dealing someone from higher up the depth chart.

    Unless they plan on running him 2RHD next season.

  21. Lowetide says:

    slopitch:
    LT, like it when you do this. You should do it more (2-3x a year).

    While your at it 1) can this cluster win a Stanley without a number 1 d (assuming Nurse, Klefbom, Larsson top out in the 2-3 range) 2) seems like you want to build for next year (year 3 McDavid ELC). I tend to agree. If that’s the case, how important is making the playoffs this year?

    I’m with ya on Hanzal.

    Thanks. I tend not to enjoy writing or posting, usually because it comes out of frustration. I didn’t feel the usual bile this year, probably means progress. 🙂

    I think they had a chance at a No. 1 in Subban, but now are probably on their own. Klefbom has a chance imo, I do not think there is another in the system trending in that way. PC probably drafts a defenseman in the first round this summer, that will be the new best bet.

  22. prairieschooner says:

    Still lots of work to do but

    How big an impact will the expansion draft have on PC’s player ins and outs
    We are already leaving some guys exposed that we would rather not

    Does the cap increase as a result of the revenue from Las Vegas buy in fee

  23. Woodguy says:

    Lowetide: That’s why the Russell quote from Chiarelli is a bit of a mystery. If you look at this roster rationally, Russell works as 3LHD, and Edmonton has better and cheaper. That said, the only way signing him makes sense imo is dealing someone from higher up the depth chart.

    Unless they plan on running him 2RHD next season.

    Peter seems to like goal metrics and Russell is still a 1018 despite running a 95.74 the last 8 games

    Speaking of PDO, this is interesting.

    Klef hasn’t had a good PDO since he got to the NHL.

    Those who (incorrectly) blame/credit the player for their PDO would consider him a poorer player due to his goal metrics.

    I posted the Expected PDO to see if Klef was leaking high danger goals or Russell was keeping them to the outside more and the answer is no, not this year.

    Player Season PDO xPDO
    KRIS.RUSSELL 13/14 – 100.06 – 99.85
    KRIS.RUSSELL 14/15 – 103.87 – 100.28
    KRIS.RUSSELL 15/16 – 100.52 – 99.99
    KRIS.RUSSELL 16/17 – 101.77 – 99.36

    OSCAR.KLEFBOM 13/14 – 95.18 – 100.09
    OSCAR.KLEFBOM 14/15 – 95.57 – 99.06
    OSCAR.KLEFBOM 15/16 – 98.46 – 99.71
    OSCAR.KLEFBOM 16/17 – 97.26 – 99.21

    I base my “Peter likes goal metrics” on what he has said since taking over the Oilers.

    It would not surprise me if he trades Klefbom.

    I find it strange that we are talking about writers and managers getting fooled by PDO almost 10 years after Brian stumbled on to it, but here we are.

    Not a couple weeks ago Spector said Russell has a 1021 PDO because of how he defended.

    Just this morning Staples said to me that Russell’s PDO tanking was due to him playing with Larsson (even though he only played 4 games with Larsson)

    Its probably the easiest fancystat to understand and the one with over whelming evidence of its regression, but nope.

    People are not open to evidence that contradicts their preconceived notions.

  24. Centre of attention says:

    Lowetide:

    Unless they plan on running him 2RHD next season.

    That’s what I was thinking.

  25. stush18 says:

    Woodguy,

    Can’t see it. Signed him to that contract. Would be very surprised.

    Woodguy:
    stush18,

    I don’t see any reason to extrend the 5th best LHD on the roster.

    Sekera
    Klefbom
    Nurse
    Davidson
    Russell

    is how I would rank them this year.

    Agreed.

    But according to Bruce Wayne’s theory (which I agree with), is to acquire more players before the expansion draft, rather than ship away.

    We can protect more of the roster, and lose less important pieces if we sign Russell.

    Personally I dunno if I would sign either. But I am not opposed to a deal like the one I mentioned.

    Tbh I think I would rather resign gryba, and try to trade for a RHD.

  26. Centre of attention says:

    Remember how McLellan LOVED Korpikoski? And Chiarelli talked about “Having time” for Korpikoski?

    Then he was suddenly bought out.

    I think there is a chance we are misreading Chiarelli’s verbal RE: Russel. Perhaps.

    “At some point there will be a discussion about an extension”

    1: Some point could mean post-expansion draft.

    2: A discussion could mean anything. They “discussed” contracts with Demers and Versteeg, but nothing actually happened.

    Pretty much, lets play this one by ear. Stauffer has mentioned more than once that he see’s Benning in the top 4 next year.

    Benning surprising this year might just save us from Russel’s retirement contract.

  27. Centre of attention says:

    Schultz now has 17 points in his last 14 games. Playing 20+ minutes a night since Letang went down and hasn’t really lost a step.

    Confidence can do wonderful things.

  28. sliderule says:

    Don’t get fooled that analytics was what turned the cup over to Pittsburg.

    It was a GM and coach going with a system that favoured their high end forwards and speedy bottom six.

    With one good defenceman they got back into their own zone quick enough to beat the forecheck and dumped the puck out to the forwards who in most cases would stay in neutral zone.

    There were not many pretty outlet passes .It was dump and let the forwards fight for it in neutral

    It was a system using speedy forwards and defencemen to beat the forecheck and advance rthe puck

    Justin Schultz who could not break the cycle with oilers didn’t have to as he got back fast and dumped it out.

    If you look at the numbers that Schultz is putting up analytics would say oilers should sign him.😄

  29. Centre of attention says:

    sliderule:
    Don’t get fooled that analytics was what turned the cup over to Pittsburg.

    It was a GM and coach going with a system that favoured their high end forwards and speedy bottom six.

    With one good defenceman they got back into their own zone quick enough to beat the forecheck and dumped the puck out to the forwards who in most cases would stay in neutral zone.

    There were not many pretty outlet passes .It was dump and let the forwards fight for it in neutral

    It was a system using speedy forwards and defencemen to beat the forecheck and advance rthe puck

    Justin Schultz who could not break the cycle with oilers didn’t have to as he got back fast and dumped it out.

    If you look at the numbers that Schultz is putting up analytics would say oilers should sign him.

    Toronto has some cash free this summer.

    They will pay Schultz and pair him with his old NCAA partner Jake Gardiner.

    Story book ending?

  30. NYCOIL "Gentleman Backpacker" says:

    Woodguy,

    We see this over and over.
    MacT on fancies and Schultz
    Rishaug on fancies and Pouliot
    Your examples, etc.
    Would you really expect much different, though? It was similar in baseball, too, and still is for some teams.
    And baseball is arguably easier to track like that than hockey.

  31. JimmyV1965 says:

    Correct me if I’m wrong, but if Chia can sign Russell to an attractive two-year contract, does this not provide valuable cover for the expansion draft? Vegas might select him or it could provide valuable depth If they take Davidson. And isn’t depth on defence always a good thing?

  32. stush18 says:

    Centre of attention,

    He would fit in well there.

    Reilly-zaistzev
    Gardner-Schultz

    They just need a lot more depth lol

  33. stush18 says:

    JimmyV1965:
    Correct me if I’m wrong, but if Chia can sign Russell to an attractive two-year contract, does this not provide valuable cover for the expansion draft? Vegas might select him or it could provide valuable depth If they take Davidson.And isn’t depth on defence always a good thing?

    Yup that was my point.

  34. Woodguy says:

    stush18: Yup that was my point.

    If you’re Russell why would you sign anything without a NMC to protect you from going to Vegas?

