• The Grinch

    by Tyler Dellow • December 25, 2012 • Uncategorized • 9 Comments

    Twitter’s @kinger999 asked me about the possibility that Jordan Eberle has a high shooting percentage because he doesn’t take shots from far away in the same way that other players do. Eberle himself has given this theory an endorsement, telling Jason Gregor:

    I think for one, I’ve really worked on my shot. I try not to waste a chance; you know if I’m in the slot I’m going to bury it. Like you said, I’m not a guy to come down the wing and fire a puck on net. I more so want be the guy that beats the defensive line, tries to make a play to a guy or turn back and try to create more out of it. The chances I usually get, I usually try to slip in to places where it is a lot easier to score. Something about my shot too, I think goalies usually have trouble with it. I think I shoot off the toe a little bit. I ask goalies, ‘What’s the difference between my shot and someone else’s?’ It just comes off differently. So, I take a lot of pride on working on it. If I get a chance, I want to score. It pisses me off when I don’t.

    There’s a fine site, Super Shot Search, that creates maps of the shots that players take, based on data gathered by the NHL. As you can see if you look up a defenceman on the site, it’s not perfect. It looks to me like either the maker of the site or (more likely, given the NHL’s involvement), the creator of the data sometimes “flips sides” and fails to employ a consistent method in terms of which side of the ice is which.

    That’s just a “mirror” error though – if a guy shoots a puck from a good spot on the ice, it will still be a good spot on the ice if the image is flipped, what with symmetry. With that said, here’s a shot map of Jordan Eberle at ES in 2010-11, when he shot 9.7% at ES…

    and here’s one from 2011-12, when Jordan Eberle shot 18.8% at ES.

    If you can see some sort of a difference that would result in his shooting percentage doubling, you’re a better man than I. Or, more likely, you’re seeing things.

    Now, this isn’t to say that Eberle doesn’t actually tend to take higher percentage shots than other players. Here, for example, is what Brian Rolston’s shot chart at ES looks like in 2011-12. You can see a significant difference.

    I don’t know of anyone who doesn’t think that Jordan Eberle is, in the long run, likely to be something better than a 9% shooter at ES. When I did my own estimate of his true offensive talent, I just assumed that he’d have the best on-ice S% of anyone outside Crosby (implicit in that is that he’s going make far more shots than average himself) when I came up with 50-60 points. I’ve no doubt that part of the reason for that is that he doesn’t tend to bomb shots from anywhere.

    The thing is though, that’s not a new development. And, while it might be a reason that Eberle shoots 9.7% or 10%, it’s not really an explanation for reeling off an 18.8% season. So…nothing new to see here.

    Email Tyler Dellow at mc79hockey@gmail.com

    About Tyler Dellow

    9 Responses to The Grinch

    1. john sonderberg
      December 25, 2012 at

      I’m wondering if

      a) the use of RFID tags in player skates and RFID readers in/around the benches, could keep track of athlete ingress/egress and therefore do the time-on-ice accounting more automatically, so that the eyes-in-the-sky could put more focus judging on-ice events (like shots).

      b) the addition of RFID tags inside the pucks and RFID readers in an under-the-ice grid, could track puck position-&-velocity-in-proximity-to-players, and therefore perform an automated (or at least assisted) tracking of shots …. where on the ice & who (most likely) took them.

    2. PDO
      December 26, 2012 at

      Must say… I wanted to believe.

    3. TigerUnderGlass
      December 27, 2012 at

      This is not meant to sispute your point here, which I agree with, but do you have any idea about the accuracy of the shot charts? I’ve never spent much time with them, but I do know that the shot charts generated by the NBA are horrendous and do not give any sort of realistic representation of shot locations. Since I’d expect shots to be easier to chart in Basketball given the relative pace of the game I’d have a hard time relying on this data for anything.

    4. TigerUnderGlass
      December 27, 2012 at

      This is not meant to dispute your point here, which I agree with, but do you have any idea about the accuracy of the shot charts? I’ve never spent much time with them, but I do know that the shot charts generated by the NBA are horrendous and do not give any sort of realistic representation of shot locations. Since I’d expect shots to be easier to chart in Basketball given the relative pace of the game I’d have a hard time relying on this data for anything.

    5. TigerUnderGlass
      December 27, 2012 at

      Sorry about double, when I tried to comment I got an error message but it looks like it went through after all.

      • Brad
        December 27, 2012 at

        The charts may not be correct, but barring a new standard of recording, these charts should be recorded consistent, at least. The Eberle charts are identical, and I think it’s safe to assume that his shots from both years reflect symmetry, no matter how the shot tracker places Eberle’s spot.

    6. Jim
      December 29, 2012 at

      Nice contribution to what has been an interesting discussion over the long term.

      Strangely enough, in Rolston’s case, I seem to recall a few of those long-distance bombs actually going in. Those were all in games against Edmonton, however. Just imagine what Eberle’s S% would look like if he played 8 games per year against the OIlers D and goaltending…

    7. Kris
      December 30, 2012 at

      The shot chart data might be awful.

      Maybe a better thing to look at would be the ration of shots to “Dennis-Scoring-Chances” that Eberle gets (at ES only, of course). A slapshot from far away with no one around and no rebound won’t get recorded as a chance by Dennis, but the shots that Eberle says he takes a lot of would get counted as chances, or at least many of them would. So if Eberle is right, he should have a higher ratio of chances over shots (at ES, of course) than other players.

    8. Jay
      January 3, 2013 at

      Coupla points…

      1. If you’re Jordan Eberle, what are you supposed to say? Gee, my shooting percentage was sky high–must’ve been a complete fluke. I expect my shooting percentage to regress to the mean. Now, where’s that contract extension we were talking about?

      2. Random theory… Shooting percentage doubles. One year sans the Nuge, the other avec Nuge…? Not to say puck luck didn’t play some component, but you’d have to expect a higher % playing with an elite passer than without. Any data on this to make me look smart for once?

    Leave a Reply

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