Staples: That Was Easy

This is Staples. I love going there, buy all kinds of stuff. Binders, computer chairs, pens, paper, gum, there’s just an appeal for me about buying new stuff and bringing it home to a (hopefully) more organized life.

But that’s not what we’re talking about right now.

Read more here.

written by

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

17 Responses to "Staples: That Was Easy"

  1. Scott says:

    I think the most helpful thing that can be done with the linguistic aspect of statistics is really to shorten the names. The problem is that most of the available names that would be convenient are already taken, so instead of goals, one says goals/60. It might be beneficial to simply refer to this as “tallying” and identify the per 60 by using two decimals afterward, so for example, 1.12 goals. Similar to “slugging” in baseball with percentage implied when one sees the number. A simple innovation, but it does in some small way simplify things for people in terms of language if not clarity on first read.

    There is so much context to provide in various situations that I might limit my list of ten to even strength as follows:

    Tallying (Goals/60)
    First Assists
    Scoring (Points/60)
    The Good (+/60)
    The Bad (-/60)
    The Ugly (+-/60)
    Ugly Corsi/Fenwick (+-/60)
    Resistance (QualComp)

    The renaming of things perhaps complicates things where none is needed. Still giving something a name that is a name and not just a description seems to me a wise endeavour. Some of the above are silly, but it’s a place to start.

    Also, thank you for choosing to put the full post on this site. It is much appreciated. Still, congratulations on your opportunity to work for the Nation of Domination.

  2. Lowetide says:

    scott: Terrific stuff. What I’ve decided to do is this:

    1. Post once a day for Oilers Nation. I post a lot and it would be ridiculous to send all of it when Jonathan’s stuff has equal value and Brownlee is contributing things that are unavailable to me. So I think that’s the best solution (although I haven’t talked ot ON about it).

    2. I will “cut off” the post here and send readers to ON for the post’s life but then re-place it here after a couple of days. I’m very interested in keeping the archive alive so I can call all you guys and girls names when I’m right about something. Don’t laugh, it’ll happen.

    Anyway, those are my thoughts on the blog, such as it is.

  3. HBomb says:

    Exceptional work LT. I do not know how I’d order the most important statistics, but I do know this: E+/- and the on-ice/off-ice numbers would be near the top of the list.

    Two things need to happen for this “revolution” in thinking to take place:

    1) They’ve got to “clean up” the plus minus stat and express it only for even-strength 5-on-5 play. 4-on-4 and ALL extra man situations (including empty-net goals in either direction) need to be filtered out and shown somewhere.

    2) Finding a way to express strength of both opposition and teammates clearly. The work Desjardins has done is nothing short of exceptional, but as has been discussed, the method needs improvement. I still find that watching the games and looking back at the shift charts qualitatively to establish trends works best. Creating numbers to show this over a single game are tough – over an entire season where every game is something different? This is a Mount Everest that Desjardins is attempting to climb here. I honestly don’t have any idea how you’d do it.

    But then again, I’m an engineer – anything beyond the most basic of statistics is not my kind of math. What I see on the ice is most important and always will be. Hockey’s a dynamic sport – baseball, it’s a lot easier to isolate events, and everything starts the same way every play, with a guy throwing a pitch.

  4. Schitzo says:

    Re: Desjardins QC numbers. If i understand the stat correctly, the biggest problems is that it’s still driven by on/off-ice +/-.

    This tends to punish players who are put into a power v. power role – in our case, Horc is doing a good job if he breaks even against the Thortons and Iginlas of the world, letting the kid line make hay. It sure makes his on-ice versus off-ice +/- suffer, though.

    If his on-ice +/- is zero (break-even) and we subtract what we hope is a positive off-ice (since the kid line is outscoring), he ends up with a negative number and is apparently “easy” to play against. Sure he is.

    I’ve never looked into it myself, but its my understanding that the chess world uses an extra degree of recursion in their rankings to account for the fact that a match of two extremely high end players tends to “look” the same as a match between two equally crappy players.
    I’d assume something similar is necessary here, but at that point the league would probably have to get involved and actually pay someone to do it.

