Nov 7, 2013; Los Angeles, CA, USA; Los Angeles Kings left wing Dwight King (74) and Buffalo Sabres left wing Johan Larsson (22) hit on the ice as they chase down the puck during the first period of the game at Staples Center. Mandatory Credit: Jayne Kamin-Oncea-USA TODAY Sports

Why the Buffalo Sabres Should Bring Johan Larsson Back

Sabres fans, I noticed that fellow editor Caitlin Campbell ran a poll the other day asking you what you think of advanced hockey stats.  A whopping 16

Oct 19, 2013; Buffalo, NY, USA; Buffalo Sabres left wing Johan Larsson (22) skates with the puck against the Colorado Avalanche during the third period at First Niagara Center. Avalanche beat the Sabres 4-2. Mandatory Credit: Kevin Hoffman-USA TODAY Sports

people voted – including me, and probably CC herself – which tells me that, even though the poll was split evenly between the choices of “Very insightful” and “They don’t hold much stock,” the fact that only 16 people voted really means that an overwhelming number of our readers find advanced statistics as useful as a poopy-flavored lollipop.

That’s too bad, because I’m about to share with you some advanced statistics that should convince you that Johan Larsson is the first prospect the Sabres should have recalled from the Rochester Americans when they began reshuffling the players on the roster a few weeks back.

For starters, if you have not read Rob Vollman’s Hockey Abstract, you may not understand what the terms Corsi and Fenwick even mean, and since those are the two advanced statistics I will be referring to the most, it helps if you know what the hell I am talking about, right?  Corsi and Fenwick ratings are used in an attempt to get a big-picture description of a team’s puck possession.  A Corsi rating takes into account all shots on goal, all shots that missed the net, and all shots that were blocked by the opposing team’s players, while a Fenwick rating leaves out blocked shots.  Essentially, these two statistical measures are a more reliable measure of what happens when a player (or an entire team) possesses the puck than the standard +/- metric, which only shows you the end result (of a shift, period, or entire game).

For most of the season, the Sabres’ Corsi rating at the end of each game has been negative, which is important to note because it not only means that the Sabres are being outshot by their opponents, but that they are losing the all-important puck possession battle, as well.  Clearly, if a team takes 60 total shots (on goal, wide, and blocked) in one game versus your 40, your opponent has an edge, as they are (usually) in possession of the puck more often and (usually) create more scoring opportunities than you are.  Recently, the Sabres have started to generate much more competitive Corsi and Fenwick ratings (in three of the past ten games, the Sabres have actually posted positive Corsi and Fenwick ratings), but there is still not one player on Buffalo’s roster who has played at least 10 games with the squad with a positive Corsi or Fenwick percentage.  In other words, if you ignore players like Matt D’Agostini (8 games), Luke Adam (9 games), and Matt Ellis (2 games), every single player for the Buffalo Sabres has seen more shots by their opponents than by the Sabres during their time on ice.

Every.  Single. Player.  No wonder the Sabres have only won 8 games.  It’s actually a miracle they have even won that many.

For a team that constantly loses the puck possession battle, then, and is routinely outshot by their opponents, you would think that the Sabres would want to stock the roster with players who have respectable Corsi and Fenwick ratings, right?   So why is Johan Larsson, who STILL has the highest Corsi and Fenwick ratings out of any Sabres’ player who has played in ten+ games, toiling down in the AHL?

You read that last sentence correctly: with the exception of the afore-mentioned D’Agostini, Adam and Ellis (Corsi ratings of 56%, 50%, and 50%, respectively), Larsson has the best Corsi rating out of any player who has played for the Sabres this season.  Larsson’s Corsi rating of 49.5% is a full two percentage points higher than super-rookie Zemgus Girgensons, and exactly 14 percentage points higher than John Scott‘s 35.5% rating.  In the 178 minutes of ice time that Larsson saw as a member of the Sabres, he was on the ice for 139 shots by Buffalo and 142 by opponents – and get this: when Larsson was on the ice, the Buffalo Sabres’ Corsi was 7.5% BETTER than when he was off the ice.  Again, that is the highest CF% rel better than anyone except Matt D’Agostini.   (If you are curious, the same can be said of his Fenwick rating.)

If everything I have written so far sounds to you like mumbo-jumbo, a bunch of foofarah, or voodoo magic, allow me to translate: when Larsson is on the ice, the Sabres barely get out-shot and possess the puck almost as much as their opponents.  More than any player who has logged significant ice time this season, Larsson has kept the Sabres competitive with their opponents.

And he’s playing for the Rochester Americans, why?

The smart hockey fan knows that there is more to a hockey player than just the number of goals he scores or assists he creates.  Larsson may not have tallied a lot of points in his time spent playing for the Sabres, but he helped his team in far greater ways while he was on the ice.  If the Sabres are evaluating the talent that is within their system, they need to bring up Larsson ASAP, to see if he can continue to help his team remain competitive at the NHL level.  I think the Buffalo Sabres front office will find that Larsson is going to be a key piece in the rebuilding of this franchise.


Tags: Buffalo Sabres Johan Larsson

  • Dano

    VERY good read, Thanks for the Article Rich.

    In the thread of CC’s you mentioned, I put forth my own little equation I use for my assessment of players. It encompasses the +/- but isnt as indepth as the Corsi/Fenwick system.

    The Problem with the Corsi system is a ‘ hole ‘. A team can have a lot of puck possession and be an offensive juggernaut. Their corsi numbers will be high. A team that takes a total of 20 shots in a game but lands 4 in the back of the net with 20mins total possession time will sour in comparison to the same team taking 40 shots, having 3 goals and using the other 40 minutes.. Even though the lower corsi team won the game.

