• Neilson Numbers

    by  • March 27, 2012 • Hockey • 47 Comments

    “All I know is that if I had Neilson chances data for the #NHL, I’d love to make trades with GM who goes by any other +/-”
    -David Staples

    If you follow the Edmonton Oilers on the internet, you’ve probably heard of Neilson Numbers. Basically, it’s scoring chance +/-, but with only those chances on which a player is involved being taken into account. There’s a fellow at the Edmonton Journal, David Staples, who writes about them a couple of times of week. He’s made a number of claims as to their superiority to scoring chances or Corsi as a measure of a player, because they capture only plays on which a player made a contribution. I’ve never seen him do any real sort of examination of them though and discuss the circumstances in which they deviate from the results produced by the team chances.

    The gist of David’s argument, as I understand it, is that team based stats, like Corsi or scoring chances produce too many false positives/negatives. Guys are getting credited for things to which they did not make a contribution. I’m sort of uncomfortable with the idea that we can assign credit/blame like that because scoring chances happen within a certain context which is created by all of the players on the ice. If the forwards do a good job keeping the puck deep on a shift, there aren’t going to be any scoring chances again, which helps everyone’s numbers, even if the defencemen didn’t contribute to it. If the puck stays in the Oilers’ end for the entire game, defencemen and centres are going to be watching the numbers climb on their scoring chances against.

    Let’s look at the scoring chances and Corsi for the Oilers’ eight most used wingers this season. I think there’s a consensus developing that scoring chance ratio and Corsi tend to track one another nicely. Vic Ferrari’s made the point and I’m pretty sure I’ve seen other people make it as well. Looking at the Oilers’ wingers, that’s broadly true this season. Hall (1st in Corsi, 1st in SC ratio), Hemsky (2nd, 2nd), Eberle (3rd, 4th), MPS (4th, 3rd), Eager (7th, 5th), Smyth (6th, 6th), Jones (5th, 7th) and Petrell (8th, 8th). Looking at this chart, you’ll notice that Hall, Hemsky, Eberle, Smyth and Jones are all pretty tightly bunched in SC against – 16.2/60, +/- 0.5. Eager, MPS and Petrell are all far lower – as I’ve written before, there tend to be fewer events when guys who bat down the lineup are on the ice, presumably because coaches instruct them to play more conservatively and they play against other lesser players who have also been instructed to play more conservatively, ensuring that time is filled with little consequence before the guys who are allowed to play interesting hockey come back on the ice.

    Let’s look at David’s Neilson data. I’m particularly interested in the percentage of scoring chances on which they got tagged with a minus. Hall and Hemsky are reasonably close to one another, with Eberle some distance back, followed by another gap into which Smyth falls and then, way further back, Jones, who is tagged with a minus on only 23.9% of the scoring chances for which he has been on the ice this year. You can probably guess where I’m going with this.

    These are the chances for which each player was on the ice and didn’t receive any credit/blame. Where I’m suspicious of Neilson numbers is this. When Hemsky’s on the ice, the Oilers have allowed 15.6 CA/60. Jones is at 15.7 CA/60. Oddly, Hemsky’s making mistakes that get him tagged with responsibility for chances against at about 156% the rate that Jones is. Logically, assuming all other things are equal (which they aren’t, but bear with me), shouldn’t Hemsky’s chance allowed number when he’s on the ice be higher than Jones? Why do the remaining Oilers make fewer chance causing mistakes when they’re out with Hemsky than they do when they’re out with Jones?

    This doesn’t seem to make sense to me. Unfortunately, we can’t really look into why this might be with the data – it seems more than possible to me that guys like Hemsky/Hall handle the puck a lot more than Jones, which means more turnovers, which leads to more chances where they’re the ones getting tagged with errors. In effect, Jones is more peripheral to what happens on the ice than is Hemsky or Hall, and looks better because of it. Looking at David’s data, he has Jones with a slightly better Neilson differential than Hemsky or Hall. I’ve got serious doubts that that’s an accurate measure of their relative contributions (leaving strength of opponent/teammates aside).

