Andrej Sekera, No. 1 Defenceman


The Edmonton Oilers have taken a defence-by-committee
approach this season. Oscar Klefbom and Kris Russell both play a lot, just
under 22 minutes per game. Adam Larsson’s numbers are skewed a little by not
taking on any kind of offensive role, but he too is over the 20-minute mark on

At the top of the depth chart is Andrej Sekera. The best
defenceman to enter unrestricted free agency in the summer of 2015, Sekera has
emerged as an all-situations workhorse for the Oilers and their top overall blueliner.

Advertisement - Continue Reading Below

There isn’t a whole lot of question as to who is driving
Sekera’s results at 5-on-5. It’s Sekera.


Sekera’s primary defensive partner this season has been
Russell. Together, the two have generally played tough opponents, having
displaced the Klefbom/Larsson pairing as the primaries in that role, though
there isn’t a lot of gap. Sekera/Russell have also been given a steady diet of
defensive zone starts.

Together, they have a 46% Corsi rating, which isn’t great.
But when they’re on the ice, the Oilers allow just over a single goal per hour
at 5-on-5, which is a tremendous number.

Advertisement - Continue Reading Below

Away from Russell, Sekera has mostly played with rookie Matt
Benning. With that pairing on the ice (in softer minutes, it must be said) the
Oilers take 58% of the Corsis and 59% of the goals. When Russell’s on the ice
away from Sekera, the Oilers take 47% of the Corsis and 45% of the goals.

I’m skeptical as to whether the good results that
Sekera/Russell have put up together can be sustained. My old Journal colleague David Staples has
made the argument
that Russell (and by extension this pairing) is so
airtight in the defensive zone that they can outperform their shot metrics over
the long haul, and doubtless some readers will find that argument convincing.

Sek Rus chart

For my part, I don’t think it’s a surprise that both Sekera
and Russell have on-ice goal totals exactly in-line with their Corsi totals when
separated from each other. Goals follow shots for the vast majority of players
the vast majority of the time.

So far, though, that pairing is working. And given the
circumstances (tough minutes, a left-shot partner on the right side) it’s difficult
to be too critical of its failings driving possession. It would be particularly foolish to be overly critical of Sekera, who has found a way to keep his head above
water with Russell, Benning and with Mark Fayne last year.

Sekera is also having a strong year on both special teams.
Edmonton’s penalty kill is about average league-wide, but both shots and goals
against improve when Sekera is out there. The “Shinpad Assassin” is somewhat
limited as a No. 1 power play defenceman, but he’s been good enough to hang on
to a top unit job on a revitalized man advantage clicking in on north of 20% of
its opportunities.

Advertisement - Continue Reading Below

PK subban

Eventually, one would hope that Sekera is asked to do a
little less of the heavy lifting. One of the big arguments for pursuing P.K.
Subban this past summer was the idea that Edmonton was going to need a marquee
talent to lead the blue line, and right now that player isn’t on the roster.

Klefbom’s defensive zone play is still a concern. Larsson’s
play with the puck limits him. Russell’s already overachieving as-is. That
leaves Sekera in the No. 1 role by default, the jack-of-all-trades who can be
used by the coach in any situation, against any opponent and with any kind of
partner with at least some expectation of success.

Those are not easy players to find. Peter Chiarelli and his
staff deserve credit for identifying Sekera in the summer of 2015 and paying
what was necessary to get him to come to a rebuilding Oilers team. Sekera in
turn deserves credit for so ably handling an incredibly difficult assignment.
He’s been a critical and often under-recognized component of Edmonton’s turnaround
this season.  

  • Been there

    I admit after last year Sekera concerned me. I am still not a huge fan, but he has played way above my expectations. I guess he took longer to adapt to Edmonton then I would have guessed. As for Klefbom I think he has done quite well for playing his first full season and the situations they throw him into. His offence is coming together, and he plays a good shutdown when required. On the whole I like our defense, they have only improved slowly but steady since November. Still missing the offfensive right hand shot from the point, but as a group I am happy with them. We have depth, something that could not be said last ten years!

