THE SECOND ROUND

A year ago, I wrote about Stu MacGregor’s second rounds. The basic premise was this: it’s almost impossible to have a rational conversation about this era’s drafting for multiple reasons. It’s a game of serve and volley, and frankly no one wins. If I say “MacGregor’s 2008 draft had no second or third round” the ball comes back over the net “STILL no second-round picks who have played over 100 games in the NHL” and on it goes. If I argue “you have to wait five years” it comes back “Lander has had five years” and we ignore the unusual progression of his career and the Oilers part in it. If I argue that the Oilers patience in regard to Martin Marincin appears to be paying dividends, well, actually no one is countering that so perhaps we’re making progress.

I’m hopeful we can have a rational conversation about the Edmonton Oilers in the second round 2008-13. No grand statements on either side, no claims of victory, just an honest assessment of the way things are, the good and bad arrows, and a discussion in regard to useful players coming out of this area of the draft.

ROY33

In order to do this, we absolutely have to agree on some things:

  • We can’t judge a draft weeks after it happens.
  • We can count arrows, and they are a good indicator.
  • We cannot expect all of the picks to succeed, that isn’t reasonable.
  • We have to come to some kind of conclusion about a line in the sand and agree to it.

In order to ensure we all enter into this discussion without bias, and in order to make sure we don’t end up back in the tired old arguments about the difficulty of deciding between Taylor and Tyler for #1, or Jordan Eberle at #22, or Oscar Klefbom at #19, we’ll wall off the first round Poe-style and walk away with our own beliefs–secure in the knowledge that a several year discussion has left us no closer to agreement. So, let’s see if we can come to an agreement about selections 31-60.

A PROPOSAL

I propose we adopt a line in the sand, we can call it the “Cullen line” after Scott Cullen of TSN. He estimated the success rate (100 NHL games or more) to be 28% back in 2009. Let’s adopt 30% as the “line” because it’s easier to remember, and let’s make “100″ closer to two seasons and use 150 games. That way, for our purposes, even a player like Anton Lander (who is in the study) can’t be claimed as a victory by the pro side. Fair? I’d love your input here, because the idea is to be fair, not bend over backwards to be fair but “I’m tuning my ’53 flathead Ford engine and need top dead center” fair.

OILERS SECOND ROUND 2008-2013

That set, let’s move on to the Oilers during the MacGregor era (2008-2013). We’re looking at one in three being a success (average) and anything more than that above average. I’ve increased the level of difficulty from one year ago, when it was one in four. The Oilers have selected seven players in Round 2 during the MacGregor era:

  • 2008: No second round selections
  • 2009: Anton Lander
  • 2010: Tyler Pitlick
  • 2010: Martin Marincin
  • 2010: Curtis Hamilton
  • 2011: David Musil
  • 2012: Mitchell Moroz
  • 2013: Marco Roy

No player in the group of 7 has reached the level (now 150 games) we’ve established as the line in the sand. And 30% of seven players is two NHL players of 150 games or more from this group. A year ago, I looked at each player and gave them a grade based on arrows:

  • Very bad arrows: Curtis Hamilton
  • Bad arrows: Tyler Pitlick
  • Lukewarm arrows: Marco Roy, Mitchell Moroz, David Musil
  • Good arrows: Martin Marincin
  • Very good arrows: Anton Lander

2014 SUMMER

  1. Anton Lander 94 games (27 in 2013-14)
  2. Martin Marincin 44 games (all games in 2013-14)
  3. Tyler Pitlick 10 games  (all games in 2013-14)

Two more players make their NHL debut, three players of the seven working toward what is now the 150 game goal. Remember, AVERAGE is two players, so all three would have to exceed 150 games to make this group above average. The good news is that some players have improved their arrows:

  • Very bad arrows: Curtis Hamilton
  • Bad arrows: Marco Roy
  • Lukewarm arrows: David Musil, Mitchell Moroz
  • Good arrows: Anton Lander, Tyler Pitlick
  • Very good arrows: Martin Marincin

Several players took a step forward this season, and two men from the 2010 draft played well enough to earn their first NHL look. Let’s take these one at a time to close:

  • L Curtis Hamilton—Craig MacTavish has far more patience than I thought, or Todd Nelson argued for him, one of the two.
  • L Marco Roy—Disappointing season due to injury. Helluva start to the year, though. Needs to avoid being Pouliot 2.0.
  • R Tyler Pitlick—Good and bad. Made his NHL debut, looked effective and scored a goal. Then he got hurt. Oy. Todd Nelson said he was just as effective on return, his final 7 NHL games didn’t kick it like his first three. May have an NHL job this time next season.
  • C Anton Lander—He enjoyed a strong AHL season and earned a one-way deal.
  • D David Musil—Enjoyed a strong first year pro, I have him in lukewarm category because he’s one year in to what is likely a three or four year process. Good arrows for sure, but this is a marathon not a sprint.
  • L Mitchell Moroz—Blossomed in his final junior season, and like Musil I’ve placed him in lukewarm despite the fine season. The men who have come before Moroz find this leap (WHL to AHL) very challenging, so it’s best we are cautious about Moroz this coming season.
  • D Martin Marincin—Arrived in the NHL and was a revelation. I’d bet on him playing all year in the NHL in 2014-15.

A year ago, this group had a total of 67 NHL games in the bank (Anton Lander) and no clear candidate to push through to the big leagues. This season? Lander has a one way deal, Marincin has a very good chance to make the roster, and Tyler Pitlick is also in the conversation. Musil completed a solid AHL debut season, Moroz is on the way.

MARINCIN

kings4.jpg

I think we have to start a new segment in this year’s look, and ask some more questions. Martin Marincin looks like he may be a rate item—a player who fills a significant roster role who was taken outside the first round. Anecdotal information suggests to me that the percentage of players taken 31-60 who end up as top 4 defensemen or top 9 forwards (or starting goalies) is less than 15%. That means that in a three-year period of second rounds (90 picks), we can expect a baker’s dozen to fall in that range. Not many. If we use 15% (and that’s an arbitrary number, this isn’t proven, but I suspect it’s high if anything) as our line in the sand, Edmonton should expect 1 player from this group of 7 to develop. If Marincin develops into such a player and the other six do not, Edmonton will have produced an average number. Fair?

