• About Those Pyrrhic Victories…

    by  • April 28, 2014 • Hockey • 4 Comments

    I’m still in the process of figuring out exactly what happened with Gagner/RNH following offensive zone faceoffs after December 7. Nate Silver had a really good chart explaining this process in his 538 manifesto, which I will borrow via Tangotiger.

    This, like all analytics, is basically a process of figuring out what’s pertinent, collecting that information, organizing it and then putting yourself into a position to make some statements about why things unfolded the way that they did. I have little belief in my own ability to just watch a bunch of information and sort it out in my head, so I think that a process like this has value. I can’t speak for other people but I note that I didn’t really hear anyone saying that the Oilers kind of became awful at generating shot attempts post-OZ faceoff with Gagner/RNH on the ice this year, which probably tells me something. I did hear lots of MOAR HITTING.

    So in this case, I kind of have to figure out all of the variables that might impact on what a team does on an OZ faceoff. The big ones, I think, are the score, the time of the game and who’s on the ice for each team. Then I’ve got to gather the necessary information, organize it and see if I can’t suss things out. It’s a bit of a pain but I think you draw better conclusions with a process like this.

    A lot of people have guessed what I’m suspicious about though, so I figured I’d throw up a quick set of stills to illustrate it. We’ll start with a faceoff that the Ducks lost against the Oilers at 17:23 of the second period in a game on December 15. The Ducks were leading 2-1 at the time. The Oilers had Yakupov, Perron, Gagner, Larsen and Ference on the ice. The Ducks had Cam Fowler and Ben Lovejoy on the ice, with Nick Bonino between Kyle Palmieri and Patrick Maroon.

    In other words, Anaheim had a lead and the Oilers had some dangerous players on the ice. Here’s how it unfolded.

    The Oilers are lined up with Yakupov and Perron on the inside, Gagner on the dot, Larsen to Gagner’s right so he can take the puck on his forehand and then Ference tucked in behind Yakupov and Perron.

    Gagner wins the draw to Larsen. Maroon begins to pressure Larsen behind the net and make sure that Larsen can’t make a quick left hand turn to get out of trouble. Palmieri could not care less about Perron cutting away from him towards the blue line.

    The pressure from Maroon and Palmieri forces Larsen to get rid of the puck. He can’t make a direct pass to Ference without getting the puck on to his backhand, so he just shovels it at the boards, trying for a bounce play. This will make it difficult for Ference to handle it cleanly.

    Without a pass that he can take in stride, Ference is forced to bleed off any speed he might have had to take the puck. Palmieri’s presence means that he risks being checked and turning over the puck if he tries to skate it around the net. The presence of the net means that he can’t make a pass to Yakupov on the boards. This seems to be a feature of what Anaheim does – they’re trying to force the defenceman with the puck to rim it by closing him down before he gets around the net.

    Rimming the puck takes the speed off of it, which means that Yakupov has to pick the puck off the boards as opposed to taking a pass and gives the Ducks defenceman time to close in on him. Note that Bonino has maintained his position above Gagner. The worst case for the Ducks here is that the pinching defenceman does not get to Yakupov in time and Yakupov is able to make a pass to Gagner. Even at that, it’s not an easy pass to make and Bonino may well be in a position to pick it off. If Yak goes off the glass and out, Bonino probably recovers.

    Yakupov tries to make the safe play by going up the boards. It’s blocked by the Ducks D, which basically creates chaos in the Oilers end.

    The Oilers end up escaping, but only because this silly pass was not intercepted. Playing the faceoff the way that they did bought the Ducks a lot of chaos in the Oilers end and, like anything else, when you spin the wheel enough, sometimes you win.

    Here’s an Oilers OZ faceoff from December 17. Edmonton’s down 1-0 in the game with 1:08 left in the first.

