A Tale of Three Cities: A Series of Charts

The Calgary Flames are a successful team, standings-wise. They enter today’s NHL action with a 16-8-2 record, good for a 65.4 points percentage despite the Flames not having an amazing Corsi For percentage, which is held by the analytics community as a predictor of wins and points.

For some added context on how the Flames are doing, let’s compare them with Edmonton and Buffalo – two truly not-good teams – on a Rolling 10 Game basis!


Okay, here’s Alberta’s PDO (shooting percentage + save percentage).

Screen Shot 2014-12-02 at 11.53.37 PM

As you can see, Calgary’s been above the 100 mark for basically the entire season. Edmonton? Well, they are not getting the bounces. At all. Ever. Like, to an epically comedic level.

Now, PDO is thought to be resistant to dropping/rising to 100 quickly if a team is appropriately awesome/awful. Meaning, Boston, with its excellent Corsi and good team, generally sticks above 100 PDO, while bad teams would be resistant to getting better.

Here’s a Corsi comparison!

Screen Shot 2014-12-02 at 11.54.03 PM

Calgary’s been pretty consistent, floating around the 44% mark all season. Their highest rolling 10-game segment creeped up to 46%. Edmonton? Variations between 48% and 52%. They’re a good possession team, while Calgary is below average.

One team is winning because they are getting bounces. The other team is Edmonton and has lost 10 in a row.


As with Edmonton, here’s a PDO comparison.

Screen Shot 2014-12-02 at 11.55.10 PM

Buffalo started off being fairly unlucky, and recently have gotten the bounces more and more. Calgary? They’ve been fairly consistent, particularly when compared to Buffalo.

How about a Corsi glimpse?

Screen Shot 2014-12-02 at 11.55.28 PM

Buffalo is bad. Like, really bad. Like, they hit 40% recently and it was insane. They’re holding parades in upstate New York for giving up “only” 60% of shot attempts in a game against. They will start losing soon. Like, sooner than Calgary might.


Are the Flames appreciably better or worse possession-wise lately?

No, not really. They’re about the same. But other teams (Edmonton) are getting terrible luck suddenly while still being good, and other teams (Buffalo) are getting incredible luck suddenly while still being fundamentally bad.

  • Kybb79

    The Avs and Flames meet tomorrow evening.

    Calgary wins 4-2.

    An unexpected phenomenon occurs during the game when both teams produce negative corsi/fenwick numbers.

    Spreadsheets around the world short circuit.

  • JMK

    I don’t believe in luck but will agree that the perception of luck can create/destroy confidence levels. Calgary may not be the most skillful team in the league but playing with confidence and togetherness in my opinion (don’t know much about PDO or advanced stats in general) can’t be quantified.

    In terms of Corsi, I would love to see an article done on all the Stanley Cup/President trophy winners in the past. Do possession teams always do the best? Has a team/manager ever changed up the tactics and won without being a possession team? I don’t know if people follow many other sports on here, and also whether or not the same way of thinking can be applied, but if you watch soccer or gaelic football (2 sports I’ve followed for years) the skillful possession based teams do generally do well and win quite often but every so often a defensive style team or counter attacking style team comes along and dominates maybe just for a year, maybe for a few years, maybe they change the face of the sport.

    Gaelic football for example is an amateur sport but in recent years it has become almost professional (in regards to training, tactics, etc. still amateur offically). For years teams won based on more skill, and thus one could correlate it with more possession. But come 2002 a defensive style (which invited teams on to them and instead of wanting to gain possession were patient and waited for the mistakes) came along and defensive teams won the top prize in 2002 & 2003. In 2004 a skillful team won it back only to be beating again by a defensive team in 2005. In 2006 that same skillful team from 2004 decided to change tactics to win, they still used skill and offence but a hockey translation might be they had a big guy crash the net all the time and thus kind of circumvented the defence. This worked for two years. Skip forward a couple of years to 2011, and again the system changed where defensive won again, and ultra defensive won in 2012. 2013 then marked a fast paced counter attacking style that won and the defensive sides were annihilated. So where teams with more possession used to win, with the changing of tactics teams with less possession began to win, and now it’s trending towards teams similar to Calgary who defend and give up a lot of possession but then attack very quickly.

    I know that’s an extreme case where the tactics have been going back and forth in a relatively short space of time but my point is they are constantly changing. The NHL is a more established game as it has been professional for a long time, but do possession teams (the Gaelic football skillful attacking teams) always win or has there been trends where a non-possession based team (the gaelic football defensive / counter attacking teams) win??

  • My issue with all the stats is how do they chart effort. Watching last nights games how many plays did the Flames make at both ends that contributed to luck or luck being produced by effort. Both Jones and Monahans goals had an element of luck but both were created by effort that created the luck and both had to be capitalized by their shooting and the scoring area the shot came from.I watch the game and I see almost on every shift guys either busting their humps to get back into the play or to get to the bench to allow fresh bodies on the ice. I’m not sure I see that when I watch the oilers play.

  • Kybb79

    Here are the things the Flames do well:

    1) The entire team seems to have bought in to shot blocking.

    2) The team is also going to the crease at the other end, and are willing to score greasy goals via “banging and crashing the net.” Most of their goals aren’t pretty, and they don’t need to be. This is not a “perimeter team.”