If you don’t like the 15% number we’ll do the math (but it has to be recent, it looks like NHL teams are getting better at second rounds, or maybe it’s just Chicago and Columbus), but Cullen’s work suggests finding a legit contributor is a fairly rare item. The 2009 draft’s second round (as an example) boasts Ryan O’Reilly and Tomas Tatar, plus some others like Jakob Silfverberg.

Either way, Marincin does seem to be trending in a good direction, a direction the other six players in the Oilers group haven’t implied or established.

LOWDOWN WITH LOWETIDE

At 10 this morning, TSN 1260, we hit the air with two hours of us! Sean Fleming will be on just after 10, he’s the head coach of Canada’s U17 soccer team, who represent the future! I’ll ask his opinion of this brilliant World Cup just over, and we’ll discuss the impressive success of his young squad and what it means for the future.

I’ll also have Alan Hull from Copper and Blue and Harrison Mooney from Puck Daddy on the show, plus a guest or two are hovering so we’ll see how things roll. @Lowetide_ or 10-1260 for comments and questions, thanks!

written by

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

65 Responses to "THE SECOND ROUND"

  1. Pouzar says:

    Wow…heckuva Monday morning post LT.

    I am late to the party on this ( I was in the Bahamas trying to stop drinking) but this
    should absolutely be a subscription site. A brother’s gotta paid!!!!!!!

  2. Wes Mantooth-11 says:

    Still bug’s me that the Oilers took Musil instead of Jenner, I hated the pick then and still hate the pick today.

    My choice was Jenner then Rattie.

  3. supernova says:

    Love this Stuff LT, excellent post

    I do think your level of 30% and 150 games by year 5 is very tough line in the sand though.

    Main reason is draft trends have teams picking more Forwards in round 1 and less goalies and D.

  4. maudite says:

    Musil and moroz are the two that stick out to me.

    Not hindsight saad and Jenner just came across as better bets at the time.

    Hindsight moroz doesn’t look so bad but I still think don’t like the pick at that spot. Too early to reach and even factoring his overage whl year, still doubt he possess the ceiling you would hope to get at that spot…

  5. maudite says:

    I like lander pol and I think in a proper organization he likely turns into a useful player. 2009 problem was not the draft. It was the subsequent lack of sensible development curves.

  6. gvblackhawk says:

    I’m confused. At the top of the article you state ” We can’t judge a draft weeks after it happens”; however, yesterday’s article was “Who won the 2014 NHL draft?”, which was two weeks ago.

  7. Lowetide says:

    gvblackhawk:
    I’m confused.At the top of the article you state ” We can’t judge a draft weeks after it happens”; however, yesterday’s article was “Who won the 2014 NHL draft?”, which was two weeks ago.

    Judging is something you do final final, there’s simply no way to know. It’s still fun to have a first blush look at things, which was the point of yesterday’s post. We’re in the fun days, folks, this is us for the next while.

  8. Lowetide says:

    maudite:
    Musil and moroz are the two that stick out to me.

    Not hindsight saad and Jenner just came across as better bets at the time.

    Hindsight moroz doesn’t look so bad but I still think don’t like the pick at that spot.Too early to reach and even factoring his overage whl year, still doubt he possess the ceiling you would hope to get at that spot…

    Moroz turned 20 in May. He didn’t play an overage year in the WHL. He wasn’t eligible to play in the AHL last fall.

  9. FastOil says:

    gvblackhawk:
    I’m confused.At the top of the article you state ” We can’t judge a draft weeks after it happens”; however, yesterday’s article was “Who won the 2014 NHL draft?”, which was two weeks ago.

    The tone I took was who made the best decisions at the table in 2014. We know enough about prospects to have a reasonable idea of how to maximize success. Yet as LT says it’s still a crap shoot outside of top 5.

  10. gvblackhawk says:

    FastOil: The tone I took was who made the best decisions at the table in 2014. We know enough about prospects to have a reasonable idea of how to maximize success. Yet as LT says it’s still a crap shoot outside of top 5.

    “Who made the best decisions at the table in 2014″ sounds like alternative wording for ‘who took the best players at the 2014 draft according to an arbitrary list’. And there was a grade assigned — you can’t get much closer to ‘judging’ than that.

    It’s a contradiction. You can’t have it both ways.

  11. Bank Shot says:

    Looking through the other teams in the NHL there are clearly some who have nailed it in the second round in the McGregor era:

    LA-Voynov,Clifford, Toffoli, Pearson
    COL-O’Reilly
    CHI-Saad
    NSH-Josi
    NYR-Stepan
    OTT-Wiercioch,Silferberg, Lehner

    There’s some that didn’t really qualify like the Flyers and Montreal who had a real dearth of second rounders in this time period.

    Then there are a lot in the middle. Based on points produced from the second round I would guess Edmonton is somewhere in the 20-25 range.

    If Marincin turns out better then anticipated, it probably puts the Oilers in the average range. Right now I think they are trending a bit below.

  12. Hammers says:

    The problem I see is if a 1st has to hit 200 games and a 2nd 150 what do you expert from the #3 ‘s or 4′s .For me once a 2nd hits 125 that’s basically a season and a half probably over a 3 year period I would say Success . So 40 games x3 years average . Just a suggestion .

  13. rickithebear says:

    The cummulative % for our 2nd round picks based on Michael Schuckers chance of a 200gm player by pick chart is

    Lander #40 20.9%
    Pitlick #31 25.5%
    Marincin #46 18.9%
    Hamilton #48 18.6%
    Musil #31 25.5% tambo total 109.4%
    ———————————— MacT VP
    Moroz #32 24.6%
    M-O Roy #56 17.8% MacT total 42.4%

    total 151.8%
    Do we get 2 200gm players from the group?

    the beauty is 3rd round
    Hesketh #71 17.1%
    Abney #82 15.1%
    Martindale #61 17.5%
    Perhonen #62 17.6%
    Ewanyk #74 16.5% tambo total 83.8%
    ———————————–MacT VP
    Khaira #63 17.6%
    ZHarkov #91 12.8%
    Yakimov #83 14.9%
    Slepyshev #89 14.7% total 60.0%

    This year:
    Draisatl #3 82.6%
    Lagesson #91 12.8%
    Nagelvoort #111 9.2%
    Coughlin #130 8.5%
    Vesel #153 7.9%
    bouchard #183 6.6%
    total 127.6%

    We are in good hands with SMB/ MacT

  14. Lowetide says:

    gvblackhawk: “Who made the best decisions at the table in 2014″ sounds like alternative wording for ‘who took the best players at the 2014 draft according to an arbitrary list’.And there was a grade assigned — you can’t get much closer to ‘judging’ than that.

