Assessing The Oilers Defence Corps

Jonathan Willis
March 05 2010 01:01PM

Edmonton Oilers v Ottawa Senators

With massive changes to the Oilers’ blue line personnel, it seemed like an appropriate time to reflect on the job they’ve done so far this season.

Personally, I like scoring chances as a metric, because it’s a larger sample than goals for and against and not as dependent on goaltender quality on either side. Unfortunately, it’s hard to break down scoring chances by defence pairings, so that won’t work for what I have in mind.

On the other hand, it turns out that Corsi numbers mirror scoring chances almost exactly. Kent Wilson of Flames Nation wrote a great article a while back on his other site explaining both what Corsi is and why it is so valuable.

The chart below uses a Corsi ratio. The number in the box indicates how many shots (including blocked and missed shots) the opposition manages at even-strength for every 100 Oilers’ shots. A low number means that a certain defence pairing has out-shot the opposition, while a high number means they’ve been out-shot by the opposition.

Take Sheldon Souray and Tom Gilbert as an example. Their combined number (88) means that for every 100 shots the Oilers have managed with them on the ice, the opposition has only managed 88 shots, which is extremely good. All combinations together for more than 100 total shots are listed below.

  Souray Visnovsky Gilbert Smid Grebeshkov Staios Strudwick Chorney Total (Ratio)
Sheldon Souray     88     126     96
Lubomir Visnovsky       100 89   105 116 97
Tom Gilbert 88       124   129 172 109
Ladislav Smid   100       148     117
Denis Grebeshkov   89 124     116 182   122
Steve Staios 126     148 116   128   126
Jason Strudwick   105 129   182 128   179 151
Taylor Chorney   116 172       179   152

(thanks to Vic Ferrari's timeonice.com for the numbers on the chart)

Looking at that chart, I see three distinct groups.

At the top end of the scale are Sheldon Souray and Lubomir Visnovsky, the Oilers most expensive and best defencemen. Souray’s now finished for the season with a hand infection and Visnovsky will be plying his trade in Anaheim, so the dive for five got a substantial boost in the last few days. I rate Visnovsky ahead of Souray because even though his total number is a hair lower, he’s also been condemned to spend time with both Chorney and Strudwick – more on those two in a moment. Meanwhile, Souray has been excellent with Tom Gilbert and poor with Steve Staios, indicating that while he’s a good defenceman who can help he can’t single-handedly carry a pairing.

In the middle are four players of varying ability. Based on what they’ve done with different partners, right now I’d rate them in order as Gilbert, Grebeshkov, Smid, Staios. Gilbert’s had a bad season but has matched well with Souray; I imagine we’ll see his stock plummet as he lines up beside inferior partners for the rest of the season. Grebeshkov’s performance has been quite good when paired with Visnovsky, poor with Gilbert and Staios, and utterly atrocious with Strudwick. Smid has had the benefit of mostly playing with Visnovsky, where he’s looked okay (but not great in comparison to Visnovsky’s other partners) and with Staios, where he’s looked lost. Finally, Steve Staios has fallen into an elevator shaft when on the ice with anyone other than Grebeshkov, with whom he has only looked moderately bad.

At the bottom end of the spectrum are a pair of matching boat anchors: Jason Strudwick and Taylor Chorney. Chorney at least has youth on his side; he isn’t an NHL defenceman at this point and I highly doubt he’ll be one next year, but it’s conceivable he could someday turn into a power play specialist who can survive on the third pairing. Right now, every player who plays with him sees his numbers drop off sharply.

Unlike Chorney, Strudwick has age as an enemy; he’s in decline at this stage of his career and should not get another NHL contract. He’s a millstone around the neck of any player he partners with, and the fact that his opponents manage 151 shots for every 100 the Oilers take is simply inexcusable, especially given that the coaching staff has taken some pains to keep him away from good opponents.

This situation bodes well in the Oilers continued quest to be the worst team in the National Hockey League: both of their best defenceman are finished for the year, the surviving members of the top-six (Gilbert and Smid) have lost their best partners and have been poor without those players (and Smid may be done anyway), and two of the worst defenceman in the NHL today will get regular minutes for the rest of the year.

74b7cedc5d8bfbe88cf071309e98d2c3
Jonathan Willis is a freelance writer. He currently works for Oilers Nation, Sportsnet, the Edmonton Journal and Bleacher Report. He's co-written three books and worked for myriad websites, including Grantland, ESPN, The Score, and Hockey Prospectus. He was previously the founder and managing editor of Copper & Blue.
Avatar
#1 David S
March 05 2010, 02:01PM
Trash it!
0
trashes
Cheers
1
cheers

Jonathan - Is it possible to re-do that chart with the players we currently have to work with? Especially now with Souray, Smid, Visnovsky and Grebs out of the picture. Or is this something your anger management therapist would discourage you from doing?

Avatar
#2 quicksilver ballet
March 05 2010, 01:07PM
Trash it!
0
trashes
Cheers
0
cheers

Is there anything left? Payrolls being cut tickets are going up....whats not to love.

