The intent here is to present evidence, not opinion. My challenge to you, the reader, is this: accept a few simple documented, verifiable facts about how Darryl Sutter works his bench.
Sounds easy, sure. But, when we’re done, you’re next challenge will be to adapt the world of advanced stats to these newly recognized facts. The effort there can never be complete; that will always be a work in progress. The facts are revealed in the NHL’s very own “Full Play By Play” logs kept by the League itself for each game played.
Before you think this will be super heavy in stats and discussion of same, fear not. This is about coaching. Indirectly, advanced stats will be affected.
For the strike-shortened season of 2012-2013, I tracked, by hand as the game played, the order and use of Forwards under Darryl Sutter. My efforts duplicated the “Full Play By Play” logs of that season, but doing it myself in real time lent a deeper understanding than can be gained easily by looking at the NHL tracking pages and Play By Play logs.
An example can be found here, from Kings.com/Schedule; choose a game Recap-click Box Score-click Full Play By Play: http://www.nhl.com/scores/htmlreports/20122013/PL020364.HTM
Just look for the “L.A. On Ice” column. At the end of this article are links to the Full Play By Play pages for every game from this past season, and some key playoff games from the past.
What you will see is a vertical list of players in the order they hit the ice thru the game. Each incident, whether hit, shot, icing or offside or other, every event is logged and shows the event itself and all players on the ice at the time of the event. Concentrate on places where the Coach’s choice is wide open, such as the beginnings of periods, after TV timeouts, and long runs (without special teams) at 5 on 5.
Here is the first page of the Play by Play log for the Kings v. Hawks, Game 7, 2014.
Variance among coaching styles and tactics matter. One size does not fit all. Advanced stats discuss play in certain situations, but usage is determined by Coaching style and Coaches decisions. There must be some adjustment, some allowance in Advanced Stats for how different coaches use their players in different ways or under different circumstances. As of now, there is not. As a result, many current advanced stats are misleading if not wholly irrelevant.
As an example: most coaches use an icing against the other team to get a line of scorers out there. Not Darryl Sutter. Here, we’re talking about “Zone Starts.” In the example linked first above and shown below, look at events #22 thru #32.
It shows Kopi on a PP with Carter, both top 2 Centers at once and they score a Power Play goal, followed by an even-strength shift from Stoll. Stoll’s line forces an icing against Calgary mid-shift. Stoll is not substituted for, he stays out to take the draw. Shortly after that, Fraser’s line, the 4th line, forces another icing. Again, Fraser is left out to take the draw deep in the Offensive zone, even though both Kopi and Carter have had at least one full minute of rest, and Carter’s line is next up in rotation.
The problem with the Zone Start stat is that Sutter does not adjust according to faceoff location, he’s usually just rolling in rotation. So where Quenneville would likely trot out Toews for an O-zone faceoff from an icing, or Vigneualt would have sent out Henrik, Sutter sends out whoever is next, regardless of zone, almost always. As in the cases of Stoll and Fraser above, Sutter didn’t even shorten their shifts in favor of “skill players.” He also hardly ever changes line order after or before TV timeouts, either. It’s egalitarian.
So the Zone Start thing for Sutter is very different than for others, we can unavoidably see.
Under Quenneville, an icing will probably show us the scorers. Under Sutter, we’ll see who is up next in line order, and how well the line before the faceoff had played. If the faceoff is not in our zone, that previous line did mighty damn well. Perhaps “Zone Finish” would be a more meaningful stat under Darryl Sutter.
Suddenly, if true, Zone Starts become nearly meaningless, or actually, only meaningful in varying degrees depending on context and coaching style. The stat is not possibly “universal.”
A “Zone Start” shows where the Kings previous shift ended. And not much else.
For the same reasons above, judging Quality of Opponent and Quality Of Teammate also just became way more challenging.
When Sutter can, when he most clearly has all options, it’s 4 lines in order. When it’s not 4 in order, it’s pretty much Power Play or Penalty Kill bleedover causing it, or rarely it is evidence of a player, or an entire line, under-performing so badly that Sutter is forced out of pattern to compensate for that player.