    It’s a contradiction.You can’t have it both ways.

    Actually, I can. It’s not impossible to expect the reader to take “instant analysis” as exactly that, while also understanding that it takes years to correctly evaluate the draft.

    I don’t think that’s asking too much of the audience.

  15. SinceTheWHADays says:

    Bank Shot,

    In the case of a lack, or absense of a pick…do we have to look at what that pick turned into? If a team trades the pick for a star….

  16. SpotTheLoon says:

    Pouzar,

    I agree Pouzar. The content that is produced for us long suffering Oilers fans is incredible. Thanks so much for all your efforts, Allan.

    I know that this won’t become a pay site but I want to make a monthly contribution to Allan. I know the idea of a recurring monthly donation was bandied about at one point. In any event, I was wondering what others thought was an appropriate amount to contribute monthly? I also would like to recommend that others also do this. Allan has worked hard producing a lot of valuable information for us for a lot of years. I understand he doesn’t do it for the money. That said, his work is certainly “top drawer” and he deserves to reap some reward for all the enjoyment his writings have brought all of us.

    Thanks again, Allan.

  17. SinceTheWHADays says:

    rickithebear,

    Ya I agree. It seems the focus on skill has disapated the walk-about coke machine picks somewhat. But small players, no matter what the skill are really tough to actualize as NHLers who can help win the cup. Mind you, that Kyle Kessy move is still weird. Jeez this is a hard exercise LT.

  18. Hammers says:

    As of this year if a player like MM plays the whole year you must say his made it . In fact any player no matter where his drafted ends up with a full season after a half season you have to figure they are here .I’m sure there are a few that fall of the bus but if we look at 750+ players in the league including injuries that should be a sign . The other point is your 150 games can’t relate the same for forwards , defencemen & goalies . Can you rethink your line in the sand .

  19. OilClog says:

    Moroz literally has to score 50 and land 200checks, to not be viewed as a terrible pick.. I think he ends up making a career for himself.

  20. SinceTheWHADays says:

    SpotTheLoon,

    Ya I’m in too. How many regular contributors? If he’s got 100..$20 bucks apiece would be $24 000/ year. Not bad. The problem is if we pay him too much he might retire.

  21. Well Oiled and Enthusiastic says:

    When I look at odds stated like you have, it really makes me wonder if a more efficient process for player procurement is to hold on to your round one and round two picks and invest significantly in understanding the best prospects available in the top 60. This would narrow the focus of player review and observation by the scouting team (thereby potential improving outcomes) versus taking such a wide swath review on other prospects that are clearly lower picks. Amp up your focus on the cream of the selections.

    Then, trade your round 3-10 selections EVERY time for emerging prospects in other organizations or established NHL’ers that fit a need on your team but may need to be moved by another team for whatever reason. On the prospect side, its important to remember that every organization is unique in where and how a prospect might fit so there has to be at any time real treasure to be mined because of circumstance. By example, if you are a young centre prospect in Colorado’s system right now, you may be caught in the numbers and circumstance. Fedun is a great example from Edmonton. These prospects are more ‘known’ then a draft selection and I think must improve your odds of locating an NHL viable player. If it improves your success rate by 1% versus drafting, you should do it no? I may be off the mark here, but I understood rd 3-10 typically has a 10% success rate….or something close to that. That seems a mugs game.

    For established NHL’ers, there seems to be inordinate value placed on mid range picks versus proven NHL capability. Lower picks seem to have strong currency value. I point to Hemsky as an example. The picks we received have low odds of creating a player and the time value to potential contribution (say 5 yrs) needs to also be factored in. I would do these trades all day long to round out your roster. This seems to be a more relevant tactic in the cap era because the cap constraints create these types of situations (Hemsky) on a regular basis.

    I’m curious about perspective on this as the historical success rates on drafting beyond the second round seems to dictate you use the pick for something else with better odds….even it is only marginal.

  22. SpotTheLoon says:

    SinceTheWHADays:
    SpotTheLoon,

    Ya I’m in too.How many regular contributors? If he’s got 100..$20 bucks apiece would be $24 000/ year.Not bad. The problem is if we pay him too much he might retire.

    I was thinking of $30/month. Let’s see if we can get at least 100 regular contributors who will commit to a monthly contribution. And if Allan retires, we’ll meet at his place in a scene reminiscent of Frankenstein. ;-)

  23. gvblackhawk says:

    Lowetide: Actually, I can. It’s not impossible to expect the reader to take “instant analysis” as exactly that, while also understanding that it takes years to correctly evaluate the draft.

    I don’t think that’s asking too much of the audience.

    Analysis and judgement are two separate things.

    “Analyzing the 2014 Draft” and “Who Won the 2014 Draft” are not the same. To answer the second question, you need more information. The five year timeline is reasonable. The graded exercise from yesterday was equivalent to “Who Picked Closest to _______’s List”. If a team strayed from the list then they were penalized. Does that mean that they drafted poorly or made poor decisions? No. Only hindsight can answer that.

  24. НИНТЕНДО⁶⁴ says:

    SinceTheWHADays: The problem is if we pay him too much he might retire

    I think he might be over 35. The cap hit if he can’t write any more could really hurt.

  25. Woodguy says:

    LT,

    A good example on why you need 5 years to judge a draft is this guy:

    Drafted as an undersized overager (19 years old) out of the USHL

    DIdn’t turn pro until the start of the 4 th season after his draft.

    A bad pick, an under ager, took too long to turn pro, walk about pick in the 7th round of the 2003 draft.

    Joe Pavelski.

  26. sliderule says:

    The second round winner is no surprise the Kings

    Kings Three players 532 games 161 pts. 5 picks

    Ducks three players 227 games 93 pts. 9 picks

    Hawks two players 182 games 101 pts. 9picks

    Oilers three players. 148 games 15 pts. 7picks

    I cherry picked the teams as I knew they had good records.
    I think it’s better to measure yourself against excellence rather than mediocrity.