  5. spOILer says:

    Mmm, Racquel. One of the tastiest grapes on the vine. (…drifts off into visions of her Mesozoic region).

    Re: Staples

    I’ve given him a pretty hard time, not often, but here and there, I’d admit. I gave him a poke just yesterday. But he does some nutty stuff. The error stat (which I give full credit for originality), and giving Gagner 28 goals as his 08-09 expectation are just two examples.

    It’s evident he loves hockey. But I just don’t see the same craft and thoughtfulness he puts into his regular articles in his blog writing. Probably a time issue as much as anything.

    Sometimes he comes across the same as that new guy to a forum who spams the board with posts that have already been discussed. Maybe he should’ve spent a year reading and commenting on the existing blogs (and thus perhaps establishing a stance or pov, not to mention street cred) before charging into the fray guns ablazing (I’m over-dramatizing, I know).

    His “highlighting” of other bloggers feels a bit ingratiating and condescending at the same time (condescending from the perspective that he can drive readership to other bloggers, which is likely true given the site he posts on, but does this need any emphasis?) Wouldn’t a blog roll suffice as endorsement?

    I expect that as Mr. Staples continues to read blogs and blog himself, his work will improve. He does suffer from our modern society’s ridiculous bias against experts exercising their opinions outside of their expertise, but I’ve always tried to take his posts on their own merits, with no consideration of source.

    Re: The Math of Hockey

    I’m not so sure the poetry is even required as much as time, familiarity, and context.

    Does something need to be done about the team bias in some of the stats? Maybe some kind of relative index, like in the Qual stats; for example, whoever takes the toughest opp on the team approaches a max of +10, the weakest a -10, and everyone else is in between. I think we’re all doing something like this mentally anyway, especially when comparing across teams (“he took the 3rd hardest mins”).

    Look at Blake’s versus McLaren’s numbers, McLaren’s are far better, yet he seems to be the odd man out. How do we account for the LAK bias to Blake’s numbers? Desjardins I believe is using relative #s, but it doesn’t seem to fully explain the story.

    Due to the more variable nature of the game, we need a top 12 not a top 10.

    My Top 12:

    1. Individual shot rate vs. Team rate
    2. Ind. shots allowed rate vs. Team rate
    3. Zone %age vs. Zone starting point %age
    4. GFON/60
    5. GAON/60
    6. Qualcomp
    7. Qualteam
    8. %age of shots allowed fr the trapezoid (quality of shots allowed)
    9. Pts/60
    10. SV %age
    11. # of giveaway rebounds
    12. Play died with #

    Honourable mention to TOI, takeaways/60, team hits per shift, and shots attempted vs. actual shots (the young Hemsky stat).

    Yeah, you’re right, some poetry is needed. Scott did a really nice job.

    Sorry for the long ramble. Can’t believe how long this took to put together.

  6. Scott says:


    Regarding the NHLE that Desjardins does. You seem to think that it works very well and frequently mention that the ice time ought to be similar in both situations. I remember in a previous post you mentioned you included a discussion of Kane, Gagner and Perron. Given that Milan Lucic came in from junior too, I’ll throw him into the mix.

    Kane went from a prediction of 26-36-62 to results of 21-51-72 in what I would think is a very similar role for the Blackhawks. The prediction understated his production by 16%.

    Gagner went from a prediction of 16-39-55 to results of 13-36-49. However, I’d imagine that in junior Gagner received probably at least a minute of extra PP time per game as well as more ice time overall. Realistically he would have been depended on a lot more in his time in junior than he was in the NHL.

    David Perron had a prediction of 13-14-27 and his results were bang on, but only over 62 games. His results would have been 17-19-36 over 82 games and I would need to think that he played a much reduced role with the Blues than he did in junior (he had about two minutes on the PP per game with the Blues). The prediction understated his prorated results by 33%.

