Sham Sharron takes over all 30 draft tables; how did he do?

That's Offside!
June 26 2014 10:52AM

prospects nhl 2003 584 

Zach Parise hasn't aged a day since 2003. Dustin Brown on the other hand was drafted at 13 years old.

So the 2014 NHL entry draft is just around the corner. NHL teams are assembling en masse in Philadelphia as we speak to prepare for the first day of the rest of their franchises' lives. Games have been watched, interviews have been conducted, numbers have been analyzed, and (as fans, we hope) that all the groundwork has been done to ensure that our favourite team comes away with a draft choice that sets themselves up for plenty of success in the future.

Of course, we know that this isn't the case. Someone is going to blow it. Someone is going to make a draft choice that everyone will look at in 5 years down the road and point and laugh at and say "HA! Can you believe the Rangers took Hugh freaking Jessiman instead of Ryan Getzlaf!?"

We just don't know who this someone will be. Yet. With hindsight being what it is, it almost makes you wonder if a potato could do better than some teams. I mean, just ask Oilers fans about the Kevin Prendergast years, or Leafs fans about any of the years. They're sure to tell you that, "no way man, my team is really the worst."

So here at Canucks Army, we decided that hey, why not take turns putting a potato named Sham Sharron in charge of amateur scouting for every NHL team in the league and see what happens? Well, as it turns out, Sham Sharron isn't half bad. But would Sham really be an upgrade for your team? Will he leave you crying bitter tears at the thought of the guys that your team should have drafted? Or will Sham have you thankful that your team's scouting staff is actually better at identifying hockey talent than an inanimate foodstuff? You'll have to read past the jump to find out.

Before we begin, a huge thank you goes out to Josh Weissbock for his invaluable help in doing this project. Quite simply, Sham Sharron 2.0 does not get started, let alone finished, without his fantastic work. Josh created the backbone of this entire project by building a database of all the 17-year olds to play in the CHL from 1999-2000 to 2009-2010, then writing scripts to simulate drafts for each team and each year. All the simulated drafts are courtesy of his fantastic Python skills, so thank you Josh. This doesn't get done without you.

Meet Sham Sharron: Your New Scouting Department

If you've read the article that inspired this project, you've already met Sham Sharron. If not, well, it's probably a good idea to get to know him since he took over your favourite team's entire amateur scouting department starting in the year 2000 until he decided to call it quits after the 2010 entry draft.

For all intents and purposes, Sham is as smart as a potato and lazy as hell. He's an unpaid summer intern who doesn't watch a single junior hockey game, ever. He conducts no interviews, speaks to no experts, and gives precisely zero craps about fitness combine results. Sham doesn't care at all about intangibles or heart or tools or style of play or organizational needs. With each draft choice, Sham will select players based only on the following rules, or some slight variation thereof:

    1. All players selected will be from the Canadian Hockey League.
    2. All players selected will be in their first draft eligible season. In other words, they will be 17 in September at the beginning of the season.
    3. Goalies are voodoo, they will not be selected at any time.
    4. Defensemen are voodoo, but they will be selected if they out-score every forward available at that time.
    5. The player that Sham selects will simply be the highest scoring first time draft eligible CHL player left on the board at the time of Sham's selection, given a very simple set of criteria.

    Further clarification on point 5: Since this is not a "stats vs. scouts" debate and Josh has created a database from which we can run pretty much any variation of Sham Sharron we want, we're able to add layers of complexity to Sham's drafting method in order to get a more nuanced and hopefully "realistic" approach to what someone may have actually done. We've still kept Sham's methods very simple as the goal here isn't to build an accurate, stats-reliant draft model, but rather to create a benchmark with which we can compare each individual NHL team's drafting record with what a potato would have done given a few simple guidelines.

    We have selected three different sets of these guidelines to test each NHL team against. They are as follows:

    1. Version #1: The player selected will simply be the available player with the most points.
    2. Version #2: As an intern, Sham doesn't want to look insane by going off the board for massive reaches, and he wants to appear sensible to his bosses. As such, Sham will select the available player with the most points from the 30 remaining highest ranked North American skaters by NHL Central Scouting Services (CSS).
    3. Version #3: Sham will incorporate CSS ranks and draft in the same manner as #2, but he also understands that age can play a factor in how a player performs. He has spent 30 minutes or so deriving a formula to adjust scoring for age, and will use "Age Adjusted Points Per Game" to decide who to pick instead of total points.

    For versions #2 and #3, Sham also knows that offense isn't equal in each different league. Brad Richards, for example, was coming off a one hundred and eighty six point season with Rimouski in 1999-2000. The QMJHL also had 21 other 100+ point scorers that season, compared to just 8 split between the OHL and WHL combined. Since an individual point is worth less in leagues where scoring is through the roof, Sham will ignore a league in a given year if total offense is 10% higher than the next highest scoring league. Fortunately, this only happens once in our sample: the 1999-2000 QMJHL season. Therefore, versions #2 and #3 of Sham will not draft QMJHLers in 2000.

    I think it's also important to note that Josh and myself have addressed the concerns about Sham's abilities of foresight that arose with the previous version of this experiment. We have ensured with this experiment that Sham only uses concrete inputs that conceivably would have been available at the time of each entry draft.

    Why Sham Exists

    To eliminate any confusion that may arise as to the purpose of this exercise, I want to make this next point perfectly clear. So I've bolded it. And put it in it's own paragraph. With italics.

    We did not design Sham Sharron to be a good or intelligent way to draft. Sham is a benchmark we have created to test each team's scouting department against publicly available information. We do not for a second believe that any NHL team should draft in the same manner as Sham.

