Navigation: Jump to content areas:


Pro Quality. Fan Perspective.
Login-facebook
Around SBN: Yu Darvish Diagnosed With Mariners Fever

Getting Defensive: Toronto Maple Leafs

"Getting Defensive" is a weekly look at the defense corps of the 30 NHL teams using the Time On-Ice (TOI) data from the 2010-11 regular season.

Having spent a lot of time thinking about the kinds of information that could help us better understand defensemen, I noticed that while forwards can be pretty well expressed by their capabilities on offense, defensemen that receive powerplay minutes almost universally outscore their fellow defensemen by a considerable margin.  This can muddy the picture of what a "good defensemen" might look like.  Tom Awad has suggested that, in terms of skaters in general, a loose but helpful definition of "good players" might be best guided by playing time, and that good players for the most part will be recognized by the coaches and given the most time.  It's an excellent read, by the way, worth your time if you haven't looked at it.

In the spirit of Awad's article, I wanted to review defense corps across the league, showing the increases and decreases in playing time and the effects of trades, injury, etc.  I chose the Toronto Maple Leafs for the first instalment because they give us a good example of a team that has experienced significant upheaval, losing two major minutes players (Francois Beauchemin and Tomas Kaberle), and seeing the increased playing time of a couple of new, younger defensemen (Keith Aulie and Carl Gunnarsson).  There will be pictures...

Star-divide

A few notes on methodology: I went by game-by-game TOI data, but to get a good idea of player usage I split them into even-strength, powerplay, and shorthanded minutes.  Also, to iron out the noise and make for a clearer chart I used polynomial trendlines; plus, I think the trendlines will closer mirror the observations and reactions a coach might make.  I included a trendline for the highest TOIs among the members of the defense unit, so as to provide a better idea of where each defenseman stands in the team.  Finally, I excluded players that played fewer than 20 games, though if there is a player that filled a gap I might throw him in there.

*Note: In the case of all of these charts, click on them for a larger version.

*Note: For TOI, 1.00 = 24:00, 0.75 = 18:00, 0.50 = 12:00, 0.25 = 6:00, and so on.

Even Strength

Getting_defensive_toronto_es_2010-11_medium

That's your general chart, so you can bounce the various TOI's off each other.  What might be more interesting are the "mainstays" and "movers and shakers" charts...

Even Strength: Mainstays

Getting_defensive_toronto_es_mainstays_medium

Few surprises here, though it does emphasize how important Beauchemin and Kaberle were.

Even Strength: Movers and Shakers

Getting_defensive_toronto_es_movers_shakers_medium

As you can see, Aulie and eventually Gunnarsson filled the gaps when Beauchemin and Kaberle left here, while Phaneuf moved up even higher.

Powerplay

Getting_defensive_toronto_pp_2010-11_medium

Powerplay: Mainstays

Getting_defensive_toronto_pp_mainstays_medium

Not surprisingly, Kaberle was almost always the team leader in PP TOI, and when he left Phaneuf became the main man.

Powerplay: Movers and Shakers

Getting_defensive_toronto_pp_movers_shakers_medium

Beauchemin was a bit of an every-minute D, but his PP numbers dipped a bit about 25 games before he was traded in favor of Brett Lebda.  Lebda's brief moment in the sun was soon eclipsed by Gunnarsson.

Shorthanded

Getting_defensive_toronto_sh_2010-11_medium

Shorthanded: Mainstays

Getting_defensive_toronto_sh_mainstays_medium

As you can see, Kaberle and Lebda are not exactly relied upon for their shutdown D.

Shorthanded: Movers and Shakers

Getting_defensive_toronto_sh_movers_shakers_medium

At least initially, it was not Phaneuf who took up the slack when Beauchemin left, but Aulie.  Since then, it appears that they've split the difference.

Combinations (data from Dobber Hockey; min. 5% of team's total strength time)

%ES Ice Time Combination
20.71 Tomas Kaberle - Luke Schenn
14.09 Dion Phaneuf - Francois Beauchemin
11.35 Dion Phaneuf - Keith Aulie
9.44 Luke Schenn - Carl Gunnarsson
7.79 Mike Komisarek - Carl Gunnarsson
5.22 Mike Komisarek - Francois Beauchemin

%PP Ice Time Combination
22.01 Tomas Kaberle
17.04 Tomas Kaberle - Dion Phaneuf
10.63 Dion Phaneuf
6.32 Dion Phaneuf - Carl Gunnarsson
5.61 Luke Schenn - Carl Gunnarsson
5.46 Luke Schenn - Francois Beauchemin
5.31 Dion Phaneuf - Brett Lebda

%SH Ice Time Combination
24.43 Luke Schenn - Carl Gunnarsson
12.02 Luke Schenn - Francois Beauchemin
8.42 Dion Phaneuf - Francois Beauchemin
7.43 Mike Komisarek - Francois Beauchemin
6.17 Luke Schenn - Keith Aulie
6.17 Dion Phaneuf - Keith Aulie

As this is the first in the series, I'd like some feedback on the structure.  Would you rather I omit the first charts with all the TOI's, and just use the "Mainstays/Movers" charts?  Or keep all three?  Would you like the TOI numbers as well?  Month-by-month rather than game-by-game, perhaps?  I purposely left them out to avoid clutter (for relatively the same reason I used the polynomial trendlines). 

P.S.  Shoot, first try and I missed a player (Mike Komisarek).  Komisarek's ES TOI figures have hung around 13 minutes (or rough 0.50 on these charts) all year, and his PP TOI is almost non-existent.  As for SH TOI, it has fluctuated from near the high-end on the team early in the year, to very low in the middle of the year, rebounding a bit with Beauchemin's departure, and now back down again to the previous lows.

Comment 13 comments  |  0 recs  | 

Do you like this story?

