Calgary's Role Changes

Robert Vollman
April 16 2012 09:11AM

 

 

Whose role changed the most from 2010-11 to 2011-12?  You'd probably say Chris Butler, who went from Buffalo's depth blue line to Calgary's top pairing, or perhaps Cory Sarich and Matt Stajan, who went from useful secondary players to the depth lines.

Fortunately, there's a way to objectively measure player's role changes from season to season using the same OZQoC charts we've reviewed every week this season (here's last year's, and this year's final result).  By looking at how the percentage of a player's offensive zone starts has increased or decreased, along with the average level of their competition, we get the insight we need to put both our subjective opinions of their play and their objective statistical results into perspective.

While it's not completely appropriate to compare Quality of Competition from one year to the next like this, especially for those who played on different teams, the chart below can show you how player's assignments changed this year.  Players who saw tougher competition are at the top, and players who saw fewer offensive zone starts are on the right, meaning that players to the top-right had tougher assignments this year, and those on the bottom left were easier.

The bottom left hand quadrant includes players that enjoyed much easier assignments this year, facing both easier competition and starting in the offensive zone more often. Since only two players qualified, and in very limited action (Guillaume Desbiens and Brett Carson), you can conclude that nobody really enjoyed the conditions for a breakout season.

The top left hand quadrant is almost as unpopulated, as it includes players that faced tougher competition, but got to enjoy more starts in the offensive zone.  It includes young T.J. Brodie and new arrivals Blake Comeau and Lee Stempniak – who were used in much more defensive-minded roles in Long Island and Phoenix respectively.

The bottom right hand quadrant is for players who started in their own zone more often – which was the case in 22 of 27 cases, but against weaker competition.  Essentially these players were demoted from higher positions on the depth chart, but were still largely used in same way. 

On defense demotions included Anton Babchuk, Derek Smith, Clay Wilson and especially Cory Sarich.  These aren't necessarily easier assignments, especially for Babchuk and Smith who previously got to play almost exclusively in the offensive zone, but had to play in more balanced situations this season.

Up front Pierre-Luc Leblond and Paul Byron were protected against all but the very lightest of competition, while the fourth line of Tom Kostopoulos, Tim Jackman and Matt Stajan were no longer regularly used against the middle lines – which makes it all the more puzzling why they all had terrible seasons.

The most interesting quadrant is the top-right, which includes players that had to start in the defensive zone more often, and also against tougher competition.

On the blue line we were correct to assume that Chris Butler had the biggest role change, but Mark Giordano also saw tougher competition, which was expected in Robyn Regehr's absence.  Scott Hannan and Jay Bouwmeester were used in fairly comparable fashion as last year, with Bouwmeester having to start in his own zone more often than the year before (although that doesn't entirely excuse his terrible -21 plus/minus).

Up front the OMG line of Olli Jokinen, David Moss and Curtis Glencross, who were historically used against 2nd or 3rd lines, became Sutter's preferred choice against opposing top lines, a move that may not have taken a bite out of their scoring, but left them with big drops in plus/minus. Meanwhile the top tier players Jarome Iginla, Mike Cammalleri and Alex Tanguay saw only slight increases in the difficulty of their assignments, so their poor seasons are more likely a consequence of age and injury. 

Blair Jones and Greg Nemisz both saw the biggest decreases in the percentage of shifts they started in the offensive zone, which fortunately didn't seem to have much of an impact on their results.  More interesting are the cases of Mikael Backlund and Lance Bouma.  Backlund went from carefully sheltered minutes in the offensive zone to top-six minutes, and subsequently went from 25 points in 73 games and +4 to 11 points in 41 games and -13.  Bouma went from 16 games of under 6 minutes of the most sheltered time possible to almost a regular shift, and scored just 3 points in 27 games with a -5.

Final Word

My personal favourite part of the weekly Black Box was the OZQoC chart because it showed at a glance how each of the players were being used. The weekly OZQoC charts were a fantastic way to know what to expect from each player, and to help explain otherwise curious changes in any of their numbers.

Looking at 2011-12's numbers in this way one final time helps put each player's individual performances into context.  As we go through the rest of the year's-end black box statistical summaries in the weeks to come, this chart should help set the right context for interpreting the results. For instance, we'll go easier on player's like Chris Butler, Mikael Backlund, Lance Bouma and the OMG line, while remaining relatively skeptical of any improvements we may accidentally discover in the play of Cory Sarich and the veteran fourth line of Tom Kostopoulos, Tim Jackman and Matt Stajan.

Next week: A look a the even-strength possession numbers. Who helped drive the play in Calgary's favour, and who helped convert shots to chances, and chances to goals?