    The Oilers have RNH, Hall and Eberle up front with Petry and Ference behind them. LA has Dwight King on the left wing, Kopitar lined up beside him and Carter taking the draw. Doughty is lined up to Carter’s right, with Jake Muzzin in behind King/Kopitar.

    Carter wins the faceoff back to Doughty. Note how Eberle has cut up and through the circle with Kopitar. This creates a huge vacuum where there’s going to be no pressure behind the net.

    Eberle is out of the picture. Doughty has moved the puck around the boards to Muzzin. There isn’t an Oiler on the right side of the ice except for possibly a defenceman who is out of the shot.

    Muzzin doesn’t handle the pass cleanly but he’s still got acres of ice and time, with two potential outlet passes and only Taylor Hall chasing him down from a distance.

    The Oilers do a reasonably good job of taking away passing lanes here but the Kings still escape cleanly, as Muzzin ends up passing the puck behind Carter. LA goes down the ice and into the Oilers end.

    At the risk of quoting Jurgen Klopp too often, I’m inclined to believe that “The best moment to win the ball is immediately after your team just lost it. The opponent is still looking for orientation where to pass the ball” applies in hockey as well.

    All of this said, anecdotes, even anecdote with lots of pictures, aren’t data. In order to prove this, I have to find some way of quantifying what’s done off the faceoff. I think that more of the Oilers faceoffs look like that one with the Kings than like what the Ducks are doing. Just a matter of collecting and organizing the data to prove it.

    Email Tyler Dellow at tyler@mc79hockey.com


    4 Responses to About Those Pyrrhic Victories…

    1. daryl
      April 28, 2014 at

      At the risk of circumventing the process, what other hypotheses are there to explain this data?

    2. verv
      April 28, 2014 at

      despite your description being very good, i think the entirety of your explanations – when accompanied by pictures – might benefit from labeling a few of the players in the first pic or two of each series (or when perspective changes)

    3. Steven
      April 28, 2014 at

      Do you have an inclination on how you plan on quantifying tactics? Presumably all of the actions done directly after the faceoff are related to coaching tactics (if not, that is even more damning than bad tactics).

      Would a good way of quantifying it be # of players who pressure the puck, or, # of players below the faceoff dot? Maybe number of players retreating vs attacking? I’m not sure which would be the best idea (or, probably none of my suggestions are the best way of categorizing).

      I’ve really been enjoying this series! Great work!

    4. Murat
      April 30, 2014 at

      There are some good things about faceoff loss tactics that should help with quantifying them.

      1) They repeat
      -By this I mean if MC79 is looking at faceoff tactics, he’s going to have a whole pile of faceoffs to go through. He’s not looking at open play or something with a lot of variables; he’s looking at a situation that repeats itself time and time again.

      2) There are relatively few faceoff alignments.
      -It’s not like there are 17 different offensive zone faceoff formations. Most teams have a couple of variants and that’s about it. This means that there will be a lot of situations which are comparable to each other because the players start in the same spot.

      3) There are relatively few different spots for the puck to be.
      -If you’re only looking at clean losses, there are only a couple of places the defending team will have the puck. This means that, over enough faceoff situations, you’re going to have a lot of repeated scenes where the same defender (LD or RD) has a puck in a similar place. The more of these you have, the easier it is to compare outcomes.

      Basically, if you combine #1, #2, and #3 and you sort out all of the comparable situations, all you then have to count is “the number of times Edmonton makes decision A” vs “the number of times Edmonton makes decisions B”. Decision A might be a winger crossing through the faceoff circle like Eberle does in MC79′s example. Decision B might be pressuring the defender behind the net like MC79 is suggesting as an improvement. Identify all of the times these situations happen, count the times Edmonton does A or B, and count the number of times Edmonton gets the puck back from A and gets the puck back from B. If Edmonton truly does have tactics, you won’t have to worry about arbitrary numbers of players pressuring/attacking, you’ll be able to isolate specific decisions and evaluate their outcomes.

      Just watch the games =)

    Leave a Reply

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