To illustrate: There are 12 Forwards. Forwards get nearly equal time 5 on 5, except for that odd mixed “Frankenstein shift” after a PK. And…
6 Forwards generally perform one PK, and 6 guys don’t; the 3 tandems do, so that means the other 6 guys lose ice time by not doing PK. But, then the Frankenstein shift is where some non-PK guys make up some ice time.
So, effectively, at least 3 guys lose a shift of ice time with every single penalty. Or, the other 9 guys gained ice time.
Now you have what looks like a 4th line’s worth of minutes. The funny thing is they won’t have played on the same actual line together. They’ll just be the 3 guys with the least minutes. But I digress.
A typical Darryl Sutter play by play log will show a pattern of using 4 lines, in a particular order. This order of lines is held steady unless interrupted by a PK or a Power Play. After such special teams shifts, there is one “Frankenstein line” made new from rested guys, and after that the normal line order resumes.
Of course, exceptions exist, and can be found often. But I am discussing a proven preponderance of evidence, what Sutter most often does, what he wants to be able to do. Circumstances can force him away from his imperatives, but he returns to them as soon as allowed.
From this, we can see, in hard numbers kept by the League itself, the distinct pattern employed by Darryl Sutter. This pattern varies greatly from Coach to Coach, in clear and easily recognized ways. Players are, for Sutter, essentially sent out according to the next guy up, and not according to where a faceoff occurs. This is clear. This is written.
So, we can verify, and therefore quantify, style of usage of players, comparatively.
We’re The Fuk-ar-wee Tribe: Where Are We, And How Did We Get Here?
All hockey teams face the same sets of questions, about system, style, personnel choices, coaching methods, roster choices and player usage. But, all those questions really boil down to one, simple question. So you want to win a Stanley Cup… How?
If you were Dean Lombardi, you came into an existing framework. Choices had been made for you, and your flexibility was limited. You stated a 5-year rebuild was needed, and you gutted the team to make room for what you hoped would become a budding new culture of winning.
But, you’ve traded away your skill players, for picks and aging short-term vets. You have a nucleus of some super promising 19 year olds, but you need to win some games without any real scoring punch. Being the genius you are, you know you won’t suddenly score more goals, so you decide to allow fewer goals scored against you.
You do this because Defense Wins Championships. Here’s a chart of the past 20 Cup winners and their rank in GAA (Goals Against) that regular season.
2015-2016 Pitt 6th
2014-2015 Chicago 2nd
2013-2014 Los Angeles 1st
2012-2013 Chicago 1st
2011-2012 Los Angeles 2nd
2010-2011 Boston 2nd
2009-2010 Chicago 5th
2008-2009 Pitt 17th then 5th in playoffs
2007-2008 Detroit 1st
2006-2007 Anaheim 7th
2005-2006 Carolina 19th then 3rd in playoffs
2004-2005 Strike year-
2003-2004 Tampa 3rd
2002-2003 New Jersey 1st
2001-2002 Detroit 3rd
2000-2001 Colorado 3rd
1999-2000 New Jersey 7th then 1st in playoffs
1998-1999 Dallas 1st
1997-1998 Detroit 7th then 6th in playoffs
1996-1997 Detroit 2nd
If you are Dean Lombardi, you know this. You decide to eventually hire a coach who can instill the harshest of Defensive discipline. Enter Terry Murray.
You hope your core can learn to play tight checking, responsible hockey, so you develop systems that focus on possession. Playing safe positional hockey (read “boring/low scoring”) you drag thru some lean years to become the most consistent shut-down defensive team in the League.
But you still can’t score to save your life, or win enough games.
So now you have choices to make. You travel more than most teams. You have very young players, physiologically, yet you’ve chosen an aggressive and draining set of systems. You play a “heavy” game with emphasis on hits and physical battles for puck possession. You don’t have PK specialists; instead your best players take Power Play and Penalty Kill minutes.
The Defense becomes great, but still no scoring. And great D alone is just not enough.
Exit Murray. Enter Sutter.