 

781445e1560a4e1d0d641bc270a7eedc
Rob Vollman of www.HockeyAbstract.com is a regular feature writer on ESPN Insider, co-author of Hockey Prospectus 2010-11 and 2011-12, and regular contributor to NHL Numbers, Flames Nation and Arctic Ice Hockey. Innovator of Player Usage Charts, Quality Starts, GVS (Goals Versus Salary), the Snepsts Projection System, and known for work in League Equivalencies (NHLE). Twitter: @robvollmanNHL
Avatar
#1 Kent Wilson
April 16 2012, 09:15AM
Trash it!
0
trashes
Props
0
props

The 4th liners relatively poor season has a lot of factors. Not only did they start in the O-zone far less, but last year Jackman and Kosto played a lot more often with David Moss and Mikael Backlund, who helped them kill other 4th lines. The best guy they saw this year was Stajan.

They were also some of the ubluckiest players in the league. Tim jackman's on-ice SH% last season was 6.57. This year? 2.72!!

Avatar
#2 Kenta
April 16 2012, 09:32AM
Trash it!
0
trashes
Props
0
props

Yawn. Just when I thought your last few posts contained some interesting analysis you have again relapsed into graphical mumbo jumbo (GmJ in your terminology).

Avatar
#3 Domebeers.com
April 16 2012, 10:37AM
Trash it!
0
trashes
Props
0
props

One of the reasons I didn't like the coach is because I had this nebulous feeling that he didn't put the players in the best position to succeed. Hard to argue, harder to prove such notions,I understand...

However, you're chart is kinda agreeing with me; at the very least, "you can conclude that nobody really enjoyed the conditions for a breakout season.". Right, exactly. And shouldn't the guy managing the ice time try and find opportunities to try and create and then exploit "conditions for a breakout season."?

At the very least, doing that allows you to have stuff to trade away and not miss at deadline day.

(one last thing in this wall of text comment. Butler may have shown he is capable of playing top minutes, but I'm not optimistic the guy is going to turn into a plus/effect the game defender with that minute load. I really hope the club gets some more depth there and pushes him down.)

Avatar
#4 mattyc
April 16 2012, 12:12PM
Trash it!
0
trashes
Props
0
props

@Kent Wilson

Kent, When you tabulate the scoring chance data, do you record who generated (i.e shot) the scoring chance, or just who was on the ice?

I ask because I wonder (and suspect) that some players convert a larger percentage of their shots into scoring chances, and into goals as a result.

I'm looking here at Jackman, who managed to score once on 103 shots, and Comeau, who scored 5 times on 137 shots.

Comeau has the longest average shot distance on the team (followed by Backlund).

Jackman's most common linemates have much higher SH% (Stajan 10%, Kostopoulos 4.5%, Horak 4.5%), which makes me think he's the weak link shooting-wise on the 4th line.

I wonder if it is 'luck' or if there is some systematic flaw in the way these two players in particular generate chances.

Avatar
#5 Kent Wilson
April 16 2012, 12:36PM
Trash it!
0
trashes
Props
0
props

@mattyc

At the edges of the league (the very best and very worst), there's a real difference in on-ice SH% due to talent level. The best guys (who usually also play with other good players) tend to outperform league average SH% over the long run. Really bad offensive players (goons, replacement level 4th liners) will probably marginally underperform average SH%...

That said, I doubt any full time NHLer is actually bad enough to sustain a 2-3% oni-ce SH% over the long run. And there's random variations around the mean in short samples, so it's often hard to separate true skill level from the noise.

Avatar
#6 mattyc
April 16 2012, 12:57PM
Trash it!
0
trashes
Props
0
props

@Kent Wilson

I guess I'm just wondering if on ice SH% is really the stat to look at for measuring a player. Sure I think it has value, and in particular I think it does do a good job of explaining Jackman's brutal -21 and his low point total, but it doesn't really prove he isn't the weak link (shooting wise) for that 4th line.

I'd imagine his on ice SH% will probably regress and I doubt his basic counting stats look as brutal again next year (it seems unlikely he shoots below 1% - does anyone?!).

Having said that though, 100 shots is not an unreasonably small sample size, there has to be some underlying reason(s) why Jackman and Comeau can shoot 100s of "blanks" and Stempniak (who played with pretty similar line mates to Comeau and had comparable shot totals) shoots 10%.

By the way, for context, what is the league average SH% and on ice SH%, and is there somewhere someone has graphed them (I'm assuming it's a normal distribution?)?

Avatar
#7 Kent Wilson
April 16 2012, 03:27PM
Trash it!
0
trashes
Props
0
props

@mattyc

Actually, 100 shots is a pretty miniscule sample size in the NHL. It takes several thousand to get a good feel for an individual players true skill level.

NHL average SH% at ES (5on5) is about 8% while corresponding mean SV% is about .920, with small differences year-to-year.

By-and-large, a good rule of thumb is to treat on-ice SH% as mostly randomness, particularly in the space of a single season. It's really hard to detect the true talent signal in all that noise.

Comments are closed for this article.