Reviewing Flames Zone Starts

One of my off-season projects this summer was to try to gauge the progress of some of Calgary’s younger or emerging players. One of the metrics you can use to try to approximate a coach’s confidence in a player is zone starts.

Zone starts measure what percentage of a player’s shifts begin in the offensive zone at even-strength, ignoring neutral-zone face-offs for simplicity. (e.g., O-Zone Start % is Offensive Starts divided by Offensive + Defensive Starts). A player that starts a lot in the offensive zone needs high ground to succeed, a player that starts mostly in the defensive zone can get high ground on their own. At least, that’s the theory.

Here’s a look at the rolling 10-game zone-start trends for a few different combinations of Flames for the entire season.


Sean Monahan was a teenage rookie, thrown into the NHL. Joe Colborne was a center on a depth line, at least initially.

Screen Shot 2014-06-03 at 2.18.47 PM

Monahan’s the dark green line, Colborne the light blue one. Monahan got the high ground early and kept it until late in the year, when he was a relative veteran and even newer rookies needed the high ground. Colborne wobbled back and forth a lot, but settled into the “middle” group as a winger, getting roughly even zone starts until late in the year. In short: Monahan was heavily protected until (a) players that needed protecting more arrived or (b) the coaches had confidence that he didn’t need it anymore. Colborne’s usage seems more context and linemate driven.


Three pending RFAs to consider: T.J. Galiardi, Lance Bouma and Paul Byron.

Screen Shot 2014-06-03 at 3.36.25 PM

Galiardi (in yellow) got buried in terms of zone starts early-on…and his counting stats weren’t amazing. So he got more and more high ground – primarily with one or both of Brian McGrattan or Kevin Westgarth – and his numbers remained not great. Take from that what you will.

Bouma (blue) was more or less started consistently in the defensive end all year. His stats were just fine considering who he played with, and honestly, his line with Matt Stajan and David Jones was really good at their role when they played together. You can bury Bouma and he’ll be fine, and it seems like the coaches had confidence in him.

Byron (red) was alternately given tons of high ground or was buried throughout his tenure. When he recovered fully from his foot injury (around where that 90% peak is), his play really picked up and he was given less and less high ground through the balance of the year. A healthy Byron appears to drive possession reasonably well regardless of zone starts.


Screen Shot 2014-06-03 at 2.25.24 PM

Backlund (in green) had his starts vary a bit during the year, as his role changed quite a bit. Note that during his early zone-start peak early on, he was a healthy scratch. Then he seemed to figure things out. When the Flames were healthy, I think the coaching staff tried to keep his starts even, but injuries forced them to give him more D-starts later on.

Brodie (red) and Giordano (yellow) stayed together most of the year, and tended to get the toughest competition all the time, so their zone starts varied depending on what end the other team’s top guys started in (and the flow of the game).

And SHOCKINGLY, late in the season, the coaching staff discovered that things tended to go well with these three guys on the ice, so their trend lines converge at the end of the year (and trend downwards, as the team’s three best possession players don’t need the high ground as much as the rookies do).


Any surprises here? Is there any combination of players you’d like us to compare or contrast? Fire us a line in the comments.

  • KH44

    What kinds of trends do we want to see as the year goes on? Obviously steady, with no severe peaks, but what should we be looking or hoping for when reading the data? Ideally, should the top d-pairing having a zone start below 50, the second pairing at 50, third pairing slightly north of 50? What constitutes growth? What can we expect next year of a player like Monahan, who was heavily sheltered to begin the year and with progressively exposed zone starts from there, and going from +60% to sub 40% in the last 22 games of the year? And for centers, what about faceoffs? A player that is more successful at draws should get more defensive zone starts, and might be pulled as soon as the puck clears the zone, how does that affect the data?

    • BurningSensation

      Indeed,but as I watched the playoffs and the Kings especially I wonder how close or far away the Flames really are; after all we did beat them twice in LA during the season. After looking at both teams we have a long road ahead of us but some of the ground work has been laid.