    David is adamant that this individual stat is better than the team based stat because of the possibility for false positives/negatives. I’m not convinced. There’s undoubtedly the possibility of false positives/negatives with team based statistics. The advantage that they have though is that they capture the things that players do that lead to scoring chances that aren’t captured in the scoring chances themselves. Why do the Oilers get outCorsi’d and outchanced so badly when Jones is on the ice if he contributes to scoring chance differential on par with a Hemsky or Hall? Is it not possible that Hemsky and Hall are doing something that tilts the ice that Jones doesn’t do, something that David isn’t capturing? Anyone got a better explanation?

    Bill James had a line about how often a stat should surprise you, something like if it was surprising you more than 20% of the time, it was probably faulty. On the basis of this, I’ve got some doubts about applying Neilson numbers to wingers, specifically, and forwards more generally.

    About

    47 Responses to Neilson Numbers

    1. March 27, 2012 at

      I haven’t read all the details as to what David Staples considers a chance or not and how he distributes the +’s and -’s but lets take this scenario. Forward#1 is in the offensive zone, he loses a puck battle along the boards to Opponent#1 allowing the opposing team to carry the puck down the ice. During the ensuing play Defenseman#1 gets out of position and it results on a scoring chance against by Opponent#2.

      My question is, who gets the Nielson# – on this play. Clearly the Defenseman#1 should get one since he got caught out of the play, but isn’t part of the responsibility a result of Forward#1 losing the puck battle in the offensive zone? Had he been able to maintain the puck, the player wouldn’t have moved into the defensive zone putting Defenseman#1 into a position to be out of position.

      Similarly, probably Opponent#2 gets credited with a scoring chance, but Opponent#2 may have just been in the right position at the right time and got a scoring chance primarily because Defenseman#1 was out of position. The real positive play was done by Opponent#1 who won the initial puck battle that resulted in the gaining control of the puck.

      In theory I think there is probably a way we could correctly distribute +’s and -’s on every scoring chance, but I believe in practice it is not easily done and probably near impossible. Hockey is a complex game and every player has an influence on the play even if they aren’t directly involved.

      This is why I think on-ice +/- stats are likely better than Nielson #’s, but of course I prefer goal +/- (with appropriate consideration to sample size) to corsi +/- because I do think players can drive and sustain elevated shooting percentages (and possibly save percentages but that is much more difficult to show).

      • March 27, 2012 at

        David gives out more than one plus or minus per play if he thinks it’s warranted, so this kind of dilemma doesn’t really happen.

        • March 27, 2012 at

          The question is more about how far back does he go? The loss of the battle along the boards may be 10 seconds prior, but may also indirectly lead to the chance.

          • Cam Charron
            March 27, 2012 at

            From what I know, every chance for gets 3 +s, and every chance against gets 3 -s. Not sure if this is consistent with every chance.

            • dawgbone
              March 27, 2012 at

              That’s the part that I don’t understand, why settle on 3? There could be 6 breakdowns that lead to a goal against, why pick 3 as the number?

            • dawgbone
              March 27, 2012 at

              My mistake, he clarifies it below, always seemed like it was limited to 2 or 3 defensively.

          • March 27, 2012 at

            David goes quite a ways back when he counts things – I’ve seen him give positives to a guy like Jeff Petry after Petry’s changed off because it was his pass originally that started an offensive play.

    2. Cam Charron
      March 27, 2012 at

      Neilson numbers introduce subjectivity into what’s supposed to be an objective statistic. It’s like “errors” in baseball.

      • March 27, 2012 at

        Subjectivity is not an inherently bad thing if it can be applied consistently and without bias. There is a certain amount of subjectivity in shots, but for the most part they are reliable (compared to no subjectivity in goals, save for maybe the video goal review judge). The problem is for most things that is really difficult to accomplish, especially if multiple people are doing the subjective evaluation. See NHL’s real time stats.

    3. March 27, 2012 at

      I’ve graded these myself, just to see what they’re like, and I don’t have an issue with the subjectivity involved. I think Tyler’s right on the two points that really troubled me with the statistic, though – guys who handle the puck a lot are bound to be more prone to this sort of thing (just like guys who make lots of passes end up with lots of giveaways) and the overall context of the play is important too and gets lost sometimes.

    4. March 27, 2012 at

      This is an excellent critique, Tyler.

