Hard To Kill

Hard to kill

Rare photo of Jim Corsi without moustache

Corsi is dead. Or at least it’s on life support and waiting for us all to say our goodbyes before the plug gets pulled. That’s in one form or another what Ken Campbell of the Hockey News wrote on the heels of the NHL giving Sportvision’s tracking technology a pilot run during the All-Star game. 

Is that true? Can the beat writers who labelled it all a fad breathe a sigh of relief?

The tracking technology, which is still incredibly expensive, will potentially one day be sewn into every sweater and embedded into every puck. This will give the NHL the ability to collect in incredible amount of detailed data. Whether it be positioning or speed or time of possession, the new information will be available quickly and accurately. 

And, of course, when there are new innovations in information collection all the data becomes super easy to understand and there will never be any more complications. Ever. All that really nerdy stuff like “Corsi” or “Fenwick” that’s just too advanced for the average person to comprehend will be left behind and replaced with straightforward stats that don’t require math classes.

Well, that’s one way things could go down. I mean, it’s not the most likely way the new stats will change the analytics landscape, but it’s a possibility. Technically. 

I, for one, don’t think Missed Shots + Blocked Shots + Shots is particularly complicated though. In fact, it’s so basic that I doubt it will disappear from the Hockey Stats Lexicon any time soon. Spearheaded largely by enthusiasts who did the work in their spare time, the Fancy Stats community has already taken something as simple as shot attempts and created a system of analysis that acts as a proxy for time of possession and scoring chances. 

Even if we get new data that gives us exact time of possession data would we want to abandon a stat that gives us total shot attempts? I hardly think so. These stats still give us impressions of offensive and defensive potency. We may be just scratching the surface of what we can do with it.


Is all of this Data getting out of control?

No, what’s most likely is that the new data will only add to the information we have, not take anything away. Soon we could be able to determine how quickly players and the puck move through the neutral zone. How fast are the Blackhawks on the transition? How slow do Pete DeBoer teams make you go in order break into the offensive zone? How long exactly does the puck stay in the offensive zone after a faceoff? Those are just questions I could come up with off the top of my head and I don’t spend my days thinking about these questions. Imagine the things guys like Tyler Dellow of the Oilers can come up with.

Additionally, it could help us execute analysis that we are already attempting much, much quicker than we can now. There will almost certainly be a revolution in Zone Entry and Zone Exit data. Right now it is incredibly time consuming to record this info but it could become very accessible. Imagine knowing a defenseman your pro scouts like contributes at a rate that is better than average at slowing down attackers as they enter the left side of the defensive zone, and you can accurately predict where he is most effective on the ice, in which situations, AND know how that relates to shot generation?

I sincerely doubt Corsi will be dead soon. The people who have pushed fancy stats to where it is today are being snatched up by NHL teams for a reason.

The information they provide is valuable.

New information won’t make the older, more basic, information useless. There will just be an expectation that analysis will become more nuanced, more complicated. Fenwick didn’t kill Corsi. Corsi didn’t kill shots. 

If there’s one thing I can be sure of, it’s that there’s no such thing as too much information. A lot of it might be noise, but I’ve never met someone who researches anything professionally who wants to do away with entire sets of data. Not everything will be used, not everything will be as important as it used to be, but the new era in stats will probably not begin with a eulogy for the last one.

There is, also, another reason Corsi might not die out so quickly: We have no idea how much of the information that is being recorded will be made available freely to the public.

How much data will be for club use only and how much will be uploaded to the NHL’s website is not set in stone. The NHL has an idea of what people want, or they should, based on what other websites are showing right now. The NHL’s website is almost useless because it lacks even basic filtering ability and they’ve been slow to adapt.

We know they’re working on it, but when the new data that is recorded by Sportvision is available will fans (and the media) have access to it?

I have my doubts.


Home Row is for Nerds

For the media guys thinking that new information will somehow make the fancy stats community go away and die of sadness in their mothers’ basements, the new era of fancy stats that’s about to begin will be a rude awakening. If shot attempts was too much to handle then what’s coming next is going to be another level altogether.

