Behold: Math!

This is Willie Nelson. Before the long hair and the outlaw manner there was a gifted songwriter trying to make it the traditional Nashville way (which involves buses and waiting your turn, and if you are a gifted songwriter handing over your songs even though you’d like to keep them for yourself).

Willie Nelson wrote “Crazy” which is a song that gets sung by someone in some bar every night of the year. Patsy Cline had the hit and if you buy me enough beer I’ll sing the whole damn thing for you a cappella.

I remember years ago hearing Steve Earle tell the story of that period in time when college kids turned on to the Redheaded Stranger’s music. Earle detailed a typical country concert hall and a clash between the working men and women who wanted to blow off some steam and the college kids who wanted to get high, lean up against the wall and listen to the words. The college kids became sport for the drinking folk (between dances) and that’s Texas on a Saturday night (or so the Earle story goes).

That’s kind of the way I feel about the NHL scouting community. The scouts go to the dance for the traditional reasons and guys like me are hanging with the college kids, maybe borrowing a jay or buying a jug of beer while we wait for the boxscore.

The problem then comes in the fact that NHL scouts are really excellent at deciding who can play. I know Henrik and Pavel get all kinds of attention, but fact is that most of the good kids are plucked in the top 50 or so every season.

I have a friend who scouts junior players (not WHL, but it’s a paying gig and he’s held the job for years) and one day decided to ask him about hockey prospects and math. I detailed my theory on junior leagues (which is basically that any organized league can be measured in terms of the next level, you simply need to know how much air to let out of the tires) and he gave me a look that told me my sanity was being questioned at that very moment.

He told me many of the things that are contained in this article but it wasn’t until today that it clicked in for me. When Hakan Andersson is asked about Wings prospect Jesper Samuelsson he says:

  • “He needs to work on several other things to just be a regular in Sweden. Statistically, he will probably never make it to the NHL, I just want to see him improve year by year and then we will see how far that will take him.”

My take on this is that Andersson may not agree specifically with my math (well it’s Gabriel Desjardins’ math, but the idea has been around for awhile) but there is a “line in the sand” for each scout based on level of play, age, etc.

When the Detroit Red Wings selected Henrik Zetterberg in the 7th round of the 1999 draft they were certainly projecting, in fact I’d suggest it was a “draft and follow” selection. By that I mean that baseball clubs sometimes select players who have classic baseball bodies even though they’re skinny or have an injury in hopes that they come around with age. “Z” was a draft and follow. If he wasn’t a draft and follow, he’d have been taken higher.

According to the Central Scouting Bureau, Zetterberg was 5.10, 154 pounds on his draft day. Daniel Sedin was 6.02, 187, Henrik Sedin was 6.03, 196 same day. Zetterberg played in a lower level in Sweden (division 1) in 98-99 and scored 37gp, 15-13-28 for Timra. Daniel Sedin played that season in the SEL, scoring 21-21-42 in 50 games for MoDo. Henrik was 12-22-34 in 49 games for the same club. As a North American hockey fan, my assumption has always been that Swe-2 is a much lower level, but I’m beginning to think the difference isn’t that great.

I think Hakan Andersson saw terrific skills in Zetterberg but would also argue that math (28 points in 37 games) and level of play (considering age) put him above the event horizon as well. How many kids who weigh 154 pounds are competing at that level at age 17 in a league where much older men battle? Timra’s leading scorer that year was Mikael Lind, who was 26 years old at the time. I think most 154 pound Swedish kids were playing in Sweden’s junior leagues, far from the older men.

Gabriel Desjardins doesn’t have a number for Swe-2 (yet), but my guess is that Hakan Andersson does. Even if he doesn’t know it.

written by

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

18 Responses to "Behold: Math!"

  1. Fake Craig McTavish says:

    The Swedish word for math is aiseenemgud.

  2. Hawerchuk says:

    For some reason I thought I sent you the Swe-1 numbers. The league is around 0.3-0.4 relative to the SEL. So you’re looking at 0.2-0.3 – Tier I junior level; better than the ECHL.

  3. DBO says:

    LT: Mirtle has a good post (http://www.fromtherink.com/) about overseas goalies. Interesting reading, especially since we lack a big time goalie prospect and have question marks going forward regarding our tenders. What kind of math does Desjardins have concerning overseas talent? And does anyone think we should be taking a chance on some of these mid 20′s goalies (ala Hiller, Backstrom, Ersberg) and see what they can do. i’d sooner have them as a backup with the big club then JDD.

  4. Quinn says:

    DBO,

    Aren’t we doing alright for goalie prospects? I know we have nothing NHL-ready now, but it seems that DD is making good progress on the farm. Plus, with the absolute glut of decent 1B-type goalies in the league right now, spending a lot of effort developing them might be putting effort in the wrong place.