      I agree that David’s opinion on the team-based stats is problematic. He fancies himself a “hockey expert” and so figures that he (and others) can correctly judge which players are (most) at fault or deserve (the most) credit correctly in almost every case, and that he will capture the most important information on the creation and prevention of scoring chances by doing so. I’m not convinced that this is true, at least in part because of our personal biases, and in part because we don’t have an intimate knowledge of the responsibilities of the players in the Oilers’ system (or that of other teams). Things like Corsi and Fenwick and Scoring Chances deal with that concern well because they remove one subjective judgment from the equation (though subjectivity is still very much present).

      Still, I think we can agree that David is generally quite rigorous, and that he will assign credit or blame pretty consistently. When the measures agree, that’s powerful confirmation. When they disagree (as with Jones), I think it makes sense to hold conclusions more loosely and to ask more questions as you’ve done here.

      Unfortunately, David seems to consistently assume that the problem must be on the side of Corsi, Fenwick and scoring chances when it could easily be that David (or one of his graders) has a systematic bias against or in favor of Jones (or someone else), or that the grader doesn’t fully understand (for example) the responsibilities of the left winger in the defensive zone when the centerman is late getting back, or even (as you’ve suggested) that individual scoring chances are not capturing important aspects of scoring chance creation and prevention.

      • Tyler Dellow
        March 27, 2012 at

        Fuck. Shoulda just sent you the data and had you describe my conclusion. ;)

        This is not intended as a shivving of David. I strongly agree with your third paragraph.

      • March 27, 2012 at

        With one exception – while David was in Boston for #SSAC, I think he’s done all the grading himself.

        • March 27, 2012 at

          He’s expressed on several occasions that there would be value in doing this league-wide. Expanding the project like that would require a team of graders, much like the scoring chance project expanding to other teams required people other than Dennis to be involved.

    5. Dennis
      March 27, 2012 at

      ///Unfortunately, we can’t really look into why this might be with the data – it seems more than possible to me that guys like Hemsky/Hall handle the puck a lot more than Jones, which means more turnovers, which leads to more chances where they’re the ones getting tagged with errors. In effect, Jones is more peripheral to what happens on the ice than is Hemsky or Hall, and looks better because of it.///

      This is an excellent point by Ty and one I can’t get past when wondering if what David does is a natural extension of what Vic has helped me to do.

      I haven’t been nearly as vociferous in promoting my efforts as has David – and probably to my own detriment – but I honestly believe it’s better to have an idea of what constitutes a scoring chance and then just log the times and let the plus and minuses fall where they may.

      Also, I always thought Roger Neilsen was a cool and innovative cat but some of the things I’ve recently read about his methods strike me as wrong. Every goal as a scoring chance? No. Hitting the post negating a scoring chance? God no. Missing the net from a prime location isn’t a scoring chance? Jesus no.

      At least from the fan’s perspective no one has logged more games than me, ie this is my fourth full season. I think I have a decent eye for what qualifies as a chance and I think the work bears out. I believe breaking it down into plus and minuses on each chance Does get too close to the old errors philosophy that David originally latched onto.

    6. David Staples
      March 27, 2012 at

      Interesting critique, and I don’t take it as a shiv job, Tyler. I’ve made some provocative comments about the superiority of Neilson plus/minus so it makes sense for those who give more weight to other plus/minus systems would respond.

      A lot of the people making comments here raised points, so instead of one mammoth comment, I’ll answer them in turn.

      First off, if you’re not sure how I grade scoring chances with this system, here’s the F.A.Q:

      http://blogs.edmontonjournal.com/2011/12/04/frequently-asked-questions-on-neilson-numbers/

    7. David Staples
      March 27, 2012 at

      David Johnson: On each scoring chance for and against, I track back to the start of the chance, then so long as the momentum on the play keeps up (even if there’s a brief change of possession) I continue to count up plus marks on chances for, minus marks on chances against.

      On most chances for, about three or four players get plus marks, though sometimes it can be six or seven, including guys on the bench who got the positive sequence going in their own end.

      On chances against, usually two guys get a minus mark, but again that number can be much higher, and include guys on bench (who might have made a crap line-change, for instance).

      • March 27, 2012 at

        So are all chances weighed evenly or are some chances worth more +’s than others. By that I mean, if one chance has 6 guys contributing and another chance has 2, do the 6 guys get 1/6 of a point and the 2 guys each get 1/2 a point? Otherwise some chances will get more points than others. If not, is it right that 6 good plays by 6 players players earn 6 +’s while 2 outstanding plays by 2 players earns just 2 +’s.