Comments

Display:

At first I thought omitting Komi was Leafs fan wish-fulfillment, and then I wondered why Lebda was still there…

What I found confusing was:

1. the charts didn’t show absences other than by trade (e.g. Phaneuf was out midseason with injury, Aulie spent a lot of time in the minors, Lebda was benched a LOT). It would have been interesting to see how others’ numbers moved in Phaneuf’s absence, for example.

2. why on some charts do the numbers for everyone dramatically climb or plunge at the beginning or end? Just looking at your first chart, how can it be that every D-man’s TOI plummeted the last few games?

Very interesting series, and very savvy to start with the Leafs.

"[Phil Kessel]'s as streaky as a flipped coin" - Shift

by The '67 Sound on Apr 19, 2011 8:28 AM EDT reply actions  

I think it’s the polynomial fit and I’m not sure why that was done instead of just presenting the raw data.

Pension Plan Puppets*
* Blog contains less than 2% puppet content by weight.

by Chemmy on Apr 19, 2011 9:36 AM EDT up reply actions  

Yeah, I’m not really a fan of the polynomials; they leave such huge tails at the ends of the series that it becomes distracting. Even worse, they make it look like Aulie was getting more shorthanded ice time than anyone else (or even the “D Unit High”) during games (40-50) where he was actually playing in the AHL.

I've been looking at the sky

by Back In Black on Apr 19, 2011 10:47 AM EDT up reply actions  

Instead of polynomial you can try a moving average which smooth for game where a player is benched or has an injury.

by jeffgm on Apr 19, 2011 10:54 AM EDT up reply actions  

Yeah, I think next time I’m going to get the gaps in the lines better to avoid that. It really got thrown off by the fact that Aulie came in and basically took on a full shorthanded TOI load for a few games.

Does he call it Luongo underwear?

Co-Manager at Behind the Net

by Bettman's Nightmare on Apr 19, 2011 12:44 PM EDT up reply actions  

The problem with the raw data is that it gets too noisy, particularly if I have about 5 or 6 of them in one chart. I also don’t want their TOI data to overreact to a single game where they might have taken a bunch of penalties, or maybe got nicked up. I’m going to try the moving average, though, and see how that works.

Does he call it Luongo underwear?

Co-Manager at Behind the Net

by Bettman's Nightmare on Apr 19, 2011 12:41 PM EDT up reply actions  

if you're using excel

then are you stuck with a 6th degree polynomial?

I mean you can increase the degree and the approximation will be more accurate. If you got up into the 10th+ degree range it’d be pretty spot on without the noise.

"Success is the ability to go from one failure to another with no loss of enthusiasm."

- Sir Winston Churchill

I'm pretty sure he's talking about the Leafs.

by Steve Burtch on May 7, 2011 6:48 PM EDT up reply actions  

Everything swings with the kinds of ice time available, so if in the last couple of games the Leafs spent a lot of time with special teams situations, then you’ll see that upswing (note that ES TOI actually goes down at the end).

Totally agreed with #1, and I’ll go back and fix up the charts a little better.

Part of the reason I started with the Leafs is because they had so many drastic changes, and they weren’t in the playoffs (so initially, we’ll see a lot of non-playoff teams in the first bunch). I plan on extending the analysis into the playoffs for playoff teams to see how things might change there, too.

Does he call it Luongo underwear?

Co-Manager at Behind the Net

by Bettman's Nightmare on Apr 19, 2011 12:49 PM EDT up reply actions  

I like the graphs and think instead of TOI you might use
- GA/60 per game
- Corsi or Fenwick per game

That way we can see a trend if say Phaneuf trend in GA is trending lower or higher.
Or if Aulie (a rookie) puck possession is improving, worsening or flat.

The last idea corsi per game trend maybe a good way to judge rookie improvement.

My 2 cents. Thx for the good work.

by jeffgm on Apr 19, 2011 10:44 AM EDT reply actions  

I actually really like using TOI here. The other stats would certainly provide us with different information, as in the examples you suggest. However, to then compare the players with each other would require incorporation of quality of teammates, quality of competition, zone starts, save %, etc. TOI gives a nice, unclouded picture of the coaching staff’s evaluation of who should get the most minutes.

I've been looking at the sky

by Back In Black on Apr 19, 2011 10:54 AM EDT up reply actions  

Its an interesting outlook but of course has its own draw backs. Aulie was being played over his head for developmental reasons and Phaneuf was bailing him out. This has been shown by advanced statistics and (dare I say it) watching the game.

I am drinking the Kule-aid!
Certified Kule lover!

by BCapp on Apr 19, 2011 3:28 PM EDT up reply actions  

I'd actually like to see

faceoff data if it’s available, but that’s sort of hard to track at the game by game level unless you can grab it from timeonice.com… which I can’t remember off the top of my head.

"Success is the ability to go from one failure to another with no loss of enthusiasm."

- Sir Winston Churchill

I'm pretty sure he's talking about the Leafs.

by Steve Burtch on May 7, 2011 6:50 PM EDT up reply actions  

Okay, I made the suggested changes to Aulie and Phaneuf’s TOI trendlines; I left Lebda’s, because the gaps are small enough to not substantially change what I’m trying to show.

Does he call it Luongo underwear?

Co-Manager at Behind the Net

by Bettman's Nightmare on Apr 19, 2011 6:01 PM EDT reply actions  

Comments For This Post Are Closed


User Tools

The finest Winnipeg Jets analysis on the internets

FanPosts


Managers

Hawerchuk_small Hawerchuk

Gary_bettman_bad_dreams_small Bettman's Nightmare

Grapes_small canadian texan

Howe_small TJCAPS

Editors

Ryan_small SO_RyanP

0_small maplestirup

Jets2_small arby_18