  • 3 Little Birds

    “Together, they have a 46% Corsi rating, which isn’t great. But when they’re on the ice, the Oilers allow just over a single goal per hour at 5-on-5, which is a tremendous number.

    Away from Russell, Sekera has mostly played with rookie Matt Benning. With that pairing on the ice (in softer minutes, it must be said) the Oilers take 58% of the Corsis and 59% of the goals. When Russell’s on the ice away from Sekera, the Oilers take 47% of the Corsis and 45% of the goals.”

    May I ask what is the goals allow per60 5 on 5 for Sekera and Benning?

      • 3 Little Birds

        Thank you for the info.

        This opinion may not stack up against the numbers, but I would lean towards lower goals against, as that is the definition of defence. With McDraiRoon I will take our chances in 1 goal games.

    • ¯\_(ツ)_/¯

      In a 5V5 world, sure. But given that Larsson doesn’t play the PP and is a risk at 3 on 3 with his speed, it’s pretty much a toss up for me. Every third game you could add Russell into the argument, the way he’s played this year.

      The nice thing is that they’re all playing well enough to make it hard to say who’s the best of the three. And with another 100 games under their belts I think we might add Nurse and Klefbom to the question, which is a good problem to have. Although I do think that Larsson has a lot more upside while Sekera is probably playing close to his potential so we may talking about this comparison in 2 years.

      The one problem I have with Sekera is that when he’s leaving the zone he seems to make the slowest pass he can every 3rd or 4th pass. It’s like he’s attempting to see how little draw weight he can put on a puck and still get it to the winger 75 feet away. Anyone else notice this? Drives me crazy when it gets intercepted.

        • ¯\_(ツ)_/¯

          I totally agree with you on anything in the Oilers’ end: Larsson all the way.

          In the opposition end, I’d go with Sekera or Klef. 3-on-3, Sekera.

          Pronger, both ends of the ice, 40 minutes a game, please.

  • Masma

    Thought this was about Sekera performing well, but every time you get a chance to knock on Russell you take it. Oilers are fighting for top spot in their division with Russell carrying his own weight on this team. Every time Oilers or players are doing well you come with corsi.

    Let me ask you this Willis , according to corsi in what position should the Oilers be in the league?

    • It’s hard to talk about Sekera’s performance without also including his defence partner, since that tandem as a unit is dramatically outperforming its shot metrics right now.

      As for your question, Edmonton is presently 9th overall in the NHL standings by points percentage.

      By all-situations Corsi, they’re 11th. By my preferred metric – score-adjusted 5v5 Fenwick – they’re 7th.

  • Makaveli

    I’m so sick of “advanced stats”.

    How do advanced stats take into consideration, lost stick on the ice in the D Zone that prevents the puck from clearing. Or how about a bad bounce off the boards. What if your D partner is playing hurt? How about ice conditions? Afternoon game vs late game? Fluky goals? there are so many changing variables within a game and from game to game that I cannot buy into advanced stats. They seem to work in baseball where you have less random events during a game, but even then baseball has its random variables.

    Case in point… look at every game wrap up on this site for the last few years. Prior to this season…. the oilers would loose but we woul “win the corsi”. Now this year, we are winning games but “loosing” the corsi… just shows you how great of an advanced stat Corsi is.

    • The Dave

      I view advanced stats as a mixed bag, but some of these rhetorical questions are easy and you are missing the most important point, so…

      “How do advanced stats take into account [weird stuff]?” By counting them indirectly. Let’s take an event that could influence a game: a ‘bad’ bounce off the boards. If it led to a goal against or a shot against then it would lead to a negative Corsi event. If the other team doesn’t get a shot off of it then who really cares?