    Now that that's out of the way, we can get in to some of the hows and whys and logic behind the Sham Sharron drafting test. 

    Sham's purpose, as stated above, is to test each team's scouting department against publicly available information. Sham only uses a simple set of rules in each iteration, as well as data that was publicly available and known at the time. If a given NHL team's scouting staff is effective and ultimately worth the money invested in them, they should be able to select players that provide more value than what a group of players selected by a simple, unthinking process would.

    We're comfortable including CSS ranks in this test (which implicitly include stuff like physical attributes and defensive ability) because this really isn't a binary "stats or scouts" debate. Instead, it's about whether a private scouting staff that travels the world on behalf of your organization is really a worthwhile investment. 

    What We Predicted

    Based on the results of the first Vancouver Canucks re-drafting project as well as knowing how Vancouver has traditionally fared on draft day, we are fairly confident that Sham's simplest iterations (which are basically indiscriminately selecting the highest scoring CHL forward with each draft choice) will not out-perform the majority of NHL teams. The talent pool that Sham is selecting from should be simply too small to find player assets at a rate that rivals teams with traditionally effective European and American amateur scouting departments, and "one-season point totals of 17-year olds" should carry a lot of noise, based on what we know about shooting percentage variance at the NHL level.

    On the other hand, while we believe that scouting and consensus opinion does hold a degree of value, we're not entirely confident that most NHL teams have been prioritizing what we know leads to future NHL success from current prospects. As a result, we also believe that Sham's methods that include a few more variables, such as CSS rank, will yield results comparative to many NHL teams.

    Results

    This Michael Shuckers paper estimates that an average small-market American team spends roughly $2 million annually on amateur scouting, for a total investment of $22 million over the period of drafts we are analyzing. Since the Shuckers piece also includes a measure of the value of one drafted NHL game at $29,300, and Sham Sharron has no cost attached to him, we can infer that, as a really rough baseline measure, a given NHL team must draft 750 games played more than Sham in order to pass the test as a rule of thumb.

    Below we'll run through the results for all 30 NHL teams, along with a quick comment on each team. More detailed macro-level analysis can be found in the "Discussion" section. The "Pass/Fail" evaluation is largely my opinion and is based on not only the number of games that Sham drafts relative to a given NHL team, but the quality of player drafted too. Since star players are harder to replace than good players, they'll be given greater weight. This component is wide open to discussion and interpretation, and we invite your feedback in the comments section below.

    Anaheim Ducks

    Actual Draft Record:

    Anaheim A

    Pretty solid, with the exception of 2003 when the Ducks landed not one, but two franchise cornerstones in the mid-late 1st round in Getzlaf and Perry.

    Sham Method #1 - Most Points:

    Anaheim B

    No Getzlaf, No Perry, less GP. Giroux, Pominville, Ennis, Dubinsky, and Gallagher are nice though.

    Sham Method #2 - Most Points of Next 30 Ranked Players by CSS:

    Anaheim C

    Again, no Getzlaf or Perry, but quite a bit of talent nonetheless. Big improvements in 2000 and 2001.

    Sham Method #3 - Highest Age Adjusted Points per Game of Next 30 Ranked Players:

    Anaheim D

    Nearly 1000 more drafted games than how Anaheim actually performed, but not as much marquee talent. Hemsky and Pominville out-produced Getzlaf and Perry in the Ducks cup year though, so the difference in player quality isn't as large as it appears.

    The Verdict: Anaheim is one of the few teams that, despite failing to match Sham's ability to find future NHLers, might get the benefit of the doubt based on finding two superstar players in one year. Sham can find Claude Giroux in '06, Mike Richards in '03, and Justin Williams in '00 though to go along with Pominville, Ennis, and Toffoli to build a deep and effective top-6 forward corps, but it's probably just not enough to match the Ducks' two superstars.

    zPass

    Boston Bruins

    Actual Draft Record:

    Boston A

    2003, 2004, and 2006 are all very strong, as is the Tyler Seguin selection in 2010. There are a couple prolonged periods of an inability to find impact players too, though.

    Sham Method #1 - Most Points:

    Boston B

    Very, very bad. Bergeron in '03 for Dawes, O'Sullivan, and Richardson? No thank you.

    Sham Method #2 - Most Points of Next 30 Ranked Players by CSS:

    Boston C

    Much better. Williams, Pominville, Stoll, Stajan, Richards, Ennis, and Brassard are probably enough to replace Marchand, Krejci, and Lucic over the last few seasons, but Bergeron leaves a void Sham doesn't fill.

    Sham Method #3 - Highest Age Adjusted Points per Game of Next 30 Ranked Players:

    Boston D

    *Note: Marchand with the 6th pick in '06 is an error. Boston did reach to select him ahead of where he was rated in real life, but we can make the assumption he would have been taken by the time of their 6th and final pick in Sham's draft. Assuming this pick played 0 NHL GP does not affect the results.

    Seguin - Couture may be as good a 1-2 punch at C as Krejci - Bergeron in a year or two, and maybe the Bruins keep Seguin if they don't have the C depth they do. Not much other top-line talent though.

    The Verdict: Once again, Sham out-drafts the Bruins in terms of games played, goals, assists, and points. Sham finds as many NHLers as Boston did over this 11-year span, but doesn't dig up an elite piece like Bergeron. It's close, but edge Boston

    zPass

    Buffalo Sabres

    Actual Draft Record:

    Buffalo A

    Under the direction of Jim Benning, Buffalo drafted extremely well in the early 2000's, finding success in many smaller, skilled players.

    Sham Method #1 - Most Points:

    Buffalo B

    Under Sham, Buffalo goes in the tank in '05, finding only two legitimate NHLers in Tyler Ennis (who BUF found anyway) and Brendan Gallagher.