      The good: the work ethic that Hartley instilled in the flames is comparable to the work ethic of the Kings and this might be the hardest thing to develop and can be the hardest thing to maintain.
      Our top defensive pairing is almost that good (neither Gio or TJB is as good as Doughty; not many are). After that things drop off.

      We have some poor man’s versions of Kopitar(Backlund), Brown(GlenX when healthy), Gaboric(Hudler) but a lot more question marks than answers. Firstly they have a world class goaltender and we do not and may not have one in the system. Two: Sutter rolls 4 lines and plays them in the appropriate places; the lines are all defensively responsible and can create offense. They have done so by developing the work ethic and providing opportunity for their young guys along side their stars. There is no separate goon or energy line. Three: there second and third pairing are more like second pairings than ours. They play as TEAM and they all put the TEAM ahead of themselves.

      So what do the Flames need to do to continue to move in the direction of the Kings and the other top teams. 1. Continue to develop the TEAM concept and work ethic they established last year and in order to have players continue to buy into this you need to have 4 lines and 3 defensive pairings you can trust. Last year our so called forth line forwards on some nights played less than 4 minutes and when they did play they we often quite ineffective leaving the hard work or other lines tired and ultimately ineffective as well. The goods new from how I see the Flames this year is that the talent level of the prospects provides the coaching staff with the opportunity to roll 4 effective, solid lines while developing the young kids.
      2. Keep Gio and TJB (the trade talk of trading either of them for Ekbald drives me crazy, mostly because if the rumors of what the Leafs offered are true then it would take more than either of them to get the pick and IMO he will not be THAT good) together, We need to upgrade the back end; Russle, Wides and Smid are fine 3rd pairing guys but we need to add some more 2nd pairing guys.

      We are long way from the playoffs and being one of those teams others want to be but there is some hope, the cupboards are not as bare as they used to be.

      • BurningSensation

        “So what do the Flames need to do to continue to move in the direction of the Kings and the other top teams.

        1. Continue to develop the TEAM concept and work ethic.
        Have 4 lines and 3 defensive pairings you can trust. Roll 4 effective, solid lines while developing the young kids.

        2. Keep Gio and TJB (the trade talk of trading either of them for Ekbald drives me crazy).
        We need to upgrade the back end; Russle, Wides and Smid are fine 3rd pairing guys but we need to add some more 2nd pairing guys.”

        Great question Gussey. Have taken license to summarize your comments above which make some great points, to which I would add the following:

        A. 2014 Draft: get a 1C/2C, plus 1RW/2RW OR Top 3 Defenders. Preference for some size to offset current roster in order to effectively compete with top 6 Western conference teams.

        B. Sell high: plan to trade 2-3 vets (retaining 1-2 for leadership) for 2015 picks this summer or at the trade deadline.

        C. Patient Development: focus this year to develop prospects; continue positive team culture and identity; rotate up top prospects meriting NHL time.

        D. Trade to fill holes: Use the growing overabundance of prospects and the strength of the Flames farm system to strategically acquire key talent by position.

        E. Promote realistic expectations by the team and fans alike. Yes to Winning; Yes to Developing; Yes to Patience; Yes all of the above is not going to happen overnight..give 1 more good developing and drafting year before making a run for the playoffs.

        • BurningSensation

          To point B; when and whom? Personally I think we need to start the season with a veteran presence but I can see making trades during the season as both our emerging players show development and the prospects make steps, however if you trade too soon much of the development could be delayed.

          To D: make smart trades but also explore character pieces via the UFA market, there are good role players available.

          and I’ve been called worse things than gussy.

          • BurningSensation

            re gussy…yeah me too. LOL

            re Vets …yes need to keep most for the first part of the year but can see by March a couple should be traded for picks and/or strategic holes, even if just to free up slots for younger roster players.