        Another question, does every chance need a + or a -? Sometimes it could be a dumb turnover that causes a chance and the offensive player really did nothing to generate that chance other than being lucky enough to be in the right place to receive the turnover pass from the opponent.

        I have thought about how I would do a system like this and I just think it gets really complicated and I wouldn’t know when or how I should credit smart plays away from the puck.

        • dawgbone
          March 28, 2012 at

          Petrell and Jones both had one of those this year.

          They were literally standing in the neutral zone and the opposition carried the puck right into them and they got shorthanded breakaways off them.

          I still think you give it to the player, I mean they were in proper position, doing what they were supposed to do.

      • Chris Abraham
        July 14, 2014 at

        What if you were to somehow calculate and factor in the amount of time they had the puck on their stick? Wouldnt that get rid of the dilemma of players who carry the puck more having a worse Nielson?

    8. Suntan Oil
      March 27, 2012 at

      While I like the Neilson approach and really appreciate David’s diligence in tracking chances (especially when I can’t catch the game myself – makes for a much better mental picture of how it came down), one downside is that the method is weighted against the creative.

      A high percentage of good SCs/goals in the NHL result from teamwork between three (often rotating) roles – the Passer, the Sniper & the Meat. David (bless him) gives credit to the screeners & forecheckers of this world – the Smyttys, the Joneses & Harskis, increasingly the Horcs – for SCFs.

      Unfortunately, when it all goes sideways only the puck handler/passer gets nailed with an SCA. Hard to pin an error on the guy battling in the crease or the one ready for the one-timer. As point guards in basketball have the most turnovers and quarterbacks in football (surprisingly) throw the most interceptions, the Hemskys/RNHs/Omarks dancing off the sidewall and trying to make something happen will end up with higher SCAs (while all get credit for SCFs).

      Further, the stat punishes those guys for making creative choices. An attempted pass to Smid creeping down for one of his guaranteed backdoor markers that gets intercepted and fed up by a D results in an SCA for the passer. A low-percentage shot from the same place that rebounds to the same D with the same result does not.

      I find it difficult to get completely behind a stat that punishes entertaining hockey.

    9. David Staples
      March 27, 2012 at

      Dennis: Neilson counted hit posts inside the scoring chance zone, but not hit posts from outside of scoring chance zone.

      Yes, he counted all goals as chances. I do, too.

      Missed shots? That’s more complicated.

      Is a rushed shot from the kill zone that fails to hit the net a scoring chance? Is a deflection that fails to hit the net a chance. I don’t think they are. Others would disagree.

      From talking to Neilson’s long-time assistant Ron Smith I now have a much better idea of exactly what Neilson did and didn’t do in making his decisions, and will soon be posting soon on this.

      There was much debate between Neilson and Smith (just as we now debae these same issues) and some changing of thinking over time over what was and wasn’t a chance.

      In the end, they decided that most missed shots aren’t chances, but some are. If a player is wide open in front of the net with time to shoot, but he misses trying to pick the corner, that is a chance. If he’s rushed, or there are players to shoot around, and he gets off a shot from the kill zone that misses the net, that’s not a chance.

      • March 27, 2012 at

        I cannot possibly fathom how anyone could defend the idea of “every goal = scoring chance”. I’d love to hear some justification of that. Taking it to its natural conclusion, Marty Brodeur dropping his stick and accidentally deflecting a puck into his own net on a long clear in the 2003 Stanley Cup Final is a “scoring chance”. I have a hard time taking anything seriously that claims scoring chances can happen on the far side of the red line, or would assign blame to a skater for such a goal. Obviously, not everything is that extreme, but there are a not-insignificant number of bad goals that happen in the NHL.

        I much prefer something closer to JJ From Kansas’ CCSI project over at Winging It In Motown. It’s fairly similar, in that it’s creating an alternative plus/minus system based off of game charting. In that project, you see things like bonus pluses or even assists given to players for things like screening the goalie on a goal scored, minuses erased when a player did his job and just happened to be on the ice when someone else screwed up (or a goalie made a boo-boo), etc. There are issues of subjectivity involved (which I think primarily shows up in goaltender analysis in that project), but it is at least consistent. Football Outsiders’ (not Tyler’s kind of football) Game Charting Project is basically a version of this on steroids, and it’s done some pretty impressive things.