  27. Ducey says:

    sliderule: The second round winner is no surprise the KingsKings Three players 332 games 161 pts. 5 picks Ducks three players 227 games 93 pts. 9 picks Hawks two players 182 games 101 pts. 9picksOilers three players. 148 games 15 pts. 7picksI cherry picked the teams as I knew they had good records.I think it’s better to measure yourself against excellence rather than mediocrity.

    Yeah, context is everything though.

    2008, the Oilers didn’t have a 2nd. For this study, this draft has the highest amount of clarity.

    The Kings picked Voynov at #32. The Ducks had 35 (Deschamps), 39(O’Dell) and 45 (Jultz). CHI had no pick. Best pick was the highest (of the 4 teams).

    2009: EDM picked Lander at 40. LA picked Clifford at 35. ANA picked Mat Clark (a bust) at 37 CHI picked Piri at 59th out of the OJHL. He was traded to FLA at the 2014 draft. Good AHL numbers but the jury is still out on him.

    You can say LA did the best, but they had the highest pick.

    2010: EDM gets lots of criticism for Pitlick, but he was highly rated. Most of the 2nd round has not turned into much. http://www.hockeydb.com/ihdb/draft/nhl2010e.html Faulk, Smith-Pelley and Toffoli look like they are about it, besides our man Marincin and maybe Merril. There is still time for some of these guys to advance. CHI had three 2nd round picks as well, and who knows, but its not looking good for them. Maybe not the best draft to stock up on picks.

    After that, its too early to tell.

  28. wheatnoil says:

    sliderule:

    I cherry picked the teams as I knew they had good records.
    I think it’s better to measure yourself against excellence rather than mediocrity.

    True, but there’s no question that the Oilers second round drafting hasn’t been excellent. One problem with measuring against the best is if you fail to meet that comparison, you may conclude that you are terrible when in fact you are merely adequate. Certainly the Oilers 2nd round drafting could be better. The question is: have they been average or below average? What needs to happen for them to be above average, even if they’re not excellent.

  29. Marcus Oilerius says:

    Interesting news. Renney takes over as Hockey Canada president, from Bob Nicholson (who came to the Oilers). Given KLowe’s undeniable influence with Hockey Canada, I feel like there’s got to be some sort of connection. Maybe an attempt at keeping Renney in orbit with the Oilers, or a “my bad” kind of move by Lowe?

    http://www.tsn.ca/nhl/story/?id=457204

  30. rickithebear says:

    Tired of al the BS:
    Michael Schukers chart is for 200 gm players from te pick.
    Ie the average.
    the question is are we above or below.

    none of the the feelings.

    http://www.youtube.com/watch?v=6ldAQ6Rh5ZI

    Just look at the chart!

  31. rickithebear says:

    wheatnoil: True, but there’s no question that the Oilers second round drafting hasn’t been excellent. One problem with measuring against the best is if you fail to meet that comparison, you mayconclude that you are terrible when in fact you are merely adequate. Certainly the Oilers 2nd round drafting could be better. The question is: have they been average or below average? What needs to happen for them to be above average, even if they’re not excellent.

    They are average!

    MacT!

  32. FastOil says:

    gvblackhawk: “Who made the best decisions at the table in 2014″ sounds like alternative wording for ‘who took the best players at the 2014 draft according to an arbitrary list’.And there was a grade assigned — you can’t get much closer to ‘judging’ than that.

    It’s a contradiction.You can’t have it both ways.

    I didn’t word that very carefully. By ‘the tone I took’ I meant the tone I understood. I of course didn’t write it.

    As an Oiler fan the Lucic hunting made it plainly clear that if players don’t have enough offense in their draft year it’s almost wasting the pick looking for the kid in a million who beats the odds. And picking bottom sixers with higher round picks shows a misunderstanding of the player markets and cap management IMO.

    It is possible to assess a poorly taken pick while not knowing the final outcome. LT will likely be wrong on a few players out of the 32 team’s picks, a few kids may beat the odds, but study has informed the process a lot in the past few years.

  33. matt says:

    I would be interested in understanding who does best when they stray from the script, and whether they do so consistently. Take the Mackenzie list, for example, and use that as an alternative draft order, taking the highest ranking player available every time. Then compare the total games played by the prospects in this alternative reality to what actually happened, and see which teams are up or down, and whether any teams are up or down consistently.

  34. VanOil says:

    In the interest of fairness does the consensus of this blog that Musil was a poor pick stand up to scrutiny?

    Yes, he was an ‘old boys kid’ pick.

    Yes, he falls into the lumbering Coke machine mold of other Tambellini picks.

    Yes, other magic beans were available at the time but he was not that big of a reach.

    But his final year Jr #’s compare favorably with Reinhart who went #4 overall and will likely play in the NHL next year.

    His first year in the AHL his performance was good and he gained his coaches trust as the year went on. Which is a heck of a feet given the crazy amount of player transactions that OKC went through.

    So we are left with a player with a very good hockey IQ, prototypical size (it does matter in making the show) and in desperate need of some figure skating lessons. (we as Oilers fans have good reason to be suspect of a defender that cannot pivot).

    I am confident Musil will have a pro career his skating will determine what league. He may not play a style of defense this many on this blog (me included) appreciate but it is one still appreciated by NHL GMs. Musil is a developing asset, who is developing well. He might never play for the Oilers but he still has value and a 10 year NHL career is still possible.

  35. Deadman Waiting says:

    These are such limited samples that it’s difficult to establish whether there’s any skill at all over and above a well-regarded consensus list, such as Bob’s Orthodox Bias.

    This is terribly difficult because a player picked 5th automatically gets more at bats than a player picked 15th. The 1st overall has nine lives, the 5th has eight lives, and the 15th has seven lives, the 60th pick has three lives based on the initial draft-day guess.

    Normally the test for skill is to look at the outcome variance curve to see if it looks the same as the null hypothesis (no skill) variance curve. Skill (and anti-skill) tend to fatten the tails.

    It’s escaping me completely right at this moment how to even to define an extraction (in a non-A/B world) of the self-fulfilling-prophecy term, without which the variance test above is highly suspect.

  36. Woodguy says:

    Interesting news on the FancyStats front.