Pressure, Possession and Percentages: The Sutter Method
You have to rest your scorers, so they can score. How? Kopi and Carter must be allowed to rest at some point in a game, but you really can’t do it with only one other line to use in a “3 lines plus fill-in 4th liners” rotation.
Asking Kopi to be a top PK man, and also skate every third shift, is asking too much. The best you can do is roll 4 lines when there are no penalties to kill. So 4 lines it is.
If you roll 4 lines, do you match which line plays against which? Some Coaches do. Sutter does not. He has said so, and the logs show it. I could post links to quotes, but rather, my entire proof here lies in the game logs.
We don’t have to interpret what Sutter means, or whether he is telling the truth. Misinformation is part of Coach’s job description, but play-by-play logs are actual empirical evidence.
So, Sutter scrambles and matches with the Defensemen, but forwards not at all. Or actually, Stevens matches his guys and Sutter does not. Sutter rolls 4 in order, except he might skip the 4th line a shift or maybe 2 toward the end of periods, especially the 3rd period. But, game in and game out, it’s 4 lines, until a Penalty Kill or a Power Play plus one extra shift to use guys that sat the Special Teams minutes, and then Sutter has his lines back in order.
And, if you are going to roll 4 lines in order, does it matter what order? Yes, it does, very much.
If we were playing against Drew Doughty, we should plan accordingly. Many Defensemen play almost 30 minutes a game, or half the game, or effectively every other shift. Holy chit BatMang, another underlining, it must mean something…
So if you put Kopi’s line out there, you’d see Doughty playing against him. Kopi gets tired, Doughty gets tired. They go off together and will come back out together later. But, if you put Carter out there right after Kopi, Doughty is tired, and unavailable. After someone like Doughty covered Kopi for a shift, we send out Carter next, and Doughty would have to be resting.
If you put another line in after Kopi, then Doughty will be rested by the time Carter comes out. Do you want Drew available for stopping both Kopi and Carter? Of course not. So you are going to play Kopi and Carter one after the other.
In reality, this is line matching before the game even starts. One of the functions of line order is actually matching the opponent’s defensemen in our own favor.
Consequently, you now have this: it’s either Kopi then Carter, or it’s Carter then Kopi. (Pick a number between 1 and 82, then go check the play by play logs for that game #. You’ll find them used as I describe in most cases. When they are not played consecutively, there is likely an interesting and obvious reason)
Next consideration is that you intend to rarely, but opportunistically skip the 4th liners. Late in the game, when you need a shift, of course you would. But that apparent contradiction, the idea that you play all 4 lines in order until you don’t, after I just told you Sutter rolls 4 lines almost all the time, is not a contradiction at all.
Rather than spending energy at every opportunity like a Quenneville, Sutter banks energy by saving his guys, and then when it’s a game-changing opportunity, or game-saving performance needed, he has the same Kopi available that he’s just been saving the whole game, for just that one important moment. As in, how-you-say, Overtime 3 on 3…
Quenneville only plays his 4th line pretty much when he is forced to, forced by top guys being tired during a long run of play without a whistle and stoppage. Sutter, nope. As I said, he generally skips the 4th line but once if at all late in the 1st and 2nd, and a bit more than that in the 3rd period.
When Darryl Sutter does skip the 4th line, it is likely not so he can get his 3rd line out there. No, he skips that line rarely, but for the same reason others do more often: to get his best players out there to make a difference in the game. Then, would it be wise to put the 3rd line after the 4th line in rotation? No, it wouldn’t. Why skip the 4th line to get the 3rd line out?
So, you now should put the 4th before either the 1st or 2nd line, so when you skip 4, you get either 1 or 2.
With these two rules, evidenced by the play by play logs, we now have our line rotation options.
Its 1,2 or it’s 2,1. And then, it’s either 4,1,2 or it’s 4,2,1. Must I say where 3 goes?
So if you are looking thru the play by play logs, you are looking for (1)Kopi, (2)Carter, (3)LeCav/Shore/3rd line Centers, and then (4) Andreoff/other 4th Center. And that is just what you will find, unless you find it’s equivalent of (2)Carter (1)Kopi (3) Lecav/Shore/Other (4)Andreoff/Other.