No sense making light of a revolving door for the next couple years, it's please shut up and pay for your tickets time.

Avatar
#3 Twiggs
March 05 2010, 01:07PM
Trash it!
0
trashes
Cheers
0
cheers

Interesting analysis on a baaaaad situation

Avatar
#4 Matt Henderson
March 05 2010, 01:21PM
Trash it!
0
trashes
Cheers
0
cheers

But Whitney will make it better, right?.....RIGHT?

*sobs uncontrollably into his Igor Ulanov signed Jersey*

Avatar
#5 The Towel Boy
March 05 2010, 01:26PM
Trash it!
0
trashes
Cheers
0
cheers

Looks like Strudwick definitely has a way of making the guys around him better.

*realizes high number is bad*

Oh. Jesus!

Avatar
#6 David Staples
March 05 2010, 01:27PM
Trash it!
0
trashes
Cheers
0
cheers

Interesting chart, and it certainly shows up Strudwick and Chorney.

Bad players drive team plus minus numbers into the gutter for all those on the ice wit them.

If one factored in quality of shot, I suspect the Viz and Souray would be ranked lower, though.

That's just my strong hunch, though, and I have no numbers to prove this point.

I don't measure quality of scoring chances -- or each individual player's responsibility for scoring chances, either -- and don't know of anyone else who does either.

Avatar
#7 Death Metal Nightmare
March 05 2010, 01:28PM
Trash it!
0
trashes
Cheers
0
cheers

how are you getting the TIMEONICE.com stuff to work? the script version? if so, how?

otherwise it looks like that sites script is old and it keeps brining up games for last year based on this years game numbers

Avatar
#8 VMR
March 05 2010, 01:43PM
Trash it!
0
trashes
Cheers
0
cheers

@David Staples

I'm with you on the quality of chances factor. On it's own this doesnt tell us a lot, if you factor in the plus minus at even strength it might tell you something. Sure Staios's numbers look pretty bad when you consider the number of chances given up but what if the other team isnt scoring on those chances? Souray and Vis were giving up the fewest chances but what if there were a larger portion of goals scored on those chances?

Avatar
#10 Death Metal Nightmare
March 05 2010, 01:49PM
Trash it!
0
trashes
Cheers
0
cheers

sweet, thanks JW

Avatar
#13 dawgbone
March 05 2010, 02:03PM
Trash it!
0
trashes
Cheers
0
cheers

"Out of curiosity, is there any reason to believe that there's a significant difference in quality of chance given up?"

Of course there isn't... though it has been used to prop up some pretty insane arguments in the past i.e. Brodeur didn't face many shots, but the ones he did were high quality or the plethora of defences given to Khabibulin after he signed with Edmonton.

And Smid is out for the season as well now, giving the Oilers Tom Gilbert, Ray Whitney and not much else. 30th should just about be a lock now, as you have to figue Giguere has at least a couple of steals left in him.

Avatar
#14 VMR
March 05 2010, 02:04PM
Trash it!
0
trashes
Cheers
0
cheers

I checked out the bit on Corsi numbers and I see a huge problem with trying to judge the defencemen on your team by it. Offensive defencemen will almost alway be better than defensive defencemen.

If Souray and Vishnovsky come out on the ice when the Oilers get a faceoff in the offensive zone they're much more likely to be on for a shot for. Conversely if Staios and Strudwick are sent out in the defensive zone they're much more likely to be on for a shot against.

The stats look interesting but in the end they're pretty easily explained.

Avatar
#15 dawgbone
March 05 2010, 02:04PM
Trash it!
0
trashes
Cheers
0
cheers

Staples, that's just cruel... why not ask him to walk barefoot across a football field made of broken glass?

Avatar
#16 Ducey
March 05 2010, 02:14PM
Trash it!
0
trashes
Cheers
0
cheers

Willis,

Wouldn't it be more likely that Lubo and Souray would get paired with a better forward group than a third pairing who might be stuck with moops like Stortini.

I don't understand how your study would be better than the Desjardains Behind the Net 5 x 5 numbers:

Plante 2.86 Arsene 1.96 Smid 1.66 Lubo .88 Gilbert .52 Struds -0.44 Souray -0.98 Staios -1.27 Chorney -1.72 Theo "Bend it like" Peckham -3.02

This tells us Struds isn't nearly as bad as you think when quality of teammates and opposition are taken into account. Doesn't it?

Avatar
#17 dawgbone
March 05 2010, 02:17PM
Trash it!
0
trashes
Cheers
0
cheers

"I checked out the bit on Corsi numbers and I see a huge problem with trying to judge the defencemen on your team by it. Offensive defencemen will almost alway be better than defensive defencemen.

If Souray and Vishnovsky come out on the ice when the Oilers get a faceoff in the offensive zone they're much more likely to be on for a shot for. Conversely if Staios and Strudwick are sent out in the defensive zone they're much more likely to be on for a shot against.

The stats look interesting but in the end they're pretty easily explained."

Not really Bruce.

Offensive Dmen may have more on the corsi for side, but at the same token they'd give up more on the corsi agaisnt side.