    • Richard Spalding

      Yes – you cannot fall in love with any one statistic, since they all have their shortcomings. Having watched Larsson, though, I remember thinking to myself that both he and Girgensons were often the only two players working hard and keeping the puck in their opponents’ zone plenty of times during Rolston’s reign. I was disappointed to see him go down, but now that the Sabres are starting to shuffle players around, he needs to come back up and spend a lot of time playing, IMO.

      • Dano

        Take your pick : Leino or Johan? Im all for Johan!

        THAT should be a poll! ;)

        • Richard Spalding

          Don’t make me pick!!!!!!!!!!!!!

          • Joe Mazurkiewicz

            What is Larssons f/o%???

          • Richard Spalding


          • Joe Mazurkiewicz

            Another reason he should be playing HERE.

          • Richard Spalding

            Yup. He had really solid f/o numbers against Anaheim and San LA, which I believe were his last two games with the Sabres.

        • Caitlin Campbell

          That could be tomorrow’s poll! Thanks Dano!

  • Jes

    I honestly do not take much stock into the Corsi or Fenwick numbers. It’s just dribble to a certain extent. The Corsi shows that they are good yet on the ice they don’t seem like they are even there for a lot of players. So ya regular stats do it for me. But yes Larsson should be up based purely on his play and style of play. It fits Nolan’s rep.

    • Richard Spalding

      True – you need to factor in other things, and I think looking at player usage charts is more insightful. However, Larsson’s Corsi and Fenwick numbers do illustrate the fact that his reputation as a really solid defensive forward is well-deserved. He would definitely help this team out now, and I think he has a great chance of being on the roster next season.

      • Jes

        Naw I like the old methods. Too much advanced stats going on which IMO didn’t even matter. Look at Luke Adam his Corsi was great yet on most nights he was a non factor so IMO Corsi doesn’t matter.

  • davidmuscalo

    Advanced statistics, to me at least, are paper hockey. Hockey is not played on paper, it is played on the ice. It is a game of played at very high speed without a lot of time for cognitive input. Players use instinct, exceptional eyesight and physical skill to succeed, but the the old bugaboo, randomness, is always rearing its ugly head. When a player scores, is more likely that he scored because another player did something wrong rather than he did something right. Statistics lead to probabilities and probabilities lead to determining odds. It is because of the randomness present in all events that gamblers die broke.

    Larrson is playing in the AHL because his coaches believe he can best develop there to the degree where he can compete successfully in the NHL. The advanced statistics proffered prove that even with all the stats in his favor, he was unable to succeed in the NHL. Yet a player with much worse statistics, Girgensons, is highly successful and would be a welcome addition to any team.

    • Richard Spalding

      I disagree. I remember watching Larsson in the early portion of this season, and found him to be very active and successful at pressuring the puck and disrupting the flow of opposing offenses. I think his stats suggest he WAS able to succeed in the NHL. Let’s not forgot that Ted Nolan doesn’t love to have rookies out there making mistakes for his team; I think the decision to re-assign the rookies was not based on anything other than a “bring in some more proven players” philosophy. Nolan has a rapport with Girgensons; otherwise, I am willing to bet he would still be in Rochester, too.

      • davidmuscalo

        You could see right away that Nolan was not interested in training raw rookies – that is what the minors are for. I agree with Nolan. Also, Nolan realizes that in Girgensons he has an exceptional athlete with the dedication and grit that will help a struggling team win games. He has been right on both counts. Larrson may eventually develop into a solid offensive player and playing in Rochester should help him get there. He didn’t exactly fill the net with pucks or develop a rapport with his line mates when playing in Buffalo and in the final analysis that is what really matters. These things come with time and patience is needed.

        • Richard Spalding

          Larsson is probably never going to be a major scorer. I can’t speak to his rapport with his teammates, and that is definitely something to take into consideration. I don’t necessarily love the mindset of having all your raw rookies spend time in the minors. Then again, I;m not the coach! Overall, though, I think he should have already been given a second look by now, so we’ll see how long it takes for him to be brought back up.

  • Ben Chalker

    Advanced stats are a thing that should be paid attention to, but there are other factors. playing the numbers might work to a point but if a kid needs time in the minors to work on whatever, then regardless of how many percentage points he might make the team better right now, he’s going to make the team even better down the line.
    He’s with the Amerks for a reason, and if, as another article says, the Sabres are in Evaluation mode, then Larsson will get his shot. Likely after the Spengler cup.

  • Pingback: Ville Leino or Johan Larsson?

  • Ludz

    I like Larsson I think he played decent for a rookie and I would welcome him back over Ellis or Porter. As far a Corsi and Fenwick advanced stats go I don’t like when people try to uses these stats on individual players. A winger at the point in his own zone has less control over the other teams shots and rebounds as the defenseman does. These statistics do a better job determining how well a defensive pairing or an entire line plays together then it explains how good an individual player is. are there statistics that looks at what happens directly before and after a player touches the puck? I would rather see stats built on that.

    • Richard Spalding

      I agree that you cannot rely on stats too much; I wouldn’t use Corsi and Fenwick ratings if I had watched Larsson play and thought to myself, “Man, he sucks!” lol However, Corsi and Fenwick are more telling stats than a simple +/-, and are ONE part of the big picture (not all, but definitely a part worth examining). Thanks for the read!!!!