    10. David Staples
      March 27, 2012 at

      Tyler (and Suntan Oil): Yes, a winger who handles the puck a lot more, such as Hemsky or Hall, might well get more errors. It’s interesting, though, that a dman like Gilbert, who also handled the puck a ton and more than anyone esle, made very few defensive errors this year, about the same rate as Smid. But other big puckhandlers such as Petry and Whitney made a ton of mistakes on chances against. So this critique likely has some validity but it only goes so far.

      https://docs.google.com/spreadsheet/pub?key=0AhMC4T3agcfmdGI4dUFSWTZ3QjlLZ2RCc0tlVEJpU1E&output=html

      I’d say while Hemsky and Hall might get hit a bit with more “mistakes on chances against” because they handle it so much that a close observer of the team might also argue that Hemsky and Hall are weak defensive players, prone not just to giveaways, but to soft and lazy backchecking now and then. I wouldn’t really trust either with a minute to go in their own end.

      My major point is this, that however you want to look at these various plus/minus numbers, in the end, the proof of all these stats is how they rank the players on their two-way play.

      Right now, if you go by Corsi on-ice plus/minus (I don’t have Dennis’ results on scoring chances) vs Neilson on-ice rankings, it goes like this for the Oilers at even strength:

      Defence (Neilson): Schultz, Gilbert, Smid, Petry, Sutton, Potter, Whitney, Barker, Peckham, Teubert, Plante, Chorney.

      Defence (Corsi): Sutton, Petry, Potter, Teubert, Smid, Plante, Peckham, Whitney, Barker, Schultz, Chorney. (Not sure where Gilbert would have fit in there)

      Comment: To my eyes, the Corsi plus/minus looks almost random here. Sutton? Potter? Teubert? Maybe I’m not accounting for what all these players do on the ice somehow, but I think Neilson nailed the order in superior fashion.

      Wing (Neilson): Eberle, Jones, Hall, Hemsky, Smyth, Paajarvi, Eager, Omark, Hartikainen, Omark, Petrell, Hordichuk.

      Wing (Corsi): Hall, Hemsky, Paajarvi, Eberle, Jones, Smyth, Omark, Hartikainen, Hordichuk, Eager, Petrell.

      Comment: I’m glad to see Eberle at top of Neilson rankings, not so keen about the high ranking of the controversial Jones. But that’s how it broke down when I broke it down.
      In fairness, hardly anything separates Jones/Hall/Hemsky/Smyth when it comes to Neilson chances plus/minus, and I’d say that’s a fair guide of their two-way play.

      https://docs.google.com/spreadsheet/pub?key=0AhMC4T3agcfmdEFQb0hLRDZSeTYyd3YxcVJXSDIwYmc&output=html

      As for Corsi plus/minus of the wingers, Paajarvi ranked ahead of Eberle and Smyth? Hmmmm…. Perhaps Paajarvi does something to impact the game that isn’t so apparent and makes him superior to Eberle and Smyth in two-way play.

      Centre (Neilson): Gagner, RNH, Lander, Horcoff, Belanger.

      Centre (Corsi): Gagner, RNH, Horcoff, Belanger, Lander.

      Comment: Both systems have it right with Gagner, I’d suggest. RNH has also done sorta, maybe OK at evs and he probably deserves second. In Neilsons, Lander gets a bump over Horcoff and Belanger. I’d say Neilsons has it right on Belanger. As for Lander being rated over Horcoff in Neilsons, that’s how the scoring chances broke down, but that’s not how I’d rate them. With Corsi I’d suggest it’s closer, but Lander did better than Belanger this year.

      Of course, with the raw data from both systems you have to factor in other things, most crucially QualComp.

      • Tyler Dellow
        March 27, 2012 at

        This is all fine and well David and I agree with a lot of it but I don’t know that it addresses my main question: why are there so many chances for which Ryan Jones isn’t at fault when he’s on the ice relative to other players? Do they get worse because of his stupid flow?

