Measuring Draft Performance

This is a photograph that is not destined to become famous. That’s Kevin Prendergast with Jesse Niinimaki, Kevin Lowe and Lorne Davis.

The Oilers have just drafted the Finn, the Habs have left the stage with Chris Higgins and a couple of picks later Los Angeles will take Denis Grebeshkov. Jarret Stoll is wondering if he’ll go on day one.

Such is the nature of the draft.

I followed drafts as a kid. I remember the 1971 draft and I anticipated the 1973 Amateur Draft. If my own son looked forward to something like a hockey league’s annual draft there’s a chance I’d slide him into a room with an “expert” in the field of social interaction but there you go I’ve confessed it.

Anyway, back then you waited for the Hockey News (I do recall a newspaper article detailing Denis Potvin as the top overall pick and that Atlanta had drafted Tom Lysiak from Medicine Hat) and they gave you the background (one of the things that all this information robs one of is waiting for something you really want. The July issue of the Hockey News was such a thing) on the draft and the trades.

The one thing I remember most about that draft was that Bob Gainey’s scoring numbers were awful for a forward. He had 22 goals in 52 games in the OHA and the guy drafted before him (Blaine Stoughton) had 58 goals in 66 games in the WHL. There was a reference to his checking ability but you’ve got to be a helluva checker to go 8th overall when you can’t pot 30 in junior at age 19. In any era.

One thing I’ve always looked for is a way to measure draft performance by team. That’s a little different than measuring it by league (I think we can all agree scouts do a damn good job in grabbing most of the top talent each season by selection 100) and it requires some “handicapping” for where a team is selecting each player.

I’ve toyed with this for decades (which really speaks volumes about how I spend my time) and “games played” has always been the line in the sand that works best because all positions are represented. Goalies get punished the most by this measure but we can make allowances for it just by being aware of it and make it close enough for jazz.

If we attempted to “tax” the high picks and let the selections after #100 remain “tax free” could we come up with a way to measure individual draft performances. What would it look like? Would it make sense?

Let’s try doing one right now. Let’s say that for a specific draft year we’ll “tax” thusly:

  1. Top 10 picks: 60 games per season after age 20
  2. Picks 11-25: 40 games per season after age 20
  3. Picks 26-50: 30 games per season after age 20
  4. Picks 51-100: 20 games per season after age 20
  5. Picks 100+ Tax free

Since we have no idea what it will look like, let’s roll with it and come back to the tax system later (I find that funny in a Brian Mulroney kind go way). I’m going to use the 2001 Entry Draft since Kevin Prendergast is going to be the subject of this blog over the next while and let’s do the entire Canadian division.

2001 Edmonton Oilers Draft (+80)

  • 13th Ales Hemsky (GP: 322-160tax= +162)
  • 43rd Doug Lynch (GP: 2-120tax= -118)
  • 52nd Eddie Caron (GP: 0-80tax= -80)
  • 84th Kenny Smith (GP: 0-80tax= -80)
  • 133rd Jussi Markkanen (GP: +128 no tax)
  • 154th Jake Brenk (GP: 0 no tax)
  • 185th Mikael Svensk (GP: 0 no tax)
  • 215th Dan Baum (GP: 0 no tax)
  • 248th Kari Haakana (GP: +13 no tax)
  • 272nd Ales Pisa (GP: +53 no tax)
  • 278th Shay Stephenson (GP: +2 no tax)

Does that look fair? We penalize them for the 3 picks in the top 100 that didn’t work out but they should be, right? Prendergast himself is always talking about not making mistakes in the top 100 and so it had a major impact. Hemsky’s emergence at 19 and Markkanen being a useful player made up for it plus a little bit. If I were going to change the system at all it might be to make the penalty for that 84th pick less severe but let’s keep it for now.

2001 Calgary Flames Draft (-16)

  • 14th Chuck Kobasew (GP: 272-160tax= +112)
  • 41st Andrei Taratukhin (GP: 0-120tax= -120)
  • 56th Andrei Medvedev (GP: 0-80tax= -80)
  • 108th Tomi Maki (GP: +1 no tax)
  • 124th Igor Shastin (GP 0 no tax)
  • 145th Jim Hakewill (GP 0 no tax)
  • 164th Yuri Trubachev (GP 0 no tax)
  • 207th Garett Bembridge (GP 0 no tax)
  • 220th David Moss (GP +71 no tax)
  • 233rd Joe Campbell (GP 0 no tax)
  • 251st Ville Hamalainen (GP 0 no tax)

Looks right to me. Like Edmonton, Calgary gets rewarded for getting a player who covered the bet in the first round and then paid heavily for the misses later in the top 100. A questions that we can think about here are “does this system penalize the players 26-100 too heavily since their chances of being anything aren’t statistically great?”