Your point about faceoffs is valid though and does have an impact, though not necessarily with this group here.

Souray 202 OZD - 182 DZD (+20) Gilbert 312 OZD - 299 DZD (+13) Staios 179 OZD - 175 DZD (+4) Grebeshkov 211 OZD - 210 DZD (+1) Strudwick 181 OZD - 207 DZD (-26) Smid 200 OZD - 227 DZD (-27) Visnovsky 225 OZD - 260 DZD (-35) Chorney 57 OZD - 104 DZD (-47)

The number in () represents whether they had more Off zone draws (positive number) or Def zone draws (negative number).

Visnovsky started out in the bad end more than the good end, but blew Strudwick out of the water. In reality, if you are within +/-30 afte this many games you are close enough to an even split for arguments sake. The only big outlier is Chorney, but I'd love to see the number of icings he's been on the ice for, I suspect it would be rather high.

Avatar
#18 Dan the Man
March 05 2010, 02:22PM
Trash it!
0
trashes
Cheers
0
cheers

Interesting stats but the only real surprise for me was that I would have thought Smid would have done a bit better.

Avatar
#19 OB1 Team Yakopov - F.S.T.N.F
March 05 2010, 02:23PM
Trash it!
0
trashes
Cheers
0
cheers

I'd like to see Whitney's #'s... who wants to bet they look just like Gilberts? ie pretty good when he's with a high end Dman and pretty poopy when he's with an average or worse dman.

Avatar
#21 JeffG
March 05 2010, 02:31PM
Trash it!
0
trashes
Cheers
0
cheers

and now even Smid is gone for the yr. (Neck surgery)

Avatar
#24 DBO
March 05 2010, 02:56PM
Trash it!
0
trashes
Cheers
0
cheers

Nice post JW. Going into next year if the d stays the same then i would expect you would see: Souray-Gilbert Smid-Whitney Johnson/Vet-Chorney/Peckham

Avatar
#25 Ender
March 05 2010, 02:58PM
Trash it!
0
trashes
Cheers
0
cheers

While everyone needs to understand that stats don't tell the whole story (~where have we heard that before?~) these numbers are resonably good indicators that prove what most of us already knew; if you were doing an office pool you'd pick Visnovsky ahead of Strudwick. I'm the kind of guy that likes to see it quantified like this, but I imagine many others will find a reason to pick it apart. Props, JW; I like the thought that went into this.

Avatar
#26 The Real Scuba Steve
March 05 2010, 03:01PM
Trash it!
0
trashes
Cheers
0
cheers

@ Johnathan

It is said all good teams need a puck moving defenceman, do the Oilers go by still having skilled defense or go to having more of a stay at home defense core?

Avatar
#28 VMR
March 05 2010, 03:14PM
Trash it!
0
trashes
Cheers
0
cheers

@dawgbone

I dont understand the bit about offensive Dmen having more on the against side, how does that make sense?

The other problem I see with these numbers is that it is just a ratio not actual numbers, correct? Doesnt that mean that there could be a wide skew if say Grebs and Struds only had a dozen shifts together. Fewer chances have a larger impact on the number. So if they end up playing together 3 teams over the course of the season and end up giving up 5 chances against and only 3 for you get a number thats going to look much worse than if they had played together 30 times and given up 2 more chances a game then they got.

Avatar
#30 Lofty
March 05 2010, 03:43PM
Trash it!
0
trashes
Cheers
0
cheers
David S wrote:

Jonathan - Is it possible to re-do that chart with the players we currently have to work with? Especially now with Souray, Smid, Visnovsky and Grebs out of the picture. Or is this something your anger management therapist would discourage you from doing?

I bet Whitney's numbers would make him look like the second coming of Paul Coffee. Playing on any other team than the Oilers this season and being put on that list will make you look like a stud.

Avatar
#31 Petr's Jofa
March 05 2010, 03:49PM
Trash it!
0
trashes
Cheers
0
cheers

Love the Bre-X comment and and I'm now waiting to see which scapegoat gets thrown from the Oiler's helicopter.

Avatar
#32 ripromer
March 05 2010, 04:17PM
Trash it!
0
trashes
Cheers
0
cheers

@dawgbone

The faceoffs per zone do not distinquish between full strength and PP. The Corsi numbers apply only to full strength.

Avatar
#33 ripromer
March 05 2010, 04:23PM
Trash it!
0
trashes
Cheers
0
cheers

Corsi analysis is the best relative assessment I have ever seen. Given the uniform save percentage of goalies around the league of approximately .92 for most quality goalies it means one additional goal for each Corsi increase of 10 shots. If Souray is at 96 and Strudwick is at 146 thats about 5 more goals against Strudwick vs. Souray for every 100 shots by the Oilers. YIKES

Avatar
#34 Rip Torn
March 05 2010, 11:54PM
Trash it!
0
trashes
Cheers
0
cheers

Off Topic maybe, (defence)

Anybody see the U of A Bears Hockey goalie hilites of tonite? 3-4 solid blocker jaw punches to the guy that ran him!!! And played awesome.

/mebbe we have a goalie right here?

Comments are closed for this article.