    11. David Staples
      March 27, 2012 at

      Finally, Scott:

      Yes, I do now fashion myself as something of a hockey expert, at least expert enough to do this work …

      I am, after all, a product of the Alberta landscape …

      “It is a country to breed mystical people, egocentric people, perhaps poetic people, but not humble ones.” Wallace Stegner

      • March 27, 2012 at

        I’m not sure you realize this, David, but I believe it’s a rule that all Westerners are humbled and kept humble by their landscape. It’s why we turn out so many enforcers.

      • March 27, 2012 at

        That little comment wasn’t really my main point, but I will say that it seems to me that that’s the kind of attitude that will foster intransigence.

    12. Dennis
      March 27, 2012 at

      Staples: I believe some of that thinking is wrong and I think in some of your more provocative defense of your work it has belittled mine.

    13. David Staples
      March 27, 2012 at

      Oh, I know it, Jon. The cold makes us what we are. Stegner was actually from the southern prairie, not my area.

      Dennis, I’d suggest all of us have felt belittled during the entirety of this stats debate from time to time. I’ve taken my share of kicks. That said, if I’ve been overly harsh (and I’m sure I have from time to time), my apologies.

      With the Oilers, we’re the only two counting chances, so a bit of rivalry can be expected, but so can some mutual respect, and I have that for you and your hard work. Your eye for the game, and for hockey talent, is as sharp as the best of them (Bruce McCurdy), and I do think I’ve mentioned that in the past as well.

    14. David Staples
      March 27, 2012 at

      But I don’t see any of us as being too humble, Jon, not even the enforcers. Grounded, maybe, but not humble. Many of us have no trouble telling Tamby and Lowe exactly how they’re screwing it up …

      • dave
        March 28, 2012 at

        It was a reference to an earlier Tyler piece

      • March 28, 2012 at

        Sorry David, I was referencing Chris Jones’ article on why the West is an enforcer factory. He quoted Stu Grimson talking about the landscape out West making one humble, and that sort of insanely weak cause/effect relationship annoyed me at the time and leads to the occasional snide remark about the premise.

    15. dave
      March 28, 2012 at

      Jones does seem like an error on this project -though not sure there are many others – which may relate to the 20% surprise rate referenced elsewhere

      But maybe if you were a goal hanger (not saying Jones is – hes been a lot better over last year) your SCA are always going to get uncounted compared to someone that did come back ?

    16. David Staples
      March 28, 2012 at

      Tyler, in response to your question: “Why are there so many chances for which Ryan Jones isn’t at fault when he’s on the ice relative to other players?”

      According to Dennis’ numbers, all the top line wingers are out for a similar number of chances against, save for Hall, who is out for a bit more, not much. Yet when it comes to their mistakes on chances, I’ve found in breaking down scoring chances that Jones (and Smyth) makes less mistakes than Hall and Hemsky.

      I’d say a few things:

      1. That Jones and Smyth do, in fact, make fewer mistakes on scoring chances against.

      2. That they handle the puck a bit less than Hall and Hemsky, so aren’t in position to make quite so many mistakes.

      3. That their lower mistake count is also impacted by their unique roles on the team: being the guys most likely to screen the goalie, they are less likely to have specific responsibility on the backcheck.

      Of course, Jones and Smyth have tougher ZoneStarts, so this would make them a bit more likely to be in position to make mistakes on scoring chances, as most (but not all) such mistakes happen in the defensive zone.

      4. Hall and Hemsky are mediocre (at best) defensive players, sometimes lazy on the backcheck, sometimes making the wrong reads, and more likely to make defensive mistakes than either Jones or Smyth. It’s worth noting that Jones and Smyth (along with another lower mistake winger in Petrell) are the coaching staffs choice for the wingers to kill penalties.

      NOTE: in any final evaluation of Jones, I’d put almost equal weight on his Neilson goals/plus minus (where he ranks behind Smyth and Eberle), and I’d bump up Hemsky a full ranking in the winger rankings based on his tougher Quality of Competition.

      • Tyler Dellow
        March 28, 2012 at

        3. That their lower mistake count is also impacted by their unique roles on the team: being the guys most likely to screen the goalie, they are less likely to have specific responsibility on the backcheck.

        If this is true then what you’re tracking isn’t so much a reflection on their defensive abilities, as it is a reflection of the role that they play. How can making fewer errors, because you’re up the ice, outweigh the fact that your team gives up piles and piles of chances?