      Seriously – who cares? A lot of people will use lines like “Corsi can’t measure heart and grit” and the answer is, it will if it helps the team get a shot, or prevents the other team from getting a shot, and we don’t really care about anything else. If a player’s heart and grit helps them attack the other team and stop the other team from attacking their own net, then yeah – it shows up as a Corsi number. If a player has lots of heart and grit but is still terrible then they’ll still have a bad Corsi.

      I’m also shocked you think Afternoon Game vs Late Game is a critical stat. It’s not like the other team isn’t playing the game at the same time of day. It might make a small difference, but I’m not sure looking at a team’s afternoon game record is going to have more predictive power than Corsi or their place in the standings.

      Here’s the big misconception about Corsi and why it matters: it is really a defensive stat, not an offensive one. Good Corsi numbers don’t correlate as highly with offense as the do with defense – in other words there are lots of ways to score goals (e.g. a breakaway after being hemmed into your zone… bad for Corsi, good for goals), but when it comes to preventing goals against it helps to just shelling the other team. It makes sense too – Corsi weirdly equates a missed shot with a shot on goal, even though that missed shot will never lead to a goal. But a missed shot probably means the puck isn’t in our end. So… a Corsi is good for tracking how well you play keep away, and that is useful for guessing how well you are defending.

      As to the issue of us winning the games but losing the Corsi – the Oilers are currently the 10th best team in the league in 5v5 Fenwick % (which is like Corsi but a bit more accurate). If you adjust for scoring effects we move up to 7th. That is also roughly where we are in the standings. We don’t win the Corsi every time we win the game, but that’s not the point – the point is that the two rankings are pretty close.

      The problem with advanced stats isn’t that they don’t summarize things well, because they do. The problem is that they aren’t really all that advanced (adding Shots + Missed Shots is somehow ‘advanced’… the average 7 year old can do this), they are often misunderstood (like thinking Corsi can overcome lack of scoring chances, a hot goalie, or crappy players), and they don’t meaningfully tell players what they should be doing on the ice.

      Corsi isn’t the end all and be all of hockey, but it’s useful for gauging part of how the team is playing. Just like shooting % is a good way of gauging how hot or cold a player is (even though it doesn’t tell the whole story), and how Save Percentage gives us some insight on a goalie. They aren’t perfect stats, but they help.

      • S cottV

        I’m sure advanced stats will continue to evolve and eventually they will nail down more useful formulas.

        In the meantime, useful but lets not get too carried away with concrete conclusions.

        I’m not sure why possession is monitored largely by shots and or shot attempts.

        If you want to measure possession, take a stop watch and break actual possession by team and by zone.

        If you gave me actual possession stats and shots / shot attempts – separately, I would probably find that more useful toward coaching a hockey club.

        As for shots and shot attempts – I would strip out low value ones. Perimeter shots with no net front presence and other useless shots – delete them. In my opinion.

        • The Dave

          Possession is measured largely by shot attempts because it is easier than tracking the ice the way you suggested and is allegedly highly correlated. I’m not so sure that is true though.

          Re: shots and shot attempts and dropping the bad ones – it wouldn’t have anything to do with possession stats, but I agree. Lots of shots have virtually no chance of scoring and are roughly on par with a dump-in. With Corsi they all count the same though.

          Incidentally I ran the numbers for this year to see how correlated Corsi was with offense… GF/60 and CF/60 are at 7% for this season. That the shot attempts could be so poorly correlated with goals makes it almost a useless stat to me. It might be just me though – apparently Corsi lovers and haters alike didn’t care for what I said in my last post.

  • Oiler Al

    Would Russells numbers improve if he was playing as a lefty.
    Is Osterle good enough to be on the roster, as No.7 used on the power play only ? Could give Sekera a bit of spell.?

  • I see similarities with Giordano. Now the offense isn’t as good, but we should also remember Gio took a hell of a long time to become the consistent 40+ point D man, and all around number one defender. Is he elite like Doughty? No. Is he a number on D man? Yes.