    Milan Lucic had a prediction of 9-11-20 and had results of 8-19-27 in what I would think was a reduced role with the Bruins (only about one minute of PP time per game with the Bruins). The prediction understated his results by 35% even allowing for only a 77 game season.

    I guess I’m saying that at least when it comes to junior the predictions seems too low and you may want to consider bumping the number up from what Desjardins has. Even in reduced roles players are beating the number which is just no good. I would hope to see players consistently underperforming by at least ten to twenty percent.

    Anyway, any thoughts on this?

  7. Lowetide says:

    Scott: Good points on Desjardins. For the purposes of this blog and what I try to look at, Desjardins is well within range. I understand the desire to make it more perfect but there is also an ebb and flow to these things.

    Perhaps a “last 5 seasons” bias, in that you’d give special weight to the kids coming out of junior recently?

  8. spOILer says:

    I should say in the above, that I’m guessing at what James’ goal was.

    I took it from the lines of individual career numbers predicting future value, contribution to team success, value in signing to another contract, that sort of thing.

    There’s so many ways to look at this ordering. If it’s about individual success over the next play, shift or game, I’d flip the shooting and scoring rates because it makes sense, to me anyway, that shooting percentage is a better indicator in the short term.

    Or if we just wanted to know the story on a guy, there’d be a different set of numbers, no?

  9. Lowetide says:

    spOILer: Your list looks good to me. I honestly don’t have a clue. I look at EV/60 points and toughness of minutes compared to teammates. Then I look at Corsi, then I like to look at PP numbers, etc. GF ON/GA ON is also a beauty.

    But I’d like to see lots of lists. It’s interesting to see point of entry for people.

  10. Vic Ferrari says:

    I spent several hours looking at baseball stuff last week, and it’s bonkers. I have a whole new respect for Bill James.

    I suspect that he’s thrown ‘runs scored’ down at the bottom of the list for effect btw. But beyond that he’s not far off on the repeatability of the numbers I would think.

    It’s all just a huge mess with MLB though. These attempts to make a picture from a billion dots, with stats that are concoctions of a bunch of things that are related to each other and couldn’t possible be divorced.

    Everything is connected to everything, at least in part because of the lack of parity in the game. Teams that can afford good players get both patience and power. The stats are laid out in nice columns, easily accessible. And not much happens in baseball, frankly, so when you’re at a game there are really only a few options for things to do between pitches (which is the vast majority of the game); ponder stats, drink, nap.

    I think that this is what really separates cricket from baseball. If you go to a cricket match and DON’T drink in the morning and nap in the afternoon, your friends will wonder why the hell you even came.

    The other thing about baseball that I’ve never liked is that the strong get stronger, sort of the opposite of hockey that way, just as part op it’s natural existence.

    Rolling on with the “Lowetide 8″ thing, a middle order hitter for Baltimore in 2007 would have faced middle relievers a lot less (over a hundred innings as a team), and good closers and starters a lot more, than a Yankee or a Red Sox player. Probably hit with the infield drawn in a lot less too, though I don’t know.

    So inevitably the rich get richer. If the Yankees trade Daniel Sedin to Baltimore for Henrik Sedin that year, they’ll get some pitching back in the deal too. Because the chances of Henrik putting up better numbers than Daniel over previous seasons must be thousands-to-one.

    I mean the shootout in hockey is like hundreds of things in baseball. You’ve got all day to think about it, patterns appear and disappear, but some stuff clearly matters. And even though it seems like there should be, there just isn’t much sustain in it.

    Baseball is like chess, hockey is like tic-tac-toe played at an hellish speed.

    Back to hockey: I can see that you’re trying to spread the word. Though I don’t understand why.

    And Staples is a good choice as a test subject, you’re looking straight into the eye of the chicken there. I suppose there probably is a way to twist simple truths into fascinating fact nuggets that would fascinate David and others like him. But I wouldn’t bother.

    Each to their own though.

  11. Ned Braden says:

    While the work of Desjardins and others is impressive, at the heart of it Hockey is not Math.