    Yes, the Flames have had luck on their side. This is not in question. But the age old adage could be applied here: the harder a team works, the “luckier” they get.

  • MWflames

    Question of the day:

    Are the 2014 Calgary flames better with stajan in the line-up?

    Trash for “no”
    Prop for “yes”

    … I’m going to trash my own comment.

    • playastation

      At the very least, he is a better possession player than all the child centers. I would at least say better than granlund if you’re gonna ask who. I think we are all worried because they’ve been winning so much without him that tinkering is bad. But major knee surgery? Can we at least get him rehabbing for a few months?

      • MattyFranchise

        I’m pretty sure he did not have major knee surgery, if he had he would be gone for much of the rest of the season (ie Ferland) Where he helps over one of the kids is face offs and this would help on all the stats.

        • MattyFranchise

          I neither trashed nor propped your post:

          I am in wait-and-see mode.

          It’s on Hartley and staff to integrate these injured players back into the lineup without disrupting team chemistry.

          It would be nice to see a spike in possession stats so people would stop telling us how bad the Flames are.

  • Not sure about this article. It doesnt really hit the mark.

    A huge part of the Oilers “bad luck” is caused by their inability to play defence and the fact that they give up way to many scoring chances. Corsi doesnt factor in quality of shots.

    The flames battle hard for every puck and do everything they can to limit quality chances.

    I get that this is a look at purely stats but I get the impression that this article was written by someone who doesnt actually watch either team; written by someone who is good at Math but knows little of the game of hockey.

    • ***sorry meant to say the oil give up way too many quality scoring chances.

      If you watch the games then you know what I am talking about. Look at the shorties they surrendered to Phoenix.

      Off topic but great to see Reider score against the team that traded him for nothing.

  • The Sultan

    Just watched the Oilers give up a 2-1 lead with less than 5 minutes left, Hall gets flattened, Jets score seconds into overtime. 11 straight losses for the losers up North.

    • McRib

      Wow I was also watching it what a collapse by the Oilers… Andrew Ferance let both players walk out front to tip point shots on the last two goals. He is really looking bad this season.

    • KiLLKiND

      Have to be careful here of those 11 loses they have lost 8 by one goal 4 in overtime/shootouts.

      They are not playing badly for the most part, just have brain cramps at the wrong time.

  • When we’re dealing with a young team like Calgary, and players like Brodie, Monahan, Jooris, Gaudreau, Granlund etc. who are improving and and still a few years away from hitting their peak, isn’t it conceivable that their possession numbers will improve as they develop?

    Heck, even Giordano is better this year than he was last year, so instead of the team as a whole regressing, isn’t it possible that this team actually gets better as the season continues?

    • de Animoe

      @flash, this is the most exciting thing about them Flames. They are only going to get better. The young players will continue to progress (Monahan is a great example) and prospects will replace players on the downward part of their career (Bennett instead of Stajan) or replac eplayers that are not than good (Wotherspoon instead of Smid).
      And we have tons of cap space.

  • KiLLKiND

    For those curious Setoguchi has 1 goal 1 assist in the AHL if he keeps it up I personally wouldn’t mind bringing him back up after 10 games-ish to see if he can convert that success to the NHL. Maybe his risk-free signing can still pan out and all he needed was a kick in the pants to get going.

  • McRib

    Mark Jankowski had two assists tonight for Providence, he also hit a post and was robbed on the doorstep on another chance. If a couple more bounces would have gone his way early this season (I have seen him hit at least three or four posts already in highlights) he would be better than a point per game player playing for a very defensive minded coach. Not to mention he clearly has battled injuries early on this season. Not a bad start at all considering as much, hope he stays healthy and starts to get a few more bounces.

    Jon Gillies numbers are starting to look more and more impressive he has only allowed one goal in his last three games (1.80 GAA / 0.941 Save %) after a slow start.

  • Quick thought: I think it would only make sense to use fenwick instead of corsi in these types of articles seeing the flames lead the league in blocked shots. Won’t change a ton but it would be more fair.

  • Percentages are useful but they can make your imagination run wild too. You can’t forget about the actual totals (ie. volume of shots taken per game). Calgary is near the bottom of the league in this stat too, it can’t be denied. But our avg shots on goal per game are 26.7, while the Oilers are sitting at 29.4. They are a whole 2 shots per game higher than us (or 7%), despite being 25% (relatively speaking, “0.25 x “40% = 10%) higher than us in Corsi. Percentages don’t really capture significance very well.

    They also can’t account for things like quality of the shots taken. A team like the Flames will no doubt make more of their opportunities than the Oilers because they are simply playing better right now.

  • MattyFranchise

    I always wondered what the advanced stats would have been for the old russian teams that would hang onto the puck forever but rarely shoot. Seems to me they would have been regularly outshot, and had bad stats…

    • RedMan

      Some people have done the analysis on this (sorry can’t cite the exact article because I don’t remember where I read it but I think Bruce McCurdy might have been the author). He also thought that he would see what you suspect (i.e. the Russians holding the puck and retaining possession but rarely shooting) but he was surprised to find that if you actually sit and count all the Corsi events in the games, the old Soviet had excellent advanced stats.