• About Last Night and Taylor Hall

    by Tyler Dellow • January 27, 2014 • Hockey • 4 Comments

    If you don’t follow my Twitter feed (admittedly, a high volume eclectic mix of Oilers, NHL, libertarianish complaints and cane waving at contumelious clouds but @mc79hockey if you’re into that sort of thing), you may have missed a graph that I put up over the weekend.

    I posted this on Twitter and all the wags came out.

    Assuming it’s none of these things (although the missing leg theory is intriguing…), it really does seem to me like something that the coaches have changed, or Hall’s failure to implement those changes correctly, is most likely to be responsible for the cratering.

    If you missed last night’s game, 93/4/14 were absolutely dominant. Hall played just 15:09 but the line absolutely crushed the Predators when the score was close – they won the shot attempts 13-5, even if they tied the Preds 1-1 at 5v5 in that time. If you had a line that won the shot attempts 13-5 every night, they’d be the greatest line in hockey and would finish the year with one of those cartoonish Bobby Orr +/-s from the 70s. The only reason that their night doesn’t look more dominant is that once the Oilers were up 3-1, the foot came off the pedal for these guys. Nashville beat them in the shot attempts 8-2 from that point forward.

    As I mentioned in my lengthy post on Hall the other day, I’ve been kind of wondering if the problem lies somewhere between what happens when the Oilers recover their puck in the defensive zone and when they hit the opposing blue line. I noted that the rate at which the Oilers get at least one shot attempt on a Hall shift this year is down from 50% to 40%, which is a pretty massive drop. Last night, Hall was fantastic in this category – when the score was close, the Oilers got shot attempts on eight of his eleven 5v5 shifts.

    I thought it might be instructive to go back and take a look at how the Oilers were breaking out and entering the offensive zone last night. To me, it looked really good. That being said, there’s a bit of a problem with doing this – it’s kind of like trying not to think about a pink elephant. If you have an idea and you go look at it, you tend to see what you want to see. So I’m still not at all sold that I’m right. The video that follows is as much about illustrating what I’m talking about as anything else.

    I want to underline something here: I make no claims to being a technical hockey brain. I don’t have the background in it. I’ve got some ability to ask questions and isolate things using data. After that I’ve watched and played enough hockey that I can think about what the data tells me and where I might look for an answer in a hockey game. That being said, a lot of the stuff that’s been done on zone entries and exits and how it plays into whether a team generates shots or not is very compelling so I think there’s good reason to be looking here for the answer to Hall’s problems.

    Here’s how last night played out:

    Dallas Eakins talked about Hall briefly on Bob Stauffer’s show today, saying “I think with Taylor, he’s taken on some of our teachings almost a little bit too far, become maybe too safe. We’re trying to give him a little bit more rope so he can get back to even a higher octane game. That says a load to me about the player, when he’s taken the coaching a little bit too far, trying to listen, trying to buy in and those are great, great signs for me.”

    While that’s a fascinating quote, I’d love to know what it means. Unfortunately for me, coaches rarely explicitly say how they’ve loosened the reins – they just say stuff like this. So I can’t really tell how they’re thinking or whether they view Hall’s numbers as a problem from it.

    To really be on firm footing, you’d want to find some way of quantifying the Oilers’ zone exits/entries with Hall on the ice this year a little better, in terms of where players are, whether the puck carrier is supported, things like that. When the NHL gets the kind of data on player motion that the NBA has, coaches that get it are going to be able to do phenomenal things with it with very little physical tracking of things. The few tests/different angles of looking at that I’ve done have me thinking that this is where the problem lies but I don’t consider the point close to proven either way yet. That said, they were flying against Nashville last night.

    A final point: a lot of people who are opposed to the use of data in hockey make the point that hockey isn’t baseball. First of all, let’s not kid ourselves. These people have not thought deeply about it and come to this realization. It’s knee jerk nonsense that ignores the tremendous results that people have achieved applying data to basketball. Second, it’s just wrong. If you look at that video I posted, things are repeating over and over and over in a hockey game. Zone entries and exits, over and over.

    What a coach is trying to do is make the chaos take place within a structure. This, to me, is far more interesting that questions about why Zack Kassian was not facewashed more times and probably has a hell of a lot more to do with winning. I highlighted two examples in that clip, one in which the Oilers had two players on the boards coming out of the defensive zone and one in which they had a player on the boards, one supporting him in the middle and one coming up behind. In both cases, the pass of the wall failed. The first time, Nashville went right back in. The second time, Nugent-Hopkins picked up the pass that Hall missed and five seconds later it’s 2-1.

    This sort of stuff repeats over and over. The pass being clean or not, that’s the chaos. Figuring out how to have your guys in the right spot so that the chaos works for you, that’s what the coaching is.

    Email Tyler Dellow at mc79hockey@gmail.com

    About Tyler Dellow

    4 Responses to About Last Night and Taylor Hall

    1. cartooncolin
      January 27, 2014 at

      Great stuff. Quite shocking to see how many decent entries were nullified by bad decisions, turnovers, little support or simple shots off the rush that had no hope. They could have dominated beyond all reason by just taking 25% of those entries and creating sustained pressure. One thing at time I suppose? Zone entry at 6:40 was effective (3 on 3) with RNH making one of those small plays in the corner that makes all the difference.

    2. Vik
      January 28, 2014 at

      The biggest thing that stood out to me in those clips is how easily Josi gets backed off the blue. Maybe it’s the Swiss rope-a-dope in him that he doesn’t want to engage but he sure seemed to be making it easier for the top line to gain the zone.

    3. Woodguy
      January 28, 2014 at

      Figuring out how to have your guys in the right spot so that the chaos works for you, that’s what the coaching is.

      Very well put sir.

      Thanks for the long hours it takes to put this stuff together for the betterment of all our knowledge of the game.

      So, when’s the video of the face washes and the ensuing shots caused by said washing of the face?

    4. Johnny
      January 28, 2014 at

      As one who isn’t “all-in” when it comes to advanced stats, I’ve used the baseball comparison in the past (and still do). I don’t use the comparison because I’m “opposed to the use of data in hockey”. I simply make the comparison because in baseball it’s very easy. Plays are easily defined and easily coded in a binary way, resulting in pretty robust data that can be analyzed.

      Basketball is not hockey. Yes, it’s more similar to hockey than baseball, but it’s easier to draw some cause-effect lines to scoring plays because they happen with such great frequency (80-120 times per game for each team — in 48 minutes). Hockey is similar in the respect that the game is fluid and the players have tendencies, styles, assignments, habits and preferences that, when tracked, will likely provide results with high reliability, but my question always boils down to: what are we measuring?

      Basketball does not have on-the-fly shift changes, they don’t ever play short-handed, match-ups remain relatively constant throughout a game, there are no face-offs (save for a handful of jump-balls), there is a shot clock and an over-and back rule (a single “zone-entry” per possession). Yes there are similarities, but the conditions and intervening variables are significant.

      I like the zone entries project and the reason I like reading your stuff is because you take a multi-faceted approach to addressing issues — you don’t accept things without question and you have an excellent mind for asking questions and synthesizing data.

      Don’t get me wrong about stats. I think the field of advanced stats is very exciting, but someone needs to continue to challenge what we are actually measuring and what conclusions can be drawn from the data. Reliability is great, validity is a tough one.

    Leave a Reply

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