    But, still, an interesting questions of math and European tenders Can we get that guy from Latvia? I hear he’s stinking it up in Germany ;)

  5. DBO says:

    Quinn: DD is doing OK, but I don’t think anyone looks at him as a blue chip prospect. The reason I like the idea of bringing in a 25/28 year old overseas goalie with starting experience is that even as a back up they may be better off then JDD or DD. you can sign them for less then $1 mill per year and they could very well work out. More importantly, it gives us someone the coaching staff may be able to trust more then JDD, and not have to play our tender 40 games in a row.

  6. Quinn says:

    I agree with you on DD and JDD, but I am not sure going overseas to find someone unknown is that great an idea. Especially when you have lots of knowns in the NHL who are available in the $1-2 mill range. Clemmensen, Anderson, Halak, and so forth.

  7. Vic Ferrari says:

    I think that there are two really significant issues with relying on the numbers too much.

    Firstly, it’s the volatility. We all see that. A guy can be a goold player, and have a lot of the bounces go his way for a month, and put up stellar goals and points stats, ones that flatter him. The next month the bounces go the other way for him, chances juuussst fail to materialize when he’s on the ice, and he’s snakebit around the net and so are his linemates.

    The guy can be just as good over both timeframes, the results came at the caprice of the hockeys Gods.

    The other is context. I mean most young players in any league aren’t given much responsibility, in the CHL they usually look completely uncoached, they are the Gretzky’s of their team. One of the DET mgmt types (I linked to the article on this site some time ago) says that even though Zetterberg was a very young player in a man’s league. He was the best and most complete player on his team, which was poor. So he played a lot against the other team’s best, was on the ice for key draws in his own end, killed 5on3 penalties. All the kinds of things that are really important to winning hockey games, of course, but they hurt your counting stats.

    (Kenta Nilsson’s favourite “one that got away” story, of course. It always stings when I think about that. Z’s numbers were poor, he was small, and Nilsson had been the only scout to see him … so Barry Fraser vetoed him off the list.

    These things are also reflected in the variance of Gabe’s predictions. He shows them at his site, and if you included them in your posts (so say a 21 year old forward projects from the KHL as a 30 point guy … it might be wise to write that as 30 points is the midpoint of expectation, and there is an 80% chance of between 15 and 45 points).

    Those are just made up numbers, Gabe may confirm, but it wouldn’t surprise me if the spread really is that wide.

    I guess the point is that hockey fans always like guys who have come off of a big year, but context and good fortune may be bigger contributors to that than skill. I’m sure scouts lean towards the ncounting numbers too, and probably don’t give enough weight to context and luck, but they still give a lot more weight to those things than us.

  8. Jonathan Willis says:

    I guess the point is that hockey fans always like guys who have come off of a big year, but context and good fortune may be bigger contributors to that than skill. I’m sure scouts lean towards the counting numbers too, and probably don’t give enough weight to context and luck, but they still give a lot more weight to those things than us.

    Some scouts give good weight to context and luck, but I’d imagine that there’s a mixed bag there too – probably as mixed a bag as we see in NHL GM’s. For every Ken Holland there’s a Doug MacLean, ready to swoop in and offer a Todd Marchant both money and term. If anything, I’d guess the quality of scouts (on average) is worse than the quality of GM’s (on average) since theoretically a greater percentage of talent should rise to the top.

  9. Mr. Bugg says:

    LT: I can shed a little light on Zetterberg’s QOP in 98-99.

    ‘Division 2′- actually called Division 1, since the league below the SEL is named the Allsvenskan- is the most challenging league for scouts. I talked to Jim Nill once a couple of years ago about a kid I really liked out of Div 1, and he echoed the same sentiments- you’ve got 40 teams spread north to south and east to west, with the level of play as diverse as the geography. It’s impossible to see everyone, and you wouldn’t want to anyway.

    Canucks Euro scout Thomas Gradin compared Alex Edler’s Div 1 team to a ‘beer league team’, and that’s pretty accurate. You’ve got teams that play for little more than recreational purposes, teams loaded with ex-SEL players who don’t want to travel anymore, and then a bunch that act like unofficial junior affiliates for Allsvenskan teams.

    I love Division 1 in that it’s the purest form of scouting- one night, you may be watching players worse than you play with on the weekends. The next, you might see… well, the next Henrik Zetterberg.

  10. HBomb says:

    Kenta Nilsson’s favourite “one that got away” story, of course. It always stings when I think about that. Z’s numbers were poor, he was small, and Nilsson had been the only scout to see him … so Barry Fraser vetoed him off the list.Barry Fraser, if you’re somewhere at a Mexican internet cafe reading this, two letters:

    F U.

  11. Vic Ferrari says:

    Re Doug MacLean:

    Don’t forget the same with Selivanov and Carter, and that’s just ex-Oilers.

    I was listening to the trade deadline day stuff, with the TV video off for the most part. Some guy was making wildly erroneous statements about player’s games. Havlat in particular rings a bell, a guy was talking about how he only brings scoring and that CHI should probably consider moving him while he’s healthy.