            Regarding the draft…my highest priority if I was Burke & Treliving (sounds like a tea brand) would be to trade for a #10-#20 1st rounder. Adding a Virtanen/Tuch…or a Fleury/Sanheim/Honka in addition to a Draisaitl/Bennett/Reinhart would be a huge bonus.

          • BurningSensation

            As the old “Wendy’s” adds used to say; “Where’s the Beef”; if the teabrand does add another high end pick I would add Ritchie to their list.

          • BurningSensation

            Sold!! Count me in if we can get Ritchie as a second selection in the 1st round…at 230lbs he would have a presence plus some skill.

            Am optimistic (a wish more than a firm analysis) for a couple of the heavies in the system to make the roster e.g. Wolf, Van Brabant, Ferland, Roy, Kanzig. Would really help overall team size without then having to resort to adding a goon or two.

          • BurningSensation

            THe key is what would be the cost? I would think our 2nd, a roster player and a serious prospect. Teams I guess we might trade with would include the Leafs, Coyotes(Burkes and trevlings contacts) and the Flyers who have expressed interest in Glenx,

            However back to your point of Wolf, Van B and Ferland; I suspect it is likely one of them makes the roster out of camp. I suspect that the opening day roster(barring trades) will look something like this up front: on LW: Glenx,Hudler,Johnny, Sven and Byron,at C: Stajan, Backs, Monahan and one of Draftpick/Grandlund/Knight, on the RW: Bouma. Colborne, DJones, McG and one of the three guys you mentioned having 14 healthy bodies up front. It would not surprise me at all if Number 4 gets his 10 games and both Granlund and Knight get sent down.

          • BurningSensation

            Agree with your general roster selection but am thinking:

            RW: Hudler, Colborne, DJones (if not bought out or traded), Wolf/Arnold/Ferland

            C: Backlund, Monahan, Granlund, Knight/Bouma

            LW: Glencross, Gaudreau, Baertschi, Byron/Bouma/VanBrabant

            Extras: McGrattan, another from above

            Hudler can play RW on the same line as Glencross. Also Poirier comes up on RW after rehabbing in Nov/Dec/Jan with the AHL Flames and DJones gets traded (so look to see his PP time pumped up etc to bump his trade value along with Hudler, Glenny and Wideman).

            I see 7 guys fighting for 3 Bottom6 spots. Also Hartley will ensure the meritocracy reigns, that top6 prospects Gaudreau/Granlund/Baertschi earn their time.

          • BurningSensation

            HUdler is listed as a LW which is why I have them there. Both Colborne and Bouma had their best moments on the wing. I would probably start the season with Colborne/Money/Hudler playing together b/c of their chemistry from last year. If you play Backs and Bouma with a scorer their possession skills and mind set could be dynamite.

          • BurningSensation

            “Adding a Virtanen/Tuch…or a Fleury/Sanheim/Honka in addition to a Draisaitl/Bennett/Reinhart would be a huge bonus.”

            To summarize: We’ve got quantity…we need top quality.

  • MattyFranchise

    These graphs are pretty much spot on for the eyetest. McGrattan and Westgarth started off alot in the OZone and ended up hemmed in their own end until BBG could get on the ice in the DZone. The difference was night and day, it was very noticeable the talent gap between those players and you can see why the coaches started them where they did.

    However, what I am curious about is a WOWY for zone starts concerning Monahan and Hudler. Was Hudler still getting the same sheltered zone starts when he wasn’t paired with Monahan? Or was he played up the rotation a bit when they were separated? Particularly when Monahan was injured.

    Basically was Hudler given tougher assignments when he wasn’t paired with a rookie, is what I’m asking.

  • Parallex

    Hey Pike,

    Great article, as always…

    I’m curious, on twitter a while back (during the reg. season) you mentioned you may start tracking controlled zone entries/exits for the flames. just wondering if you ever got around to it and what/if anything you pulled from that.