    Sham Method #2 - Most Points of Next 30 Ranked Players by CSS:

    Buffalo C

    The Sabres draft in a weird position where Sham's results are among the worst in the league. The 5378 future games Sham drafts with his 2nd method is the 3rd worst performance Sham has, ahead of only Toronto and Detroit.

    Sham Method #3 - Highest Age Adjusted Points per Game of Next 30 Ranked Players:

    Buffalo D

    Again, every draft post-2003 is essentially a write-off.

    The Verdict: Buffalo drafted more games played into the NHL (9804) than any other team, largely thanks to the Jim Benning era. Given the same draft picks, Sham makes a mess of them. Easy win for Benning, Darcy Regier, and the Sabres.

    zPass

    Calgary Flames

    Actual Draft Record:

    Calgary A

    Dion Phaneuf, then a bunch of pluggers and busts. Eugh.

    Sham Method #1 - Most Points:

    Calgary B

    Still not good, but about 400 more points drafted. Ennis, Gallagher, Dubinsky, and Pominville probably destroy Backlund, Prust, Moen, and co. too.

    Sham Method #2 - Most Points of Next 30 Ranked Players by CSS:

    Calgary C

    The early years of the Flames really carry their drafting. Sham doesn't find much post-2003, but neither do the real Flames.

    Sham Method #3 - Highest Age Adjusted Points per Game of Next 30 Ranked Players:

    Calgary D

    It still doesn't look great, but that's essentially equal to an average NHL team's draft record from 2000-2010. 

    The Verdict: Sham's results aren't sparkling, but he drafts more games and points into the NHL than Calgary, taking them from a poor team on draft day to an average drafting team. Sham gets his first win, meaning the Flames are literally worse than inanimate foodstuff.

    zFail

    Carolina Hurricanes

    Actual Draft Record:

    Carolina A

    Just three guys taken by the Canes can be considered slam-dunk above average NHL players, though there's an argument to be made for Justin Faulk too.

    Sham Method #1 - Most Points:

    Carolina B

    Sham Method #2 - Most Points of Next 30 Ranked Players by CSS:

    Carolina C

    Sham really likes this Jordan Weal guy. To be fair, Weal probably should be in the NHL though - he's just shy of a point per game #1C in the AHL in his sophomore season on an extremely good Manchester Monarchs team.

    Sham Method #3 - Highest Age Adjusted Points per Game of Next 30 Ranked Players:

    Carolina D

    The decision essentially comes down to Ennis, Pominville, Stoll, and Stajan vs. E. Staal and Ladd.

    The Verdict: Carolina has drafted poorly, that much is obvious, and Sham produces results that are at least comparable to the Hurricanes. Eric Staal is very good, but Andrew Ladd was traded away for Tuomo Ruutu, not allowing Carolina to realize his full value. Edge to Sham, but just barely and it's open for debate.

    zFail

    Chicago Blackhawks

    Actual Draft Record:

    Chicago A

    Between Toews, Kane, Seabrook, Keith, Byfuglien, Hjalmarsson, and Wisniewski, the Blackhawks have found some premium talent at the draft table. Toews was an NCAA player, so Sham must pass on him. Finding a replacement will be hard.

    Sham Method #1 - Most Points:

    Chicago B

    Not even close, Sham.

    Sham Method #2 - Most Points of Next 30 Ranked Players by CSS:

    Chicago C

    Sham has a good 2004, but can't replace the quality of player CHI found in their defensemen like Keith, Seabrook, Hjalmarsson, and Byfuglien.

    Sham Method #3 - Highest Age Adjusted Points per Game of Next 30 Ranked Players:

    Chicago D

    Given Chicago's massive number of picks, Sham finds a gaudy amount of NHL talent. He drafts over 10,100 games into the NHL, which would be by far and away the most of any team. Unfortunately, the quality of player just isn't there.

    The Verdict: Part of drafting is luck, and part is science. And let's face it: had Chicago or anyone known that Duncan Keith would become Duncan Keith, he wouldn't have been passed over entirely in his draft year, and taken late in his 2nd year of eligibility. Sham finds an astounding number of NHL-calibre pieces, but no superstars like the Hawks did. Win for Chicago.

    zPass

    Colorado Avalanche

    Actual Draft Record:

    Colorado A

    Not much great quality in the early 2000's, but have an amazing 2009.

    Sham Method #1 - Most Points:

    Colorado B

    Fewer games played, but Claude Giroux, Evander Kane, Brendan Gallagher, Brandon Dubinsky, and a good 2002 makes this crop tantalizing.

    Sham Method #2 - Most Points of Next 30 Ranked Players by CSS:

    Colorado C

    Again, less quantity than the Avs' draft record, but Giroux, Kane, and Williams all carry serious weight.

    Sham Method #3 - Highest Age Adjusted Points per Game of Next 30 Ranked Players:

    Colorado D

    Sham really needs a Francois Bouchard shirt. As soon as he takes age into account, he's all "more like Claude Gir-who?"

    The Verdict: In the end, foregoing Paul Stastny, Kevin Shattenkirk, Chris Stewart along with Duchene, O'Reilly et. all is far too large a price to pay for Giroux, especially when Matt Duchene may get to his level anyways. The Avs weren't good early in the 2000's, and Sham would have helped then, but their drafting of late has been impressive.

    zPass

    Columbus Blue Jackets

    Actual Draft Record:

    Columbus A

    Between Filatov, Picard, Brule, and Leclaire, Columbus has a lot of guys that never really lived up to their lofty potential - and you can arguably throw Nikolai Zherdev in that group too. Let's see if Sham can do better.