As always, how much any individual fan or writer wants to acknowledge or use these stats will be up to them. That said, as someone who writes about hockey, I would think that understanding the basics that we have now would have already been a pre-requisite for the job.

The anti-intellectual, anti-stat sentiment that more than a few writers express is foreign to me. I like to write. I like narratives. I like fights in hockey and human stories. I also like exploring new ideas and knowing more than basic platitudes that anyone can say regardless of any actual hockey knowledge.

Will Corsi die? It has already survived several attempts on its life by those who are so eager to pronounce its time of death. I don’t think it’s going anywhere.

  • Britts94

    Mike Richards is on waivers. I don’t care what kind of numbers he has this year or what his cap hit is. Pick him up.
    1. Nuge.
    2. Richards.
    3. Roy.
    4. Gordon.
    Actually looks like a capable lineup at centre.

      • Quicksilver ballet

        Who cares about his salary? Katz will pay anything his little buddies want. Cap hit is what I care about, but I’ll take Richards at 5.75 over Nikitin at 4.5 or Pouliit at 5 or Jultz at whatever he’s gonna be at anyday!

        There is zero reason not to grab Richards

    • Yeah, it’s like, he’s a perfectly capable centerman, who cares what he’s making, or that he puts up hardly any points! Oiler fans would never try to run a guy out of town based purely on making more money than they think he’s worth!

    • ubermiguel

      Contract notwithstanding Richards is the kind of offensive-minded veteran on the backend of his career I would love to have onboard mentor the kids, especially Yak.

    • Ever the Optimist

      I wouldn’t pick up that contract on waivers but if you look at nikitin for Richards and they keep a million of his contract then it is more doable. I think he is young enough to regain some of his former glory in a larger role with better wingers than he has had in LA for a long time ….. plus he knows what it takes to win and that is something that needs to be taught to our kids.

    • nugeformayor

      Looks like a lineup that would just make it outside playoffs and we would be overpaying a second line center. We can get Connor mcdavid or as i like to cal him baby mcjesus for entry lvl dollars for three years and he is a second line center off the hop and in three years or less will contest nuge for top spot or we will run two first lines sounds like a better plan to me franchise player!!!!

  • Lowe But Now High Expectations

    All these advanced stats are killing the game. Hockey is supposed to be about speed, creativity & body contact with the odd fight thrown in. Now it’s overcoached, all about matchups and clogging up passing lanes and stopping the puck from even reaching the net.

    The players today are far superior to any that played in the past. Bigger, faster, stronger & easily as skilled. But the product that’s presented is not very entertaining.

    • nugeformayor

      True these people probably have never even played hockey it is supposed to be a fun game the best game not robot players worried about stats that dont necessarily mean wins i hope this trend dies soon like with coachs like eakins that use these stats to ruin the game!

  • Rob...

    Until that chip can tell the referee whether or not the puck crossed the line I won’t be impressed. Absolute BS that in this day and age we are relying on a game of peak-a-boo with the puck.

  • Ed Hod

    Corsi is corny.

    It a justification for overpay on above average players that want superstar money.

    Being hit by a puck is not always a blocked shot, and starting on the ice against a line doesn’t mean you played against them the whole shift.

  • Ed Hod

    The only stats that matter are Wins and Losses. Corsi is just a stat that can provide an excuse for not getting the wins, or hope that wins may come “one day” because these fancy stats are telling a different story. Let’s all remember that in Moneyball the team lost at the end, despite a great run midseason. I tried to get behind Corsi, but at the end of the day, learning the stat didn’t make the losses more bearable, and we still gave up “GRADE A” chances (ie: dman loses a battle behind his own net, opposing player passes the puck to a guy in front for a tap in that the goalie didn’t have time to see or prepare for). I’ll take the lower Corsi but better win percentage of a Nelson than the other way around with Eakins.