  35. who says:

    Good post LT. Love your list. Would like to comment on your 10 points.
    1) Can’t start Talbot 75 times so they need to find a back up the coach will trust. I would give LB a spin before spending much on another back up.
    2)Agreed. Like what Russell has brought so far but don’t want him signed to an expensive long term extension, and they are so deep at left d it just makes sense to get something for him.
    3)Not sure I agree but I think the decision has already been made. The only move left is to bring Lander back up and let him play 4th line center.
    4)Sort of agree but I would change it to read “Recognize that he no longer HAS to make the high risk trade” but don’t be afraid to pull the trigger if it makes sense. I think Chia has shored up the right defense enough that it is no longer a black hole. I don’t see a glaring need that requires a high risk trade at this point.
    5)Give JP some decent minutes or send him down. If they load up the top two ines it only makes sense to send him down. I think he is good enough to play on a scoring line right now but obviously TMac doesn’t.
    6)Agreed. Would make a nice skilled linemate for JP.
    7)If Chia can find a young righty d with some offense, Great. I just don’t see many teams giving this player up. Don’t really think we need a righty at the top of the umbrella and Letestu is doing fine on his off side. Slepy and JP would be credible options here as well.
    8)Agreed. Hate rentals at the best of times and the only way the Oilers should give up a propsect or draft pick is if they are getting a long term, possible core player back.
    9)Agreed. Same line of thinking as trading Russell. If they are not going to be back next year might as well get something for them.
    10)Don’t have an opinion. Don’t know enough about analytics or the Oilers managment to make a comment but I don’t think this is as important as some of your other points.
    The big theme I would be pushing here is patience. Push for the playoffs but don’t sell out to make it, or make a deep run, this year.
    On an unrelated note: Man does Barzal ever look good for Team Canada

  36. Woodguy says:

    sliderule:
    Don’t get fooled that analytics was what turned the cup over to Pittsburg.

    It was a GM and coach going with a system that favoured their high end forwards and speedy bottom six.

    With one good defenceman they got back into their own zone quick enough to beat the forecheck and dumped the puck out to the forwards who in most cases would stay in neutral zone.

    There were not many pretty outlet passes .It was dump and let the forwards fight for it in neutral

    It was a system using speedy forwards and defencemen to beat the forecheck and advance rthe puck

    Justin Schultz who could not break the cycle with oilers didn’t have to as he got back fast and dumped it out.

    If you look at the numbers that Schultz is putting up analytics would say oilers should sign him.

    Running a system designed to play to your player’s strengths (speed) and minimize their weaknesses (puck battles) is…….wait for it……good analytical thinking.

    Schultz is doing well in a 3rd pairing role (although has been elevated lately)

    He has some of the least TOI% vs Elites in the WoodMoneys this year.

    Not making, $4MM, playing 3rd pair.

    That’s what a lot of us wanted for him here, but they slammed him into 1/2 RD off the hop and here we are.

    Also,

    The withering of 3rd and 4th liners in the EC who don’t bring any offence is the influence of the fancystaters.

    I agree with you that Sullivan’s system was probably not influenced by fancystaters, but disagree on the removal of puck optional players. That’s a fancystat hallmark and while Rutherford never employed a lot of puck optional players, he did employ them during his history as a GM.

  37. Centre of attention says:

    Woodguy: If you’re Russell why would you sign anything without a NMC to protect you from going to Vegas?

    This is why I think they have a hand shake agreement in place, where they will revisit things in the summer.

    I don’t think Chiarelli is in a huge rush to sign him immediately, just my opinion.

  38. stush18 says:

    Woodguy: If you’re Russell why would you sign anything without a NMC to protect you from going to Vegas?

    Because you couldn’t get a contract last summer?

  39. Ryan says:

    sliderule:
    Don’t get fooled that analytics was what turned the cup over to Pittsburg.

    It was a GM and coach going with a system that favoured their high end forwards and speedy bottom six.

    With one good defenceman they got back into their own zone quick enough to beat the forecheck and dumped the puck out to the forwards who in most cases would stay in neutral zone.

    There were not many pretty outlet passes .It was dump and let the forwards fight for it in neutral

    It was a system using speedy forwards and defencemen to beat the forecheck and advance rthe puck

    Justin Schultz who could not break the cycle with oilers didn’t have to as he got back fast and dumped it out.

    If you look at the numbers that Schultz is putting up analytics would say oilers should sign him.😄

    Did you watch the Pens in the SCF?

    Justin Schultz didn’t have to break the cycle because he didn’t play against any good hockey players.

    Sullivan ran the Schultz/Letang bait-and-switch game against the Sharks.

    Schultz used his speed to full advantage to skate to the bench whenever anyone who could score against him was on the ice. Then Letang jumped over the boards.

    I’m not sure we’ve ever seen such a cat and mouse game for match ups.

  40. Coffeys_Messy_eh says:

    Speaking of Matt Benning, does anyone know why he isn’t playing? Did he get hurt? Did I miss an announcement from the team/media?

  41. Centre of attention says:

    Bear scores on a slap shot.

  42. Chachi says:

    Woodguy: Running a system designed to play to your player’s strengths (speed)and minimize their weaknesses (puck battles) is…….wait for it……good analytical thinking.

    Schultz is doing well in a 3rd pairing role (although has been elevated lately)

    He has some of the least TOI% vs Elites in the WoodMoneys this year.

    Not making, $4MM, playing 3rd pair.

    That’s what a lot of us wanted for him here, but they slammed him into 1/2 RD off the hop and here we are.

    Also,

    The withering of 3rd and 4th liners in the EC who don’t bring any offence is the influence of the fancystaters.

    I agree with you that Sullivan’s system was probably not influenced by fancystaters, but disagree on the removal of puck optional players.That’s a fancystat hallmark and while Rutherford never employed a lot of puck optional players, he did employ them during his history as a GM.

    Meh, without Crosby and Malkin Pittsburgh would be awful.

  43. Lowetide says:

    Coffeys_Messy_eh:
    Speaking of Matt Benning, does anyone know why he isn’t playing?Did he get hurt?Did I miss an announcement from the team/media?

    He had a head hit in a game, has not returned (iirc) since. Could be nothing, could be everything.

  44. fifthcartel says:

    I’m a bit skeptical of a Russell extension. He doesn’t really fit and with Benning’s play you have Larsson/Benning as 2/3 of the RHD.

    Then they have left handed defensemen for days and although I fear they trade Klefbom because of their unwavering belief in Darnell Nurse, it still seems like a stretch, even for the Oilers.

  45. stush18 says:

    Lowetide: He had a head hit in a game, has not returned (iirc) since. Could be nothing, could be everything.

    That’s the best/most frusterating thing about this team. Nothing leaks out, which is maddening as a fan.

  46. stush18 says:

    fifthcartel,

    Ya I can’t see him being moved. He’s clearly our most offensive dman right now I think, signed cheap long term, and is gobbling up minutes.

    Klef-Larsson
    Nurse-sekera
    Davidson-benning

    That’s where I see things going next year.

  47. Georges says:

    Woodguy: Peter seems to like goal metrics and Russell is still a 1018 despite running a 95.74 the last 8 games

    Speaking of PDO, this is interesting.

    Klef hasn’t had a good PDO since he got to the NHL.

    Those who (incorrectly) blame/credit the player for their PDO would consider him a poorer player due to his goal metrics.

    I posted the Expected PDO to see if Klef was leaking high danger goals or Russell was keeping them to the outside more and the answer is no, not this year.