    Sham Method #1 - Most Points:

    Columbus B

    Sham Method #2 - Most Points of Next 30 Ranked Players by CSS:

    Columbus C

    Sham Method #3 - Highest Age Adjusted Points per Game of Next 30 Ranked Players:

    Columbus D

    Sham drafts an astounding 10,498 games of future NHL talent during his tenure with the Blue Jackets, by far and away the most of any team, real or Shammed.

    The Verdict: Even though they'd lack a true #1 centre with Dubinsky - Brassard - Bailey - Stajan down the middle, the Jackets would boast absurd depth on the wings under Sham. Hemsky - Voracek/Skinner - Pominville/MacArthur - Hartnell is as deep a top-9 W group as you'd find in the NHL today. Oh what should have been, Jackets fans.

    zFail

    Dallas Stars

    Actual Draft Record:

    Dallas A

    Sham can't draft Jamie Benn. Might as well give the Stars the win right now.

    Sham Method #1 - Most Points:

    Dallas B

    Sham Method #2 - Most Points of Next 30 Ranked Players by CSS:

    Dallas C

    Sham Method #3 - Highest Age Adjusted Points per Game of Next 30 Ranked Players:

    Dallas D

    Some nice pieces, but missing Benn definitely hurts.

    The Verdict: Any time a team gets top-end talent outside the 1st round from a non-traditional source, it's a boon for a franchise and extremely hard to match. Sham once again drafts more NHL games than the Stars do, but can't quite match Jamie Benn and/or James Neal, though it's close with guys like Williams, Pominville, and Dubinsky.

    zPass

    Detroit Red Wings

    Actual Draft Record:

    Detroit A

    Sham Method #1 - Most Points:

    Detroit B

    That's... really not good. Detroit has had a lousy draft position that they've had to contend with, and it shows in Sham's results.

    Sham Method #2 - Most Points of Next 30 Ranked Players by CSS:

    Detroit C

    Sham Method #3 - Highest Age Adjusted Points per Game of Next 30 Ranked Players:

    Detroit D

    There's really not a lot there. As good as Sham was with Columbus, he is as bad with Detroit.

    The Verdict: The last time we visited Detroit's drafting record, I personally cast some doubt on if they were really as good as the narratives held them up to be, or if they were still dining off Datsyuk and Zetterberg. As it turns out, they really are good at finding talent despite an awful draft position. Sham makes a mess of the Wings, so Detroit gets an easy pass.

    zPass

    Edmonton Oilers

    Actual Draft Record:

    Edmonton A

    Sham Method #1 - Most Points:

    Edmonton B

    The Oilers have managed to draft a lot of guys who sneak into 100 or so NHL games before disappearing off the face of the Earth. Sham doesn't find as many of those guys, nor does he find Jordan Eberle, but he still turns up quite a few quality forwards.

    Sham Method #2 - Most Points of Next 30 Ranked Players by CSS:

    Edmonton C

    Sham Method #3 - Highest Age Adjusted Points per Game of Next 30 Ranked Players:

    Edmonton D

    Take away Paajarvi, Eberle, and Cogliano, but add Boyes, Pominville, Clowe, Stajan, Talbot, Gorges, Bolland, Dubinsky, Ennis, and Ellis? I think that's a good deal.

    The Verdict: lolilers.

    zFail

    Florida Panthers

    Actual Draft Record:

    Florida A

    When the best player you've drafted since 2000 is Jay Bouwmeester, you're doing something wrong.

    Sham Method #1 - Most Points:

    Florida B

    Jason Pominville would be Florida's all time leading scorer by over 100 points.

    Sham Method #2 - Most Points of Next 30 Ranked Players by CSS:

    Florida C

    Pierre-Marc Bouchard and Matt Stajan would also be among the top-5 all-time leading scorers in Panthers history.

    Sham Method #3 - Highest Age Adjusted Points per Game of Next 30 Ranked Players:

    Florida D

    Jeff Skinner would be 9th all time in goals for Florida. Jeff Skinner just turned 22 years old.

    The Verdict: With a large selection of high picks, it's a real head-scratcher as to why Florida hasn't done better at the draft table. Sham drafts far more future games for them, and Skinner, Little, Pominville, Hemsky, Dubinsky, and MacArthur is probably the strongest top-6 forwards the Panthers have had.

    zFail

    Los Angeles Kings

    Actual Draft Record:

    Los Angeles A

    Los Angeles has consistently been one of the best teams at the draft table, even before Dean Lombardi took over. It's going to take a huge showing from Sham to out-do Kopitar and Doughty.

    Sham Method #1 - Most Points:

    Los Angeles B

    Bryan Little and Claude Giroux in 2006 is a fantastic haul, but Alex Bourret in '05 over Kopitar and Josh Bailey and Tyler Ennis in '08 over Doughty is less than ideal.

    Sham Method #2 - Most Points of Next 30 Ranked Players by CSS:

    Los Angeles C

    Sham Method #3 - Highest Age Adjusted Points per Game of Next 30 Ranked Players:

    Los Angeles D

    Adjusting for age causes LA to whiff on Francois Bouchard over Claude Giroux in '06, but it instead finds Patrice Bergeron in 2003.

    The Verdict: Despite not being able to select Anze Kopitar of Drew Doughty, Sham makes it surprisingly close. Switching Bouchard for Claude Giroux in 2006 of method #3 may have even tipped the scales in Sham's favour.

    Colour me shocked then that the general manager of a two-time cup winner is smarter than a literal potato on a keyboard.

    zPass

    Minnesota Wild

    Actual Draft Record:

    Minnesota A

    The Wild cleaned up in the early part of the decade, but their drafts have been pockmarked with high strikeouts since Brent Burns in 2003.