    Eric Tulsky, who is one of the best minds and writers about Hockey FancyStats had this announcement today:

    Eric T. @BSH_EricT · 3h
    I’m pleased to announce that I’m going to be working part-time for an NHL team this year.

    You can read his stuff here: http://www.sbnation.com/outnumbered

    He also wrote for nhlnumber and most of his stuff started on the Flyers blog Broad Street Hockey.

    I think LT had Eric on the show a few times.

    Eric has a PHd (Chemistry I think) and has a good full time gig using his degree, but this is still cool news.

    What happened in the NBA is that all the teams got smart quickly and snapped up all the best FancyStats writers/thinkers.

    NBA executives usually had a business background and not a playing backgrouind, so the uptake on the analytics was very quick once they proved their worth. The info the NBA got from SportsVu was a clincher for all the teams to open analytics departments.

    Hockey is much, much slower on the uptake and the data is about to explode with the SportsVu coming online in a few years.

    Ex-players haven’t been exposed to uses of “big data’ that normal business executives have so I imagine there will still be a competitive advantage in having an analytics dept. in the NHL for years yet.

    Good for Eric. He’s one of the good guys and a very good writer.

  37. Lowetide says:

    Deadman Waiting:
    These are such limited samples that it’s difficult to establish whether there’s any skill at all over and above a well-regarded consensus list, such as Bob’s Orthodox Bias.

    This is terribly difficult because a player picked 5th automatically gets more at bats than a player picked 15th.The 1st overall has nine lives, the 5th has eight lives, and the 15th has seven lives, the 60th pick has three lives based on the initial draft-day guess.

    Normally the test for skill is to look at the outcome variance curve to see if it looks the same as the null hypothesis (no skill) variance curve. Skill (and anti-skill) tend to fatten the tails.

    It’s escaping me completely right at this moment how to even to define an extraction (in a non-A/B world) of the self-fulfilling-prophecy term, without which the variance test above is highly suspect.

    Absolutely. We used to talk for years back when the Oilers finished 8th or 9th or 10th in the conference about how good it would be to have a No. 1 overall pick. That’s the way to beat the odds.

  38. VanOil says:

    Lowetide: Absolutely. We used to talk for years back when the Oilers finished 8th or 9th or 10th in the conference about how good it would be to have a No. 1 overall pick. That’s the way to beat the odds.

    And as we all now know the price of beating the odds with No.1 overall picks is only soul sucking, utterly demoralizing, last place hockey.

  39. smellyglove says:

    LT:

    You establish a quantitative framework for judging successful picks with # of games and a timeframe but use the “eye” to establish good and bad arrows. Not sure if there is a better way to do this?

    Also, so much about development is about opportunities, which don’t necessarily reflect the quality of the puck. Would Lander gave the games he does under his belt if he played for a well managed club?

    Great way to start framing the evaluation though or!
    Ps: any chance you could turn on the mobile site on your blog so its easier to access the blog on your phone?

  40. wheatnoil says:

    I was thinking about Anton Lander the other day and had a few thoughts rolling around in my head… now seems as good a time as any to share them, given we’re talking about second round Oilers picks.

    Here’s the thing. Anton doesn’t have an offensive bat and he never did. We need to agree on what the Oilers expected him to become in order to decide if he has been successful.

    Let’s go back in a time machine. Back to December 2009. Here’s what Lowetide wrote about Lander then, ranking him #10 in the summer top 20.

    http://lowetide.ca/blog/2009/12/10-prospect-anton-lander-2.html
    “Do you remember Doug Jarvis?… Doug Jarvis had terrific skills, all pointed in one direction. Defensive play. Aggressive forecheck, a demon in the faceoff circle, outstanding penalty killing. Jarvis wasn’t a great goal scorer… but he was a valuable player for two strong teams during his career. … [Lander's] going to be an NHL coaches dream if he makes the show but offense is going to be a problem. Already regarded as a team leader, every scouting report talks about his prowess as a checker and “two-way” forward. … Desjardins NHLE for this season shows Lander at 82gp, 10-10-20. … I think Lander has a chance to make it as a specialist.”

    If you read the whole post, it pretty much sums up everything we know about Lander and LT had him pinned five years ago!

    Here’s the thing, you can quibble about whether it makes sense to draft a defense-first bottom sixer early in the draft. However, the way the league is going (and indeed the Oilers based on stated verbal) is 3 offensive lines and 1 defensive specialist line. With that forward make-up, I would argue the most important player outside of the first line, is your 4C. Your 4C and his ability to handle the heavy defensive load is what allows you to carry 3 scoring lines. Today, Gordon is the most important Oiler forward after RNH, Hall, and Eberle. Lander was drafted to become what Gordon is now.

    So what does that mean? That means that Lander doesn’t have to bring the bat of a 2C or 3C. He needs to meet his NHLE from his draft year (20 points). Gordon has produced the following point totals (adjusted to 82 games played to make for easy comparison): 33, 20, 18, 22, 12, 25, 24, 23. That averages out to 22 points per 82 game season (or 0.27 ppg). So if Lander can get about 20 points over 82 games while playing a Gordon role, I would argue that’s an acceptable level of production for the role.

    This is a long comment and what sparked it is this question: “What does Lander need to do this year for us to be able to agree that he was a draft success?” Lander’s potential role on this team hasn’t changed since the day he was drafted. If he can take over Gordon’s duties in 2 years when Gordon’s contract is up, that is a huge win for the Oilers. I would argue that in today’s NHL, a centre that can play the defensive specialty line is an impact player, regardless of point totals.

    Based on all this, I think this is what we need to see from Lander this year to say he’s been a draft success:
    1) He needs to play the vast majority of the season at the NHL level.
    2) He needs to bring enough of a defensive game with him that he can play on the Gordon-Hendricks line, face crippling defensive zone starts against tough opposition, and not get destroyed.
    3) He needs to bring a small amount of offense. Anything. If he can hit 0.25ppg (20 points in 82 games), we sing in the street and call it a huge success, celebrate the acquisition of an impact player in the second round of the draft and look at succession planning for Lander to take Gordon’s role when the latter’s contract is up.