    PlayerSeasonPDOxPDO
    KRIS.RUSSELL13/14 –100.06 –99.85
    KRIS.RUSSELL14/15 –103.87 –100.28
    KRIS.RUSSELL15/16 –100.52 –99.99
    KRIS.RUSSELL16/17–101.77 –99.36

    OSCAR.KLEFBOM13/14 –95.18 –100.09
    OSCAR.KLEFBOM14/15 –95.57 –99.06
    OSCAR.KLEFBOM15/16 –98.46 –99.71
    OSCAR.KLEFBOM16/17 –97.26 –99.21

    I base my “Peter likes goal metrics” on what he has said since taking over the Oilers.

    It would not surprise me if he trades Klefbom.

    I find it strange that we are talking about writers and managers getting fooled by PDO almost 10 years after Brian stumbled on to it, but here we are.

    Not a couple weeks ago Spector said Russell has a 1021 PDO because of how he defended.

    Just this morning Staples said to me that Russell’s PDO tanking was due to him playing with Larsson (even though he only played 4 games with Larsson)

    Its probably the easiest fancystat to understand and the one with over whelming evidence of its regression, but nope.

    People are not open to evidence that contradicts their preconceived notions.

    I think PDO has reached the status of preconceived notion for the fancystats community.

    It comes down to whether PDO is capturing a skill element of the game because teams/players have varying levels of talent or whether teams/players are close enough in talent that the percentage results we observe are just random draws from a common league-wide distribution.

    If I’m reading you correctly, fancystats has decided that there is overwhelming evidence that PDO is not reproducible and therefore does not capture skill. It’s luck. And luck evens out.

    I posted earlier in the season on how teams with elite talent can be relied upon to beat the median for PDO stats for long stretches. And how expansion teams with a skill deficit can be counted on to perform below the median for a number of years.

    I’m going to expand on my critique of the PDO is luck and not reproducible view in the next bit. (Tulsky’s articles on the topic are a good place to start.) But I need some time to gather the pieces I’ve worked on besides the elite talent/no talent teams spreadsheet I shared earlier.

    For now, I’d like to make an observation on the numbers you posted for Klefbom. Those numbers indicate to me that he’s probably been losing on GF% for all of those years. He’s never popped his head above 100 on PDO. Shouldn’t that be a concern? At what point should it be a concern? Never? Is there a timeline for the PDO correction? How can we be sure Klefbom, who has never posted a 100+ PDO, will post a 100+ PDO?

    If you were a coach, who would you give more ice time to, all other things being equal: the guy who’s losing you games or the guy who’s winning you games (GF%)? If your job was on the line? Isn’t the NHL a great big sifting machine for results? Aren’t the PDO patterns that fancystats relies on the product of the NHL’s sifting machine, i.e., data produced by the survivors? If so, can we apply the patterns we’ve found among the successful survivors to all NHL players? Like Klefbom 143 games into his career?

    (WG, I know you said you’re not a fancystats evangelist, and I’m not holding you to that. I hope this would be an enjoyable back and forth for you, that’s all. Not looking to score points, just trying to share another perspective and sharpen our thinking as a community. If I’m wrong, then the fancystats thinking on PDO will have withstood my little assault. That’s a good result in my book as well.)

  48. rickithebear says:

    Davidson
    Klefbom
    Sekera

    WG tsk tsk

  49. Centre of attention says:

    Benson has an assist so far tonight, giving him 38 points in 30 games played.

  50. Lowetide says:

    rickithebear:
    Davidson
    Klefbom
    Sekera

    WGtsk tsk

    Nice to see you, bear.

  51. Gordies Elbow says:

    rickithebear,

    Happy holidays Mr. Bear!

    Happy to see you back around these parts!

  52. fifthcartel says:

    stush18,

    Benning gives them a nice option as a strong third-pairing RHD. I wonder if its McLellan liking the player (Russell, that is) a lot more than Chiarelli. He can’t really say he’s not interested in re-signing him.

    A Russell extension wouldn’t shock me but it’d be a little surprising and a curious decision.

  53. Woogie63 says:

    No way is PC trading Russell for a 2nd round pick. Our young guys need +100 games more before we willingly give up a 4/5 dman with +600 NHL games.

    Sekera. 603 games
    Russell. 602 games

    Fayne. 389 games
    Larsson. 310 games
    Gryba 235 games

    Klefbom 143 games
    Nurse 96 games
    Benning 24 games
    Davidson 69 games

  54. digger50 says:

    I see everybody looking at who we are likely to lose in the expansion draft. What about opportunities to gain a valuable player? Will there not be a flurry of activity as teams attempt to turn a likely LV pick into something prior to the draft? I do not know enough to offer a suggestion but it would seem that opportunities will emerge.

    This should make draft picks more valuable, and exempt players such as Benning more valuable.

  55. Centre of attention says:

    Watching the Giants game and wow is Benson greasy.

    Takes a 3 stride run at a guy and creams him into the boards. No call.

    Scrum ensues at the next whistle and you bet your ass he’s right in the middle of it.

  56. Centre of attention says:

    Benson on first unit penalty kill.

    Interesting observation: He is taking face offs.

    Wonderful play by Benson short handed, pokes the puck off the defenders stick then chases it down the ice, beats his man there and throws a dandy of a pass all the way across the zone to a trailing player for a great short handed chance.

  57. Gordies Elbow says:

    Centre of attention,

    Hasn’t changed since his play with SSAC Lions – one of the few players that I’ve had the privilege of watching before the jump to the WHL. You could see why Bob Green lost a night of sleep when he was available in 2nd round that draft.

    Reminds me of Erik Cole.

  58. Centre of attention says:

    Giants short handed again.

    Benson again taking the face off.

  59. stush18 says:

    rickithebear:
    Davidson
    Klefbom
    Sekera

    WGtsk tsk

    Welcome back mr bear.

  60. VOR says:

    LT, I am not sure you want an analytics guy at all. Yes, I know that is heresy. But hear me out.

    For those who don’t want to read a wall of text, and I have included one, my thesis is very simple, the first ten points listed below are the likely keys to building a winning hockey operation with a true competitive advantage. Point 11 is my neat summary of the conversations that go on at every analytics conference amongst the professional sports statisticians. None of the ten points requires an analytics guru. Point 11 suggests a competitive advantage might be achieved by not doing what everyone else is doing, if everybody is doing analytics you want to be the odd man out.

    Let me break this down to a series of simple statements. 1. Get good players. 2. Develop good players. 3. Improve good players. 4. Help good players become a cohesive unit. 5. Keep good players. 6. Develop winning team dynamic. 7. Improve coaching skills. 8. Help every one in organization to make smart choices. 9. Learn to think about hockey in way that increases live options. 10. Incorporate AI.
    11. Refuse to get caught up in some ridiculous statistical pissing contest that just boils down to a zero sum game. Please feel free to tell me where analytics in the sense we use it here to refer to sabre metrics comes into any of these points.

    1. Is obviously reliant on drafting, both amateur and professional. It is hard to miss the emerging use of demographics in improving draft outcomes, Michael Shuckers for example is doing fascinating work on the use of demographics to improve NHL draft outcomes. In the last two years there have been numerous papers on improving draft outcomes with better psychological testing and profiling. There is the use of improved cognitive skills in your scouting staff, particularly to overcome various scouting and coaching observational biases. Darryl Morey, the Houston Rockets GM is using this approach. Then there is interesting work in psycho neuro economics and behavioural economics. And while you may need a nerd to help you understand this stuff you don’t need an analytics guy. The most interesting thing I read this year was a brilliant analysis of the rates at which athletes’ skills deteriorate. Think how useful the ability to predict the decline in an aging Milan Lucic might be to making smart decisions. No technical facility was required to understand the key concepts.