    Sham Method #1 - Most Points:

    Minnesota B

    Sham Method #2 - Most Points of Next 30 Ranked Players by CSS:

    Minnesota C

    *Clutterbuck in round 6 of 2006 is a glitch. We can assume he was off the board by the 6th round and therefore unavailable.

    Sham Method #3 - Highest Age Adjusted Points per Game of Next 30 Ranked Players:

    Minnesota D

    The Verdict: The Wild won't find Koivu, Gaborik, or Burns, but Hemsky, Little, Pominville, MacArthur, and Hartnell probably make up for it. Sham finds slightly fewer games, but compensates with some deeper and better talent up front.

    zFail

    Montreal Canadiens

    Actual Draft Record:

    Montreal A

    Montreal's 2007 draft was fantastic. McDonagh - Subban may be the best two D taken by one team in one draft in years. Then they traded one for Scott Gomez. Oops.

    Sham Method #1 - Most Points:

    Montreal B

    Claude Giroux would be a god in francophone Canada. Even though he's from Hearst, Ontario.

    Sham Method #2 - Most Points of Next 30 Ranked Players by CSS:

    Montreal C

    2005 goes from a pivotal year in deciding the future of the Habs to a Gilbert Brule-related washout. One wonders if Brule grows into his talent by not breaking his leg with Columbus.

    Sham Method #3 - Highest Age Adjusted Points per Game of Next 30 Ranked Players:

    Montreal D

    The Verdict: Montreal's 2007 is simply too good to pass up. Now about that Gomez trade...

    zPass

    Nashville Predators

    Actual Draft Record:

    Nashville A

    Sham is, in a lot of ways, the anti-Predators. The Preds love their D-men. Sham thinks they're voodoo.

    Sham Method #1 - Most Points:

    Nashville B

    Sham Method #2 - Most Points of Next 30 Ranked Players by CSS:

    Nashville C

    Sham Method #3 - Highest Age Adjusted Points per Game of Next 30 Ranked Players:

    Nashville D

    Both Scott Hartnell and Jason Pominville would lead Nashville in all-time goals. They would also be 2nd and 3rd in all time points, behind David Legwand.

    The Verdict: With a dearth of star forwards in their franchise's history, there was a good chance for Sham to out-draft the Preds. Unfortunately, Sham wasn't able to dig up any real star power. As a result, it would be too much to fill the shoes left by Weber, Suter, and co.

    zPass

    New Jersey Devils

    Actual Draft Record:

    New Jersey A

    We have Zach Parise aaaaand no one else. Well, Zajac and Henrique exist, but still. That's ugly.

    Sham Method #1 - Most Points:

    New Jersey B

    Sham Method #2 - Most Points of Next 30 Ranked Players by CSS:

    New Jersey C

    *Note: Henrique with the 9th pick in 2008 is a glitch. He would be taken by that time.

    Sham Method #3 - Highest Age Adjusted Points per Game of Next 30 Ranked Players:

    New Jersey D

    Sham's record isn't great with the Devils, partly because they've had bad draft positions for a while now. Still, Sham likely finds more useful forwards in Williams, Ennis, Richards, etc.

    The Verdict: Come to think of it, Lou Lamoreillo kinda does look like a potato.

    zFail

    New York Islanders

    Actual Draft Record:

    New York Islanders A

    Sham is pretty much 100% guaranteed to take John Tavares, so it's just a matter of whether he can find a better group than Okposo, Bailey, Nielsen, Torres, and assorted other stuff.

    Sham Method #1 - Most Points:

    New York Islanders B

    Sham Method #2 - Most Points of Next 30 Ranked Players by CSS:

    New York Islanders C

    Sham Method #3 - Highest Age Adjusted Points per Game of Next 30 Ranked Players:

    New York Islanders D

    Do you trade John Tavares for Evander Kane and Jeff Skinner? What if Little is thrown in? That's an entire 1st line. I tend to lean "no," but it's actually an interesting question.

    The Verdict: Tavares makes it close, but Sham ultimately drafts a deeper, and presumably more competitive, roster for the Isles (although goaltending is still an issue). Party on, Garth.

    zFail

    New York Rangers

    Actual Draft Record:

    New York Rangers A

    Sham Method #1 - Most Points:

    New York Rangers B

    Sham Method #2 - Most Points of Next 30 Ranked Players by CSS:

    New York Rangers C

    Somewhere, in an alternate universe, Claude Giroux plays for the Rangers.

    Sham Method #3 - Highest Age Adjusted Points per Game of Next 30 Ranked Players:

    New York Rangers D

    The Verdict: The real Rangers out-draft Sham in terms of games played in each version, and also quality too. They find Henrik Lundqvist in 2000, and a good number of impact players after round 1. Ol' Slats has still got it.

    zPass

    Ottawa Senators

    Actual Draft Record:

    Ottawa A

    Sham Method #1 - Most Points:

    Ottawa B

    Sham Method #2 - Most Points of Next 30 Ranked Players by CSS:

    Ottawa C

    Sham Method #3 - Highest Age Adjusted Points per Game of Next 30 Ranked Players:

    Ottawa D

    The Verdict: Ottawa drafted extremely well that the start of the 2000's, and was largely matched by Sham. Where the Sens really pull away is when they land Erik Karlsson in 2008. Sham simply cannot find a player of that quality outside the top-3 picks in North America.

    zPass

    Philadelphia Flyers

    Actual Draft Results:

    Philadelphia A

    Seeing as Sham seems to really like Nigel Dawes in 2003, it's going to be hard to match the Flyers that year.