      • dawgbone
        March 28, 2012 at

        I think #3 is flat out wrong.

        You don’t lose defensive responsibilities because you are screening the goalie, anymore than you lose them by being the first forechecker.

        If I’m too busy trying to screen the goalie and don’t offer proper puck support and they get the puck out down the ice, I have a responsibility in that.

        One of the things that makes guys like Holmstrom and Smyth good effective players is that they know when they should be in front of the net and when they shouldn’t be. You can’t plop a guy in front and do a 2 man cycle in the corner.

    17. Darren
      March 28, 2012 at

      I will always struggle with a scoring chance metric that enables the ‘safe play’ in that a player will pass sideways rather than attempt the pass out of the zone to a streaking forward.
      It’s easier to look good when you don’t look to create and consequently shift the burden to the teamates you are playing with.

    18. Adam Dyck
      March 28, 2012 at

      Could it not be that the reason Jones is a part of fewer mistakes than Hemsky, and yet is on the ice for more than Hemsky, because Hemsky typically has stronger linemates?

      (Note: I am playing the Devil’s Advocate here, and actually put more stock into Corsi and its variants/Dennis’ work than Neilson Numbers)

    19. March 28, 2012 at

      Adam: I don’t rail as much as I could just because I think my work stands. David has more of a platform to pimp his work so that’s why he gets his MO out in front of mine; and that’s his right.

      Still, I think he’s looking for something that either isn’t to be found or doesn’t really need to be discovered

    20. David Staples
      March 29, 2012 at

      @Dawgbone. Scoring chances develop quickly very often, a mad dash up the ice. So if a guy is doing his job and screening, the puck squirts loose, then the high man fails to backcheck, yes the guy who is typically high man has more defensive responsibility.

      As for me having more of a forum, to “pimp” my work, whatever …

      • dawgbone
        March 29, 2012 at

        Granted, but if the puck is in position to be a mad dash up the ice the other way, you probably shouldn’t be planted in front of the net, you should be in a puck support position.

        Being the guy to stand in front of the net goes well beyond just standing in front of the goaltender. There was a play last night where Smyth and Jones were both on the ice on the PP. Jones started in front of the net and never moved. The puck started at the right point, went down to the left corner, up to the left point, across to the top middle of the zone and on net.

        The Sequence by Ryan Jones, stand in front of the goalie on the right half of the ice, stand in front of the goalie on the right side of the ice, and get boxed out from screening the goalie. Smyth on the other hand won 2 puck battles in opposite sides of the rink to keep the play alive and almost made it to the front of the net before Jones did.

        Scoring chances don’t just happen, there’s often breakdowns that lead to open space that cause further breakdowns. Being below a puck battle in the offensive zone is generally a poor position to be in.

      • Preeti
        May 31, 2013 at

        The fans who have done this really enjoy gttnieg inside the glass and meeting, seeing and hearing the players while they go about their pre-game routines.

    21. Captain Obvious
      March 29, 2012 at

      The issue with the Jones/Hemsky/Hall comparison is that there doesn’t appear to be an absolute scale against which these errors are measured.

      If we grant everything that David says is true there is still a problem so long as there are a different number of errors assigned per chance. Which is to say the since hockey is a zero-sum game, assignment of individual scoring chances needs to be bound by the absolute number of scoring chances while on the ice.

      For instance, if Jones is on the ice for the same number of scoring chances against as Hemsky while he is given an error on only 15% of those as compared to 37% by Hemsky then as a matter of logical necessity whomever is playing with Jones has to pick up the other 22%. Otherwise the consequence of the system is that those players that appear to be making more errors will be assigned blame for errors that, by definition, do not end in goals.

      This means there are two choices. You could assign an equal number of errors on every play, which while rationally coherent does not reflect empirical reality. Or you could weigh individual errors by the total number of errors on the play.

      The latter is the right way to do this because it correctly weighs the value of an individual error against the absolute cost of that errors in goals.

    22. Coach Pb
      March 31, 2012 at

      Great takedown, Tyler.

    23. Pingback: The Blues and The Abstract Truth: Corsi, Facts vs. Norms - The Oilers Rig - The Oilers Rig

    Leave a Reply

    Your email address will not be published. Required fields are marked *