    2. Is about best practice and fitting the player and the development program to each other. This is critical and makes your Director of Player Development a key role on your management team. I would argue this job has to be filled by a former player with both strong strategic skills and actual training in the psychological, neurological, and physiological development of elite athletes. Sadly, that is very rarely the case.

    3. Probably appears identical to two but this is where a specialist in human performance armed with state of the art biometrics, a strong coaching background, fitness and strength training qualifications, and state of the art biomechanics software comes in. They work with the player, the coaching staff including your conditioning coach, nutritionist, and skating instructor and all your resources to help the player maximize their performance.

    4. Is a role for a good sports psychologist. One with serious training in organizational theory, building teamwork, cultural constructs etc.

    5. This is cap management, negotiation, and psychology rolled into one. A professionally trained negotiator, with a tremendous background in game theory, arbitration, mediation and legal training would be ideal. Sounds remarkably like Tyler Dellow. Just skip the analytics part.

    6. The sports psychologist again. Good mental health and superb cognitive skills are critical to managing in complex environments with large numbers of constantly changing unknowns. Cognitive skills are trainable. Live option maximization and forced perspective changes are just two of dozens of tools in a vast toolkit that started with Edward de Bono and has grown into a little industry all its own.

    7. Mentor coaches and professional adult educators/trainers.. Your coaching staff should be continuously upgrading or they should be gone. There are some great educational programs in coaching theory and practice. Other pro sports are insisting their coaches constantly learn and expand their thinking. Hockey needs to do the same. Make all the coaching staff at all levels go to Sloan/MIT if you want to get analytics established in your program. If they fall in love with the field let them run with it. You can learn statistics and math before, during, or after you become an elite coach. The idea that analytics or tracking needs to be its own silo because the math is just so complicated mere mortals can’t hope to understand lies at the very shakey foundations of sports analytics corroding everything it touches. The math is trivial, the statistics rudimentary. You don’t need a Professor from Carnegie Mellon, you need a janitor with some time on his hands.

    8. Cognitive therapist/thinking coach/cognitive psychologists. There are lots of tools for learning to think about drafting, and everything else without bias. Teams just need to insist their staff learn and practice these techniques.

    9. What you really need is an economist with a screwy world view, a tremendous sense of humour and stunning communication skills. We could be using tools ranging from NPV, depletion curves, non-linear
    regressions, unc-unc management and Lord knows what else. The sky is the limit. But teams need someone to help them think outside the box. Hell I know one team that employs a professional clown to help them think of new options. Not a hockey team of course, but innovation and NHL are mutually exclusive.

    10. Incorporate AI into all the places in the system where efficiency matters. Leave the humans to do what we do best waste time and resources doing what we do best, wasting time. There are some very interesting examples of AI deployed in both basketball and soccer.

    11. Just exactly why given all the things the average NHL team could upgrade would any of them devote time and energy to trying to be the next Bill James? Why as fans would we want them to be the next Bill James?

  61. Woogie63 says:

    Push the coach to play Jesse

    I see Jesse as not near strong enough. The next time he plays count the number of one on one battles he loses and the number of times the puck dies on his stick.

    He looks like a gangly teenager trying to grow into body.

  62. stush18 says:

    fifthcartel:
    stush18,

    Benning gives them a nice option as a strong third-pairing RHD. I wonder if its McLellan liking the player (Russell, that is) a lot more than Chiarelli. He can’t really say he’s not interested in re-signing him.

    A Russell extension wouldn’t shock me but it’d be a little surprising and a curious decision.

    If they do, and it doesn’t have a NMC, I think it’s signals to them going to a 7-3-1 protection plan, rather than a 4-4-1. They’ll lose one of Davidson/Russell, and keep the other on the third pair.

    If they don’t sign him, they go 4-4-1 and protect Davidson. Imo

  63. Woodguy says:

    rickithebear:
    Davidson
    Klefbom
    Sekera

    WGtsk tsk

    This year’s results RB.

    Good to see you back and breaking my balls.

  64. Alpine says:

    Centre of attention,

    Benson is a very good hockey player. The Dub is an interesting league because there’s some great NHL players who have come out of there recently that didn’t quite torch the league offensively. There may be some valid concern that his offensive numbers aren’t elite, but he’s come back from injury and has been scoring at a good rate on a weak team among other things.

  65. Woodguy says:

    Chachi: Meh, without Crosby and Malkin Pittsburgh would be awful.

    A team without 2 superstars wouldn’t be good?

    The hell you say!!

    Amazing what having two to drive a team does eh?

    Also,

    For the record, when Schultz is on the ice and neither Crosby or Malkin is on the ice he has a CF% of 53.7 and a GF% of 61.9, so the rest of the team is hockeying pretty good too.

  66. Woogie63 says:

    Russell strongly prefers to play for an Alberta team and at 31 the player is not in a good negotiation place.

  67. NYCOIL "Gentleman Backpacker" says:

    Personal opinion:
    Hall, Yak, a 1st, a high 2nd for Larsson, Reinhart and a 3rd

    The asset deterioration from the vault built after 10 years of losing that began under MacT has accelerated under PC.

    Now I am not saying the team is not better. It is. And it is going in the right direction, but the team can’t afford to bleed more talent in this manner. Have to win or at least saw off future trades.

    Now Chia has won some deals…Kassian and Talbot come to mind. But there weren’t high end assets going from the Oilers. On the big trades he has lost here, as he did with Seguin in Boston.

    Team going better but running out of tradeable bullets that are meaningful. We don’t want this team to be McDavid, Draisaitl, JP and scraps. Not saying that will happen. Just mindful of the possibility.

  68. Side says:

    Woodguy: If you’re Russell why would you sign anything without a NMC to protect you from going to Vegas?

    If I were a player who was taken at the tail end of free agency who isn’t ready to retire and I wasn’t getting a lot of interest from teams and ended up signing a cheap, short term contract. I would be extremely happy to be given a longer contract for more money and wouldn’t really care where I played, as long as it isn’t in the minors or having to retire.

  69. Centre of attention says:

    Alpine:
    Centre of attention,

    Benson is a very good hockey player. The Dub is an interesting league because there’s some great NHL players who have come out of there recently that didn’t quite torch the league offensively. There may be some valid concern that his offensive numbers aren’t elite, but he’s come back from injury and has been scoring at a good rate on a weak team among other things.

    He’s looked dynamic to me tonight, I definitely see a top 9 forward there when he turns pro.

    His team is definitely holding him back.

    Vancouver had a 5-on-3 powerplay and despite everything Benson could do nobody could cash. At one point he took the puck himself directely into the slot, beating a guy and getting a good shot on net. The rebound laid there for what seemed like ages for a 5-on-3.

    If Benson played for the London Knights he would be a 100+ point junior player no doubt in my mind.

  70. Chachi says:

    Woodguy: A team without 2 superstars wouldn’t be good?

    The hell you say!!

    Amazing what having two to drive a team does eh?

    Also,

    For the record, when Schultz is on the ice and neither Crosby or Malkin is on the ice he has a CF% of 53.7 and a GF% of 61.9, so the rest of the team is hockeying pretty good too.

    Yup, just playing the stupid stupid game people seem to be fond of playing with the Oilers lineup. Yes, if you take two of the superstars (either two forwards or a forward and the goalie) off of a team they would be worse. That isn’t “analytics” its common sense. Yet, when someone tweets out that the Oilers would be awful without McDavid and Talbot they act like it is some kind of epiphany.

  71. Oilspill says:

    Centre of attention:
    Schultz now has 17 points in his last 14 games. Playing 20+ minutes a night since Letang went down and hasn’t really lost a step.