    Having noted the pairing with Russel, I still would not hate the idea of them bringing in Hamonic if that is an option without overpaying. I really believe the Power Play right shot quarterback D man is non existent in the NHL right now. As in, not available. Or at least not available for any reasonable price.

    I don’t mind the defense by committee approach. I think Tampa won that way in 2004, and Carolina in 2006. It is possible.

    • Randaman

      With all due respect, Gio is not a bonafied #1. He had one spectacular year and that’s it. Like the rest of his team, he has seriously failed to make the grade this year.

      • 43 pts in 82 games in 2010, 47 and 48 points in 64 n 61 games in 2014 and again in 2015. 56 pts in 82 games last year. Overall 323 pts in 644 career games.

        Not just one lucky year. Currently he’s on pace for his usual 40 + points.

        I agree with you the whole team has under-performed, including Gio this year, but he is a legitimate number one D man, and I would say a little underrated. His leadership and do it all style play was tantamount to any success the team had two years ago.

        I think this year Galutzen is having a tough time getting the players to play the roles he needs for success. Right now the Backlund line has really become the top line. Monahan has been bad. Gadreau was injured. And they turned Brodie into a stay at home shut down D man.

        So… as an Oiler fan I am familiar with how a coach and his systems can destroy players and the overall production of a team.

  • S cottV

    Corsi is skewed to the more typical d man that is built like a wolf.

    Sekera and Russell are more like foxes. They have to take a longer and more measured approach toward achieving the kill.

    The wolf sizes up attacks or cycles or puck possessions – and tends to say screw this – using commanding size / strength and agility to bust things up – sooner rather than later. In their haste or over confidence, they sometimes give up the odd glaring chance – but most times its shut down quickly with the puck going the other way. Not much zone time is given up to allow quality corsi shots and or outside low value shots designed to pad the corsi.

    The foxes know their strengths and weaknesses. They know they cant pull a screw this and shut things down immediately. They require more patience, angling, good stick work, positioning, team work, timing, wait for the right opportunity to strike – etc etc. The last thing they want to do is get drawn into an unnecessary physical tussle, where they run the risk of being thrown around and out of position. Russell in particular, avoids boxing out net front threats by fronting them. He turns to the incoming shot and tries to block it, which is why he has so many blocks.

    Done properly – it can be effective. Not too many glaring chances (re Staples) and a lot of outside shots that look worse than is reality re the Corsi.

    If I had my choice – I’d have more wolves than foxes, but it can be done in a different way.

    Corsi and Fenwick don’t capture different ways, all that well.

  • OilCan2

    Sekera has been gold since he got here. It’s understated because he lacks the big, shiney numbers or gaudy point totals. As a team we are far better with him.

  • Big Jacks Meat

    I Hate the Wild. Need to start road trip off with a win. I wish this was a road game actually.

    DO IT OILERS, The next 10 will give us a good picture.

  • Makaveli wrote:

    Case in point… look at every game wrap up on this site for the last few years. Prior to this season…. the oilers would loose but we woul “win the corsi”. Now this year, we are winning games but “loosing” the corsi… just shows you how great of an advanced stat Corsi is.

    That’s simply not true.

    • 2009-10: 44.2% score-adjusted Corsi (30th, NHL)
    • 2010-11: 45.2% (28th)
    • 2011-12: 46.8% (28th)
    • 2012-13: 44.4% (30th)
    • 2013-14: 43.7% (28th)
    • 2014-15: 47.3% (23rd)
    • 2015-16: 48.1% (21st)

    The Edmonton Oilers have been a miserable team by every advanced metric for the entirety of their rebuild.

    I had to listen to optimistic fans complaining “why don’t you ever say anything nice about the Oilers” for the last seven seasons. I’m certainly not going to sit quietly by while you pretend the stupid analytics guys loved the team all this time.

    • Makaveli

      Touche Willis!

      I guess I would have to go back through all the wrap ups and note everytime Baggedmilk or Jeanshorts mentioned winning the corsi after losing a game… or maybe they were just trolling us.