In some games, Andreoff is treated as #3 and Shore relegated to being skipped, in the #4 slot. In that case, Shore is counted as #4 and Andreoff #3. Through the 1st period we can see Sutter deciding which line is “on” and which line is playing poorly in that game. Nonetheless, line order evolves to that as expected. He alternates plugging in Shore and Andreoff until settling on an order of Kopi/Carter/Andreoff/Shore thru the 1st period and onward.
Rolling 4 lines is a fact. Line order discipline is a fact. And with just 2 major considerations, line order choices become self-defined, and are done so most simply.
The “why” of it we must deduce; my belief is that it speaks to long-term strategy. Sutter plays the long con. He’s preserving the energy level of his top players for late in the game, and he’s saving minutes per game to be stronger late in the season. When considering the Kings style and game plan, it becomes even more obvious.
What’s My Line? Musical Chairs Centers
We know what Sutter has done. We know how he uses Forwards. He tells us, and we see it. Somehow, most of this escapes visiting pundits; “Sutter the line magician, the scramble-mad coach searching for the magical new line combo.”
Most often, perceived “radical line changes” are merely because one Center got caught on an extended shift with a second set of Wingers who had changed. So one Center is out too long and two Wingers are out with not-their-usual Center.
The Center that did not get to come out is still on the bench ready for his turn, so he finally goes out, but by then it’s with the next set of Wingers ready to go, the ones that would have come out after him.
So if a Center gets caught out, rather than skip anybody, the wingers just go out with a different Center. The Wingers play in the same order as they had been, and so do the Centers, just not with each other on the same lines.
It’s not line scrambling, it’s more like rows of Musical Chairs 3 across, where the Wingers stayed seated but every Center moved back one row. It’s always the “next guy up.”
Los Angeles Ram-ifications
Earlier, you read this:
“A typical Darryl Sutter play by play log will show a pattern of using 4 lines, in a particular order. This order of lines is held steady unless interrupted by a PK or a Power Play. After such special teams shifts, there is one “Frankenstein line” made new from rested guys, and after that the normal line order resumes.”
If the goal is to use 4 lines in order so everybody gets the max rest, you do two things, at least.
First, you give up the idea of matching forward lines against your opponent. You don’t care. You have your order, and you let the opponent match however they want. This means your 4th line must be able to defend. And since talent costs money, whether it’s D talent or O talent, you have a bit less to spend on top guys because you need competent 4th liners to play 10 to 13 minutes a night against whomever the other coach wants.
So Clifford sees Getzlaf or whoever, Nolan, or Mersch, or anybody, sees Toews or Kane, or anybody…
Your 4th line must hold it’s own. If they break even, it’s a win. Now we glimpse the need for balance on the 4th line to include at least one solid checker, plus a solid defending Center. While Salary Cap may prevent 4 lines of scoring, it cannot be allowed to prevent 4 lines of defense.
From this we appreciate the flexibility built in to Sutter’s method. If he has tandems for lines, then the third player can be chosen according to need; opponent size or speed, injuries on your own team, but especially anticipated matchups chosen by the other Coach. This method of line building is well-documented.
You can help your 4th line, you can “protect” players on weaker lines, by matching your D against their opponents. Match Forwards no, match D-men, yes.
The second thing rolling 4 lines does is impact your PK pairs. You can’t send out Kopi and Carter together to PK really, you just got your 2 top Centers tired at the same time. You’d have to scramble in a Center to take a shift for one of them, which scramble leads to other substitutions, also. So you avoid that somehow. How?
You make your PK pairs play on the same line, if you can. They go out together full strength, and for PK. Same order, same linemates, no confusion.
This is what we have seen, predominantly. Carter and Toffoli are played the most together, for each player, 5 on 5, and they also make a PK tandem.
Brown and Kopitar have been a tandem since they have both been Kings, meaning they played together both 5 on 5 and for the PK. This they did in the playoffs, most reliably. Over time and acquisitions (Gaborik, M.), the Brown/Kopi tandem has had decreased time together 5 on 5 to now just 36% of Brown’s time. (article with graphs of Ice Time percentages by player)
When the Kings were setting up the Reverse Sweep against San Jose, they lost 3 games with Brown and Kopi skating on different lines 5 on 5.