    Sham Method #1 - Most Points:

    Philadelphia B

    Sham Method #2 - Most Points of Next 30 Ranked Players by CSS:

    Philadelphia C

    Sham Method #3 - Highest Age Adjusted Points per Game of Next 30 Ranked Players:

    Philadelphia D

    Francois Bouchard screws up another good draft.

    The Verdict: Sorry, Sham. Steve Bernier is not the next Cam Neely.

    zPass

    Phoenix Coyotes

    Actual Draft Record:

    Phoenix A

    The 'Yotes had the worst 2003 of any team in the NHL. In the best draft in league history, they found a grand total of 0 NHL GP.

    Sham Method #1 - Most Points:

    Phoenix B

    Sham Method #2 - Most Points of Next 30 Ranked Players by CSS:

    Phoenix C

    Sham Method #3 - Highest Age Adjusted Points per Game of Next 30 Ranked Players:

    Phoenix D

    The Verdict: Keith Yandle and Oliver Ekman-Larsson are good, but not good enough to make up for the 2,000 more GP and 700 more goals that Sham drafts into the NHL. Hemsky and Pominville are also the best two forwards the Desert Dogs draft since Shane Doan.

    zFail

    Pittsburgh Penguins

    Actual Draft Record:

    Pittsburgh A

    The Pens draft record is pretty unspectacular. Except for those Crosby and Malkin guys.

    Sham Method #1 - Most Points:

    Pittsburgh B

    lol Alex Picard over Evgeni Malkin.

    Sham Method #2 - Most Points of Next 30 Ranked Players by CSS:

    Pittsburgh C

    Sham Method #3 - Highest Age Adjusted Points per Game of Next 30 Ranked Players:

    Pittsburgh D

    The Verdict: As with any team fortunate enough to land a European superstar, it's tough for Sham to keep pace. The CHLers he drafts are comparable to those from Pittsburgh, but it certainly seems like eschewing "enigmatic" Russians in favour of more gritty North American skaters is a losing recipe.

    zPass

    San Jose Sharks

    Actual Draft Record:

    San Jose A

    Sham Method #1 - Most Points:

    San Jose B

    Claude Giroux in a Sharks jersey terrifies me.

    Sham Method #2 - Most Points of Next 30 Ranked Players by CSS:

    San Jose C

    Sham Method #3 - Highest Age Adjusted Points per Game of Next 30 Ranked Players:

    San Jose D

    The Verdict: If it wasn't for Francois Bouchard having a better age adjusted points per game in his draft year than Claude Giroux (or Brett MacLean having 0.03 more pts/GP than Logan Couture), this would most likely a win for Sham. As it stands though, Sham foregoes a bunch of really good D like Marc-Edouard Vlasic, as well as Pavelski and Couture to take Giroux and Mike Richards. Advantage goes to the real Sharks.

    zPass

    St. Louis Blues

    Actual Draft Record:

    St Louis A

    Sham Method #1 - Most Points:

    St Louis B

    Sham Method #2 - Most Points of Next 30 Ranked Players by CSS:

    St Louis C

    Sham Method #3 - Highest Age Adjusted Points per Game of Next 30 Ranked Players:

    St Louis D

    The Verdict: There is simply no one here that is the calibre of an Alex Pietrangelo or a David Backes. Sham wins again on the basis of games played, but the Blues draft better players.

    zPass

    Tampa Bay Lightning

    Actual Draft Record:

    Tampa Bay A

    Tampa didn't really get a legitimate above average NHLer until 2008 - and Steven Stamkos has scored half of all goals Bolts picks from this 11-year span have scored.

    Sham Method #1 - Most Points:

    Tampa Bay B

    Add Evander Kane, Claude Giroux, and co. and Tampa Bay is already getting killed by Sham.

    Sham Method #2 - Most Points of Next 30 Ranked Players by CSS:

    Tampa Bay C

    Sham Method #3 - Highest Age Adjusted Points per Game of Next 30 Ranked Players:

    Tampa Bay D

    The Verdict: While the Lightning get Hedman as well as Stamkos, Sham finds at least two other star forwards as well as a number of other useful depth players. Bolts goal scoring by draft picks quadruples under Sham, and he finds twice as many future games played.

    zFail

    Toronto Maple Leafs

    Actual Draft Record:

    Toronto A

    Until recently, that's actually not that bad.

    Sham Method #1 - Most Points:

    Toronto B

    Sham Method #2 - Most Points of Next 30 Ranked Players by CSS:

    Toronto C

    Kessel - Giroux - Williams is a pretty sick 1st line IMHO. Pominville - Bailey - Gallagher is a pretty solid L2 too.

    Sham Method #3 - Highest Age Adjusted Points per Game of Next 30 Ranked Players:

    Toronto D

    Francois Bouchard is significantly less sick, IMHO.

    The Verdict: Interestingly enough, the Leafs biggest issues haven't been at the draft table, though I suspect if we run this test again in 10 years, these last few drafts could end up looking pretty awful. Instead, Toronto has traded some really good pieces like Alex Steen and Tuukka Rask for pennies on the dollar.

    Still, the allure of Claude Giroux is too much. While you beat Sham in games played, you still didn't find as good players as the potato did. Potato for HHOF 2015.

    zFail

    Vancouver Canucks

    Actual Draft Record:

    Vancouver A

    We've been down this road before. More than once. Multiple times even. So many times. Here we go again.

    Sham Method #1 - Most Points:

    Vancouver B

    First Line Centre Henrik Sedin, Second Line Centre Claude Giroux.

    Sham Method #2 - Most Points of Next 30 Ranked Players by CSS:

    Vancouver C

    Third Line Centre Mike Richards. Fourth Line Centre David Bolland.