    Confidence can do wonderful things.

    And a Stanley Cup team around you.

  72. Alpine says:

    Centre of attention: He’s looked dynamic to me tonight, I definitely see a top 9 forward there when he turns pro.

    His team is definitely holding him back.

    Vancouver had a 5-on-3 powerplay and despite everything Benson could do nobody could cash. At one point he took the puck himself directely into the slot, beating a guy and getting a good shot on net. The rebound laid there for what seemed like ages for a 5-on-3.

    If Benson played for the London Knights he would be a 100+ point junior player no doubt in my mind.

    I’ve only seen a few games of him but he can definitely run a PP at that level. I saw him a couple times as a 16 year old and he was already doing then.

    I peg Benson in the sort of Tyler Ennis/Brendan Gallagher range of Western kids in terms of how good I think he could be as a pro. He doesn’t quite skate like them but he’s bigger and very well rounded and that should help the transition a bit.

    He’s also not too dissimilar to our own #14 at this stage of his career, as Eberle at the same age wasn’t considered a great skater nor was he posting eye popping offensive numbers (that came in draft +2). The raw talent isn’t quite on Eberle’s level, but I think Benson is succeeding in areas that Eberle was and somewhat remains to be weaker in.

  73. stush18 says:

    NYCOIL “Gentleman Backpacker”,

    So chiarelli won three trades? How are we bleeding? Assets? 😜

  74. stush18 says:

    Oilspill: And a Stanley Cup team around you.

    When you think about it, the Penguins team is perfect for him.

    The wingers have more than enough speed to drive play forward, and Crosby and making can button hook and find him late all day long.

    Someone here did some work showing his was actually pretty elite at that aspect of the game with the oilers.

  75. NYCOIL "Gentleman Backpacker" says:

    stush18:
    NYCOIL “Gentleman Backpacker”,

    So chiarelli won three trades? How are we bleeding? Assets?

    Turning higher value assets into lower ones. Sometimes it is necessary to take a step back in terms of value and talent in order to make the team better but one only has a certain number of bullets.

  76. Woodguy says:

    Georges,

    I think PDO has reached the status of preconceived notion for the fancystats community.

    Every goal for is a goal against. League wide PDO is 1000

    It comes down to whether PDO is capturing a skill element of the game because teams/players have varying levels of talent or whether teams/players are close enough in talent that the percentage results we observe are just random draws from a common league-wide distribution.
    If I’m reading you correctly, fancystats has decided that there is overwhelming evidence that PDO is not reproducible and therefore does not capture skill. It’s luck. And luck evens out.

    That depends.

    For the vast majority of players they will cluster very very close to 1000 over their careers as they do not drive ONSH% much, and almost no one has been found to drive ONSV%.

    Here’s and interesting little exercise someone did: http://www.pensburgh.com/2010/6/23/1531707/pdo-and-what-it-means

    You have it right in the lack of PDO’s ability to predict itself on an individual level leads people to call it “luck”

    However, there are two big caveats:

    1) Very good goalies post higher than league average SV% and those who play with them will generally put up higher PDO due to higher SV%.

    2) Elite offensive talents can drive ONSH%. Those players and the ones that play with them often will also see a slight boost in PDO.

    I’m actually looking at this on a team-wide level and figure that this year’s version of NYR has a “baseline PDO” of 1024 (all situations) which is massive. I used their goalie’s career SV% and their foward’s career SH% to figure it out.

    I think that teams who stop employing stick optional forwards will create a very slight advantage in this, but I’m not sure that if they are not good possession teams and get more shots if it will matter that much.

    You will also see 3rd pairing Dmen usually post higher than average ONSV%. This is because they play more against stick optional players. The downside is that they also play more with stick optional players on their own side, so their ONSH% tends to be lower than average.

    For now, I’d like to make an observation on the numbers you posted for Klefbom. Those numbers indicate to me that he’s probably been losing on GF% for all of those years. He’s never popped his head above 100 on PDO. Shouldn’t that be a concern? At what point should it be a concern? Never? Is there a timeline for the PDO correction? How can we be sure Klefbom, who has never posted a 100+ PDO, will post a 100+ PDO?

    That’s why I look at “expected PDO” to give me some direction.

    I also look at the goalies and their SV% that Klef has played with and its not good.

    Klefbom has essentially been a 1st pairing Dman on an very bad team with some bad goalies.

    The Oilers 5v5 SV% in the last 282 games has been .9122, which is 30th over that time (last 4 seasons including this one)

    Their PDO as a team is 982, and since Klef plays against the best offensive forwards, his ONSV% will be lower than team average.

    He could be the worst Dman in the league (he has the worst PDO among 177 Dmen who have played 2000 minutes since 13-14), but his other metrics suggest he’s not.

    Other Dmen in the bottom 15 or so: Justin Faulk, Seth Jones, Dustin Byfuglien, Jake Muzzin.

    They could all be awful defenders, but I don’t think so. Doesn’t pass the smell test.

    If you were a coach, who would you give more ice time to, all other things being equal: the guy who’s losing you games or the guy who’s winning you games (GF%)? If your job was on the line? Isn’t the NHL a great big sifting machine for results? Aren’t the PDO patterns that fancystats relies on the product of the NHL’s sifting machine, i.e., data produced by the survivors? If so, can we apply the patterns we’ve found among the successful survivors to all NHL players? Like Klefbom 143 games into his career?

    That’s a good question, but a coach will drive himself crazy trying to chase PDO.

    Here’s a relevant example from this year:

    Russell PDO per 7gms:
    1st 7: 107.7
    2nd 7: 104.1
    3rd 7: 99.83
    Last 8: 95.74

    He went from Jesus Christ Superstar to Jesus Christ Get Off The Ice in 20 or so games.

    I think that’s why coach always talk about “process, process, process” because variance can affect results sooooo much that you’d be a fool for trying to read too much into it.

    Also,

    Here’s Klef’s PDO in 4 9 game segments

    1st 9 – 102.54
    2nd 9 – 91.46
    3rd 9 – 99.30
    4th 9 – 96.20

    You’d kill yourself trying to chase that.

  77. Centre of attention says:

    Alpine: I’ve only seen a few games of him but he can definitely run a PP at that level.I saw him a couple times as a 16 year old and he was already doing then.

    I peg Benson in the sort of Tyler Ennis/Brendan Gallagher range of Western kids in terms of how good I think he could be as a pro. He doesn’t quite skate like them but he’s bigger and very well rounded and that should help the transition a bit.

    He’s also not too dissimilar to our own #14 at this stage of his career, as Eberle at the same age wasn’t considered a great skater nor was he posting eye popping offensive numbers (that came in draft +2). The raw talent isn’t quite on Eberle’s level, but I think Benson is succeeding in areas that Eberle was and somewhat remains to be weaker in.

    Benson is a much, much, MUCH different player than Eberle.

    Eberle feasts on the powerplay, and at even strength he slithers around the net like a garden snake.

    Benson is more of a utility player. Plenty of penalty kill time. He’s much more engaged physically too. Like I mentioned earlier in the thread, Benson took a nasty run at an Everette player tonight. A good three stride plastering. Got right in the middle of the scrum at the next whistle too, talking and shoving away.

    Benson’s shot is pretty average but he can really dish. Off the rush or during the cycle Benson knows where his options are, and can deliver the puck accurately and quickly.

    Heck, Benson is even taking face offs tonight. I don’t know how common that is but its another element to his game.