    I started focussing on this area during the tail end of the season and I really think this is a great gauge of a players overall effectiveness particularly on the backend. Guys like Gio, Brodie and Backs obviously seemed to kill at this and even Wotherspoon seemed pretty adept at it.

    just curious for yours and everyone else’s thoughts on the subject.

    • Zone entries is the next phase of the advanced stats movement in hockey. Problem is, they are a real pain to track. Someone needs to develop a tracking app similar to the scoring chance one we use to make collecting the data a lot less painful.

      • DragonFlame

        All you analytics guys have to do is convince the NHL to go digital.

        Each player has a chip in their uniform, and there are chips in the various zones on the ice.

        It must be hell trying to track any sport that changes on the fly. Computers and tracking chips could change it all and really enhance the game and how we view it.

        The NHL is way behind in analytic information to the public. Watch an NFL broadcast and they blow the viewer away with graphics and statistical information. Why hasn’t the NHL caught on that people love stats?

        • BurningSensation

          The NBA is already crushing this with ‘SportsVu’ cameras that track player movements everywhere on the floor and provide information at a level of detail unheard of previously.

          Things like;

          ‘How well does player X pass the ball after three dribbles’?

          ‘What is player X’s shooting % when dbl teamed in the low post on the left side vs the right?’


          Using SportsVu for something as simple as a zone entry would be child’s play.

          It’s coming soon, and I’d bet Burke is an early adopter.

          • mk

            The biggest issue with technology with SportsVu is that same with that UBC research: the puck is significantly more difficult to track visually when compared to a basketball.

            The passing information is something I’ve always been interested in, especially once I tried to wrap my head around Tanguay’s value a few years ago.

        • mk

          If you look at FoxTrax, the technology exists. However, it didn’t get serious buy-in from ‘the establishment’ and the potential usefulness for analysis was wasted on glowing puck graphics. Sounds like lots of teams complained about the difference in ‘puck feel’ and didn’t get to practice with the chipped pucks because of the cost.

          At UBC, there is work into using visual tracking methods to automate compilation of statistics, but it is still in its infancy. Their algorithm to determine player possession is very basic. Not to mention, they face some restraints given visual restrictions (players/refs/boards/netting/nets blocking lines of sight, etc.). One of the papers is in the link below.


        • Byron Bader

          Agreed! Sensors in pucks and nets so you can see if a puck’s in without having to rely on the human eye is also something I’d like to see but maybe that makes the game too “robotic”.

          Tough convincing some of those old school gate keepers but in 5-10 years or so I think we’ll see some pretty interesting in-game stats stuff. Corsi went from non-existent to a staple on most game broadcasts in a matter of months last year. It’s coming.

          • MattyFranchise

            Speaking of Corsi, it blew my mind when I saw them mention it on in an article early this season. I thought this stuff was for pantsless bloggers?

  • ChinookArchYYC

    Really interesting study here Ryan.

    The zone starts by Hartley are a welcomed change from the B. Sutter years. And, if he has more players like Bouma, and less like McGrattan and Westgarth he may be able to put the more offensively skilled guys in the o-zone where they can actually put some points up.

    It was maddening to watch Sutter march out the Flames worst players, for an o-zone start time-after-time, so that they wouldn’t be a defensive liability. Vigneault is very good at putting players in positions where they’ll exceed. In short, more Vigneault and less Sutter.

    Also, I’m so glad Vigneault is no longer in Vancouver.

  • FeyWest

    I really like these, paints a clearer picture for a full season of play. Curious as to what you use to input your data that spits these out, or is it Excel? I don’t recognize the graphs is all.

    Keep up the great work!

    I think Byron covers the guys I’m really interested in, and maybe Butler to give an idea about where he’s played I know he’s been shuffled around a bunch.

  • Byron Bader

    Good things, Pike. Cammy, Hudler, Stajan and Russell are ones that I would like to see. Also Glencross (even though he only played about a 3rd of the year). His offensive stats were better in the games that he played than I would have thought … curious if he was getting all gravy zone starts when he would return from injuries.