      Whether you use the eye test or fancy stats, a crap hockey team is still a crap hockey team.

      Kris Russell is the reason I’ve really began to question advanced stats. Since becoming an Oiler, I’ve seen him make more good hockey plays than bad. And with the way McLellan utilizes him, I would think he agrees and probably doesn’t pay too much attention to advanced stats either. However, Matt Henderson says the Oilers should replace him ASAP which I think would be nuts (I also think signing him longer than 2 years would be nuts too due to age and other moving parts).

      • Makaveli

        50.2% (17th in NHL)

        Minny is 49.7% yet they are first in the West.

        Boston is first in the league for Corsi with 54.6% yet they are 14th in league standings…

        so ya Corsi… great stat (giant fart sound)

        *source =

    • YEGswede

      And in those miserable years Corsi made sense, it captured what I was seeing. It made me believe that corsi (or shot metrics in general) actually was a reasonable proxy for possession. I guess Eakins did his best to show me how it is flawed, but in general it made sense. This year though, being less one-dimensional, has highlighted that something more than counting shots is needed for “advanced stats” to be meaningful.

  • toprightcorner

    The Oilers may not have a #1 dman but what has made there defense so solid is that have three #2 dmen in Sekera, Klefbom and Larsson. I think tree #2’s and a #4 in Russell is fairly close to the traditional #1,#2,#3 and #4.

    If Edmonton can sign a #4 dman with a right shot and some offense, that would be a big step of improvement. I think Klefbom or Larsson could develop into an avg #1 dman in a year or two.

  • Jordan88

    The only metric I really care about is goal differential and the Oilers are doing exceptionally well in that compared to last year.

    Even against teams like Phoenix and New Jersey, the most boring teams to watch in the NHL the Oilers were able to muster an effective offense.

    All I can say, is this. I remember during Eakins the Advanced stats for us showed a good team. The on ice product was terrible. The reason for that I will argue is playing the players so they look good on paper but not to the best of their abilities.

    Case in point. Yakupov being a 3rd liner… A 3rd liner. The most offensively gifted shooter to come out of Sarnia since Stamko’s and his line mates are essentially AHL call ups. It’s a little hard to tread water when you have 2 boat anchors on your ankles.

    McLellan is honestly a great coach amazing even. I think he is not getting nearly the credit he deserves. Chiarelli gave him the pieces but McLellan plays those pieces to their strengths.

    • S cottV

      I realize that Corsi and Fenwick are useful numbers to consider, as part of the process of evaluating players and teams.

      Just don’t think the measurement captures all that some want to crack it up to be.

      Really don’t like the shots and shot attempts connection to possession. I get that junk shots are balanced somewhat by turnovers and counter attack shots. Still – to me, there will eventually be a better way to track possession. Like software that easily clocks possession by player and team in all zones and tallied.

      Take real possession numbers and blend with other criteria like shots, scoring chances, high grade scoring chances etc and maybe we something even more meaningful.

      I’m big on possession and I wish we would stop squandering it so much. Sure – improved by Corsi measurement and by eye over the one and done wonder years. However McL’s apparent (according to Drew) desire to mass shoot at net, has to be inflating the Corsi – somewhat and I think to the detriment of improved actual o zone possession and the possibility of subsequent high grade scoring chances.

      We may be at 51.5 but will never get to 55, the way it’s going.

  • Anton CP

    Couple of years ago when he was with Canes that he helped Faulk looked like elite offensive dman and ever since he left that Faulk is a minus generator. Sekera is underrated top pairing dman.

  • Spoils

    not a lot of people know this, but Klefbom is a WJC All Star. Only 2 of those each year, which is to say Klefbom was one of the 2 best DMen in the world for his age, at one point. others – Karlsson, Hedman, doughty etc.

    he is 23. D typically come online around this age…

    I believe Klefbom has another gear and I bet he hits it next year.