  78. D'oh-ilers says:

    I’m neither for nor against extending Russell, but I would be reluctant to move him this season if the Oil are still in the playoff hunt at the deadline. Regardless of what the fancy stat analysts or seen him good crowd have to say on the subject, right now Russell is performing some sort of voodoo that keeps the points coming. I’d rather see a hard playoff push and lose him this summer for nothing, than to jinx the good thing he’s got going.

    Obviously, if the team is out of contention by the deadline, move him for what you can get, but until then, keep playing the hot hand.

  79. Centre of attention says:

    Bear now has 1 goal and 1 assist tonight.

    Takes another slapper and T-bird forward bangs in rebound.

    That slap shot is going to get him paidddd

    (flipping back and forth between Giants game and T-birds game)

  80. Alpine says:

    Centre of attention: Benson is a much, much, MUCH different player than Eberle.

    Eberle feasts on the powerplay, and at even strength he slithers around the net like a garden snake.

    Benson is more of a utility player. He’s much more engaged physically too. Like I mentioned earlier in the thread, Benson took a nasty run at an Everette player tonight. A good three stride plastering. Got right in the middle of the scrum at the next whistle too, talking and shoving away.

    Benson’s shot is pretty average but he can really dish. Off the rush or during the cycle Benson knows where his options are, and can deliver it accurately and quickly.

    Heck, Benson is even taking face offs tonight. I don’t know ho common that is but its another element to his game.

    Yeah I just mean they’re similar in how they were highly touted but slightly risky prospects that were ranked and drafted in the same range of the draft.

    It was more of a compliment to Benson as he is sometimes knocked for a lack of elite offense, but even Eberle hadn’t quite shown he could be a dominant scorer until in his last season of junior.

    They have similar trajectory as prospects, similar natural talent levels, and aren’t far off in size (Benson is even a little bigger!). So what I was initially getting that is while they play differently, they have a lot of the same arrows at the same age, and that should give Benson doubters some room to breathe. I have seen a fair amount of sentiment about him that’s almost unfairly skeptical.

  81. Woodguy says:

    VOR:
    LT, I am not sure you want an analytics guy at all. Yes, I know that is heresy. But hear me out.

    For those who don’t want to read a wall of text, and I have included one, my thesis is very simple, the first ten points listed below are the likely keys to building a winning hockey operation with a true competitive advantage. Point 11 is my neat summary of the conversations that go on at every analytics conference amongst the professional sports statisticians. None of the ten points requires an analytics guru. Point 11 suggests a competitive advantage might be achieved by not doing what everyone else is doing, if everybody is doing analytics you want to be the odd man out.

    Let me break this down to a series of simple statements. 1. Get good players. 2. Develop good players. 3. Improve good players. 4. Help good players become a cohesive unit. 5. Keep good players. 6. Develop winning team dynamic. 7. Improve coaching skills. 8. Help every one in organization to make smart choices. 9. Learn to think about hockey in way that increases live options. 10. Incorporate AI.
    11. Refuse to get caught up in some ridiculous statistical pissing contest that just boils down to a zero sum game. Please feel free to tell me where analytics in the sense we use it here to refer to sabre metrics comes into any of these points.

    1. Is obviously reliant on drafting, both amateur and professional. It is hard to miss the emerging use of demographics in improving draft outcomes, Michael Shuckers for example is doing fascinating work on the use of demographics to improve NHL draft outcomes. In the last two years there have been numerous papers on improving draft outcomes with better psychological testing and profiling. There is the use of improved cognitive skills in your scouting staff, particularly to overcome various scouting and coaching observational biases. Darryl Morey, the Houston Rockets GM is using this approach. Then there is interesting work in psycho neuro economics and behavioural economics. And while you may need a nerd to help you understand this stuff you don’t need an analytics guy. The most interesting thing I read this year was a brilliant analysis of the rates at which athletes’ skills deteriorate. Think how useful the ability to predict the decline in an aging Milan Lucic might be to making smart decisions. No technical facility was required to understand the key concepts.

    2. Is about best practice and fitting the player and the development program to each other. This is critical and makes your Director of Player Development a key role on your management team. I would argue this job has to be filled by a former player with both strong strategic skills and actual training in the psychological, neurological, and physiological development of elite athletes. Sadly, that is very rarely the case.

    3. Probably appears identical to two but this is where a specialist in human performance armed with state of the art biometrics, a strong coaching background, fitness and strength training qualifications, and state of the art biomechanics software comes in. They work with the player, the coaching staff including your conditioning coach, nutritionist, and skating instructor and all your resources to help the player maximize their performance.

    4. Is a role for a good sports psychologist. One with serious training in organizational theory, building teamwork, cultural constructs etc.

    5. This is cap management, negotiation, and psychology rolled into one. A professionally trained negotiator, with a tremendous background in game theory, arbitration, mediation and legal training would be ideal. Sounds remarkably like Tyler Dellow. Just skip the analytics part.

    6. The sports psychologist again. Good mental health and superb cognitive skills are critical to managing in complex environments with large numbers of constantly changing unknowns. Cognitive skills are trainable. Live option maximization and forced perspective changes are just two of dozens of tools in a vast toolkit that started with Edward de Bono and has grown into a little industry all its own.

    7. Mentor coaches and professional adult educators/trainers.. Your coaching staff should be continuously upgrading or they should be gone. There are some great educational programs in coaching theory and practice. Other pro sports are insisting their coaches constantly learn and expand their thinking. Hockey needs to do the same. Make all the coaching staff at all levels go to Sloan/MIT if you want to get analytics established in your program. If they fall in love with the field let them run with it. You can learn statistics and math before, during, or after you become an elite coach. The idea that analytics or trackingneeds to be its own silo because the math is just so complicated mere mortals can’t hope to understand lies at the very shakey foundations of sports analytics corroding everything it touches. The math is trivial, the statistics rudimentary. You don’t need a Professor from Carnegie Mellon, you need a janitor with some time on his hands.

    8. Cognitive therapist/thinking coach/cognitive psychologists. There are lots of tools for learning to think about drafting, and everything else without bias. Teams just need to insist their staff learn and practice these techniques.

    9. What you really need is an economist with a screwy world view, a tremendous sense of humour and stunning communication skills. We could be using tools ranging from NPV, depletion curves, non-linear
    regressions, unc-unc management and Lord knows what else. The sky is the limit. But teams need someone to help them think outside the box. Hell I know one team that employs a professional clown to help them think of new options. Not a hockey team of course, but innovation and NHL are mutually exclusive.

    10. Incorporate AI into all the places in the system where efficiency matters. Leave the humans to do what we do best waste time and resources doing what we do best, wasting time. There are some very interesting examples of AI deployed in both basketball and soccer.

    11. Just exactly why given all the things the average NHL team could upgrade would any of them devote time and energy to trying to be the next Bill James? Why as fans would we want them to be the next Bill James?

    This whole post is about analytics.

    As you know analytics is coming to conclusions or hypothesises (sp?) through the analysis of data and pretty much all of your points are exactly that.

    As far as what guys like Tyler or Sunny Mehta or Sam Ventura etc do, from what I’ve seen its a lot more than the simple shot metric analysis that we kick around here or I do on my blog.

    That said having a guy on staff who does good simple shot metric analysis and listening to him would have, at the very least, saved MacT and eventually Nonis from the Clarkson contract, Chiarelli and Nill from the Korpikoski contracts (both Peter and Jim are paying him this year, good times for him)

    Your first point will always be the most important one in hockey – get good players.