    That offensive bat may be asking a lot… he only got 1 point in 27 NHL games last year… but he also got 52 points in 46 games in the AHL last year. Now, according to Vollman, translating AHL points to NHL points changes drastically at age 22. 19-21 year olds have a NHLE factor of 0.46, where as at 22 that drops off to 0.31 (an effect that disappears at lower scoring rates). So, based on this, Lander’s NHL-E should be 0.35 ppg or 29 points next year. (Link: http://www.hockeyprospectus.com/puck/article.php?articleid=1139)

    However, looking deeper at that 52 points, he got exactly half on the powerplay (10-16-26). He’s unlikely to get any powerplay time in the majors, though credit must be given to Lander for making the most of his special teams opportunity. This means he scored 8-18-26 at even strength and short-handed (0.57 ppg). It’s difficult to make a Vollman translation on this number because the translations don’t separate out special teams. If we use the 0.31 of 22 year olds, we’d get 0.18 ppg or 14 points, but that is an under-estimate because the drop off for 22 year old AHLers diminishes with lower point totals.

    All of this seems to indicate that, at least based on his AHL work last season, Lander took a major offensive step forward, mostly fuelled by powerplay time, but even at even-strength he should be able to bring some offense to the NHL level. Why he didn’t is up for question. Was there a mental block? An inherent problem with his play? Bad luck? I don’t know. However, hoping for 20 points over 82 games isn’t completely out of reach for Lander based on his AHL play. If he can do that… and then transition into the Gordon role over the next two seasons? Well that’s a massive win for Stu and the Oilers and a fantastic second round pick.

  41. Factotum Pochemuchka says:

    OT, but can someone tell me where I can find career faceoff data (not just FO%, but total number of draws taken, won, and lost per season) for individual players? Thanks.

  42. doritogrande says:

    I’m wanting to nitpick about the 150 games line in the sand, but need some clarity first LT.

    Are you using “150 games” as a threshold they have to attain before they reach the end of their 5 year mark, or “150 games” as a benchmark for their career to be a success?

    Because if it’s the former, it’s too damn high, and the latter, well I didn’t think that far through, and needs a fancystats helper.

  43. Lowetide says:

    doritogrande:
    I’m wanting to nitpick about the 150 games line in the sand, but need some clarity first LT.

    Are you using “150 games” as a threshold they have to attain before they reach the end of their 5 year mark, or “150 games” as a benchmark for their career to be a success?

    Because if it’s the former, it’s too damn high, and the latter, well I didn’t think that far through, and needs a fancystats helper.

    I started with Cullen’s 100, but don’t really like that number, it’s not enough really. On top of that, we’re 6 games away from calling Lander a “success” and that’s not right either. If Lander can play 56 games this season, Edmonton will have found some way to make him useful (or that’s the idea).

    So, not 150 games in 5 years, but 150 games (hopefully on the way to more) as a threshold.

  44. Bruce McCurdy says:

    Lowetide: Actually, I can. It’s not impossible to expect the reader to take “instant analysis” as exactly that, while also understanding that it takes years to correctly evaluate the draft.

    I don’t think that’s asking too much of the audience.

    Well done on giving your readership credit for having a little intelligence. Should be a given everywhere there IS a readership, but sadly, it’s not.

  45. RexLibris says:

    Lowetide: I started with Cullen’s 100, but don’t really like that number, it’s not enough really. On top of that, we’re 6 games away from calling Lander a “success” and that’s not right either. If Lander can play 56 games this season, Edmonton will have found some way to make him useful (or that’s the idea).

    So, not 150 games in 5 years, but 150 games (hopefully on the way to more) as a threshold.

    I still prefer the 200 NHL games threshold because of what it implies, that the player was considered useful and/or promising for the better part of two and a half NHL seasons, barring injury.

    I agree, though, that Cullen’s cut off is far too low. Sometimes a high-ranked prospect will get 100 NHL games from one, two or even three teams before they all realize he is a bust. Krys Kolanos netted 149 NHL games from four teams (Phoenix, Edmonton, Minnesota and Calgary) in part due to the fact that he just looked like a hockey player and the Coyotes had taken him 19th overall back in 2000 so there just had to have been a hockey player in there somewhere.

    Turns out there really wasn’t.

    The differentiation between 150 and 200 NHL games makes a subtle, but I feel important, difference in the numbers returned and helps to whittle down the results.

  46. Romulus Apotheosis says:

    Against my better judgment… got in another tangle with Staples today, wrote about it here

    http://www.theoilersrig.com/2014/07/corsi-facts-vs-norms/

  47. Lowetide says:

    RexLibris: I still prefer the 200 NHL games threshold because of what it implies, that the player was considered useful and/or promising for the better part of two and a half NHL seasons, barring injury.

    I agree, though, that Cullen’s cut off is far too low. Sometimes a high-ranked prospect will get 100 NHL games from one, two or even three teams before they all realize he is a bust. Krys Kolanos netted 149 NHL games from four teams (Phoenix, Edmonton, Minnesota and Calgary) in part due to the fact that he just looked like a hockey player and the Coyotes had taken him 19th overall back in 2000 so there just had to have been a hockey player in there somewhere.

    Turns out there really wasn’t.

    The differentiation between 150 and 200 NHL games makes a subtle, but I feel important, difference in the numbers returned and helps to whittle down the results.

    I’d prefer 200 too, but then we’re really stretching Cullen’s 28% number (which I’ve stretched to 30%).

  48. doritogrande says:

    I think I’m more a fan of tracking games played during the 5 year window as an accurate measurement of success. It’s going to wind up with a much lower percentage, but we’ll likely have the “Actual NHLers” surrounded rather than the (aforementioned) “Krys Kolanos pretenders”

    I’d devise a model that features cup of coffee (3-4 games) in year 3, shuttle squad (10-25 games) in year 4, and minimum half a season (50 games) in year 5 post draft.

    Gives something like 65-80 games in the 5+ draft year. I think by then we’ll have a reasonable expectation of whether we’re talking about an actual NHLer rather than a fringe guy.

    This model does exclude the late bloomers, and certainly the collegians, so maybe it’s something to be employed for those that follow the typical progression model (two years of junior or Europe, three-year entry level contract.)

  49. Romulus Apotheosis says:

    I always liked 200 games for Fs; 150 for D and 100 for G.

    But, I think setting any specific bar is less important, than getting everyone to agree to something reasonable, esp. regarding the drawing of conclusions.