    Sham Method #3 - Highest Age Adjusted Points per Game of Next 30 Ranked Players:

    Vancouver D

    I really want to know what held Francois Bouchard back. Same everything as Giroux, but one turned into a star and the other spent the last season with Oskarshamn IK.

    The Verdict:

    Sedin - Sedin - Burrows

    Pominville - Giroux - Williams

    Ennis - Richards - Gallagher

    Higgins - Bolland - Torres

    zFail

    Washington Capitals

    Actual Draft Record:

    Washington A

    No Ovi? No problem. I hear he's overrated anyways.

    Sham Method #1 - Most Points:

    Washington B

    Scratch that. Big problem.

    Sham Method #2 - Most Points of Next 30 Ranked Players by CSS:

    Washington C

    *Note: Mike Green with the 13th pick in 2004 is a glitch. Assume he is not taken by WSH that year.

    At the very least, the Caps have a lot more grit with Richards, Ladd, Bolland, Talbot, Stoll, and Williams.

    Sham Method #3 - Highest Age Adjusted Points per Game of Next 30 Ranked Players:

    Washington D

    *Note: Green in '04 is also a glitch here.

    The Verdict: So, uh, maybe the Capitals aren't better off without Ovi after all.

    zPass

    Winnipeg Jets

    Actual Draft Record:

    Winnipeg A

    Maybe it's a good thing if the then-Thrashers fail. Success may have meant a longer stay in Atlanta.

    Sham Method #1 - Most Points:

    Winnipeg B

    Sham Method #2 - Most Points of Next 30 Ranked Players by CSS:

    Winnipeg C

    Sham Method #3 - Highest Age Adjusted Points per Game of Next 30 Ranked Players:

    Winnipeg D

    The Verdict: Although Sham doesn't really find a second star to replace either Heatley or Kovalchuk, he makes up for it by adding roughly 2000 more GP of NHL talent, as well as 400 more goals and almost 1,000 more assists. It's close, but Sham gets the decision.

    zFail

    Discussion

    In all, we had 17 teams pass the Sham Sharron drafting test, and 13 teams fail. However, as I mentioned right at the beginning, these grades are largely based on my opinion and are wide open to debate and disagreement. Building a concrete quantitative analysis off of these findings is definitely in the plans for the future, but we'll save the detailed numerical breakdowns for another day.

    We do have some data to share though. The following table contains each team's total drafted games played, goals, assists, and points under their actual scouting departments and the different variations of Sham. Green indicates that the team beat Sham by more than 750 games (remember, 750 games = $2 m/yr investment the average small market American team spends on amateur scouting annually), yellow indicates the team beat Sham by less than 750 games, while red indicates the team lost to Sham (click to make bigger):

    z Results Table


    You'll notice that the more "complex" versions of Sham's method perform at around the same level as the average NHL team. This indicates that, on average, NHL teams are deriving no value from keeping a private scouting staff. Obviously there are exceptions - the Kings, Sabres, and Red Wings come to mind - but it's fairly clear that whatever the Canucks, Lightning, Panthers, and Coyotes have been doing has been a waste of resources. All four teams would have been better off ignoring half the world's talent pool and drafting via a simple formula.

    If that sounds insane, that's because it should sound insane. These are multi-million dollar businesses whose success depends on being able to find hockey talent and ice a competitive team. They should, in theory, kick a layman's ass when it comes to this stuff. Josh and myself shouldn't be able to build a system this easily that rivals any NHL team. Sham really shouldn't be close to anyone. But yet, he is. 

    Keep in mind that Josh and myself designed a system where nothing that couldn't have been known to scouts at the time was used, we closed ourselves off from half of the world's hockey talent, and we were biased against defensemen. We addressed the concerns about old Sham's magical foresight and came to the same conclusions as we did back then: the Canucks are traditionally really bad at this, and whatever "traditional" amateur scouting consists of is broken and needs to be fixed.

    Sham is, at the end of the day, still a poor way to scout players. It relies on data that we know is widely variable from one year to the next, and doesn't account for most of the variables that make up a hockey player. I am confident that we can refine Sham into something reliable and useful, and maybe that happens one day down the road. Until then, teams like the Canucks and Bolts and Panthers and Oilers provide a massive market inefficiency that other teams can and have repeatedly capitalized on. I introduced the concept of opportunity costs at the draft in this post, and the more that poor drafting teams screw up, the easier it is for other, smarter, teams to realize a net benefit. You're essentially giving your opponents a free head start on you.

    As a fan, all we can do is hope that if your team has traditionally been bad, that they've learned from their past mistakes. And if they've been good, we can hope that they're committed to getting even better. The first day of the rest of your franchise's life starts tomorrow, so here's to hoping it goes well (especially for the Canucks).

    With that, I'll leave you with Sham Sharron's top 30 CHL prospects, ranked by most points in the CSS next 30 highest rated players. Thanks for reading:

    z Sham Draft Rankings

    10df053f41f4bdaf4d65eca6e982e46d
    Rhys Jessop - CanucksArmy's resident CHL buff/hater of prospects, depending on who you ask. Follow my ramblings on Twitter: @Thats_Offside. Email contact: thats.offside.hockey@gmail.com
    Avatar
    #1 Jason
    June 26 2014, 11:09AM
    Trash it!
    8
    trashes
    Props
    17
    props

    This is obviously a terrible idea, even though you are insisting that "any NHL team should draft in the same manner as Sham." I can't believe you even said "We did [...] design Sham Sharron to be a good or intelligent way to draft."

    Obviously you are fools and this is ridiculous. The Canucks have drafted perfectly and are sure to win the Cup this year.