2001 Vancouver Canucks Draft (+172)

  • 16th RJ Umberger (GP: 203-160tax= +43)
  • 66th Fedor Fedorov (GP: 18-80= -62)
  • 114th Evgeny Gladskikh (GP 0 no tax)
  • 151st Kevin Bieksa (GP 132 no tax)
  • 212th Jason King (GP 59 no tax)
  • 245th Konstantin Mikhailov (GP 0 no tax)

Nice draft by the Canucks, they got a nice first round pick and then a very useful player in the no tax zone. Plus Jason King who has played in the NHL about as much as Marc Pouliot.

2001 Ottawa Senators Draft (+621)

  • 2nd Jason Spezza (GP: 293-240tax= +53)
  • 23rd Tim Gleason (GP: 237-160tax= +77)
  • 81st Neil Komadoski (GP: 0-80tax= -80)
  • 99th Ray Emery (GP: 123-80tax= +43)
  • 127th Christoph Schubert (GP 189 no tax)
  • 162nd Stefan Schuauer (GP 0 no tax)
  • 193rd Brooks Laich (GP 204 no tax)
  • 218th Jan Platil (GP 0 no tax)
  • 223rd Brandon Bochenski (GP 113 no tax)
  • 235th Neil Petruic (GP 0 no tax)
  • 256th Gregg Johnson (GP 0 no tax)
  • 286th Toni Dahlman (GP 22 no tax)

That was a nice draft. Spezza points out an interesting problem for this. If we were going to do this over a long period (say 25 years) would it be a good idea to end the tax at some point? Say after 10 years? I don’t know, just thinking out loud.

2001 Montreal Canadiens Draft (+5)

  • 7th Mike Komisarek (GP: 272-240tax= +32)
  • 25th Alexander Perezhogin (GP: 125-160tax= -35)
  • 37th Duncan Milroy (GP: 5-120tax= -115)
  • 71st Tomas Plekanec (GP 202-80tax= +122)
  • 109th Martti Jarventie (GP 1 no tax)
  • 171st Eric Himelfarb (GP 0 no tax)
  • 203rd Andrew Archer (GP 0 no tax)
  • 266th Viktor Ujcik (GP 0 no tax)

A draft that is better than it looks currently. Habs got 3 nice players in the top 100 who will more than cover their bet and Milroy should play some eventually.

2001 Toronto Maple Leafs Draft (-69)

  • 17th Carlo Colaiacovo (GP: 81-160tax= -79)
  • 39th Karl Pilar (GP: 90-120tax= -30)
  • 65th Brendan Bell (GP: 48-80tax= -32)
  • 82nd Jay Harrison (GP: 13-80tax= -67)
  • 88th Nicolas Corbeil (GP: 0-80tax= -80)
  • 134th Kyle Wellwood (GP: 166 no tax)
  • 168th Maxim Kondratiev (GP 40 no tax)
  • 183rd Jaroslav Sklenar (GP 0 no tax)
  • 198th Iva Kolozvary (GP 0 no tax)
  • 213th Jan Chovan (GP 0 no tax)
  • 246th Tomas Mojzis (GP 13 no tax)
  • 276th Mike Knoepfil (GP 0 no tax)

I think everyone associated with this draft has been fired. I should point out that injury has affected their draft but that goes for all teams and certainly affected the Oilers in the case of Doug Lynch.

Okay a few points and then I would like to ask you some questions (if you’ve read this far please keep going). Based on this one season I would say the Oilers draft department did well but could have done better. I would put them in the mid-level area with Vancouver. Of course the cream of the crop on this list is Ottawa, with Montreal off the pace of the mid-level teams, plus Calgary Toronto in trouble. The Leafs would have a very nice draft the following season, although I haven’t put it through this “tax system” I’m looking at here.

Okay, some questions for you.

  1. Does this kind of system tell us anything we wouldn’t know just by adding up the GP?
  2. Does it tell us anything worth knowing?
  3. Is the tax system a fair one? If not, where could it be improved?
  4. Should we give special consideration for players who are clearly at a level above the rest?
  5. Could we use this over a 25-year period without changing the tax?
  6. Any other ways to improve the system?

As an aside, the Habs draft is on display right now on HNIC if you didn’t know.

EDIT TO ADD: Fixed the Calgary Moss mistake, thanks to toqueboy for catching it.

written by

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

10 Responses to "Measuring Draft Performance"

  1. toqueboy says:

    i think that your calgary numbers are wrong. moss has only played 71 games…not 220 … so the flames would be in the minuses.

    i like the tax system but i think there has to be some relation to who draftees have played for…ie under milbury, the isles drafted quite well but traded them all…so raffi playing a significant portion of his ‘taxed’ games as an oiler…ought to be reflected somehow…i think…no idea how though…

    nice start on this though

  2. Lowetide says:

    Good catch, TB. Thank you.

    I’m not so much worried about the players and where they play and we can see that VAN and CGY actually got way less because the Canucks didn’t sign Umbarger and the Flames sent away Kobasew for less than 100 cents on the dollar.

    I see this kind of thing answering questions like “Is Kevin Prendergast and his scouting staff worth keeping?”