    This includes avoiding bad ones and given how many below replacement NHL players get contracts, or below average NHLers get good contracts each year, I’d say the NHL as a whole is still miles away from “everyone does analytics so there is no advantage”

  82. Alpine says:

    D'oh-ilers:
    I’m neither for nor against extending Russell, but I would be reluctant to move him this season if the Oil are still in the playoff hunt at the deadline. Regardless of what the fancy stat analysts or seen him good crowd have to say on the subject, right now Russell is performing some sort of voodoo that keeps the points coming. I’d rather see a hard playoff push and lose him this summer for nothing, than to jinx the good thing he’s got going.

    Obviously, if the team is out of contention by the deadline, move him for what you can get, but until then, keep playing the hot hand.

    He is not performing any sort of voodoo, he’s just a useful player who’s not sticking out in too many ways, either good or bad. I see an argument for having as much defensive depth as you can for the stretch run, but I don’t believe his inclusion is causing anything magical to happen. McDavid does that himself.

  83. Centre of attention says:

    Alpine,

    Talking box cars, yes I see what you mean.

    I think Benson will be a short study in the AHL.

  84. Centre of attention says:

    Benson wins the face off in overtime, helps set up Ty Ronning for the OT winner.

    MacT is coaching this team I swear. Benson is now a center apparently.

    Finishes the night with 2 assists, now 39 pts in 30 games

  85. JimmyV1965 says:

    With the expansion draft coming up, there’s no GM in the league giving Russell a NMC. As much as I lack confidence in Chia right now, I would be shocked if he gave Russell a NMC. It just doesn’t make sense to give a 4D a NMC.

  86. NYCOIL "Gentleman Backpacker" says:

    Burns with a beauty of an OT winner. 3 point game Sharks over Ducks not helping the Oilers any, but it is fun to scoreboard watch again.

  87. LadiesloveSmid says:

    Cam Talbot is a 53.3% CF away from Russell (45.3% with), Gustavsson 53.8% without (46.1% with). Some of that is competition but come on. Get rid of him.

  88. Centre of attention says:

    NYCOIL “Gentleman Backpacker”:
    Burns with a beauty of an OT winner. 3 point game Sharks over Ducks not helping the Oilers any, but it is fun to scoreboard watch again.

    This game against the Kings Thursday is going to be a doozey isn’t it.

  89. russ99 says:

    Can we at least agree that CF% doesn’t tell the whole story, especially for defensemen, and that we shouldn’t expect the team to make decisions solely based on that metric? Sure it’s the best predictor for goals that we have now, but we have to think about offense and defense if we want to see playoff dates, not just offense.

    I get the idea of using advanced metrics to get a clearer picture, but why are we using that one as the hill to die on, when others have more context, and again, that old Earl Weaver vs. Whitey Herzog thingy…

    So here’s mine:

    1. Send Puljujarvi to Bakersfield, it’s really the best place to send him to find his offense, He’s not going to find it without Connor, and isn’t finding it with Connor either, Russia is really bad idea, he needs to get acclimated to North American hockey. Bakersfield may not be the most welcoming city, and there may be apprehension with the coaching staff there, but let him play with Lander while he’s there.

    2. Keep Kris Russell right where he is, and no contract offers until the expansion draft, if we can find someone better, let him go, if not, he’s an NHL defensman and has arguably done the most to limit opposition chances after entry, though Larsson seems to be coming around too.

    3. Matt Benning has had a nice start, but he’s still a rookie, let’s not repeat old mistakes pushing a kid into a role that he’s not ready for, especially since the only basis for doing so is CF%. Let him develop in a third line role and we’ll be happier with what we have next season.

    4. Enough with the line blender every single game, find groups that work and stick with them until someone either needs a break or gets hurt,

    5. Play Gustafsson more if you’re going to keep him otherwise replace him.

    6. Find a NHL right winger with a good shot.

    7. Beg borrow or steal to get takers for MacT’s free agents (Pouliot, Fayne) so we have more salary flexibility for expansion trades and free agency.

  90. Georges says:

    Woodguy,

    Here’s and interesting little exercise someone did: http://www.pensburgh.com/2010/6/23/1531707/pdo-and-what-it-means

    Because Dellow’s site is down, I can only go by the description of his work on the pensburgh site.

    Dellow took the top 20 and bottom 20 PDO teams for the first quarter of a season’s games going back to 2003-2004 and then looked at their PDO results for the remainder of the season. He found that the top 20 teams did worse and the bottom 20 teams did better and the average PDO of the two groups got pretty close.

    1. PDO is going to vary more in smaller samples (i.e., more variance in 20 games than in 62 games).

    2. The top 20 PDO results in the first quarter of games are likely to be the top 20 PDO results in the entire dataset (similar observation for the bottom 20 PDO results).

    3. Because of the way Dellow set up the experiment, it’s near certain that the top 20 will do worse because he chose the group that can’t do any better (similar observation for the bottom 20).

    4. He reported that the sh% of the top 20 in the remaining 3 quarters was .087 and the sh% of the bottom 20 was .082.

    5. Given that there are 20 teams in each group and each team plays 60 games in the sample with, let’s say, 30 shots on average per game, the data supports the conclusion that the two groups of teams differ in their shooting percentage in the last 3 quarters of their seasons.

    6. So, I would be open to the idea that teams that strongly outperform on sh% early in the season are likely to be better on sh% than teams that badly underperform on sh% early in the season.

  91. New Improved Darkness says:

    VOR:
    The most interesting thing I read this year was a brilliant analysis of the rates at which athletes’ skills deteriorate.

    I filed your post under “VOR meh analytics manifesto”. Nice contrarian synopsis.

    However, please name the “thing”, or what he answers to, whether it’s URL or dog biscuits (or worse).

  92. Georges says:

    Woodguy,

    You have it right in the lack of PDO’s ability to predict itself on an individual level leads people to call it “luck”

    However, there are two big caveats:

    1) Very good goalies post higher than league average SV% and those who play with them will generally put up higher PDO due to higher SV%.

    2) Elite offensive talents can drive ONSH%. Those players and the ones that play with them often will also see a slight boost in PDO.

    These two big caveats are so big they’re enormous.

    When hockey fans talk about regressing to the mean, I think they’re thinking that all players have similar talent levels and their performances are random draws from a common performance distribution. This is what PDO is luck implies. If you accept that view, you don’t think that percentage data is going to tell you anything. It’s just noise. You don’t use it in your decision-making.

    The two big caveats show you that sometimes talent gaps are wide enough even at the NHL level that they give you a sustained edge in performance. These exceptions are really valuable if you can detect them before they’re apparent to everyone (i.e., get good players). And it’s important to value these edges appropriately once you believe you’ve detected them (i.e., keep good players).

  93. Georges says:

    Woodguy:
    Georges,

    That’s a good question, but a coach will drive himself crazy trying to chase PDO.

    Here’s a relevant example from this year:

    Russell PDO per 7gms:
    1st 7: 107.7
    2nd 7: 104.1
    3rd 7: 99.83
    Last 8: 95.74

    He went from Jesus Christ Superstar to Jesus Christ Get Off The Ice in 20 or so games.

    I think that’s why coach always talk about “process, process, process” because variance can affect results sooooo much that you’d be a fool for trying to read too much into it.

    Also,

    Here’s Klef’s PDO in 4 9 game segments

    1st 9 – 102.54
    2nd 9 – 91.46
    3rd 9 – 99.30
    4th 9 – 96.20

    You’d kill yourself trying to chase that.

    Do you think there’s any relationship between time-on-ice and PDO? The coach controls time-on-ice. If he’s not chasing PDO, we should see no relationship. If he’s chasing PDO and he’s no good at it or it’s hopeless, we should see no relationship. Correct?