    Avatar
    #2 Peachy
    June 26 2014, 11:37AM
    Trash it!
    2
    trashes
    Props
    8
    props

    Very impressive piece of work.

    Avatar
    #3 Ragnarok Ouroboros
    June 26 2014, 11:54AM
    Trash it!
    3
    trashes
    Props
    4
    props

    Good article, but the images for the drafts from ST. Louis onward are not appearing. I hope Benning brings a better approach for drafting for the Canucks.

    Avatar
    #4 Chungus
    June 26 2014, 12:01PM
    Trash it!
    2
    trashes
    Props
    10
    props

    @Jason

    Nice try... Fell a little flat.

    Avatar
    #5 Lee
    June 26 2014, 12:07PM
    Trash it!
    0
    trashes
    Props
    2
    props

    The images after the last one for SJS are all very small and unreadable, and clicking on them takes you to postimg.org homepage. Anyone else seeing this?

    Avatar
    #6 NM00
    June 26 2014, 01:01PM
    Trash it!
    2
    trashes
    Props
    13
    props

    First off, I want to applaud Rhys and Josh for their tremendous work on this project.

    Rhys (and Josh): I clearly spoke too soon when deeming Cam the most realistic blogger on here...

    A few things though.

    While I respect that games played/points is a simple proxy for value and you openly admit that it is imperfect, could this be tweaked (minutes?) so that teams aren't being dinged for drafting good players that don't rack up the points (such as a top goalie, for example)?

    As an example, perhaps a 1.33 adjustment for defenseman compared to forwards (average forward plays 15 mins, average defenseman plays 20 mins).

    Since Sham is almost exclusively taking forwards, it's going to be much easier for his selections to rack up the games played/points.

    Are his players actually more valuable, though?

    Second, while I repeat that I respect and appreciate the effort put into this, I continue to disagree that beating less than half the NHL teams is an accomplishment for this or any other system.

    As I inititally suspected, you may be able to beat quite a few teams with Sham (or any system, really).

    But teams that do well outside of the CHL (such as Detroit) is where this system is limited.

    Finally, in response to Cam's post about beating a "complex scouting system" as well as the repeated assertions that you are not advocating that teams actually use Sham...

    How much real world value to an NHL front office is there if you have not found a system that can beat (significantly) more than half the teams in the NHL?

    Some teams are inevitably going to do better than they "should" have (relative to draft position) and some teams are going to do worse.

    30 teams are fighting over the same pool of players in a zero sum game.

    Somebody wins and somebody loses.

    I guess my main interest in this is if you stumble upon something that would be useful to an NHL front office that they do not already know...

    I'm curious about two more things.

    1. How would Sham do by simply taking the highest player left on the board of a publicly available ranking system.

    2. Are teams getting better? Was Sham more useful from 2000-2004 (or from 1990-1999) than from 2005-2010?

    Very good post!

    Avatar
    #7 TheShadierTwin
    June 26 2014, 01:06PM
    Trash it!
    0
    trashes
    Props
    4
    props
    Jason wrote:

    This is obviously a terrible idea, even though you are insisting that "any NHL team should draft in the same manner as Sham." I can't believe you even said "We did [...] design Sham Sharron to be a good or intelligent way to draft."

    Obviously you are fools and this is ridiculous. The Canucks have drafted perfectly and are sure to win the Cup this year.

    HAH! More fool you. With our new Potato GM and with head coach Toastbuster keeping our picks in line, we are guaranteed to see the Leafs take the cup in 2015!

    Avatar
    #8 Mack
    June 26 2014, 03:54PM
    Trash it!
    0
    trashes
    Props
    4
    props

    great work guys!

    Avatar
    #9 Locky
    June 26 2014, 08:47PM
    Trash it!
    0
    trashes
    Props
    2
    props

    Much better article than the first iteration. Good job.

    "If that sounds insane, that's because it should sound insane."

    It should. It illustrates very well the old boys club tradition that permeates hockey management and, especially, the media.

    Reminds me of this: http://www.youtube.com/watch?v=EEneggSHic4

    Avatar
    #10 Goon
    June 26 2014, 08:54PM
    Trash it!
    0
    trashes
    Props
    0
    props

    This is awesome work but holy crap some of these tables are hard to read - can you embed links in the images to full-sized versions so we're actually able to read the Chicago ones and a few others?

    Thanks!

    Avatar
    #11 Dan
    June 27 2014, 09:23AM
    Trash it!
    0
    trashes
    Props
    1
    props

    First off: love it.

    Second: I think you're giving NHL teams a little bit too much credit for my liking in the subjective pass/fail , but it's still incredible how close Sham comes to beating the NHL teams. With some minor tweaks to fix Sham's systematic exclusion of certain players - for example, the only reason Sham likes Derick Brassard, Rob Schremp, and Josh Bailey so much is because guys like Evgeny Malkin, Jonathan Toews, Nicklas Backstrom, Phil Kessel, and Drew Doughty were all systematically excluded - Sham would have surely been a better drafter.

    In other words, I'm saying that given how close things are as they stand now, a potato with a slightly less exclusive system would have surely been a better drafter than real NHL teams. The only reason that this particular potato didn't out-draft most of the NHL is because you forced him - side note: do potatoes take on a gender? - to take all of his draft picks from Canada.

    Third: in the interest of helping Sham succeed, do the results change if we use points/game rather than total points? We'd of course need some sort of reasonable minimum game threshold, but the reasoning behind this suggestion is that the top players frequently leave their teams to participate in international tournaments (World Juniors, Ivan Hlinka, etc.). Using total points handicaps the best scorers because they have 10-20 fewer games to produce points in and, by extension, handicaps Sham's ability to draft the players that are truly the best scorers.

    Comments are closed for this article.