  3. speeds says:

    For me, it’s the same problem I always have with using GP as a measure of success – 13 games of Ales Hemsky is worth more than the 13 games played by Kari Haakana.

  4. Bruce says:

    Interesting post, LT. As for suggested improvements, to me the category 51-100 seems a little too broad to be all valued the same. How about making it 20 GP per season for 51-75 and 10 GP for 76-100?

    Another suggestion is to lower expectations slightly for picks 6-10 to 50 GP per. Taken together these suggestions would yield:

    1-5 = 60 GP per
    6-10 = 50 GP
    11-25 = 40 GP
    26-50 = 30 GP
    51-75 = 20 GP
    76-100 = 10 GP
    ——

    You could also align it more closely by draft round, e.g.:

    1-5 = 60 GP per
    6-15 = 50 GP
    16-30 = 40 GP
    31-60 = 30 GP
    61-90 = 20 GP
    91-120 = 10 GP
    All picks from fifth round on “tax free”.
    —–

    Another option would be to change the age threshold, e.g. GP after age 21 or 22 for the later round picks. But I think it’s cleaner to leave the age a constant and simply lower the expected GP for the deeper rounds.

  5. Lord Bob says:

    I think the most important change would be to make it a ‘per pick’ measurement. The Sens had a draft for the ages but they were sure helped by the fact that they picked a jillion guys.

    Games played is obviously a hideously imperfect measurement; to paraphrase Winston Churchill out of context, this is the worst system in the world except for all the other ones. I think we have to make do with GP for the foreseeable eternity.

  6. Lowetide says:

    Well we can do points per game but then we’re left with goalies completely out of the mix and defensemen severely punished.

  7. Eetu says:

    Since there is an exact number for each draft pick, why not make the “tax” more specific as well? The one suggested by Bruce seemed better to me, but it could be even more specific. What would be the expected number of games for a first pick? What about fourth, sixth, thirteenth or fifteenth (notice a pattern there ;-?)?

    Shouldn’t you also count in the quality of the pool of players each year as well? Ok, maybe this kind of measurement isn’t good for comparing drafts from different years, just for comparing the performance of clubs in the same draft.

    Then, of course, you should eliminate the advantage of having more picks by using the average (which isn’t a perfect way either since with more picks, you’re “allowed” to gamble more…)

    About that Niinimäki guy (he’s back in Finland once again, btw), I can see what they saw in him. He has all the talent in the world, crazy skillset, and he has a nice “frame”, as they say. But how on earth did they miss all the negatives? That he’s a lazy, inconsistent, weak, arrogant piece of shit who only plays for himself and hardly knows what “backcheck” means.

    Then again, the experts were right about 2002, it was a top-heavy draft with little depth. There weren’t too many great picks made after the the Oilers ruined theirs (Hannu Toivonen to Boston comes to mind).

    They were right about 2003 as well. Only one first round pick from that draft has no NHL games under his belt, so it was a relatively deep draft. You could argue that Pouliot in 2003 was a bigger mistake than Niinimäki in 2002. It isn’t true, but you could argue…

  8. Lord Bob says:

    About that Niinimäki guy (he’s back in Finland once again, btw), I can see what they saw in him. He has all the talent in the world, crazy skillset, and he has a nice “frame”, as they say. But how on earth did they miss all the negatives? That he’s a lazy, inconsistent, weak, arrogant piece of shit who only plays for himself and hardly knows what “backcheck” means.

    The same way that the previous administration missed the exact same signs with Jason Bonsignore and the same way that fans across the Oilogosphere were clamouring for the Oilers to take one of Esposito or Cherepanov last year.

    Skill is deadly. You do things for a player with skill that you wouldn’t do for anybody else in the world.

  9. digger says:

    I have to admit LT, I looked at this post and the only thing I could think of was:

    “Damn, Lowe’s sure aged over the past 5 years.”

  10. anthony bremond says:

    Dear Mr,

    I’m an intern with startup widget 1000 and I created a news widget for Los Angeles Kings News.
    It’s updated every hour with Los Angeles Kings headlines from several different sports websites like ESPN, FOX Yahoo,etc. (Take a look: http://www.widget1000.com/)

    Can we add Los Angeles Kings blog to the widget,too? That means that every time you post to your blog the headline of the post will show up on the widget and link directly to your post.
    Anyone who puts this widget on their start page or on their website will see your Los Angeles Kings posts too. It’s a powerful and free way to publicize your site.

    Finally, would you add your Los Angeles Kings News Widget to your site so that others can see it?
    Everyone who grabs the widget will automatically link to you so it’s great for your search engine rankings. Plus, it shows a lot of stories about Los Angeles Kings and since it’s updated every hour it is fresh and will make your site even more appealing than it already is.

    What do you think?

    Go Los Angeles Kings

    Anthony Bremond

Leave a Reply

Want to join the discussion?
Feel free to contribute!

Leave a Reply

© Copyright - Lowetide.ca