  94. Georges says:

    Woodguy,

    With Woodmoney, do the three groups have the same SF60 on average? When I ran a quick calculation, it seemed that the numbers were close and that the gritensity crowd couldn’t push back against the elites. But I didn’t separate the goalies. Illumination would be appreciated.

  95. Gret99zky says:

    Centre of attention: This game against the Kings Thursday is going to be a doozey isn’t it.

    Will the Oilers be rested or rusty?

    We wait.

  96. Georges says:

    Woodguy,

    On Oscar, I don’t know what to think. You had a very nice post about Schultz finding his place in Pittsburgh. Maybe Klef needs 3rd pair minutes. He’s played against tough opponents in the past with an inexperienced defense partner and a shaky goaltender and he’s come up short. Now he’s playing with an experienced partner and a capable goaltender and the results are still coming up short. And the forward corps is better. But all eyes are on Larsson because of the anger around the Hall trade.

  97. J-Bo says:

    Middle of the night indigestion gave me time to read through all the comments. Nice thread! The Klefbomb trade talk made me wonder if we would be better off if we could trade Klefbomb for that RHD we need. For instance, is this better:

    Sekera – Larsson
    Nurse – Trouba
    Davidson – Benning

    This is misleading I suppose as it isn’t possible, but it also makes me wonder what Klefbom’s value is on the market. Could he bring in a legit RHD like Trouba at this point?

  98. hunter1909 says:

    Dubnyk, Gagner, Schultz… all starring for other NHL teams than the Edmonton retarded Oilers.

    Taylor Hall will be hell bent to destroy the Oilers in a brace of upcoming New Year’s games.

    MacT’s “challenging” Petry needs it’s very own wing in Toronto’s Hockey Hall of Fame.

    On the bright side: Connor McDavid. Then, Connor McDavid, and Connor McDavid. As well as Connor McDavid, Connor McDavid, and Connor freaking McDavid.

    Also, Draisraitl looks to be our Oiler’s own version of Jean Beliveau. Also, the Oilers currently sit in solid playoff position. When exactly was that happening?

    Happy New Year from the Hunter1909 Death March – which will return in 2017 bigger than better than ever imagined before.

  99. spoiler says:

    hunter1909,

    The one bright side is that during trade negotiations Chia can point to their success when getting lowballed because “Oiler losering”.

  100. Centre of attention says:

    hunter1909,

    You had me down for 90 points right?

    I still feel like that’s a solid prediction 3 months in.

  101. Woodguy says:

    Georges: Do you think there’s any relationship between time-on-ice and PDO? The coach controls time-on-ice. If he’s not chasing PDO, we should see no relationship. If he’s chasing PDO and he’s no good at it or it’s hopeless, we should see no relationship. Correct?

    Probably not.

    The players who play the most tend to have the highest SH%, so it *might* be smidge higher (when controlled for SV%)

    These players also tend to play the most against the players with the highest SH%, so it might wash out.

  102. Woodguy says:

    Georges,

    . Now he’s playing with an experienced partner and a capable goaltender and the results are still coming up short

    When I showed Oscar’s PDO per 9 games, he was well over 100 for one section, way, way below for another, about even in another, and then slightly below again.

    In front of the same goalie, Russell was way, way above in one, above in a second, way below in 2.

    You’re chasing noise imo.

  103. Woodguy says:

    russ99,

    Can we at least agree that CF% doesn’t tell the whole story, especially for defensemen, and that we shouldn’t expect the team to make decisions solely based on that metric? Sure it’s the best predictor for goals that we have now, but we have to think about offense and defense if we want to see playoff dates, not just offense.

    I think everyone here can agree with that.

    We also see with teams like PIT and CBJ that the best defence can be not letting the other guys have the puck much.

  104. PhrankLee says:

    Hopefully they aren’t buyers at the deadline.

    Hopefully they aren’t sellers at the deadline.

    I would be happy if they let this group fight it out with the exception of Hendricks.

    If he is up to speed by the deadline I bet you could get a second for him.

    It would be sad to see him go. We desperately need picks.

    I did not have them making the playoffs this year but have seen great progress toward a winning product. The goal differential, so far, is enough to bring tears to my eyes.

    I hope they work the kinks out of Klefbom pronto because I have not liked the timing of his mistakes…way more than the mistakes themselves.

  105. Woodguy says:

    Georges,

    If your hypothesis is that “players control their ONSH% and ONSV% and therefore PDO is an indication of player ability, the first thing you should do is look at how your hypothesis ranks players.

    Here’s the top 10 PDO players over the last 4 seasons (including this one) with the seasons aggregated:

    GUENIN, NATE 102.6
    DUMBA, MATT 102.4
    MILLER, KEVAN 102.1
    MAATTA, OLLI 101.9
    SPURGEON, JARED 101.8
    MCDONAGH, RYAN 101.7
    LINDHOLM, HAMPUS 101.7
    MARKOV, ANDREI 101.6
    HOLDEN, NICK 101.6
    RUSSELL, KRIS 101.6

    The best Dman is a AHL journeyman who stuck with the Avs for the years Varly was red hot?

    Nick Holden also played most of his hockey in front of Varly.

    Etc.

    Does it pass the smell test off the hop?

  106. Woodguy says:

    Georges:
    Woodguy,

    These two big caveats are so big they’re enormous.

    When hockey fans talk about regressing to the mean, I think they’re thinking that all players have similar talent levels and their performances are random draws from a common performance distribution. This is what PDO is luck implies. If you accept that view, you don’t think that percentage data is going to tell you anything. It’s just noise. You don’t use it in your decision-making.

    The two big caveats show you that sometimes talent gaps are wide enough even at the NHL level that they give you a sustained edge in performance. These exceptions are really valuable if you can detect them before they’re apparent to everyone (i.e., get good players). And it’s important to value these edges appropriately once you believe you’ve detected them (i.e., keep good players).

    I think the best things to do are:

    1) Have Carey Price in net (heh)
    2) Hire high career SH% players and stop using checkers

  107. Professor Q says:

    It would be pretty crazy indeed if Minnesota and Columbus met in the Cup Finals.

    Gagner vs. Dubnyk…

  108. norm_klassen says:

    Woodguy:
    Georges,

    If your hypothesis is that “players control their ONSH% and ONSV% and therefore PDO is an indication of player ability, the first thing you should do is look at how your hypothesis ranks players.

    Here’s the top 10 PDO players over the last 4 seasons (including this one) with the seasons aggregated:

    GUENIN, NATE102.6
    DUMBA, MATT102.4
    MILLER, KEVAN102.1
    MAATTA, OLLI101.9
    SPURGEON, JARED101.8
    MCDONAGH, RYAN101.7
    LINDHOLM, HAMPUS101.7
    MARKOV, ANDREI101.6
    HOLDEN, NICK101.6
    RUSSELL, KRIS101.6

    The best Dman is a AHL journeyman who stuck with the Avs for the years Varly was red hot?

    Nick Holden also played most of his hockey in front of Varly.

    Etc.

    Does it pass the smell test off the hop?

    </block quote
    Most those player's play with Price, Dubnyk; and Lundquist as there goalies. U know the best of the best goslies of the nhl.

  109. Woodguy says:

    norm_klassen,

    Most those player’s play with Price, Dubnyk; and Lundquist as there goalies. U know the best of the best goslies of the nhl.

    Yessir, add to that Guenin, Holden and Barrie (just outside the top 10) played with Varly when he single handedly won them the Central.

  110. Roughneck says:

    Woodguy:
    Georges,

    Here’s some very good reading on whether or not Dman effect SV%

    https://hockey-graphs.com/2014/07/07/defensemen-still-have-no-sustainable-control-over-save-percentage/

    Good read.

Leave a Reply

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