    ——–

    I’d be really interested to know if Nelson had a hand in the Hamilton qualification, as LT suggests may have happened. I expect Bill Scott was involved in some manner.

    speaking of qualifications… any word on when these kids might actually sign something?

  50. Ducey says:

    wheatnoil,

    Great comments.

    But for all the fancy stats, I would settle for a few “saw him goods” in order to hold out hope.

    Pitlick has the skating and aggression to maybe make it. We saw that for a few games before he got submarined. It didn’t see much in the way of puck skills but hey at least he had an impact.

    With Lander, I was really disappointed last year. He seemed to be really tentative on all fronts. Didn’t really seem to flash any of the stuff he apparently had on the farm. If he doesn’t show up to camp with a burr under his saddle, run into a few people, and generate some turnovers and shots, I will have given up on the man.

    As for Musil. I still think he is a good prospect. He won’t be a puck mover, but every team needs a 5-6 who can blocks shots, plays the PK, breaks up the cycle, lays the lumber in front of the net and handles the heavy stuff. Some simple mistake free defence would go a long way in these parts.

  51. Lowetide says:

    Romulus Apotheosis:
    I always liked 200 games for Fs; 150 for D and 100 for G.

    But, I think setting any specific bar is less important, than getting everyone to agree to something reasonable, esp. regarding the drawing of conclusions.

    ——–

    I’d be really interested to know if Nelson had a hand in the Hamilton qualification, as LT suggests may have happened. I expect Bill Scott was involved in some manner.

    speaking of qualifications… any word on when these kids might actually sign something?

    They’ll announce all of them same day. MacT might be trying to trade one or more, that’s another reason to qualify him. Oilers qualified Omark once without loving him even a little.

  52. nycoil says:

    Romulus Apotheosis:
    Against my better judgment… got in another tangle with Staples today, wrote about it here

    http://www.theoilersrig.com/2014/07/corsi-facts-vs-norms/

    When I read the article on CoH I was wondering if someone here was going to take that on.

    Lowetide,

    Love the idea for this article/examination of the 2nd round drafting record. I know the discussion still centres around what thresholds and lines are “fair,” but looking ahead a little, I was wondering if you/knowledgable posters here could address whether there is a gap between perception and reality with regards to how the Oilers’ rebuild has been portrayed.

    I believe there has been a discourse of the majority that has purported one of the main failings of the Oilers’ rebuild to-date to be the lack of success of the draft picks outside the 1st overalls. If I elaborate a bit further on what I am getting at, the narratives I hear are, and please feel free to add to them if I am forgetting any:

    1) The Oilers were unlucky to be bad in bad years (No Tavares, Stamkos, Crosby, Mackinnon when the Oilers got to pick 1st).
    2) The Oilers took the obvious players at 1st overall and deserve no credit for making those picks.
    3) The Oilers have failed to produce any players outside the 1st overalls while teams like LA and Chicago have drafted good players with their late 1sts or with 2nd rounders or later.
    4) A 31st overall pick is pretty much a late 1st and the Oilers should have done better with their selections.

    So we’re looking at items 3 and 4 here, clearly. I’d add to that perhaps looking at the Chicago and Pittsburgh rebuilds of high lottery picks, did those teams really find success with their early 2nd round picks (Duncan Keith comes to mind) while the Oilers have failed?

    With all that said, whether the line is 28% or 30%, 150 games or 200, is there a way we can use this study to either prove or disprove statements 3 and 4, and also contrast with the Chicago and Pittsburgh 2nd round drafting record during their bottom-feeding years (or better yet, compare with other teams that were at the time in similar lottery situations to increase the sample size–Colorado, LA, Philly, Anaheim, etc.), in order to address point 4?

    It would be wonderful to be able to point to this study whenever people make comments about the Oilers’ drafting record on points 1) thru 4) and show definitively that the evidence (hopefully) doesn’t support those claims perpetuated by the media and by perception around the hockey world.

    Thanks again.

  53. RexLibris says:

    Lowetide: I’d prefer 200 too, but then we’re really stretching Cullen’s 28% number (which I’ve stretched to 30%).

    The funny thing is, before I had started down the rabbit hole of reviewing draft success %s, I had kind of a gut feeling that one in three was a good way to go about looking at picks. Take three prospects who fit within a certain range (Fs, drafted 2nd round, similar development) and you could probably guess that one would become an NHLer, one a minor leaguer and one would fail to develop into a pro.

    I’ve argued for this approach as a rough rule of thumb for Flames fans with their recent influx of prospects in order to try and ground expectations there.

    Many remain convinced that 80% of their prospects are going to become full-time NHL players who fill out a complete roster.

    I remember those days.

  54. wheatnoil says:

    I think if you move the line in the sand to 150 or 200 games, you have to lower the success rate off 30%. It’s true that keeping Cullen’s 100 games is a low bar that would include Lander in the mix, but Cullen’s 28% includes guys like Lander. So if you push the marker to 150, there’s going to be guys in Cullen’s 28% that no longer make the cut, lowering the success rate for all teams.

    The cut off doesn’t matter really but it has to be fair for all teams. If “average” is a 30% chance at 100 games then 150 means something less than that.

  55. kooler says:

    nycoil,

    I would trade for EDM 2015 2nd round pick for LAK ‘s head scout. just kidding…but why not.

  56. G Money says:

    Deadman Waiting: These are such limited samples that it’s difficult to establish whether there’s any skill at all over and above a well-regarded consensus list, such as Bob’s Orthodox Bias.

    On the topic of limited samples and trying to determine who’s better at drafting and who isn’t…

    A while back I did my own analysis of draft success. My approach was a bit different from others – I didn’t filter the data based on games played, I just used total games and then did a curve fit. Given the way the data works, the curve fit should give a more “true” assessment of success rates, and I think it does.

    As a side experiment, I took the probabilities of success that I calculated and plugged them into a Monte Carlo simulation that compared two scouts, one with a 10% higher probability of success than the other (that’s a pretty massive edge in most mostly-random processes. You’d get rich awful quick with an edge like that in poker, horse racing, the stock market, etc.).

    I had to push the simulation to almost 300 years before a visible (on the histograms) difference showed up between the scouts. Or to put it another way, a scout who is 10% better than another would not conclusively demonstrate his prowess until 300 drafts had gone by.

    The idea that “Team X is a better team than Team Y” because they hit five players while another hit only two is pretty much exactly like declaring that you are more skilled at dice than I because you rolled a five while I rolled a two.