    That’s just wildly wrong, by eye and by any numbers. Hell, without looking at numbers right now, I’d still be pretty sure that you could make a hell of a case for Havlat as a Selke candidate.

    I just assumed it was an eastern reporter commenting, someone who hadn’t seen many WC games, because the same guy was making provably wrong statements about Dumont (loved him) and Kim Johnsson (wayyy overpaid!).

    Turns out it was Doug MacLean. Damn!

    You know, I think he got famous for FLA’s run as a 3rd or 4th year franchise. But that wasn’t just luck, and it wasn’t just Vanbiesbrouck (sp?). That team had terrific outshooting numbers for an expansion team right from the get go. And were well in the black the year that MacLean took over and they went to the finals.

    Clarke, probably relying very heavily on Roger Neilson’s advice, built a hell of a team out of largely older guys who never had much in the way of counting stats, but did a lot of little things to help you win games.

    MacLean was just a right-place/right-time guy I think. Let’s all hope that he ends up as a GM in Minny or Colorado. In my dream scenario Flames ownership cans Sutter and brings in Doug on a lucrative, 10 year GMing contract. That would be sweet.

  12. Mr. Bugg says:

    Also, you’ll notice that Hakan’s drafting philosophy has changed slightly under the new CBA. He knows you can’t stash 17/18 year-olds anymore and wait for them to work their way up the letter. Instead, he’s grabbing 20/21 year-olds who have been proven too good for their respective levels and have been promised ice-time at the SEL or NCAA level. Examples of this are Samuelsson, U. Maine standout Gustav Nyquist and Dick Axelsson.

  13. Vic Ferrari says:

    Just on the general theme of using data to assess prospects and minor leaguers.

    There is a company that makes software called “Time on Ice” that sells software, really just a quick and nasty spreadsheet program, that lets you build you own shift charts, TOI tables, etc. If you had time to burn and a bit of money to spend you could do the same for your beer league team.

    Others surely sell this too, it wouldn’t take much to write something better than this program. But it serves the purpose, and I happen to know about it because I own a domain by the same name.

    Anyhow, they have sold this package to several leagues, but only four AHL teams: Manitoba, Cincinnati, Portland, Toronto.

    The thing those four teams have in common? All were under the charge of Brian Burke at some point. Odd that, I think of Burke as being more old school.

    In any case, I suspect that he knew a whole lot more about the context of Lupul’s and Smid’s ice time in the minors than Lowe did on that dark June day in 2006.

  14. Jonathan Willis says:

    Vic: Wow, just wow. Supposedly, MacLean’s in the discussion for the Minnesota job (that would be amazing) but I can’t see any ownership group being stupid enough to hire him again, despite his remarkable ability to self-promote.

    He pretty much singlehandedly drove attendance down in Columbus (a great market) and he’s spreading his stupidity at the national level now, thanks to Sportsnet.

    About the only thing he ever did that was worthwhile on the larger scale was letting Gare Joyce write that book.

  15. Jonathan Willis says:

    The thing those four teams have in common? All were under the charge of Brian Burke at some point. Odd that, I think of Burke as being more old school.

    I’ve spent some time digging into Burke’s career and if he doesn’t use the numbers than he’s damn good at evaluating with the eyes because I don’t think he gets it wrong very often.

    Witness his public statements about Alexei Ponikarovsky and Pavel Kubina; two guys everyone thought were finished when he took the job. He comes across as stupid from time to time because of his public persona (and I’m sure he isn’t an easy guy to get along with) but he’s done a lot of brilliant things behind the scenes, even before he got his first GM job in Hartford.

    I don’t know how much of the Canucks stuff was him (when he was director of Hockey Ops under Quinn) but I suspect it was a lot and it’s good to remember that he’s got serious ties to Lou Lamoriello too. I don’t like him, but he’s brilliant.

  16. Coach pb9617 says:

    … it might be wise to write that as 30 points is the midpoint of expectation, and there is an 80% chance of between 15 and 45 pointsIf you think that people’s eye glaze over now, start hinting at distribution in your posts. A widespread mob of angry narcoleptics will ensue.

  17. Hawerchuk says:

    Vic – the spread really is that huge. If all you know about a player is his G-A-P and maybe his age, then some huge percentage of your outcomes the next season are zeroes because the guy wasn’t good enough to make the jump.

    But even if you know that a guy’s going to get a regular shift, the middle 80% range is 4.5:1 in PPG. The middle 50% is 2:1.

  18. Vic Ferrari says:

    Thanks Gabe. At least it’s a line in the sand, though, Lowetide’s NHLE stuff beats the hell out of any other prospect forecasting on the internet.

    We just have to keep in mind that the line in the sand is really wide.

Leave a Reply

Want to join the discussion?
Feel free to contribute!

Leave a Reply

© Copyright - Lowetide.ca