When Brown and Kopi were re-united 5 on 5, the Kings then went on to win 4 in a row. So there.
If Brown and Kopi don’t get tired at the same time, they’re not likely to be ready to go again at the same time, either. And it’s hard to play together only at certain times and not at other times. It’s like expecting two perpendicular orbits of fatigue to meet in sync. But, in 2015-2016 Kopi and Brown were not on the same line 5 on 5 as much as Carter and Toffoli. Why?
Trevor Lewis, that’s why. Or actually, because of Jarrett Stoll. Lewis with Stoll was once our #1 PK unit, and they skated on the same line, too. (Coincidence? No.) But then Stoll left, and Sutter needed to shop for a new tandem partner that could play 5 on 5 as well as PK with #22.
That shopping did not quite ever succeed. The Kings never successfully replaced the Stoll and Lewis PK tandem. As I said, in 2012 and 2014 the first out for a PK was Stoll and Lewis, then in 2015-16 the first out was changed to Kopi and Brown. Since Kopi and Brown did not also skate on the same line 5 on 5 most of the time, this PK tandem solved one problem but created another, that being the makeshift (pardon the pun) line of guys playing together temporarily.
Sutter mostly settled on Pearson as a PK tandem with Lewis, but not as a 5 on 5 tandem, so now you had 2 lines worth of guys mixing up the line constitution and order of being played. The 5 on 5 stranded Brown from Kopi, and the PK stranded Pearson from his usual linemates. Brown got used to “cover” for weaker 3rd and 4th line Centers at 5 on 5, and Pearson lost ice time on his usual line 5 on 5 when he was tired from the PK pairing.
Contrast this with the gold standard of line construction; the 2012 L.A. Kings. Sutter had 3 main PK tandems. He had 3 main even-strength tandems. They were the same.
Stoll/Lewis, Kopi/Brown and Richards/Carter killed penalties. These tandems skated together 5 on 5, and killed penalties together. Stoll and Lewis were the first tandem out for a PK, and coming out of the PK it was easy to ice an intact line.
So, the “ideal” for Sutter is tandems that can both PK and play 5 on 5 together, so he can roll 4 lines in order even thru Penalty Kill situations.
If you want to discuss his players, you need to understand the objectives of their deployment. And, since roles and expectations vary greatly from game to game, from Coach to Coach and Team to Team, non-tailored Advanced Stats tell a story that does not translate equally among varied characters. If you want to discuss Sutter’s coaching, you really have to start with the PK tandems and the commitment to using 4 lines.
That’s what he does, that’s what we should do. Seems obvious, but to judge him and his players you have to at least understand his bench-use approach and tactics, no?
Would Darryl Sutter use players just to make stats look good? Does he decide how to use players to make stats look bad? No, it’s neither. Sutter has a method and the current stats do not tell the actual story.
So we see now; usage can impact some advanced stats more than competence of play.
Zone starts are of varied meaning from team to team, from Coach to Coach. Quality of Teammate and Opponent must be evaluated in context of usage. The full effect on advanced stats is uncharted, but until coaching styles are identified and quantified as to effect, we may run low on grains of salt with which we take many of the advanced stats.
And, we haven’t even started talking about compensating for the variance in systems used from team to team, from coach to coach. Maybe next time I feel like writing another novel, I’ll start thinking about it.
For now, this is enough to get us started.
Game 1, Canucks 2012 http://www.nhl.com/scores/htmlreports/20112012/PL030151.HTM
Game 4, Phoenix 2012 http://www.nhl.com/scores/htmlreports/20112012/PL030325.HTM
Game 1, Devils 2012 http://www.nhl.com/scores/htmlreports/20112012/PL030411.HTM
Game 4, San Jose 2014 http://www.nhl.com/scores/htmlreports/20132014/PL030184.HTM
Game 7, Hawks 2014 http://www.nhl.com/scores/htmlreports/20132014/PL030327.HTM
Regular Season 2015-2016