    Baseball is Math. Hockey, like Basketball is a game of constant motion, where as Baseball in played in a series of states and arithmetic is its outcomes. Ball Players can be parsed into their value offensively and defensively. Bill James, Pete Palmer and others showed the true hidden values and rolled it into Runs Created, Win Shares and numerous other valuable tools, Billy Beane had a leg up until everyone else caught on, and teams like the Sox hired James.

    Hockey isn’t Math, its more like Language Arts. Because players play offense and defense at the same time and they play alongside other skaters and their goalie. Their value is interdependant, it aint 1 + 1 = 2.

    That said Desjardins and Corsi’s like James 20 years ago is providing a new perspective on how players are valued. Just like Klein and Reif did when they pealed the first layer off hockey numbers more than a decade ago.

    Anyways, I think you have to be able to summarize into one number for easy measure, the hockey equivalent of Win Shares or Runs Created, which is not quite there yet.

    And while I’m a huge fan of Lowetide, Desjardins and Mirtle, I’m not a fan of Staples.

  12. Ribs says:

    Damn Staples (the store). I don’t think I’ve ever went in there and not spent over $20 worth of stuff I probably don’t really need.

    I just want to applaud all of you “Stats Guys” out there for your efforts. Before reading this blog I never even imagined there were so many tools out there that can be used for projections and player ratings.
    Desjardins is mind-blowing. Even the purest traditionalist out there can not deny its effectiveness. I also find the quality of opposition stats quite fascinating these days.

    I don’t have the time or patience to really delve into these sorts of things but it’s good to know that they exist to help reinforce opinions and provide some solid reference when evaluating players and various situations.

  13. breakerdog says:

    My jumping off point for using stats to evaluate a player is

    1)goals for at EV per 60min vs. goals against at EV per 60min.

    Basically, I want to know if he is an outscorer

    2)Qualcom /Qualteam.

    What situation does he get used in and how will that affect his outscoring ability

    3) Team save %

    How has goal tending affected his outscoring ability

    4) Shooting % vs. historical shooting %

    Is he having a “Ribeiro” year??

    5)PP points per 60min. PK points vs. per 60min. minutes played on special teams

    What kind of contributer is he to special teams

  14. oilswell says:

    FWIW I see it a somewhat shallow and rambling analysis by Staples, understanding the main question to be the methods one uses to know hockey. If trust in intuition is childhood, now seems like the awkward part of early puberty in which the means of statistics are treated as ends. I’m waiting for the models.

    And NedB, if hockey is beyond the grasp of math/logic, I would think it is beyond the grasp of comprehension. You may be misunderstanding what math is?

    As for LT’s question, I don’t know how close we can get without properly accounting for the fact that it is a “luck”-soaked team game played over 60-70 minutes of different sub-games stuffed into 3-5 frames over unequal seasons of 82 matches. Until I know inter-player factors and the impact of the several layers of strategy, I know of only individualized aggregated measures, and I’m very interested in total quality of chances scaled by opposition and linemate quality. VF’s work on posession/shot proxies seems to be getting pretty fucking close. Well, so I hope.

  15. Lowetide says:

    Ah, oilswell. I love your posts. First two paragraphs are genuinely funny and you finish up with something that moves the conversation forward.

    I agree that Vic’s stuff is all around it. His last few posts over there have been especially interesting.

  16. sexy says:




    色情A片,A片下載,色情遊戲,色情影片,色情聊天室,情色電影,免費視訊,免費視訊聊天,免費視訊聊天室,一葉情貼圖片區,情色視訊,免費成人影片,視訊交友,視訊聊天,言情小說,愛情小說,AV片,A漫,av dvd,情色論壇,視訊美女,AV成人網,情色文學,成人交友,成人電影,成人貼圖,成人小說,成人文章,成人圖片區,成人遊戲,愛情公寓,情色貼圖,成人論壇,色情


  17. boogleboy says:







Leave a Reply

Want to join the discussion?
Feel free to contribute!

Leave a Reply

© Copyright -