    Yet we persist in our delusions of pulling meaningful data out of random noise…

  57. RexLibris says:

    Romulus Apotheosis,

    Good article. The dialogue over Corsi v. Neilsen is important, if arcane to many casual observers.

    On advanced stats, I had wondered a few years back when I was just starting to look at some of the BtN numbers about the value of investigating how much of a stat, like Corsi, is self-fulfilling.

    We can use Manny Malhotra as an example. Vigneault absolutely buried the man in ZS in order to give the Sedin’s the lighter assignments, and he proved very capable in this regard, therefore he continued to get the Def ZS.

    Then we have a self-perpetuating cycle that becomes reflected in the numbers but may not be an entirely fair evaluation of Malhotra as a player.

    Consider another example. Andy Sutton arrived here appearing, for all intents and purposes, like the proverbial Coke Machine on D who would somehow hold opposition players accountable for any number of dastardly acts. Yet, we were pleasantly surprised at how well he could pass, skate and indeed shoot the puck for such a large man. He wasn’t Coffey, but he was a darned sight better than many had anticipated based on some preliminary data.

    To look at it another way, your boss assigns you to tackle some crisis outside of your usual duties. You do it and perform well. He notices and therefore assigns you another crisis because he can’t seem to run an operation smoothly. Soon enough, despite being entirely competent in other areas of your job, your workload becomes dominated by putting out organizational fires because you’ve shown the slightest bit of proficiency doing so and management needs you to continue so that the rest of the group can strive towards their goals.

    Meanwhile, by most metrics it would appear as though you accomplish less than your colleagues.

    So, kind of a roundabout way of explaining what I was mulling over, but you can see why I never really tackled the idea – I just don’t think we (myself, mostly) are privy to enough information on deployments and coaching perspectives to discuss the topic appropriately.

  58. nycoil says:

    G Money: On the topic of limited samples and trying to determine who’s better at drafting and who isn’t…

    As a side experiment, I took the probabilities of success that I calculated and plugged them into a Monte Carlo simulation that compared two scouts, one with a 10% higher probability of success than the other (that’s a pretty massive edge in most mostly-random processes.You’d get rich awful quick with an edge like that in poker, horse racing, the stock market, etc.).

    I had to push the simulation to almost 300 years before a visible (on the histograms) difference showed up between the scouts.

    Yes! Monte Carlo simulations! You’re speaking my language. There’s an awful lot of randomness masquerading as skill in the sports world, and I think especially true in hockey.

  59. G Money says:

    Factotum Pochemuchka: OT, but can someone tell me where I can find career faceoff data (not just FO%, but total number of draws taken, won, and lost per season) for individual players? Thanks.

    When I desire to dive deeply into details and data (can I get an alliteration woop woop), I eschew the web interfaces and go straight to the real poop: Rob Vollman’s detailed spreadsheet data. You can find ‘em at hockeyabstract.com. What a freakin’ gift.

  60. RexLibris says:

    This is a little off topic, but good Gord, the Oilers finally have a chance to make something happen, to be good, to accomplish something of note, and then the apocalypse has to happen!

    http://images.christianpost.com/full/72803/the-official-movie-poster-for-left-behind-starring-nicolas-cage.jpg

    Yep, sure sign the end times are here.

  61. godot10 says:

    RexLibris: I still prefer the 200 NHL games threshold because of what it implies, that the player was considered useful and/or promising for the better part of two and a half NHL seasons, barring injury.

    I agree, though, that Cullen’s cut off is far too low. Sometimes a high-ranked prospect will get 100 NHL games from one, two or even three teams before they all realize he is a bust. Krys Kolanos netted 149 NHL games from four teams (Phoenix, Edmonton, Minnesota and Calgary) in part due to the fact that he just looked like a hockey player and the Coyotes had taken him 19th overall back in 2000 so there just had to have been a hockey player in there somewhere.

    Turns out there really wasn’t.

    The differentiation between 150 and 200 NHL games makes a subtle, but I feel important, difference in the numbers returned and helps to whittle down the results.

    Isn’t Kris Kolanos a concussion casualty?

  62. RexLibris says:

    godot10: Isn’t Kris Kolanos a concussion casualty?

    I hadn’t heard that anywhere. He may have suffered a concussion early in his career, but I remember him being essentially a “taxi squad” guy for the Oilers during the 2006 playoffs and then struggling on the Heat when he signed with Calgary.

    If he did have chronic concussion issues there was nothing in the items I read at the time that suggested that his performance was being impacted by it (which isn’t to suggest that it wouldn’t have).

  63. maudite says:

    Lowetide,

    Sorry, draft +1 is what I meant. Wrong term but don’t think he had the numbers to reach at that spot. End of 2nd round maybe, likely could have got him with 3rd pick I bet. Worth the “gamble” to wait until then.

  64. Lowetide says:

    maudite:
    Lowetide,

    Sorry, draft +1 is what I meant.Wrong term but don’t think he had the numbers to reach at that spot.End of 2nd round maybe, likely could have got him with 3rd pick I bet.Worth the “gamble” to wait until then.

    Agreed.

  65. gcw_rocks says:

    My biggest criticism of the Oilers draft strategy in round two is that they don’t give themselves the best probability to succeed, and the lead time is long. By the time you get to round two, the odds of the player making it are dropping dramatically, so getting enamoured with a player for need is a dumb way to draft, especially since by the time they make it the team needs could have changed dramatically.

    Very few people thought Musil was the highest probability choice at 31. Saad was ranked 22 on Mackenzie’s list and was still on the board. I don’t care how much weak the defence was at that point, drafting a defender with footspeed issues over Saad was foolish. Nobody thought Moroz was the highest probability choice at 32 either. Finn was still on the board and he was ranked 21. Neither Finn nor Moroz could make it, but the pick was still a bad pick.

    As a bonus, they just locked up a guy for five years to play in the spot the team would have aspired to see Moroz play in at 3LW. Because if the aspiration for drafting Moroz was for him to play 4LW then that makes the pick even worse.

    For me, games played fits not tell the whole story, and in fact it leaves out some critical elements of it.

Leave a Reply

Want to join the discussion?
Feel free to contribute!

Leave a Reply

© Copyright - Lowetide.ca